ijms winter 2006 web

VOLUME 175 NUMBER 4 OCTOBER, NOVEMBER, DECEMBER 2006 CONTENTS INCLUDE:

Heart Failure: Elevated BNP Home vs Hospital Initiated ThrombolysisStructuring Diabetes CareHypertension in Type 2 DiabetesFDG-PET Scanning : Cancer of the OesophagusHRT: Have We Changed?

IRISH JOURNAL OF MEDICAL SCIENCE • VOLUME 175 • NUMBER 4 1

This journal is indexed by Current Contents, Embase and is included in the abstracting and indexing of the Bio Sciences Information Service of Biological Abstracts. It is available in microfilm from University Microfilms Ltd.

All communications to the Editor should be addressed to: Irish Journal of Medical Science, 2nd Floor, Frederick House, 19 South Frederick Street, Dublin 2 Tel: 00353-1-633 4820 Fax: 00353-1-633 4918 Email: [email protected] Website: www.rami.ie www.iformix.com

Annual Subscription: Ireland and EU Countries E 156 Non-EU E 192 Single Copy E 42

Published by The Royal Academy of Medicine in Ireland

ISSN 0021-1265

Designed by Austin Butler Email: [email protected]

Editor-in-Chief David Bouchier-Hayes

Editor Brian Sheppard

Editorial Assistant Helen Moore

Editorial Consultant John Daly

Statistical Consultant Alan Kelly

Information Systems Consultant C Shields

JMS Doctor Awards Editor TN Walsh

Editorial Advisers OS Breathnach CF Donegan J Fenton ADK Hill F Howell H O’Connor S Sreenan S Tierney

EXECUTIVE OF THE ACADEMY President FD O’Kelly

General Secretary J O’Connor

Immediate Past President D Bouchier-Hayes

Members TN Walsh K O’Boyle E Kay D McCormack D Curtin

EDIT

ORI

AL

BOA

RD

� IRISHJOURNALOFMEDICALSCIENCE•VOLUME175•NUMBER4

Original PaPers5 ElevatedBNPwithnormalsystolicfunctioninasymptomaticindividualsat-riskforheartfailure:

AmarkerofdiastolicdysfunctionandclinicalriskSKaruppiah,FGraham,MLedwidge,CConlon,JCahill,CO’Loughlin,JMcManus,KMcDonald

14 FeasibilityandlongtermoutcomeofhomevshospitalinitiatedthrombolysisBMcAleer,MPSVarma

�0 Theexperiencesandattitudesofgeneralpractitionersandhospitalstafftowardspre-hospitalthrombolysisinaruralcommunityDTedstoneDoherty,JDowling,PWright,JCuddihy

�6 LowmolecularweightheparinprophylaxisindaycasesurgeryJShabbir,PFRidgway,WShields,DEvoy,JBO’Mahony,KMealy

30 ValidationofapointofcarelipidanalyserusingahospitalbasedreferencelaboratoryMCarey,CMarkham,PGaffney,GBoran,VMaher

36 Lipidloweringtargetsareeasiertoattainthanthosefortreatmentofhypertensionintype�diabetesMSherlock,DMylotte,JMacMahon,KBMoore,CJThompson

4� Structuringdiabetescareingeneralpractices:Manyimprovements,remainingchallengesSJennings,DLWhitford,DCarey,SMSmith

48 FDG-PETscanninginthemanagementofcanceroftheoesophagusandoesophagogastricjunction:Earlyexperiencewith100consecutivecasesVMalik,MKeogan,CGilham,GDuffy,NRavi,JVReynolds

55 Poorawarenessofcolorectalcancersymptoms;apreventablecauseofemergencyandlatestagepresentationATManning,RWaldron,KBarry

58 HRT:Havewechanged?OConlon,KMcKinney

6� AreviewofneonatalattendancesoutofhoursinaDublinmaternityhospitalLFAWong,KTLim,ATwomey,JMurphy

granD rOUnD66 Managementofspontaneousruptureoftheoesophagus(Boerhaave’ssyndrome):Singlecentre

experienceof18casesRPrichard,JButt,NAl-Sariff,SFrohlich,SMurphy,BManning,NRavi,JVReynolds

71 EndovascularaneurysmrepairinapatientwithahorseshoekidneyandimpairedrenalfunctionMSSajid,NAhmed,JTibbals,GHamilton

74 EncrustedCystitis—AnUnusualCauseofRecurrentFrankHaematuriaOO’Sullivan,OClyne,JDrumm

76 ComminutedfractureofthethoracicspineJPCashman,FLCarty,MRyan,KMahalingham

79 Smallbowelvolvulussecondarytoamesentericlipoma:acasereportandreviewoftheliteratureASMcCoubrey,RLEThompson

COrresPOnDenCe81 ExtensortendonruptureoffingerwhileplayingUileannpipe

SMAli,DO’Farrell8� UltravioletphototherapyrisksinIreland

CDupont8� Repetitivetranscranialmagneticstimulation:Apsychiatrictreatmentofthefuture?

HO’Connell,RGoggins,PGDoyle

BOOK reVieWs83 TheMenopause:whatyouneedtoknow

EVMocanu84 PrivatePractice—IntheearlytwentiethcenturymedicalofficeofDr.RichardCabot

CSBreathnach

CO

NT

EN

TS

ORI

GIN

AL

PAPE

R


INTRODUCTION B-type natriuretic peptide (BNP) is a member of the family of genetically distinct natriuretic peptides, synthesized and released by cardiomyocytes in response to myocyte stretch due to volume expansion and pressure overload.1-3 It is predominantly released from ventricular myocytes as NT-BNP (76 amino acid fragment) and BNP (32 amino acid fragment). In addition to natriuretic effects, BNP has been shown to relax vascular smooth muscle and exert anti-proliferative and antifibrotic effects.4, 5

Increases in plasma BNP concentration have diagnostic and prognostic implications in selected populations. This was shown initially in the presence of heart failure due to left ventricular systolic dysfunction (LVSD) and subsequently in both early stage and asymptomatic LVSD.6-14 More recently, the diagnostic and prognostic value of BNP has been underlined in a range of settings including the emergency room,10 in heart failure due to diastolic dysfunction,11 post-MI 12,13 and in an at-risk renal population.14 McDonagh and colleagues were among the first to explore the value of BNP screening in

the general population, and while they found BNP to be an independent predictor of mortality, such screening programs may be limited by the low screening return and event rates.15 Recent data from the Framingham Offspring Study support the role of BNP in predicting the risk of death, cardiovascular events, heart failure and stroke, independently of traditional risk factors.16 Therefore, while BNP is widely accepted in the evaluation of LVSD and heart failure, there are emerging data on its prognostic benefits in the general population.15-18 However, there has been little analysis of the drivers, and clinical implications of elevated BNP levels in a community population possessing risk factors for heart failure. The aim of this study was to investigate the prevalence, drivers, and clinical implications of elevated BNP levels in asymptomatic individuals with established cardiac risk factors for HF with proven normal systolic function.

METHODSThis was a collaborative study between St Vincents University Hospital Heart Failure Unit and a large General Practice in the South Eastern Area. The study

Elevated BNP with normal systolic function in asymptomatic individuals at-risk for heart failure: a marker of diastolic dysfunction and clinical riskABSTRACTBackground B-type natriuretic peptide (BNP) is widely accepted in the evaluation of left

ventricular systolic dysfunction and heart failure. However, little is known of the implications of elevated BNP levels in individuals with preserved systolic function (PSF).

Aims To investigate the drivers and clinical implications of elevated BNP levels in asymptomatic individuals with established PSF.

Methods We enrolled 154 individuals who all underwent physical examination, BNP evaluation and Doppler-echocardiographic studies. They were divided into those above and below the median BNP level (50pg/ml).

Results Independent predictors of higher BNP were older age, more severe left ventricular hypertrophy (LVH), reduced E/A ratio and ischaemic heart disease. Survival and multivariable analysis demonstrated more death and/or admission in those above the median BNP (HR: 4.79, p=0.007).

Conclusions Elevated BNP is the strongest, independent predictor of serious adverse cardiovascular outcomes in this population and requires closer clinical follow-up.

S Karuppiah, F Graham, M Ledwidge, C Conlon, J Cahill, C O’Loughlin, J McManus, K McDonaldHeart Failure Unit, St Vincent’s University Hospital, Dublin and *Carlton Clinic, Bray

ELEvAtEd BNP wIth NORMAL systOLIC fuNCtION IN AsyMPtOMAtIC INdIvIduALs At-RIsk fOR hEARt fAILuRE: A MARkER Of dIAstOLIC dysfuNCtION ANd CLINICAL RIsk

Springer and the Royal Academy of Medicine of Ireland enter into publishing partnership Irish Journal of Medical Science added to growing medical portfolio

London/New York, 17 November 2006

Springer, one of the world’s leading scientific publishers, has announced a partnership with the Royal Academy of Medicine of Ireland (RAMI) to publish their journal, the Irish Journal of Medical Science (IJMS), starting in 2007. The addition of this journal to Springer’s portfolio accentuates its growing strength in medical publishing.

Established in 1832, the Irish Journal of Medical Science is one of the first and foremost scientific journals in the Anglo-Celtic islands. With articles written by international specialists, it covers all branches of medicine and publishes papers relevant to the daily practice of the medical professional. IJMS will be published quarterly and will contain reviews, original articles and commentaries. The Editor-in- Chief is Professor David Bouchier-Hayes, a surgeon with a strong reputation in Ireland and internationally.

Dr. Fergus O’Kelly, President of the RAMI, said, “This is an exciting and most important time for the Royal Academy of Medicine of Ireland and its official organ, the Irish Journal of Medical Science. The Executive Board of the Royal Academy is most positive about our association with Springer.” Dr. John O’Connor, General Secretary, continued, “We look forward to this partnership and the many benefits that Springer’s publishing experience will bring to the journal.”

Christiane Notarmarco, Executive Editor at Springer London, echoed their sentiments, “We are proud to be associated with such a respected and historic body as the RAMI and are delighted to be supporting their mission. Through global distribution and electronic publishing on SpringerLink, IJMS will continue to grow.”

Springer will publish the Irish Journal of Medical Science in both electronic and print formats to serve the needs of librarians and readers around the world. The journal will be available via SpringerLink,

Springer’s fully integrated, online information platform. Both Online First™, a feature where articles are published online before they appear in print, and Editorial Manager, an online peer review and author submission system, will be fully implemented for the journal. The Springer Open Choice program offers all potential IJMS authors the option of publishing their articles using the open access model once they have been accepted for publication.

ABOUT RAMIThe Royal Academy of Medicine of Ireland (www.rami.ie) was formed in 1882, following an amalgamation of the four main medical societies in Ireland. At present there are over 1,200 fellows and members of the Academy representing 22 medical specialty sections. The primary role of the Academy is to provide a forum for the exchange of scientific information and to promote the academic discussions essential to scientific progress.

ABOUT SpRINgERSpringer (www.springer.com) is the second-largest publisher worldwide in the science, technology, and medicine (STM) sector and publishes on behalf of more than 300 academic associations and professional societies. Springer is part of Springer Science+Business Media, one of the world’s leading suppliers of scientific and specialist literature. The group owns 70 publishing houses, together publishing a total of 1,450 journals and more than 5,000 new books a year. The group operates in over 20 countries in Europe, the USA, and Asia, and has some 5,000 employees.

Media Contact: Joan [email protected] +49 (0) 6221 487 8130

Editorial Contact: Christiane [email protected] +44 (0) 1483 734620

PREs

s RE

LEA

sE

4 IRISH JOURNAL OF MEDICAL SCIENCE • VOLUME 175 • NUMBER 4

PREss RELEAsE


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


conducted using Cox Proportional Hazards regression statistical model. Receiver Operated Characteristic (ROC) analysis was carried out on different cutoff levels of BNP for the prediction of primary endpoints. The adequacy of these curves is assessed by comparing the Area Underneath the Curve (AUC) statistics (Area, P Value and 95% Confidence interval) and the sensitivity and specificity of selected BNP cutoffs were calculated. All analyses were carried out using SPSS Vs. 11 statistical software.

resUlts The study schedule is outlined schematically in Figure 1. A total of six individuals were found to have LVSD (population rate 3.6% of which 1 (0.6%) had a BNP <50 pg/ml) and along with a further three individuals (two with mild symptomatic heart failure

and a third with moderate mitral regurgitation) were excluded leaving a study population of 154 (age 67.4 ± 9.7 years, 57% male, 40% history of ischaemic heart disease, 62% history of long-standing hypertension, 7% history of valvular heart disease, 10% cardiac arrhythmia and 20% diabetes mellitus). Participants were followed-up for an average of 282 ± 149 days.

The median BNP for the study population was 50pg/ml and the cohort was divided into those above and below this value. There were 79 patients (51.2%) with BNP levels above the median value (EB Group) and 75 patients (48.2%) with BNP levels below the median value (C Group). The demographic characteristics of the population are presented in Table 1. Univariate analysis demonstrated that in addition to higher BNP, individuals in the EB group were older with a more

Figure 1—SCHEME OF INDIVIDUAL SCREENING AND INCLUSION

was approved by the St Vincent’s University Hospital Medical Ethics Committee and conforms to the Declaration of Helsinki.

All patients >55 years of age with at least one risk factor for the development of HF were deemed suitable for enrollment: long-standing hypertension, diabetes and coronary artery disease. Patients with a documented history of heart failure, documented left ventricular systolic dysfunction or any individual with documented non-cardiac conditions that could significantly alter BNP levels (e.g. pulmonary hypertension, pulmonary embolism) were excluded.

Database query from an estimated general practice population of 17,000 people identified 816 individuals considered suitable for enrollment. From those identified, approximately a third of the sample were randomly selected using a computer generated protocol and invited to partake in the study (n=254). Those giving informed consent attended the General Practice to give a detailed medical history and undergo physical examination, 12-lead electrocardiography, Chest X-Ray, phlebotomy for BNP level evaluation and urinalysis. These individuals were subsequently referred to the St Vincent’s University Hospital Heart Failure Service for Doppler echocardiographic studies.

BNP levels were measured using the Triage Meter point-of-care assay (Biosite, Ca).19, 20 Doppler echocardiographic analyses were performed using 2.5 and 3.5-MHz transducers (Hewlett-Packard Sonos 5500). Preserved left ventricular systolic function was defined as a LVEF ≥ 45% 21 Left ventricular hypertrophy (LVH) was assessed as a mean value of interventricular wall thickness and posterior wall thickness assessed at end-diastole. Values were subsequently graded as Grade 1 (mean value <11mm); Grade 2 (mean value 11-15.9mm); Grade 3 (mean value 16.0 to 19.9mm) and Grade 4 (mean value ≥ 20mm). Parameters of left ventricular diastolic function were made using standard methods and included: peak velocities of both the early (E) and atrial (A) diastolic filling and the derived E/A ratio; E-wave deceleration time (DT); isovolumetric relaxation time (IVRT). For individuals in atrial fibrillation, five Doppler complexes were sampled for measurements of E-wave deceleration time and IVRT at an average ventricular rate of >90 beats/min. Premature or aberrant ventricular complexes were ignored and modal values were accepted as being representative of diastolic filling.

Operators and physicians interpreting echocardiography were blinded to BNP score. Individuals identified at this stage with LVSD (LVEF <45%)21 and those with left ventricular chamber enlargement and normal ejection fraction were excluded from further analysis leaving a population with preserved systolic function (PSF).

The population was divided into two groups, those above and below the median BNP value for the total cohort. These two groups were termed ‘Elevated BNP’ (EB Group) and ‘Controls’ (C Group).

All participants were subsequently reviewed at the general practice, at a mean of 10 months following the initial analysis with regard to the primary endpoints of death and/or unplanned hospital admission for cardiovascular causes. The secondary endpoints evaluated were new diagnoses of heart failure and/or new cardiac diagnoses/events not requiring hospitalisation.

STATISTICAL ANALYSISComparisons between EB and C Groups were conducted using independent sample t-tests for continuous variables and Mann-Whitney test for non-normal distributions (two-sided, α = 0.05). Chi-squared analysis was used for discrete variables. Since the BNP values are positively skewed, the univariate and multivariable analysis to determine predictors of higher BNP was carried out using log-transformed BNP values. Data are presented as the mean value ± the standard deviation (SD) for continuous variables and absolute or relative frequencies for discrete variables. Univariate and multivariable analyses were conducted using binary (or binomial) logistic regression using death and/or unplanned hospital admission as the outcome variable. The multivariable model included theoretically reasonable variables and those with univariate p-values of ≤ 0.25. The likelihood ratio test was conducted to identify independent variables with low explanatory power. In addition, the Hosmer and Leme show goodness of fit indices were compared to assess the fit of each specification of the multiple regression model. The Kaplan-Meier product limit method of survival analysis was used to generate and adjust survival curves (death and/or unplanned cardiovascular hospital admission) using the EB Group variable. People lost to follow-up or known to be still alive were censored. The Mantel-Haenszel log-rank test was used to test the equality of the survivor functions. Multivariable analysis incorporating both continuous and discrete predictors of survival was

ELEVATED BNP wITH NORMAL SYSTOLIC FUNCTION IN ASYMPTOMATIC INDIVIDUALS AT-RISk FOR HEART FAILURE: A MARkER OF DIASTOLIC DYSFUNCTION AND CLINICAL RISk

S Karuppiah et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


Table 2DOPPLER ECHOCARDIOGRAPHIC CHARACTERISTICS OF THE EB GROUP (N=79) AND THE C GROUP (N=75)

VARIABLE TOTAL C GROUP (BNP<50pg/ml) EB GROUP (BNP>=50pg/mL) P

DT 263.01±57.04 244.81±52.45 280.29±56.13 0.00008

IVRT 92.02±39.29 84.65±35.43 99.10±41.68 0.020

E/A Ratio 1.00±0.46 1.19±0.44 0.83±0.42 0.000001

LVH Score 1.71±0.59 1.45±0.60 1.95±0.48 <0.000001

Table 1DEMOGRAPHIC CHARACTERISTICS OF THE EB GROUP (N=79) AND THE C GROUP (N=75)

Variable Total C Group(BNP<50pg/ml)

EB Group (BNP>=50pg/mL) P

Population 154 75 79

Age (Years) 67.4 ± 9.7 64.4 ± 10.0 71.1 ± 8.29 0.00001

Gender: Male:Female 88:66 47:29 41:37 0.206

BMI (Kg/m2) 27 ± 4 28 ± 4 27 ± 4 0.127

LVEF (%) 63 ± 9 64 ± 9 62 ± 9 0.157

BNP (pg/ml) 99 ± 178 23 ± 14 171 ± 226 N/A

Ischaemic Heart disease (Y:N) 61:93 24:51 37:42 0.060

Valvular Heart Disease (Y:N) 11:143 3:72 8:71 0.211*

Diabetes (Y:N) 31:123 17:58 14:65 0.444

Hypertension (Y:N) 95:59 51:24 44:35 0.116

Arrythmia (Y:N) 15:139 2:73 13:66 0.005*

HR (BPM) 69.8 ± 12.9 68.5 ± 11.3 71.0 ± 14.2 0.248

SBP (mmHg) 150 ± 22 147 ± 19 152 ± 24 0.162

DBP(mmHg) 84 ± 12 85 ± 11 84 ± 13 0.621

Pulse Pressure (mmHg) 44 ± 10 43 ± 9 45 ± 10 0.135

Creatinine (µmol/L) 90.5 ± 30.6 89.5 ± 26.6 91.6±34.3 0.685

ACE Inhibitor (Y:N) 50:104 22:53 28:51 0.418

AII Antagonist (Y:N) 14:140 6:69 8:71 0.646

Beta Blocker (Y:N) 72:82 32:43 40:39 0.322

Calcium Antagonists (Y:N) 31:123 18:57 13:66 0.243

Nitrate (Y:N) 28:126 13:62 15:64 0.790

Alpha Blocker (Y:N) 1:153 0:75 1:78 N/A

Diuretic (Y:N) 50:104 28:47 22:57 0.209

Antiplatelet (Y:N) 70:84 34:41 36:43 0.977

Statin (Y:N) 38:116 22:53 16:63 0.191

Anti-arrythmic (Y:N) 1:153 0:75 1:78 N/A

Warfarin (Y:N) 11:143 4:71 7:72 0.535*

Digoxin (Y:N) 10:144 2:73 8:71 0.099*

*Fishers Exact Test used. (Y:N)=Yes:No

Figure 2—kaplan Meier event-free (death and/or unplanned hosptialisation) survival curves of people with elevated BNP levels and normal systolic function (EB Group, normal EF and elevated BNP ≥50pg/ml, n=79) and Controls (C Group, normal EF low BNP <50pg/ml, n=75).

Figure 3—Receiver Operated Characteristic (ROC) curve using different BNP cutoff levels for the prediction of (primary endpoints) death and/or unplanned readmission. Sensitivities and specificities of selected cutoff levels are presented in Table 6.

frequent history of arrhythmia. The only difference in medications noted between the groups was in usage of digoxin, albeit in small numbers of individuals.

Doppler-echocardiographic analyses of both groups are presented in Table 2. They demonstrate longer DT and IVRT, reduced E/A ratio and higher LVH score consistent with more marked diastolic dysfunction and evidence of hypertensive heart disease in the EB Group. The differences in E:A ratio and LVH score were independent of age.

We analysed the dataset to determine the predictors of a higher BNP levels using log transformed values. Univariate results, presented in Table 3, suggest that the influencing variables are age, DT, E/A ratio, LVH and histories of ischaemia, hypertension and arrhythmia. However, multivariable analysis identified only four independent predictors of elevated BNP levels in the following order of decreasing significance: age (p<0.0001), LVH (p<0.0001), E/A ratio (p=0.026) and ischaemic heart disease (p=0.046).

During an average follow-up of 282 ± 149 days, 16.5% of individuals in the EB Group had reached the primary endpoints which comprised of four deaths (two cancer, one sudden death, one fall with neck fracture), nine emergency CV admissions (six arrhythmias, one preserved systolic function heart failure, one MI, one worsening angina). In contrast, 4% of the C Group had reached the primary endpoints comprising of one death (peri-operative for valvular heart disease) and two emergency cardiovascular admissions (unstable angina). Analysis of survival curves using death and/or unplanned cardiovascular readmission showed a significantly poorer outcome for people in the EB Group (p=0.004, Figure 2).

Univariate analysis using death and/or unplanned cardiovascular readmission as outcome measures is presented in Table 5. Multivariable analysis demonstrated that the only independent predictors of outcome were the categorical EB Group variable (p=0.032; HR: 4.0, 95% CI 1.12- 14.25) and ischaemic heart disease (p=0.046; HR: 2.9, 95% CI 1.08-8.70).

Secondary endpoints in the EB group were five diagnoses of PSF heart failure (four outpatient diagnoses and one during an emergency admission for heart failure), and three reported cardiovascular events not requiring hospitalization (two reported

symptomatic episode of atrial fibrillation and one reported incidence of angina). In the C group there were two new diagnoses of PSF heart failure and one episode of syncope not requiring hospitalization. In total, 24.1% of the individuals in the EB group had events in the follow-up period compared to the 8.0% in the C group (p=0.0004).

The utility of BNP as a predictor of primary endpoints (death and/or emergency hospital admission) was evaluated using receiver operating characteristic (ROC) curves (Figure 3). The area underneath the curve (AUC) statistics (Area 0.720, Std Error 0.49, P Value 0.002 and 95% Confidence interval 0.624 to 0.817) demonstrate that BNP is a highly significant predictor of the primary endpoints, although its overall performance as a clinical screen is somewhat limited by compromise between sensitivity and specificity. Accordingly, the sensitivity and specificity of a range of selected BNP cut-off values is presented in Table 6.


S Karuppiah et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


Table 6SENSITIVITY AND SPECIFICITY VALUES OF DEFINED BNP THRESHOLDS FOR THE PREDICTION OF PRIMARY ENDPOINTS (DEATH AND/OR EMERGENCY HOSPITAL ADMISSION)

BNP LEVEL THRESHOLDS SENSITIVITY (%) SPECIFICITY (%)

BNP: ≥25 pg/ml 100 31

BNP: ≥50 pg/ml 81 52

BNP: ≥75 pg/ml 63 68

BNP: ≥100 pg/ml 44 77

BNP: ≥125 pg/ml 25 82

DisCUssiOnThese data indicate that there is a high prevalence of elevated BNP levels in asymptomatic individuals with preserved systolic function possessing risk factors for heart failure. In this cohort, elevated BNP levels are associated with older age, LVH, diastolic abnormalities and ischaemic heart disease. Furthermore, a clinically important proportion of these individuals had serious adverse event within one year of follow-up. An elevated BNP level was the strongest independent predictor of serious adverse outcome in this population.

Data from the Framingham Offspring Study have shown that elevated BNP is strongly associated with adverse outcome in the general population.16 This observation suggests that BNP testing may be of benefit as a screening tool for occult structural and functional abnormalities of the heart, in particular, the left ventricle. 16,17,22,23

The Doppler-echocardiographic data from this study support this concept by demonstrating a close relationship between BNP and both LVH and parameters of diastolic dysfunction. Individuals in the EB Group had a significant increase in DT and IVRT and decrease in E/A ratio in comparison with the C Group. Additionally, the E/A ratio was found to be an independent predictor of elevated BNP levels in multivariable analysis. Moreover, the entire incident cases of heart failure during follow-up in the EB Group (6.25% rate) were associated with preserved systolic function.

Other studies have shown that BNP can identify individuals with asymptomatic LVSD 24, 25 although they provide no information on subsequent cardiovascular events. The data by Wang et al 16 demonstrate a high correlation between BNP levels and outcome in a community population, but give

few Doppler-echocardiographic data. Our study further develops these observations by identifying a high prevalence of diastolic abnormalities, strongly associated with elevated BNP levels, which is in turn associated with a high risk of subsequent cardiac events.

The clinical implication of identifying these at-risk individuals remains unclear. However, it likely includes the need to manage more intensively the risk factors of individuals with elevated BNP levels, paying particular attention to LVH, the strongest modifiable driver of BNP in the EB Group. This relates not solely to hypertension management, including 24-hour control of blood pressure, but also potentially to vascular compliance and myocardial fibrosis. Interestingly a lower rate of defined history of hypertension was observed in the EB Group compared to controls. However, there were no differences in blood pressures at baseline between the groups or in antihypertensive medications taken. It has been shown that BNP has anti-fibrotic properties 5 and may therefore be produced in response to an early fibrotic process. Such a process could explain the subtle abnormalities in diastolic function observed in the EB Group. In this regard, it is also of note that BNP can reduce the expression of aldosterone synthase, an enzyme involved in the production of aldosterone, a hormone closely linked to fibrosis.27,28 Furthermore, it is of note that the most common clinical events observed in this study were arrhythmia and diastolic heart failure, both of which can be potentially explained by an accelerated fibrotic process.

There are a number of limitations to this study which warrant further comment. Firstly, although representative of a large, asymptomatic, at-risk population, these are preliminary data in a relatively small sample. The ability of BNP to reliably predict outcome in this particular population or to add

Table 3UNIVARIATE PREDICTORS OF HIGHER BNP LEVELS FOR THE ENTIRE STUDY COHORT AT BASELINE (N=154)

Variable R2 F β (Unstandardised) P CI-Lwr CI-UprAge 0.172 31.602 0.415 (0.057) <0.00001 0.037 0.076Gender: Male=1 0.002 0.278 -0.043 (-0.115) 0.599 -0.546 0.316BMI 0.008 1.136 -0.088 (-0.027) 0.288 -0.077 -0.023EF% 0.006 0.855 -0.075 (-0.011) 0.357 -0.035 0.013DT 0.063 10.256 0.251 (0.006) 0.002 0.002 0.009IVRT 0.011 1.710 0.106 (0.004) 0.193 -0.002 0.009E/A Ratio 0.076 12.565 -0.276 (-0.792) 0.001 -1.234 -0.351LVH 0.242 48.403 0.491 (1.099) <0.00001 0.787 1.411Ischaemia 0.053 8.467 0.230 (0.621) 0.004 0.199 1.043Diabetes 0.009 1.330 -0.093 (-0.307) 0.251 -0.833 0.219Hypertension 0.031 4.873 -0.176 (-0.479) 0.029 -0.908 -0.050Arrhythmia 0.047 7.457 0.216 (0.964) 0.007 0.267 1.662Creatinine 0.011 1.686 0.106 (0.005) 0.196 -0.002 0.012

Table 5UNIVARIATE PREDICTORS OF PRIMARY END-POINT

VARIABLE MODEL χ2 -2LL P HR CI-LWR CI-UPRAge 0.099 143.02 0.753 0.992 0.943 1.044Male Gender 0.469 142.44 0.496 1.423 0.516 3.929BMI 0.461 132.02 0.497 1.040 0.928 1.167EF 0.072 143.04 0.789 0.993 0.943 1.046EB vs C group 7.274 135.37 0.015 4.790 1.363 16.838DT 0.002 143.12 0.966 1.000 0.992 1.008IVRT 2.12 140.69 0.151 0.992 0.980 1.003E/A 0.477 142.68 0.491 1.389 0.547 3.524LVH 3.920 139.07 0.048 2.294 1.009 5.215DD 0.309 142.81 0.579 1.307 0.508 3.364Hx of Ischaemia 6.601 136.76 0.016 3.665 1.269 10.583Hx of Diabetes 0.986 142.22 0.326 1.698 0.590 4.891Hx of Hypertension 3.25 140.05 0.081 0.414 0.154 1.114Hx of Arrhythmia 0.016 143.10 0.899 1.101 0.248 4.897Creatinine 1.111 131.64 0.288 0.992 0.978 1.007

Table 4FINAL MULTIVARIABLE MODEL OF PREDICTORS OF HIGHER BNP LEVELS FOR THE ENTIRE STUDY COHORT (N=154)

Variable R2 F β (Unstandardised) P CI-Lwr CI-UprAge 0.391 23.908 P<0.00001 0.309 (0.042) 0.000006 0.024 0.060LVH 0.389 (0.870) <0.000001 0.575 1.165E/A Ratio -0.149 (-0.427) 0.026 -0.803 -0.052Ischaemic 0.132 (0.356) 0.046 0.006 0.707


S Karuppiah et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


14. Mallamaci F Zoccali C Tripepsi G et al Diagnostic potential of natriuretic peptides in dialysis individuals. Kidney Int 2001;59:1559-63.

15. McDonagh TA, Robb SD, Murdoch DR et al. Biochemical detection of left-ventricular systolic dysfunction. Lancet 1998;351:9-13.

16. Wang TJ, Larson MG, Levy D, Benjamin EJ, Leip EP Omland T, Wolf P and Vasan RS. Plasma Natriuretic Peptide Levels and the Risk of Cardiovascular Events and Death. N Engl J Med 2004;350:655-663.

17. Nakamura M, Endo EB, Nasu M, Arakawa N et al Value of B type natriuretic peptide measurement for heart disease screening in a Japanese population. Heart 2002;87:131-135.

18. McKie PM, Rodeheffer RJ, Cataliotti A, Martin FL, Urban LH, Mahoney DW, Jacobsen SJ, Redfield MM, Burnett JC Jr. Amino-terminal pro-B-type natriuretic peptide and B-type natriuretic peptide: biomarkers for mortality in a large community-based cohort free of heart failure. Hypertension 2006; 47 (5): 874-880.

19. Dao Q, Krishnaswamy P, Kazanegra R et al. Utility of B-Type Natriuretic Peptide in the diagnosis of congestive heart failure in an urgent care setting. J Am Coll Cardiol 2001;37:379-385.

20. Morrison LK, Harrison A, Krishnaswamy P et al Utility of a rapid B-natriuretic peptide assay in differentiating congestive heart failure from lung disease in individuals presenting with dyspnoea. J Am Coll Cardiol 2002;39:202-209.

21. Remme WJ, Swedberg K. Task Force Report. Guidelines for the diagnosis and treatment of chronic heart failure. Task Force for the Diagnosis and Treatment of Chronic Heart Failure, European Society of Cardiology. Euro Heart Journal 2001; 22: 1527-1560.

22. Struthers AD. Introducing a new role for BNP: as a general indicator of cardiac structural disease rather than a specific indicator of systolic dysfunction only. Heart 2002;87:109-10.

23. Kranelund C, Gronning B, Kober L, Hildebrandt P, Steffensen R. N-Terminal Pro-B-Type Natriuretic Peptide and Long-Term Mortality in Stable Coronary Heart Disease. New Eng J Med 2005;352:666-675.

24. Yu CM Sanderson JE Shum IO et al Diastolic dysfunction and natriuretic peptides in systolic heart failure: higher ANP and BNP levels are associated with restrictive filling pattern. Eur Heart J 1996;17:1694-702.

25. Lubien E, De Maria A, Krishnaswamy P et al. Utility of B-natriuretic peptide in detecting diastolic dysfunction: comparison with Doppler velocity recordings. Circulation 2002;105:595-601.

26. Yamamoto K Burnett JC Jougasaki M et al Superiority of Brain natriuretic peptide as a hormonal marker of ventricular systolic and diastolic dysfunction and left ventricular hypertrophy. Hypertension 1996;28:988-94.

27. Tsutamoto T Wada A Maeda K et al Effect of spironolactone on plasma brain natriuretic peptide and left ventricular remodelling in individuals with congestive heart failure. J Am Coll Cardiol 2001;37:1228-1233

28. Holmes SJ, Espiner EA, Richards AM, Yandle TG, Framoton C. Renal, endocrine and haemodynamic effects of human brain natriuretic peptide in normal man. J Clin Endocrinol Metab 1993;76:91-96.

29. Redfield M, Rodeheffer MD, Jacobsen S et al. Plasma concentration of Brain Natriuretic Peptide: impact of age and gender. J Am Coll Cardiol 2002;40:976-82.

30. Wu AH, Packer M, Smith A, Bijou R, Fink D, Mair J, Wallentin L, Johnston N, Feldcamp CS, Haverstick DM, Ahnadi CE, Grant A, Despres N, Bluestein B, Ghani F. Analytical and clinical evaluation of the Bayer ADVIA Centaur automated B-type natriuretic peptide assay in patients with heart failure: a multisite study. Clin Chem. 2004;50:867-873

Correspondence to: Dr Ken McDonald, Director, Heart Failure Unit, St. Vincent’s University Hospital, Elm Park, Dublin 4, Ireland. Tel: 00353 1 2304629, Fax: 00353 1 2304639 e-mail. [email protected]


value over conventional and newer risk stratification methods must be examined in larger studies.

Secondly, more work is needed to define optimal BNP cut-off levels depending on screening objectives,17,22 population assessed,22,29 and BNP assay used.30 It is interesting to note that McDonagh et al demonstrated that a truly “normal” community population will have BNP values below 20 pg/ml,15 Wang et al identified values above 20 pg/ml for males and 23 pg/ml for females as conferring substantial risk 16 and the present work demonstrates 100% sensitivity for subsequent serious clinical events using a cut-off of 25pg/ml. Although asymptomatic, the present study population is more selected and has a higher prevalence of cardiovascular disease than the general community populations described in the aforementioned studies and may, therefore warrant a higher screening threshold. Therefore, the BNP cut-off of 50pg/ml in this setting may offer a reasonable compromise between sensitivity and specificity in predicting subsequent serious clinical events.

Thirdly, a cut-off of LVEF <45% may have allowed for inclusion of patients with established ischaemic heart disease and mild but important degrees of LVSD. Furthermore, the error of measurement of echocardiography may also have resulted in the inclusion of some patients with LVSD.

Finally, although the observation of diastolic abnormalities and increased risk among individuals in the EB Group is made in this study, we have not clarified in detail any causative mechanism. Clearly this population requires further evaluation to identify these mechanisms of increased risk and other means of reducing this risk apart from aggressive risk factor management.

COnClUsiOnsElevated BNP levels in the setting of preserved systolic function in asymptomatic, at-risk individuals is associated with older age, ischaemic heart disease, more severe LVH and abnormalities of diastolic function. Elevated BNP is the strongest, independent predictor of serious adverse outcome in this population. More work is required to clarify whether BNP testing may be a suitable screening tool for echocardiography and to establish its value as means of risk stratifying community populations with cardiovascular risk factors for intensive therapy and close clinical follow-up.

ACkNOwLEDGEMENTS

The authors wish to acknowledge the dedication and hard work of Tracey Reynolds in the organisation of this study.

REFERENCES1. Chen HH, Burnett JC. The natriuretic peptides in heart

failure: diagnosis and therapeutic potentials. Proc Assoc Am Physicians 1999;111:406-16

2. Cheung BMY, Kumana CR. Natriuretic peptides-relevance in cardiac disease. JAMA 1998;280:1983-4.

3. Maeda K, Takayoshi T, Wada A, Hisanaga T, Kinoshita M. Plasma Brain Natriuretic Peptide as a biochemical marker of high left ventricular end-diastolic pressure in individuals with symptomatic heart failure. Am Heart J 1998;135:825-32.

4. Clarkson PB, Wheeldon NM, McLeod C, Coutie W, Mac Donald TM. Brain natriuretic peptide: effect on left ventricular filling patterns in healthy subjects. Clin Sci 1995;88:159-164

5. Cao L, Fardner DG. Natriuretic peptides inhibit DNA synthesis in cardiac fibroblasts. Hypertension 1995;25:227-234.

6. Cowie MR, Struthers AD, Wood DA et al. Value of natriuretic peptides in assessment of individuals with possible new heart failure in primary care. Lancet 1997;350:1349-1353.

7. Tsutamoto T, Wada A, Maeda K et al. Attenuation of compensation of endogenous cardiac natriuretic peptide system in chronic heart failure: prognostic role of plasma brain natriuretic peptide concentration in individuals with chronic symptomatic left ventricular dysfunction. Circulation 1997;96:509-516.

8. Maeda K, Tsutamoto T, Wada A et al High levels of plasma brain natriuretic peptide and interleukin-6 after optimised treatment for heart failure are independent risk factors for mortality. J Am Coll Cardiol 2000;36:1587-1593.

9. Omland T, Aakvasg A, Vik-Mo EB Plasma cardiac natriuretic peptide determination as a screening test for the detection of individuals with mild left ventricular impairment. Heart 1996;76:232-237.

10. Maisel AS, Krishnashwamy P, Nowak RM et al Rapid measurement of B-Type Natriuretic Peptide in the emergency diagnosis of Heart Failure. N Eng J Med 2002; 347:161-167.

11. Maisel AS, McCord J, Nowak RM et al. Bedside B-type natriuretic peptide in the emergency diagnosis of heart failure with reduced or preserved ejection fraction. J Am Coll Cardiol 2003;41:2010-17.

12. Omland T, Ankvaang A, Bonarjee VV et al. Plasma brain natriuretic peptide as an indicator of left ventricular systolic function and long term survival post myocardial infarction: comparison with plasma atrial natriuretic peptide and NT-proatrial natriuretic peptide. Circulation 1996;93:1963-1969.

13. de Lemos JA, Morrow DA, Bentley JH et al The prognostic value of brain natriuretic peptide in individuals with acute coronary syndromes N Engl J Med 2001;345:1014-21.

S Karuppiah et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnIn recent years, several studies have shown that thrombolytic therapy used in acute myocardial infarction leads to a reduction in infarct size, improved left ventricular function and consequent improvement in mortality.1-4 Much of this work, however, has been carried out in large centres serving mainly urban populations and with ready access to invasive facilities for investigation and further treatment of ischaemic heart disease. Use of pre-hospital thrombolysis is limited, but encouraging improvement in left ventricular function has been shown following its use in an urban area.5

We have studied the feasibility and outcome of pre-hospital thrombolysis in an area that is almost entirely rural and these patients have been compared with a similar group who received thrombolysis in the coronary care unit of a district general hospital.

geOgraPHYThe study was carried out in the Erne hospital, a 212 bed district general hospital. There is a dedicated coronary care unit comprising 12 beds

and the hospital is the only acute hospital in County Fermanagh, which is located in the western part of Northern Ireland. The catchment area is over 800 square miles with communities up to 30 miles radius from the hospital. The total population is approximately 60,000. The nearest tertiary referral centre is located in Belfast, a distance of 90 miles.

Patients anD MetHODsPATIENT ELIGIBILITYOver a four-year period (1988-1992) all patients with a suspected acute myocardial infarction were seen by a mobile coronary care unit (MCCU). This unit was staffed by a physician (SHO level) and a senior coronary care trained nurse. Those patients whose major symptoms were of less than six hours duration were considered for thrombolysis. ECG evidence of myocardial infarction was required: ST-segment elevation of at least 1 mm in two or more standard leads and/or at least 2 mm in two or more praecordial leads.

After initial assessment by the MCCU staff, patients were randomly allocated to receive treatment with

Feasibility and long term outcome of home vs hospital initiated thrombolysisaBstraCtBackground Thrombolytic therapy improves mortality in acute myocardial infarction

especially in those who receive treatment early. Pre-hospital therapy can reduce the time to treatment.

Methods Open, randomized study of patients with acute myocardial infarction of less than six hours duration in a rural community. Pre-hospital thrombolysis was administered using a mobile coronary care unit (MCCU) and all patients received IV streptokinase.

Results Two-hundred and forty-eight patients were studied, 82 in the MCCU and 166 in the hospital group. The mean delay time to treatment was 136 minutes (MCCU group) and 196 minutes (hospital group) (p < 0.001). Reperfusion time was 116 minutes for the MCCU group and 118 minutes for the hospital group. Mortality at 30 days was 4.9% for the MCCU group and 15.7% for the hospital group (p = 0.014). Mortality at one year was 9.8% for the MCCU group and 23.5% for the hospital group (p = 0.009). Mortality for patients followed up to five years was 17.7% for the MCCU group and 35.2% for the hospital group (p = 0.005). There were no significant adverse events in either treatment group.

Conclusion Pre-hospital thrombolysis by MCCU is feasible and allows significant reduction in the delay time to treatment initiation. There are encouraging improvements in short- and long-term survival with no apparent reduction in safety profile.

B McAleer MPS VarmaCardiac Unit, Erne Hospital, Enniskillen, N. Ireland

FEASIBILITY AND LONG TERM OUTCOME OF HOME VS HOSPITAL INITIATED THROMBOLYSIS B Mcaleer & MpS VarMa

thrombolysis at home (administered by the MCCU staff) or alternatively to wait until transfer to the hospital coronary care unit. Randomisation was determined by using the on-call rota – one particular SHO commenced treatment at the patient’s home, whereas the other SHOs waited until arrival in the hospital CCU.

Patients were included irrespective of age, prior infarction, cardiogenic shock, pulmonary oedema, or life-threatening arrhythmias. Clinical grounds for exclusion were bleeding disorders, concurrent anticoagulant therapy, active peptic ulceration, recent stroke (<3 months ), recent surgery (<4 weeks), severe hypertension (BP>220/120 mmHg), and any patient whose life expectancy for other reasons was less than two years. The protocol was approved by the hospital ethical committee and appropriate patient consent was obtained.

PROTOCOLBefore initiation of therapy, all patients had a 12-lead ECG performed. The time of onset of the acute symptoms was recorded. Blood samples were taken for group and hold, full blood count, prothrombin time, partial thromboplastin kaolin time, fibrinogen degradation products, total creatinine kinase (CK), and CK-MB fraction.

All patients received streptokinase infused in a dosage of 1.5 million units in 100 ml of N saline over a 30-minute period, and the time of commencement of the infusion was recorded. This was immediately followed by an IV infusion of heparin in a dosage of 1000-1500 U/hr such that the PTTK ratio was maintained at 1.5 - 3 times the baseline value. After five days of heparin therapy, oral anticoagulation was substituted, and this was continued for at least three months in the vast majority of cases. Ancillary therapy (aspirin, betablockers, ACE inhibitors, etc) was prescribed along standard post-infarct guidelines.

RECOGNITION OF REPERFUSIONIn the absence of readily available facilities for coronary angiography, we used indirect evidence of reperfusion. The criteria used were:-

1. Rapid relief of chest pain and improvement in haemodynamic parameters.

2. Regression of ST-segment elevation (50% reduction in ST elevation).

3. Early CK and CK-MB peak.4. Occurrence of specific reperfusion arrhythmias,

i.e. idioventricular rhythm, ventricular fibrillation, ventricular tachycardia and ventricular ectopic beats.

POST-TREATMENT ASSESSMENTFollowing treatment, cardiac enzymes were taken six-hourly for the first 24 hours and then 12-hourly for the next 48 hours. ECGs were recorded every 15 minutes for the first two hours and then hourly for six hours. Haemodynamic parameters were recorded hourly for the first 12 hours and thereafter at the usual times. Some patients (<35% in each group ) subsequently had coronary angiography performed, but this was decided on the basis of clinical criteria and in many cases there was a prolonged delay. Ejection fraction was determined by radionuclide ventriculography, but only in those patients who had coronary angiography performed. All patients were followed-up for five years.

STATISTICAL METHODSFor quantitative variables, differences in mean values were tested by double-tailed t-tests employing the 5% level to distinguish statistically significant differences. In the case of qualitative variables, standard χ2 tests of significance (again using the 5% level) were applied, except where the small number of cases in an individual cell of the contingency table necessitated the use of Fisher's exact probability test.

resUltsDuring the study period 82 patients received thrombolysis via the MCCU and 166 were treated in the hospital coronary care unit. This represents 29% of all MCCU calls received in the study period. Table 1 shows the baseline characteristics of both groups and it can be seen that they are similar as regards mean age and percentage of male patients.

The numbers with a prior history of myocardial infarction are lower in the pre-hospital group but this is not a significant difference. The type of infarct sustained is similar in both groups. Although there is a higher percentage of anterior infarcts in the hospital group, this difference does not reach significance. Prior to treatment, similar numbers in each group had life-threatening arrhythmias. In most cases, these patients had ventricular fibrillation and all were successfully resuscitated before administration of lysis.

Table 2 shows the results of reperfusion data and the delay times to treatment. Delay time to treatment was defined as the difference between time of symptom onset and the time of commencement of the streptokinase infusion.

As expected, by administering thrombolytic therapy at home, we achieved a significant reduction in delay time to treatment (136 minutes vs 196 minutes), a difference


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


of one hour (p < 0.0001). The mean time to reperfusion (as judged by indirect evidence) was marginally reduced in those patients treated by the mobile unit. Over 92% of those treated by the mobile unit showed indirect evidence of reperfusion, as opposed to 81% in

those treated in the hospital coronary care unit and this difference was significant (p < 0.01).

In Table 3, comparisons of LV function are shown. However, ejection fraction was determined in under

Table 1BASELINE CHARACTERISTICS

MCCU HOSPITAL

Male (%) 76.8 76.5

Age (mean, years) 61.0 60.5

Age (range, years) 37 – 78 37 – 95

Previous infarction (%) 17.1 22.3

Anterior infarct (%) 40.2 41.6

Angina (%) 43.9 41.0

VF prior to Rx (%) 6.1 7.2

Table 2DELAY TIMES / REPERFUSION DATA

MCCU HOSPITAL

Delay time to Rx (mean, mins) 136 196 p < 0.0001

Distance from hospital (km) 16.1 15.8

Reperfusion (%) 92.7 81.3 p < 0.01

Reperfusion time (mean, mins) 116 118 p = NS

Table 3 CARDIAC ENzYME / LV FUNCTION DATA

MCCU HOSPITAL

Peak CK level (mean) 2474 2128

Time to peak CK (hrs) 13.9 16.4 P < 0.01

Peak CKMB level (mean) 234 247

Time to peak CKMB (hrs) 12.2 14.0 P = 0.012

Ejection fraction (%) 53 52

Heart failure on CXR ( %) 43 47

Table 4MORTALITY DATA

MCCU HOSPITAL

30 days 4/82 (4.9%) 26/166 (15.7%) P = 0.014

1 year 8/82 (9.8%) 39/166 (23.5%) P = 0.009

2 years 9/82 (11.0%) 46/166 (27.7%) P = 0.003

5 years 14/81 (17.3%) 58/165 (35.2%) P = 0.005One patient lost to follow-up in each group.

30% of each group and was rarely determined in those who died in the acute phase. This almost certainly is the reason for the non-significant difference in the ejection fraction data. The times to peak CK and CKMB are shorter in the MCCU group reflecting the earlier reperfusion.

Mortality figures are shown in Table 4 and it can be seen there are significant differences in both short term and long term results up to five years post treatment.

There were no significant bleeding events in either treatment group, but minor bleeding episodes (bruising at venepuncture, haematuria) were noted in 16% of the MCCU group and in 15% of the hospital group. Blood transfusion was not required. One patient in the hospital treated group suffered a stroke (infarct). There were no significant allergic or hypotensive events in either group.

DISCUSSIONSeveral large studies1-4 have shown that thrombolytic agents can reduce mortality from acute myocardial infarction, especially when given within six hours of the onset of symptoms. The importance of early treatment is well shown in the GISSI study,2 in which the mortality of those treated within the first hour was reduced by nearly 50%. Koren et al5 had previously reported significant improvement in left ventricular function following the use of thrombolysis in nine patients treated by a mobile coronary care unit.

Most of these papers, however, relate to work in tertiary centres in predominantly urban areas where facilities for invasive investigation and treatment are readily available. The majority of patients with acute myocardial infarction are treated in district general hospitals and our results relate to the use of pre-hospital thrombolysis in a community served by a district general hospital. By administering the agent at the place of onset of attack by means of a mobile coronary care unit, significant reduction in delay time to treatment has been achieved, as compared to a similar group who received treatment in hospital. Reperfusion was assessed noninvasively and is high in both groups. A number of studies have shown that noninvasive markers of reperfusion, such as ST-segment changes6 and creatinine kinase isoenzymes,7 are useful predictors of reperfusion. However other studies8,9 have shown that their usefulness to determine reperfusion early is questionable and

are only accurate if there is concordance of several criteriae, as is the case with our study.

Mortality is encouragingly low in both the short term and the long term for those treated in the pre-hospital phase and the difference is significantly lower than in the hospital group. Although the hospital treated group appears to have a high mortality when compared to the large studies of thrombolysis, this may be explained by the fact that patients in our study were included irrespective of age, the presence of cardiogenic shock or pulmonary oedema. A more valid comparative group might be that of MacLennan et al,10 who used intravenous streptokinase in a group of 50 patients in a tertiary cardiac centre with similar hospital mortality rates. Recently, other studies11-14 with pre-hospital thrombolysis have shown the feasibility of this treatment and the time gains that can be achieved. Morrison and colleagues published a meta-analysis12 of six trials of pre-hospital thrombolysis carried out by paramedics, general practitioners or mobile intensive care units in Europe or the UK. The “call-to-needle time” was reduced by 33 to 130 minutes and short-term hospital mortality was reduced by 17%.

Because of delays in admission to hospital, less than 30% of patients with chest pain are suitable for thrombolysis. Methods of improving the percentage of patients who might benefit from this treatment have been suggested: (a) reducing the delay in the admission of patients with chest pain, primarily by encouraging such patients to contact the emergency services directly, thus bypassing the general practitioner, (b) initiation of treatment in the pre-hospital phase by ambulance crews, general practitioners, or mobile coronary care units and (c) reducing in-hospital delays such as time spent in A&E departments awaiting bedspace.Although significant reductions in delay to admission have been achieved in some areas, notably Brighton, general practitioners in general are reluctant to relinquish their role in the management of acute myocardial infarction and they would contend that they are ideally placed to institute thrombolytic therapy at home. However, Rawles15 in a survey of GPs in the Aberdeen area has highlighted the problems that may occur should this approach be used - only 18% of GPs carried an ECG machine on call and only 30% carried a defibrillator. The vast majority stated that they were reluctant to use any drugs other than opiates or atropine in the treatment of patients with chest pain.



ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


In the Grampian Region Early Anistreplase Trial (GREAT)16 it was shown that general practitioner use of thrombolysis was feasible and safe and resulted in a time saving of about two hours in the delay to treatment and there was an associated decline in mortality. Indeed the five-year mortality results of the GREAT study17 are similar to our results using a mobile coronary care unit. However, some authorities, notably Julian,18 qualify their support for GP use of thrombolytics and question whether the GP should use thrombolysis without an electrocardiogram. Even Rawles19 in a follow-up study of the attitudes of general practitioners who had taken part in the GREAT study found that less than 20% of respondents had used thrombolysis in the previous year.

Administration of thrombolytic therapy via ambulance personnel has also been suggested, but many clinicians are still unhappy about the legal responsibilites with this approach, as well as the fact that this would almost certainly involve thrombolysis being administered without ECG evidence of infarction.

Treatment given by means of MCCU is a third alternative. Geddes20 in a review of 20 years of pre-hospital coronary care has highlighted the benefits of using such a service at a time before thrombolysis became widely used. Enthusiasm for such units is not widespread in the UK, except for Northern Ireland. However, the trial performed by the European Myocardial Infarction Project Group21 has shown that if the staff and equipment are available then pre-hospital thrombolysis by mobile coronary care units can reduce mortality, especially in those patients for whom the delay to hospital treatment is longest. Our study, based in a rural area where a delay to hospital treatment is invariably more than 60 minutes, suggests that a mobile coronary care unit can reduce mortality in the short and the long term.

In summary, pre-hospital thrombolysis using a mobile coronary care unit:

1. Is feasible and safe;2. Allows significant reduction in delay time to

treatment;3. Reduces mortality in both short term and long

term; Use of MCCUs should be developed further in order to optimize the number of patients who could benefit from thrombolytic therapy.

ACkNOwLEDGEMENTS

We thank the Nursing Staff at the Erne Hospital CCU for their help with this study; Dr E Turkington for the statistical analysis and Mrs JM Cecil for the preparation of the type script.

REFERENCES1. AIMS Trial Study Group. Effect of intravenous APSAC on

mortality after acute myocardial infarction: Preliminary report of a placebo-controlled trial. Lancet 1988; 1:545-549.

2. Gruppo Italiano per lo Studio della Streptochinasi nell’infarto miocardico (GISSI). Effectiveness of intravenous thrombolysis in acute myocardial infarction. Lancet 1986; 1:397-402.

3. ISIS Steering Committee. Intravenous streptokinase given within 0-4 hours of onset of myocardial infarction reduced mortality in ISIS-2. Lancet 1987; 1:502.

4. ISAM Study Group. A prospective trial of intravenous streptokinase in acute myocardial infarction (ISAM). Mortality, morbidity and infarct size at 21 days. N Engl J Med 1986; 314:1465-1471.

5. Koren G, Weiss AT, Hasin Y, et al. Prevention of myocardial damage in acute myocardial ischaemia by early treatment with intravenous streptokinase. N Engl J Med 1985; 313:1384-1389.

6. Hogg KJ, Hornung RS, Howie CA, et al. Electrocardiographic prediction of coronary artry patency after thrombolytic treatment in acute MI: Use of the ST-segment as a non-invasive marker. Br Heart J 1988; 60:275-280.

7. Tsukamoto H, Hashimoto H, Matsui Y, et al. Detection of myocardial reperfusion by analysis of serum creatine kinase isoforms. Clin Cardiol 1988; 11:287-291.

8. McCaliff MR, O’Neill W, Stack RS, et al. Failure of simple clinical measurements to predict perfusion status after intravenous thrombolysis. Ann Intern Med 1988; 108:658-662.

9. Kircher BJ, Topol E, O’Neill W, et al. Prediction of infarct coronary artery recanalisation after intravenous thrombolytic therapy. Am J Cardiol 1987; 59:513-515.

10. MacLennan BA, McMaster A, Webb SW, et al. High dose intravenous streptokinase in acute myocardial infarction - short and long term prognosis. Br Heart J 1986; 55:213-239.

11. Sauval P, Artigou JY, Cristofini P, et al. Pre-hospital thrombolysis with rcombinant tissue plasminogen activator in acute myocardial infarction (Fren). Arch Mal Coeur Vaiss 1989; 82:1957-1961.

12. Morrison LJ, Verbeek PR, McDonald AC, et al. Mortality and pre-hospital thrombolysis for acute myocardial infarction. JAMA 2000;283:2686-92.

13. Dubous-Rande JL, Herve C, Dubal-Moulin AM, et al. Pre-hospital thrombolysis in acute myocardial infarction: Preliminary results in the Val-de-Marne department (Fren). Arch Mal Coeur Vaiss 1989; 82:1963-1966.

14. Cazaux P, Leclercq G, Vahanian A, et al. Intravenous thrombolysis with tissue-type plasminogen activator (rt-PA) in the pre-hospital phase of acute myocardial infarction. 49 cases (Fren). Arch Mal Coeur Vaiss 1989; 82:1967-1971.

15. Rawles JM. General practitioner management of acute myocardial infarction and cardiac arrest: Relevance to hrombolytic therapy. Br Med J 1987; 295:639-640.

16. GREAT Group. Feasibility, safety, efficacy of domiciliary thrombolysis by general practitioners. BMJ 1992; 305:548-53.

17. Rawles JM. Quantification of the benefit of earlier thrombolytic therapy: five year results of the Grampian Region Early Anistreplase Trial (GREAT). Journal of the American College of Cardiology 1997. 30(5):1181-6.

18. Julian DG. Thrombolysis, the general practitioner, and the electrocardiogram. Br Heart J 1994; 72:220-221.

19. Rawles J. Attitudes of general practitioners to pre-hospital thrombolysis. BMJ 1994; 309:379-82.

20. Geddes JS. Twenty years of pre-hospital coronary care. Br Heart J 1986; 56:491-495.

21. The European Myocardial Infarction Project Group. Prehospital thrombolytic therapy in patients with suspected acute myocardial infarction. N Engl J Med 1993; 329:383-9.

Correspondence to: Dr B McAleer, Erne Hospital, Enniskillen, Co. Fermanagh BT74 6AY



ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnThe Cardiovascular Health Strategy has recommended that eligible patients receive thrombolysis within 90 minutes of alerting medical or ambulance services.1 Previous research in the Donegal area has shown a median call to needle time of 200 minutes, which clearly exceeds this guideline.2 The aim of DARTS was to determine the feasibility of the administration of domiciliary thrombolysis by rural general practitioners.4 Pre-hospital thrombolysis resulted in a median call to needle time of 62 minutes – a time saving of 142 minutes compared to a control group thrombolysed in the hospital (median call to needle time of 204 minutes). The report recommended that pre-hospital thrombolysis be extended to other areas where the recommended call to needle times cannot be achieved.

Few studies have assessed the general practitioners’ or hospital staff attitudes to pre-hospital thrombolysis. One study in the Grampian region concluded that general practitioners were more likely to use thrombolysis if they were encouraged to do so by the local cardiologist.5 A follow-up of this study6 to investigate the adoption of a

pre-hospital thrombolysis policy argued that the attitudes of hospital consultants were generally negative and that they lacked confidence in the general practitioners’ ability to read and analyse an elecrocardiogram (ECG). It was felt that this was a significant barrier to the adoption of a pre-hospital thrombolysis policy by the general practitioner as they felt that encouragement and support from local health authorities was necessary.

Given the previous findings, there is a need to examine the attitudes of general practitioners who took part in a pre-hospital thrombolysis study (DARTS) within an Irish context. If a pre-hospital thrombolysis policy is to be instigated, it is beneficial to be aware of the possible barriers to the use of thrombolytic treatment. The aim of this study sought to determine the participating general practitioners’ attitudes to pre-hospital thrombolysis following completion of DARTS. A questionnaire was designed to identify possible advantages for and barriers to the administration of pre-hospital thrombolysis in a rural community. Furthermore, the study extended previous work by investigating the attitudes of hospital staff including consultants and nursing staff.

The experiences and attitudes of general practitioners and hospital staff towards pre-hospital thrombolysis in a rural communityaBstraCtBackground In rural areas it is impossible for eligible patients presenting with acute

myocardial infarction (AMI) to receive thrombolysis within the recommended 90 minutes unless administered in the community by the general practitioner.

Aims The aim of this study was to describe the attitudes of hospital staff and general practitioners towards pre-hospital administration of thrombolysis.

Method General practitioners, consultant physicians and nursing staff participated in the survey.

Results General practitioners were convinced of the added benefits of administration of thrombolysis in the community and believed the hospital had a role to play. Likewise the hospital staff agreed with the benefits of pre-hospital thrombolysis. However, they felt that the decision to thrombolyse patients should be made in consultation with the hospital.

Conclusions Pre-hospital thrombolysis programmes must be continuously monitored and evaluated to identify important factors that may prevent wider use of thrombolytic treatment.

D Tedstone Doherty1, J Dowling2, P Wright1, J Cuddihy1

Dept of Public Health Medicine, NSE–NWA, Bishop Street, Ballyshannon, Co Donegal1; North West Immediate Care Programme2

THE EXPERIENCES AND ATTITUDES OF GENERAL PRACTITIONERS AND HOSPITAL STAFF TOwARDS PRE-HOSPITAL THROMBOLYSIS IN A RURAL COMMUNITY

MetHODAll general practitioners involved in DARTS were asked to complete a questionnaire designed to assess attitudes towards thrombolysis and problems that may be associated with administering pre-hospital thrombolysis. It contained questions relating to benefits and safety of pre-hospital thrombolysis, education, training and support in pre-hospital thrombolysis, and confidence in the treatment of acute myocardial infarction. All of the consultant physicians and nursing staff of the ED and CCU were administered questionnaires assessing the benefits and safety of pre-hospital thrombolysis.

The questionnaire relating to benefits and safety of pre-hospital thrombolysis was comprised of statements requiring participants to indicate agreement, disagreement, or neutrality. Questions relating to confidence in the administration of thrombolysis were on a five-point Likert type scale and general practitioners indicated how confident they were by circling the appropriate response from ‘not at all confident’ to ‘extremely confident’. Following completion of the questionnaires, participants were given the opportunity to elaborate on answers that they had given. The researcher recorded these responses by hand.

resUltsRESPONSE RATEA total of 82% (13 / 16) of the participating general practitioners completed the questionnaires. Of these, 38% (5 / 13) administered thrombolysis to patients in the community during the study period. All the consultant physicians (100%; n = 5), 54% (25 / 46) of the nursing staff from CCU and ED completed the questionnaires. Table 1, Table 2 and Table 3 show the percentage of the general practitioners’ responses to the individual questions, while Table 4 shows the percentage of the hospital staffs’ responses to the individual questions.

GENERAL BENEFITS OF THROMBOLYSIS AND EARLY ADMINISTRATIONAll general practitioners (100%) were convinced of the benefits of thrombolysis in the treatment of acute myocardial infarction and were convinced of the additional benefits derived from administration in the community at the earliest opportunity. A total of 97% (29 / 30) of the hospital staff were convinced of the benefits of thrombolysis and 3% (1 / 30) remained neutral. Ninety-four per cent (28 / 30) were convinced of additional benefits from early administration in

the community. Only 3% (1 / 30) disagreed with the additional benefits of pre-hospital thrombolysis and 3% (1 / 30) remained neutral.

TIME, COST AND SAFETYA total of 77% (10 / 13) of general practitioners felt that administering thrombolysis in the community would result in appreciable time saving and 85% (11 / 13) felt that they could make time to give thrombolysis. The majority of the general practitioners (92%; 12 / 13) and hospital staff (97%; 29 / 30) disagreed that thrombolysis was too expensive for general practice. All the general practitioners (100%) agreed that it was not too difficult to administer in general practice and 69% (9 / 13) agreed that it was not inconvenient for use in general practice. The majority of the hospital staff also agreed that eligible patients are not too difficult to diagnose for pre-hospital thrombolysis (90%; 27 / 30). A total of 61% (8 / 30) of general practitioners and 67% (20 / 30) of the hospital staff agreed that it was safe to administer thrombolysis in the community. One general practitioner disagreed while three hospital staff disagreed and 30% (4 / 13) of the general practitioners remained neutral, as did 23% (7 / 30) of the hospital staff.

EqUIPMENTA total of 92% (12 / 13) of the general practitioners had an ECG machine that they could use on call and all (100%) were willing to record an ECG in cases of suspected AMI. A total of 85% (11 / 13) of the general practitioners reported being able to interpret the ECG, while two reported that they would not be able to interpret the ECG recording. A total of 77% (10 / 13) of the general practitioners always carried the thrombolytic kit while on duty. Regarding defibrillation, 92% (12 / 13) of general practitioners had access to a defibrillator while on duty and 85% (11 / 13) agreed that they knew how to use a defibrillator.

TRAINING, EDUCATION AND SUPPORTTable 2 shows the percentage responses from the general practitioners in each of the categories for the individual questions pertaining to training, education and support of prehospital thrombolysis. In terms of the training received prior to the initiation of DARTS, 61% (8 / 13) of the general practitioners agreed that training was sufficient. However, 62% (8 / 13) agreed that more training in pre-hospital emergency cardiac care is necessary. A total of 62%(8 / 13) agreed that training in ECG interpretation is sufficient, but 54% (7 / 13) felt that more training in the use of ECG equipment was necessary. In terms of the telemetry

D teDStone Doherty et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


to transmit the ECGs to the hospital, just over half the general practitioners (54%; 7 / 13) felt that training was sufficient. The majority (62%; 8 / 13) also felt that more training in administering thrombolysis was warranted.

RESPONSIBILITY AND SUPPORTThe majority of the general practitioners (61%; 8 / 13) felt that the decision to thrombolyse the patient was entirely the general practitioner’s and over half (59%; 7 / 13) thought that the decision should not be made in consultation with the hospital. In contrast, over half of the hospital staff felt that the

Table 1GP ATTITUDES TOwARDS PREHOSPITAL THROMBOLYSIS. THE PERCENTAGE OF GENERAL PRACTITIONERS RESPONDING IN EACH OF THE CATEGORIES (N=13)

AGREE DISAGREE NEUTRAL

I am convinced of the benefits of thrombolytic treatment in acute myocardial infarction 100 (13)

I am convinced that there are additional benefits from giving thrombolytic treatment in the community at the earliest opportunity after symptom onset

100 (13)

From the point of view of my practice, giving thrombolytic treatment at home would not result in any appreciable time saving 15 (2) 77 (10) 8 (1)

I could make time to give thrombolytic treatment to patients with an acute myocardial infarction 85 (11) 8 (1) 8 (1)

Thrombolytic treatment is too expensive for use in general practice 92 (12) 8 (1)

Thrombolytic treatment is safe for use in general practice 61 (8) 8 (1) 30 (4)

Thrombolytic treatment is too difficult for use in general practice because it needs to be given intravenously 100

Thrombolytic treatment is too inconvenient for use in general practice because you may have to travel to the hospital with the patient 8 (1) 69 (9) 23 (3)

I do not have an ECG machine that I could use on call 8 (1) 92 (12)

I would be willing to record an ECG in cases of suspected AMI 100 (13)

I could interpret an ECG in cases of suspected AMI 85 (11) 15 (2)

The decision to thrombolyse a patient is entirely the general practitioner’s 61 (8) 31 (4) 8 (1)

The general practitioner’s decision to thrombolyse patients should be made in consultation with the hospital 23 (3) 59 (7) 23 (3)

I do not have access to a defibrillator 8 (1) 92 (12)

I know how to use a defibrillator 85 (11) 8 (1) 8 (1)

I always carry a defibrillator with me when on duty 92 (12) 8 (1)

I always carry a thrombolytic kit with me when on duty 77 (10) 15 (2) 8 (1)

I am not willing to use thrombolytic treatment unless encouraged to do so by the Department of Health and Children 8 (1) 77 (10) 15 (2)

I am not willing to use thrombolytic treatment unless encouraged to do so by the ICGP 77 (10) 23 (3)

I am willing to use thrombolytic treatment if it is promoted by the drug manufacturers for use in general practice 8 (1) 77 (10) 15 (2)

I am not willing to use thrombolytic treatment without further training 15 (2) 77 (10) 8 (1)

I am not willing to thrombolyse patients other than my own patients 92 (12) 8 (1)

decision to thrombolyse patients was not solely the responsibility of the general practitioner (53%; 16 / 30) and that the decision to thrombolyse patients should be made in consultation with the hospital (53%; 16 / 30). The majority of the general practitioners did not feel that support for the use of thrombolysis in the community from the Department of Health and Children (77%; 10 / 13), the Irish College of General Practitioners (77%; 10 / 13), or the drug manufacturers was necessary (77%; 10 / 13). Of the hospital staff, 97% (29 / 30) agreed that general practitioners should be allowed to administer thrombolysis with sufficient training, and that the hospital had a role to play in pre-hospital thrombolysis. A total of 92% (12 / 13) of the general

practitioners were willing to thrombolyse patients other than their own and 77% (10 / 13) were willing to administer thrombolysis without further training.

CONFIDENCE IN PREHOSPITAL CARDIAC CARE TREATMENTTable 3 shows the percentage of responses in each of the categories pertaining to confidence in the pre-hospital treatment of AMI patients. The majority (69%; 9 / 13) of the general practitioners reported that they were extremely confident in the ability to record an ECG, while 23% (3 / 13) were very confident and 8% (1) were moderately confident. Regarding interpretation, 46% (6 / 13) were moderately confident, 46% (6 / 13) were very

Table 2GENERAL PRACTITIONERS’ ATTITUDES TOwARDS TRAINING, EDUCATION AND SUPPORT. PERCENTAGE OF GENERAL PRACTITIONERS RESPONDING IN EACH OF THE CATEGORIES (N=13)


The training received prior to the initiation of the study was sufficient 61 (8) 15 (2) 23 (3)

I feel that more training in prehospital emergency cardiac care is necessary 62 (8) 23 (3) 15 (2)

I feel that the training in ECG interpretation was sufficient 62 (8) 30 (4) 8 (1)

I feel that more training in administering thrombolytic treatment is necessary 62 (8) 23 (3) 15 (2)

I feel that more training in the use of the ECG equipment is necessary 54 (7) 38 (5) 8 (1)

I feel that the training given on ECG telemetry was sufficient 54 (7) 31 (4) 15 (2)

I receive sufficient support from the project team in terms of training and education 54 (7) 15 (2) 30 (4)

I feel that there is a need for a refresher course in cardiac care 77 (10) 23 (3)

Table 3qUESTIONNAIRE ASSESSING THE PARTICIPATING GP’S CONFIDENCE IN THE ABILITY TO RECORD AND INTERPRET AN ECG AND TO ADMINISTER THROMBOLYTIC THERAPY. THE PERCENTAGE OF GENERAL PRACTITIONERS RESPONDING IN EACH OF THE CATEGORIES (N=13)

1a 2 3 4 5

My ability to record an ECG 8 (1) 23 (3) 69 (9)

My ability to interpret an ECG 46 (6) 46 (6) 8 (1)

My ability to transmit the ECG to the CCU 15 (2) 23 (3) 31 (4) 31 (4)

My awareness of the three indications to thrombolyse patients as outlined by DARTS 23 (3) 46 (6) 8 (1) 23 (3)

My knowledge of the contraindications to thrombolysis 8 (1) 46 (6) 23 (3) 23 (3)

My ability to administer the current thrombolytic agent 23 (3) 31 (4) 31 (4) 15 (2)

My ability to provide post thrombolytic Treatment 8 (1) 23 (3) 31 (4) 31 (4) 8 (1)

My ability to deal with possible adverse reactions 15 (2) 31 (4) 31 (4) 23 (3)

1a=not at all confident; 2=slightly confident; 3=moderately confident; 4=very confident; 5=extremely confident




ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


confident and one (8%) general practitioner was extremely confident. Confidence in the transmission of the ECG to the hospital showed that 15% (2 / 13) were not at all confident, 23% (3 / 13) were slightly confident, 31% (4 / 13) were moderately confident and 31% (4 / 13) were very confident. General practitioners reported that they were slightly confident (23%; 3 / 13), moderately confident (46%; 6 / 13), very confident (23%; 3 / 13) or extremely confident (23%; 3 / 13) in their knowledge of the three indications for thrombolysis as outlined by the study protocol. Knowledge of the contraindications to thrombolysis showed that only one general practitioner reported slight confidence, 46% (6 / 13) reported moderate confidence and 23% (3 / 13) reported very confident and extremely confident. A total of 23% (3 / 13) of general practitioners reported that they were slightly confident in their ability to administer the current thrombolytic agent, while 31% (4 / 13) reported moderate confidence, 31% (4 / 13) were very confident, and 15% (2 / 13) were extremely confident. Finally, a total of 15% (2 / 13) of the general practitioners were not at all confident in the ability to deal with adverse reactions to thrombolysis while 31% (4 / 13) felt slightly confident, 31% (4 / 13) felt very confident and 23% (3 / 13) felt extremely confident.

DisCUssiOnAcknowledgement of the benefits of rapid treatment for AMI patients was presented in the Cardiovascular Health Strategy and the report recommended that eligible patients receive pre-hospital thrombolysis within 90 minutes of alerting medical help.1 In rural areas, however, it is often impossible to achieve this guideline. DARTS was the first study in Ireland to introduce general practitioner administered thrombolysis. This study was a follow-up of general practitioners involved in DARTS and the aim was to investigate the attitudes towards pre-hospital thrombolysis of both the participating general practitioners and relevant hospital staff.

BENEFITS OF THROMBOLYSIS AND EARLY ADMINISTRATIONResponses to the questionnaire suggest that both the general practitioners and hospital staff believe in the efficacy of thrombolysis and of the added benefits of early administration. The general practitioners also felt that appreciable timesavings would be incurred if thrombolysis were to be administered in the community.

Table 4TABLE SHOwING THE PROPORTION % (N) OF RESPONSES FROM HOSPITAL STAFF IN EACH OF THE CATEGORIES FOR THE INDIVIDUAL qUESTIONS (CONSULTANT PHYSICIANS N = 5; NURSING STAFF N = 25; TOTAL N=30)


I am convinced of the benefits of thrombolytic treatment in acute myocardial infarction 97 (29) 3 (1)

I am convinced that there are additional benefits from giving thrombolytic treatment in the community at the earliest opportunity after symptom onset

94 (28) 3 (1) 3 (1)

Thrombolytic treatment is too expensive for use in general practice 97 (29) 3 (1)

Thrombolytic treatment is safe for use in general practice 67 (20) 10 (3) 23 (7)

Thrombolytic treatment is too difficult for use in general practice because it is to difficult to diagnose eligible patients 90 (27) 10 (3)

The decision to thrombolyse a patient is entirely the general practitioner’s 37 (11) 53 (16) 10 (3)

The general practitioner’s decision to thrombolyse patients should made in consultation with the hospital 53 (16) 27 (8) 20 (6)

Only certain thrombolytic treatments are suitable for use in the community 67 (20) 26 (8) 7 (2)

General practitioners should be allowed to administer thrombolytic treatments with sufficient training 97 (29) 3 (1)

EMT’s should be allowed to administer thrombolytic treatment with sufficient training 27 (8) 63 (19) 10 (3)

The hospital has a role to play in prehospital thrombolysis 86 (26) 7 (2) 7 (2)

COST, ADMINISTRATION AND SAFETYBoth groups agreed that it was safe to administer thrombolysis in the community. In terms of the cost of thrombolysis, the general practitioners and hospital staff both agreed that thrombolysis is not too expensive for use in general practice. Likewise, the general practitioners did not see thrombolysis as too difficult or too inconvenient to administer. The majority of the hospital staff had confidence in the general practitioners’ ability to assess and diagnose eligible patients for thrombolysis. This is in contrast to the findings from Scotland.6

EqUIPMENTThe possession of equipment such as an ECG and defibrillator are necessary for domiciliary thrombolysis. Practically all the general practitioners had access to an ECG and were willing to use it in the case of suspected AMI. Likewise, all but one had access to a defibrillator. Only two of the general practitioners reported being unable to interpret an ECG and two reported being unable to use a defibrillator.

TRAINING AND EDUCATION NEEDSPrevious work has estimated that the general practitioners in this area will on average see three AMI cases per year6. Given the rare occurrence of AMI, continued training and education in pre-hospital cardiac care must become an integral part of any pre-hospital thrombolysis policy. The majority of the general practitioners felt that training was sufficient prior to the initiation of the project. However, it is clear that the general practitioners feel that refresher courses in pre-hospital cardiac care and in interpretation of the electrocardiogram are needed. Regarding ECG telemetry, only half the general practitioners felt that training was sufficient. Findings from the DART study4,8 showed that none of the general practitioners made use of the facility to transmit the ECG to the hospital prior to the arrival of the patient. At the initiation of the project there was a shortage of hospital on-call staff making it difficult to obtain assistance from the hospital for diagnostic support. The general practitioners were confident in their diagnostic ability and felt that the decision to thrombolysis was entirely their own. Using the ECG telemetry especially when not competent in its use would only increase the delay to pre-hospital thrombolysis.

COnClUsiOnThese results show that the general practitioners are a willing, enthusiastic and confident group for the provision of pre-hospital thrombolysis. Furthermore, the hospital staff acknowledges the important role of general practitioners and are willing to provide support to those general practitioners who wish to provide this service. General practitioners should be supported in terms of equipment, training and education in pre-hospital thrombolysis and incentives should be offered to those who are willing to provide this service.

ACkNOwLEDGEMENTS

We wish to acknowledge and express our appreciation to the general practitioners, nursing staff and consultants for their time and enthusiasm to participate in this study.

REFERENCES1. Cardiovascular Health Strategy Group (1999). Building

Healthier Hearts. Department of Health and Children.

2. O’Neill J, Dowling J, Wright P et al. Patients presenting with acute myocardial infarction to a district general hospital: Baseline results and effect of audit. Irish Med J 2003; 96; 3.

3. GREAT Group. Feasibility, safety and efficacy of domiciliary thrombolysis by general practitioners: Grampian region early anistreplase trial. Br Med J 1992; 305: 548 – 53.

4. North West Immediate Care Programme (2003). Donegal Area Rapid Treatment Study (DARTS): Final Report. North Western Health Board.

5. Rawles J. Attitudes of general practitioners to prehospital thrombolysis. Br Med J 1994; 309: 379

6. Rawles J, Ritchie L. Thrombolysis in peripheral general practices in Scotland: another rule of halves. Health Bull 1999; 57:1: 10 – 16.

7. Doherty D, Dowling J, Wright P, Murphy AW, Bury G, Bannan L. The potential use of prehospital thrombolysis in a rural community. Resus 2004, 61: 303 – 307.

8. Doherty D, Dowling J, Wright P, Murphy AW, Bury G, Bannan L. Prehospital thrombolysis in a rural community: A case series and clinical follow-up. Submitted to Irish Journal of Medical Sciences.

Correspondence to: Dr Donna Tedstone Doherty, Health Research Board, Third Floor, Knockmaun House, 42 - 47 Lower Mount Street, Dublin 2.Email: [email protected]




ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnVenous thromboembolism (VTE) remains an important complication amongst surgical inpatients. Before the widespread introduction of heparin prophylaxis, as many as 25% of patients developed deep venous thrombosis (DVT).1,2 Due to the largely silent nature of DVT, the true incidence and prevalence rates is probably substantially higher.3 Several meta-analyses have revealed that Low Molecular Weight Heparins (LMWH) are as effective as unfractionated heparins for VTE prophylaxis in general surgical patients and have practically replaced unfractionated heparin for VTE prophylaxis as well as treatment for inpatients.4,5

The prevalence of VTE after day case and short stay surgeries is less clear. The literature base upon which putative prophylaxis regimens may be based is largely absent. The authors identified one large study conducted in Denmark studying VTE post day case herniorrhaphy.6

The study period was between 1982 and 1992 and documented rates of VTE were less than 1%. Intuitively, these patients should be at inherently lower risk as they tend to be a well population undergoing short duration surgery. Surprisingly, there

are no other correlative studies confirming incidence in day case herniorrhaphy or other common short stay procedures.

Thus as surgeons we practice in a time where most units have clear guidelines in place for their surgical inpatients, whereas for our day case or short stay patients no clear guidelines for VTE prophylaxis exist. In the absence of higher-level evidence, the authors’ sought to elucidate a national consensus with respect to VTE prophylaxis for four specific day case and short stay procedures.

MetHODsA standardised anonymous postal questionnaire was utilised. The operations selected for review were day case herniorraphy, minor anorectal surgery and varicose vein surgery. Day case or overnight laparoscopic cholecystectomy was also included. The questionnaire was posted to all Consultant Surgeons on the active register in Ireland in 2003. Replies within one month of posting were included in the sample. One postal reminder was sent to those who did not respond within four weeks.

Consultant surgeons were asked whether standardised protocols for prophylaxis against

Low molecular weight heparin prophylaxis in day case surgeryaBstraCtBackground The role of Low molecular weight heparins (LMWH) in day case/ short-stay

surgery is unknown. Aim To characterise the current national use of LMWH prophylaxis in specific day and

short stay surgeries.Methods A standardised anonymous postal questionnaire was sent to all consultant

general surgeons in Ireland. The operations selected were herniorraphy, anorectal, varicose vein and laparoscopic cholecystectomy.

Results Questionnaires were sent to 82 surgeons in 2003. There was a response rate of 68.3% (56). Fifty-four per cent of respondents said there was a protocol in place for administration of LMWH in day case surgery. Of these 41% were not confident that their protocols were being adhered. Fifty-nine per cent of all respondents said they stratified patients according to individual risk. Thirteen per cent reported occurrence of VTE post day case surgery.

Conclusion This study demonstrates a heterogeneous pattern of administration of LMWH. In the absence of published validated protocols, the authors suggest a consensus day case protocol.

J Shabbir, PF Ridgway, W Shields, D Evoy, JB O’Mahony, K MealyDept of Surgery, Wexford General Hospital, Wexford

VTE were in place and adhered to for each of the procedures. Usage and dosage, timing of LMWHs were investigated as well as attitudes toward risk stratification. Surgeons’ individual recollections of whether they had known cases of VT post day case surgery were assessed. The use of mechanical aids such as intermittent compression or thromboembolic event deterrent (TED) stockings was recorded.

resUlts Questionnaires were sent to 82 general surgeons. There was an overall response rate of 68.3% (56). Sixty one per cent (50) responded within first four weeks and further 7% (6) following reminders. Fifty-four per cent of respondents said there was a protocol in place for administration of LMWH in day case surgery. Of these 41% were not confident that their protocols were being adhered to leaving 59% with working protocols (Figure 1).

Only one respondent utilised TED stockings in prophylaxis for patients undergoing laparoscopic cholecystectomy. Individual procedures investigated in this study with presence or absence of protocols are shown in (Figure 2 and Table 1). Enoxaparin and Tinzaparin were the two commonest LMWH preparations in use. Dosages ranged from low dose (20 mg Enoxaparin or 3,500 IU Tinzaparin) to high dose (40mg Enoxaparin or 4,500 IU Tinzaparin). Timing where stated, ranged from within an hour prior to surgery to greater than two hours (Table 2). Interestingly, 59% of all respondents said they stratified patients according to individual risk, although this appeared to be on an individual basis as protocols were not structured to include stratification.

In the varicose vein group, 43% have protocols in place, although 16% are not confident they are in use. The majority select low dose LMWH within an hour prior to surgery. Thirty-two per cent of day case herniae have LMWH prophylaxis, usually within one hour of surgery. Once again, high dose LMWH is only used in a small per centage. Anorectal day case surgery is rarely given prophylaxis. Working protocols are seen in 13%. Low dose LMWH is used exclusively. Day case and short stay laparoscopic cholecystectomy had the highest per centage of protocols in place (46%). Once again the commonest dosage is low dose LMWH given within one hour prior to the procedure.

Thirteen per cent of surgeons reported documented occurrence of VT post day case surgery within a

Figure 1 —PRESENCE AND PERCEIVED ADHERENCE TO THROMBOPROPHY-LAXIS REGIMENS FOR DAY CASE SURGERIES

Figure 2 —PRESENCE OR ABSENCE OF PROTOCOL PER PROCEDURE (%).

“?Do” refers to those with protocols who are not confident of adherence

Table 1TIMING FOR PROPHYLAXIS (HOURS BEFORE SURGERY)

% <1 1 TO 2 >2 NOT STATED

Varicose Veins 60 20 8 12

Hernia 50 22 6 22

Anorectal 25 25 13 37

Lap Chole 44 20 12 24

Table 2LMwH DOSAGE REGIMEN FOR INDIVIDUAL PROCEDURES

VARICOSE VEINS

LAPAROSCOPICCHOLECYSTECTOMY HERNIA ANORECTAL

LD LD None None

LD LD None None

HD HD IPB IPB

LD=Low dose, HD= High dose, IPB=Individual patient basis

LOw MOLECULAR wEIGHT HEPARIN PROPHYLAXIS IN DAY CASE SURGERY J ShaBBir et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


one-month period post operatively. There was no statistically significant increased reports of known VTE in the group without a protocol (14% versus 11% in protocol group; p=0.76 Mann Whitney U test).

DisCUssiOnPrevalence of VTE is not known with respect to day case and short stay surgical patients. The last two decades have seen a consensus regarding prophylactic regimens reducing the risk of post operative VTE amongst longer stay patients.1, 2 This is further supported by investigation of patients dying as a consequence of VTE, showing that prophylaxis was omitted in a high proportion.4 Thus most authors extrapolate from surgical inpatient data which may or may not be relevant to the short stay setting. Increasing pressures on hospital inpatient beds due to long waiting times for surgery has led to more procedures being carried out as day case or short stay. Most institutions now perform varicose veins and hernia surgery as day cases and some hospitals perform laparoscopic cholecystectomies as day case although in the main, an overnight stay is mandated.

Against this backdrop, the current survey showed that a protocol for VTE prophylaxis for day case surgery was in place in a little over half of respondents’ hospitals. Furthermore compliance with the protocols was doubted in over a third. Considerable variation was noted within similar postal districts although due to the anonymous nature of the questionnaire further conclusions with respect to this point could not be made. Almost universally, where protocols were in place, LMWHs

were used as thromboprophylaxis. Selection of the lower dose was common with some timing differences noted between the individual procedures.

Surgeons’ anecdotal experience of over 1 in 10 surgeons reporting knowledge of VTE post day case/ short stay surgery is interesting. This is probably significant as the local referral pattern for VTE is medical and the attending surgeon is not always informed of the occurrence. Hence, this personal experience may be underestimated, and the true figure being higher than the 13%. There are no day case incidence rates published to give a reliable incidence of VTE per procedure apart from a study of over 2000 day case herniae conducted in Denmark in the 1980s.6 The authors found a non-significant rate of VTE and although assertions are made regarding specific thromboprohylaxis, mention is made of encouraging ambulation at discharge and advice thereafter. It is therefore uncertain whether this reflects a true rate without thromboprophylaxis or indeed whether these data can be extrapolated to laparoscopic herniorrhaphy.

Minimally invasive surgery, such as laparoscopic cholecystectomy, might be considered to reduce the rate of post-operative VTE. However, the raised intra abdominal pressure reduces the venous return, leading to venous pooling in the legs.7 This may be associated with increased risk of DVT and subsequent Pulmonary Embolus. A survey of members of ASGBI looking particularly into VTE prophylaxis for laparoscopic cholecystectomy among UK and Irish surgeons showed that VTE prophylaxis was used as a routine in 74% of patients undergoing

LOw MOLECULAR wEIGHT HEPARIN PROPHYLAXIS IN DAY CASE SURGERY

Table 3wEXFORD GUIDELINES FOR DAY CASE/ SHORT STAY SURGERY

VARICOSE VEINS LAPAROSCOPICCHOLECYSTECTOMY HERNIA ANORECTAL

Low risk LD LD None None

Moderate risk LD LD None None

High Risk HD HD IPB IPB

LD: low or standard dose LMWH (e.g. Tinzaparin 3,500IU) given 1 hour preoperatively (as per consensus timing)HD: high dose LMWH (e.g. Enoxaparin 40mg) given at least 2 hours preoperatively (as per manufactures’ safety guidelines)IPB: Individual patient basisRisk Stratification:Low: <30minute surgeryModerate: >30minute surgery High: >30minute surgery and additional risk factor(s) for VTE*(*Oral contraceptive pill, Obesity, Medical Co morbidity, Immobility (4/7+), Cancer, hx DVT, PE, Thrombophillia)

J ShaBBir et al

laparoscopic cholecystectomy and selectively in another 20% of patients.7 This is contrary to our results which show that under half of surgeons use routine prophylaxis for short stay laparoscopic cholecystectomy.

There is a well-known association between varicose veins and VTE and a number of studies have shown an increased risk of deep vein thrombosis of calf in patients with varicose veins.9 A survey of Welsh surgeons showed that 30% routinely use VTE prophylaxis for patients undergoing unilateral varicose vein surgery and 44% for bilateral varicose veins.9 In our survey 43% of surgeons use routine VTE prophylaxis for varicose veins surgery which is similar to the Welsh experience. As expected fewer surgeons use prophylaxis for hernia surgery and anorectal procedures because duration of surgery is short and VTE risk is perceived as minimal.

Risk stratification is now supported for all thromboprophlaxis protocols (THRIFT) guidelines.10 Although nearly two thirds reported they stratified patients according to risk, little evidence was seen in the protocols regarding formalising this tenet. Hence, it is assumed that this is carried out on a patient-by-patient basis which may have clinical as well as medicolegal implications for units with working protocols.

This survey highlights the uncertainties regarding VTE and day case/short stay surgeries that are present in today’s practice in Ireland. We acknowledge postal surveys have several limitations including the possibility of responder bias towards the use of prophylaxis. We believe our response rate of 68.3% is adequate to draw conclusion about currents trends.

Given that so many issues are unclear including actual incidence figures, the heterogeneity of departmental protocols and guidelines is perhaps unsurprising. It is nevertheless alarming and requires to be addressed, as VTE and PE are serious but preventable complications of surgery. In the absence of high level evidence in the existing literature and based on the findings of this survey, we recommend that a national audit or prospective study should be carried out to find out the true incidence of VTE following day case/short stay surgeries. The authors also suggest a consensus protocol based on the survey findings but recognise this represents a summation of current standard care rather than a prospectively validated regimen (Table 3).

COnClUsiOnThis national survey of LMWH prophylaxis in day case/short stay surgery demonstrates a heterogeneous pattern of administration of thromboprophylaxis by surgeons. VTE post day case occurs, and in the absence of published validated protocols for day case surgery, the authors suggest a consensus day case protocol.

REFERENCES1. Claggett GP, Reisch JS. Prevention of venous

thromboembolism in general surgical patients. Result of meta-analysis. Ann Surg 1988;208:227-40

2. Lowe GDO, Greer IA, Cooke TG, et al. Risk of and prophylaxis for venous thromboembolism in hospital patients. BMJ 1992;305:567-74

3. Gillies TE, Ruckley CV, Nixon SJ. Still missing the boat with fatal pulmonary embolism. Br J Surg 1996; 83:1394-95

4. Nurmohamed MT, Rosendaal FR, Buller HR, et al. Low molecular weight heparin versus standard heparin in general and orthopaedic surgery: a meta-analysis. Lancet 1992; 340:152-56

5. Hollingsworth J, Ghadban W, Kelly A, et al. Unfractionated heparin, time for a change? IMJ 1999; 92:302-6

6. Riber C, Alstrup N, Nymann T, et al. Postoperative thromboembolism after day-case herniorrhaphy. Br J Surg. 1996;83(3):420-21

7. Bradbury AW, Chan YC, Darzi A, et al. Thromboembolism prophylaxis during laparoscopic cholecystectomy. BJS 1997;84:962-64

8. Lowe GD, Osborne DH, McArdle BM. Prediction and selective prophylaxis of venous thrombolism in elective gastrointestinal surgery. Lancet 1982;i: 409-12

9. Williams EV, Williams RS, Hughes JL, et al. Prevention of venous thrombolism in Wales: results of a survey among general surgeons. Postgrad Med J 2002;78:78-88

10. Risk of and prophylaxis for venous thromboembolism in hospital patients. Thromboembolic Risk Factors (THRIFT) Consensus Group. BMJ 1992;305:567-74.

Correspondence to: Mr J Shabbir, SpR Colorectal Surgery, Gloucestershire Royal Hospital, Greatwestern Road, Gloucester GL1 3NNTel: 084 5422 2222/ Fax: 084 5422 5612Email: [email protected]


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnIreland has a very high incidence of premature coronary heart disease (CHD) mortality 1. Risk factor assessment and in particular lipid screening has become increasingly important. Since the primary care setting is where the main battle against CHD will be fought,2 risk assessment and prevention needs to be optimized at this point. However, in order to measure lipid levels, patients often have to travel to hospitals or their blood samples are sent to laboratories. This can result in excess patient travel, sample loss and a repeat visit to their GP. The latter can result in a delay in initiation of treatment and loss of continuity of care. A point of care lipid measurement would avoid these potential problems and afford better patient care.

Point of care lipid analysers have been available for some time in Ireland 3 but the accuracy of this method has yet to be established. We therefore conducted a study to compare the results of full lipid profiles using the point of care lipid analyser (Cholestech LDX) to that of the reference hospital laboratory.

SUBJECTS AND METHODS100 subjects were recruited. These subjects were either attending the Cardiac Risk Factor Clinic at

the Adelaide Meath incorporating the National Children’s Hospital because of coronary heart disease or hypertension, or were hospital staff members who volunteered to participate in the study. Seventy per cent of the subjects were female and 30% male. All subjects were Caucasian.

Fasting (80%) and non-fasting lipid profile samples (20%) were taken so that a broad range of lipid measurements could be compared. Following informed consent, subjects donated both a finger-prick sample and a venous whole blood sample. The finger prick sample and a portion of the venous sample were subjected to analysis in the point of care (POC) analyser. The rest of the venous sample was sent to the reference laboratory. All POC samples were analysed for a full lipid profile (total cholesterol, triglyceride, low-density lipoprotein cholesterol, high density lipoprotein cholesterol) and glucose level. Laboratory samples were analysed for lipid (n=100) and glucose levels (n =40).

Performance characteristics of the Cholestech LDX analyser – analytical limits, imprecision, interference factors are outlined in the Cholestech LDX System Procedure manual (www.cholestech.com).

Validation of a point of care lipid analyser using a hospital based reference laboratoryaBstraCtBackground Lipid measurements are essential in cardiovascular disease management in

primary care. However, utilizing hospital laboratories may result in excess patient travel, sample loss and repeat clinic visits. Point of care (POC) lipid measurement would overcome these difficulties but has not been validated in an Irish setting.

Aim To compare POC lipid profile measurements using a cholestech LDX analyser with a hospital reference laboratory (Lab).

Method One-hundred subjects (30 men, 70 women) participated. Finger prick and venous samples were analysed directly by Cholestech LDX and the Lab.

Results A broad range of lipid values were measured. Absolute differences between POC and Lab measurements were insignificant except for a small over-estimation by the POC method of triglyceride 0.25mmol/l (95% CI 0.17 to 0.24), and an underestimation of HDLc – 0.11mmol/l (95% CI – 0.143 to –0.078). There were significant correlations between POC and Lab. levels; total cholesterol r=0.92, triglyceride r=0.93, HDLc r=0.92 and LDLc r=0.86 (all p< 0.0001).

Conclusion These results validate the use of the Cholestech LDX® analyser for point of care lipid measurements in clinical practice, provided well trained operators are supported by a hospital laboratory delivering quality assurance support.

M Carey,1 C Markham,1 P Gaffney,2 G Boran,2b V Maher1

Depts of Cardiology1 and Chemical Pathology2, The Adelaide and Meath Hospital incorporating The National Children’s Hospital, Dublin

VALIDATION OF A POINT OF CARE LIPID ANALYSER USING A HOSPITAL BASED REFERENCE LABORATORY M carey et al

Two research nurses who were trained in the use of the Cholestech machine performed all the blood sampling and data collection. Finger-prick sampling involved ensuring the subject was seated and their hands warmed to ensure good circulation. This was achieved by subjects washing their hands with warm water or gently massaging their fingers from the base to the tip to help the circulation in their fingertips. The finger was cleansed with an alcohol swab and dried thoroughly with a gauze strip before pricking the side of the finger (to reduce pain) with a lancet. To determine if alcohol swabbing interfered with the lipid results, a number of subjects (n=30) did not have the finger site swabbed with alcohol.

The first drop of blood from the puncture site was wiped away to reduce the potential influence of interstitial fluid 4. The second drop of blood was then collected in a capillary tube. The capillary tube was filled with 35µl of blood for analysis within 10 seconds of blood sampling. This was to avoid blood coagulation interfering with the capillary action. Care was taken to avoid the collection of air bubbles. It was necessary to insert the blood from the capillary tube into the sample well of the cassette within 5 minutes to avoid blood clotting. The cassettes were placed in a cassette holder and analysed by a fully automated procedure in less than 10 minutes. The concentration of each lipid fraction was proportional to the intensity of the coloured product obtained and measured by reflectance photometry 5. Results were automatically printed (Figure 1). The storage and handling of the cassettes is important and clear guidelines accompany each analyser. Cassettes were stored in a fridge at 2 to 8ºc and only removed 10 minutes before use. Unused cassettes were never returned to the fridge for future use.

The hospital laboratory staff were blinded as to which samples were part of this validation study. Total cholesterol, HDL cholesterol, triglycerides and plasma glucose were analysed in the laboratory using a Hitachi Modular P® analyser and Roche® reagents. All methods have a between run imprecision of <3%. Results outside the analytical limits of the methods are automatically diluted by the analyser and the corrected result reported. Both analyser and laboratory LDL cholesterol was calculated by the Friedewald equation6 ([LDL-chol] = [Total chol] - [HDL-chol] - ([TG]*.458). Laboratory results were downloaded from the hospital computers.

This study had ethical approval from Trinity College ethical committee.

STATISTICAL ANALYSISAll statistical analysis was performed using JMP statistical software 5.1.2, www.jmp.com. Univariate correlations between like variables (measured by POC or Laboratory) were performed using the Spearman rank correlation method. Absolute differences between samples were compared by paired t-tests. P values less than 0.05 were deemed significant. Bland Altman difference plots were used to evaluate differences of values between methods. Passing Bablock Regression was used to compare the methodologies used for lipid measurements.

resUltsSeventy women and 30 men mean age 46 ± 13 years participated in the study. Their range of lipid values were as follows: total cholesterol (3.3 to 9.2 mmol/l), triglyceride (0.38 to 6.32 mmol/l ), high density lipoprotein cholesterol (HDLc) (0.84 to 2.97 mmol/l) and low density lipoprotein cholesterol (LDLc) (0.9 to 7.3 mmol/l).

Figure 2 demonstrates the strong correlations between lipid levels measured by POC and the hospital laboratory; total cholesterol: r = 0.92 (p <0.0001), triglyceride: r = 0.93 (p <0.0001) HDLc r = 0.92 (p < 0.0001) and LDLc r = 0.86 (p < 0.0001) . The LDLc values were determined indirectly using the Friedewald equation except for two samples where triglycerides exceeded 4 mmol/l. From the graphs, it is evident that the correlations were true throughout the range of measured values. However, there were two outliers evident in triglyceride measurements.

The outliers from the finger prick sample analysed were no longer evident when the venous samples from the same two patients were analysed in the Cholestech LDX (Figure 3). Both patients were female and on recollection admitted to using hand cream before the lipid testing had been carried out.

In absolute terms, the mean difference between the Cholestech LDX values of finger prick samples and the laboratory values were small.

The methodologies for lipid measurements were also compared using Passing Bablock Regression with the following results for finger prick vs laboratory : cholesterol: slope 1.02 (95% CI 0.95 to 1.11), intercept -0.145 (95% CI - 0.53 to 0.23), triglyceride: slope 0.93 (95% CI 0.88 to 0.98), intercept -0.086 (95% CI -0.15 to -0.02), HDLc: slope 1.07 (95% CI 1.0 to 1.16), intercept -0.001 (95% CI -0.123 to 0.105), LDLc slope 0.96 (95%

32 IRISHJOURNALOFMEDICALSCIENCE•VOLUME175•NUMBER4

ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R

IRISHJOURNALOFMEDICALSCIENCE•VOLUME175•NUMBER4 33

CI0.87to1.06),intercept–0.07(95%CI-0.21to0.29).Theseresultsareallhighlysignificant.Table1Ademonstratesthattotalcholesterolmeasurementsusingthesemethodsdifferedby0.037(CI-0.03–0.11mmol/l)whichweresmallandnotstatisticallysignificant.TriglyceridevalueswerehigherusingPOCthanlaboratoryby0.25mmol/l(CI0.17to0.24mmol/l)(p<0.001)andHDLcholesterolvalueswerelowerusingPOCthanthelaboratoryby-0.11mmol/l(CI-0.143to-0.078mmol/l)(p<0.001).Therefore,therewasasmallbutsignificantoverestimationoftriglyceridevaluesandasmallbutsignificantunderestimationofHDLcholesterollevels.AsevidentinTable1B,thedifferenceinHDLmeasurementwasinsignificantifavenoussamplewasdirectlyanalysedintheCholestechLDX.Differencesintriglyceridelevelswerealsolessmarkedbutstillsignificant.Thiswouldindicatethatthesmalldifferenceswereduetodifferencesinthelaboratorymethods.ThedifferencesbetweenthefingerprickandvenoussamplesanalysedintheCholestechLDXmostlikelyreflecteddifferencesduetotissuefluidlipoproteinspresentinthefingerpricksample.

Useornon-useofanalcoholswabbeforesamplingdidnotinfluencethedegreeofcorrelationbetweenthemethodsorabsolutedifferences.Thecorrelationsbetweenmethodsandtheabsolutedifferencesbetweenmethodswereunchangedwhethersubjectswerefastingornot.ThedifferencesinlipidvaluesusingPOCandlaboratorydidnotcorrelatewithsubjectslipidlevels.Thiswouldindicateaccuracyatalllipidlevels.Thisisevidentfromthedifferenceplotsforthelipidmeasurementsshowingthatthedifferenceswereequallyspreadacrosstheconcentrationrange(Figure4).

Thelearningcurvesofthetwodifferentoperatorsdidnotresultinsignificantlygreaterdifferencesbetweenthemethodologiesintheearlypartofthestudycomparedtothelipidmeasurementsofsubjectsrecruitedlaterinthestudy.

In40subjects,correspondinghospitallaboratoryglucosemeasurementsweremade.Therewasasignificantcorrelationbetweenfingerprickandhospitallaboratoryglucoselevelsr=0.54p<0.0005.Absolutedifferencesbetweenanalyserwholebloodglucoselevelsandlaboratoryplasmaglucoselevelsweresmall0.2(0.05–0.34)mmol/landnotstatisticallysignificant.

DISCUSSION ThisstudyconfirmsthatnearpatientlipidtestingusingtheCholestechLDXmachine,inahospitalclinicsetting,usingdedicatedandmotivatedoperatorswithcentrallaboratorysupport,yieldsresultsthatcorrelatehighlysignificantlytothosemeasuredindependentlyinastandardizedhospitallaboratory.Thiswasevidentfortotalcholesterol,triglyceride,HDLcholesterolandLDLcholesterollevels.TherewasastatisticallysignificantoverestimationoftriglyceridelevelsandanunderestimationofHDLcholesterollevelsbythePOCmethodbutthesedifferenceswerequitesmall.Inpractice,thesedifferencesareunlikelytobeclinicallysignificant.

ThecorrelationcoefficientforLDLcrvaluewaslowerthanthatfortotalcholesterol,triglycerideandHDLc.ThisispossiblyduetothefactthatLDLcwascalculatedusingtheFriedewaldequationandthevariationsintriglycerideandHDLvaluesbetweenmethodswouldhaveexaggeratedLDLdifferencesandhencereducedtheLDLccorrelationbetweenmethods.

Anumberofpossibleconfounderswereconsideredsuchassamplingtechnique,Haemoglobinlevels,useornon–useofalcoholswabs,operatorlearningcurveanddifferencesbetweenoperators.Noneofthesefactorsmadeanysignificantimpactontheresultsobtained.Theonlyrecognizedconsiderationwastheinfluenceofhandcreamonsampletriglyceridelevels.

Theimplicationsoftheseresultsarethatarelativelysimplepointofcareassayprovidesaccuratelipidprofileresultsforclinicalpractice.Utilizationofthistechniquecouldprovidegeneralpractitionersandpracticenurseswithusefulmeasurementswithwhichtheycanmakeclinicaljudgmentsandinstigatetreatmentatthetimeoffirstpatientcontact.Thiswillnaturallyfreeupdoctorandpatienttimeandensurebettercontinuityofcare.Itshouldbenoted,however,thatanyPOCserviceshouldonlybeimplementedinaccordancewithnationalandinternationalguidelines7-8.

Pointofcaretestingcanalsobeusedintheout-patientsettinginhospitalswheredecisionsneedtobemadeonpharmacologicalchangesatthetimeofthepatientvisitthusreducingvisitfrequency.

Whilethisstudyprimarilyexaminedlipidmeasurements,asub-populationof40patientsrevealedthattherewasasignificantcorrelationbetweenlaboratorymeasuredandpointofcare

Table 1 (a)

LIPIDS MEAN DIFFERENCE BETWEEN METHODS (POC - LAB.) 95% CI P VALUE

Cholesterol mmol/l 0.037 (-0.030to0.110) n.s.

Triglycerides mmol/l 0.200 (0.170to0.240) p<0.001

HDLc mmol/l -0.110 (0.143to-0.078) p<0.001

POC=pointofcarefingerpricksample;Lab.=Laboratorymeasured;n.s.=nonsignificant

Figure 2 —COMPARISON OF LABORATORY AND POINT OF CARE TESTING

Figure 1 —CHOLESTECH LDX ANALYZER

Table 1 (b) — Absolute differences between Lipid measurements

A DifferencebetweenfingerpricklipidmeasurementsusingLDXcholestechandLaboratorymeasurements

mmol/l 95% CI p valuesCholesterol 0.037 (-0.37to0.11) nsTriglyceride 0.25 (0.18to0.33) p<0.001HDLc -0.11 (-0.143to-0.078) p<0.001

B DifferencebetweenvenouslipidmeasurementsusingLDXCholestechandLaboratorymeasurements

mmol/l 95% CI p valuesCholesterol 0.024 (-0.05to0.95) nsTriglyceride 0.155 (0.08to0.23) p<0.001HDLc -0.015 (-0.065to-0.03) ns

VALIDATION OF A POINT OF CARE LIPID ANALYSER uSING A HOSPITAL BASED REFERENCE LABORATORY M Carey et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


Figure 3 —TRIGLYCERIDE MEASUREMENT

Figure 4 —DIFFERENCE PLOTS OF LABORATORY AND FINGER PRICk LIPID MEASUREMENTS

measured glucose levels. Absolute differences in glucose measurements were not statistically significant. However, this was not the primary aim of this study and a more detailed study of a larger sample of glucose measurements would be needed to confirm its accuracy.

This study did not directly test how accurate the point of care results would be in the primary care setting as comparative samples would have had to be transported to hospital laboratories. Prolonged transport time, altered sample temperature and may have added variables which could have confounded the results. For this reason it was thought that it would be most appropriate to perform the study directly in the hospital setting to avoid sample transportation difficulties but the results would be applicable in all settings. One important consideration in the primary care setting is operator training which may differ from practice to practice. However, with careful instruction, as was the case for the two nurses who performed this study, there was no evidence of any measurement inaccuracies or of a learning curve effect. All operators need to be aware of possible confounders as already outlined. There is a need for a validation of the Cholestech LDX or other point of care lipid analysers in individual clinical practices before this service could be offered.

There should be sufficient training of staff in the safe operation and maintenance of the equipment with standard operating procedures fully documented. Quality assurance measures9 both internal and external need to be established to ensure quality control in the clinical setting. Support structures with a local clinical laboratory are advocated for the long-term success of the service, particularly for the ongoing training and quality assurance. Use of point of care analysis in settings where medical input for result interpretation is not present or where appropriate medical guidelines10 are unavailable, should be cautioned.

The cost-effectiveness of the service needs to be examined further taking into account the current cost of each cassette (E12 for each full lipid and glucose cassette) versus the cost to the patient for travel (E1 to E40 private car versus taxi) parking (E5) and hospital laboratory costs (E5) per sample for phlebotomy, reagents and staffing. This would nearly break even for a person with their own transport (E11 hospital lab attendance versus E12 per cassette at GP surgery). However, the cost of repeat GP visits (E20 – E50,

GMS or Private) has not been considered. In addition, the inconvenience for patients, the potential for sample and result loss and the time delay in starting treatment need to be factored into the equation.

When used appropriately, point of care testing will improve patient management and fortify the services that are provided for patients at the primary care level.

ACkNOwLEDEMENTS

The authors would like to thank all the patients and hospital personnel who volunteered to participate in this study. While this study was initiated and controlled entirely by the authors, a small amount of funding was received from Cardiac Services to cover the staff nurse time for patient visits and sample management and the Cholestech cassettes were provided free of charge. We would also like to thank Elaine Hand, Biostatistician, Health Science centre for performing the Passing Bablock Regression analysis.

REFERENCES1. 50 years of Heart Disease in Ireland. Mortality,

morbidity and health services implications. Dublin: Irish Heart Foundation, 2001.

2. Heartwatch. The national programme in general practice for the secondary prevention of cardiovascular disease in Ireland. Annual Report 2004 -2005 ICGP

3. Point of Care Diagnostic testing – world markets report Guinness Centre, Taylor’s Lane, Dublin 8

4. CH Sloop, L Dory, PS Roheim. Interstitial fluid lipoproteins J Lipid Res, 1987; Vol 28, 225-237,

5. Cummings MH, Watts GF, Mazurkiewicz JC, Slavin BM. An evaluation in the diabetes clinic of the Cholestech L-D-X analyser. Practical Diabetes July/August 1994; 11(4):151-153.

6. Friedewald, WT, et al., Estimation of the Concentrations of Low-Density Lipoprotein Cholesterol in Plasma, Without Use of the Preparative Ultracentrifuge, Clin.Chem., 1972; 18/6:499.

7. UK Joint Working Group on Quality Assurance. Near to Patient or Point of Care Testing Guidelines. Association of Clinical Biochemists, Institute of Biomedical Sciences, Royal College of Pathologists January 1999

8. Policy Guidelines on Near Patient Testing. Academy of Medical Laboratory Science 1998

9. Additional essential criteria for quality systems of medical laboratories. Jansen R.T.P., Blaton V., Burnett D., Huisman W., Queralto J.M., Zerah S. Allman B. Clin. Chem Lab. Med. 1998; 36(4):249-252.

10. European guidelines on Cardiovascular disease prevention in clinical practice. Eur Heart J 2003; 24 , 1601 – 1610.

Correspondence: Vincent Maher, Dept of Cardiology, Adelaide Meath Hospital, Tallaght, Dublin 24 Ireland; Tel. +353 1 414 2112; Fax +353 1 414 3052; Email: [email protected]

VALIDATION OF A POINT OF CARE LIPID ANALYSER USING A HOSPITAL BASED REFERENCE LABORATORY M carey et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnType 2 diabetes confers a 2-4 fold increased risk of cardiovascular disease and over 60% likelihood of death from macrovascular complications.1, 2 Recent studies have demonstrated that interventions to reduce vascular risk, including the control of hypertension and hyperlipidaemia, and aspirin therapy, can reduce morbidity and mortality in people with type 2 diabetes.3-9 Vascular risk factor reduction is possible in the setting of an intensified multifactorial intervention trial,5 but published data highlight the difficulty in achieving risk reduction targets in routine clinical practice.10-13

In patients with type 2 diabetes, aggressive blood pressure control is more effective then intensive glycaemic control in reducing macrovascular events.4, 14 Hypercholesterolaemia doubles cardiovascular risk in diabetes,15 and the relative cardiovascular protection from treatment with statins is greater in patients with diabetes than in non-diabetic controls.8, 16, 17 The vascular risk is such that diabetes is regarded as a cardiovascular disease equivalent,18 and American Diabetic Association (ADA) guidelines recommend the use of statins as primary prevention for people with diabetes and

hypercholesterolaemia,19 a stance that has been justified by the results of the Heart Protection Study.16

Aspirin is of proven value for both primary and secondary prevention of macrovascular disease,3,20

and is recommended by the ADA for all patients with diabetes over the age of 40 years, who have no contraindication to treatment.19 Epidemiological data have demonstrated that hyperglycaemia is clearly a risk factor for the development of macrovascular disease, 1 though the evidence that tight glycaemic control confers protection against vascular events in people with type 2 diabetes was not provided by the results of the UKPDS.14

We wished to see how effectively the information from clinical trials and current clinical guidelines are implemented in routine diabetes care. The purpose of our study was to establish the prevalence of cardiovascular risk in our patient population, to assess the success of vascular risk factor control and to examine the efficacy of a strategy to improve risk factor management in patients with type 2 diabetes in a routine annual review diabetes out patient clinic.

Lipid lowering targets are easier to attain than those for treatment of hypertension in type 2 diabetesaBstraCtAim To assess the impact of therapeutic strategies to reduce cardiovascular risk in

patients with type 2 diabetes.Methods Five-hundred patients with type 2 diabetes were studied, using retrospective

case note analysis in 1997 (followed by a unit policy targeting vascular risk) and again in 2001.

Results The mean BP of the hypertensive patients was unchanged, 151/83 ± 23/12mmHg (1997) and 149/84 ± 19.1/9.8mmHg (P=0.2) (2001) (despite increase in patients receiving ≥3 antihypertensives (4.2% to 18.0%, P<0.01). The mean cholesterol improved from 5.34 ± 1.1 mmol/L to 4.72 ± 0.94 mmol/L (P<0.01). 2.9% compared with 44.6% (P<0.01) of hypercholesterolaemic patients, achieved target cholesterol. Antiplatelet therapy increased from 27.6% to 61.2% (P<0.01). Reduced mean HbA1c, 7.91 ± 1.61% to 7.12 ± 1.41% (P<0.01).

Conclusion Improved lipid profiles, aspirin uptake and glycaemic control, but no improvement in blood pressure targets were achieved. Additional strategies are required to achieve cardiovascular risk factor targets.

M Sherlock, D Mylotte, J Mac Mahon, KB Moore, CJ ThompsonDept of Diabetes and Endocrinology, Beaumont Hospital, Dublin

MetHODsCLINIC STRUCTUREThe Beaumont Hospital Diabetes Service provides care for 4000 people with diabetes under the supervision of one consultant diabetologist. There are 4000 outpatient clinic appointments available per annum for review of people with diabetes, so the vast majority of patients with Type 2 diabetes attend on an annual basis for complication screening and review of treatment strategies. At the time of the initial audit, there was no structured diabetes care in general practice, though interim care for glycaemic control was provided in the nurse-led diabetes day centre. At each annual review, full history and examination including blood pressure and ECG were recorded, early morning urine was collected for albumin creatinine ratio, and blood was taken for lipid profile, haemoglobin A1c (HbA1c), urea and electrolytes and liver function tests.

INITIAL AUDITA retrospective case note analysis was performed in 1997 of 500 consecutive patients with type 2 diabetes attending the diabetes review clinic. Data from the medical notes, including HbA1c, total cholesterol (TC), HDL cholesterol (HDL), LDL cholesterol (LDL), triglycerides, ECG, albumin excretion rate and blood pressure were recorded. Previous vascular events and intervention procedures, along with medications, were recorded on a standard form and entered into a database for processing.

Hypertension was defined as two consecutive blood pressure readings greater then 140/90mmHg, or previously documented and treated hypertension. Hypercholesterolaemia was defined as a TC > 5mmol/L or a documented diagnosis of hypercholesterolaemia. Patients not prescribed aspirin, but prescribed warfarin or clopidogrel were deemed to be on antiplatelet therapy.

Exclusion criteria included patients with type 1 diabetes, duration of diabetes less than one year, incomplete clinical information and failure to attend for the previous two clinic visits. Patients on insulin therapy were considered to have type 2 diabetes if they had been treated with diet and/or oral hypoglycaemic agents for more than two years prior to the commencement of insulin.

INTERVENTIONAfter analysis of the results of the audit in 1997, a unit policy for risk factor reduction was adopted, on

the basis of the ADA guidelines at that time and also the results of relevant clinical trials.21

1. Patients with any additional cardiovascular risk factors were treated with aspirin, in the absence of contraindications.

2. All patients with total cholesterol >5mmol/L were treated with a statin, with target total cholesterol of <5mmol/L.

3. Antihypertensive medication was increased in order to achieve a target of 140/90mmHg (this target has decreased in recent years to <130/80mmHg).19

4. Treatment of hyperglycaemia was intensified in order to achieve a target HbA1c of <7%.

5. All patients received advice regarding diet and exercise.

6. Staff underwent continuous structured education on risk factor management for patients with diabetes along with reinforcement of unit targets. Targets for aspirin, cholesterol, blood pressure and glycaemic control were given equal importance.

CLOSING THE AUDIT LOOPIn 2001 an audit of 500 consecutive patients with type 2 diabetes who were attending the diabetes review clinic was conducted in an identical fashion to that of 1997. analYsisBlood pressure readings were measured using a standard mercury sphygmomanometer. Readings were taken with patients seated and the lowest of three measurements was selected. HbA1c was measured by HPLC using a Hi Auto A1C (A MENARINI DIAGNOSTICS, BERKSHIRE, UK). This system, although not DCCT aligned, has been found to consistently perform 0.6% lower than DCCT aligned machines.

TC, HDL and triglycerides was measured using an Olympus AU640 / AU2700 (NAGASUMI, JAPAN). LDL was calculated using the Friedwald formula (LDL= cholesterol – (TG/2) – HDL-c). Urine albumin was measured using immunoturbidometric assay, (ROCHE DIAGNOSTICS, NEW JERSEY USA), with microalbuminuria defined as ACR >21mg/gCr on an early morning urine sample.

All information was placed on a database. The database was designed using Microsoft Access 2000. All statistical calculations were performed using Microsoft Excel 7, and results were compared using unpaired t tests.

LIPID LOwERING TARGETS ARE EASIER TO ATTAIN THAN THOSE FOR TREATMENT OF HYPERTENSION IN TYPE 2 DIABETES M SherlocK et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


resUltsPATIENT DEMOGRAPHICSMedian age of population was 68 years (range 31 to 91) in 1997 and 69 years (range 34-91) in 2001. All patients were Caucasian with 58% and 60% of patients of male sex in 1997 and 2001 respectively (p=0.55). Duration of diabetes was 8.0 + 6.12 and 7.7 + 6.36 years (Mean +SD) respectively (p=0.46).

PREVALENCE OF VASCULAR RISk FACTORS AND VASCULAR DISEASEThe prevalence of individual vascular risk factors and established macrovascular conditions are demonstrated in Table 1. The prevalence of established macrovascular disease was 32.2% in 1997 and 29% in 2001 (P=0.27).

HYPERTENSION DATAThe number of patients with hypertension was higher in 2001 (51.0%) than in 1997 (42.6%) (p = 0.01). The mean BP of the hypertensive patients was 151/83 + 23/12mmHg in 1997 and was unchanged at 149/84 + 19.1/9.8mmHg in 2001 (p= 0.2). The number of hypertensive patients achieving target BP did not vary significantly, at 41% and 45.6% (p =0.31) respectively. When the blood pressure results were divided into quartiles (Figure 1), there was no difference in the proportion of patients in each quartile in 1997 and 2001. This is despite an increase in the prescribing of antihypertensive medication with an increase in the number of patients on three or more antihypertensive agents from 4.2% to 18.0% (p<0.001) (Table 2).

Table 1PREVALENCE OF VASCULAR RISk FACTORS AND VASCULAR DISEASE AND NUMBER OF PATIENTS ON ANTIPLATELET THERAPY IN 1997 AND 2001. FIGURES EXPRESSED AS TOTAL NUMBER (PERCENTAGE). * = P < 0.01 COMPARED TO 1997

CARDIOVASCULAR RISK FACTOR OR MACROVASCULAR DISEASE

NO. OF PATIENTS (%)

(N=500)

NO. OF PATIENTS ON

ANTIPLATELET RX (%)

NO. OF PATIENTS (%)

(N=500)

NO. OF PATIENTS ON

ANTIPLATELET RX (%)

1997 1997 2001 2001

Hypertension 213 (42.6) 64/213 (30.0) 255* (51.0) 183/255* (71.8)

Albuminuria 200 (40.0) 78/200 (39%) 242* (48.4) 141/242* (58.3)

Hyperlipidaemia 304 (61.0) 81/304 (26.6) 311 (62.2) 206/311* (66.2)

Angina 56 (11.2) 35/56 (62.5) 62 (12.4) 47/62* (75.8)

MI 67 (13.4) 46/67 (68.7) 38* (7.6) 31/38 * (81.6)

PVD 29 (5.8) 22/29 (75.8) 35 (7.0) 27/35 (77.1)

TIA/ CVA 32 (6.4) 28/32 (87.5) 29 (5.8) 24/29 (82.7)

Primary prevention group 339 47/339 (13.9) 355 198/355* (55.8)

Secondary prevention group 161 91/161 (56.5) 145 108/145* (74.5)

Table 2ANTIHYPERTENSIVE PRESCRIBING IN 1997 AND 2001. FIGURES EXPRESSED AS TOTAL NUMBER (PERCENTAGE) OF THE SUBSET OF PATIENTS wITH HYPERTENSION. NS = NON SIGNIFICANT

ANTIHYPERTENSIVE TREATMENT

NO. OF HYPERTENSIVE PATIENTS ON TREATMENT 1997 (%)

NO. OF HYPERTENSIVE PATIENTS ON TREATMENT 2001 (%) P VALUE

ACE inhibitor 121 (56.8) 163 (63.9) 0.08

B Blocker 53 (24.9) 84 (32.9) 0.057

Ca Channel Blocker 88 (41.3) 133 (52.1) 0.02

Other classes 40 (18.8) 106 (41.5) <0.01

2 Agents 54 (25.4) 76 (29.8) NS

>2 agents 9 (4.2) 46 (18.0) <0.001

Glycaemic control and mean plasma cholesterol were similar in the patients with hypertension achieving target compared with those with uncontrolled hypertension, however patients failing to achieve target blood pressure were more likely to be male (p < 0.01) and to have albuminuria (P<0.01).

HYPERCHOLESTEROLAEMIA DATAThe prevalence of hypercholesterolaemia in 1997 and 2001 was 60.8% and 62% respectively (p=0.74). The percentage of hypercholesterolaemic patients treated with lipid lowering agents increased from 13% in 1997 to 78.7% in 2001 (p<0.01). The mean cholesterol of the total group decreased from 5.34 (+/-1.1) mmol/l in 1997 to 4.72 (+/-0.94) mmol/l in 2001 (p=<0.01). Target cholesterol was achieved in 2.9% of hypercholesterolaemic patients in 1997 and 44.6% in 2001 (p<0.01) (figure 2). In 2001, the breakdown of statin prescribing showed that 39.4% of patients were on the minimum dose, whereas only 4.5% were receiving maximum dose for the respective agents.

ANTIPLATELET THERAPYTotal number of patients prescribed antiplatelet agents rose significantly from 27.6% in 1997 to 61.2% in 2001 (p<0.01). In the secondary prevention group, prescribing increased from 56.5% of patients in 1997 to 74.5% in 2001 (p=<0.01) and in the primary prevention group, from 13.9% to 55.8% (p<0.01). The rate of antiplatelet prescribing in the individual risk factor groups is demonstrated on Table 1.

GLYCAEMIC CONTROL DATAThe mean HbA1c in the total group decreased from 7.91 + 1.61% to 7.12 + 1.41% (p<0.01). The number of patients achieving target HbA1c increased from 142 (28.4%) in 1997 to 243 (48.6%) in 2001 (p<0.01). There was an increase in the number of patients on insulin from 8% to 19.8% (p<0.01), but no other difference in treatment.

DisCUssiOnThe results of a closed loop audit of our vascular risk management strategy in patients with type 2 diabetes show improvements in glycaemic control, lipid management and aspirin prescribing, but no improvement in blood pressure control within the setting of a routine annual review clinic.

Baseline prescribing of cholesterol-lowering agents was low in 1997, with very few hypercholesterolaemic patients reaching recommended targets. As there was a clear need to increase the use of statins, we adopted

Figure 1—DISTRIBUTION OF BLOOD PRESSURE READINGS AMONGST PATIENTS wITH HYPERTENSION IN 1997 AND 2001. Results have been divided into quartiles. Patients were allocated to the highest quartile applicable on the basis of either the systolic or the diastolic reading.

Figure 2—DISTRIBUTION OF CHOLESTEROL READINGS AMONGST PATIENTS wITH HYPER-CHOLESTEROLAEMIA IN 1997 AND 2001. Patients were divided into quartiles on the basis of total cholesterol.

a target of total cholesterol < 5mmol/l, rather than LDL, for simplicity of message. The ADA now suggest a target LDL cholesterol of <2.6mmol/l as LDL correlates better with the risk of atherosclerosis when compared to total cholesterol and therefore is a better marker of cardiovascular risk than total cholesterol in patients with type 2 diabetes.19 After implementation of our intervention strategy, there was a clear increase in prescribing of statins and in the numbers of patients reaching defined cholesterol targets. Mean plasma cholesterol fell significantly, though less than half of our hypercholesterolaemia patients achieved target cholesterol. Further improvements in attaining target cholesterol could therefore be achieved with a more aggressive approach to prescribing. Data from Gaede et al indicate that intensive treatment of hypercholesterolaemia, as part of a multifactorial intensive interventional trial in type 2 diabetes, can reduced the mean total cholesterol to <4.0mmol/L.5

However in this trial patients were reviewed four times a year, a frequency of review, which is not currently possible in our unit, without a substantial increase in resources.

In 2001 the ADA position statement on antiplatelet therapy was updated,22 to recommend aspirin for



ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


all patients with diabetes over 40 years of age, who have no contraindication to treatment. This would apply to over 97% of patients. Our data demonstrated increased antiplatelet treatment in both primary and secondary prevention groups, which compares favourably to some published studies, and is similar to the results of the VA health care group study, which demonstrated 66% uptake of aspirin among patients with diabetes who responded to a questionnaire.11

The most disappointing aspect of the programme was the failure to impact on blood pressure. Poor compliance with antihypertensive medication was an unlikely explanation for this as the group with uncontrolled hypertension had similar glycaemic control and similar improvement in plasma cholesterol to both those with well-controlled hypertension and those with normal blood pressure. Blood pressure reduction to achieve recommended targets for type 2 diabetes is difficult, particularly in those with albuminuria.23 The UKPDS study also experienced difficulties in controlling blood pressure in patients with hypertension and type 2 diabetes, though the mean blood pressure in the intensively treated group was 144/ 82 mmHg with 29% of patients requiring three or more antihypertensive medications. Our data demonstrated a higher mean BP of 149/84mmHg with only 18% of patients on three or more agents. Extrapolation from this data suggests that in order to achieve a mean blood pressure in clinical practice of less than 140/90mmHg, approximately 50% of patients will need to be on three or more antihypertensive agents. Current ADA guidelines recommend a target blood pressure of 130/80mmHg for patients with diabetes.19 Gaede et al reported that less than 50% of patients achieved their target systolic BP of <130mmHg amongst intensively treated type 2 diabetic patients with microalbuminuria,5 whereas in the ALLHAT study approximately 60% of patients with diabetes achieved a target BP of 140/90mmHg.24 The most impressive data for blood pressure control in people with diabetes come from the HOT study,7 which specifically targeted and reported treatment of diastolic blood pressure. Our patient group, similar to cohorts in other studies of type 2 diabetes,5, 12 demonstrated a greater prevalence of systolic hypertension; the results of the HOT study may only specifically apply to a minority of hypertensive patients with type 2 diabetes.

The treatment of diabetic hypertension is clearly a challenge, with blood pressure targets difficult to

achieve, even in the setting of a clinical trial. It is unlikely that annual review of patients will be able to achieve blood pressure targets in patients with type 2 diabetes and more frequent review with aggressive prescription of antihypertensive agents seems to be required for high risk patients with diabetes and hypertension.

Our data, in common with those of other studies demonstrate the difficulty in translating the evidence from clinical trials into routine clinical practice, even within clinic structures designed to implement specific strategies.10, 12, 13 It has been suggested that patient education and diabetes nurse specialist input improve patient outcomes,25,

26 and we have now introduced a nurse-led clinic to specifically target patients with uncontrolled cardiovascular risk factors. The failure to implement risk factor management recommendations may reflect unrealistic targets.27 The results achieved in well-resourced clinical trials are difficult to emulate in clinical practice, where staffing levels and resources are stretched. Even in the setting of clinical trials, not all patients achieve target,5, 24 in clinical practice polypharmacy is inevitable and even with intensive patient education and frequent review, it is unlikely that greater than 70-80% success in attaining targets is possible.27

In conclusion, in the setting of a programme for vascular risk reduction in routine clinical practice, we have demonstrated improved glycaemic control, lipid profiles and aspirin uptake, but no improvement in blood pressure. Our results still fall short of targets and highlight the difficulties in implementation of best practice risk factor management in patients with type 2 diabetes.

ACkNOwLEDGEMENTS

This research was support by a research grant from Bristol Myer Squibb (Ireland).

REFERENCES1. Kannel WB, McGee DL. Diabetes and cardiovascular

risk factors: the Framingham study. Circulation 1979;59(1):8-13.

2. Nathan DM, Meigs J, Singer DE. The epidemiology of cardiovascular disease in type 2 diabetes mellitus: how sweet it is …or is it? Lancet 1997;350 Suppl 1:SI4-9.

3. Collaborative overview of randomised trials of antiplatelet therapy--I: Prevention of death, myocardial infarction, and stroke by prolonged antiplatelet therapy in various categories of patients. Antiplatelet Trialists’ Collaboration. BMJ 1994;308(6921):81-106.

4. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. UK Prospective Diabetes Study Group. BMJ 1998;317(7160):703-13.

5. Gaede P, Vedel P, Larsen N, et al. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med 2003;348(5):383-93.

6. Goldberg RB, Mellies MJ, Sacks FM, et al. Cardiovascular events and their reduction with pravastatin in diabetic and glucose-intolerant myocardial infarction survivors with average cholesterol levels: subgroup analyses in the cholesterol and recurrent events (CARE) trial. The Care Investigators. Circulation 1998;98(23):2513-9.

7. Hansson L, Zanchetti A, Carruthers SG, et al. Effects of intensive blood-pressure lowering and low-dose aspirin in patients with hypertension: principal results of the Hypertension Optimal Treatment (HOT) randomised trial. HOT Study Group. Lancet 1998;351(9118):1755-62.

8. Pyorala K, Pedersen TR, Kjekshus J, et al. Cholesterol lowering with simvastatin improves prognosis of diabetic patients with coronary heart disease. A subgroup analysis of the Scandinavian Simvastatin Survival Study (4S). Diabetes Care 1997;20(4):614-20.

9. Tuomilehto J, Rastenyte D, Birkenhager WH, et al. Effects of calcium-channel blockade in older patients with diabetes and systolic hypertension. Systolic Hypertension in Europe Trial Investigators. N Engl J Med 1999;340(9):677-84.

10. Agha A, Dillon D, Corbett M, et al. Target blood pressure for patients with type 2 diabetes is difficult to achieve in the setting of a busy diabetes clinic. Ir J Med Sci 2003;172(4):168-70.

11. Krein SL, Vijan S, Pogach LM, et al. Aspirin use and counseling about aspirin among patients with diabetes. Diabetes Care 2002;25(6):965-70.

12. McFarlane SI, Jacober SJ, Winer N, et al. Control of cardiovascular risk factors in patients with diabetes and hypertension at urban academic medical centers. Diabetes Care 2002;25(4):718-23.

13. Rolka DB, Fagot-Campagna A, Narayan KM. Aspirin use among adults with diabetes: estimates from the Third National Health and Nutrition Examination Survey. Diabetes Care 2001;24(2):197-201.

14. Intensive blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). UK Prospective Diabetes Study (UKPDS) Group. Lancet 1998;352(9131):837-53.

15. Stamler J, Vaccaro O, Neaton JD, et al. Diabetes, other risk factors, and 12-yr cardiovascular mortality for men screened in the Multiple Risk Factor Intervention Trial. Diabetes Care 1993;16(2):434-44.

16. MRC/BHF Heart Protection Study of cholesterol lowering with simvastatin in 20,536 high-risk individuals: a randomised placebo-controlled trial. Lancet 2002;360(9326):7-22.

17. Shepherd J, Cobbe SM, Ford I, et al. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. West of Scotland Coronary Prevention Study Group. N Engl J Med 1995;333(20):1301-7.

18. Haffner SM, Lehto S, Ronnemaa T, et al. Mortality from coronary heart disease in subjects with type 2 diabetes and in nondiabetic subjects with and without prior myocardial infarction. N Engl J Med 1998;339(4):229-34.

19. Clinical Practice Recommendations 2005. Diabetes Care 2005;28 Suppl 1:S1-79.

20. Colwell JA. Aspirin therapy in diabetes. Diabetes Care 1997;20(11):1767-71.

21. American Diabetes Association: clinical practice recommendations 1997. Diabetes Care 1997;20 Suppl 1:S1-70.

22. American Diabetes Association Clinical Practice Recommendations 2001. Diabetes Care 2001;24 Suppl 1:S1-133.

23. Bakris GL. Maximizing Cardiorenal Benefit in the Management of Hypertension: Achieve Blood Pressure Goals. J Clin Hypertens (Greenwich) 1999;1(2):141-147.

24. Major outcomes in high-risk hypertensive patients randomized to angiotensin-converting enzyme inhibitor or calcium channel blocker vs diuretic: The Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT). Jama 2002;288(23):2981-97.

25. New JP, Mason JM, Freemantle N, et al. Specialist nurse-led intervention to treat and control hypertension and hyperlipidemia in diabetes (SPLINT): a randomized controlled trial. Diabetes Care 2003;26(8):2250-5.

26. Renders CM, Valk GD, Griffin S, et al. Interventions to improve the management of diabetes mellitus in primary care, outpatient and community settings. Cochrane Database Syst Rev 2001(1):CD001481.

27. Winocour PH. Effective diabetes care: a need for realistic targets. Bmj 2002;324(7353):1577-80.

Correspondence to: CJ Thompson, Dept of Diabetes, Beaumont Hospital, Dublin.Tel: 00-353-18093000/Fax: 00-353-18376501 E-Mail: [email protected]



ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnDiabetes is an important chronic condition which is set to increase in prevalence in the coming decades. The complications of type 2 diabetes can be significantly reduced by appropriate treatment.1-3 For people with type 2 diabetes to enjoy these benefits, care needs to be organised in a systematic and integrated way. This has been shown to be effective in general practice settings4 and sustainable over 10 years.5 A trial in North Dublin6 showed that structured care in general practice can produce significant improvements in diabetes care delivery in the Irish healthcare environment.

Ten general practices in the south inner city of Dublin agreed to provide structured diabetes care in their practices in 2001. Care was shared with one local hospital diabetes service for patients attending that service. This study was carried out to determine whether delivery of care and intermediate outcomes of care for patients in these practices have altered after the move to structured care for diabetes.

Structuring diabetes care in general practices: Many improvements, remaining challengesaBstraCtBackground For people with type 2 diabetes to enjoy improved longevity and quality of

life, care needs to be organised in a systematic way.Aim To test if processes and intermediate outcomes for patients with type 2 diabetes

changed with the move to structured care in general practice shared with secondary care.

Methods An audit of process and intermediate outcomes for patients with type 2 diabetes before and after the change to structured care in 10 Dublin general practices shared with secondary care four years on.

Results Structured diabetes care in general practice has led to more dedicated clinics improved processes of care and increased access to multidisciplinary expertise. Improvement in blood pressure control, the use of aspirin and the use of lipid lowering agents indicate a significant decrease in absolute risk of vascular events for this population.

Conclusions Structured care in general practice improves intermediate outcomes for people with type 2 diabetes. Further improvements need to be made to reach international targets.

S Jennings,1 DL Whitford,2 D Carey,1 SM Smith3

Dept of Public Health Medicine1, Health Service Executive, Dr Steevens Hospital, Dublin; Dept of General Practice2, Royal College of Surgeons in Ireland, Dublin; Dept. of Public Health and Primary Care3, Trinity College Dublin, Trinity Health Sciences Institute, Tallaght Hospital, Dublin

STRUCTURING DIABETES CARE IN GENERAL PRACTICES: MANY IMPROVEMENTS, REMAINING CHALLENGES

1. Registration of patients2. General Practitioners (GP) and practice nurse

training3. Diabetes care guidelines adapted from

national (ICGP) and international (Diabetes UK) guidelines

4. Regular review of patients in general practice (consisting of quarterly visits to GP with annual assessment in hospital)

5. Annual ophthalmological review6. Support staff (dietician, specialist nurse and

chiropodist) were employed and available in the community to participating practices

Figure 1—SYSTEM OF CARE IN PARTICIPATING PRACTICES FOR PATIENTS wITH DIABETES

MetHODsA baseline audit of diabetes care was carried out in 2000 in ten general practices and repeated in 2004 after the introduction of structured diabetes care for type 2 diabetes. Two practices declined to participate in the re-audit. System of care for patients with diabetes is outlined in Figure 1. All patients, over 16 years of age, attending the participating practices, were included in the baseline (n= 293) and repeat audits (n = 324) with the exception of pregnant

S JenningS et al

women. Data on patients with type 2 diabetes are presented here. Practice diabetes registers were compiled in autumn 2000 and were used for the baseline audit (n = 234). In summer 2004 the registers were reviewed prior to carrying out the re-audit (n = 271). Patients with type 2 diabetes comprise 80% and 84% of the diabetes population in the participating practices in 2000 and 2004 respectively.

As the GP practices had patients from a wide catchment area, attending a number of hospitals, all type 2 patients were analysed in the re-audit though only 167 (61.6%) type 2 patients had care formally shared with the one participating hospital.

The audit instrument was finalised after consulting other Irish and international audits and recommendations7;8 and employed variables in the following categories: demography, evidenced based measures of process and intermediate outcome as well as activity data in the 12 months prior to the audits. A nurse researcher collected the data from practice medical records. Patient confidentiality was maintained with only anonymous patient data held on computer. Ethical approval was obtained from the Irish College of General Practitioners.

Data were analysed using both Excel and SAS packages. Chi-square, Wilcoxon rank sum and z tests were used for independent data analysis. A cohort of patients, common to both audits, was identified and McNemar’s and paired t-tests were used for these paired data. One sample z tests were used for age and gender comparisons of the cohort and main study group. All tests were two-tailed and the significance level was set at 5%.

resUltsCharacteristics of the patients in the baseline and in the re-audit are shown in Table 1. Mean age, gender and duration of diabetes remained the same in both audits. More patients were covered by general medical services (GMS) in 2004.

PROCESS MEASURESRecording of the majority of processes of care increased between 2000 and 2004 (Table 2). However, the recording of examination of foot pulses and fundi within the previous 12 months decreased. In the latter case there was an increase in specialist ophthalmic screening. Notably, when all eye examinations are combined there was no change (p=0.95) in the overall proportion of patients recorded as receiving

eye examination from 131 in 2000 (56.0%) to 151 in 2004 (55.7%). The proportion of patients prescribed therapeutic agents of proven benefit improved between 2000 and 2004 as did referrals to the dietician, chiropodist and ophthalmologist.

wORkLOADThe GP consultation rate of patients with type 2 diabetes prior to July 2000 was 9.23 consultations per patient per annum. This declined to 6.02 consultations per patient per annum in the 12 months prior to October 2004 (p< 0.0001). Admissions to hospital for all causes rose from 0.21 admissions per patient per annum to 0.40 per patient per annum for the same periods (p=0.0001).

INTERMEDIATE OUTCOME MEASURESThere was no change in the proportion of recorded smokers (Table 3) and no change in diabetes control as measured by HbA1c. There were significant improvements in blood pressure control between 2000 and 2004.

Paired data for processes and intermediate outcomesPaired data on patients present in both audits (n=58) are shown in Table 4. The cohort of patients common to both audits was similar in age and gender to the main study group in 2000. Similar changes in recording of process measures, referrals and prescribing were seen for this group as for the total population with the exception of the recording of BP, smoking and dietician visit which remained unchanged. However there was a non-significant decrease in the proportion of patients recorded as receiving any eye examination. There were significant improvements in blood pressure and no change in HbA1c over the four-year period in these patients.

CHANGES AT PRACTICE LEVELThe eight practices grew in size during the period of observation with the total patient population estimated to have grown by 68%. The GMS population increased by 45% from 7,490 to 10,890 patients. The number of doctors in the practices increased from 17 to 24 and the number of nurse hours from 171 to 233. The prevalence of type 2 diabetes (and both type 1 and 2 together) among the GMS population showed a small but non-significant increase from 1.76% (2.08%) in 2000 to 1.93% (2.23%) in 2004.

Of the eight practices six now run dedicated diabetes clinics with the two others following the agreed proforma of care in ordinary appointments. Also


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


in most practices (n=6) there is both medical and nursing input to the diabetes review quarterly. The degree of computerisation increased in the practices between 2000 and 2004 with increasing use for the consultation, problem lists and disease registers.

DisCUssiOn This study has shown that a planned approach to structured diabetes care for patients with type 2

diabetes in general practice has led to more dedicated clinics, improved processes of care and increased access to multi-disciplinary expertise. The latter is evidenced by increased number of professionals involved and increased referral. In particular, there have been significant improvements in blood pressure control, the use of aspirin and the use of lipid lowering agents. At the same time, there has been no deterioration in diabetes control.

Table 1CHARACTERISTICS OF PATIENTS wITH TYPE 2 DIABETES IN 2000 AND 2004 AUDITS

BASELINE 2000(N=234)

RE-AUDIT 2004(N=271) P VALUE

Number of Practices 10 8Mean age – years (SD) 67.75 (13.98) 66.84 (12.91) NSNumber males (%) 120 (51.3%) 145 (53.5%) NSNumber GMS Status (%) 156 (67.8%) 210 (77.8%) 0.01Mean duration of diabetes – years (SD) 6.12 (6.18) 6.52 (6.30) NS

SD = standard deviation, NS = not significant.Missing data excluded from analysis as follows: (a) Age 3 cases for 2000; (b) GMS status 4 and 1 cases for 2000 and 2004 respectively, c) Duration of diabetes 59 and 73 cases for 2000 and 2004 respectively.

Table 2PROCESS MEASURES RECORDED IN 2000 AND 2004 FOR PATIENTS wITH TYPE 2 DIABETES

BASELINE 2000 (N=234)

RE-AUDIT 2004 (N=271) P VALUE

Smoking status 132 (56.4%) 179 (66.1%) 0.03Blood pressure 204 (87.2%) 252 (93.0%) 0.03Weight 66 (28.2%) 187 (69.0%) <0.0001Body mass index 13 ( 5.6%) 80 (29.5%) <0.0001Examination of foot pulses 146 (62.4%) 41 (15.1%) <0.0001HbA1c 116 (49.6%) 193 (71.2%) <0.0001Microalbuminuria 36 (15.4%) 92 (34.0%) <0.0001Serum creatinine 74 (31.6%) 152 (56.1%) <0.0001Total cholesterol 63 (26.9%) 149 (55.0%) <0.0001Fasting triglycerides 39 (16.7%) 130 (48.0%) <0.0001HDL 30 (12.8%) 124 (45.8%) <0.0001LDL 28 (12.0%) 118 (43.5%) <0.0001Taking aspirin 73 (31.2%) 145 (53.5%) <0.0001On anti-hypertensives 125 (53.4%) 184 (67.9%) 0.001On lipid lowering agents 49 (20.9%) 131 (48.3%) <0.0001Dietician seen in last 12 months 49 (20.9%) 91 (33.6%) 0.002Chiropody referral in last 12 months 64 (27.4%) 155 (57.2%) <0.0001Examination of fundi 60 (25.6%) 14 ( 5.2%) <0.0001Referral for fundal screening 81 (34.6%) 144 (53.1%) <0.0001

Table 3COMPARISON OF INTERMEDIATE OUTCOME MEASURES IN PATIENTS wITH TYPE 2 DIABETES

BASELINE 2000(N=234)

RE-AUDIT 2004(N=271) P VALUE

Smokers (%) 42 (18.0%) 54 (19.9%) NS

HbA1c % < 7% % > 8% Mean (SD)

(n=114)32.5% 34.2%

7.70 (1.41)

(n=193)39.9%28.5%

7.40 (1.23)

NSNSNS

Systolic blood pressure % < 140 mmHg Mean (SD)

(n=181)40.9%

142.17 (22.01)

(n=252)62.7%

133.03 (19.01)<0.0001<0.0001

Diastolic blood pressure % < 90 mmHg Mean (SD)

(n=181)73.4%

81.91 (10.15)

(n=252)89.7%

75.83 (10.82)<0.0001<0.0001

NS=not significant, SD = Standard deviation

Table 4PAIRED COMPARISON OF PROCESS MEASURES AND INTERMEDIATE OUTCOMES IN THE SURVIVING COHORT OF PATIENTS wITH TYPE 2 DIABETES (2000 – 2004)

PROCESS MEASURES RECORDED N (%)

BASELINE 2000 (N=58)

RE-AUDIT 2004(N=58) P VALUE*

Smoking status 37 (63.8%) 36 (62.1%) NS

Blood pressure 55 (94.8%) 55 (94.8%) NS

Body mass index 2 (3.4%) 16 (27.6%) 0.001

Cholesterol 14 (24.1%) 32 (55.2%) 0.001

Serum creatinine 17 (29.3%) 35 (60.3%) 0.002

% on aspirin 10 (17.2%) 24 (41.4%) 0.0001

% on anti-hypertensives 30 (51.7%) 41 (70.7%) 0.003

% on lipid lowering agents 10 (17.2%) 26 (44.8%) 0.0001

Dietician seen in last 12 months 14 (24.1%) 14 (24.1%) NS

Chiropody referral in last 12 months 15 (25.9%) 32 (55.2%) 0.001

Eye referral or fundi check 35 (60.3%) 33 (56.9%) NS

Outcomes

Smokers (%) (n=25) 48.0% 36.0% NS

HbA1c (n=24) % < 7% (N) Mean (SD)

37.5% (9)7.45 (1.13)

29.2% (7)7.60 (1.10)

NSNS

Systolic blood pressure (n=47) % < 140 mmHg (N) Mean (SD)

38.3% (18)142.40 (21.52)

61.7% (29)131.96 (15.44)

0.020.002

Diastolic blood pressure (n=47) % < 90 mmHg (N) Mean (SD)

66.0% (31)83.38 (8.75)

89.4% (42)75.77 (10.08)

0.003<0.0001

NS=not significant, SD=Standard deviation.* McNemar’s tests were used for comparisons of proportions and paired t-tests for comparisons of means.

STRUCTURING DIABETES CARE IN GENERAL PRACTICES: MANY IMPROVEMENTS, REMAINING CHALLENGES S JenningS et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


The increase in the proportion of patients with GMS cover is most likely due to the introduction of GMS cover for all patients aged over 70 years in April 2001. The mean age of the two groups in 2000 and 2004 did not differ in spite of the passage of four years. This highlights one difficulty in monitoring the care of chronic illness. In every day practice there is an influx of newly diagnosed patients with earlier disease and the exit of patients who die or move to more dependant care. This leads to a problem in interpreting audit data over time. Consequently, a strength of this study was the ability to identify a cohort of patients who participated in both audits. Changes in this cohort were on paired data from the same patients and therefore demonstrate true change.

The design of this study was that of a repeat audit of care and intermediate outcomes. We are therefore not able to attribute the changes demonstrated in this study to the introduction of structured care. However, the evidence that general practice can provide structured diabetes care and achieve intermediate outcomes similar to this study is now overwhelming.4;9 This is an inner city group of practices and so results may not be generalised to or reflect the whole country. However, this study, in tandem with international literature and the North Dublin randomised control trial, is a further demonstration of the need for structured community based diabetes care in Ireland.

The improvements in processes of care demonstrated in this audit after the introduction of structured diabetes care are similar to those demonstrated elsewhere.10;11 However, the level of recording of processes of care in these patients is still below those advocated by the Irish College of General Practitioners and those achieved in these other studies. Both North Tyneside5;10 and Cardiff11 demonstrated the documentation of all relevant process measures in more than 70% of patients. On the other hand the data on controls in an Irish randomised control trial12 of diabetes shared care reveal similar levels of documentation to this study. This would indicate that even after the introduction of structured diabetes care, general practice in Ireland has still to improve the management of this chronic disease. Interestingly, prevalence has remained the same for the subgroup of GMS patients indicating a lack of improvement in disease detection compared to other studies.11;13 Contributory factors to consider in understanding these comparative deficiencies are a) the lack of planning and resourcing of diabetes care in Irish

general practice, b) the deskilling of Irish general practitioners in diabetes care over many years and c) the absence of a systematic screening programme for fundus examination. These need urgent attention in order to further improve care and outcomes for patients with diabetes. Indeed as a result of the audit one initiative undertaken was an outreach ophthalmological screening programme for this group of patients in the south inner city.

The most striking feature of this audit is that of improvement in blood pressure. In both the main group, and more particularly in the paired group, the improvement in systolic and diastolic blood pressure over the four years is a major achievement with significant implications for reduction in cardiovascular risk for this group of at risk patients as demonstrated by UKPDS.2 Over the ten-year period the cohort saw a reduction in mean systolic blood pressure of 10mmHg. This equates to a risk reduction of 15% for deaths related to diabetes, 11% for myocardial infarction, and 13% for microvascular complications.14 However, if the cohort is compared with the UKPDS group, absolute levels of blood pressure control are similar. This would equate to a risk reduction of 32% for deaths related to diabetes, 44% in strokes, and 37% in microvascular end points.2 Furthermore, the increase in prescribing of lipid lowering agents and aspirin alongside the reduction of blood pressure indicates a significant decrease in absolute risk of vascular events for this population with type 2 diabetes. There was no change demonstrated in levels of diabetes control. This would initially appear disappointing but UKPDS demonstrated an inevitable decline in diabetes control.1 The maintenance of diabetes control in this study is therefore an achievement and would contribute to further risk reduction for this group of patients.

The decrease in consultation rates with general practitioners as demonstrated in this study has not been shown elsewhere and merits further investigation. Potential reasons for this finding are more efficient care at each visit, better patient satisfaction with consultation or measurement error in regard to recording GP consultations at practice level. Routine care by nurses only was not a feature of this programme. The increase in hospital admissions is at odds with a previous study that showed a reduction in hospital referrals in patients attending general practice structured diabetes care.15 We did not have a breakdown of the reasons for admission

and this would also need further research as the aim of structured care is to improve outcomes and reduce the costs of diabetes that are largely hospital based.

This study is in line with previous studies showing that the introduction of structured diabetes care in general practice and supported by secondary care is associated with improvements in both processes and intermediate outcomes of care. However, there is a need for further improvements in order to bring the standards up to international norms. Consequently, we point to the urgent need for planning and resourcing of initiatives of this nature as well as a programme of retinal screening as part of high quality community based diabetes care in Ireland. The aim must be to provide care to the whole population with diabetes in order to achieve improved long term outcomes for these patients.

ACkNOwLEDGEMENTS

We would like to acknowledge the hard work and cooperation of the ten general practices who have had the commitment to initiate this programme of care, the two nurses (Bronagh Peelo and Ruth Bowie) who dedicated time and energy to accurate data collection and to the staff of the SIC Partnership Office who contributed in many ways to the change management programme.

REFERENCES 1. UK Prospective Diabetes Study Group. Intensive

blood-glucose control with sulphonylureas or insulin compared with conventional treatment and risk of complications in patients with type 2 diabetes (UKPDS 33). Lancet 1998;352:837-53.

2. UK Prospective Diabetes Study Group. Tight blood pressure control and risk of macrovascular and microvascular complications in type 2 diabetes: UKPDS 38. BMJ 1998;317:703-13.

3. Gaede P, Vedel P, Larsen N, et al. Multifactorial intervention and cardiovascular disease in patients with type 2 diabetes. N Engl J Med. 2003;348:383-93.

4. Griffin S. Diabetes care in general practice: meta-analysis of randomised controlled trials. BMJ 1998;317:390-6.

5. Whitford DL, Roberts SH, Griffin S. Sustainability and effectiveness of comprehensive diabetes care to a district population. Diabet Med 2004;21:1221-8.

6. Smith S, Bury G, O’Leary M, et al. The North Dublin randomized controlled trial of structured diabetes shared care. Fam Pract 21(1):39-45, 2004.

7. British Diabetic Association. Recommendations for the management of diabetes in primary care. 1997. London, BDA.

8. Brennan, C., De La Harpe, D., and Harkins, V. Audit Report of the Midland Health Board: Diabetes Shared Care Project 1998-1999. 2001. Midland Health Board.

9. Olivarius NdF, Beck-Nielsen H, Andreason AH, et al. Randomised controlled trial of structured personal care of type 2 diabetes mellitus. BMJ 2001;323:970-5.

10. Whitford DL, Southern AJ, Braid E, et al. Comprehensive diabetes care in North Tyneside. Diabet Med 1995;12:691-5.

11. Butler C, Smithers M, Stott N et al. Audit-enhanced, district-wide primary care for people with diabetes mellitus. Eur J Gen Pract 1997;3:23-7.

12. Smith S, Bury G, O’Leary M, et al. The North Dublin Diabetes Shared Care (DiSC) Project: a profile of current diabetes care in Ireland. IMJ 2001;94:240-3.

13. Whitford DL,.Roberts SH. Changes in prevalence and site of care of diabetes in a health district 1991-2001. Diabet Med 2004;21:640-3.

14. Adler AI, Stratton IM, Neil HAW, et al. Association of systolic blood pressure with macrovascular and microvascular complications of type 2 diabetes (UKPDS 36): prospective observational study. BMJ 2000;321:412-9.

15. Farmer A, Coulter A. Organization of care for diabetic patients in general practice: influence on hospital admissions. Br J Gen Pract 1990;40:56-8.

Correspondence to: Dr Siobhan Jennings;Email: [email protected]

STRUCTURING DIABETES CARE IN GENERAL PRACTICES: MANY IMPROVEMENTS, REMAINING CHALLENGES S JenningS et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnCarcinoma of the oesophagus and gastroesophageal junction represent aggressive diseases with a poor prognosis even in patients undergoing curative resection.1 The risk of treatment-related, in particular surgical morbidity and mortality is high, and accurate staging guiding appropriate treatment plans has considerable value and clinical importance.2 Staging of oesophageal cancer incorporates the UICC or AJCC staging which is based on TNM.3,4 The objective of accurate staging is to precisely select patients in whom a curative approach based principally on an aggressive loco-regional therapy, including surgery, is appropriate, accordingly, patients with M1 disease must be excluded. A further consideration, although controversial, is that patients with locally advanced disease, such as T4, bulky T3 and possibly node-positive disease are better considered for multmodality approaches rather than surgery alone. The traditional armamentarium for staging oesophageal cancer is cross-sectional imaging with computerized tomography (CT) or magnetic resonance imaging (MRI), both standard for identifying T4 or M1 disease.5 Endoscopic ultrasonography (EUS) has increased the accuracy of T staging, and this may be of value in the staging, but

it offers no greater advantage to good quality cross-sectional imaging in discriminating patients suitable for a palliative rather than a curative regimen.6

Positron emission tomography (PET) imaging provides qualitatively different information from CT and MRI, as it relies on the metabolic function of tumours instead of tumour size alone. PET utilizes the glucose analogue FDG (2-[18Fluorine]-Fluro-2-deoxy-D-glucose), providing physiological information that enables the diagnosis of malignancy to be based on altered tissue metabolism. In the process of transformation from normal cells, malignant cells develop significant alterations in metabolism. The major changes occurring include increases in DNA synthesis, amino-acid use and glycolysis.7,8 Tumours studied to date, where PET has shown a high diagnostic accuracy, include carcinoma of the lung, breast, colon, head and neck, brain as well as lymphoma, melanoma and sarcoma.9-15

Oesophageal cancer has a high glucose metabolism and thus has an avid FDG consumption, and there is an emerging literature on the use of PET imaging in the staging of tumours of the oesophagus and oesophago-gastric junction.16-21 This has been

FDG-PET scanning in the management of cancer of the oesophagus and oesophagogastric junction: Early experience with 100 consecutive casesaBstraCtBackground The aim was to evaluate the impact of FDG-PET scan on tumour staging and

management decisions in oesophageal cancer. Methods One-hundred consecutive patients referred for consideration of surgery

underwent a whole body FDG-PET scan in addition to CT imaging. Results Based on CT scan, a curative approach could be considered in 62 patients. The

PET scan altered regional nodal (N) staging in 16 patients overall, but did not alter management decisions. Metastatic status (M) was increased in 14 patients, with altered management in 10/62 (16%). Nine were downstaged, with management changed in 3/38 (8%). Seventeen patients underwent 19 additional tests to clarify findings on PET, in 15 patients (88%) the tests revealed no pathology.

Conclusion FDG –PET alters M stage in 23% of patients and may impact on surgical decision-making. The spurious investigations and cost of the high false-positive rate of further tests is of concern.

V Malik, M Keogan1, C Gilham2, G Duffy 3, N Ravi, JV ReynoldsDept of Clinical Surgery, Dept of Radiology1, Dept of Radiotherapy2, St James’s Hospital and Trinity College Dublin & Dept of Nuclear Medicine3, Blackrock Clinic, Dublin

FDG-PET SCANNING IN THE MANAGEMENT OF CANCER OF THE OESOPHAGUS AND OESOPHAGO-GASTRIC JUNCTION: EARLY EXPERIENCE wITH 100 CONSECUTIVE CASES

associated with decreased use of invasive procedures including EUS-directed nodal biopsies, thoracoscopy, and laparoscopy.22-24 This study aimed to assess the clinical impact of the introduction of staging FDG-PET on the management of oesophageal cancer at a single institution, in particular the number of patients who had surgery either deferred or facilitated based on PET findings, as well as impact of PET in generating further investigations and the outcome of those additional studies.

Materials anD MetHODsSTUDY DESIGNA prospective study was performed from November 2003 to December 2004 on 100 consecutive patients with biopsy proven carcinoma of the oesophagus or the gastro-oesophageal junction.

COMPUTED TOMOGRAPHYEach patient underwent a spiral CT examination, 91 at this institution and nine at the referring centre. Spiral CT scanning at this institution was carried out on a Siemens Somatom scanner, following administration of oral and intravenous contrast. In all patients imaging included neck, thorax and abdomen to the iliac crest, using a slice thickness of 8mm. Intravenous contrast medium (100ml) with concentration of 300ml iodine/l (Omnipaque, Amersham) was injected by power injector with a 30-second delay before scanning the neck/thorax and 65-second delay before scanning the liver.

CT images provided by the referral hospital (slice thickness 3mm-10mm) were re-evaluated by an experienced on-site radiologist (MK). These were read independent of the PET findings. The CT analysis included extent of primary tumour, tumour invasion of adjacent structures, lymph node enlargement and distant metastases. Perioesophageal invasion was considered present when direct invasion into the surrounding tissue was noted or when there was absence of fat cleavage planes between the tumour and adjacent organs. Lymph nodes were considered positive when the short axis was greater than 1 cm in diameter. Lymph nodes less than 1 cm in diameter were considered to be indeterminate for metastases (Nx). Lesions in the liver were considered suspicious for metastases (Mx) if not characteristic of a cyst or haemangioma.

Positron Emission Tomography ScanningAll patients underwent a whole body FDG-PET

scan using a PET Advance full ring PET scanner (General Electric). Patients fasted for a minimum of four hours. Blood glucose was measured and thus additional blood glucose manipulation was not required. Patients were given 5mg Valium orally 20 to 35 minutes prior to the FDG injection and were well hydrated orally in the two-hour period prior to FDG injection. Between 237 and 490 MBq of FDG was injected intravenously and patients remained lying in a quiet dark room for 60 minutes prior to scanning. Patients voided before going on to the scanner. The scan extended from the mid thigh to the base of the skull; six to eight bed positions were obtained, five minutes per emission scan and three minutes per transmission scan per bed position. Images were reconstructed using an interactive reconstructed algorithm with segmented attenuation correction. Scan images were viewed in the transaxial, coronal and sagittal plains, and cine projections on a dedicated work station and reported by a consultant physician in nuclear medicine with experience in PET interpretation.

FDG-PET images were compared with the most recent CT images. Maximum standardized uptake values (SUV) were obtained and a maximum SUV of 3.5 was used to assist the visual interpretation in differentiating inflammatory tissue and normal tissue from malignant tissues (malignant SUV > 3.5). Metastatic lesions were divided into local, regional and distant metastases. Further CT imaging, MRI, Ultrasound (U/S), ENT consultations and invasive procedures such as colonscopy were performed where PET identified equivocal metastatic lesions and patients were considered fit for surgery.

STAGINGStaging was based on TNM classification, depending on the localization of the primary tumour. 4 Suspected nodes were classified as locoregional (N1), and non-local lymph nodes were considered as M1a. According to this system, the definition of N changes depending on the location of the primary tumour (Table 1).

A curative approach to treatment involved either surgery alone or chemotherapy and radiation therapy prior to surgery. A curative resection denotes wide clearance of the primary tumour and a radical lymphadenectomy, including peri-oesophageal nodes as well as N2 nodes on the left gastric, common hepatic and splenic arteries, along with sub-carinal and mediastinal lymph nodes.

V MaliK et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


resUltsPATIENTS (TABLE 2)Seventy-eight male and 22 female patients with oesophageal carcinoma were enrolled prospectively. Their age ranged from 37 years to 86 years. Mean age at time of diagnosis was 61.5 years. Preoperative diagnosis was confirmed by pathological examination of endoscopic biopsy specimens, as adenocarcinoma in 56 patients and squamous cell carcinoma in 40 patients.

Based on the CT imaging, 62 (61%) were considered to have resectable disease and suitable for a curative approach involving either resection alone or preceded by chemotherapy and radiation therapy.

NODAL STAGEA total of 16 patients (16%) of the 100 enrolled had their N stage altered as a result of the FDG-PET scan. Five patients (5%) had upstaging of nodal status, four from N0 to N1, and one from Nx to N1. In two patients, surgery was not performed because of evidence of both nodal and systemic (bony) metastases. Two patients with nodal upstaging underwent curative resections, and one had multimodality therapy. In no case was management altered based on FDG-PET N upstaging. Eleven patients had decreased nodal staging, eight from Nx and three from N1, in no case did this influence the decision to proceed with or defer surgery.

M STAGINGA total of 23 patients of the 62 enrolled with potentially resectable disease had their M stage altered as a result of FDG-PET. Fourteen patients had upstaging of M status by FGD-PET (Table 3), and nine patients had their M status downstaged. Of the 14 upstaged patients, six did not undergo surgery because of systemic metastases, four because of distant nodal disease, two underwent surgery with radical lymphadenectomy, and one patient with small cell cancer continued with chemotherapy, and one patient was deemed unfit for any treatment. Thus M upstaging impacted on management in 10 of these cases, or 10/62 (16%) considered treatable with curative intent.

Nine patients had their M stage downstaged by FDG-PET, 5 from M1b to M0, and four from M1a to M0. Three underwent surgery, in four other patients surgery was not performed as they were unfit for the procedure, one patient developed intercurrent brain metastases, and one patient refused surgery. Thus

M downstaging may have impacted in 9% of cases overall, or 9/38 (24%) patients initially CT-staged as palliative. However, in fact the management was altered in just three cases, 3% overall or 8% deemed to be in the palliative cohort.

Although not strictly upstaging or downstaging, six further patients with initial M0 staging became Mx and a further 16 patients initially categorized as Mx became M0 following FDG-PET. In the former group further tests including focused liver ultrasound and bone scans were required, but management was not altered in any case, all of whom went on to have surgery.

ADDITIONAL INVESTIGATIONSOf the 100 patients enrolled in our study, 17 patients had FDG-PET scans which identified possible metastatic lesions not seen on CT scans. These lesions were indentified in the liver, the bone, the nasopharynx, the colon, the tongue, the thyroid, the pharynx and larynx, the tongue and in the leg. While the nature of many abnormalities on PET were not problematic (i.e. focal uptake in the liver was taken to represent metastatic disease, and widespread bony uptake was accepted as M1b), other areas were more equivocal, particularly focal colonic uptake. As a result 19 additional procedures and scans were performed (Table 4). Six patients were identified by FGD-PET to have possible colonic lesions, as a result they all underwent colonscopy, revealing benign changes (four polyps, two normal) in six patients. Six underwent further imaging investigations for suspected isolated bony metastases, in just one confirmed by MRI. In one patient a second primary in the nasopharynx was identified. Thus the additional investigations were falsely positive in 15 of 17 (88%) patients.

DisCUssiOnOesophageal cancer is increasing in incidence, it carries a poor outlook, it is often advanced at diagnosis, and the treatment modalities utilized in the curative approach to this disease have considerable attendant risks of morbidity and mortality.1 Only patients with localized regional disease can be treated with curative intent, either surgery alone, surgery as a component of a multimodality regimen, or combination chemotherapy and radiation therapy. In this era the rate of unnecessary surgical exploration should be continually decreasing, and some have advocated reporting of one-year survival rates as an index of the appropriateness of staging and management decisions.25,26

Table 1OESOPHAGEAL CARCINOMA: THE TNM STAGING SYSTEM

PRIMARY TUMOUR (T)TXT0TisT1T2T3T4

Primary tumour cannot be assessed No evidence of primary tumour Carcinoma in situ Tumour invades lamina propria or submucosa Tumour invades muscularis propria Tumour invades adventia Tumour invades adjacent structures

REGIONAL LYMPH NODES (N)NXN0N1

Regional lymph nodes cannot be assessed No regional lymph node metastasesRegional lymph node metastasis

DISTANT METASTASIS (M) MXM0M1

Distant metastasis cannot be assessed No distant metastasis Distant metastasis

TUMOURS OF THE LOWER THORACIC OESOPHAGUSM1aM1b

Metastasis in celiac lymph nodes Other distant metastasis

TUMOURS OF THE MIDTHORACIC OESOPHAGUSM1aM1b

Not applicable Non-regional lymph nodes and/or other distant metastasis

TUMOURS OF THE UPPER THORACIC OESOPHAGUS M1aM1b

Metastasis in cervical nodes Other distant metastasis

Table 2DEMOGRAPHICS

PATIENTS NO.

Demographics Male Female

7823

Mean age at diagnosis 61.5

Site Gastro-oesophageal junction Thoracic oesophagus Upper oesophagus Middle oesophagus Lower oesophagus

34

41646

Histology Adenocarcinoma Squamous cell carcinoma Small cell Intramucosal

564022

Since its introduction in 1974, computerised tomography has been the imaging modality of choice for the preoperative evaluation of gastrointestinal malignancies. Detection of malignancy is based only on structural change however, thus there are inherent limitations in specificity, particularly false positive findings in enlarged inflammatory lymph nodes, and sensitivity, in particular false negative findings in non-enlarged invaded lymph nodes.27 Increased metabolism associated with malignant transformation was first described by Warburg and forms the basis for imaging of cancer with FDG-PET.28,29 The ability to generate images on the basis of tissue metabolism allows PET to complement CT imaging which provides anatomical focus. In addition, the generation of whole body images in three dimensions, offers the option of replacing multiple imaging investigations, such as brain CT / MRI or bone scans with a single study.30

Several studies to date have reported a high sensitivity of FDG-PET for staging cancer of the oesophagus and of the gastroesophageal junction


V MaliK et al


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


Table 3MANAGEMENT IN PATIENTS wITH INCREASED M STAGE

PATIENT CHANGE IN M STAGING MANAGEMENT CHANGE DID PET INFLUENCE SURGERY?

1 M1a--------------M1b Yes, (distant mets in liver) Yes

2 M0---------------M1b Yes (Liver mets) Yes

3 M0---------------M1b No, (chemotherapy for small cell histology) No

4 M0---------------M1a No, Pt underwent surgery No

5 M0---------------M1a Yes, (distant nodal spread) Yes

6 Mx---------------M1a Yes, (distant nodal spread) Yes

7 M0---------------M1a No, (Pt not fit for surgery) No

8 Mx---------------M1b Yes (liver mets) Yes

9 Mx---------------M1b Yes, (distant mets in liver) Yes

10 Mx---------------M1a No, Pt underwent surgery No

11 Mx---------------M1a Yes, (distant nodal spread) Yes

12 M1a--------------M1a Yes, (distant nodal spread) Yes

13 Mx---------------M1b Yes, (distant mets in liver) Yes

14 M0---------------M1b Yes, (distant mets in bone) Yes

Table 4FDG-PET: ADDITIONAL INVESTIGATIONS, AND IMPACT ON MANAGEMENT

PATIENT FDG-PET LESION ADDITIONAL INVESTIGATION IMPACT OF FDG-PET ON SURGICAL DECISION

1 Bony metastases CT pelvis- normal No, Other (pt unfit for surgery)

2 Lesion in nasopharynx ENT review No, Other (DTX over surgery)

3 Lesion in posterior triangle of neck

CT-indeterminateU/S & Bx-normal

No, Other ( Small cell CA, chemotherapy

4 Lesion in colon Colonscopy-normal No, Pt underwent surgery

5 Bone-marrow & colonic uptake MRI & colonscopy - normal No, Pt underwent surgery

6 Lesion in nasopharynx ENT review normal No, Other (pt unfit for surgery)



9 Bony metastases MRI-normal No, Pt underwent surgery

10 Lesion in tongue ENT review No, Pt underwent surgery

11 Thyroid uptake U/S thyroid-normal No. Pt underwent surgery

12 Lesion in colon Colonscopy-normal No, Other (pt unfit for surgery)

13 Lesion in leg Clinical assessment No, Other (pt unfit for surgery)

14 Bony metastases Bone scan-normal No, Pt underwent surgery

15 Bony and Adrenal metastases Bone scan-normal Yes

16 Bony uptake and colonic Colonscopy-normal Bone scan- normal No, Pt underwent surgery

17 Bony metastases MRI-abnormal Yes, (distant metastases)

but few large studies of the impact of this modality in day-to-day practice have been undertaken.21,30,31,32 Previous studies have reported a high sensitivity (72% - 90%) and specificity for nodal disease (82% - 90%).27, 30 In this study, sensitivity and specificity were not addressed, but the practical impact of a change of N staging by PET scanning in 16 patients was negligible, being directly responsible for withholding surgery in just one case. This in part relates to the modern staging nomenclature which defines extra-regional nodes such as cervical, celiac and para-aortic as M1a disease. It is the Unit policy to advocated a radical curative approach involving lymphadenectomy even where N1 disease exists. For nodal assessment, the authors recognize that a limitation of the study with respect to nodal involvement is the lack of EUS availability for all patients, and EUS-directed nodal biopsies, and this modality may have reduced the marked impact of FDG-PET on CT-determined management decisions in this cohort. The use of preoperative chemotherapy alone or in combination with radiation therapy is controversial but it may be that this will become increasingly standard in patients with gross T3 or T4 disease, and patients with N1 disease, and perhaps the combination of PET with CT/EUS may better select patients in the node-positive subset.33,34

For M staging PET has been shown to be superior to CT both in sensitivity (69% vs. 46%) and specificity (93.4% vs 73.8%).17,31 In this study, 23 patients had their M stage altered, with 14 upstaged. Ten patients had surgery avoided by the PET finding, representing 16% of the group considered suitable for a curative approach. The M stage was downstaged in nine patients, four of whom went on to curative surgery.

The limitations in specificity of PET relate to the uptake of glucose by inflammatory, hyperplastic and lymphoid tissue. In this study 24 patients had possible metastatic lesions on PET not seen on CT scan, with uptake identified in the liver, bone, nasopharynx, colon, tongue, thyroid, pharynx and larynx, tongue and in the lower limb. Additional investigations in 16 patients included colonoscopies, MRI, CT pelvis, ultrasound of neck and bone scans. Metastatic disease was confirmed in just one of the 16 (6%) and in another tumour, a second primary in the nasopharynx altered the management decision. Focal activity in the colon was a common cause of concern leading to colonoscopies for clarification in six cases, all negative for significant pathology. Six

patients underwent further investigations of FDG-PET – suspected bone metastases, and in five of six cases (86%) this was negative. While it is possible that the introduction of CT-PET fusion imaging may help to resolve these issues it is likely that false positive findings will remain an unavoidable limitation of PET scanning until tumour-specific isotopes are developed.

This study set out to study the practical implications of PET in a tertiary centre. Several conclusions can be drawn. The positive message is that in 62 patients considered for curative treatment based on CT-imaging, M upstaging through PET impacted on management in almost 16% of cases. Downstaging, although impacting in this study in just three cases overall, had the potential to influence management in nine of 38 (24%) patients CT-staged with metastatic disease. The principal negative of PDG-PET evident in this study is the frequent need for further investigations with the attendant anxiety from predominantly false positive uptake. Advances in PET scanner technology and the introduction of fused PET/CT scanners may help to reduce costs and increase availability of PET scanning. Notwithstanding the cost-benefit debate, outside the focus of this study, the improved M staging where PET is performed in combination with CT imaging may support the routine use of FDG-PET in patients with apparent localized disease.

REFERENCES1. Enzinger PC, Mayer RJ. Esophageal Cancer. N. Engl J Med,

2003;349: 562-572

2. Hulscher JB, Tijssen, Obertop H, van Lanschot JJ. Transthoracic versus transhiatial resection for carcinoma of the esophagus: A meta-analysis. Ann Thorac Surg 2001; 72: 306-13

3. Beahrs OH, Henson DE, Hunter RVP, et al: American Joint Committee on Cancer: Manual for staging of cancer (ed 3). Philadelphia, PA, JB Lippincott, 1998,99 63-67.

4. Sobin LH, Wittekind C: TNM Classification of Malignant Tumours, 6th ed. New York: John Wiley & Sons, 2003.

5. Kumbasar B. Carcinoma of esophagus: radiologic diagnosis and staging. Eur J Radiol 2002;42:170-180.

6. Kelly S, Harris KM, Berry E, et al A systematic review of the staging performance of endoscopic ultrasound in gastro-oesophageal carcinoma. Gut 2001; 49; 534-9.

7. Bar-Shalom R, Valdiva AY, Blaufox MD. PET imaging in oncology. Semin Nucl Med 2000;30:150-85.

8. Lightdale CJ. Positron emission tomography: Another useful test for staging esophageal cancer. J Clin Oncol 2000;18:3199-01.


V MaliK et al


ORI

GIN

AL

PAPE

R

9. Lewis P, griffin S, Marsden P, et al Whole body F-fluorodeoxyglucose positron emission tomography in the preoperative evaluation of lung cancer. Lancet 1994; 344: 1265-1266.

10. Adler LP, Crowe JP, al-Kaisi NK, et al Evaluation of breast masses and axillary lymph nodes with (F-18) 2-deoxy-2-fluoro-d-glucose PET. Radiol 1993;187:743-750.

11. Rege S, Maass A, Chaiken L, et al Use of positron emission tomography with fluorodeoxyglucose in patients with extracranial head and neck tumors. Cancer 1994;74:3047-3058.

12. Newman JS, Francis IR, Kaminski MS, et al Imaging of lymphoma with PET with 2-(F-18)-fluoro-2-deoxy-D-glucose: correlation with CT. Radiol 1994;190:111-116.

13. Di Chiro G. Positron emission tomography using (F-18)fluorodeoxyglucose in brain tumors. A powerful diagnostic and prognostic tool. Invest Radiol 1987;22:360-371.

14. Steinert HC, Huch RA, Buck A, et al Malignant melanoma: staging with whole body positron emission tomography and 2-(F-18)-fluoro-2-deoxy-D-glucose. Radiol 1995;195: 705-709.

15. Nieweg OE, Pruim J, van Ginkel RJ, et al Fluorine-18-fluorodeoxyglucose PET in soft tissue sarcoma. J Nucl Med 1996;37:257-261

16. Henderik L, van Westreenen MD, Heeren PAM, van Dullemen HM, et al Positron emission tomography with F-18-fluorodeoxyglucose in a combined staging strategy of esophageal cancer prevents unnecessary surgical explorations. J Gastrointest. Surg 2005;9:54-61.

17. Wallace MB, Nietert PJ, Earle C, et al An analysis of multiple staging management strategies for carcinoma of the esophagus: computed tomography, endoscopic ultrasound, positron emission tomography and thoracoscopy/laparoscopy. Ann Thorac Surg 2002;74:1026-32.

18. Kato H, Hiroyuki K, Nakajima M, et al Comparison between positron emission tomography and computed tomography in the use of the assessment of esophageal carcinoma. Cancer 2002;94:921-928.

19. Kim K, Park SJ, Kim BT, Lee KS, Shim YM. Evaluation of lymph node metastases in squamous cell cancer of the esophagus with positron emission tomograpgy. Ann Thorac Surg. 2001;71:290-294.

20. Wren SM, Stijns P, Srinivas S. Positron emission tomography in the initial staging of esophageal cancer. Arch Surg. 2002; 137: 1001-1007.

21. Luketich JD, Friedman DM, Weigel TL, et al Evaluation of distant metastases in esphageal cancer: 100 consecutive positron emission tomography scans. Ann Thorc Surg 1999;68:1133-7.

22. Eloubeidi MA, Wallace MB, Reed CE, et al The utility of EUS and EUS-guided fine needle aspiration in detecting celiac lymph node metastasis in patients with esophageal cancer: A single center experience. Gastrointest. Endosc. 2001;54:714-9.

23. Parmar KS, Zwischenberger JB, Reeves AL, Waxman I. Clinical impact of endoscopic ultrasound-guided fine needle aspiration of celiac axis lymph nodes (M1a disease) in esophageal cancer. Ann Thorac Surg 2002;73:916-20.

24. Meno KV, Dehn TC. Multiport staging laporoscopy in esophageal and cardiac carcinoma. Dis Esophagus 2003;16:295-300.

25. Jamieson GG, Mathew G, Ludeman R, Wayman J, Myers JC, Devitt PG. Postoperative mortality following esophagectomy and problems in reporting its rate. Br J Surg 2004; 91: 943-97

26. Ellis FHJr., Heatley GJ, Krasna MJ, Williamson WA, Balogh K. Esophagogastrectomy for carcinoma of the esophagus and cardia: A comparison of findings and results after standard resection in three consecutive eight-year intervals with improved staging criteria. J Thorac Cardiovasc Surg 1997;113:836-46.

27. Flamen P, Lerut A, Van Cutesm E, et al: Utility of positron emission tomography for the staging of patients with potentially operable esophageal carcinoma. J Clin Oncol 2000;18:3202-10.

28. Warburg O. On the origin of cancer cells. Science 1956;123:309-14.

29. Wahl RL, Hutchins GD, Buchsbaum DJ, et al (F-18)-2-deoxy-2-fluoro-d-glucose uptake into human tumor xenografts. Feasibility studies for cancer imaging with positron emission tomography. Cancer 1991;67:1544-50.

30. Block MI, Patterson GA, Sundaresan S, et al Improvement in staging of oesophageal CA with the addition of positron emission tomography. Ann Thorac Surg 1997;64:770-7.

31. Fukahagan T, Okazumi S, Koide Y, Zsoho K, Zwnazeki K. Evaluation of esophageal cancers using fluorine-18-fluorodeoxyglucose PET. J Nucl Med 1998;39:1002-7.

32. Flanagan FL, Dehdashti F, Seigel BA, et al Staging of esophageal cancer with (F-18) fluorodeoxyglucose postron emission tomography. AJR 1997;168:417-24.

33. MRC Trial. Surgical resection with or without preoperative chemotherapy in oesophageal cancer: A randomised controlled trial. Lancet 2002;359:1727-1733

34. Fiochha F, DiBona D, Schepis F, Licata A, Shahied L, et al Preoperative chemoradiotherapy for oesophageal cancer: a systematic review and meta-analysis. Gut 2004; 53:925-930.

Correspondence to: Professor John V. Reynolds, Dept of Clinical Surgery, Trinity Centre, St.James’s Hospital, Dublin 8. Tel: (01)4162500; e-mail: [email protected]


ORI

GIN

AL

PAPE

R


intrODUCtiOnColorectal cancer (CRC) is the most frequently diagnosed non-cutaneous cancer in Ireland and Europe,1,2 and the second most common oncological cause of death. Approximately one in every twenty men and one in every thirty women will develop this malignancy before the age of 75 with a documented 50% mortality rate.3 Prognosis in CRC is closely related to disease stage at diagnosis, and patients with early disease may be candidates for curative operative treatment. Unfortunately a large percentage of patients diagnosed with CRC already have advanced disease at presentation, and their survival is therefore greatly reduced.

It is expected that screening for CRC may effect a reduction in mortality of up to 15%.4 With the high incidence of CRC in Europe it is likely that the demand for screening is going to increase. According to the criteria for an effective screening programme it is necessary that all cost-effective primary prevention interventions should be implemented as far as practicable before commencement of screening for a condition. Primary prevention for CRC would include increasing public awareness of risk factors, relevant symptoms, as well as dietary recommendations and lifestyle modifications. It is our hypothesis that

late stage presentation may be attributed at least in part to poor public awareness of CRC symptoms. In this study we analysed the level of knowledge of CRC and its symptoms among a random population of outpatient attendees and compared this with knowledge of other common malignancies. We also performed a review of all patients diagnosed with CRC at our unit over a two-year period to determine duration of symptoms prior to presentation, the percentage of emergency presentations, and disease stage at the time of diagnosis.

Patients anD MetHODs This prospective study was performed at Mayo General Hospital between January and March 2004. All attendees at outpatient clinics were asked to take part in an anonymous questionnaire survey in which their knowledge of CRC symptoms was evaluated. Knowledge of lung cancer and breast cancer symptoms was evaluated for comparison. The term “bowel cancer” was also used as well as CRC in the questionnaire, and participants were asked to write down any symptoms they knew for each of the cancer types. Acceptable symptoms included change in bowel habit, bleeding per rectum and abdominal pain for CRC, persisting cough, haemoptysis and shortness of breath for lung cancer, and a breast

Poor awareness of colorectal cancer symptoms; a preventable cause of emergency and late stage presentationaBstraCtAims To assess knowledge of colorectal cancer (CRC) symptoms among outpatient

attendees, and to review disease stage, presentation and duration of symptoms in patients diagnosed with CRC.

Methods A questionnaire survey was used to evaluate knowledge of symptoms of CRC and other malignancies. A review of patients diagnosed with CRC during a two-year period was performed.

Results Of 350 survey participants 26.6% could name a CRC symptom, compared to 53.4% for lung cancer and 71.5% for breast cancer. Of 102 patients diagnosed with CRC 3.9% had Dukes A disease, 32.4% had Dukes B, 39.2% had Dukes C and 24.5% had distant metastases. Forty per cent of patients presented acutely. The mean duration of symptoms was 24 weeks.

Conclusions Knowledge of CRC symptoms is poor and is reflected in the percentage of late stage and emergency presentations. Increasing public awareness of CRC may lead to earlier presentation and improved survival.

AT Manning, R Waldron, K BarryDept of Surgery, Mayo General Hospital, Castlebar, Co. Mayo

POOR AwARENESS OF COLORECTAL CANCER SYMPTOMS; A PREVENTABLE CAUSE OF EMERGENCY AND LATE STAGE PRESENTATION



ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


early presentation is well known with 90% five-year survival for patients with Dukes A tumours, however this unfortunately applies to only 3.9% of our patients. It is generally accepted that increased awareness of the disease among the public will lead to earlier presentation, diagnosis and an increased likelihood of curative treatment,6 and our results suggest that the low level of awareness of CRC in this population may in part explain the augmented percentage of late stage and emergency presentation. McArdle et al analysed 3,200 CRC patients in Scotland documenting that post-operative mortality is augmented and long term survival is decreased for patients with CRC who present as an emergency compared to patients with the same disease stage undergoing elective surgery.7 It is interesting to note that only 30.8% of patients presented as emergencies compared to 42% in our patient group, again possibly reflecting the poor public awareness of CRC and its symptoms in our population.

While randomized controlled trials have demonstrated that screening results in a significant reduction in mortality,4,8,9 there is in addition suggestive evidence from epidemiological and experimental studies that dietary and lifestyle factors are likely to impact on the risk of developing CRC.10,11 It has been postulated that educational strategies directed at primary prevention may effect a reduction in mortality equal to that achieved by screening and in a more cost-effective manner,12 but there is little evidence of effort to address this issue. Increasing public awareness of CRC may therefore in itself lead to reduced mortality in terms of increasing primary prevention, but also the success of any future screening programme for CRC will depend on this. It is known that mortality reduction by screening can only be achieved by high compliance rates which would require widespread awareness of the disease. With the current level of knowledge this is unlikely and recent trials of CRC screening in UK and Norway report compliance rates between only 38% and 67%.

This study shows that the level of knowledge of CRC and its symptoms is surprisingly low. This lack of public awareness may in part explain the high percentage of late stage and emergency presentations as seen in this study. In order to begin to reduce mortality from CRC by primary prevention and subsequently by screening, it is essential that both local and national educational strategies be put in place to increase public awareness of this disease.

REFERENCES1. Cancer in Ireland 1994 – 2001; incidence, mortality and

treatment. A report from the National Cancer Registry of Ireland June 2005

2. Parkin DM, Bray F, Ferlay J et al. Estimating the world cancer burden: Globocan 2000. Int J Cancer 2001; 94: 153-156

3. Cancer in Ireland, 1997; incidence and mortality. A report from the National Cancer Registry Sept 2000

4. Hardcastle JD, Chamberlain JO, Robinson MHE et al. Randomised controlled trial of faecal-occult-blood Lancet 1996; 348:1472-1477.

5. Pullyblank AM, Dixon N, Dixon AR. The impact of bowel cancer awareness week. Colorectal Disease 2002; 4: 483-485

6. McLennan I, Hill J. How can doctors diagnose colorectal cancer earlier? By increasing patient’s awareness of the disease an investigating them promptly when they present. BMJ 1993; 306: 1707

7. McArdle CS, Hole DJ. Emergency presentation of colorectal cancer is associated with poor 5-year survival. Br J Surg 2004; 91: 605-609

8. Kronborg O, Fenger C, Olsen J et al. Randomised study of screening for colorectal cancer with faecal-occult blood test. Lancet 1996; 348: 1467-1471

9. Mandel JS, Bond JH, Church TR et al. Reducing mortality from colorectal cancer by screening for fecal occult blood. N Eng J Med 1993; 328: 1365-1371

10. Martinez ME. Primary prevention of colorectal cancer: lifestyle, nutrition, exercise. Recent Results Cancer Res. 2005; 166: 177-211

11. Giovannucci E. Modifiable risk factors for colon cancer. Gastroenterolol Clin North Am. 2002 Dec; 314: 925-43

12. Agrez MV, Coory M, Cockburn J. Population screening for colorectal carcinoma with fecal-occult blood testing, Are we sufficiently informed? Cancer 1998; 82: 1803-1807

Correspondence to: Mr K Barry, Consultant General Surgeon, Mayo General Hospital, Castlebar, Co. Mayo; Tel: +353 94902 8044; Fax: +353 94902 8044Email: [email protected] / [email protected]

POOR AwARENESS OF COLORECTAL CANCER SYMPTOMS; A PREVENTABLE CAUSE OF EMERGENCY AND LATE STAGE PRESENTATION

lump, nipple discharge and skin changes for breast cancer. Those attending colorectal or breast clinics were not asked to participate. Details including age, sex and relevant family history were recorded for each participant. Using SPSS software (v12.0) a chi-squared analysis was used to test for differences between knowledge of the three cancer types, and a p-value of less than 0.05 was accepted as significant.

All patients diagnosed with CRC between 2002 and 2004 at Mayo General Hospital were identified using the hospital in-patient enquiry database (HIPE). A retrospective chart review was performed. Details of mode of presentation (accident and emergency department or out-patient referral) and total duration of symptoms prior to presentation were determined from the medical notes. Histological reports were examined to determine the Dukes’ staging for each case. Operation notes and radiological investigations were reviewed to determine if distant metastases were present at diagnosis.

resUlts In the questionnaire survey data were collected from 350 outpatient attendees. The mean age of participants was 43 years (range 18 – 84) and 25.4% were male. As shown in Figure 1, only 26.6% of participants could name one symptom of CRC, compared with 53.4% for lung cancer and 71.1% for breast cancer (p = 0.0001 vs. CRC). Patients with a family history of colorectal cancer in a first degree relative were more likely to name a symptom compared to those with no family history (38.6% versus 24.8%; p = 0.053). Female respondents were significantly more likely to name a symptom of colorectal cancer compared to males (31% versus 12.4%; p = 0.0001). Seventy-three per cent of those questioned were unaware that CRC is a common disease. Ninety-seven per cent of participants stated that they would participate in a screening programme for colorectal cancer.

A total of 102 patients were diagnosed with CRC over the two-year period from January 2002 to December 2003. The mean age at diagnosis was 67 years (range 33 – 90), and 65 patients (63.7%) were male. Forty-two patients (41.2%) presented acutely to the Accident and Emergency department, the remainder were first seen in the out-patient department. Of the 102 patients diagnosed with CRC, histological examination revealed that 3.9% had Dukes A tumours, 32.4% had Dukes B, 39.2% had Dukes C, and 24.5% had distant metastases at the time of diagnosis (Figure 2). Patients

Figure 1 —PERCENTAGE OF OUT-PATIENT ATTENDEES (N = 350) CAPABLE OF NAMING ONE SYMP-TOM OF COLORECTAL, LUNG AND BREAST CANCER. P<0.001 V’S CRC.

who presented acutely were more likely to have distant metastases at time of diagnosis (38.1% versus 15%). The mean duration of symptoms for all patients was 24 weeks. There was no significant difference in disease stage at diagnosis or duration of symptoms between male and female patients.

DisCUssiOnWe have demonstrated that despite being the most commonly diagnosed non-cutaneous cancer in this country, awareness of CRC and knowledge of its symptoms is poor, with almost three quarters of those questioned unable to name a symptom of the disease. In a similar study in the United Kingdom, 44% of general practice attendees could name a CRC symptom whereas 85% could name a breast cancer symptom.5 Although this shows a better knowledge of CRC symptoms than seen in our survey, the unfavourable comparison with breast cancer is very similar.

In this study 63.7% of patients diagnosed with CRC had advanced disease at presentation (Dukes C or distant metastases) and 42% presented as emergencies. The possibility of long-term survival with

at Manning et al

Figure 2 —DISEASE STAGE AT DIAGNOSIS OF 102 PATIENTS DIAGNOSED wITH CRC OVER A TwO YEAR PERIOD


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOnIn the last few years, Hormone Replacement Therapy (HRT) prescribing has changed. This has been largely due to the publication of recent studies. The ‘Million Women Study’ in the United Kingdom recruited 1084110 women aged 50-64 years between 1996 and 2001.1 This observational study provided information about their use of HRT and other personal details. These women were followed-up for cancer incidence and death.1

The ‘Women’s Health Initiative’ in the United States enrolled 161809 postmenopausal women aged 50 to 79 years.2 This study focused on defining the risks and benefits of strategies that could potentially reduce the incidence of heart disease, breast and colorectal cancer, and fractures in postmenopausal women. The trial of combined oestrogen and progestogen in women with a uterus was stopped early based on health risks that exceeded health benefits over an average follow-up of 5.2 years. A parallel trial of oestrogen alone in women who have had a hysterectomy was allowed to continue.2

The aim of this study was to establish whether prescribing of HRT in Northern Ireland has changed since the publication of these studies and to identify the current prescribing patterns.

MetHODA structured questionnaire was posted to all hospital based Consultants, Specialist Registrars and Senior

House Officers in Obstetrics and Gynaecology in Northern Ireland. Questions included the main indications for HRT prescribing, (sequential or continuous combined preparations), duration of use, presence of associated risk factors and discussion regarding alternative methods for symptom control and prophylaxis. A stamped addressed envelope was included to aid reply.

resUltsSeventy-five per cent of Consultants (39/52), 42% of Specialist Registrars (16/38) and 26% of Senior House Officers (16/60) replied. Overall 54% of respondents indicated that they had changed their prescribing practice on HRT in the past year; 52% of Consultants, 56% of Specialist Registrars and 63% of Senior House Officers (Figure 1).

The most common primary indication for prescribing HRT was vasomotor symptoms (93%). This was not dependant on changed prescribing practices. Sixty-six per cent of respondents felt it was justified to prescribe HRT for osteoporosis prevention in patients with a strong family history of osteoporosis in the absence of vasomotor symptoms. This was more common among the Consultants and Specialist Registrars who had not changed their prescribing practice in the last year.

Fifty-six per cent of respondents recommended 1-5 years treatment as the optimum duration of use; 45% of the Consultants, 78% of the Specialist

HRT: Have we changed?aBstraCtBackground In recent years, medical controversy, has surrounded the prescribing of

hormone replacement therapy (HRT). Aim This study aimed to establish whether prescribing of HRT in Northern Ireland has

changed and what the current prescribing patterns are.Method A structured questionnaire was sent by post to all medical staff in Obstetrics and

Gynaecology in Northern Ireland. A stamped addressed envelope was included.Results Overall 54% of respondents indicated that they had changed their prescribing

practice on HRT in the past year. The primary indication for prescribing HRT was vasomotor symptoms (93%). Fifty-six per cent of doctors recommended 1-5 years as duration of use. Oral preparations were those most commonly prescribed (57%). The dose and type chosen were the same whether prescribing had changed or not.

Conclusions More than half of all doctors who responded had changed their prescribing practices on HRT, yet some respondents still preferred more traditional prescribing.

O Conlon, K McKinneyDept of Obstetrics and Gynaecology, Daisy Hill Hospital, Newry

HRT: HAVE wE CHANGED?

Registrars and 88% of the Senior House Officers. Twenty-three per cent of doctors recommended 5-10 years as duration of use and only 4% of doctors recommended > 10 years of use. Consultants who stated they had not changed their prescribing practices indicated a longer duration of use. The most commonly quoted short-term risk for HRT use was venous thromboembolism (79%). The most common long-term concern with HRT use was breast carcinoma (94%), with 26% of respondents quoting endometrial carcinoma as a long-term concern. Ten per cent of respondents indicated that ovarian carcinoma was a long-term risk of HRT.

Oral preparations were the most commonly prescribed method of HRT (57%) (Figure 2). The dose and type were the same, whether or not prescribing practices had changed. There was a wide variation in the duration of use of sequential preparations. Seventy-six per cent of respondents used a low starting dose i.e. 1mg of oral oestradiol / 50mcg patch of oestradiol, 16% of doctors indicated their starting dose was high i.e. 2mg of oral oestradiol / 100mcg patch of oestradiol and 26% of doctors indicated tibolone as their initial prescription of HRT (some respondents indicated both a low dose preparation and tibolone). Ten per cent of respondents indicated that they would not change to a continuous combined preparation until the age of 54 years. This was not dependant on whether prescribing practice had changed.

Fifty-seven per cent of respondents indicated that they would recommend other preparations in place of HRT. The most common preparation mentioned was edidronates (37%) (Figure 3). Others included selective serotonin reuptake inhibitors (24%), soya/ phyto-oestrogens (21%), black cobosh (19%), red clover (14%), progestins (9%), clonidine (1%), cimetidine (1%) and calcium (1%) (some respondents quoted more than one alternative preparation).

DisCUssiOnIn the UK the ‘Million Women Study’ and in the US the ‘Women’s Health Initiative’ (WHI) have caused considerable debate in the field of HRT prescribing.1,2 The Chairman of the Committee on Safety of Medicines (CSM) in the UK issued further advice on the use of HRT on 3rd December 2003, three months prior to issuing this questionnaire.3 The recommendation was that the minimum effective dose should be used for the shortest duration. This was reflected in the prescribing practices of those who had changed their practice within the last year. However many doctors, especially Specialist

o conlon & K McKinney

FIGURE 3 —OTHER PREPARATIONS PRESCRIBED IN PLACE OF HRT

FIGURE 1 —CHANGE IN PRESCRIBING PRACTICE (%)

FIGURE 2 —TYPES OF HRT PREPARATIONS MOST COMMONLY PRESCRIBED

Registrars, who had not changed their practice were using small doses for short durations anyway prior to the publication of these recommendations.

The recommendations for the use of oestrogen replacement for osteoporosis and hip fractures challenged the recent randomised controlled trials.


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


Long-term data show a reduced risk of fractures in women who used oestrogen for 10-20 years.4 Cauley et al also showed a reduction in vertebral, hip fractures and other non-vertebral fractures with HRT in healthy post menopausal women not at risk of osteoporosis.5 There are no long-term studies on use of alternatives to HRT for osteoporosis prevention. Most studies on bisphosphonate use in the treatment of osteoporosis are directed towards older women who already have osteoporosis or those at high risk of developing it.6 Long-term use of bisphosphonates may cause bones to be more brittle due to suppression of bone formation, which may increase risk of fractures.7 In this study it was the majority of doctors who had not changed their prescribing practice who indicated osteoporosis as an indication for HRT use. The pre-mentioned statement from the Committee on Safety of Medicines issued a few months earlier may have changed prescribing practice related to osteoporosis. Raloxifene was indicated as a treatment for osteoporosis prevention by 2% of doctors. Ettinger et al have shown that raloxifene only reduces vertebral fractures and not fractures of the hip.8

Overall one in three women in the UK are affected by osteoporosis during their lifetime.9 The majority of women by the age of 80 years will have sustained at least one fracture.9 The cost is not only human cost and quality of life issues, but also financial cost. The National Health Service in the UK spends approximately £1.7 billion on peri-fracture care. 9

Therefore, although HRT is now not recommended for osteoporosis prevention, it may still be considered in selected patients.

Breast cancer remains the most serious long-term risk with HRT. Ninety-four per cent of respondents indicated this. The Million Women Study asked women who had mammography to complete a questionnaire on HRT.1 It must be remembered that this was an observational study in healthy women and did not provide direct evidence on those women who have breast cancer and their use and requirement of HRT.

Dr John Stevenson in a press release from the British Menopause Society concluded that the Million Women Study had a very small number of deaths from breast cancer and the follow-up was too short and that the statistical significance of the finding is borderline.10 He also concluded that the increased risk of breast cancer remains much the same as

the risk of breast cancer from drinking alcohol and that it is still less than the increased risk for breast cancer associated with being overweight.10 Marsden and Sacks showed that short-term exposure to HRT in symptomatic patients with a personal history of breast cancer might not significantly increase recurrence rates.11

Another gynaecological cancer mentioned in this study was the risk of endometrial cancer. This was indicated by 26% of doctors. The risk of endometrial cancer though, is reduced in those patients using continuous combined preparations of HRT. Only 10% of doctors indicated a long-term risk of ovarian cancer. La Vecchia et al showed a statistically significant risk of ovarian carcinoma in women using oestrogen only preparations and not those using continuous combined preparations.12 The questionnaire was limited on this topic as it was a ‘tick’ box and the duration or type of HRT was not specified to change risk.

There are many alternatives to HRT, including tibolone. This was indicated as a first line treatment by 26% of respondents. There was no significant difference in those whose prescribing practices had changed. The Million Women Study was the first study to provide risk data about tibolone and breast cancer. This suggests that the reported lack of effect of tibolone on mammographic density may not be an accurate indicator of its effect on the breast.13

Historically sequential preparations were used for longer durations and this is reflected in the varied responses. The risk of use with sequential preparations is endometrial carcinoma, which has previously been discussed. Another aspect is quality of life issue where women who change to continuous combined preparations will not have a monthly bleed.

The British Menopause Society Council Consensus statement on HRT attempts to clarify the recent controversial findings of the WHI and Million Women Studies.14 It concludes that important points to consider before prescribing HRT are:

• long-term use needs to be assessed for each individual at regular intervals,

• women with premature menopause should normally be offered HRT until the age of 51 years,

• the effects of HRT on cardiovascular disease and Alzheimer’s disease are still uncertain and HRT should not be solely prescribed for prevention of these.

The final question asked was the use of alternatives to HRT. Fifty-seven per cent of doctors indicated they would prescribe an alternate to HRT. The most common alternative was the use of edidronates and the next was selective serotonin reuptake inhibitors. Phyto-oestrogens were the most common ‘herbal’ preparation advised, which gives some benefit to some women. One consultant commented, ‘if it helps, use it.’ Other comments included advice on exercise and to stop smoking. Finally a Senior House Officer concluded that ‘the menopause is a self limiting, non-lethal condition mostly requiring tender loving care (TLC)’. Many doctors and patients would regard this as trivialising a condition causing serious morbidity and quality of life problems for many women. The questionnaire was designed to be brief and easy to complete. It gives us some insight into the prescribing practices and wide variations on HRT in Northern Ireland.

There are several areas within the study which could be explored in more detail in future research projects.

REFERENCES1. Million Women Study Collaborators. Breast cancer and

hormone- replacement therapy in the Million Women Study. Lancet 2003; 362: 419-427.

2. Writing Group for the Women’s Health Initiative Investigators. Risks and benefits of oestrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomised controlled trial. JAMA 2002: 288: 321-33.

3. Further confusion in postmenopausal health: Committee on Safety of Medicines Message 3rd December 2003.

4. Cauley JA, Zmuda JM, Ensrud KE, Bauer DC, Ettinger B. Timing of oestrogen replacement therapy for optimal osteoporosis prevention. J Clin Endocrinol Metab 2001; 86: 5700-5.

5. Cauley JA, Robbins J, Chen Z, et al Women’s Health Initiative Investigators. Effects of oestrogen plus progestin on risk of fracture and bone mineral density: the Woman’s Health Initiative randomised trial. JAMA 2003; 290:1729-38.

6. Marcus R, Wong M, Heath H, Stock JL. Antiresorptive treatment of postmenopausal osteoporosis: comparisons of study designs and outcomes in large clinical trials with fracture as an endpoint. Endocr Rev 2002; 23: 16-37.

7. Chavassieux PM, Arlot ME, Reda C et al Histomorphometric assessment of the long-term effects of alendronate on bone quality and remodelling in patients with osteoporosis. J Clin Invest 1997; 100: 1475-80.

8. Ettinger B, Black DM, Mitlak BH, et al Reduction in vertebral fracture risk in postmenopausal women with osteoporosis treated with raloxifene: results from a 3-year randomised clinical trial. Multiple Outcomes of Raloxifene Evaluation (MORE) Investigators. JAMA 1999; 282: 637-45.

9. Royal College of Physicians. Osteoporosis: clinical guidelines for prevention and treatment. Update on pharmacological interventions and an algorithm for management. London, Royal College of Physicians 2000.

10. Stevenson J. Hormone Replacement Therapy and Breast Cancer: a Response to the Million Women Study from Women’s Health Concern. Press Release from the British Menopause Society; August 2003.

11. Marsden J, Sacks N. The national randomised trial of hormone replacement therapy in women with a history of early stage breast cancer: an update. J B Menopause Soc 2002; 8: 129.

12. La Vecchia C, Brinton LA, McTiernan A. Cancer risk in menopausal women. Best Pract Res Clin Obstet Gynaecol 2002; 16: 293-307.

13. Lundstrom E, Christow A, Kersemaekers W et al Effects of tibolone and continuous combined hormone replacement therapy on mammographic breast density. Am J Obstet Gynaecol 2002; 186: 717-722.

14. Managing the Menopause. British Menopause Society Council Consensus Statement on Hormone Replacement Therapy. January 2004.

Correspondence to: Dr Orla Conlon, 6 Portadown Road, Armagh, BT61 9EETel: 04837523890, Mobile: 00447767271208Email: [email protected]

HRT: HAVE wE CHANGED? o conlon & K McKinney


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


intrODUCtiOn The maternity hospitals in Dublin have a long tradition in caring for the mothers and neonates. All neonates have free open access to the Baby Clinic on weekdays at the maternity hospitals in Dublin for routine assessment of all neonatal health issues. Recent observation, however suggested an increase in number of neonates attending the hospital outside the scheduled 9 am to 1 pm Baby Clinic hours. This has added considerable workload to the acute neonatal service and the hospital staff on duty. Despite this service being run for many years, it is not formally funded by the Health Service Executive (HSE). This paper examines the number and nature of neonates attending the maternity hospital outside the Baby Clinic hours and to identify potential solutions for improvement and rationalization of the services provided.

MetHODsThis study was conducted at The National Maternity Hospital, Holles Street in Dublin. Neonates attending outside the clinic hours of 9 am to 1 pm were seen as un-booked emergency cases. This acute neonatal service was staffed by a senior house officer (SHO), a registrar and a consultant in neonatology (off-site). A peadiatric nurse will assist up to 8 pm. Thereafter

a midwife with peadiatric experience will come on duty until 8 am the following day.

There were 8,466 live births recorded by the hospital in 2003. The total number of neonates who attended the emergency department during this period out of hours were obtained from the acute neonatal register book.

The nature of the presenting complaints and subsequent diagnoses were ascertained by reviewing all case notes retrospectively from January to December 2003. Neonatal attendances were categorized according to the presenting complaints. These data were collected onto a spreadsheet and analyzed using SPSS package 11.1 for Microsoft Windows.

resUltsThere were 736 acute neonatal attendances over the twelve-month period. Four attendances were excluded from the study because of incomplete data. Out of the 732 attendances, 106 neonates (14.4%) required further investigation and treatment. Eighty-one neonates were admitted to the neonatology unit at the maternity hospital and the other 25 were transferred directly to a tertiary paediatric hospital.

A review of neonatal attendances out of hours in a Dublin maternity hospitalaBstraCtBackground All neonates have free open access to the Baby Clinic at the maternity

hospitals in Dublin for assessment of neonatal health issues. Through observation, however there is an increase in number of neonates attending the hospital outside the Baby Clinic hours.

Aims To determine the number of neonates attending the acute neonatal service out of hours and to identify the percentage of neonates treated as true emergency.

Methods Retrospective chart review over a twelve-month period.Results Seven hundred and thirty-two neonates attended the hospital out of hours. The

majority were diagnosed with gastrointestinal problems (228 / 31%), jaundice (101 / 13.7%), respiratory problems (82 / 11.1%) and skin disorders (79 / 10.7%). Only 106 (14.4%) attendances warranted admissions.

Conclusions A large number of neonatal attendances did not require acute assessment out of hours and were managed by reassurance and maternal education. A centralized phone-in-triage system was suggested to relieve the strain on the acute neonatal service.

LFA Wong1, KT Lim2, A Twomey1, J Murphy1

Dept of Neonatology, National Maternity Hospital, Dublin,1 Dept of Surgery, Scunthorpe General Hospital, United Kingdom2

A REVIEw OF NEONATAL ATTENDANCES OUT OF HOURS IN A DUBLIN MATERNITY HOSPITAL lFa Wong et al

Of the 732 attendances, 681 (93.0%) were term babies and 51 (6.97%) were preterm babies. Three hundred and nineteen (43.3%) neonates were female while 417 (56.7%) were males. The median age of the neonates on presentation was 13.0 days (range 1-167 days).

Only 132 neonates (17.9%) were either referred by their general practitioner or public health nurse to the emergency department and 18 (13.6%) of these were admitted. Six hundred (81.5 %) attendances presented directly to the emergency department and 88 (14.7%) of these were admitted. A minority of these had telephoned the SHO prior to presentation. The main presenting complaints or diagnoses are outlined in Table 1 along with those admitted.

DisCUssiOn This paper identified about 8.7% of the live neonates delivered at The National Maternity Hospital, Holles Street in Dublin would attend the acute neonatal service out of hours. Only 14.4% of these 732 neonates were admitted and treated as true emergency that required further investigation and treatment. The majority of these neonates presented to the acute neonatal service out of hours did not require urgent intervention and could be assessed and reviewed the following morning at the Baby Clinic. About a quarter of the neonates who required admission had to be transferred to a specialist paediatric hospital. The admission rate of neonates out of hours referred by the general practitioners or public health nurses was not much different to the

Table 1THE NUMBER OF ATTENDANCES THAT wERE ADMITTED ACCORDING TO THE MAIN PRESENTING COMPLAINTS OR DIAGNOSES (n=732)

PRESENTING COMPLAINTS OR DIAGNOSES NUMBER OFATTENDANCES (%)

NUMBER OFADMISSIONS (%)

Gastrointestinal Vomiting/reflux diarrhea constipation/colic poor feeding oral candidiasis

228 (31)88 (12.0)17 (2.32)

70 (9.56)40 (5.46)13 (1.78)

26 (24.6)

Jaundice 101 (13.7) 15 (14.2)

Respiratory snuffles transient tachypnoea/ periodic breathing cough/sneeze apnoea cyanosis

82 (11.1)36(4.12)10 (1.37)22 (3.01)7 (0.96)7 (0.96)

10 (9.4)

Skin erythma toxicum bacterial skin infection others

79 (10.7)43 (5.87)18 (2.46)18(2.46)

8 (7.5)

Umbilical cord concerns 38 (5.2) 1 (0.9)

Neurological irritability/lethargic seizures/twitches/jittery

35 (4.8)13 (1.78)22 (3.01)

12 (11.3)

Eye Discharge/ Sticky Eye 34 (4.6) 3 (2.8)

Weight Loss 12 (1.6) 7 (6.6)

Pyrexia 24 (3.3) 16 (15.1)

Musculoskeletal Injury 8 (1.1) 0 (0)

Cardiac 2 (0.3) 1 (0.9)

Others 89 (12.1) 7 (6.7)

TOTAL 732 (100) 106 (100)


ORI

GIN

AL

PAPE

R

ORI

GIN

AL

PAPE

R


walk-in/phone-in attendances. More importantly, this paper showed the majority of these neonatal attendances out of hours were dealt with by giving advice, education and reassurance. It is sensible to encourage parents to bring their neonates to the scheduled Baby Clinic. A telephone triage system is suggested to identify the category of neonates who do not require urgent assessment and treatment and to book them for the scheduled Baby Clinic.

In our institution, neonates are discharged routinely at day 2 in uncomplicated vaginal deliveries and at day 5 after caesarean sections. Our results are in line with other institutions, that early discharge post partum at day 2 does not increase readmission rates.1,2,3 Readmission was less likely in neonates who had home visits or telephone contacts by the public health nurse within the first three days after discharge.4 The readmission rate were more frequent in neonates with feeding difficulties, jaundice and pyrexia at presentation. This is in contrast with the low 2.8% readmission rate in the study conducted by Oddie.2 The causes of concern for readmission in their study were somewhat similar which includes suspected infection, colic, convulsions, vomiting and jaundice.

The majority of breastfeeding mothers were worried about neonatal loose stools, which in fact were normal breastfed consistency stools. All breast fed neonates should either be visited by the public health nurse or booked into the health centre for regular weight checks and breast feeding technique checks upon discharge from the maternity hospital. This will help to alleviate fear and anxiety and to provide some continuity of care to the mother and neonates. Support for breastfeeding mothers in the community is an important area for improvement of maternal and neonatal health.

Most parents would have thought neonatal noisy breathing, snuffles and periodic breathing equates to chest infection and jitteriness being seizures. These symptoms are considered normal in neonates and can be reassured. Most rashes are due to erythema toxicum and not from meningitis and can be reassured unless there is a doubt.

Our institution now adopts a pre-discharge education on common neonatal symptoms and signs, feeding techniques, umbilical cord care, rash, eye care, snuffles, umbilical cord care, constipation and colic in an effort to reduce these presenting complaints out of hours. This information can be

reinforced by the public health or community nurses and at the six-week postnatal check. In an article by Nelson et al on six-week postnatal check, 26% planned to attend hospital clinics and 55% planned to attend the GP. The remainder planned to attend a consultant privately or had been referred to a specialized paediatric clinic. Factors influential for those choosing GPs were convenience, familiarity and opportunity for discussion. Hospital instructions and the expected quality of examination were important for those choosing the hospital.5

Upon discharge from the maternity hospital, the next point of contact for mothers should be their general practitioners or local public health nurse for any concern on their neonates. General practitioners should receive adequate training and exposure in neonatal care through their continuing medical education in lectures and workshops. It is important to note that newborn training in a paediatric residency program is insufficient for primary care.6

With the nationwide practice of early discharge, this study clearly identifies the need for supportive organized community care provided by the general practitioners, public health nurses and community midwives. Resources should be allocated and tailored to provide adequate training to healthcare professionals in the community. Issues of neonatal care should be addressed by the respective health boards under the HSE by means of education at a community level by means of leaflets or web education. Usage of credible web information can augment parental education on healthcare issues.7

During out of hours, the feasibility of a telephone triage system could reduce the need of presentation out of hours. These findings would support a triage system and telephone advice line, handled by trained paediatric nurses or doctors on call. Large scale after-hours telephone coverage systems is shown to be effective and achieved satisfaction of 96% to 99% among parents. 8 A 24-hour telephone service similar to the NHS Direct 9 in the United Kingdom could be modelled here. Its concept would reduce parents from turning up out of hours for advice. Advice given is based upon the skills and experience of nurses assisted with a comprehensive computer system.

Many studies on telephone advice were based in paediatric emergency department settings, a situation which could be easily be adapted in our institution.

Majority of attendances to our emergency department were not acute, did not require admissions and could have been dealt with through appropriate structured telephone call advice. A centralized neonatal health line advice system could be implemented offering advice to new parents on neonatal issues. This could be implemented centrally in the three Dublin maternity hospitals or nationwide and could potentially reduce out of hours healthcare costs, workload and staffing levels.

Appropriate support mechanisms are likely to play a key role in ensuring the safety of early discharge and emphasis should be placed on the application of scientific research methods for the development and refinement of guidelines around neonatal care.10

In conclusion, a significant number of neonates attend out of hours for various reasons. Majority of neonates who presented did not require admission but managed with advice, education and reassurance. Current resources are not designed to deal with these attendances, with the need of more staff on duty and a properly equipped assessment area. An alternative neonatal care delivery through a centralized phone-in triage system could benefit both parents and staff.

REFERENCES1. Escobar GJ, Greene JD, Hulac P, et al. Rehospitalisation

after birth: patterns among infants of all gestations. Arch Dis Child 2005; 90(2):125-31.

2. Oddie SJ, Hammal D, Richmond S, Parker L. Early discharge and readmission to hospital in the first month of life in the Northern Region of the UK during 1998: a case cohort study. Arch Dis Child 2005; 90(2):119-24.

3. Scott-Jupp R. Readmission of neonates. Arch Dis Child 2005; 90(2):111-2.

4. Nelson E, Brannagain DO, McCabe B, Moran D. The six week postnatal check: women’s choice of service provider. Ir Med J 2002; 95(10): 302-4.

5. Cappleman J. Community neonatal nursing work. J Adv Nurs 2004; 48(2):167-74.

6. Walton DM, Edwards MC. Nationwide survey of paediatric residency training in newborn medicine: preparation for primary care practice. Paediatrics 2002; 110(6):1081-7.

7. Peters R, Sikorski R. Paediatrics on the PC. JAMA 1998; 279(19):1583-4.

8. Poole SR, Schmitt BD, Carruth T, Peterson-Smith A, Slusarski M. After hours telephone coverage: the application of an area-wide telephone triage and advice system for paediatric practices. Paediatrics 1993; 92(5):670-9.

9. http://www.nhsdirect.nhs.uk

10. Danielsen B, Castles AG, Damberg CL, Gould JB. Newborn discharge timing and readmissions: California, 1992-1995. Paediatrics 2000; 106(1 Pt 1):31-9.

Correspondence to: Dr LFA Wong E-mail: [email protected]

A REVIEw OF NEONATAL ATTENDANCES OUT OF HOURS IN A DUBLIN MATERNITY HOSPITAL lFa Wong et al


CASE

REP

ORT


CASE

REP

ORT

INTRODUCTIONBoerhaave’ssyndrome,alsoknownasspontaneousperforationoftheoesophagus,isararecondition.HermannBoerhaave,ProfessorofMedicineatLeidenUniversity,initiallydescribedthesyndromein1724followingtreatmentoftheGrandAdmiraloftheDutchfleet,BaronJanvanWassenaer.1TheBaronhadfeastedonduckandbeer,beganvomitingandthendevelopedseverepain,complainingtohisservantsthatsomethingneartheupperpartofhisstomachhadtorn.Deathoccurred18hourslater,andatautopsyBoerhaaveobservedemphysemaofthechestwallandabdomen,andaruptureoftheleftposterolateralwalloftheoesophagus3inchesabovethediaphragm,withoutevidenceofoesophagealpathology.In1946NormanBarrettreportedthefirstearlydiagnosisandsuccessfulrepair.2Themortalityratewasapproximately50%uptothe1980s,butmorerecentlythemortalityrateisapproximately30%becauseofearlierdiagnosis,surgicalrepairandimprovementsinintensivecare.3-7

Thetypicalpattern,so-calledMackler’sTriad,ofspontaneousoesophagealperforationisviolentretchingorvomitingfollowingfoodandveryoftenalcohol,whichisthenfollowedbyseverechestpainandsurgicalemphysema.8Asuddenriseintheintraoesophagealpressureleadstoatransmural

perforationoftheoesophagus.Anatomicallytheoesophagushasnoserosallayer,whichallowsgastriccontentsdirectaccessintothemediastinumandpleuralcavity,resultinginearlymediastinitis,sepsisandfrequentlymulti-organfailure.Adelayindiagnosisiscommon.Becauseofitsrarity,theclassicalpicturemaybemissed,andinatleast40%ofcasesthepictureisnotclassical,anditmayhavebeenmistakenforamyocardialinfarction,pancreatitis,perforatedulcer,pericarditis,mesentericthrombosis,dissectinganeurysmandothercausesofsevereupperabdominalorchestpain.

Thereisgeneralconsensusonmanagementprincipleswherethediagnosisisestablishedearly,andsurgerywithin24hoursofpresentationhasabetterchanceofsuccessfuloesophagealrepairandpatientsurvivalthansurgerybeyondthistimepoint.3-7Themanagementofpatientspresentinglateduetoamisseddiagnosisiscontroversial,withadvocatesforsurgicalrepair,repairoveraT-tube,oesophagectomy,oraconservativeapproach.9-11Thefirstsuccessfuloutcomefromaconservativeapproachwaspublishedin197512,andtheadventofcoveredendoprostheticdeviceshasimprovedthetherapeuticoptions.13

Caseseriesintheliteraturearefew,andmainlyemanatefromtertiaryoesophagealunits.Thetwo

Managementofspontaneousruptureoftheoesophagus(Boerhaave’ssyndrome):Singlecentreexperienceof18casesABSTRACTBackgroundSpontaneousoesophagealrupture(Boerhaave’ssyndrome)israre,and

carriesahighattendantriskofmortality.MethodsAretrospectiveeight-yearreviewfromatertiaryunit.Results Eighteenpatientsweremanaged,withameanageof57(39-88years).Eight

patientspresentedearlyandunderwentsurgery,sevenwithprimaryclosureandonewithexclusionanddiversion.Therewasonedeathinthisgroup.Tenpatientsweremanagedconservatively.Inthisgroup,twounderwentanoesophagectomybecauseoffailedconservativemeasures,andfourhadanendoprosthesisinserted.Onepatientdiedinthisgrouponthefirstadmission,buttwopatientswithstentsinsitudiedfrommassivebleedingrelatingtoanaorto-oesophagealfistulaat39daysand189daysrespectivelyfollowingpresentation.

Conclusions Surgicalinterventionremainsthegoldstandardwhenthediagnosisismadeearly.Forlatediagnoses,thisseriessuggestscautionintheuseofendoprostheses.

RPrichard,JButt,NAl-Sariff,SFrohlich,SMurphy,BManning,NRavi,JVReynoldsDeptofClinicalSurgery,StJames’sHospitalandTrinityCollegeDublin

MAnAgEMEnT Of SPOnTAnEOuS RuPTuRE Of ThE OESOPhAguS (BOERhAAvE’S SyndROME): SinglE CEnTRE ExPERiEnCE Of 18 CASES

mostrecent,an18-caseseriesfromSiewertetalfromMunichin1997,anda14-caseseries,publishedin2000,fromGriffinandcolleaguesinNewcastle,underlinethispoint.6,7WepresenthereintheexperienceofthisUnitinthemanagementof18cases,andhighlightthelessonslearned.

METHODSAretrospectivereviewofallpatientswithadiagnosisofspontaneousoesophagealperforationswhopresentedorwhowerereferredtoStJames’sHospitalfrom1997to2004wasundertaken.PatientswereidentifiedthroughthehospitalHIPEsystem.Caseswereexcludediftheoesophagealperforationwasasaresultofinstrumentation,traumaoranunderlyingmalignancy.

RESULTSThedetailsofall18patientswhofulfilledthecriteriaaredetailedinTable1.FivefurthercaseswerereferredinthistimeperiodwithsuspectedBoerhaave’ssyndrome,allyoungmaleswithsubcutaneousemphysema,chestdiscomfort,andnohistoryofvomiting,andadiagnosisofmediastinalruptureofbullaewasconfidentlymadeafteranegativecontrastswallow.

Eightcasespresentedwithclassicalsymptomatologyandwerediagnosedandarrivedinthiscentrewithin24hoursoftheirfirstsymptom.Inthreecases,thesiteofperforationwasnotintheclassicallowerleftposition,butinthemiddlerightsideoftheoesophagus.Theoperationineightcasesinvolvedathoracotomy,debridementanddrainageofthemediastinumandpleuralspaces,visualisationandrepair(withinterrupted3.0Polydioxanone,Ethicon)oftheentiremucosaldefect,andapleural(n=5)ordiaphragmatic(n=2)flap.Allpatientshadawide-borednaso-gastrictubesecurelypositioned,receivedbroad-spectrumantibiotics,andwerefedpostoperativelybytotalparenteralnutrition.Inonecase(no.9)whowasprofoundlyillonarrival,theperforationwasattheoesophagogastricjunctionwithinalargehiatushernia,withcontaminationofboththeabdominalandthoraciccavity,andpartialgastrectomywithbluntoesophagectomy,feedingjejunostomyandcervicaloesophagostomywasrequired.Shesurvivedagainstconsiderableodds,andhasalwaysdeclinedanyconsiderationofreconstruction.Onepatientdied,a60-year-oldmanwhohadpresentedprofoundlyshockedandwithacutetubularnecrosisandhaemodynamicfailure,succumbingtoARDSandoverwhelmingmultipleorganfailureonthe19thpostoperativeday.

Twoof10casesassignedforconservativemanagementunderwentoesophagectomyafterconservativemeasureswereconsideredtohavefailed.Inonecase,consideredtoounstableatinitialpresentationforthoracotomy,alargeleakpersistedafteronemonthdespiteastent.Thepatientsurviveda3-stageoesophagectomy.Oneotherpatient,afemaleaged58,underwenta2-stageoesophagectomyforpersistentsepsis,anoesophago-bronchialfistula,andfailureofthestent.

Anendoprosthesiswasconsideredinallcasesandusedwithapparentinitialsuccessincontrollingtheleakandongoingsepsisinfourcases.Strikingly,however,twopatientshavesurvivedtheearlyeffectoftheinsulttodielaterfrommassiveuppergastrointestinalhaemorrhagefromanaorto-oesophagealfistula.Oneelderlypatientalsodiedofsepsis,andthusthreeoftenpatientsconservativelymanagedsuccumbedtoeithersepsisorhaemorrhage.

DISCUSSIONThetermspontaneousperforationoftheoesophagus,althoughcommonlyused,isaninappropriateterm,astheruptureisrarelyspontaneousandalmostinvariablyfollowsbarotraumafromasuddenpost-emeticriseinoesophagealpressure.AlthoughtheoriginalreportofBoerhaavedescribesatransversetearinthelowerleftoesophagus,mosttearsarelongtitudinal,varyinginlengthfrom0.5to20cm,locatedontheleftposterolateralwalloftheoesophagus,2to6cmabovethediaphragmin80%ofcases.1Thepredilectionforthissiteisunclear,butitmayresultsfromsplayingofmusclefibresandtheentranceofbloodvesselsandnervesatthissite.Inthisseries,sevenpatientshadspontaneousperforationintotherightthorax.Inasmallseries,thisisprobablyachancefinding,butitdoesemphasisetheimportantpointthatperforationsarenotalwaysintheclassicallowerleftsite,andthatacontrastswallowismandatoryintheinvestigationofallsuspectedcasestodeterminethesiteofperforationandguidetheoperativeapproach.

Thecommonestageofpresentationisbetween40and60yearsofage,withaslightmalepredominancewithmaletofemaleratioof2:1.Inourseriesofpatientsthemeanageofpresentationwas57yearswitha2.2:1maletofemalepredominance.Clinically,Boerhaave’ssyndromepresentsin60%ofcaseswithatriadofsymptoms,whichisknownasMackler’sTriad.Thisencompassesahistoryofvomitingfollowedbythesuddenonsetofsevere

R PRichaRd et al


CASE

REP

ORT


CASE

REP

ORT

Table 1

PATI

ENT

AGE

SEX

LEN

GTH

OF

STAY

IN D

AYS

PRES

ENTA

TIO

N

CON

TRAS

T S

WAL

LOW

OG

D

DIAG

NO

SIS

DELA

Y

SUSP

ECTE

D DI

AGN

OSI

S

OU

TCO

ME

1 39 M 17 Haematemesis,leftsidedChestPain Lowerthird,left Yes <24h Nocomplications

2 60 M 19 Vomiting,leftsidedchestpain,shock O-Gjunction,left yes <24h Death,multiple

organfialure

3 52 M 91Vomiting,left

sidedchestpain,peritonitis,shock

10cm,middlethird,right

Stentplacedon

day30<24h

Oesophagectomyat6weeksfor

failedconservativemanagementand

stentfailureMildCVA

4 58 F 136 Leftsidedpain 5cm,right stent 96h Stentfailure,lateoesophagectomy

5 52 M 30 Retrosternalchestpain Lowerthird,left no <24h Nocomplications

6 58 F 49 Vomitingandrightsidedchestpain

10cm,lowerthird,right no <24h EmpyemaAtrial

fibrillation

7 64 M 28 Vomitingandrightsidedchestpain

5cm,mid-oesophagus,right no <24h Atrialfibrillation

8 53 M 37 Vomiting,leftpleuriticchestpain

4cm,loweroesophagus,left no <24h ARDS,Atrial

fibrillation

9 58 F 34 Vomiting,chestpain 5cm,midoesophagus,right no <24h Empyema

10 79 F 89 Vomiting,abdominalandchestpain

OGjunctioninhiatushernia,left no <24h

Proximalgastrectomy,cervical

oesophagostomy,andoesophagectomy

11 88 F 27 Leftsidedchestpain,hypoxia

Completedisruptionofoesophagus

no 5days pneumoniaDeathfromPneumonia

andMOF

12 72 F 86 Vomiting,leftsidedchestpain,shock

Leakofcontrastleftmiddlethird

yes–andcovered

stentdeployed

3days cholecystitis pneumonia

13 74 F 39 Unresolvingrightsidedempyema

Fistulabetweenoesophagusand

empyema

yes–andcovered

stentdeployed

10days pneumonia Death,aorto-

oesophagealfistula

14 48 M 58Vomiting,epigastricandleftsidedchest

pain

Leakloweroesophaguswith

contrastintoRightside

no 10days

Pleuraleffusion

Empyemaradiologicallydrained

15 71 M 150 Vomitingandcentralchestpain

Loweroesophagus,

left

yes–covered

stentdeployed

10days

pleuraleffusion

Deathat6months,aorto-oesophageal

fustula

16 55 M 37 Vomiting,epigastricdiscomfort

Loweroesophagus,left

yes–nostentplaced

19days pancreatitis

EmpyemaRadiologically

drained

17 73 M 98 Vomiting,Chestpainandorthnopea

mid-oesophagus,right

yes–toolargetostent

10days pneumonia Pleuraleffusion

18 28 M 10 Vomiting,subcut.emphysema Nonevident no 2

days pneumonia Nocomplications


lowerthoracicorepigastricpainandthefindingofsubcutaneousemphysemaonexamination.Twoofthreeelementsofthetriadwerepresentinallbutoneofthiscurrentseries.PatientsareoftenshockedasaresultofassociatedsepsisandmayexhibitHamman’scrunch(astheheartbeatsagainstair-filledtissues)onpalpationofthethorax.Auscultationrevealssignsconsistentwithachesteffusionorpneumothorax.Thecommonestpresentingsymptomsinourstudywereofthesuddenonsetofseverechestorepigastricpainandassociatedvomiting.InitialinvestigationsincludeachestX-ray,whichmaydemonstratethepresenceofahydropneumothorax,pneumomediastinumortheV-signofNacleriowhichismediastinalwideningduetoretrocardiacemphysematousdissectionoffascialplanes.Thisdescribesthelateraltrackingofairwithinthemediastinum.However,theinitialchestX-raywillbenormalin10%ofpatients.Acontraststudywithawater-solubleagentisperformedtodelineatethelocationandsizeoftheperforation.Ifnegative,andtheindexofsuspicionishigh,thiscanberepeatedwithbarium.Contrastradiographyisfalselynegativeinupto25%ofcasesoftransmuraloesophagealtears.ACTscanmayalsobeperformednotonlytoconfirmtheperforationandextentofcontamination,butalsotoruleoutthepresenceofalternativepathology.Anendoscopyprovidesausefuladjunctbeingbothdiagnosticandtherapeutic.Eightpatientsinthisseriesunderwentearlyendoscopy.Thisisnotonlyanadjuncttoradiologyindeterminingthelocationoftheperforation,itcanalsoassesstheextentofmucosalinjury,thesuitabilityforanendoprosthesis,andenablethoroughlavageofthemediastinumthroughtheperforation.7

Surgeryisthemainstayofmanagementforthemajorityofpatientspresentingwithinthefirst24hours.Inthisseries,alleightpatientsundergoingearlysurgeryhadaprimarymucosalclosureandbuttressingofthesuturelinewitheitherpleuraordiaphragm.Nopatientexperiencedbreakdownofthisrepairasevidencedclinicallyorradiologically.Onepatientdied,apatientalreadyprogressingtomultipleorgandysfunctionatthetimeofpresentation,andthesedatasupportanearlyaggressiveoperativeapproach.Nopatientinthisserieshadsurgeryplannedastheprimarytreatmentoptioniftheypresentedbeyond24hours,butitshouldbenotedthatsomeseriesreportexcellentresultsfromoesophagealrepairinpatientspresentingbeyond24hours.9,14

Aconservativeapproachisinappropriateforthemajorityofpatientswhohaveextensivemediastinalcontaminationfrompost-emeticperforation.Aconservativeapproachwasnonethelessthechosenapproachintenpatientsinthisseries,andthecasestudiesrepresentthespectrumofthosethatcanbeconsideredforconservativemanagement.Sevencaseswerediagnosedinitiallywithanothercondition,inparticularpneumonia,pancreatitis,orcholecystitis,andweretransferredonlywhenpleuralaspirationordrainageclearlydemonstratedgastriccontents.Thisisclearlyaself-selectedgroup,anumeratorfigurewhohavenotmanifestedaprofoundimmunoinflammatoryorsepticresponsetomediastinalcontamination.Aconstantfeatureinthesecasesisthebreachofthemediastinalpleurawhichprobablypreventedfulminantmediastinitisandnecrosis.Onefurthercase,an88-year-oldfemale,couldnotbeconsideredforthoracotomy,anddiedofsepsis.A28-year-oldmalepresentedaftertheclassicalhistory,butonpresentationtothisunitafter48hourshehadsubcutaneousemphysema,amildtachcardia,andasmallpleuraleffusion.Agastrografinfollowedbyabariumstudyfailedtoshowevidenceofleakage,andhewastreatedsimplywithantibiotics,theassumptionbeingthatthetearhadsealed.Inotherseries,patientshavedemonstratedadequatedrainageofamediastinalabscessbackintotheoesophagusandhavesettledwithoutanyintervention.

Thetherapeuticroleofendoscopicapproaches,eitherthejudicioususeoffibringlueorofoesophagealstentshasgainedincreasingprominence.13,15,16Fibringluesealantorcyanoacrylatehasbeensuccessfullyreportedforsmallleaksandfistulae.Wide-diameterself-expandingcoveredmetallicstentsmaybeusedforperforations,andfourpatientsinthisserieshadacoveredoesophagealstentinsertedearlyfollowingpresentation.Theexperienceisunsatisfactoryandcautionary;inthreepatientsthestentsuccessfullycontrolledthefistulaandmayhaveaidedrecovery,twoofthesehoweverwhohadbeendischargedfromhospitalsuccumbedtomassiveuppergastrointestinalbleedingfromanoesophageal-aorticfistula.Inonefurtherpatientthestentdidnotcontrolthefistula,andduetothisandprogressivesepsisthatpatientunderwentanoesophagectomy.Anotherpatientwhohadastentinsertedonday30followingadmissionhadpersistentleakageandunderwentanoesophagectomy.

Inconclusion,earlysurgicalinterventionineightpatientswasassociatedwithgoodoutcomes

R PRichaRd et al


CASE

REP

ORT


CASE

REP

ORT

comparedwithmodernbenchmarks,onedeathinapatientalreadymanifestingmultipleorganfailure.Aconservativeapproachisjustifiedinthosedelayedinpresentationwhoareperhapsself-selectedtosurvive.Theexperienceoftheunittodatecautionsagainsttheliberaluseofendoprosthesesinthiscohort,and,ifused,theyshouldberemovedifpossiblewithinthreetofourmonths.Consistentwithotherseries,oesophagectomycanbeconsideredinselectedcases.

REfEREnCES1. BoerhaaveH.Atrocis,NecDescriptiPruis,Morbid

Historia:SecundumMedicaeArtisLegesConscripta.LugduniBatavorum:Boutesteniana,1724.

2. BarrettNR.Spontaneousperforationoftheoesophagus;reviewoftheliteratureandreportof3newcases.Thorax 1946;1:48-56.

3. LawrenceDR,OhriSK,MoxonREetal.PrimaryoesophagealrepairforBoerhaave’ssyndrome.Ann Thorac Surg1999;67:818–20

4. BufkinBL,MillerJI,MansourKA.Esophagealperforation:Emphasisonmanagement.Ann Thorac Surg1996;61:1447–52

5. PateJW,WalkerWA,ColeFHJretal.SpontaneousRuptureoftheoesophagus:a30yearexperience.Ann Thoracic Surg1989;47:689–692

6. BrauerRB,Liebermann-MeffertD,SteinHJ,BartelsH,SiewertJR.Boerhaave’sSyndrome:analysisoftheliteratureandreportof18newcases.Dis Esophagus1997;10:64–8

7. ShenfineJ,DresnerSM,VishwanathY,HayesN,GriffinSM.Managementofspontaneousruptureoftheoesophagus.Br J Surg2000;87:362-73.

8. MacklerSA.Spontaneousruptureoftheesophagus:Anexperimentalandclinicalstudy.Surg Gynaec Obstet1952;92:345-50.

9. JougonJ,McBrideT,DelcambreFetal.PrimaryoesophagealrepairforBoerhaave’ssyndromewhateverthefreeintervalbetweenperforationandtreatment.Eur J Cardithorac Surg2004;25:475–479

10. AltorjayA,KissJ,VorosA.Theroleofesophagectomyinthemanagementofoesophagealperforations.Ann Thorac Surg1998;65:1433–6

11. OjaimaH,KuwanoH,SasakiS,FujisawaT,IshibashiY.SuccessfullatemanagementofspontaneousesophagealruptureusingT-tubemediastinoabdominaldrainage.Am J Surg2001;182:192-6

12. LarrieuAJ,KiefflerR.Boerhaave’ssyndrome:reportofacasetreatednon-operatively.Ann Surg1976;183:401-408

13. TsunodaS,ShimadaY,WatanabeG,NakauM,ImamuraM.Coveredmetallicstentofapatientwithspontaneousruptureoftheesophagus.Dis Esophagus2001;14:254-7.

14. LawrenceDR,OhriSK,MoxonREetal.PrimaryoesophagealrepairforBoerhaave’ssyndrome.Ann Thorac Surg1999;67:818–20

15. HarriesK,MasoudA,BrownTH,RichardsDG.Endoscopicplacementoffibrinsealantasatreatmentforlong-standingBoerhaave’sfistula.Dis Esophagus2004;17:348-50

16. DaviesAP,VaughanR.ExpandingmeshstentintheemergencytreatmentofBoerhaave’ssyndrome.Ann Thorac Surg1999;67:1482-3

Correspondenceto:ProfessorJohnVReynolds,DeptofSurgery,TrinityCentre,StJames’sHospital,Dublin8Tel:014162500;Fax:01-4546534;e-mail:[email protected]


CASE REPORTAn80-year-oldfemalewasreferredtoourunitwitha6.3cmAAAandcoexistentHSK.Shepresentedtoadistrictgeneralhospitalwithsixmonthsoflowerbackache,worseningovertherecentfewweeks.Co-morbiditiesincludedmorbidobesity,NIDDM,hyperlipidemiaandmoderaterenalimpairment(Urea:14mmol/l;Creatinine:175µmol/l)secondarytohypertensivenephropathy(BP:140/90mmHg)andmildleftrenalarterystenosis.

Acomputedtomography(CT)angiogramrevealedcomplexrenalarteryanatomy;juxtapositionoftheoriginsofthemainrenalarteries;mildleftrenalarterystenosisandatleasttwoaccessoryrenalarteries(2mmdiametereach)supplyingthebridgeoftheHSK.Theinferiormesentericarteryoriginatedfromtheaortaimmediatelysuperiortotherenalbridge.Beyondtheaneurysmalaorta,thedistalaorticsegmentwasstenotic(11mm),aswereboththeexternaliliacarteries(6mm).CTscandidnotindicateanyobviousAAArupture.However,duetopressuresymptomsarisingfromtheaneurysmsacshewaspreparedforaneurysmrepair.

Pulmonaryfunctiontestingandstressechocardiographyrevealedsevererespiratoryandcardiaccompromise.ShewasconsideredhighriskforopenAAArepairandwasthereforeofferedanEVAR.

PROCEDUREPreliminary,thepatientunderwentballoonangioplastyofherdistalaortafrom11mmto18mm.Similarly,bothexternaliliacarterieswereangioplastiedfrom6mmto9mm.However,thedistalaorticsegmentwasstillnotwideenoughtoaccommodateabifurcatedendograft;thereforeanaorto-uni-iliacdevicewasdeployedinstead.

Simultaneouson-tableaortographyconfirmedthatthemajorsourceofrenalperfusionwasfromtheanatomicalrenalarteriesandthattherewouldbelowchanceofcompromiseafteranEVAR.

Followingthisprimaryprocedure,theadviceofareno-vascularphysicianwassoughtinordertoprovidethebestrenalprotectivestrategy.Sodiumbicarbonate wasthereforeadministeredbeforeandaftertheEVAR.Shewasalsogivenasalineinfusionfrommidnight(6hourlyperlitre)forplentifulrenalhydration.Therenovascularteamdidnotrecommendleftrenalarteryangioplastyasstenosiswasmild,thoughtheconcludingdecisionwasdeferreduntilpostoperativerenalfunctionassessment.(Indeed,postoperativelythepatient’srenalprofileremainedunchangedandherbloodpressurewascontrolled,sorenalarteryangioplastywasnotnecessary).

EndovascularaneurysmrepairinapatientwithahorseshoekidneyandimpairedrenalfunctionABSTRACTBackground Wepresentthecaseofan80-year-oldfemalewithathree-weekhistoryof

lowerbackacheandapulsatilemassinherepigastrium.Method UsingCTangiogram,thepresenceofaninfra-renalabdominalaorticaneurysm

(AAA)andco-existenthorseshoekidneywasconfirmed.Results Duetoassociatedmultipleco-morbiditiesofhypertension,moderaterenalfailure,

COPDandmorbidobesity,sheunderwentendovascularaneurysmrepair(EVAR).Conclusion Thepresenceofahorseshoekidney(HSK)posessignificantdifficultyduring

openrepairofanAAA,assatisfactoryanatomicalexposureoftheaortamayprovepotentiallyhazardous.Thisproblemmaybecircumventedinsuitablecasesbyendovascularrepair,althoughtreatmentisnotriskfree.EVARisusuallyonlyrecommendedifrenalfunctionisnormal.WereportacaseofAAAcoexistentwithanHSKandmoderaterenalimpairment,whichwastreatedsuccessfullyusinganEVARtechnique,andwithnosubsequentrenalcomplication.

MSSajid,NAhmed,JTibbals*,GHamiltonDeptofVascularSurgery&Radiology*,RoyalFreeHospital,London,UK

EndOvASCulAR AnEuRySM REPAiR in A PATiEnT wiTh A hORSEShOE kidnEy And iMPAiREd REnAl funCTiOn


CASE

REP

ORT


CASE

REP

ORT

ENDOVASCULAR ANEURYSM REPAIR IN A PATIENT wITH A HORSESHOE kIDNEY AND IMPAIRED RENAL FUNCTION

horseshoe kidney is primarily via anatomical arteries, endovascular repair may be the preferred method provided the aneurysm is morphologically suitable and renal function is normal5. However, in our patient, renal function was pre-operatively moderately deranged and post-operatively it did not deteriorate. Based on our experience of this case, it has been shown that EVAR can be offered to such patients with mild to moderate renal impairment and is probably the best modality for treatment of an AAA with a coexistent horseshoe kidney.

REFERENCES1. Greenberg PK, Chuter TA, Lawrence-brown M, Haulon S,

Nolte L. Analysis of renal function after aneurysm repair with Zennith AAA Endovascular graft in contrast to open repair. J Vasc Surg. 2004 Jun; 6:1219-28

2. Noguchi M, Hirashima S, Eishi K et al. Surgical treatment of AAA associated with horseshoe kidney and coagulopathy J Cardiovasc Surg. (Torino).2004 Oct; 5: 505-9.

3. Davidovic LB, Kostic DM, Jakovljevic NS et al. Abdominal aortic surgery and horseshoe kidney. Ann Vasc Surg. 2004 Nov; 6: 725-8

4. Jackson RW, Fay DM, Wyatt MG, Rose JD. The renal impact of aortic stent-grafting in patients with horseshoe kidney. Cardiovasc Intervent Radiol. 2004 Nov-Dec; 6: 632-6

5. Ruppert V, Umscheid T, Reiger J et al. Endovascular aneurysm repair: Treatment of choice for AAA coincident with horseshoe kidney? J Vasc Surg. 2004 Aug; 2: 367-70.

Correspondence to: Mr MS Sajid, Senior Clinical Fellow Vascular Surgery, Dept of Vascular Surgery, Royal Free Hospital, Hampstead, London NW3 2QG; Email: [email protected]); Tel. 0044 789 166 7608

MS SaJiD et al

Ten days after the initial procedure the EVAR was possible. Under combined spinal and epidural anaesthesia both femoral arteries were exposed, controlled and cannulated in the standard manner. Under image intensification, a single-limbed, 20mm aorto-uni-iliac stent graft was deployed (Zenith suprarenal fixation device) with preservation of the anatomical renal arteries. The procedure was completed with endovascular deployment of a contralateral, 10mm common iliac artery occluder (Zenith) and femoro-femoral bypass, using a synthetic graft (Supported Vascutek).

Her postoperative recovery was quick, uneventful and unremarkable, apart from left groin wound dehiscence and subsequent superficial wound infection. Her renal function (Urea: 11-15mmol/l; Creatinine 150-180 µmol/l) and blood pressure (140-150mmHg systolic; 85-95mmHg diastolic) remained unchanged. Eight weeks after the procedure, there was no evidence of endoleak or renal infarction on CT angiogram and duplex scan. Her renal function was equal to her pre-operative base line.

DisCUssiOn There are very few cases in the literature describing repair of an AAA coexistent with a horseshoe kidney and impaired renal function. Detailed pre-operative assessment of renal anatomy and its vasculature by various imaging modalities is crucial. Anomalous renal arteries are found in 67 % of patients of abdominal aortic aneurysm coexistent with horseshoe kidney preoperatively. In most cases anatomical renal arteries predominantly supply renal tissue. Accessory renal arteries are always present but are minor and can be sacrificed without developing significant renal infarction or renal impairment. After careful preoperative assessment and optimization, patients with mild to moderate renal impairment can be offered endovascular repair of an AAA coexistent with a horseshoe kidney. Aortic stent-grafting is a good alternative to open surgery for such patients but it is not without difficulties. In all patients the aneurysm can successfully be excluded with occlusion of between one and four anomalous renal arteries without significant renal impairment or renovascular hypertension4. EVAR provides several potential advantages. It is a minimally invasive procedure and reduces respiratory and cardiovascular stress compared to open surgery. It can be performed under spinal/epidural anaesthesia which assists rapid recovery, decreasing requirements for major anaesthesia.

Renal dysfunction occurs in both groups of patients regardless of type of repair (open or endovascular with supra-renal fixation)1. Management of our patient and a successful outcome provides evidence to support treatment of an AAA with coexistent horseshoe kidney and renal impairment, using the endovascular method.

Postoperative causes of renal impairment are probably multifactorial. The observed dysfunction occurs in a small number of patients (12 % at 12 months); the effect in the endovascular group appears to be transient5. The initial dysfunction, apparent in both groups over 12 months of follow-up, stabilizes or improves at 12 to 24 months. This may be due to the development of intra-renal collateral blood flow5.

In the presence of multiple comorbidities as in our patient, outcome after open repair is usually much compromised. If the blood supply to the

FIGURE 2 —DSA DEPICTINGa & b: ANATOMICAL RENAL ARTERIESc: ABDOMINAL AORTIC ANEURYSMd: LOwER AORTIC STENOSIS

a

bc

FIGURE 2 —DSA DEPICTINGa & b: ANATOMICAL RENAL ARTERIESc: ABDOMINAL AORTIC ANEURYSMd: LOwER AORTIC STENOSIS

d

ab

c


CASE

REP

ORT


CASE

REP

ORT

intrODUCtiOnFrank haematuria is a common symptom of carcinoma of the urinary tract; 85% of patients with bladder cancer present with intermittent haematuria. It is also associated with benign conditions such as urinary tract infections and calculi. We report a rare cause of recurrent frank haematuria.

Case rePOrtA 51-year-old male presented with recurrent haematuria associated with minimal exercise. He also complained of urinary frequency, poor stream and penile pain. Investigations included ultrasound (normal), IVP (normal), MSU and cystoscopy. MSU revealed leucocytes but no organisms were isolated. At cystoscopy two small areas of inflammation with overlying calcification were demonstrated (Figure 1). Subsequent biopsy and histological examination revealed a chronic inflammatory infiltrate with areas of calcification, and no evidence of malakoplakia (Figure 2).

However the haematuria persisted and repeat cystoscopy showed extensive ulceration of the bladder midline. Transurethral resection of the lesions was performed. Histology again confirmed the presence of chronic inflammation and calcification. In spite of complete resection his symptoms recurred and the decision was taken to carry out a partial cystectomy. Histology of the gross specimen confirmed a diagnosis of encrusted cystitis. One year post- surgery he is asymptomatic and there is no evidence of recurrence on cystoscopy.

DisCUssiOnEncrusted cystitis is a rare inflammatory condition characterised by the formation of calcified plaques adherent to or embedded in the bladder mucosa.1

It is associated with infection by urea-splitting organisms and thereby alkaline urine.2 Ammonium produced from the breakdown of urea damages the glycosaminoglycan layer of the bladder mucosa. This facilitates bacterial adherence, tissue inflammation, production of an organic matrix, and crystal matrix interaction. Gram positive and negative bacteria along with mycoplasma and yeasts produce urease. The most common organism associated with encrusted cystitis is corynebacterium urealyticum. However it can also develop in the absence of these states.

Other predisposing factors for encrusted cystitis are long-term indwelling catheterisation and previous renal transplantation.

Symptoms include chronic urinary discomfort and macroscopic haematuria.

Treatment is with antibiotics, local diathermy and/ or surgical resection of the encrusted lesions.

Encrusted cystitis — An unusual cause of recurrent frank haematuriaaBstraCtBackground Encrusted cystitis associated with malakoplakia is described in the literature

as a rare condition characterized by the formation of calcified plaques adherent to or embedded in the bladder mucosa.

Aims We present the case of a 51-year-old male, presenting with recurrent haematuria associated with minimal exercise. Despite normal ultrasound, IVP, two calcified lesions on his bladder were found on cystoscopy. Conservative management was ineffective thereby necessitating a curative partial cystectomy.

Conclusion We would like to illustrate/discuss this rare but problematic condition with gross and microscopic pictures.

ENCRUSTED CYSTITIS — AN UNUSUAL CAUSE OF RECURRENT FRANk HAEMATURIA

O O’Sullivan, O Clyne, J DrummDept of Urology, Mid-Western Regional Hospital, Limerick

1(a)

o o’SulliVan et al

REFERENCES1. Aguado JM, Morales JM, Salto E et al. Encrusted Pyelitis

and Cystitis by Corynebacterium Urealyticum (CDC group D2): a New and Threatening Complication Following Renal Transplant. Transplantation 1993;56:617-22.

2. Huguet Perez J, Salvador Bayarri J, Vicente Rodriguez J. Encrusted Cystitis. Is it Always Alkaline? [Spanish]. Arch Esp Urol 1999;52:157-64.

Correspondence to: Orfhlaith O’Sullivan, Annapurna, Lower Park, Corbally, LimerickTel: 087 9272894; E-mail:[email protected]

Figure 1a —ANTERIOR BLADDER wALL

Figure 1b —POSTERIOR BLADDER wALL

1(b)

2(a) 2(b) Figure 2a —MICROSCOPY OF BIOPSY

Figure 2b —MICROSCOPY OF GROSS SPECIMEN


CASE

REP

ORT


CASE

REP

ORT

intrODUCtiOnComminuted fractures of the spine are an uncommon group of injuries that most frequently occur following high-speed collisions. Thoracic spinal fractures occur whenever forces exceed the strength and stability of the spine. They are a major cause of morbidity, particularly with regard to neurological disability and carry huge social and economic implications. While many options are available, scant evidence exists supporting operative over non-operative treatment.

In this case report, we describe a patient who suffered a comminuted fracture of the sixth, seventh and eighth thoracic vertebral bodies without neurological deficit following a road traffic accident with a favourable outcome.

Case rePOrtA 17-year-old girl was brought by ambulance to Emergency Department following a road traffic accident in which she was a restrained front seat passenger. She was cut free from the car by the emergency services. She complained of mid-thoracic back pain and chest pain. She was tachycardic and dyspnoeic. Clinical assessment confirmed that she was tender over her left chest and over her mid-thoracic spine. There was no distal neurological deficit. Her Glasgow Coma Score was 14 out of 15.

Chest x-ray confirmed three broken ribs, a pulmonary contusion, a widened mediastinum and

a mid thoracic fracture. CT of Thorax confirmed these findings and also demonstrated a small haemothorax. CT of Thoracic spine showed comminuted fractures of the bodies of the sixth, seventh and eighth thoracic vertebrae, without violation of the spinal canal or impingement of the spinal cord. Aortography outruled associated thoracic aortic injury. CT brain showed an undisplaced occipital fracture but no intra-cranial injury.

Following resuscitation, she was nursed overnight on a Turning Frame. Thereafter, she was treated surgically. Under general anaesthetic, she was positioned prone on a Wilson frame. Instrumented reduction and internal fixation of the fracture was carried out from the fourth thoracic vertebra to the tenth, using AO pedicle screws and rod fixation through a posterior approach. She had an uneventful post-operative period, being maintained on bed rest until the third day. She was mobilised in a spinal orthosis under physiotherapy supervision. She was discharged on the tenth day mobilising with crutches. At four-month follow-up, she was functioning normally and had no neurological deficit.

DisCUssiOnThis case demonstrates a very dramatic injury with good functional outcome. In managing such an injury, it is important to consider the mechanism of injury and to classify the injury so as to determine appropriate management. We highlight difficulties in selecting treatment for unstable thoracolumbar fractures and discuss pertinent economic issues.

Comminuted fracture of the thoracic spineaBstraCtBackground Road deaths fell initially after the introduction of the penalty points but

despite this, the rate of spinal injuries remained unchanged. Aims We report a patient with a dramatic spinal injury, though without neurological

deficit. We discuss the classification, management and economic impact of these injuries.

Methods We describe the management of a patient with a comminuted thoracic spinal fracture without neurological injury. We conducted a literature review with regard to the availability of literature of the management of these injuries.

Results This 17-year-old female was managed surgically and had a good functional outcome. There is no clear consensus in the published literature on the management of these injuries.

Conclusions Comminuted thoracic spinal factures are potentially devastating. Such a patient presents challenges in determining the appropriate treatment.

JP Cashman1, FL Carty2, M Ryan2, K Mahalingham1

Dept of Trauma and Orthopaedics1, Dept of Radiology2, Cork University Hospital, Wilton, Cork

COMMINUTED FRACTURE OF THE THORACIC SPINE

The introduction of the penalty points system in Ireland in 2001 saw an initial reduction in overall trauma figures, although there was no change in the rate of spinal injuries.1 Spinal injury occurs in approximately 11% of motorcyclists and 14% of car occupants involved in motor vehicle accidents. The majority of victims are young, with mean ages of 30.2 years for motorcyclists and 37.8 years for car occupants, and there is a male predominance (motorcycle 88.9%, car 60.6%).2

Formal classification allows for assessment of fracture pattern stability and the need for operative intervention. Denis proposed that for assessment of stability, the spine be considered as three columns - anterior, middle and posterior.3 The Denis classification, while relatively simple, is incomplete and incomprehensive. The AO group use a comprehensive classification system which includes three types of fractures with 55 different subtypes. While more complete, it is cumbersome for daily practice. Both systems have been found to have significant inter-observer variability in their use.4 This raises difficulties in determining which system to apply, and importantly, which fractures to operate on.

There is a paucity of evidence from high quality randomised trials on the relative effectiveness of operative and conservative treatment of unstable traumatic thoracolumbar fractures. Retrospective studies have found favourable radiological outcomes following operative intervention with better resorption of retropulsed fractures in that group.5

However, it has been found that surgical intervention carries a higher risk of wound infection (8% vs 0%), while nonoperative treatment results in longer hospital stay (24 days).6 With regard to thoracolumbar spinal fractures without neurological injury, some studies indicate favourable short-term pain control with surgery, though there is no appreciable difference in long-term outcome in either group.7

There is considerable range in hospital costs incurred depending on how such an injury is treated. In one review of the economic impact of these, stable fractures without neurological deficits which were treated nonoperatively inferred a cost of €5,100. Unstable fractures without neurological deficits which were treated nonoperatively had an average cost of €12,500, while those treated operatively had an average cost of €19,700. Unstable fractures with neurological deficits were usually treated surgically at an average cost of €31,900.8 These factors

Jp caShMan et al

Figure 1—3D RECONSTRUCTION OF CT THORACIC SPINE SHOwING A COMMINUTED FRACTURE OF THE 6TH, 7TH AND 8TH THORACIC VERTEBRAE

Figure 3—AXIAL SECTION OF CT THORACIC SPINE SHOwING DEGREE OF COMMINUTION OF THE ANTERIOR AND MIDDLE COLUMNS, AND A PATENT SPINAL CANAL

Figure 2—SAGITTAL SECTION OF CT OF THORACIC SPINE SHOwING THE EXTENT OF THE TRANSLATION OF THE FRAGMENTS, THOUGH wITHOUT IMPINGING ON THE SPINAL CANAL


CASE

REP

ORT


CASE

REP

ORT

contributeinselectingtreatmentwithincreasingfrequencyinamoreconsumerdrivenhealthsystem.

CONCLUSIONThiscasehighlightsdifficultieswithregardtooptimaltreatmentofcomminutedspinalfractureswithoutneurologicaldeficitandillustratesthehigheconomicandpersonalcostassociatedwiththistypeofinjury.Theoptimalmanagementoftraumaticthoracicinjuriesremainsacontroversialissue.Adequateassessmentisparamountinselectingtheappropriatemanagement.Thereisnoclearconsensusonthetreatmentofunstablethoracicfractureswithoutneurologicalinjury.Intheabsenceofconcomitantneurologicalinjury,itispossibletoachieveagoodoutcomeregardlessofhowthefractureistreated.

REFERENCES1. LenehanB,StreetJ,BarryK,MullanGImmediateimpact

of‘penaltypointslegislation’onacutehospitaltraumaservices.Injury2005;36(8):912-6.

2. RobertsonA,BranfootT,BarlowIF,GiannoudisPV. Spinalinjurypatternsresultingfromcarandmotorcycleaccidents.Spine.2002;27(24):2825-30.

3. DenisFThethreecolumnspineanditssignificanceintheclassificationofacutethoracolumbarspinalinjuries.Spine.1983;8(8):817-31.

4. WoodKB,KhannaG,VaccaroAR,ArnoldPM,HarrisMB,MehbodAAAssessmentoftwothoracolumbarfractureclassificationsystemsasusedbymultiplesurgeons. J Bone Joint Surg Am.2005;87(7):1423-9.

5. GotzenL,PuplatD,JungeA.Indications,techniqueandresultsofmonosegmentaldorsalspondylodesisinwedgecompressionfractures(gradeII)ofthethoracolumbarspine.Unfallchirurg199295:445–454

6. RechtineGR,CahillD,ChrinAMTreatmentofthoracolumbartrauma:comparisonofcomplicationsofoperativeversusnonoperativetreatment.J Spinal Disord199912:406–409

7. ShenWJ,LiuTJ,ShenYSNonoperativetreatmentversusposteriorfixationforthoracolumbarjunctionburstfractureswithoutneurologicdeficit.Spine200126:1038–1045

8. VanderRoerN,deBruyneMC,BakkerFC,vanTulderMW,BoersM.Directmedicalcostsoftraumaticthoracolumbarspinefractures.Acta Orthop.2005;76(5):662-6.

Correspondenceto:JamesCashman,DeptofOrthopaedics,CorkUniversityHospital,Wilton,Cork;Tel:(021)4546400;Email:[email protected]

COmmiNuTEd FRACTuRE OF ThE ThORACiC SPiNE

INTRODUCTIONSmallbowelvolvulusisaconditioninwhichthereistorsionofthesmallbowelabouttheaxisofitsmesentery.1,2Itisrarelydescribedinadultsalthoughiswellrecognisedinchildren.2Thisreportdescribesacaseofsmallbowelvolvulusinanadultsecondarytoalipomainthesmallbowelmesentery.

CASE REPORTA40-year-oldmanpresentedwithaone-weekhistoryofconstipationassociatedwithcolickyabdominalpainandvomiting.Hecontinuedtopassflatus.Therewasnopastmedicalhistoryofnote.Onexaminationhewasclinicallydehydrated.Hisabdomenwasdistendedbutnon-tenderwithincreasedbowelsoundsandnoevidenceofanyherniae.WhitecellcountwasnormalbutaC-reactiveproteinlevelwasraisedat67.AplainabdominalX-raywastakenandrevealedmarkedlydistendedsmallbowelloopsinkeepingwithhighgrademechanicalsmallbowelobstruction.

Acomputedtomography(CT)scanwascarriedouttotrytoelucidatethecauseofthesmallbowelobstruction.ThissimplyconfirmedtheX-rayfindingsanddidnotshowanyobviouscause.Ashissymptomswerenotimprovinghewastakenforlaparotomy.

Atlaparotomy,therewasrotationofmidsmallbowelaboutamesentericlipomaresultinginobstruction.Thelipomameasured16x8x5cmandweighed302g.Elevencentimetresofsmallbowelwereresectedalongwiththelipoma.Histologyconfirmedoedemaofthesmallbowelconsistentwithsomedegreeoftorsionandmatureadiposetissueinkeepingwithalipoma.

RetrospectivereviewoftheCTscanrevealedthecharacteristic‘whirlpool’appearanceofthemesentericvesselsassociatedwithsmallbowelvolvulus(Figure1)and10mmbelowthis,alowdensitylesioninkeepingwithalipomawasshowntobeleadingintothevolvulus(Figures2and3).

Thepatienthadanuneventfulpost-operativerecoveryandwasdischargedhomeonday8post-operatively.

DISCUSSIONSmallbowelvolvulusisrareinadultsandcanbecategorisedaseitherprimary(nopredisposingfactors)orsecondary(predisposinganatomicalabnormalitiespresent).2,3,4Thiscasedescribessecondarysmallbowelvolvuluscausedbyamesentericlipoma.Mesentericlipomasarerare,withlessthan50casesdescribedintheEnglishlanguageliterature.5,6Alipomaisabenigntumourofmatureadiposetissue.7,8Theycanoccuranywhereinthebodybutintraperitoneallipomataareveryrare.5,7,8Anincreasedincidenceoflipomataisassociatedwithobesity,diabetesmellitus,hypercholesterolaemia,familialtendency,traumaandchromosomalabnormalities.7,8

Mesentericlipomataareusuallyasymptomaticbutcancausesymptomssecondarytodistensionoftheabdomen7orsymptomsofsmallbowelobstructionsecondarytovolvulus.5,6,8,9,10

Primarymesenterictumoursaredifficulttodetectandareoftennotdiagnosedpreoperatively.6,8Imagingmodalitiessuchascomputedtomography(CT)ormagneticresonanceimaging(MRI)canbeuseful1,2,6,8showingacharacteristic‘whirlpool’patternofsmallbowelloopsaroundthesuperior

Smallbowelvolvulussecondarytoamesentericlipoma:acasereportandreviewoftheliteratureABSTRACTBackground Smallbowelvolvulusisrareinadultsandoftenhasaprecipitatingfactor.methods Thisreportdescribesacaseofsmallbowelvolvulussecondarytoamesenteric

lipomaandreviewstheliteraturedescribingthiscondition.Conclusion Mesentericlipomaisarareprecipitatingcauseofsmallbowelvolvulusin

adults.Computedtomographyscanningmaybeusefulfordiagnosispreoperatively.Thetreatmentofchoiceissurgerywithcompleteexcisionofthelipoma.

ASMcCoubrey,RLEThompsonDeptofGeneralSurgery,AltnagelvinAreaHospital,Londonderry,NI

SmAll BOwEl vOlvuluS SECONdARy TO A mESENTERiC liPOmA: A CASE REPORT ANd REviEw OF ThE liTERATuRE


CASE

REP

ORT


CASE

REP

ORT

mesentericarteryconsistentwithvolvulus2,3,8ora‘peacock’stail’signduetotorsionofthebowelaroundthemesentericaxis.3CTscanningmayalsobehelpfulinhighlightingpotentialsmallbowelischaemia.3MRIprovidessimilarinformationtoCTscanningbutcandefinetissuecharacteristicswithgreatersensitivity.4Therearenodatatocomparethetwoimagingmodalitiesatpresent.TheradiographicimagingshouldalsoidentifythemesentericlipomaasahomogenousfatattenuationmassonCTscanningorwithahomogenoussignalintensityidenticaltofatonMRI.7,11

Surgeryisthetreatmentofchoicetorelievethevolvulusandexcisethelipoma.1,2,8Completeexcisionofthelipomaisrecommendedduetothepotentialformalignanttransformation(especiallyinlargerlipomata)andlowrecurrencerate(<5%).5,6,7,8

CONCLUSIONSmallbowelvolvulusshouldbeconsideredinthedifferentialdiagnosisofsmallbowelobstructioninadults.CTscanningmaybeusefulinelucidatingthediagnosisandsurgeryisthetreatmentofchoicewithcompleteexcisionofthemesentericlipomarecommended.

REFERENCES1. KatisPG,DiasSM.Volvulus:araretwistonsmall-bowel

obstruction.CMAJ2004;171(7):728.

2. HuangJC,ShinJS,HuangYTetal.Smallbowelvolvulusamongadults.JGastroenterolHepatol2005;20:1906-1912.

3. IwuagwuO,DeansGT.Smallbowelvolvulus:areview.JRCollSurgEdinb1999;44:150-155.

4. ChouCK,Tsai,TC.Smallbowelvolvulus.AbdominalImaging1995;20:431-435.

5. SignerRD,BregmanD,KlausnerS.Giantlipomaofthemesentery:reportofanunusualcaseandreviewoftheliterature.AmSurg1976;42(8):595-597.

6. WongHI,ChenCY,LiuGC.Primarymesentericlipomacausingclosedloopbowelobstruction:acasereport.KaohsiungJMedSci2005;21(3):138-141.

7. IlhanH,TokarB,IsiksoyS,KokuN,PasaogluO.Giantmesentericlipoma.JournalPediatrSurg1999;34(4):639-640.

8. SheenAJ,DrakeI,GeorgePP.Asmallbowelvolvuluscausedbyamesentericlipoma:reportofacase.SurgToday2003;33(8):617-619.

9. TahlanRN,GargP,BishnoiPKSinglaSL.Mesentericlipoma:anunusualcauseofsmallintestinalvolvulus.IndianJGastroenterol1997;16(4):159.

10. ShahSA,SaaiqM,WaqarSHRashidR,JamalS.Mesentericlipomaleadingtosmallgutstrangulationandshortgutsyndrome.JCollPhysiciansSurgPak2005;15(6):371-372.

AS Mccoubrey And rLe ThoMpSon

11. PereiraJM,SirlinCB,PintoPS,CasolaG.CTandMRimagingofextrahepaticfattymassesoftheabdomenandpelvis:techniques,diagnosis,differentialdiagnosisandpitfalls.Radiographics2005;25:69-85.

Correspondenceto:MissAlisonMcCoubrey,DeptofGeneralSurgery,AltnagelvinAreaHospital,GlenshaneRoad,LondonderryBT476SB,UKTel:[email protected]

Figure 1—CT SCAN SHOWING ‘WHIRLPOOL’ APPEARANCE OF MESENTERY (ARROW)

Figure 2—LIPOMA (L) FOUND BELOW APEX OF VOLVULUS

Figure 3—CORONAL IMAGE SHOWING LIPOMA (ARROW) LEADING INTO VOLVULUS

1

2

3

DearEditor

ExtensortendonruptureiscommonlyseenaftertraumaorinassociationwithdiseaseslikerheumatoidarthritisorKienbock’sdisease.HerewedescribeacaseofextensortendonrupturecausedwhileplayingUlileannpipe(traditionalIrishmusicalpipes).Thisunusualmechanismofinjuryisnotbeenreportedbeforeintheliterature.

A34-year-oldmaleright-handedcarpenterbyoccupation,presentedtoinjurycliniccomplainingofinabilitytoextendhisleftmiddlefinger,swellingandmildpainonthedorsalaspectofhandforthreeweeks.HissymptomsstartedsuddenlyafterplayingtheUileannpipesatamusicfestival.Heexperienceda‘popping’sensationonthedorsumofthehandandwasunabletocontinueplaying.Onexaminationflexiondeformityoftheleftmiddlefingerwaspresentwithswellingonthedorsalaspectofhandandunabletoextendthefinger.Ultrasoundofhand(Figures1and2)showedhaematomaatdorsalaspectofhandandretractededgesofextensortendon.Hehadexplorationofthedorsalaspectofhandwhichshowedacompletetearofextensortendondistaltowrist,andedgeswereretractedleavingagapofmorethan4cm.AgraftwasobtainedfrompalmarislongusandthetendonwassuturedwithEthiboundsutureandwristwasimmobilizedincastforaperiodoffourweeksandthenaphysiotherapyandrehabilitationprogramwasstarted.Threemonthsaftersurgery,hecouldfullyextendthemiddlefingerandreturnedtohisoccupationasacarpenter.Heisalsorecentlyreturnedtoplayingmusic.

Extensortendoninjuriesaredividedintoeightdifferentzones;I-InjuryatDIPJ,II-InjuryatM.P,III-InjuryatPIPJ,IV-InjuryatPP,V-InjuryatMPJ,VI-InjuryatM.C,VII-InjuryatDistalRetinaculum,VIII-Injuryatdistalforearm/wrist.ThisinjurywasinzoneVI,whichhasabetterprognosisbecauseoffollowingreasons:

1 Unlikelyitisassociatedwithjointinjuries.2 Decreasetendonareainthiszonelessensthe

potentialofadhesionformation.3 Increasedsubcutaneoustissueinthiszonelessen

thepotentialforadhesion.4 Greatertendonexcursioninthiszone.5 Complextendonimbalanceislesslikelytooccurin

thiszone.Therearemanydifferentoptionsoftreatmentwhichhavebeendescribedintheliterature.

ForthiscaseweusedPalmarislongustendongraft,whichachievedfavorableresultandthepatientwasbacktoworkwithin12weeksofinjury.Inconclusion,ultrasoundhasabenefitofidentifyingtheinjuryindelayedpresentationforthepre-operativeplanning.Earlymobilizationwillhelpinachievingagoodoutcome.

REFERENCES1. EarlZ.Earlydynamicsplintingforextensortendon

injuries.J Hand Surg1089;14A,(1);72-6.

2. MuraseT.ExtensortendonsruptureduetoKienbock’sdisease.J Hand Surg-Brit 1997;22(5):597-8.

3. NewportM.Longtermresultsofextensortendonrepair.J Hand Surg.199015A,(6):961-6.

4.WadaT.ClosedruptureofafingerextensorfollowingtheSauve–Kapandjiprocedure:casereport.J Hand SurgAm.1997;22(4):705-7.

5. W.Bradford;ExtensorTendon;Anatomy,InjuryandReconstruction;J Plast Reconstr Surg;2000;106-4:1592-1603.

Correspondenceto:SMAli,10LeafAvenue,HamptonHargate,Peterborough,PC78EF,UKE-mail:[email protected]

ExtensortendonruptureoffingerwhileplayingUileannpipe

SMAli,DO’FarrellDeptofOrthopaedics1,Mid-WesternRegionalOrthopaedicHospital,Limerick

CORRESPONDENCE

Figure 1 — DORSUM OF LEFT HAND, TRANSVERSE SECTION12mhzultrasoundprobeshowslowechogenicfluid(broadarrow)consistentwithhaemorrhagicexudatessurroundsproximalextensortendon(narrowarrow).

Figure 2 — DORSUM OF LEFT HAND, LONGITUDINAL SECTION demonstratingrupturedproximaloriginofthirdextensortendon(arrow).

1

2


CorrespondenCe

Corr

espo

nd

enCe


Boo

K r

eVIe

W

BooK reVIeW

UltravioletphototherapyrisksinIrelandDearEditor

ThereweretwoarticlesofdermatologicalinterestinyourJune2006issue.O’ConnorandO’Hare’sanalysisofultravioletphototherapyrisksinIrelandwasmostinformative.Couldtheyextendtheirfindingstotherisksofsunparloursinthiscountry?HazardsassociatedwiththeseemporiahavebeenfoundinGreatBritainandtheUnitedStates.

ThearticleonVitaminDintakeanditsimportanceinrespectofsunscreenusagewasalsonoteworthy.1,2ObviouslytheamountofVitaminDinthedietwillvaryconsiderablyandproductionbysunlightwilldependonpigmentation,durationofexposure,sunprotectionetc.For

example,howmuchVitaminDisproducedintheskinofamiddle-agedpatientwitha‘normal’dietandhowmuchisobtainedfromfood?LikealldermatologistsIhavebeenproselytisingforsunscreensforyears.Aremypatientstoexchangericketsforskincancer?

referenCes1. MarksR,FoleyPA,JolleyDetal.Theeffectofregular

sunscreenuseonVitaminDlevelsinanAustralianpopulation.ResultsofarandomisedcontrolledtrialArch Derm19951314

2. GlerupH,MikkelsenKl,PoulsenLetal CommonlyrecommendeddailyintakeofVitaminDisnotsufficientifsunlightexposureislimitedJ Internal Med2000247;260

Repetitivetranscranialmagneticstimulation:Apsychiatrictreatmentofthefuture?

CDupontConsultantDermatologist18MerlynRoad,Dublin

HO’Connell1,RGoggins2PGDoyle3

DeptofOldAgePsychiatry,StCamillus’Hospital,1Limerick;St.Senan’sHospital2,Enniscorthy,Co.Wexford;TevereDayHospital,3Limerick

DearEditor

Repetitivetranscranialmagneticstimulation(rTMS)hasreceivedincreasedattentioninrecentyearsasapotentiallysafeandeffectivealternativetoelectroconvulsivetherapy(ECT).Itisanon-invasivemeansofelectricallystimulatingorindeedinhibitingneuronsinthecortexanditcanmodifyneuronalactivitylocallyandatdistantsiteswhendeliveredinaseriesofpulses.Thetypeofexcitabilityonthecortexdependsonthefrequencyofstimulationused.WhilethespecificmodeofactionofrTMSisunclear,ithasbeendemonstratedtohaveeffectsonmetabolisminthefrontallobesleadingtoaconsequentelevationinmood.Painlessmagneticpulsesaredeliveredtothecortexofafullyawakepatient,usingahand-heldstimulatingcoilapplieddirectlytothehead.IncontrasttoECTwhereelectricitygetsdiffusedbybone,highintensitymagneticpulsespassreadilythroughtheskull.Thisintheorymakesitpossibletoattempttofocusmagneticpulsesonparticularbrainstructures.

ThereisanemergingevidencebasethatsuggeststheremaybebenefitsassociatedwithuseofrTMS.However,thenumberandqualityofclinicaltrialsarelowandtherearenoestablishedguidelinesontheoptimalclinicaluseofrTMS.Asystematicreviewandmeta-analysisbyMartinetal1includedonly14smallrandomisedcontrolledtrialsdeemedtobeofsufficientquality.ThesetrialscomparedrTMSwithshaminpatientswithdepression.PooledanalysisshowedaneffectinfavourofrTMScomparedwithshamaftertwoweeksoftreatment,butthiswasnotsignificantatthe2-weekfollow-up.TheauthorsconcludedthatthecurrentevidencebasewasofapoorqualityanddidnotsupportuseofrTMSinthetreatmentofdepression.Furthermore,arecentstudycomparingrTMSandECTinmajordepressionfoundECTtobemoreeffective,especiallyintheshort-term.2

Despitethesefindingshowever,thereareenoughpotentialadvantagesassociatedwithrTMSuseoverECT,andevensomepharmacotherapeuticapproaches,tojustifyfurther

morerigorousresearchinthisarea.Suchadvantagesincludethelackofarequirementforgeneralanaestheticandthemedicalrisksandfinancialcostsofthis,thepotentialforadministrationinanoutpatientsettingandtherecentlydemonstratedlackofcognitiveimpairmentinrTMSpatientswhencomparedtopatientsreceivingECT,despitesimilarclinicalbenefitsfrombothtreatmentsinthatstudy.3Furthermore,TMSmayprovetobebeneficialfortreatmentforotherpsychiatricillness,andfordiagnosticpurposes.

Ideally,futurerTMSresearchshouldinvolvelargenumbersofparticipantsinmultiplecentres,comparabletodrugtrials.Randomisingandblindingofparticipantsshouldbeoptimised.Effortsmustbemadetoaddresstheoptimalintensity,duration,intertrainfrequencyandfrequencyofthemagneticstimulusapplied,andclearguidelinesonpotentialadverseeffectsshouldbeclarified.Questionshavealsobeenraisedaboutthenatureofshamtreatmentandhowthisisbestdelivered;double-blindingofbothpatientandrTMSadministratordoesnotappeartobepossibleatpresent.SoitappearsthatthequestionsaboutrTMScurrentlyoutnumbertheanswers,butageneralviewamongexpertsintheareaisthatfurtherresearchisrequiredbeforerTMScanbediscountedasapsychiatrictreatmentofthefuture.

referenCes1. MartinJLR,BarbanojMJ,SchlaepferTE,etal.(2003)

Repetitivetranscranialmagneticstimulationforthetreatmentofdepression:systematicreviewandmeta-analysis.Brit J Psyc,182:480-91.

2. ErantiS,MoggA,PluckGetal,Arandomised,ControlledTrialWith6-MonthFollow-UpofRepetitiveTranscranialMagneticStimulationandElectroconvulsiveTherapyforSevereDepression.Am J Psych.2007;164(1):73-81.

3. Schulze-RauschenbachSC,HarmsU,SchlaepferTE,etal.(2005)Distinctiveneurocognitiveeffectsofrepetitivetranscranialmagneticstimulationandelectroconvulsivetherapyinmajordepression.Brit J Psyc 186:410-416.

TheMenopause:whatyouneedtoknowMargaretRees,DavidWPurdie,SallyHopePublishedbyRSMPressISBN1-85315-672-8PriceStg£10.95

EVMocanuConsultantGynaecologist,ReproductiveMedicineSpecialist

Thisbookiseasytoreadandcoversthemenopauseinacomprehensivemannerwithoutgoingintomuchmedicaljargon.Thespaceallocatedisproportionatewiththetopic’simportance,thebookcontaining14chaptersin104pages.

Thereisanelegantbalancebetweenscientificevidenceandpatientrelevance,yetsometimestoomuchdetailleavesthenon-medicalreaderoverwhelmedbythecomplexityofthiscondition,itstreatmentandresearchrelatedtoit.Otherwise,theinformationprovidedisuptodateandhelpful.

Theboxesatthebeginningofeachchapterallowforeaseofuseandillustrationsareofhighquality.Thebookcoversverywellissueslikealternativemedicinesandtheirinteractionwithother“orthodox”medications,aswellas,warningonlackofqualitycontrolwithnon-HRTmedicines.Thebibliographyiswellorganised,offeringbothpaperandwebresources.Thenecessarywordlistisverycomprehensive.

Onecriticism,consideringthetargetedaudience,isthegeneralorganisationofthebook.Itwouldhavebeenmoreappropriatetohavethenon-medicalchapters(i.e.lifestyle,alternativestoHRT)beforetheHRTrelatedandmedicalones.Forexample,chapter14containsexcellentadviceonwhatthedoctorintheclinicwouldliketoknow,whichchronologically

staysbeforethevisittothedoctor.Ontheplusside,anexcellentchapteronearlymenopause,veryrelevant,consideringtheincreasingincidenceandlongtermhealthandreproductiveconsequences.

Thisisabookwrittenbyexpertsinthefield,targetingthelaypublic.Ifyouareapatientyouwillfindallaspectsofthemenopauseclearlycovered.Eachpublicandhospitallibraryshouldhaveacopy.Medicalstudentsandjuniordoctorswillfindthisavaluablecompanion.

PublishedbytheRoyalSocietyofMedicinePressLtd.witharetailpriceof£10.95,itisgoodvalueforthetargetedaudience.Itisasmall,light,well-boundpaperbackthatcouldbeeasilyreadanywhere.


BOOk REVIEw

BOO

k R

EVIE

w

Private Practice — In the early twentieth century medical office of Dr. Richard Cabot

Christopher CrennerPublished by Johns Hopkins University Press, Baltimore and London, 2005ISBN: 0-8018-8117, pp. xv + 303Price Stg £32

C S BreathnachSchool of Medicine and Medical Science, NUI UCD

The subtitle tells more about this book than the title. To pursue his analyses of private practice Dr. Crenner has used the extensive collection of records built up by Richard Cabot in his office, after first establishing himself as a haematologist, in Boston from 1897 to 1936. His story as a physician therefore portrays the growth of technological medicine, not to be confused with scientific medicine. The author makes useful contrasts between illness and disease, referrals and consultations, ‘physiological’ versus symptomatic treatment, and frankness versus ‘deception’ in hopeless cases. Cabot vehemently proclaimed the unassailable right of the patient to absolute truth in regard to prognosis, but when Ella, his wife of forty years, was dying from lung cancer he did not tell her, even in her last days, an omission highlighting the foolhardiness of his intransigence.

Notes for an intended biography made in 1936 three years before his death were retained in the cache. They reveal that in 1893, when he was barely a year qualified, he used chloroform to anaesthetise and finally suffocate his eldest brother who had lapsed into recurring bouts of diabetic ketosis. Unrepentant, Cabot’s boundless hubris never wavered; throughout his long career he delighted in scourging the American medical profession, and he was loud in applause of Bernard Shaw’s Doctor’s Dilemma when it was published in the U S A in 1911. Establishing the clinico-pathological conference (CPC) at Massachusetts General Hospital in 1909 was ‘perhaps his most lasting and significant contribution to medicine’ according to Crenner (p. 78); in fact the first series began in 1900 at the behest of a medical student, Walter B Cannon (The use of clinical records in teaching medicine, Bull. Amer. Acad. Med. 1900-02; 5: 203-213).

Not as elegantly written perhaps as the Autocrat Oliver Wendell Holmes would have demanded for the Arch-Brahmin, the monograph captures, if the random walk into anthropology in the closing chapter is ignored, the realities of the transition from patient to client to consumer in medical practice with enviable clarity and concision.

ijms winter 2006 web

Documents