‘takotsubo’ cardiomyopathy in a maintenance hemodialysis patient

7
Therapeutic Apheresis and Dialysis 10(1):94–100, Blackwell Publishing Asia Pty Ltd © 2006 International Society for Apheresis 94 Blackwell Publishing AsiaMelbourne, AustraliaTAPTherapeutic Apheresis and Dialysis1774-99792005 International Society for Apheresis? 2005 101 94100 Case Report Takotsubo Cardiomyopathy and HemodialysisM Fukui et al. Received April 2005; revised June 2005. Address correspondence and reprint requests to Dr. Yasukiyo Mori, Division of Hypertension and Nephrology, Kyoto Prefec- tural University of Medicine, 465 Kajii-cho, Kawaramachi- Hirokoji, Kamigyo, Kyoto 602-8566, Japan. Email: [email protected] ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient Masayoshi Fukui, Yasukiyo Mori, Satoshi Tsujimoto, Kazuya Takehana, Norihiko Sakamoto, Noriko Kishimoto, Takanobu Imada, Hirofumi Maeba, Atsuko Nose, Hideki Yamahara, Yasuaki Kijima, Tetsuya Kitamura, Takanao Ueyama, Sanae Kikuchi, Toshiko Tokoro, Hiroya Masaki, Mitsushige Nishikawa, and Toshiji Iwasaka Department of Medicine II, Kansai Medical University, Osaka, Japan Abstract: An 84-year-old woman undergoing maintenance hemodialysis presented with chest discomfort lasting sev- eral days and electrocardiographic abnormalities. She had stopped smoking 2 weeks earlier and was experiencing irri- tability. Upon admission, electrocardiography showed ST- segment elevation in leads I, II, aVF, and V 2-6 and an abnor- mal Q wave in leads II, III, and aVF. Ultrasound cardiog- raphy showed left ventricular anteroapical akinesia and basal hyperkinesia. The chest discomfort disappeared with- out specific therapy. During hospital days 1–5, the ST- segment elevation gradually improved. Giant negative T waves then developed. The left ventricular asynergy resolved by day 8. Radionuclide imaging with iodine-123- beta-methyl-p-iodophenyl pentadecanoic acid, but not with technetium-99 m-sestamibi, showed an apical defect. Elective coronary angiography showed no stenosis. ‘Takot- subo’ cardiomyopathy was diagnosed. After discharge, the patient continued regular dialysis without cardiac symp- toms. We concluded that endogenously activated sympa- thetic nerve action in hemodialysis patients, especially those under emotional or physical stress, might be a caus- ative factor for Takotsubo cardiomyopathy. Key Words: Chronic kidney disease, Emotional stress, Hemodialysis, Sympathetic nerve activity, Takotsubo cardiomyopathy. Chronic kidney disease (CKD), particularly end- stage renal disease, is associated with increased cardiovascular morbidity and mortality (1). Many factors such as volume overload, hypertension, ane- mia, arteriovenous fistula, and uremic toxins might contribute to the development of cardiovascular events. In addition, hemodialysis patients exhibit sus- tained activation of the sympathetic nervous system, which contributes to increased cardiovascular mor- bidity (2,3). ‘Takotsubo’ cardiomyopathy is a newly described cardiac syndrome characterized by transient left ven- tricular dysfunction with chest discomfort, electro- cardiographic changes, and minimal release of myocardial enzymes, together mimicking acute myo- cardial infarction (4). Although the exact mechanism of the left ventricular dysfunction has not been clar- ified, catecholamine cardiotoxicity induced by phys- ical and/or emotional stress is a suspected trigger of this syndrome (5). Most hemodialysis patients com- plain of physical and mental stresses (6,7). However, there have been few reports of Takotsubo cardiomy- opathy associated with hemodialysis. We present a case of Takotsubo cardiomyopathy in a hemodialysis patient and discuss the plausible relation between this novel cardiac syndrome and end-stage renal disease. CASE REPORT An 84-year-old woman was referred and admitted to our hospital because of several days of sustained chest discomfort and the presence of electrocardio- graphic abnormalities not detected on any previous electrocardiograms (ECGs). Acute myocardial inf-

Upload: masayoshi-fukui

Post on 21-Jul-2016

215 views

Category:

Documents


1 download

TRANSCRIPT

Page 1: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

Therapeutic Apheresis and Dialysis10(1):94–100, Blackwell Publishing Asia Pty Ltd© 2006 International Society for Apheresis

94

Blackwell Publishing AsiaMelbourne, AustraliaTAPTherapeutic Apheresis and Dialysis1774-99792005 International Society for Apheresis? 2005

101

94100

Case Report

Takotsubo Cardiomyopathy and HemodialysisM Fukui et al.

Received April 2005; revised June 2005.Address correspondence and reprint requests to Dr. Yasukiyo

Mori, Division of Hypertension and Nephrology, Kyoto Prefec-tural University of Medicine, 465 Kajii-cho, Kawaramachi-Hirokoji, Kamigyo, Kyoto 602-8566, Japan. Email:[email protected]

‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

Masayoshi Fukui, Yasukiyo Mori, Satoshi Tsujimoto, Kazuya Takehana, Norihiko Sakamoto, Noriko Kishimoto, Takanobu Imada, Hirofumi Maeba, Atsuko Nose, Hideki Yamahara, Yasuaki Kijima, Tetsuya Kitamura, Takanao Ueyama, Sanae Kikuchi, Toshiko Tokoro,

Hiroya Masaki, Mitsushige Nishikawa, and Toshiji Iwasaka

Department of Medicine II, Kansai Medical University, Osaka, Japan

Abstract: An 84-year-old woman undergoing maintenancehemodialysis presented with chest discomfort lasting sev-eral days and electrocardiographic abnormalities. She hadstopped smoking 2 weeks earlier and was experiencing irri-tability. Upon admission, electrocardiography showed ST-segment elevation in leads I, II, aVF, and V2-6 and an abnor-mal Q wave in leads II, III, and aVF. Ultrasound cardiog-raphy showed left ventricular anteroapical akinesia andbasal hyperkinesia. The chest discomfort disappeared with-out specific therapy. During hospital days 1–5, the ST-segment elevation gradually improved. Giant negativeT waves then developed. The left ventricular asynergy

resolved by day 8. Radionuclide imaging with iodine-123-beta-methyl-p-iodophenyl pentadecanoic acid, but notwith technetium-99 m-sestamibi, showed an apical defect.Elective coronary angiography showed no stenosis. ‘Takot-subo’ cardiomyopathy was diagnosed. After discharge, thepatient continued regular dialysis without cardiac symp-toms. We concluded that endogenously activated sympa-thetic nerve action in hemodialysis patients, especiallythose under emotional or physical stress, might be a caus-ative factor for Takotsubo cardiomyopathy. Key Words:Chronic kidney disease, Emotional stress, Hemodialysis,Sympathetic nerve activity, Takotsubo cardiomyopathy.

Chronic kidney disease (CKD), particularly end-stage renal disease, is associated with increasedcardiovascular morbidity and mortality (1). Manyfactors such as volume overload, hypertension, ane-mia, arteriovenous fistula, and uremic toxins mightcontribute to the development of cardiovascularevents. In addition, hemodialysis patients exhibit sus-tained activation of the sympathetic nervous system,which contributes to increased cardiovascular mor-bidity (2,3).

‘Takotsubo’ cardiomyopathy is a newly describedcardiac syndrome characterized by transient left ven-tricular dysfunction with chest discomfort, electro-cardiographic changes, and minimal release of

myocardial enzymes, together mimicking acute myo-cardial infarction (4). Although the exact mechanismof the left ventricular dysfunction has not been clar-ified, catecholamine cardiotoxicity induced by phys-ical and/or emotional stress is a suspected trigger ofthis syndrome (5). Most hemodialysis patients com-plain of physical and mental stresses (6,7). However,there have been few reports of Takotsubo cardiomy-opathy associated with hemodialysis. We present acase of Takotsubo cardiomyopathy in a hemodialysispatient and discuss the plausible relation betweenthis novel cardiac syndrome and end-stage renaldisease.

CASE REPORT

An 84-year-old woman was referred and admittedto our hospital because of several days of sustainedchest discomfort and the presence of electrocardio-graphic abnormalities not detected on any previouselectrocardiograms (ECGs). Acute myocardial inf-

Page 2: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

Takotsubo Cardiomyopathy and Hemodialysis 95

© 2006 International Society for Apheresis Ther Apher Dial, Vol. 10, No. 1, 2006

arction was suspected. Because the patient had alsoexperienced occasional post-herpes zoster neuralgiaover the prior several years, the precise onset of thecurrent cardiogenic chest discomfort was difficult todetermine. For the prior 40 years, she suffered fromchronic nephritic syndrome and renal parenchyma-tous hypertension. Hypercholesterolemia had beenpresent for the prior 5 years, but there was no his-tory of diabetes mellitus. There was no family his-tory of coronary artery disease. Although thepatient had been treated daily with antihypertensivemedications and statin (enalapril 2.5 mg, nifedipine40 mg and pravastatin 10 mg/day), her renal func-tion had declined gradually over the prior 3 years.She had been on hemodialysis for the previous2 years. According to her medical record, themaintenance hemodialysis was relatively stable; andthe appropriate dry weight was maintained. Kt/Vurea was 1.577. For the previous 60 years, she hadsmoked one pack of cigarettes per day. However,she stopped smoking abruptly 14 days prior toadmission, and consequently experienced emotionalirritability and sleep disturbance concomitant withthe chest discomfort.

Upon admission, the patient’s blood pressure was160/82 mm Hg, heart rate was 98 bpm, and tempera-ture was 36.8°C. Her level of consciousness was nor-mal. The appearance of the palpebral conjunctivaindicated slight anemia. No extra sound or signifi-cant heart murmur was heard upon auscultation. Noabdominal abnormalities were found. Neurologicaltesting showed no abnormalities. Blood tests yieldedthe following: a white blood cell count of 4700/µL(normal range: 3000–8500); hemoglobin level of10.3 g/dL (normal range: 11.3–15.4); platelet count of34.3 × 104/µL (normal range: 14.0–34.0); c-reactiveprotein level of 0.89 mg/dL (normal range: 0.00–0.30); AST level of 41 U/L (normal range: 12–31);LDH level of 386 U/L (normal range: 105–210); CKlevel of 115 U/L (normal range: 39–160); CK-MBlevel of 2.6 ng/mL (normal range: 0.5–3.2); and car-diac troponin I level of 2.32 ng/mL (normal range:0.00–0.10). Chest X-ray examination yielded a car-diothoracic ratio of 55% and no other remarkablefindings (Fig. 1). The ECG showed ST-segment ele-vation in leads I, II, aVF, and V2 through V6 and anabnormal Q wave in leads II, III, and aVF (Fig. 2a).Ultrasound cardiography showed anteroapical aki-nesia of the left ventricle but hyperkinesia of thebasal region (Fig. 3). Together with the non-recipro-cal changes in the ST-segment on the ECG and apeculiar wall motion in the left ventricle, these find-ings led us to consider other diagnoses such asTakotsubo cardiomyopathy. However, because we

could not rule out acute myocardial infarctionentirely, empiric therapy with intravenous adminis-tration of nitrates was started on hospital day 1 toimprove the patient’s symptoms. Hemodialysis wassuccessfully carried out three times per week duringthe hospital stay. The ST-segment elevation gradu-ally normalized during the first 5 days without devel-opment of critical arrhythmia (Fig. 2). Because herchest discomfort disappeared quickly, the patientwas treated with oral medications alone (enalapril2.5 mg, aspirin 100 mg and pravastatin 10 mg/day)beginning on hospital day 2. Reevaluation by ultra-sound cardiography on hospital day 8 showed thatthe left ventricular asynergy was completelyresolved. Also, the serum cardiac troponin I levelwas significantly decreased (0.16 ng/mL). On hospi-tal day 14, giant negative T waves developed(Fig. 2c). Elective cardiac catheterization was carriedout on hospital day 21 to rule out coronary arterystenosis. Coronary angiography showed normal cor-onary arteries (Fig. 4a,b), and the results of a coro-nary spasm provocation test with ergonovine werenegative (data not shown). Left ventriculographyshowed normal kinesis at that time (Fig. 4c,d). Theleft ventricular ejection fraction calculated by theSimpson method was 77%. Technetium-99m-sesta-mibi (99mTc-MIBI) quantitative gated single-photonemission computed tomography (SPECT) was car-ried out on hospital day 10. 99mTc-MIBI uptake indi-cated normal perfusion (Fig. 5a). On hospital day 25,iodine-123-beta-methyl-p-iodophenyl pentadecanoicacid (123I-BMIPP) SPECT was carried out. 123I-BMIPP images showed reduced uptake at the apexof the left ventricle (Fig. 5b). In light of these find-

FIG. 1. Chest X-ray film obtained upon admission. The cardiotho-racic ratio and lung field are normal.

Page 3: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

96 M Fukui et al.

Ther Apher Dial, Vol. 10, No. 1, 2006 © 2006 International Society for Apheresis

ings and the hospital course, the final diagnosis wasTakotsubo cardiomyopathy. The patient was dis-charged on hospital day 26 and continued toundergo regular dialysis in the clinic without any car-diac symptoms. The ECG abnormalities were stillpresent 3 months later. However, the T wave inver-sion in the precordial leads had decreased, and theabnormal Q wave was no longer detected in leads II,III, and aVF (Fig. 2d). 123I-BMIPP SPECT imagesobtained 3 months after the onset of symptoms againshowed reduced uptake in the apical area, but thereduction was significantly less than that on hospitalday 24 (Fig. 5c).

DISCUSSION

A novel cardiac syndrome manifesting as tran-sient left ventricular dysfunction without coronaryartery stenosis was recently described in Japan (8)and subsequently confirmed by reports from Europe

(9). Left ventriculograms obtained after onset showapical ballooning with basal hyperkinesia. Becausethis gives the left ventricle a peculiar shape (i.e. around bottom and narrow neck) that resembles atype of bottle used in Japan for trapping octopus,the syndrome was termed Takotsubo cardiomyopa-thy, derived from the Japanese words tako, meaningoctopus, and tsubo, meaning bottle (4,10). It has alsobeen called ampulla or amphora cardiomyopathy(11).

Despite the dramatic initial presentation, therelease of cardiac markers are disproportionatelylimited to the extent of akinesia followed by the com-plete recovery of ventricular function within severalweeks in most cases (9). In addition, the electrocar-diographic changes are quite drastic. In patients withTakotsubo cardiomyopathy, the admission ECG usu-ally shows ST-segment elevation, especially in leadsV3 through V6. T wave inversion deepens progres-sively to its first negative peak, which occurs approx-

FIG. 2. Time course of electrocardiographic changes. Electrocardiogram obtained upon admission showed ST-segment elevation in leadsI, II, aVF, and V2 through V6 and an abnormal Q wave in leads II, III, and aVF (a). By hospital day 5, the ST-segment elevation hadimproved (b). Deep inverted T waves developed in leads I, II, aVL, and V2 through V6 by hospital day 14 (c). The T wave amplificationwas reduced in the precordial leads, and the abnormal Q wave had disappeared in leads II, III, and aVF 3 months after the onset ofsymptoms (d).

(a) (b) (c) (d)

Page 4: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

Takotsubo Cardiomyopathy and Hemodialysis 97

© 2006 International Society for Apheresis Ther Apher Dial, Vol. 10, No. 1, 2006

imately 3 days after onset. The T wave is shallow forseveral days and then deepens again, and a secondnegative peak occurs at approximately 2–3 weeks(12); less frequently, there is an abnormal Q waveand prolongation of the QTc interval (13). The recip-rocal changes resembling acute myocardial infarctionare rarely seen in Takotsubo cardiomyopathy (4,14).The above-mentioned characteristic features werepresent on ultracardiographic and electrocardio-graphic images in the present case.

Takotsubo cardiomyopathy has two distinguishingcharacteristics: the syndrome is found predominantlyin elderly (post-menopausal) women (in 75% to 93%of cases); and onset is triggered by emotional orphysical stress (in 33% to 86% of cases) (15). Ourpatient was an 84-year-old woman. In addition, theabrupt cessation of smoking 14 days prior to admis-sion to our hospital had caused mental irritability.Because chronic hemodialysis patients are subject tovarious stresses caused by their health status, finan-cial status, employment status, and/or family respon-sibilities (6), it is difficult to quantify the degree ofmental stress suffered by a hemodialysis patient.However, our patient did not complain of any partic-ular emotional stress other than that caused by thecessation of smoking just 14 days prior to the onsetof Takotsubo cardiomyopathy.

Although the precise pathophysiologic basis ofTakotsubo cardiomyopathy remains unknown, cate-cholamine overload is thought to be one possiblecause (5,16,17). Indeed, radionuclide images in thissyndrome show a specific pattern. As observed in ourpatient, myocardial fatty acid metabolism measuredon 123I-BMIPP images was more severely impairedthan myocardial perfusion measured on 99mTc-MIBIor thalium-201 images (16), suggesting dysfunction ofthe cardiac catecholamine dynamics. In addition, thesympathetic nerve activity might be inappropriatelyincreased in our patient because of CKD (2,3). Neu-man et al. indicated that increased sympathetic activ-ity in patients with CKD is not dependent uponvolume status and decreased number of nephrons(18). The sympathetic nerve activity is also indepen-dent of circulating uremia-related toxins (19), indi-cating that hemodialysis patients are always prone toincreased sympathetic nerve activity. Myocardialischemic reperfusion caused by microvascular spasmis another possible cause of Takotsubo cardiomyop-athy (4). However, provocation testing with ergono-vine proved positive in only 21% of cases (8); in ourpatient, it was negative.

Although we cannot rule out the possibility thatTakotsubo cardiomyopathy occurred spontaneouslyin our patient, we suggest that Takotsubo cardiomy-

FIG. 3. The echocardiogram up-on admission showed left ven-tricular anteroapical akinesia (a)and basal hyperkinesia (b).

(a) (b)

Page 5: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

98 M Fukui et al.

Ther Apher Dial, Vol. 10, No. 1, 2006 © 2006 International Society for Apheresis

opathy might be easily induced in hemodialysispatients enduring only mild emotional stress becauseof the increased sympathetic nerve activity thatoccurs with CKD. Interestingly, four of 88 patientswith Takotsubo cardiomyopathy in a Japanese multi-center study were undergoing maintenance hemodi-alysis (8), suggesting that the incidence of Takotsubocardiomyopathy in hemodialysis patients might behigher than that in the general population. Contin-

ued accumulation of data is needed to clarify thisissue.

In summary, we encountered Takotsubo cardiomy-opathy in a hemodialysis patient. The causes of chestpain or discomfort in hemodialysis patients are many,including acute myocardial ischemia, acute pericardi-tis, pleuritis, air embolism, gastroesophageal reflux,herpes zoster and musculoskeletal disorder (20).Takotsubo cardiomyopathy should be considered in

FIG. 4. Coronary angiogramsand left ventriculograms ob-tained on hospital day 21. Therewas no epicardial coronary ar-tery stenosis (a,b). Left ventricu-lography showed normokinesia(c,d).

(a)

(c)

(b)

(d)

Page 6: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

Takotsubo Cardiomyopathy and Hemodialysis 99

© 2006 International Society for Apheresis Ther Apher Dial, Vol. 10, No. 1, 2006

the differential diagnosis of cardiac symptoms inhemodialysis patients.

REFERENCES

1. Sarnak MJ, Levey AS, Schoolwerth AC et al. Kidney diseaseas a risk factor for development of cardiovascular disease: astatement from the American Heart Association Councils onKidney in Cardiovascular Disease, High Blood PressureResearch, Clinical Cardiology, and Epidemiology and Preven-tion. Hypertension 2003;42:1050–65.

2. Converse RL Jr, Jacobsen TN, Toto RD et al. Sympatheticoveractivity in patients with chronic renal failure. N Engl JMed 1992;327:1912–8.

3. Zoccali C, Mallamaci F, Parlongo S et al. Plasma norepineph-rine predicts survival and incident cardiovascular events inpatients with end-stage renal disease. Circulation2002;105:1354–9.

4. Ishikawa K. ‘Takotsubo’ cardiomyopathy A syndrome charac-terized by transient left ventricular apical ballooning thatmimics the shape of a bottle used for trapping octopus inJapan. Intern Med 2004;43:275–6.

5. Akashi YJ, Nakazawa K, Sakakibara M, Miyake F, Sasaka K.Reversible left ventricular dysfunction ‘takotsubo’ cardiomy-opathy related to catecholamine cardiotoxicity. J Electrocar-diol 2002;35:351–6.

6. Mapes DL, Bragg-Gresham JL, Bommer J et al. Health-related quality of life in the Dialysis Outcomes and PracticePatterns Study (DOPPS). Am J Kidney Dis 2004;44:54–60.

7. Heiwe S, Clyne N, Dahlgren MA. Living with chronic renalfailure: patients’ experiences of their physical and functionalcapacity. Physiother Res Int 2003;8:167–77.

8. Tsuchihashi K, Ueshima K, Uchida T et al. Transient leftventricular apical ballooning without coronary artery stenosis:a novel heart syndrome mimicking acute myocardial infarc-

tion. Angina Pectoris-Myocardial Infarction Investigations inJapan. J Am Coll Cardiol 2001;38:11–8.

9. Desmet WJ, Adriaenssens BF, Dens JA. Apical ballooning ofthe left ventricle: first series in white patients. Heart2003;89:1027–31.

10. Nyui N, Yamanaka O, Nakayama R, Sawano M, Kawai S.‘Tako-Tsubo’ transient ventricular dysfunction: a case report.Jpn Circ J 2000;64:715–9.

11. Kawai S, Suzuki H, Yamaguchi H et al. Ampulla cardiomyop-athy (‘Takotusbo’ cardiomyopathy)—reversible left ventricu-lar dysfunction: with ST segment elevation. Jpn Circ J2000;64:156–9.

12. Kurisu S, Inoue I, Kawagoe T et al. Time course of electro-cardiographic changes in patients with tako–tsubo syndrome:comparison with acute myocardial infarction with minimalenzymatic release. Circ J 2004;68:77–81.

13. Ogura R, Hiasa Y, Takahashi T et al. Specific findings of thestandard 12-lead ECG in patients with ‘Takotsubo’ cardiomy-opathy: comparison with the findings of acute anterior myo-cardial infarction. Circ J 2003;67:687–90.

14. Moriya M, Mori H, Suzuki N, Hazama M, Yano K. Six-monthfollow-up of takotsubo cardiomyopathy with I-123-beta-metyl-iodophenyl pentadecanoic acid and I-123-meta-iodo-benzyl-guanidine myocardial scintigraphy. Intern Med2002;41:829–33.

15. Ueyama T. Emotional stress-induced Tako-tsubo cardiomyop-athy: animal model and molecular mechanism. Ann N Y AcadSci 2004;1018:437–44.

16. Owa M, Aizawa K, Urasawa N et al. Emotional stress-induced‘ampulla cardiomyopathy’: discrepancy between the meta-bolic and sympathetic innervation imaging performed duringthe recovery course. Jpn Circ J 2001;65:349–52.

17. Abe Y, Kondo M, Matsuoka R, Araki M, Dohyama K, TanioH. Assessment of clinical features in transient left ventricularapical ballooning. J Am Coll Cardiol 2003;41:737–42.

18. Neumann J, Ligtenberg G, Klein II, Koomans HA,Blankestijn PJ. Sympathetic hyperactivity in chronic kidney

FIG. 5. Nuclear cardiographicevaluations showed a clear dis-crepancy between perfusion(99mTc-MIBI) (a) images andmetabolic (123I-BMIPP). (b)images mainly at the left ven-tricular apex in the initial phase.After 3 months, the defect inthe apical region on 123I-BMIPPimages was significantlyimproved (c).

(a) (b) (c)

Page 7: ‘Takotsubo’ Cardiomyopathy in a Maintenance Hemodialysis Patient

100 M Fukui et al.

Ther Apher Dial, Vol. 10, No. 1, 2006 © 2006 International Society for Apheresis

disease: pathogenesis, clinical relevance, and treatment. Kid-ney Int 2004;65:1568–76.

19. Hausberg M, Kosch M, Harmelink P et al. Sympathetic nerveactivity in end-stage renal disease. Circulation 2002;106:1974–9.

20. Modi KS, Gross D, Davidman M. The patient developingchest pain at the onset of haemodialysis sessions—it is notalways angina pectoris. Nephrol Dial Transplant 1999;14:221–3.