cpg fluids

44
  PJSS Vol. 55, No. 4, October-December , 2000 PJSS PHILIPPINE COLLEGE OF SURGICAL SPECIALTIES  Evidence-based Clinical Practice Guidelines on Some Important Aspects of the Care of Critically-ill Surgical Patients Part I : Fluids for Resuscitation, Blood Transfusion, Assessment of Volume Resuscitation, Nutritional Support, Pharmacologic Cardiovascular Support Adriano V. Laudico, M.D., F.P.C.S. 1  ; M. Francisco T. Roxas, M.D., F.P.C.S. 1 ; Eduardo R. Bautista, M.D., F.P.A.T.A.C.S.I. 1 ; Ma.Luisa D. Aquino, M.D., F.P.C.S. 2  ; Arturo S. De la Peña, M.D., F.P.C.S. 1 ; Armando C. Crisostomo, M.D., F.P.C.S. 1  ; Narciso S. Navarro Jr., M.D., F.P.C.S. 3 ; Camilo C. Roa, M.D., F.P.C.P. 4  ; Abundio A. Balgos, M.D., F.P.C.P 4 .; Donato R. Marañon, M.D., F.P.C.P. 5 1  Department of Surgery, College of Medicine and Philippine General Hospital, University of the Philippines Manila 2  Department of Surgery, Jose R. Reyes Memorial Medical Center, Manila 3  Department of Surgery, Santo Tomas University Hospital, Manila 4  Section of Pulmonary Medicine, College of Medicine and Philippine General Hospital, University of the Philippines Manila 5  Section of Cardiology, College of Medicine and Philippine General Hospital, University of the Philippines Manila EXECUTIVE SUMMARY The clinical area identified by the Philippine College of Surgeons (PCS) for the second evidence-based clinical practice guidelines (EBCPGs) was on the care of the critically ill surgical patient. A research grant was provided  by GlaxoWellcome Philippin es, Inc. and a Technical Working Group (TWG) was formed, composed of 6 general surgeons, one thoracic/cardiovascular surgeon, one cardiologist and two pulmunologists. The TWG was tasked to identify the clinical questions, and to adhere to the PCS-approved method of developing EBCPGs. The TWG began work on May 1, 2000. The literature search, limited to English  publicatio ns, used both electronic and manual methods. Three electronic databases were used: 1) The Cochrane Library, Issue 2, 2000; 2) National Library of Medicine - Medline (PubMed, no time limit); and HERDIN (Health Research and Development Information  Network) V ersion 1, 1997 of DOST-PCHRD. Titles of all articles were printed and all members of the TWG went over the list and was composed of one thoracic/cardiovascular surgeon and one cardiologist. Group III was assigned to appraise the evidence on ventilatory support, and Group IV looked at  pharmacolog ical cardiovascu lar support. Groups I and II appraised the evidence on fluid resuscitation, blood transfusion, monitoring and nutritional support. The printed abstracts were given to the members of each group, who then decided which articles were to be included for full text retrieval. The full texts were obtained from the University of the Philippines Manila Library, and were appraised using standard forms.The TWG then compiled, summarized and classified the evidence according to 3 levels and proposed a first draft of recommendations according to 3 categories: LEVELS OF EVIDENCE I. Evidence from at least one properly designed randomized controlled trial, meta-analysis. II. Evidence from at least one well-designed clinical trial without proper randomization, from prospective or retrospective cohort or case-control anal tic studies referabl from

Upload: dsalazarparas

Post on 07-Oct-2015

57 views

Category:

Documents


0 download

DESCRIPTION

GS

TRANSCRIPT

  • PJSS Vol. 55, No. 4, October-December, 2000

    PJSS PHILIPPINE COLLEGE OF SURGICALSPECIALTIES

    Evidence-based Clinical Practice Guidelines on Some Important Aspects of the Care ofCritically-ill Surgical Patients Part I : Fluids for Resuscitation, Blood Transfusion,Assessment of Volume Resuscitation, Nutritional Support, Pharmacologic CardiovascularSupport

    Adriano V. Laudico, M.D., F.P.C.S.1 ; M. Francisco T. Roxas, M.D., F.P.C.S.1;

    Eduardo R. Bautista, M.D., F.P.A.T.A.C.S.I.1; Ma.Luisa D. Aquino, M.D., F.P.C.S.2 ;

    Arturo S. De la Pea, M.D., F.P.C.S. 1; Armando C. Crisostomo, M.D., F.P.C.S.1 ;

    Narciso S. Navarro Jr., M.D., F.P.C.S.3; Camilo C. Roa, M.D., F.P.C.P.4 ;

    Abundio A. Balgos, M.D., F.P.C.P4.; Donato R. Maraon, M.D., F.P.C.P.5

    1 Department of Surgery, College of Medicine and Philippine General Hospital, University ofthe Philippines Manila

    2 Department of Surgery, Jose R. Reyes Memorial Medical Center, Manila

    3 Department of Surgery, Santo Tomas University Hospital, Manila

    4 Section of Pulmonary Medicine, College of Medicine and Philippine General Hospital,University of the Philippines Manila

    5 Section of Cardiology, College of Medicine and Philippine General Hospital, University of thePhilippines Manila

    EXECUTIVE SUMMARY

    The clinical area identified by the PhilippineCollege of Surgeons (PCS) for the secondevidence-based clinical practice guidelines(EBCPGs) was on the care of the critically illsurgical patient. A research grant was providedby GlaxoWellcome Philippines, Inc. and aTechnical Working Group (TWG) was formed,composed of 6 general surgeons, onethoracic/cardiovascular surgeon, onecardiologist and two pulmunologists. TheTWG was tasked to identify the clinicalquestions, and to adhere to the PCS-approvedmethod of developing EBCPGs.

    The TWG began work on May 1, 2000. Theliterature search, limited to Englishpublications, used both electronic and manualmethods. Three electronic databases wereused: 1) The Cochrane Library, Issue 2, 2000;2) National Library of Medicine - Medline(PubMed, no time limit); and HERDIN (HealthResearch and Development InformationNetwork) Version 1, 1997 of DOST-PCHRD.

    Titles of all articles were printed and allmembers of the TWG went over the list andchecked the titles of articles whose abstract

    was composed of one thoracic/cardiovascularsurgeon and one cardiologist. Group III wasassigned to appraise the evidence onventilatory support, and Group IV looked atpharmacological cardiovascular support.Groups I and II appraised the evidence on fluidresuscitation, blood transfusion, monitoringand nutritional support.

    The printed abstracts were given to themembers of each group, who then decidedwhich articles were to be included for full textretrieval. The full texts were obtained from theUniversity of the Philippines Manila Library,and were appraised using standard forms.TheTWG then compiled, summarized andclassified the evidence according to 3 levelsand proposed a first draft of recommendationsaccording to 3 categories:

    LEVELS OF EVIDENCE

    I. Evidence from at least one properly designedrandomized controlled trial, meta-analysis.

    II. Evidence from at least one well-designedclinical trial without proper randomization,from prospective or retrospective cohort orcase-control analytic studies (preferably from

  • checked the titles of articles whose abstractthey felt should be read. The abstracts of allchecked articles were printed.

    The TWG was divided into 4 groups: Group Iwas composed of 3 general surgeons; Group IIwas composed of 2 general surgeons; GroupIII was composed of onethoracic/cardiovascular surgeon and 2pulmunologists; and Group IV

    case-control analytic studies (preferably fromone center), from multiple time-series studies,or from dramatic results in uncontrolledexperiments.

    Care of Critically-ill Surgical Patients

    III. Evidence from opinions of respectedauthorities on the basis of clinical experience,descriptive studies, or reports of expertcommittees.

    CATEGORIES OFRECOMMENDATIONS

    Category A : Recommendations that wereapproved by consensus (75% of themultisectoral expert panel).

    Category B : Recommendations that were

    somewhat controversial and did not meetconsensus.

    Category C : Recommendations that causedreal disagreements among members of thepanel.

    The following were the clinical questions:

    1. What is the appropriate type ofintravenous fluid to be used in theresuscitation of hypovolemic critically illsurgical patients?

    2. What are the adverse effects of bloodtransfusion?

    3. What are the indications for bloodtransfusion in critically ill surgical patients?

    4. What are the clinical and/or diagnosticindicators that have been shown to be usefulin assessing the adequacy of resuscitation ofhypovolemic critically ill surgical patients?

    5. Are there patients who will benefit frompostoperative nutritional support?

    6. What is the most effective way ofdelivering postoperative nutritional

    Shangri-la Hotel. The guidelines wereapproved by the PCS Board of Regents 13December 2000.

    RECOMMENDATIONS

    FLUIDS FOR RESUSCITATION

    1. Crystalloids are the appropriate intravenousfluids to be used in the resuscitation ofhypovolemic surgical patients. The use ofcolloids does not decrease mortality, and mayresult in higher mortality in trauma patients.

    Albumin may increase the risk of death inpatients with hypovolemia, burns andhypoalbuminemia. ( Level 1, Category A)

    RED BLOOD CELL TRANSFUSION

    1. Aggressive and early control of hemorrhageshould be considered an integral part ofresuscitation. (Level II, Category A)

    2. In a patient who is hypovolemic secondaryto acute hemorrhage, volume resuscitationshould be initiated. (Level I, Category A)

    3. In a normovolemic patient whose level ofanemia is associated with evidence of anincreased risk of major adverse outcomes, themajor indication of RBC transfusion is toincrease oxygen delivery. This should becorrected to achieve a level of anemia whereinthere is evidence that the increased risk is nolonger present.

    The traditional "10/30 rule", or a transfusiontrigger of an Hb concentration of less than 10 g/ dL or a hematocrit concentration of less than30 per cent should no longer be utilized. (Level1, Category A)

    4. There is evidence that there is no benefit of

  • delivering postoperative nutritionalsupport?

    7. What is the most effective composition ofpostoperative nutritional support?

    8. When is pharmacological cardiovascularsupport beneficial among critically illsurgical patients?

    9. What is the appropriate method ofpharmacological cardiovascular support ofthe critically ill surgical patient?

    A total 3,846 titles were printed, out ofwhich 546 abstracts were reviewed, 114 fulltext articles appraised, and 71 full textarticles used.

    The TWG prepared a first draft of themanuscript which consisted of a summaryof the strongest evidence associated with theclinical questions and suggestedrecommendations. The first draft wasdiscussed and modified by a Panel ofExperts convened by the PCS on November4 and 11, 2000 at the PCS building. A seconddraft was prepared by the TWG and thiswas discussed in a Public Forum onDecember 13, 2000 during the 56th ClinicalCongress of the PCS held at Edsa

    4. There is evidence that there is no benefit ofRBC transfusion for patients whose Hbconcentration is equal or more than 10 g / dL.(Level I, Category A)

    5. In general, there may be an increased risk ofadverse outcomes in normovolemic patientswith acute anemia whose Hb concentration isless than 7 g / dL. (Level I, Category A)

    6. Normovolemic patients with acute anemiawho have cardiac disease or are at risk ofcardiac disease may benefit from RBCtransfusion when their Hb concentration is lessthan 7 g / dL. Factors that increase the risk ofpostoperative cardiac events, and thecorresponding evidence are contained in PCSEBCPG on Seeking Referral for PreoperativeCardiac Evaluation for Elective NoncardiacSurgery (PJSS 1999; 54 (4): 171 - 233). (LevelI, Category A)

    PJSS Vol. 55, No. 4, October-December, 2000

    Patients with acute myocardial infarction orunstable angina may benefit from a transfusiontrigger higher than 7 g / dL. (Level II, CategoryA)

    7. Packed red blood cells is preferred whenblood transfusion is necessary. (Level II,Category A)

    ASSESSMENT OF RESUSCITATION OFHYPOVOLEMIA

    1. The clinical judgment of a well-trained andexperienced surgeon, utilizing clinicalparameters, such as blood pressure, heart rateand urine output should still be the best"monitor" of the adequacy of resuscitation.(Level II, Category A)

    2. At present there is no evidence that theroutine use of tests such as serum lactate

    PHARMACOLOGICCARDIOVASCULAR SUPPORT

    1. When volume resuscitation is consideredadequate based on clinical judgment and theclinical response is suboptimal, vasoactivedrugs may improve hemodynamics, oxygentransport and clinical condition. (Level I,Category A)

    2. Hemodynamic monitoring using apulmonary artery catheter (PAC) may improveclinical decision making in selected criticallyill patients. (Level II, Category A)

    3. When vasoactive drugs are used, the goalsof therapy remain normalization ofhemodynamics, oxygen transport, andimprovement in the clinical manifestations ofshock, but not supranormal resuscitationendpoints. (Level I, Category A)

  • routine use of tests such as serum lactatelevels, base deficits and pulmonary arterycatheters (PAC) improve major clinicaloutcomes and therefore are not recommended.(Level I, Category A)

    3. Indications for the use of tests should bebased on specific circumstances in which thereis strong evidence that their use is associatedwith an improvement in major clinicaloutcomes, such as a decrease in mortality andserious complications. (Level I, Category A)

    POSTOPERATIVE NUTRITIONALSUPPORT

    1. Early nutritional support (within 24 hours)of critically ill surgical patients should beinstituted because it decreases the risk ofpostoperative morbidity, particularly septiccomplications, and more so in malnourishedpatients (greater than 10% preoperative weightloss). (Level I, Category A)

    2. Among critically ill surgical patients, there isevidence that enteral nutrition (EN) deliveredinto the jejunum and started within 24 hoursafter operation results in lower septiccomplications compared to total parenteralnutrition (TPN). TPN is used only when theenteral route is not feasible. (Level I, CategoryA)

    The clinical trials that had demonstratedbenefits of early TEN had used commercialTEN formulations.

    3. TEN using commercial immune-enhancingformulae may be associated with a greaterdecrease in the risk of septic complicationscompared to standard TEN formulae inselected critically ill patients. (Level I,Category A)

    4. The routine use of postoperative enteralnutrition (EN) in noncritical, nonmalnourishedpatients is not recommended.(Level I,Category A)

    endpoints. (Level I, Category A)

    4. After restoring volume, clinical judgmentshould be exercised in the particular choice ofa vasoactive drug, with consideration ofhemodynamic parameters including heart rate,blood pressure, cardiac index, volume statusand cardiac filling pressures, vascularresistance, and preexisting cardiac and medicalstatus of an individual patient.

    4a. Dobutamine is the drug of choice toincrease cardiac index and oxygen delivery ifblood pressure is acceptable but there are stillsigns of inadequate tissue perfusion. (Level II,Category A)

    4b. Dopamine should be used initially forpersistent mild hypotension > 70 and < 90mmHg and shock to support blood pressure;and dobutamine may then be added to optimizecardiac index and oxygen delivery. (Level II,Category A)

    4c. Dopamine and/or norepinephrine shouldgenerally be used for severe hypotension withsystolic blood pressure

  • Abundio A. Balgos (pulmonologist)

    Camilo C. Roa (pulmonologist)

    Group IV Eduardo R. Bautista (thoracic andcardiovascular surgeon)

    Donato R. Maraon (cardiologist)

    Research Assistant Jaimelly P. Buenbrazo

    Panel of Experts

    1. Joel U. Macalino (general surgeon, PCS)

    2. Nestor E.dela Cruz (general surgeon, PCS)

    3. Adriano V. Laudico (general surgeon, PCS)

    4. Narciso S. Navarro Jr. (generalsurgeon,PCS)

    5. M. Francisco T. Roxas (general surgeon,PCS)

    6. Ma. Luisa D. Aquino (general surgeon,PCS)

    7. Edgardo R. Cortez (general surgeon, PCS)

    8. Camilo C. Roa (pulmonologist, PhilippineCollege of Chest Physicians)

    9. Eduardo Bautista (thoracic andcardiovascular surgeon, PCS)

    10. Abundio A. Balgos (pulmonologist,Philippine College of Chest Physicians)

    11. Donato R. Maraon (cardiologist,Philippine College of Physicians)

    12. Ramon S. Inso (general surgeon, SurgicalOncology Society of the Philippines)

    13. Teresita S. de Guia (pulmonologist,Philippine College of Chest Physicians)

    14. Myrna N. Baares (pulmonologist,Philippine College of Chest Physicians)

    15. Edgar A. Baltazar (general surgeon,Philippine Society of General Surgeons)

    16. Edgardo F. Fernandez (general surgeon,Philippine Society of General Surgeons)

    17. Guillermo B. Barroa Jr. ( thoracic andcardiovascular surgeon, Philippine Associationof Thoracic and Cardiovascular Surgery, Inc.)

    18. Teodoro G. Herbosa (general surgeon,Philippine Society for the Surgery of Trauma)

    19. Eric A. Tayag (epidemiologist, Departmentof Health)

    ACKNOWLEDGEMENT

    The Philippine College of Surgeons (PCS) hadidentified the development, dissemination andimplementation of evidence-based clinicalpractice guidelines (EBCPG), as an importantstrategy in improving surgical care, trainingand research. The Philippine Council forHealth Research and Development (PCHRD)of the Department of Science and Technology(DOST) had also identified the development ofEBCPGs as one of the top priorities in thenational research agenda, and in 1999 theDOST-PCHRD, PCS and the Department ofSurgery of the University of the PhilippinesManila College of Medicine signed a trilateralMemorandum of Agreement on thedevelopment of EBCPGs on certain areas ofsurgical care in the Philippines. These areas ofsurgical care should be those wherein currentpractice may not be truly evidence-based, andhave a large potential of improving majoroutcomes and even decreasing costs, and therecommended EBCPGs can be implementednationwide in both government and privatehealth facilities. The first PCS-EBCPG was onseeking referral for perioperative cardiacevaluation for noncardiac surgery and on whenthe intraoperative presence of acardiologist/internist would be beneficial.1

    The fourth Surgical Forum, also known as the"Subic Bay Forum" was held on 25-27 July1997 and dealt with the "ImmuneConsequences of Trauma, Shock and Sepsis,"co-sponsored by the Philippine College ofSurgeons' Surgical Infection Committee andMSD Philippines. This new concept, alsocalled SIRS/MODS/MOF (systemicinflammatory response syndrome / multipleorgan dysfunction syndrome/ multiple organfailure) which is extremely crucial to the betterunderstanding of the prevention and successfulmanagement of critical surgical illnesses, hassince been, and will continue to be, widelydisseminated through continuing medical

    education activities nationwide.2-4

    On 29-30 July 1999, a symposium on surgicalcritical care was held in Phuket, Thailand, co-sponsored by the PCS and GlaxoWellcomePhilippines Inc., and the symposium revealedthat there was still no general agreement oncertain key aspects in the management ofcritically ill surgical patients, and that therewere current practice variations, some ofwhich may not be cost-

  • The project was supported by a research grantto the Philippine College of Surgeons fromGlaxoWellcome Philippines Incorporated.

    PJSS Vol. 55, No. 4, October-December, 2000

    effective and may also lead to higher mortalityand morbidity. The symposium participantsunanimously recommended that the PCS comeup with an EBCPG, and that GlaxoWellcomePhilippines support the project.5

    Methods

    The PCS appointed a Technical WorkingGroup (TWG) composed of 6 general

    surgeons, one thoracic/cardiovascular surgeon,2 pulmunologists and 1 cardiologist. The groupwas instructed to adhere to the methods used inthe development of the guidelines onpreoperative cardiac evaluation (Roxas 1999)and were given a free hand to formulate thespecific clinical questions on areas consideredimportant in the perioperative care of thecritically ill surgical patient such asintravenous fluids for resuscitation, bloodtransfusion, monitoring of resuscitation,nutritional support, pharmacologicalcardiovascular support and ventilatory support.

    The TWG began work on 1 May 2000 andinitial clinical questions were listed, thenrevised several times as the search of theliterature progressed.

    The literature search, limited to Englishpublications, used both electronic and manualmethods. Three electronic databases were used:1) The Cochrane Library, Issue 2, 2000; 2)National Library of Medicine - Medline(PubMed, no time limit); and HERDIN (HealthResearch and Development InformationNetwork) Version 1, 1997 of DOST-PCHRD.Manual searching of the reference lists ofreview articles and some important meta-analyses and randomized controlled trials(RCTs) were also done. The search terms usedwere: 1) Cochrane Library : critical injuries,benefits, blood volume, crytalloids, drugs,hypotension, hypovolemia, surgical intensivecare, central venous pressure, nutrition,survival, colloids, complications, dopamine,

    drugs, hemoglobin, congestive heart failure,intravenous fluids, monitoring, plateletconcentrate, surgical intensivist, trauma,resuscitation, packed red blood cells,laboratory tests, nutrition, respirator,surveillance, ventilatory support, surgery,pharmacologic support, medications,nutritional support, respiratory failure,transfusion, work-up, plasma, surgicalintensive care, 3) HERDIN : colloids,

    crystalloids, nutrition, critical injuries, bloodtransfusion, central venous pressure,resuscitation, surgery, blood volume, intensivecare, fluid resuscitation. In the case of subjectswhere there were relatively few or absentRCTs, cohort studies, case-control studies, caseseries reports and review articles wereincluded, and their references manuallysearched. The primary outcome of interest wasmortality.

    Titles of all articles were printed and allmembers of the TWG went over the list andchecked the titles of articles whose abstractthey felt should be read. The abstracts of allchecked articles were printed.

    The TWG was divided into 4 groups: Group Iwas composed of 3 general surgeons; Group IIwas composed of 2 general surgeons; GroupIII was composed of onethoracic/cardiovascular surgeon and 2pulmonologists; and Group IV was composedof one thoracic/cardiovascular surgeon and onecardiologist. Group III was assigned toappraise the evidence on ventilatory support,and Group IV looked at pharmacologicalcardiovascular support. Groups I and IIappraised the evidence on fluid resuscitation,blood transfusion, monitoring and nutritionalsupport.

    The printed abstracts were given to themembers of each group, who then decidedwhich articles were to be included for full textretrieval (Appendix 1). The full texts wereobtained from the University of the Philippines

  • blood transfusion, hypovolemic shock, fluidresuscitation, 2) Medline : anemia, cardiacfailure, colloids, critically injured, diagnosis,fresh whole blood, hemorrhagic shock, bloodcomponents, cardiogenic shock, critical care,crystalloids, diagnostic tests, hematocrit,hypovolemia, blood transfusion, cardiovascularfailure, critically ill, inotropic drugs,

    obtained from the University of the PhilippinesManila Library, and were appraised usingstandard forms (Appendices 2,3). The TWGthen compiled, summarized and classified theevidence according to 3 levels and proposed afirst draft of recommendations according to 3categories:

    LEVELS OF EVIDENCE

    I. Evidence from at least one properly designedrandomized controlled trial, meta-analysis.

    Care of Critically-ill Surgical Patients

    II. Evidence from at least one well-designedclinical trial without proper randomization,from prospective or retrospective cohort orcase-control analytic studies (preferably fromone center), from multiple time-series studies,or from dramatic results in uncontrolledexperiments.

    III. Evidence from opinions of respectedauthorities on the basis of clinical experience,descriptive studies, or reports of expertcommittees.

    CATEGORIES OFRECOMMENDATIONS

    Category A: Recommendations that wereapproved by consensus (75% of themultisectoral expert panel).

    Category B: Recommendations that weresomewhat controversial and did not meetconsensus.

    Category C: Recommendations that causedreal disagreements among members of thepanel.

    The following were the clinical questions:

    1. What is the appropriate type of intravenousfluid to be used in the resuscitation ofhypovolemic critically ill surgical patients?

    2. What are the adverse effects of bloodtransfusion?

    3. What are the indications for bloodtransfusion in critically ill surgical patients?

    4. What are the clinical and/or diagnostic

    8. When is pharmacological cardiovascularsupport beneficial among critically ill surgicalpatients?

    9. What is the appropriate method ofpharmacological cardiovascular support of thecritically ill surgical patient?

    The TWG prepared a first draft of themanuscript which consisted of a summary ofthe strongest evidence associated with theclinical questions and suggestedrecommendations. The first draft wasdiscussed and modified by a Panel of Expertsconvened by the PCS on November 4 and 11,2000 at the PCS building. A second draft wasprepared by the TWG and this was discussedin a Public Forum on December 13, 2000during the 56th Clinical Congress of the PCSheld at Edsa Shangri-la Hotel. The guidelineswere approved by the PCS Board of Regents13 December 2000.

    Results

    A total of 3,846 titles were printed, out ofwhich 546 abstracts were reviewed. A total of114 full text articles were critically appraised,of which 71 were included in the manuscript.

    FLUIDS FOR RESUSCITATION

    1. What is the appropriate type of intravenousfluid to be used in the resuscitation ofhypovolemic critically ill surgical patients?

    There were 5 systematic reviews (meta-analyses) that assessed the effects on mortalityof colloids compared to crystalloids for fluidresuscitation in critically ill patients.

  • indicators that have been shown to be useful inassessing the adequacy of resuscitation ofhypovolemic critically ill surgical patients?

    5. Are there patients who will benefit frompostoperative nutritional support?

    6. What is the most effective way of deliveringpostoperative nutritional support?

    7. What is the most effective composition ofpostoperative nutritional support?

    Alderson, et al6 (2000) using the CochraneInjuries Group methods for search, studyselection, assessment of methodologicalquality, data collection and analysis7, looked atall randomized and quasi-random trials of anytype of colloids (either alone or in combinationwith crystalloids) compared to crystalloids(any type) in patients requiring volumereplacement (excluding

    PJSS Vol. 55, No. 4, October-December, 2000

    neonates, pregnant women and cross-overtrials). Forty eight studies, involving 2,884patients, were included. The authors concludedthat: "There is no evidence from randomizedclinical trials that resuscitation with colloidsreduces the risk of death compared tocrystalloids in patients with trauma, burns andfollowing surgery." A sample of the meta-analysis, looking at hydroxymethyl starch,modified gelatin and dextran is shown inFigure 1.8-55

    Choi, et al.56 (1999), looked at 105 clinicaltrials of adult patients requiring fluidresuscitation comparing isotonic crystalloids(lactated Ringer's solution, normal saline,Ringer's acetate and balanced electrolytesolution) versus colloids. They included 17studies involving 814 patients, and concluded:"Overall, there is no apparent difference inpulmonary edema, mortality or length of staybetween isotonic crystalloids and colloidresuscitation. Crystalloid resuscitation is

    associated with a lower mortality in traumapatients" (Figures 2,3).10,13,18,23,24,25a,25c,26d,30,31b,34,36,39,41,48,57,58

    Schierhout and Roberts59 (1998) reported theirsystematic review of 19 unconfounded RCTsthat compared mortality and the use of purecolloids with crystalloids (N= 1,315) forvolume replacement of critically ill patients,and stratified according to patient type andquality of allocation concealment. Colloidswere associated with an increased absolute riskof mortality of 4% (95% confidence interval0%, 8%) or 4 extra deaths for every 100

    confidence intervals are wide and do notexclude significant differences betweencolloids." 8-9,17,35,36c,42,62-93,94a-94b,95-102

    The Albumin Reviewers of the CochraneInjuries Group focused on the benefits or harmin using human albumin solutions forresuscitation and volume expansion incritically ill patients (1998103, 2000104). Theyincluded 24 RCTs (N=1,204). They placed thepatients into 3 categories (hypovolemia, burns,hypoalbuminemia), and for each category therisk of death was higher in the albumin groupcompared to the comparison group (Figure 5).The authors concluded that: "There is noevidence that albumin administration reducesthe risk of death in critically ill patients withhypovolemia, burns or hypoalbuminemia, anda strong suggestion that it may increase the riskof death." 10,18,19b,22d,25a,26d,36c,38,42,48,54-55,105-116

    Recommendations

    1. Crystalloids are the appropriate intravenousfluids to be used in the resuscitation ofhypovolemic surgical patients. The use ofcolloids does not decrease mortality, and mayresult in higher mortality in trauma patients.Albumin may increase the risk of death inpatients with hypovolemia, burns andhypoalbuminemia. (Level 1, Category A)

    RED BLOOD CELL TRANSFUSION

    The HIV/AIDS scare generated a globalinterest in blood transfusion, and in developedcountries, the last two decades had seen

  • 0%, 8%) or 4 extra deaths for every 100patients resuscitated, when only trials withadequate allocation concealment wereincluded. There was no evidence fordifferences among patients with different typesof injuries (trauma, surgery, burns, others)(Figure 4). The authors concluded that themeta-analysis "does not support the continueduse of colloids for volume replacement incritically ill patients."10,13,18-20,22-23,25,30-32,36,42,45,47-49,52,60

    Bunn, et al61 (1999) reported a meta-analysisthat compared the effects of different colloidsolutions on mortality in patients that werethought to need volume replacement:hydroxyethyl starch (HES), albumin, plasmaprotein fractions (PPF), gelatin and dextran.They included 46 studies (N=2,884), andconcluded that: "From this review, there is noevidence that one colloid solution is moreeffective or safe than any other, although the

    countries, the last two decades had seenheightened activities on the safety and properutilization of blood transfusion. In thePhilippines, two additional factors have to bestrongly considered in the decision process.First, there still is a relative scarcity of bloodproducts, and the deficits can vary widely notonly between different areas in the country, buteven between hospitals in the same area.Second, the technical and procedural standardsused in developed countries so that transfusionrisks may be minimized are still not evenlyenforced owing to a variety of reasons.

    Care of Critically-ill Surgical Patients

  • PJSS Vol. 55, No. 4, October-December, 2000

  • Care of Critically-ill Surgical Patients

  • PJSS Vol. 55, No. 4, October-December, 2000

    Figure 5. Fixed effects model of relative risks (95% confidence interval) of death associatedwith intervention (fluid resuscitation with albumin or plasma protein fraction) compared withcontrol (no albumin or plasma protein fraction or resuscitation with a crystalloid solution) incritically ill patients. (Cochrane Injuries Group Albumin Reviewers, BMJ 1998; 317)

  • Care of Critically-ill Surgical Patients

    2. What are the adverse effects of bloodtransfusion?

    There were several reviews on the manyreported perils associated with allogeneicblood transfusion (ABT). The term ABT was

    used to mean the infusion of either wholeblood or any red blood cell (RBC) productfrom an unrelated donor. (Autologoustransfusion refers to the transfusion of thepatient's own blood, obtained through variousmethods such as intraoperative/postoperativesalvage, preoperative donation, and isovolemichemodilution). Three articles are cited.

    In the United States, the National Institutes ofHealth (NIH) together with other agenciesconvened a Consensus DevelopmentConference on Perioperative Red Blood CellTransfusion on June 1988, which cautionedthat "the need for possibly improvedoxygenation must be weighed against the risksof adverse consequences, both short-term andlong-term." Transmission of infectious agentsare the most feared which consists of not onlya host of viruses (known and unknown) such asHIV, HBV, HCV, cytomegalovirus, HTLV,EBV and parvoviruses, but othermicroorganisms such as malaria, Trypanozomacruzi and bacteria. Noninfectious morbiditiesare mainly attributed to immune mechanismsand include hemolytic and nonhemolyticreactions, immunosuppression and graft-versus-host disease.1

    Goodnough and Schuk followed with a moreextensive review in 1990.2 They reported thatof a total of 121,171 AIDS cases in the UnitedStates by 1990, 3,120 cases (2.6%) weretransfusion-associated. Even with state-of-the-art screening, it was estimated that 67 to 227

    which were invariably related to blood groupincompatibility resulted mainly from clericalerrors, such as transfusing the wrong unit to amisidentified patient.

    Of special interest to surgeons is the strongpossibility that ABT, particularly thosecontaining components other than RBC, mayhave immunomodulatory consequences andmay increase the risk of immediate and long-term postoperative morbidity and evenmortality.6-11

    There is some amount of fear that ABT may beassociated with an increased risk ofpostoperative infections, as well as increasedrecurrence and mortality among cancer patients.It is not easy to compare the postoperativesequelae of patients who were transfused andthose who did not receive blood transfusionbecause of many real and potential confoundingvariables present among the two groups, thewide variations in transfusion practices, and thedifficulty of undertaking a large RCT. Some ofthe difficulties may be reduced by focusing onthe patients who were in fact transfused, andcomparing those who received autologous

  • art screening, it was estimated that 67 to 227persons might acquire transfusion-associatedAIDS yearly.3 This was similar to the risk ofdying from a fatal hemolytic transfusionreaction, which was 30 to 60 deaths yearly outof 4 million blood transfusion recipients.4 TheJoint Committee on Accreditation of HealthCare Organizations required the audit of eachtransfusion event to ensure that transfusionpractice guidelines are implemented. TheAmerican Association of Blood Banks hadrecommended that informed consent beobtained and documented so that the recipientis made fully aware of the potential hazards ofblood transfusion.

    Klein published a review on ABT in 1995which included the estimated risks in theUnited States of some adverse events (Table1).5 Fatal hemolytic reactions

    comparing those who received autologousblood, or leucocyte-depleted RBC (since theimmunomodulatory effect of ABT seems to bewhite cell mediated).

    Duffy and Neal12 (1996) published a meta-analysis of reports on postoperative infectionrates comparing patients who receivedautologous blood compared with allogeneicblood. Out of 9 studies published after 1989, 7met their inclusion criteria, with a total of 1,060patients.13-19 The 5 retrospective studies and 2RCTs did

    PJSS Vol. 55, No. 4, October-December, 2000

    not have any significant statisticalheterogeneity between them and removing anyone study from the meta-analysis changedneither statistical significance nor the directionof the effect. The risk of postoperativeinfection was greater in the allogeneic group,odds ratio 2.37 (95%Cl, 1.6, 3.6, p

  • heterogeneity in the 6 RCTs that reported

    postoperative infections, and when 2 studieswhich differed sufficiently from the rest withregards patient mix and study design wereexcluded from the meta-analysis, the 4 RCTs(N=1,316) showed a risk ratio of 1.00 (95%

  • Meanwhile, the notion of a "universal trigger"was increasingly questioned. The experience ofthe United States military during the KoreanWar revealed that maintaining a patient's Hbconcentration at around 7g/dL was notassociated with serious outcomes as long as theblood volume was adequately restored.50Anecdotal experiences and case reportsinvolving Jehovah's Witnesses added impetusto the re-evaluation of the "10/30 rule".51-54

    At long last, the HIV/AIDS scare in the UnitedStates spurred the National Institutes of Healthto convene a Consensus DevelopmentConference on Perioperative Red CellTransfusion on June 27, 1988, since aroundtwo thirds of all RBC transfusions were thengiven in the perioperative period.

    Among the statements were:1

    1. Available evidence does not support the useof a single criterion such as a hemoglobinconcentration of less than 100 g/dL. No singlemeasure can replace good clinical judgment asthe basis for decision-making regardingperioperative transfusion.

    2. There is no evidence that mild-to-moderateanemia contributes to perioperative morbidity.

    3. Otherwise healthy patients with Hb valuesof 100 g/dL or greater rarely requireperioperative transfusion, whereas those withacute anemia with resulting Hb values of lessthan 70 gm/dL frequently will require redblood cell transfusion.

    number of homologous transfusions should bekept to a minimum.

    8. There are being developed a variety ofpromising alternatives to homologoustransfusion.

    On January 12-14, 1995, a Blood ManagementPractice Guidelines Conference was convened,involving 15 physicians that includedsurgeons, anesthesiologists, intensivists and alawyer, as well as 2 spokespersons for theJehovah's Witness. The recommended BloodManagement Surgical Practice Guidelinesincluded transfusion practice policies, sectionson risks of allogeneic blood transfusion,morbidity risk assessment, alternatives to ABT(erythropoietin, autotransfusion, perioperativedonation, acute normovolemic hemodilution),and specific surgical situations (cardiovascular,orthopedic and urologic surgery).55 The NIHguidelines were reiterated, includingindications: "transfusion should not be used asa primary means of restoring blood volume orsimply to "raise hematocrit" in the absence of aclinically defined need for improved DO2".They recommended a transfusion trigger of 7g/dL in a healthy low risk patient with noevidence of cardiopulmonary disease, and 10g/dL for patients with cardiopulmonarydisease.56 They reiterated the risks for ABT57,and recommended that a "higher thresholdshould be used in patients who have or are atrisk of cardiac or pulmonary disease".58 Unlikethe NIH guidelines, the authors included a fairamount of references, although

    PJSS Vol. 55, No. 4, October-December, 2000

    it was neither mentioned if an electronic searchwas used nor how the evidence was appraised.

    The Canadian Medical Association sponsoredthe most extensive review of the literature onABT. Electronic search of Medline fromJanuary 1966 to July 1996 revealed 189articles (111 observational studies and 78interventional studies). The authors identified 6RCTs (N=813) comparing two transfusionstrategies.60-65 From their systematicevaluation, they were able to draw 19inferences, which were graded according to thestrength of supporting evidence (Table 2).59

    1. Transfusion is rarely indicated when thehemoglobin concentration is greater than 10g/dL and is almost always indicated when it isless than 6 g/dL, especially when the anemia isacute.

    2. The determination of whether intermediatehemoglobin concentrations (6-10 g/dL) justifyor require RBC transfusion should be based onthe patient's risk for complications ofinadequate oxygenation.

    3. The use of a single hemoglobin "trigger" forall patients and other approaches that fail to

  • strength of supporting evidence (Table 2).59Some of the inferences were:

    1. A transfusion threshold of [Hb] 100 g/L isoptimal in high risk patients (RCTs,heterogeneity, CIs all on one side of thresholdNNT)

    2. A transfusion threshold between [Hb] 70 g/Land 80 g/L is optimal in all patients,independent of risk (RCTs, heterogeneity, CIsoverlap threshold NNT).

    The Canadian Medical Society sponsored asecond systematic review, on publishedrecommendations and guidelines on ABT. Atotal of 8,426 titles were identified, of which1,424 were initially deemed relevant, and 42review articles and 17 guidelines were includedin the review. Only 2 articles were consideredas systematic reviews.66,67 There were 8guidelines that fulfilled all aspects of the CPGdefinition. The most rigorously developed

    guidelines were those of the American Societyof Anesthesiologists68, as it was the only onethat specified an electronic search, a literatureselection criteria, and graded the evidence.69

    The American Society of Anesthesiology TaskForce on Blood Component Therapy beganwork in 1994 to develop evidence-basedindications for transfusion in perioperative andperipartum settings. Recommendations werefor typical surgical and obstetric patients, andwere based on scientific evidence and expertopinion on effectiveness of the intervention,which considered clinical benefits, adverseeffects, and costs. Regarding RBC transfusion,5 recommendations were given.

    The task force bases its recommendations onavailable category II-2 and II-3 evidence andexpert opinion. The task force concluded that:

    all patients and other approaches that fail toconsider all important physiologic and surgicalfactors affecting oxygenation are notrecommended.

    4. When appropriate, preoperative autologousblood donation, intraoperative andpostoperative blood recovery, acutenormovolemic hemodilution, and measures todecrease blood loss (deliberate hypotensionand pharmacologic agents) may be beneficial.

    5. The indications for transfusion ofautologous RBCs may be more liberal than forallogeneic RBCs because of the lower (but stillsignificant) risks associated with the former.

    The most current RCT was reported by Hebert,et al. in 1999, of the Transfusion Requirementsin Critical Care Investigators for the CanadianCritical Care Trials Group. They enrolled 838critically ill normovolemic patients who hadhemoglobin concentration of less than 9 g/dLwithin 72 hours after admission to the ICU,and randomly assigned 418 patients to arestrictive transfusion group (R) in which redcells were transfused if the Hb dropped below7 g/dL and were maintained at Hbconcentrations of 7-9 g/dL, and 420 patients toa liberal strategy group (L) in whichtransfusions were given when Hb fell below 10g/dL and were maintained at Hb concentrationsof 10-12 g/dL. The primary diagnosis wererespiratory disease (242,28.9%),cardiovascular disease (170, 20.3%), andtrauma (165, 19.7%), comprising 577 cases(68.9%). Overall, the 30-day mortality wassimilar (R=18.7%, L=23.3%, p=0.11).Mortality rates were significantly lower in theR group among those who were less ill(APACHE II score of less/equal 20), andamong those who were less than 55

    Care of Critically-ill Surgical Patients

    years old. There was no mortality differenceamong patients who had significant cardiacdisease (R=20.5%, L=22.9%, p=0.69) (Table 3).The authors concluded that "A restrictive strategyof red-cell transfusion is at least as effective andpossibly superior to a liberal transfusion strategyin critically ill patients, with the possibleexception of patients with acute myocardial

  • exception of patients with acute myocardialinfarction and unstable angina."70

    Recommendations:

    1. Aggressive and early control of hemorrhageshould be considered an integral part ofresuscitation. (Level II, Category A)

    2. In a patient who is hypovolemic secondary toacute hemorrhage, volume resuscitation shouldbe initiated. (Level I, Category A)

    3. In a normovolemic patient whose level ofanemia is associated with evidence of anincreased risk of major adverse outcomes, themajor indication of RBC transfusion is toincrease oxygen delivery. This should becorrected to achieve a level of anemia whereinthere is evidence that the increased risk is nolonger present.

    The traditional "10/30 rule", or a transfusiontrigger of an Hb concentration of less than 10 g/ dL or a hematocrit concentration of less than20 per cent should no longer be utilized. (Level1, Category A)

    4. There is evidence that there is no benefit ofRBC transfusion for patients whose Hbconcentration is equal or more than 10 g / dL.(Level I, Category A)

    5. In general, there may be an increased risk ofadverse outcomes in normovolemic patientswith acute anemia whose Hb concentration isless than 7 g / dL. (Level I, Category A)

    PJSS Vol. 55, No. 4, October-December, 2000

    6. Normovolemic patients with acute anemiawho have cardiac disease or are at risk ofcardiac disease may benefit from RBCtransfusion when their Hb concentration is lessthan 7 g / dL. Factors that increase the risk ofpostoperative cardiac events, and thecorresponding evidence are contained in PCSEBCPG on Seeking Referral for PreoperativeCardiac Evaluation for Elective NoncardiacSurgery (PJSS 1999; 54 (4): 171 - 233). (Level

    clinical picture of shock (uncompensatedshock) emerges, and it used to be believed thatwhen the clinical picture improved and evenbecame "normal" and "stable" that adequateperfusion had been restored and thatresuscitation was adequate.4 However, aprofusion of reports in the past 2 decades byinvestigators using different tests haddemonstrated that some of those hydratedpatients with an apparently normal clinical

  • Surgery (PJSS 1999; 54 (4): 171 - 233). (LevelI, Category A)

    Patients with acute myocardial infarction orunstable angina may benefit from a transfusiontrigger higher than 7 g / dL. (Level II,Category A)

    7. Packed red blood cells is preferred whenblood transfusion is necessary. (Level II,Category A)

    ASSESSMENT OF RESUSCITATION OFHYPOVOLEMIA

    4. What are the clinical and/or diagnosticindicators that have been shown to be usefulin assessing the adequacy of resuscitation ofhypovolemic critically ill surgical patients?

    Shock may be viewed as "a series of sequelaeof tissue perfusion that is inadequate tomaintain normal metabolic and nutritionalfunctions."1 This recent review by Dabrowski,et al., and those by Henry and Scalea2, andMcGee, et al.3, looked at the publishedevidence on the accuracy and usefulness ofclinical and physiological parameters indetermining the existence of shock, and theability of resuscitation measures to improvetissue perfusion.

    Early on, patients are able to compensate for adecreasing blood volume through a number ofevolutionary responses, and many may appearnormal even though decreased tissue perfusionmay already be present (compensated shock).This compensatory ability varies betweenathletes, healthy but sedentary individuals, andthose with co-morbid conditions particularlycardiac and pulmonary diseases. Whenincreasing volume loss exceeds this initialcompensatory ability, the classical

    patients with an apparently normal clinicalpicture were in fact still hypoperfused (incompensatory shock again), and ifunrecognized and not corrected may notexhibit all the classical clinical picture ofuncompensated shock even when nearingdeath (irreversible shock). This had led someto advocate the almost routine use of certaintests, which are indicators of tissue perfusion,in the assessment of volume resuscitation. Thedifficulties are in the additional costs, theinvasive nature of some of the methods, and inthe result's lag time of some tests. But mostimportantly, many of the evidence used topromote the utility of these tests identifiedphysiological parameters as surrogate ofclinical outcomes, and there is a relativepaucity of Level I evidence on major clinicallyrelevant outcomes. " Given the clear record ofsuccessful treatment of most patients in shockusing nothing more than clinically evidentoutcomes, however, whether one need progressbeyond the use of bedside tests to resuscitatepatients in shock, whatever the cause, remainscontroversial" (Dabrowski, et al.).

    Serum lactate5-7 is not as rapidly available asits surrogate, the arterial blood base deficit(BD) which is readily computed from bloodgas analysis results.8-13.

    Pulmonary artery catheterization (PAC) usingthe Swan-Ganz catheter was introduced in197014, rapidly became widely used inintensive care areas in developed countries,and by 1996 more than 2 million catheterswere being sold worldwide every year.15 Evennow, the PAC stands as a symbol of the state-of-the-art in the monitoring of critically illpatients, highly desirable but unaffordable inmost developing countries. However, because" A possibility exists that the accuracy of dataobtained from the PAC is often variable,potentially inaccurate, and the informationobtained may not be applied appropriately",and the fact that "unambiguous proof thatinformation derived from the PAC improvespatient outcome measures is scarce", led theSociety of

    Care of Critically-ill Surgical Patients

    Critical Care Medicine (SCCM) to host, in1996, a Pulmonary Artery Catheter Consensus

    Further research is needed to define the properrole of the PAC in geriatric patients.

  • 1996, a Pulmonary Artery Catheter ConsensusConference, with participation andendorsement of: American Association ofCritical Care Nurses; American College ofChest Physicians; American Thoracic Society;and European Society of Intensive CareMedicine. The process followed thatrecommended for EBCPGs, including anelectronic search and assigning levels ofevidence. Specific questions were identified,and the multidisciplinary expert panelconvened on December 6-8,1996.16 Some ofthe pertinent recommendations were:

    1. The PAC may be useful in the managementof patients with shock unresponsive to fluidresuscitation and use of vasopressors. Clinicaltrials are needed to determine whethermanagement of various types of shock with thePAC leads to better outcomes thanmanagement using less invasive means ofmonitoring.

    2. Until further data are forthcoming, routineuse of the PAC does not appear to be necessaryin low-risk patients undergoing cardiacsurgery. The PAC may be useful in high-riskpatients undergoing cardiac surgery, especiallyin those patients with clinically important leftventricular dysfunction. Research is needed todetermine whether high-risk patients managedwith the PAC have better outcomes fromcoronary artery bypass grafting than thosepatients managed by less invasive means.

    3. Use of the PAC may lead to fewercomplications in patients undergoingperipheral vascular surgery. Further study ofPAC use is needed in this population toconfirm the reduced rate of complications andto determine the effect of the PAC on otheroutcomes especially mortality.

    4. The PAC may be useful in the managementof some patients undergoing aortic surgery,although recent studies have identifiedpopulations of patients that can be safelymonitored by less invasive means. Furtherstudy is needed to determine whether PAC useimproves patient outcomes in patientsundergoing aortic surgery.

    5. Routine perioperative use of the PAC doesnot appear to be appropriate because of agealone.

    role of the PAC in geriatric patients.

    6. Until data are forthcoming it is not possibleto accurately assess the overall impact of PACuse on complications and mortality in patientsundergoing neurosurgical procedures.However, use of the PAC to monitor and treatair embolism in this group of patients does notappear to be appropriate.

    7. Based on expert opinion, the PAC may behelpful in selected traumatically injuredpatients. Randomized, controlled trials of PAC-guided therapy are needed to validate theindications and effectiveness of PAC use in thetraumatically injured patient.

    8. The PAC may be useful in patients withseptic shock who have not responded to initialaggressive fluid resuscitation and low doseinotropic/vasoconstrictor therapy. Variousmanagement strategies for sepsis and septicshock (intravenous fluids, vasoactivemedications, inotropic medications, etc.)should be evaluated in prospective,randomized, controlled trials. Patient groupsshould be carefully defined, by source ofsepsis, severity of illness and organdysfunction.

    9. PAC-guided hemodynamic intervention to

    augment DO2 to supranormal values inpatients with systemic inflammatory responsesyndrome-related organ dysfunction fromsepsis, trauma, or postoperative complicationsis not recommended. Carefully designed,multicenter, randomized, controlled trials areneeded to establish whether augmenting DO2improves organ-specific outcomes and survivalin patients with systemic inflammatoryresponse syndrome-related organ dysfunctionafter sepsis, trauma, acute inflammation andpostoperative complications.

    10. Further research must be performed beforea recommendation can be made about goal-oriented hemodynamic intervention utilizingthe PAC to augment DO2 to supranormallevels before high-risk surgery. There is a needfor well-designed, multicenter, randomized,controlled trials focusing on whetherprospective

  • PJSS Vol. 55, No. 4, October-December, 2000

    preoperative augmentation of DO2 tosupranormal levels compared with normativevalues in stable ICU patients decreasessubsequent organ dysfunction and death aftersurgical operation in high risk patients and, ifso, by what mechanism(s).

    There are other technologies being investigatedas alternatives or adjuncts to serum lactate,base deficit and PAC (which are globalindicators of perfusion). Since there is a

    preferential shunting of blood flow in thehypovolemic state, away from the viscera andsoft tissues and towards the vital organs,determination of perfusion in these "sacrificialareas" has attracted attention as they may givemore accurate information and may be simplerto use. Gastric tonometry allows the estimationof mucosal PO2 and PCO2 and calculation ofintramucosal pH, wherein a pH lower thannormal (cutoff pH 7.3) is interpreted asinadequate tissue perfusion.17-22 Tissueoximetry had also been investigated as amethod of determining whether measurementsof regional perfusion (subcutaneous tissue,muscle) have advantages over, or maycomplement, global perfusion parameters indetermining the adequacy of resuscitation.23-26More research on the instrumentation andtechniques, and more importantly on whetherthe use of regional perfusion measurementswill actually improve clinical outcomes ofpatients compared to those in whom the testswere not used are still needed. Dabrowski etal.1 summarized their review: "Because no onetest or device can be relied on to identify thepresence of shock in all situations, the best"tool" is a well-trained clinician."

    Recommendations:

    1. The clinical judgment of a well-trained andexperienced surgeon, utilizing clinicalparameters, such as blood pressure, heart rateand urine output should still be the best"monitor" of the adequacy of resuscitation.(Level II, Category A)

    2. At present, there is no evidence that theroutine use of tests such as serum lactatelevels, base deficits and pulmonary arterycatheters (PAC)

    improve major clinical outcomes and thereforeare not recommended. (Level I, Category A)

    3. Indications for the use of tests should bebased on specific circumstances in which thereis strong evidence that their use is associatedwith an improvement in major clinicaloutcomes, such as a decrease in mortality andserious complications. (Level I, Category A)

    POSTOPERATIVE NUTRITIONAL

    SUPPORT

    5. Are there patients who will benefit frompostoperative nutritional support?

    6. What is the most effective way of deliveringpostoperative nutritional support?

    7. What is the most effective composition ofpostoperative nutritional support?

    Dabrowski and Rombeau1 published a reviewwith some very practical clinical aids on thenutritional management of critically ill traumapatients, and gave a relevant definition ofmalnutrition as " a nutritional deficit associatedwith an increased risk of adverse clinicalevents and with a decreased risk of such eventswhen corrected." The catabolic,hypermetabolic response initiated by injury(including surgical procedures), shock andsepsis is a beneficial, evolutionary adaptiveresponse that had enabled our ancestors tosurvive, at least those who had access to fluidsuntil help arrived, and they could begin healingthemselves under the care of family andfriends. Severe malnutrition has long beenrecognized to increase the risk of adversepostoperative sequelae.2-4 More recently, therole of the "gut barrier" and the importance ofearly enteral feeding and maintaining theintegrity of the gastrointestinal mucosa inpreventing or modulating SIRS, MODS, MOFhas been repeatedly emphasized.5-12

    Nevertheless, many surgeons continue tofollow the traditional practice of starting oralalimentation only in the presence of flatus.Since the practice causes delay in startingenteral feeding, particularly among critically illsurgical patients in whom flatus is commonly

  • Care of Critically-ill Surgical Patients

    delayed, total parenteral nutrition (TPN) becamethe most popular method for deliveringnutritional needs for almost 3 decades.

    Heyland, et al.13 (1998) published a meta-analysis on the effects of TPN versus standardcare (usual oral diet plus intravenous glucose)on major clinical outcomes of critically ill adultpatients and adult surgical patients. Theyconducted a computerized search (MEDLINE1980-1998, English) as well as a manual searchfrom review articles. The authors used explicitcriteria for study selection and a methodologicquality assessment scoring system. Theyincluded 26 RCTs (N=2211), and aggregateddata showed that TPN had no effect on mortality(RR 1.03, 95% CI, 0.8,1.31) (Figure 6) andsome tendency to have a lower majorcomplications rate (RR=0.84, 95% CI 0.64,1.09) (Figure 7). They also performed five-subgroup analyses, each based on an a priorihypothesis (Figure 8)14-39

    PJSS Vol. 55, No. 4, October-December, 2000

    1. Regarding premorbid nutritional status, therewere no differences in mortality among eithermalnourished patients (RR=1.13, 95% CI 0.75-

    toward an increase in the rate (RR=2.4, 95%CI 0.88, 6.58), while RCTs on surgical patientsshowed lower rates (RR=0.76, 95% CI 0.48,

  • malnourished patients (RR=1.13, 95% CI 0.75-1.71), or adequately nourished patients(RR=1.00, 95% CI 0.71, 1.39). Majorcomplications were significantly lower amongmalnourished patients who received TPN(RR=0.52, 95% CI 0.30,0.91), while there wasno difference among adequately nourishedpatients (RR=1.02, 95% 0.75, 1.40).

    2. They also compared studies with highermethodological quality scores ((7) to thosewith lower scores. Regarding mortality, RCTswith higher scores showed no effect of TPN(RR=1.17, 95% CI 0.88, 1.56), and a trend tolower mortality in RCTs with lower scores(RR=0.76, 95% CI 0.49, 1.19). Forcomplication rates, RCTs with higher scoresshowed no treatment effects (RR=1.13, 95% CI0.86-1.50), while RCTs with lower scoresshowed a significant reduction (RR=0.54, 95%CI 0.33, 0.87).

    3. They also compared studies published in1988 or earlier with those published later.Regarding mortality, earlier RCTs showed atrend toward a lower rate with TPN (RR=0.70,95% CI, 0.44-1.13), while later RCTs showedno treatment effects (RR=1.18, 95% CI0.89,1.57). For major complications, earlierRCTs showed lower rates with TPN (RR=0.49,95% CI 0.29-0.81), while later RCTs showedno effect (RR=1.19, 95% CI 0.93-1.53).

    4. They compared trials in which the TPNcontained lipids with those that did not. Therewas no difference in mortality in either RCTsthat used lipids (RR=1.03, 95% CI 0.78,1.36),or those that did not (RR=0.98, 95% CI 0.49,1.95). For complications, rates in RCTs thatused lipids showed no effect (RR=0.96, 95%CI 0.69, 1.34), but in RCTs that did not uselipids the rate was significantly lower(RR=0.59, 95% CI 0.38, 0.90).

    5. They compared studies of critically illpatients with those on surgical patients. Formortality, the rate of critically ill patients onTPN were higher (RR=1.78, 95% CI 1.11,2.85), while there was no treatment effect onsurgical patients (RR=0.91, 95% CI 0.69,1.21). For complications, RCTs on critically illpatients showed a trend

    showed lower rates (RR=0.76, 95% CI 0.48,1.0).

    The authors concluded: "Total parenteralnutrition does not influence the overallmortality rate of surgical or critically illpatients. It may reduce the complication rate,especially in malnourished patients, but studyresults are influenced by study population, useof lipids, methodological quality and year ofpublication."

    Moore, et al.40 (1989) published a RCT(N=59) comparing TPN with enteral feeding(TEN) following major abdominal trauma,initiated within 12 hours after operation. TEN(Vivonex TEN) via a needle catheterjejunostomy was well tolerated. Though notstatistically significant (p=0.5644), there was atendency for lower major septic complications(1/29 or 3%) in the TEN group compared to20% (6/30) in the TPN group (RR=0.17, 95%CI 0.02, 1.35).

    Moore, et al.41 then published a meta-analysisof 8 RCTs that compared TPN and TEN(Vivonex TEN by tube/needle catheterjejunostomy) in high-risk surgical patients(N=230). There was negligible heterogeneityacross and within studies. The time to startpostoperative nutritional support was similar(TEN, 32.5 + 1.8 hours; TPN, 32.8 + 1.7hours), and 85 per cent of patients toleratedTEN well. Intention-to-treat analysis revealedhigher septic complications in the TPN group(35%) compared to the TEN group (16%) (p =0.01) (Table 4) among trauma patients. Therewas no difference among nontrauma patients.

    There were also no significant differences in10-day and 30-day mortality rates.42-49

    Meanwhile, more evidence from animal aswell as human studies strengthened the factthat enterocytes derived most of theirnutritional needs from nutrients absorbed fromthe intestinal lumen. Some investigators thenexpanded the concept of TEN to formulas thatincluded specific nutrients that had been shownto improve immune function and would helpmodulate the etiopathogenesis ofSIRS/MODS/MOF now commonly called"immunonutrition". Beale, et al.50 (1999)published a meta-analysis of 12 RCTs(N=1,482) that compared outcomes ofcritically ill patients (medical, surgical,trauma) requiring TEN (nasoentericjejunostomy), wherein one group receivingstandard TEN and another group had TENwith arginine with or without glutamine,

  • Care of Critically-ill Surgical Patients

    nucleotides andomega-3 fattyacids (IMPACT,Immune-aid)and analyzed onan intention-to-treat basis.There was noeffect ofimmunonutritionon mortality(RR=1.05, 95%CI 0.78, 1.41;p=0.76).

    There were significant reductions ininfection rate (RR=0.67, 95% CI 0.1,51;p=0.04), and hospital length of stay (2.9days, 95% CI 1.4, 4.4; p=0.0002) in theimmunonutrition group (Figure 9).51-62

  • PJSS Vol. 55, No. 4, October-December, 2000

    The authors concluded, "The benefits ofimmunonutrition were most pronounced insurgical patients, although they were present in

    all groups. The reduction in hospital length ofstay and infections has resource implications."

    With the increasing evidence that postoperativeTEN did indeed improve major outcomes ofcritically ill surgical patients, the next issue tobe investigated was if the improved outcomescould also be achieved in noncritical patients.Carr, et al.63 (1996) published a small RCT(N=30) on patients who underwent electivegastrointestinal resection for chronicgastrointestinal disease. The treatment group,upon returning from the operating room, weregiven an isovolemic feed (Fresubin) via anasojejunal tube. The control group receivedintravenous fluids and nothing by mouth untilflatus returned. TEN was well tolerated in allpatients who received it, resulting in a meandaily intake of 1622 kcal and positive nitrogenbalance before oral feeding was started. Therewas also no increase in gut mucosalpermeability in the TEN group. There were nosignificant differences in clinical outcomes.

    A larger RCT (N=195) was published byHeslin, et al.64 involving patients whounderwent resection for upper gastrointestinalcancer at the Memorial Sloan-Kettering CancerCenter. The treatment group received TEN viaa jejunostomy tube using an immune-enhancing formula (IEF) (IMPACT) which wasstarted within 24 hours after operation. Thecontrol group received crystalloids until oralfeeding resumed. As expected, all of thephysiological nutritional parameters weresignificantly better in the TEN group. Therewere no differences in clinical outcomes(mortality, morbidity, length of hospital stay),and the authors concluded "Early enteralfeeding with an IEF was not beneficial andshould not be used in a routine fashion aftersurgery for upper GI malignancies." The paperwas presented at the 117th Annual Meeting ofthe American Surgical Association in Quebecand was thoroughly discussed by a panel ofnotables. Important points included: 1) thepatients were generally not malnourished. Thesubset who were malnourished (greater then10% perioperative weight loss) also did notbenefit from IEF; 2) outcomes were also

    reflective of the staff's surgical skills; and 3) itwas a good RCT.

    Recommendations:

    1. Early nutritional support (within 24 hours)of critically ill surgical patients should beinstituted because it decreases the risk ofpostoperative morbidity, particularly septiccomplications and more so in malnourishedpatients (greater than 10% preoperativeweight loss). (Level I, Category A)

    2. Among critically ill surgical patients, thereis evidence that enteral nutrition (EN)delivered into the jejunum and started within24 hours after operation results in lower septiccomplications compared to total parenteralnutrition (TPN). TPN is used only when theenteral route is not feasible. (Level I, CategoryA)

    The clinical trials that had demonstratedbenefits of early TEN had used commercialTEN formulations.

    3. TEN using commercial immune-enhancingformulas may be associated with a greaterdecrease in the risk of septic complicationscompared to standard TEN formulae inselected critically ill patients. (Level I,Category A)

    4. The routine use of postoperative enteralnutrition in noncritical, nonmalnourishedpatients is not recommended.(Level I, CategoryA)

    PHARMACOLOGICCARDIOVASCULAR SUPPORT

    8. When is pharmacological cardiovascularsupport beneficial among critically illsurgical patients?

    Cardiovascular failure is the inability of theheart to provide sufficient blood to meet themetabolic demands of the body, resulting inlow flow, hypoxia and metabolic acidosis. Thisfailure as a single cause of mortality in thepost-injury period is unusual because mosttrauma patients are young. Cardiovascularcollapse is a component of the multiorganfailure syndrome and is most commonly due toinadequate resuscitation.

  • Care of Critically-ill Surgical Patients

    Reducing the risk of cardiovascular failuremust start with the basic principles of traumacare, including adequate treatment ofhypovolemic shock. Treatment principlesshould emphasize an overall concern forphysiologic responses rather than just treatingthe heart as an isolated organ.

    Understanding the Frank-Starling Law iscrucial in managing cardiovascular failure incritically ill surgical patients. This lawdemonstrates that cardiac output is directlyproportional to filling pressures. As pressureincreases, myocardial fiber length increases,producing increased tension and greatercontractile force.1

    The most important concept of support forcardiovascular function is that fluids should begiven in the appropriate volume. Vasopressorshave no role in the initial treatment ofhypovolemic shock. Studies have shownbenefit from vasopressors only after volumehas been restored. Furthermore, theadministration of vasopressors duringhypovolemia reduces the already depletedplasma volume.2,3

    Central venous pressure (CVP) monitoringmay be used as a measure of right sided fillingpressure or right sided preload. CVPmonitoring provides data for appropriate fluidmanagement especially in young patients withnormal cardiac function.4,5

    Central venous pressure reading however isaffected by blood volume, distensibility andcontractility of the right heart, venomotor toneand intrathoracic pressure. The measured CVPis the result of all these factors and does notalways reflect the adequacy of the circulatingblood volume or the competence of the rightand left ventricle.6

    Trends and pressure response to fluidadministration are much more important thanthe absolute value. In general, a low CVPsuggests that hypovolemia may exist and thatfluid challenge is not likely to be harmful. Therate of infusion should be gauged by CVPresponse. A rapid rise in pressure suggests anadequate blood volume or poor rightventricular reserve. A minimal responsesuggests hypovolemia. CVP monitoring can be

    The clinical challenge of pulmonary arterycatheter use is to properly interpret theinformation provided. It should be used withan understanding of its assumptions andlimitations.7

    Pulmonary artery pressure monitoring may beuseful in selected critically ill trauma patientswhere tissue perfusion is less than optimal andthe balance between intravascular volume andcardiac competence is not known.8 Theappropriate titration of volume (preload),inotropic support and afterload agents can beeffectively controlled with these hemodynamicmeasurements. Frequent changes in amountand type of treatment is possible if pulmonarypressure readings are available.

    Quinn and Quebbman used PAC for unknownfluid status in 43 patients , for hypotension in28 patients and for sepsis in 13 patients.Therapy was altered by pulmonary arterypressure and cardiac output measurements in30 per cent of these patients. Eight patientswere thought to have cardiac failure butactually required volume replacement, whereas7 patients thought to have volume depletionhad volume overload.9

    Under optimal conditions, a ventricularresponse curve can be generated by measuringthe cardiac output sequentially as the volume isincreased. When the cardiac output no longerincreases or begins to decrease, optimal

    volume loading has been attained.

    When the PCWP (pulmonary capillary wedgepressure ) is at 12-18 and the cardiac output isstill low, the next step is to improve cardiacoutput by giving inotropic agents. Unless theperipheral vascular resistance is greater than1800 dynes/sec/cm2, no afterload unloadingagents are given.10 On the other hand whensystemic vascular resistance is low,vasopressors are indicated.

    The choice of inotropic agents should beindividualized to each patient and hishemodynamic status at a given time. Thepatient should be frequently examined andpharmacologic drugs titrated based on thehemodynamics and the overall clinical pictureof the patient.

  • suggests hypovolemia. CVP monitoring can beuseful in many patients if its limitations arerecognized. However, in severly ill surgicalpatients where the CVP does not reflectchanges in blood volume, Swan Ganzpulmonary artery catheter placement may beperformed.

    A lot of controversy has been generated by theappropriate use of pulmonary artery catheters(PAC).

    of the patient.

    There has been a controversy whether"supranormal" delivery of oxygen bysupranormal cardiac index resuscitation goalsimproves outcome in patients with trauma,septic shock as well as in heterogeneousgroups of critically ill patients. This topic willbe discussed in the next section.

    PJSS Vol. 55, No. 4, October-December, 2000

    9. What is the appropriate method ofpharmacological cardiovascular support ofthe critically ill surgical patient?

    In the Philippines, the adrenergic agonistsdopamine, dobutamine and epinephrine are

    widely utilized for pharmacologiccardiovascular support in critically ill patients,while norepinephrine is only available in somehospitals. More recently, milrinone, aphosphodiesterase III inhibitor with inotropicand vasodilatory properties has been releasedin the country.

    When there is clear evidence of cardiogenicshock based on preexisting or complicatingacute, chronic or mixed acute and chroniccardiac disease in a critically ill surgicalpatient (e.g. postoperative acute myocardialinfarction with pulmonary congestion andhypotension after aortic surgery), theseinotropic agents generally in combination withdiuretics with or without vasodilators areappropriate initial therapy. Further discussionof the recognition and management of thisclinical situation is beyond the scope of thisguideline.

    In the setting of shock with relative or absolutehypovolemia in the critically ill surgicalpatient, as occurs in hemorrhagic conditions,third-spacing, trauma, sepsis or SIRS,vasoactive agents may favorably affecthemodynamic and oxygen transport and tissueutilization when volume resuscitation isadequate or at least ongoing. Vasopressors in avolume-depleted patient may cause cardiacdecompensation and hemodynamicdeterioration, particularly in patients withischemic heart disease and therefore should not

    on the one hand, or noncompensatorytachycardia on the other hand, needs to havethe heart rate problem appropriately treated aswell. A full vagolytic dose of atropine, anddopamine and/or epinephrine drips areappropriate agents in the setting ofbradycardia-related hypotension.

    In a volume-replete patient, persistent mildhypotension (systolic blood pressure less than90 but not less than 70 mmHg) and shockindependent of heart rate problems shouldgenerally be treated with dobutamine (with orwithout dopamine) in an attempt to increasethe cardiac index and oxygen delivery. In thelower range of systolic blood pressure (70-80mmHg) and especially if there are moreprofound signs and symptoms of shock,dopamine should be started first, anddobutamine then added when the bloodpressure improves. Dopamine and/ornorepinephrine should generally be used forsevere hypotension with systolic bloodpressure less than 70 mmHg.13 Milrinone hasbeen shown to increase cardiac index andoxygen delivery in critically-ill patients,14 andmay be regarded as a therapeutic alternative todobutamine, or when dobutamine fails. Special

  • ischemic heart disease and therefore should notbe administered prior to fluid replacement.13

    The choice of vasoactive agent (Table 5)depends on the blood pressure, cardiac index,heart rate and systemic vascular resistance, andit may be appropriate to change from one drugto another or combine drugs as thehemodynamic parameters and clinical situationchanges. Close patient monitoring cannot beoveremphasized. Consideration of intraarterialinvasive monitoring of blood pressure in apatient with extreme peripheralvasoconstriction and cool extremities, andinsertion of a central venous or pulmonaryarterial catheter in this situation have beendiscussed above.

    With simultaneous volume replacement, ahypotensive patient with absolute or relativebradycardia

    situations involving unique pathophysiologicmechanisms of cardiovascular collapse mayrequire particular vasoactive agents in additionto volume replacement, such as dopamine forspinal shock; epinephrine for anaphylacticshock, or to counteract the excessive effects ofbeta-blocker or alpha-blocker drugs; andnorepinephrine to offset the untoward effectsof alpha-blocker drugs.13 Phenylephrine,isoproterenol, glucagons, and recentlyvasopressin24 have been reportedly useful forspecial

    Care of Critically-ill Surgical Patients

    cardiovascular collapse situations, but arecurrently unavailable in our setting. The end-effect of all these drugs on survival remainsunclear however at this time.

    While the pharmacologic approach outlinedabove is generally accepted by mostcardiologists and intensivists, it is remarkablethat there are only a limited number of reportedrandomized trials involving the use of thesevasoactive agents, primarily dobutamine, toaugment hemodynamic and oxygen transportparameters in critically ill patients. There arealso few case-control trials in this regard. Therandomized trials that have looked at the mostimportant outcomes of in-hospital morbidityand mortality focus on alternative resuscitationstrategies, i.e. achieving supranormal targetlevels of cardiac index and oxygen transportvs. conventional hemodynamic targets,especially normal cardiac index and bloodpressure. Because of its hemodynamic effectson cardiac index and oxygen delivery,15 moststudies largely utilized dobutamine as primaryvasoactive drug, on top of adequate fluidresuscitation and other add-on vasopressoragents. Shoemaker et al. earlier observed thatsurvivors of high-risk surgical operations hadsignificantly higher mean values of cardiacindex (CI), oxygen delivery (DO2) and oxygen

    r= 0.94 (p=0.016).23 Beyond fluidreplacement, the contribution of vasopressortherapy to achievement of goals and its impacton mortality is less clear. It is possible that theability to achieve desired levels of cardiacoutput, oxygen delivery and consumption(regardless of therapy employed) identifies apopulation with a larger physiologic reserve,less severe illness, and consequently a betterprognosis.15 It remains unproved thatsupranormal hemodynamic values should serveas endpoints of resuscitation.

    The tabulated studies also reported a variableeffect of supranormal goals of resuscitation onsecondary clinical outcomes. Compared tocontrols, the treatment group had reducedcomplications,16,17,11 length of ICU stay;16,23and length of hospitalization and hospitalcosts.16 Other studies, however, reported nodifference in these secondary outcomes ofinterest.18,19 One other randomized studyinvolving critically ill surgical patients foundthat the frequency of myocardial infarction wasnot higher among patients in whomsupranormal values were achieved withinotropic therapy, although analysis was notperformed according to the "intent to treat"principle.20 Another randomized trial onpostoperative cardiac surgical patients found

  • consumption (VO2) than nonsurvivors.16 Sincethese supranormal hemodynamic and

    metabolic patterns might represent circulatorycompensations for increased postoperativemetabolism necessary for survival, it washypothesized that supranormal CI, DO2 andVO2 values as resuscitation goals mightimprove the outcome in critically ill patients.The randomized trials testing this hypothesisamong various surgical patient populations,tabulated in Table 6, have had conflictingresults. Smaller studies have shown a favorableeffect of supranormal goals on mortality16,17but many other trials and the largestrandomized study reported a negative effect.18-20,23 One trial reported excess mortality withsupranormal resuscitation goals.12 Except forthe study by Gattinoni et al20, a moreconsistent finding in secondary analysesamong the various studies was that survivalwas significantly improved among patientswho reached supranormal values of DO2whether treated or self-generated as comparedwith patients who reached only normal DO2values.15,18,19,23 The correlation betweenpostresuscitation oxygen delivery andmortality was a strongly inverse one with acorrelation coefficient of

    postoperative cardiac surgical patients foundthat aiming for supranormal oxygen delivery to

    achieve normal SvO2 values and lactateconcentration during the immediatepostoperative period after cardiac surgery canshorten the length of hospital stay.22

    There are no randomized trials directlycomparing the effects of various vasoactiveagents on mortality in critically ill surgicalpatients. Furthermore, while the hemodynamiceffects of various vasoactive agents have beenwell characterized in short-term controlledclinical studies, not one randomized trialcomparing the effect of vasoactive drugs vs.placebo comprising volume resuscitation aloneon mortality and morbid outcomes wasretrieved from an exhaustive literature search.There may actually be some difficulties orlimitations compromising optimal design ofclinical trials in this area, including aheterogeneous patient population comprisingcritically ill surgical patients - nature ofprimary illness (e.g. trauma, sepsis, electivesurgery), type of surgery, timing ofpresentation; coexisting medical illnesses andcomplications; variable appropriate timing ofaggressive intervention - before, during or aftersurgery; and ethical issues regardingwithholding of conventional treatment.

    PJSS Vol. 55, No. 4, October-December, 2000

  • Recommendations:

    1. When volume resuscitation is consideredadequate based on clinical judgment and theclinical response is suboptimal, vasoactivedrugs may improve hemodynamics, oxygentransport and clinical condition. (Level I,Category A)

    2. Hemodynamic monitoring using apulmonary artery catheter (PAC) may improveclinical decision making in selected criticallyill patients. (Level II, Category A)

    3. When vasoactive drugs are used, the goalsof therapy remain normalization ofhemodynamics, oxygen transport, andimprovement in the clinical manifestations ofshock, but not supranormal resuscitationendpoints. (Level I, Category A)

    4. After restoring volume, clinical judgmentshould be exercised in the particular choice ofa vasoactive drug, with consideration ofhemodynamic parameters including heart rate,blood pressure, cardiac index, volume statusand cardiac filling pressures, vascularresistance,

    Care of Critically-ill Surgical Patients

    and preexisting cardiac and medical status ofan individual patient.

    4a. Dobutamine is the drug of choice toincrease cardiac index and oxygen delivery ifblood pressure is acceptable but there are stillsigns of inadequate tissue perfusion. (Level II,Category A)

    4b. Dopamine should be used initially forpersistent mild hypotension > 70 and < 90mmHg and shock to support blood pressure;and dobutamine may then be added to optimizecardiac index and oxygen delivery. (Level II,Category A)

    4c. Dopamine and/or norepinephrine shouldgenerally be used for severe hypotension withsystolic blood pressure

  • and abnormal cardiac rhythm should beaddressed by appropriate pharmacologictherapy and/or adjunctive devices (such astemporary pacer, external cardioverter-defibrillator). (Level II, Category A)

    Technical Working Group

    Group I Armand C. Crisostomo (generalsurgeon) Arturo S. De la Pea (generalsurgeon) and M. Francisco T. Roxas (generalsurgeon)

    Group II Narciso S. Navarro Jr. (generalsurgeon), Ma. Luisa D. Aquino (generalsurgeon) and Adriano V. Laudico (generalsurgeon)

    Group III Eduardo R. Bautista (thoracic andcardiovascular surgeon), Abundio A. Balgos(pulmonologist) and Camilo C. Roa(pulmonologist)

    Group IV Eduardo R. Bautista (thoracic andcardiovascular surgeon) and Donato R.Maraon (cardiologist)Research Assistant: Jaimelly P. Buenbrazo

    Panel of Experts

    1. Joel U. Macalino (general surgeon, PCS)

    2. Nestor E.dela Cruz (general surgeon, PCS)

    Philippine College of Physicians)

    12. Ramon S. Inso (general surgeon, SurgicalOncology Society of the Philippines)

    13. Teresita S. de Guia (pulmunologist,Philippine College of Chest Physicians)

    14. Myrna N. Baares (pulmunologist,Philippine College of Chest Physicians)

    15. Edgar A. Baltazar (general surgeon,Philippine Society of General Surgeons)

    16. Edgardo F. Fernandez (general surgeon,Philippine Society of General Surgeons)

    17. Guillermo B. Barroa Jr. ( thoracic andcardiovascular surgeon, Philippine Associationof Thoracic and Cardiovascular Surgery, Inc.)

    18. Teodoro G. Herbosa (general surgeon,Philippine Society for the Surgery of Trauma)

    19. Eric A. Tayag (epidemiologist, Departmentof Health)RECOMMENDATIONS

    Fluids for Resuscitation

    1. Crystalloids are the appropriate intravenousfluids to be used in the resuscitation ofhypovolemic surgical patients. The use ofcolloids does not decrease mortality, and mayresult in higher mortality in trauma patients.Albumin may increase the risk of death inpatients with hypovolemia, burns andhypoalbuminemia. ( Level 1, Category A)

    Red Blood Cell Transfusion

    1. Aggressive and early control of hemorrhageshould be considered an integral part ofresuscitation. (Level II, Category A)

    PJSS Vol. 55, No. 4, October-December, 2000

    2. In a patient who is hypovolemic secondaryto acute hemorrhage, volume resuscitationshould be initiated. (Level I, Category A)

    3. In a normovolemic patient whose level ofanemia is associated with evidence of anincreased risk of major adverse outcomes, the

    should still be the best "monitor" of theadequacy of resuscitation. (Level II, CategoryA)

    2. At present, there is no evidence that theroutine use of tests such as serum lactatelevels, base deficits, and pulmonary artery

  • major indication of RBC transfusion is toincrease oxygen delivery. This should becorrected to achieve a level of anemia whereinthere is evidence that the increased risk is nolonger present.

    The traditional "10/30 rule", or a transfusiontrigger of an Hb concentration of less than 10g / dL or a hematocrit concentration of lessthan 30 per cent should no longer be utilized.

    (Level 1, Category A)

    4. There is evidence that there is no benefit ofRBC transfusion for patients whose Hbconcentration is equal or more than 10 g / dL.(Level I, Category A)

    5. In general, there may be an increased risk ofadverse outcomes in normovolemic patientswith acute anemia whose Hb concentration isless than 7 g / dL. (Level I, Category A)

    6. Normovolemic patients with acute anemiawho have cardiac disease or are at risk ofcardiac disease may benefit from RBCtransfusion when their Hb concentration is lessthan 7 g / dL. Factors that increase the risk ofpostoperative cardiac events, and thecorresponding evidence are contained in PCSEBCPG on Seeking Referral for PreoperativeCardiac Evaluation for Elective NoncardiacSurgery (PJSS 1999; 54 (4): 171 - 233). (LevelI, Category A)

    Patients with acute myocardial infarction orunstable angina may benefit from a transfusiontrigger higher than 7 g / dL. (Level II,Category A)

    7. Packed red blood cells is preferred whenblood transfusion is necessary. (Level II,Category A)

    Assessment of Resuscitation of Hypovolemia

    1. The clinical judgment of a well-trained andexperienced surgeon, utilizing clinicalparameters, such as blood pressure, heart rateand urine output

    catheters (PAC) improve major clinicaloutcomes and therefore are not recommended.(Level I, Category A)

    3. Indications for the use of tests should bebased on specific circumstances in which thereis strong evidence that their use is associatedwith an improvement in major clinicaloutcomes, such as a decrease in mortality andserious complications. (Level I, Category A)

    Postoperative Nutritional Support

    1. Early nutritional support (within 24 hours)of critically ill surgical patients should beinstituted because it decreases the risk ofpostoperative morbidity, particularly septiccomplications, and more so in malnourishedpatients (greater than 10% preoperativeweight loss). (Level I, Category A)

    2. Among critically ill surgical patients, thereis evidence that enteral nutrition (EN)delivered into the jejunum and started within24 hours after operation results in lower septiccomplications compared to total parenteralnutrition (TPN). TPN is used only when theenteral route is not feasible. (Level I, CategoryA)

    The clinical trials that had demonstratedbenefits of early TEN had used commercialTEN formulations.

    3. TEN using commercial immune-enhancingformulae may be associated with a greaterdecrease in the risk of septic complicationscompared to standard TEN formulae inselected critically ill patients. (Level I,Category A)

    4. The routine use of postoperative enteralnutrition (EN) in noncritical, nonmalnourishedpatients is not recommended.(Level I, CategoryA)

    Pharmacologic Cardiovascular Support

    1. When volume resuscitation is consideredadequate based on clinical judgment and theclinical response

    Care of Critically-ill Surgical Patients

    is suboptimal, vasoactive drugs may improve

  • is suboptimal, vasoactive drugs may improvehemodynamics, oxygen transport and clinicalcondition. (Level I, Category A)

    2. Hemodynamic monitoring using apulmonary artery catheter (PAC) may improveclinical decision making in selected criticallyill patients. (Level II, Category A)

    3. When vasoactive drugs are used, the goalsof therapy remain normalization ofhemodynamics, oxygen transport, andimprovement in the clinical manifestations ofshock, but not supranormal resuscitationendpoints. (Level I, Category A)

    4. After restoring volume, clinical judgmentshould be exercised in the particular choice ofa vasoactive drug, with consideration ofhemodynamic parameters including heart rate,blood pressure, cardiac index, volume statusand cardiac filling pressures, vascularresistance, and preexisting cardiac andmedical status of an individual patient.

    4a. Dobutamine is the drug of choice toincrease cardiac index and oxygen delivery ifblood pressure is acceptable but there are stillsigns of inadequate tissue perfusion. (Level II,Category A)

    4b. Dopamine should be used initially forpersistent mild hypotension > 70 and < 90mmHg and shock to support blood pressure;and dobutamine may then be added to optimizecardiac index and oxygen delivery. (Level II,Category A)

    4c. Dopamine and/or norepinephrine shouldgenerally be used for severe hypotension withsystolic blood pressure

  • b. Grundmann R, Meyer H. The significance of colloid osmotic pressuremeasurement after crystalloid and colloid infusions. Intensive Care Med1982;8(4):179-186.

    20. a. Hall K, Sorensen B. The treatment of burn shock. Scand J PlastReconstr Surg 1973;7:67-73.

    b. Hall K, Sorensen B. The treatment of burns shock. In: Vrabec R,Konickova L, Moserova J, editor(s). Basic problems in burns. Berlin:Springer-Verlag, 1975.

    c. Hall KV , Sorensen B. The treatment of burn shock : results of a 5-yearrandomized, controlled clinical trial of Dextran 70 vs. Ringer lactatesolution. Burns 1978;5(1):107-112.

    21. Hartmann M, Jonsson K, Zederfeldt B. Effects of dextran andcrystalloids on subcutaneous oxygen tension and collagen accumulation. Arandomized study in surgical patients. Eur Surg Res 1993;25:270-277.

    22. a. Jelenko C 3rd. Fluid therapy and the HALFD method. J Trauma1979;19 (11 Suppl):866-867.

    b. Jelenko C 3rd, Solenberger RI, Wheeler ML, Callaway BD. Shock andresuscitation. III. Accurate refractometric COP determinations inhypovolemia treated with HALFD. JACEP 1979;8(7):253-256.

    c. Jelenko C 3rd, Wheeler ML, Callaway BD, Divilio LT, BucklenKR,Holdredge TD. Shock and resuscitation. II: Volume repletion withminimal edema using the HALFD (Hypertonic Albuminated FluidDemand) regimen. JACEP 1978; 7(9):326-333.

    d. Jelenko C 3rd, Williams JB, Wheeler ML, et al. Studies in shock andresuscitation, I: use of a hypertonic, albumin-containing, fluid demandregimen (HALFD) in resuscitation. Crit Care Med 1979;7(4):157-167.

    23. Karanko M, Klossner J, Laksonen V. Restoration of volume bycrystalloid versus colloid after coronary artery bypass: haemodynamics,lung water, oxygenation and outcome. Crit Care Med 1987;15:559-566.

    24. Ley SJ, Miller K, Skov P. Crystalloid versus colloid fluid therapy aftercardiac surgery. Clin Studies Cardiac Care 1990;19(1):31-40.

    25. a. Lowe RJ, Moss GS, Jilek J, Levine HD. Crystalloid vs. colloid in theetiology of pulmonary failure after trauma: a randomized trial in man.Surgery 1977;1(6):676-683.

    b. Lowe RJ, Moss GS, Ji