Effect of warm heart surgery on perioperative management of patients undergoing urgent cardiac surgery

ORIGINAL ARTICLES

Effect of Warm Heart Surgery Undergoing

on Perioperative Urgent Cardiac

Management Surgery

of Patients

Brian P. Kavanagh, MB, BCh, BSc, MRCPI, C. David Mazer, MD, FRCPC, Anthony Panos, MD, MSc, FRCSC, and Samuel V. Lichtenstein, MD, PhD, FRCSC The anesthetic management and outcome data were examined in a retrospective case-controlled study that compared a conventional hypothermic cardioplegic technique with the recently described method of warm heart surgery, in patients undergoing urgent cardiac surgery. Hypothermic continuous oxygenated blood crystalloid cardioplegia with systemic hypothermia was used for 37 patients who underwent cardiac surgery by the same surgeon over a 16-month period from July 1988 (group 1). whereas normothermic continuous oxygenated blood crystalloid cardioplegia with systemic normothermia was used on 56 patients over the following 16.month period until March 1990 (group 2). The groups were similar in terms of age, sex, ASA status, NYHA classification, and preoperative left ventricular function. Defibrillation following cardiopulmonary bypass was required in only 3.8% of the warm heart surgery patients (group 2) compared with 83.8% in group 1 (P < 0.0001). and use of warm heart surgery (group 2) eliminated the need for rewarming. There was a

C vide

ARDIOPLEGIA WAS originally developed to proa still operating field, to allow intracardiac procedures under direct vision, and to protect the myocardium during periods of arrest and cessation of coronary blood flow.’ Whether continuous or intermittent blood or crystalloid cardioplegia is superior is a matter of controversy.2-5Although there are several conventionally accepted components to a successful cardioplegic technique, including the presence of potassium and magnesium, the single most important component is thought to be hypothermia.6-‘0 Lichtenstein et al developed the concept of continuous normothermic blood cardioplegia on the basis that the metabolic demands of the normothermic arrested heart are exceedingly low, and are only minimally further decreased with profound hypothermia.“~” These demands could easily be supplied by continuous perfusion of a normothermic blood-crystalloid solution rather than a cold perfusate, particularly in terms of oxygen delivery to the myocardium.“,” Previously published data point to several advantages of continuous normothermic, compared with continuous hypothermic, blood-crystalloid cardioplegia.” The data were derived from a prospective analysis of consecutive patients who underwent a coronary artery bypass graft (CABG), and suggested that there were beneficial effects in terms of reduction in perioperative myocardial infarction, low output syndrome, increased cardiac output (CO) following cardiopulmonary bypass (CPB), and reduced reperfusion

Journal of Cardiothoracic and VascularAnesthesia,

trend towards a reduced incidence of myocardial infarction (19% in group 1 vs 9% in group 2). low cardiac output syndrome (40% vs 29%). and use of the intraaotiic balloon pump (16% vs 9%) in warm heart surgery patients, but these differences did not reach statistical significance. There were no differences between the two groups in terms of anesthetic drug usage, total heparin or protamine doses, blood loss, transfusion requirements, or duration of ICU stay. These results suggest that: (1) hypothermia is not an absolute requirement for myocardial protection; (2) warm cardioplegia is a useful and safe technique in high-risk patients undergoing urgent cardiac surgery; (3) warm cardioplegia is associated with a reduced requirement for postbypass defibrillation; and (4) introduction into clinical practice of warm heart surgery resulted in no significant changes in perioperative management of patients undergoing urgent cardiac surgery. Copyright o 1992 by W.B. Saunders Company

time following aortic cross-clamp removal. A major advantage of the method may be that greatly prolonged crossclamp times are possible, thus allowing for more complex or longer surgical procedures.‘3 Although these initial reports suggest improved surgical outcome, the anesthetic implications of normothermic blood cardioplegia have not been studied. To determine the effect of the introduction into clinical practice of warm heart surgery on perioperative anesthetic management, patients undergoing urgent CPB using continuous normothermic blood-crystalloid cardioplegia with systemic normothermia were compared with a similar group receiving cold blood-crystalloid cardioplegia and systemic hypothermia.

This article is accompanied by an editorial. Please see: Ix11 WA: Hot or Cold, Continuous or Intermittent? What Goes Around Comes Around. J Cardiothorac Vast Anesth 6:125-126,1992

From the Departments of Anesthesia and Surgery, Universi~ of Toronto, St Michael’s Hospital, Toronto, Ontario, Canada. Address reprint requests to C. David Mazer, MD, FRCPC, Department of Anesthesia, University of Toronto, St Michael’s Hospital, 30 Bond St, Toronto, Ontario, MSB I W8 Canada. Copyright 0 1992 by W.B. Saunders Company X053-0770192/0602-0002$0.00/0

Vol6, No 2 (April), 1992: pp 127-131

127

KAVANAtiH

128

MATERIALS

AND METHODS

The perioperativc records of 93 patients who underwent urgent or emergency cardiac surgery by the same surgeon over a 3-year period were reviewed. Group 1 consisted of 37 patients from July I986 to November 19X7, all of whom received continuous hypothermic blood-crystalloid cardioplegia. All the patients in group 2 (n = 56) were operated on between November 1987 and March 1990, and all received continuous normothermic blood-crystalloid cardioplegia. All patients were ASA class IV or V, and were seen in preanesthetic consultation by a staff anesthesiologist. Patients were premeditated with narcotics and benzodiazepines as appropriate. Intraoperative monitoring consisted of a two-lead (II and V,) electrocardiogram (ECG), pulse oximeter, capnograph, and transduced intra-arterial and pulmonary artery pressure (PAP). Anesthesia was induced with high-dose fentanyl and pancuronium at dosage regimens determined by the attending anesthesiologist, and supplemented with narcotics, benzodiazepines, and/or low concentrations of isoflurane or enflurane. CO was measured at regular intervals using the thermodilution technique with rapid injection of room temperature normal saline. All hemodynamic data were recorded immediately pre-CPB and post-CPB. Patients were anticoagulated with heparin, 390 IUikg, before CPB plus supplemental doses as required to maintain the activated clotting time (ACT) of more than 400 seconds during CPB. In addition, the bypass circuit prime contained 5,000 IU of heparin in both groups. During CPB isoflurane, nitroglycerin, sodium nitroprusside, and/or phenylephrine were used to maintain the mean arterial pressure in the range of 50 to 90 mmHg. The following data were also recorded: serial ACT. blood loss, transfusion requirements, fluid balance, urinary output, aortic cross-clamp time, bypass time, core temperature. and dosages of inotropes, heparin. and protamine sulphate. In all cases, cardiac arrest was induced by infusion of a high potassium (K = 20.0 mmol/L) oxygenated blood-crystalloid (4:l) solution into the aortic root, proximal to the aortic cross-clamp. The cardioplegic solution was infused in all cases at a rate of 300 mL/min for a period of 5 minutes. In group I, the cardioplegia temperature was 10°C; iced saline was inserted into the pericardial well to maintain the myocardial temperature at approximately 15”C, with systemic hypothermia to 27°C to 32°C. In group 2, the cardioplegia temperature was 37°C; body temperature was maintained at 33°C to 37°C. Cardiac arrest was invariably achieved in less than I minute. Following perfusion with the high potassium solution for 5 minutes, the perfusate was switched to a low potassium (K = 6.0 mmol/L) oxygenated blood-crystalloid solution, which was identical in all other respects to the solutions previously described. These cardioplegic solutions, hypothermic in group 1 and normothermic in group 2, were continued at an infusion rate of 50 to 150 mL/min. If electrical activity was detected on the ECG, the high potassium solution was temporarily infused until asystole was again achieved. Following discontinuation of CPB, circulatory support with either inotropic drugs or intra-aortic balloon pump (IABP) counterpulsation was provided if necessary, and patients were transferred to the cardiovascular intensive care unit (ICU). Immediately following admission to the ICU, hemoglobin and serum potassium were measured, and hemodynamic evaluation (arterial blood pressure, central venous pressure, PAP, pulmonary capillary wedge pressure [PCWP], CO) was undertaken. Temperature was recorded, as was the time taken to reach the ICU following discontinuation of CPB. The chest tube drainage. fluid balance, and requirement for blood products were recorded for 12 hours following ICU admission. The low CO syndrome (LCOS) was defined as the need for cardiovascular support with either ino-

t’ ~1.

tropeh or lAHP to mainlain ay\tolic blood pl-esaure abo\c Yll mm Hg, in association with a cardiac index less than 2 L:min/ m head I rale greater than 80 beats/min. and PCWP greater than I8 mmfly. Perioperalive myocardial infarction was defined as the appearance of new Q waves on the ECG and CK-MB greater than 30. Duration of mechanical ventilation and ICI1 stay was recorded. as w
The age, sex, ASA status, NYHA classification, and preoperative left ventricular function were similar in both groups (Table 1). In group 1.30 patients underwent CABG, 3 had CABG combined with another cardiac procedure, and 4 had valvular procedures alone. In group 2,50 patients had CABG alone, 4 had combined CABG and valve replacement, and 2 had only valve replacement. The same anesthetic technique (fentanyl/diazepam) was used for both groups. The total intraoperative dose of fentanyl was 5.2 t 0.3 mg in group 1 and 5.4 k 0.2 mg in group 2, whereas the total diazepam dose was 15.3 k 1.8 mg in group 1 and 14.5 t 1.5 mg in group 2 (Table 2). Supplementation with an inhaled anesthetic at any time intraoperatively occurred in 56.8% of patients in group 1 and in 57.1% of patients in group 2. Similarly, there was no difference in the number of patients receiving supplemental doses of fentanyl or benzodiazepines while on CPB. There was also no difference in the total doses of heparin or protaminc given to both groups. Although the percent of patients in either group (72% warm, 52% cold) receiving supplemental heparin during CPB was not significantly different, the total dosage of supplemental heparin given during bypass was higher in the warm group (10,960 k 1,309 IU vs 6,310 + 1,461 IU, P = 0.04). Intraoperative data analysis (Table 3) showed that 31 patients (83.8%) in group 1 required defibrillation following CPB compared with only 2 patients (3.6%) in group 2 (P < 0.0001). In addition, there seemed to be a trend towards a lower incidence of pcrioperative myocardial Table 1. Demographic Data

Age (mean +- SEMI Male (%l Female (%) ASA Class IV (%) ASA Class V (%) NYHA Class II (%) NYHA Class III(%) NYHA Class IV (%) Pre-OP LV Grade I/IV (%) Pre-OP LV Grade II/IV (%) Pre-OP LV Grade III/IV(%l Pre-OP LV Grade IV/IV (O/o) Abbreviations:

Pre-OP. preoperative;

Group 1

Group

(ll = 37)

(n = 56)

61.1 i; 1.7 30 (81) 7 (19) 34 (92)

62.6 + 1.4 46 (82) 10 118) 54 (96)

3 (8) 0 (01 5 (13.5) 32 (86.5) 2 (5.4) 13 (35.1) 13 (35.1) 9 (24.3)

2 (41 l(l.8) 7 (12.71 47 (85.5) 5 (9.11 12 (21.8) 19 (34.6) 19 (34.6)

LV, left ventricular

2

ANESTHESIA

129

AND WARM HEART SURGERY

Table 4. ICU Data

Table 2. Anesthesia Data

Mean 2 SEM Fentanyl (mg - total dose)

Group 1

Group

(n = 37)

(n = 56)

2

5.2 f 0.3

5.4 + 0.2

15.3 + 1.8

14.5 + 1.5

21 (56.8)

32 (57.1)

29,233 +- 2,042

33,267 2 1,622

439 f 20

442 2 21

Diazepam (mg - total dose) Inhaled anesthetic supple-

Group 1 (Mean f SEM)

Admission temperature (“C)

Heparin (IU) Protamine dosage (mg)

infarction (MI), LCOS, and requirement for IABP in group 2, but these differences did not achieve statistical significance. There were no significant differences between the two groups with respect to the total transfusion requirements of packed red blood cells, platelets, or fresh frozen plasma intraoperatively, operative fluid balances, maximum serum potassium levels, or time taken to reach the ICU following weaning from CPB. More patients in the cold group (31% cold vs 9% warm, P = 0.01) received vasodilators (nitroglycerin or nitroprusside) during CPB. Sixtythree percent of warm patients received vasopressors during bypass compared with 48% of the cold group (P = 0.27). There was no difference in inotrope use in either group. Data obtained following admission to the ICU (Table 4) showed no significant difference between the two groups in terms of the following: core temperature on admission, serum potassium on admission, hemoglobin on admission, chest tube drainage in the first 12 hours, number of patients requiring blood transfusion, and CO on admission and at 12 hours. Furthermore, there was no difference between the two groups in the incidence of major sepsis, reoperation, in-hospital mortality, postoperative cerebrovascular accident, duration of ventilation, or length of stay in the ICU. DISCUSSION

The introduction of warm heart surgery into clinical practice represents a major change in the philosophy of management of patients undergoing CPB.” To the authors’ knowledge this is the first study to compare the perioperaTable 3. lntraoperative

Data

Group 1

Group 2

(n = 37)

(n = 56)

(U) Platelets (U)

2.0 2 0.4

1.8 ? 0.3

1.0 + 0.4

1.2 zk 0.3

Frozen plasma (U)

0.6 + 0.2

0.8 + 0.2

(Mean r SEM)

Red blood cell transfusions

4.6 -t 0.3

4.5 2 0.3

Bypass time (min)

Fluid balance (L)

93.2 IT 4.6

108 + 8.0

Cross-clamp time (min)

73.0 + 5.0

84.0 + 5.0

CPB-ICU

62.4 2 3.8

73.7 2 4.0

time interval (min)

Maximum serum potassium 5.6 k 0.18

5.6 + 0.12

LCOS (X)

(mmol/L)

15 (40)

16 (29)

IABP (%)

6 (16) 43.2

5 (9) 42.9

lnotrope use (% of patients) Patients requiring defibrillation postbypass (%) ?? P < 0.0001 (Group 1 vs Group 2).

2 (3.6)* 31 (83.8)*

36.2 + 0.1

Group 2 tn = 56)

35.9 2

0.1

Admission potassium (mmol/L)

mentation (% of patients)

(n = 37)

Admission Hb

3.9 2

0.9

3.8 + 0.6

99.6 2 3.3

102.7 + 2.4

Patients transfused with blood (%)

22 (62.9)

32 (59.3)

Admission CO (L/min)

5.0 + 0.3

4.9 + 0.2

CO at 12 hrs (Limin)

5.0 2 0.2

5.7 + 0.2

0.41 + 0.07

0.34 2 0.05

2 (5.4)

2 (3.6)

Reoperation (%)

2 (5.4)

1 (1.8)

Major sepsis (%)

2 (5.4)

Chest tube drainage in 12 hrs (L) Hospital mortality (9’0)

l(l.8)

Time ventilated (hr)

31.1 * 4.9

29.6 2 9.4

Time In ICU (hr)

41.1 + 5.3

60.6 + 16.2

CVA (X)

1 (2.7)

l(l.8) 5 (9.1)

Ml (%)

7 (18.9)

tive management of patients before and after introduction of warm heart surgery. It was found that the change from cold to warm heart surgery produced no clinically significant change in the management of anesthesia, coagulation, fluid administration, or postoperative care of patients undergoing urgent cardiac surgery, but dramatically reduced the occurrence of ventricular fibrillation on reperfusion, eliminated the need for rewarming before weaning from CPB, and suggested improved outcome in terms of mortality, MI, and the LCOS. Ventricular fibril ation following aortic cross-clamp removal is a common but undesirable event. Ventricular fibrillation is known to markedly increase myocardial oxygen requirements.‘” In addition, Gaba et al have shown that defibrillation is associated with increased myocardial lactate production even in the absence of ischemia.15 Although the addition of verapamil to the cardioplegia priming solution may decrease the number of countershocks needed to successfully delibrillate, it does not reduce the actual incidence of ventricular fibrillation.16 In this study, the use of normothermic blood cardioplegia almost always resulted in spontaneous onset of sinus rhythm after aortic crossclamp release. This dramatic reduction in the incidence of ventricular fibrillat.ion suggests a greater degree of cardioprotection with warm heart surgery. Myocardial hypothermia has traditionally been considered to be an absolute prerequisite for successful cardioplegia.7,8,‘oThe technique of continuous normothermic bloodcrystalloid cardioplegia differs significantly from others,‘7-19 which combine intermittent cold cardioplegia with either a normothermic induction, terminal “hot shot,” or the addition of various substrates. These data, in addition to previous reports, suggest that the need for hypothermia as part of the cardioplegic regimen has been overstated, thus changing the perspective on the essential elements of perioperative cardlioprotection. Advances like this set the stage for critical reappraisal of other elements of cardioplegia, such as blood versus crystalloid solutions or the

130

KAVANAGH

addition of various metabolic or nonmetabolic additive compounds. No difference was found between groups in the doses of anesthetic drugs administered or in the number of patients given supplemental volatile anesthetic agents. It cannot be determined whether this reflects no difference in actual anesthetic requirements or simply the established practice of individual anesthesiologists that did not change with the introduction of warm heart surgery. Hypothermia is well known to decrease MAC of volatile anesthetics.” However, a recent study by Hall et al in dogs demonstrated an approximately 30% reduction in enflurane MAC during normothermic CPB.” Those authors speculate that the reduction in MAC during bypass may be explained by factors such as altered cerebral hemodynamics or complement activation. Further studies of anesthetic requirements during CPB are required. The alterations in the pharmacokinetic and pharmacodynamic parameters of many drugs during hypothermic CPB have recently been reviewed.“~** Many studies have documented a decrease in plasma concentration and prolongation of elimination half-life of fentanyl resulting from a combination of factors including increased plasma dilution, adsorption onto the CPB membranes, and diminution in hepatic perfusion.” Similar reduction of plasma concentrations have been documented with lorazepam following CPB.23 Although studies that critically examine the relative contributions of both hypothermia and CPB are lacking, the present study found no difference in intravenous anesthetic requirement on clinical grounds, between two CPB regimens that differed only in perfusate temperature of both the heart and the body. However, measurement of plasma drug levels was not undertaken. Systemic hypothermia during CPB is thought to prolong the activity of heparin.z4 Although the warm patients required more supplemental heparin while on CPB, no other clinically significant difference was found in the management of anticoagulation during warm heart surgery, as measured by total doses of heparin or protamine or the peak ACT during bypass. Furthermore, there was no difference in intraoperative or postoperative blood loss or transfusion requirements between groups, suggesting no clinically significant differences in management. Future studies of heparin levels or coagulation factor assays will be of interest to correlate with these clinical observations. One advantage of systemic normothermia during warm heart surgery is the elimination of the need for rewarming before the discontinuation of CPB, although this did not result in a shorter bypass time or a temperature difference on arrival in the ICU. Whether there is a difference in the incidence of postoperative shivering or its management remains to be determined. The study also did not find an adverse effect of normothermia in terms of postoperative major neurovascular deficits.

ET &L

Thcrc arc several distinct limitations to a study such 1:~ this. The sample sizes arc small: therefore, the beneficial trends toward improved outcome (lower mortality, MI. and LCOS), which did not achieve statistical significance, may have been concealed as type 2 errors. To demonstrate a difference in these outcomes. a minimum sample size of 21fl patients per group would be required (assuming a P value of 0.05 and a power of SO%).” However, urgent cardiac cases arc less plentiful than elective cases, and acquiring larger numbers over a prolonged period may increase the risk of including uncontrolled methodological differences that may gradually develop over time. To reduce the incidence of operator variability, it was chosen to restrict this study to patients operated on by only one surgeon. The study is a retrospective review of medical records; thcrcfore, it has inherent limitations. Notwithstanding thcsc limitations, the groups were well matched in terms of selection and management. Preoperative assessment was similar; intraoperative surgical management was similar in that the same surgeon performed all the procedures; and the anesthetic management was conducted by the same group of cardiac anesthesiologists. In addition, postoperative care was similar in that there were not any significant alterations in ICU management of patients in this institution following cardiac surgery during the study period. A recent larger study of patients undergoing elective CABG showed a significant reduction in perioperative MI, reduced use of the IABP, reduced incidence of the LCOS. and reduced postoperative mortality.” That study compared 121 patients treated with continuous normothermic cardioplegia to a historical cohort of 133 patients treated with the continuous cold technique. Although similar trends are observed in this study. these did not reach statistical significance because of differences in the study periods chosen, and inclusion in this study of all urgent cardiac cases as opposed to only patients undergoing CABG in the former study. A randomized controlled prospective trial comparing continuous cold with continuous normothermic cardioplegia in patients undergoing elective and urgent CABG surgery is underway. The findings suggest that the technique of warm heart surgery is at least as safe and efficacious as a conventional cold technique in patients undergoing a variety of urgent cardiac surgical procedures using CPB. The use of warm oxygenated blood-crystalloid cardioplegia dramatically reduced the incidence of ventricular fibrillation after aortic cross-clamp removal. No significant alterations in the conduct of anesthesia were required with the introduction of the normothermic technique, and similar anesthetic drug and dosage regimens were used with equal success in the patients receiving warm cardioplegia. This suggests that standard anesthetic management techniques can be used during warm heart surgery with excellent results.

REFERENCES 1. Melrose

DG, Dreyer B, Bentall HH, et al: Elective cardiac

arrest. Lancet 2:21-22, 1955 2. Bernhard WF, Schwarz HF, Mallick NP: Intermittent coronary perfusion as an adjunct Gynecol Obstet 111:744-748,1960

to open

heart

surgery.

cold Surg

3. Bomfim V, Kaijser L. Bendz R, et al: Myocardial protection during aortic valve replacement. Cardiac metabolism and enzyme release following continuous blood cardioplegia. Stand J Thorac CardiovascSurg 15:141-147, 1981 4. Hearse PJ, Brainbridge MV, Jynge

P: Protection

of the

ANESTHESIA

131

AND WARM HEART SURGERY

Ischemic Myocardium: Cardioplegia. New York, NY, Raven Press, 1981 5. Noble WH, Lichtenstein SV, Mazer CD: Cardioplegia Controversies. Can J Anaesth 38:1-3,199l 6. Bigelow WG, Lindsay RC, Harrison RC, et al: Oxygen transport and utilization in dogs at low body temperatures. Am J Physiol160:125-137,195O 7. Bigelow WG, Lindsay WK, Greenwood WF: Hypothermia: Its possible role in cardiac surgery: An investigation of factors governing survival in dogs at low body temperatures. Ann Surg 132:849-8651950 8. Buckberg GD: A proposed “solution” to the cardioplegic controversy. J Thorac Cardiovasc Surg 77:803-815,1979 9. Roe BB: A history of clinical cardioplegia, in Engleman RM, Levitsky S (eds): A Textbook of Clinical Cardioplegia. New York, NY, Futura, 1982, pp l-7 10. Roberts AJ: Preface, in Roberts AJ (ed): Myocardial Protection in Cardiac Surgery. New York, NY, Dekker, 1987, pp v-viii 11. Lichtenstein SV, Ashe KA, El Dalati H, et al: Warm Heart Surgery. J Thorac Cardiovasc Surg 101:269-274,199l 12. Lichtenstein SV, Fremes SE, Abel JG, et al: Technical aspects of warm heart surgery. J Cardiac Surg 6:278-285,199l 13. Lichtenstein SV, El Dalati H, Panos A, et al: Long crossclamp time with warm heart surgery. Lancet 1:1443,1989 14. Buckberg GD, Brazier JR, Nelson RL, et al: Studies of the effects of hypothermia on regional myocardial blood flow and metabolism during cardiopulmonary bypass. J Thorac Cardiovasc Surg 73~87-94, 1976 15. Gaba DM, Maxwell MS, Merlone S, et al: Internal countershock produces myocardial damage and lactate production without myocardial ischemia in anesthetized dogs. Anesthesiology 66:477482,1987 16. Baraka A, Usta N, Baroody M, et al: Verapamil pretreat-

ment before aortic cross-clamping in patients undergoing artery bypass graft. Anesth Analg 66:560-564,1987

coronary

17. Teoh KH, Mickle DAG, Weisel RD, et al: Improving myocardial metabolic and functional recovery after cardioplegic arrest. J Thorac Cardiovasc Surg 95:788-798,1988 18. Rosenkranz ER, Vinten-Johansen J, Buckberg GD, et al: Benefits of normothermic induction of blood cardioplegia in energy-depleted hearts, with maintenance of arrest by multidose cold blood cardioplegic infusions. J Thorac Cardiovasc Surg 841667-677, 1982 19. Rosenkranz ER, Buckberg GD, Mulder DG, et al: Warmglutamated blood cardioplegic induction in inotropic, intra-aortic balloon-dependent coronary patients in cardiogenic shock: Initial experience and operative strategy. J Thorac Cardiovasc Surg 86:507-518,1986 20. Hall RI, Sullivan enflurane requirements 255,199O

JA: Does cardiopulmonary bypass alter for anesthesia? Anesthesiology 73:249-

21. Buylaert WA, Herregods LL, Mortier EP, et al: Cardiopulmonary bypass and the pharmacokinetics of drugs. Clin Pharmacokinet 17:10-26, 1989 22. Holley FO, Ponganis KV, Stanski DR: Effect of cardiopulmonary bypass on the pharmacokinetics of drugs. Clin Pharmacokinet 7:234-251,1982 23. Aaltonen cardiopulmonary Acta pharmacol

L, Kanto J, Arola M, et al: Effect of age and bypass on the pharmacokinetics of lorazepam. et toxic01 51:126-131, 1982

24. Cohen JA, Frederickson EL, Kaplan JA: Plasma heparin activity and antagonism during cardiopulmonary bypass with hypothermia. Anesth Analg 56:564-570,1977 25. Colton T: Statistics 1974

in Medicine.

Boston,

MA, Little, Brown,