Effects of a paf-receptor antagonist on hemodynamics during and after cardiopulmonary bypass

Effects of a Paf-Receptor Antagonist on Hemodynamics During and After Cardiopulmonary Bypass N. Nathan, MD, DrSc, P. Mercury, MD, Y. Denizot, DrSc, E. Cornu, MD, M. Laskar, MD, M. Lathehze, MD, B. Arnoux, DrSc, and P. Feiss, MD

Objectives: To assess after cardiopulmonary bypass (CPB) the role of paf-acether (paf}, a phospholipid mediator whose injection in animal mimics the hemodynamics observed after CPB. Design: Prospective double-blind randomized study, Setting: Single institutional university hospital. Participants: 18 patients scheduled to undergo coronary artery bypass graft. Interventions: 18 patients randomly recewed a placebo (n = 8) or 120 mg BN52021 (n = 10), a paf-receptor antagonist injected twice just before vascular cannulation and before cross-clamp release. Measurements and Main Results; Hemodynamic measurements were performed with a pulmonary artery and a radial artery catheter before and after the first injection of BN52021 or placebo, at the end of CPB, 1, 15, and 30 minutes after protamine infusion, then 6 hours and 24 hours postoperatively. BN52021 infusion, did not affect hemodynamic parameters. After CPB, the pulmonary artery pressures, the

cardiac index, and the pulmonary artery occlusion pressures were statistically the same between groups. By contrast, the pulmonary vascular resistances (1.5 -+ 0.5 IU v 4.5 -+ 0.6 IU, p < 0.05), the right ventricular systolic work index (5.33 _+ 0.91 g m m-2 v 9.37+ 1.02 g m m-2, p < 0 . 0 5 ) and the transpulmonary gradient (4.7 -+ 1.1 mmHg v 12.0 -+ 1.2 mmHg, p < 0.05) were lower in the BN52021 group as compared with the placebo group. After protamine infusion, these differences between groups disappeared. Conclusion; Because the inotropic and vasodilator therapy and the volume loading were the same between groups, this study suggests that pretreatment with a paf-receptor antagonist improves post-CPB pulmonary resistance. Nevertheless, this beneficial effect is transient without consequences on left ventricular function indices.

HE ROLE OF R I G H T ventncular function in the maintenance of cardioclrculatory stability after cardmc surgery has recently been stressed ~-3 Right ventricular failure may be observed at the end of cardlopulmonary bypass (CPB) and has been related to the increase m afterload induced by acute pulmonary artery hypertension. 1,3 The particular anatomy of the right coronary artery can lead to inadequate myocardial protection and right ventricular dysfunction. 1 Many compounds such as thromboxane A2 and C5a complement fractions, acting on pulmonary hemodynamics, are released during CPB and coronary reperfusion and have been beheved to participate in the post-CPB dysfunction. 4-7 The role of the phospholipld mediator paf-acether (paf) has been suggested m a m m a l models of lschemla-reperfusion, s13 Paf may be released from platelets, neutrophds, endothelial vascular cells, and cardlomyocytes actwated by pathologic stimuli.14-16 Infusion of subnanomolar doses of paf in ammals IS followed by a sharp increase in pulmonary arterml pressure and by systemlc hypotenslon resulting chiefly from right venmcular failure. ~7 Paf also induces pulmonary edema and left ventricular dysfunction through a direct reduction of contractIhty and coronary blood flow. 18"19 It has prewously been shown that pal can be released from the human heart at the end of CPB. 2° Variations of blood paf concentrations after protamlne reversal of heparin m human and rabbit have been reported. -'1-22Nevertheless, in the pig, no increase of blood paf levels and no effect of the paf receptor antagonist BN52021 have been observed on pulmonary hemodynamics 23 BN52021 is a parenteral form of gmgkolide B, an extract of the dried leaves of the gingko biloba Chinese tree It possesses natural antagomstic properties on paf-mduced hemodynamlc effects. It displaces paf from its speofic receptors on human cells, inhzbits its binding, 24 and has a long half-life (approximately 10 hours in normal volunteers). Pretreatment with BN52021

protects the animal against cardiovascular effects induced by high doses of paf. 25 In order to evaluate the role ofpaf on hemodynamics after CPB and protamine reversal of heparin m humans, the effects of the pal-receptor antagonist BN 52021 (IHB, France) have been assessed in comparison with a placebo infused in a double-bhnd randomized manner.

T

Copyright © 1995by W.B, Saunders Company KEY WORDS: cardiopulmonary bypass, paf-acether, pulmonary hypertension, protamine, reperfusion

PATIENTS AND METHODS

After informed written consent and approval of the ethics committee, 18 patients were included in this prospective double-blind placebo-randomized study. Atopy, emergency surgery, nonsteroid anti-inflammatory therapy, or platelet anta-aggregant therapy during the 15 days before surgery were considered as exclusion criteria. The patients scheduled to undergo coronary artery bypass grafting were randomly allocated before surgery into two groups. One group of patients received a placebo (n = 8) Intravenously, and another group of patients received 120 mg of BN 52021 (n = 10) just before vascular cannulanon and before crossclamp release during rewarming. The patients' demographic data are displayed in Table 1. All patients were operated on and managed by the same surgical and anes-

From the Department of Anesthestology, CHU Dupuytren, the Laboratozre d'H~matologle Expenmentale, Facult~ de M~decme, the Department of Cardtothoractc Surgery, CHU Dupuytren, and CJF INSERM 92-10, Hopttal A De Vdleneuve, Montpelher, France This work was performed m the Department of Anesthesiology, CHU Dupuytren, 2 Avenue Martm Luther King, 87042 Limoges COdex, France. Address reprint requests to N Nathan, Department d'Anesth~sw R&mmatton Ch~rurgtcale, CHU Dupuytren, 2 Avenue Martm Luther King, 87042 Ltmoges Cgdex Copyright © 1995 by W B Saunders Company 1053-0770/95/0906-000553 00/0

JournalofCardlothoraclc and VascularAnesthesla, Vol 9, No 6 (December), 1995 pp 647-652

647

648

NATHAN ET AL

Table 1, Demographic Data

Age (years) Sex (M/F) WeNht (kg) Length of surgery (mm) Length of CPB (mm) Delay 1 (rain) Delay2(mm) Protamme doses (rag) Heparm doses (IU) Number of grafts

BN 52021 Group (n = 10)

Placebo Group (n = 8)

62 5 -+ 1 8 9/1 70 +- 2 1 318 -+ 25 130 _+ 10 6 7 -+ 1 8 48 6 -+ 4 4 260 5 +- 11 3 35,450 -+ 150 3 2 -+ 0 2

55 9 -+ 2 6* 8/0 76.8 -+ 3 0 258 -+ 20 88 - 9* 6.1 -+ 1 1 47 3 _+ 8 2 248 8 -+ 16 5 31,880 -+ 253 3 1 -+ 0 2

Mean _+ SEM *p < 0 0 5 Delay 1, delay between the first dose of BN 52021 or placebo and vascular cannulatmn, Delay 2, delay between the second dose of BN 52021 or placebo and protamme mfusmn

thetlc teams. No patient had clinical ewdence of heart failure at the time of surgery Anesthesia and surgical procedures were standardized and were similar for all patients After intramuscular premedlcatlon with morphine (0.02 mg/kg) and flunitrazepam (0.02 mg/kg), anesthesia was performed usmg tltrated doses of flunltrazepam, fentanyl, and pancuronlum. Additional doses were administered before sternotomy and during the rewarmmg period CPB was achieved using moderate hypothermla. Blood oxygenation was performed using a bubble oxygenator (William Harvey, H 1700: Bard, France) and a continuous flow. Myocardial preservation was accomplished with cold Ranger's lactate solution supplemented wtth Kcl, and the pump prime conststed of crystalloid solution After CPB weaning, all patients were ventilated with 100% oxygen and ZEEP (tidal volume = 10 mL/kg and frequency = 14 BPM). Blood harvested from the surgical field and from the oxygenator after the end of bypass using a cell saver (Haemonetlcs, France) approximated blood loss and was remfused to all patients. Hemodynamic parameters were monttored using a 7F fiberoptlc pulmonary artery catheter (PAC) (Swan-Ganz; Optlcath, Abbott, France) and a radial artery catheter mserted just after induction of anesthesia. The PAC positron was controlled by inspectton of the curves and fluoroscopy. All the measures were performed at the endexpiratory point. Heart rate (HR), mean artertal systemic (MAP) and pulmonary artery pressures (PAP), right atrial (RAP), pulmonary artery occlusion pressures (PAOP), and mixed venous oxygen saturation (SvO~) were recorded before (TO) and immediately after (T1) the first dose of BN 52021 or placebo infusion and lmmedtately after weaning from CPB (T3) Additional measurements were made 1, 15, and 30 minutes after protamine infusion (T4, T5, T6), then 6 and 24 hours after surgery (T7, T8). Three consecutive measurements of cardiac output by the thermodilution method were obtained for each determination Cardiac index (CI), systolic index (SI), left and right ventrlcular stroke work index (LVSWI, RVSWI), systemic and pulmonary vascular resistances (SVR, PVR), and transpulmonary

gradient were calculated using standard formulas. MAP and CI were obtained 2 minutes after the beginning of CPB (T2). Pharmacologic support and fired volumes were selected by anesthesiologists not revolved m the protocol. At the end of CPB, volume loadmg was accomplished with hydroxyethylstarch when PAOP and MAP were below 10 and 90 mmHg. When MAP was less than 90 mmHg and the PAOP above 10 mmHg, dobutamme was administered at incremental doses and supplemented by epinephrine when dobutamlne was ineffective to normalize the hemodynamic parameters. Measurements made at T3 were recorded as soon as hemodynamic parameters were considered stable enough to allow protamine infusion. Cumulative doses of vasoactive drugs were calculated at the 24th postoperative hour. Results were expressed as mean - SEM. Demographic data were analyzed using t tests and chi-square tests. A one-way analysis of variance (ANOVA) for repeated measures and subsequent t test (with Bonferroni corrections when needed) were performed on hemodynamlc parameters obtained at different times. When the variances were not homogeneous (Bartlett test), a Kruskall-Wallis analysis was performed. At the end of CPB and at the 6th and 24th hours, hemodynamlc data were compared with basehne values. After protamine infusion, values were compared with T3 values. Hemodynamac data were compared between groups using a Kruskal-Wallis analysis. A covariance analysts studied the effect of age on hemodynamlCS. Blood loss, volumes of fluid, and drugs mfused were compared with Mann-Whitney U tests. A p < 0.05 was considered significant. Statistical analysis was performed using the statistic package "Statgraphlc" (Statistical Graphtcs Corporation, Rockvllle, MD). RESULTS

The two groups of patients did not differ significantly in regard to sex, weight, duration of surgery, the number of grafts performed, heparin and protamlne doses, or time elapsed between the BN 52021 infusion and the vascular cannulation (delay 1) or the protamine infusion (delay 2) (Table 1). The patients of the BN52021 group were older, and the length of CPB was longer than in the placebo group (Table 1) At baseline, hemodynamic data did not statistically dJffer between groups. No stattstlcal hemodynamlc variation followed BN 52021 infusion (T1) and there was no difference between groups (Figs 1 through 3) After CPB weaning, PAP, CI, HR, RVSWI, and the transpulmonary gradient increased in the placebo group, whereas they remained unchanged In the BN52021 group. PVR, RVSWI, and the transpulmonary gradient were higher in the placebo than in the BN52021-treated patients (Fig 3). The PVR increased slightly at T3 in only one patient of the BN52021 group, whereas tt increased m six patients of the placebo group (not shown). Three patients of both groups received etther dlltlazem or nitroglycerin (Table 2). These placebo-treated patients also experienced increased PAP, PVR (except one), and RVSWI despite the use of vasodilators

PAF-RECEPTOR ANTAGONIST AND CPB

649

Mean arterial p r e n u r e

Pulmonary artery oo¢luelon pregmure

MAP

PAOP

120 15

100

J

80 10 60

I

40

5

20

i T2

T3

T4

T5

T6

T7

0 TO

T0

T~

T1

T,

Time --Plapebo

group

~BN

Placebo group

52021 group

Right atrium prenuree (RAP) 14

T,

~T,

T5

Time ~

BN 52021 group

Pulmonary arterial prealure

RAP

PAP

12 10

211

8

18 g

Fig 1 MAP, PAOP, RAP, PAP (mmHg) Mean -+ SEM *p < 0.05 as compared with TO values **p < 0.05 as compared wtth T3 values, ~p < 0 05 as compared with placebo group

4 2

%

T1

T3

T4

T5

T5

T7

T1

T3

T4

~P~oebo

group

T5

T8

T7

T8

Timl

~BN

--P~©ebo

52021 group

In the BN 52021 group, hemodynamac parameters did not change over the 30 minutes after protamme infusion. except the PVR that increased at the 30 th minute after protamme. In the placebo group, PAP, CI, and RVSWI began to decrease at the 15 th minute after protamme

group

--8N

52021 group

infusion. After 30 minutes, SVR was increased in the placebo group (Fig 3) without dafference between groups. Six hours postoperatively, the HR and the CI increased an the placebo group as compared with the TO values. These hemodynamic variations were not large enough to produce Laft ventrloM lyetolk; work Index

Systemic vaecular resfetancea 3.5

TO

T8

TIr~l

LVSWI L

SVR

2,5

1~

1,5

2O

~t

tl

10

0.5 TO

T1

T3

T4

T5

T8

T7

T8

OTO

,

i

I

Tt

T3

T4

Time - - P l a c e b o group

~SN

52021 group

--

Plaoebo group

i

i

1

T8

T7

T8

~

BN 52021 group

Right ventricle eyatolio work index

Pulmonary vaaoular reelatancea RVSWl

PVR

12

I

TO

i

T5 Time

T1

T3

10

T4

T5

T6

Time --

Placebo group

--

T7

TO

~

T1

"\

T3

T4

T5

T6

Time BN 52021 group

--

PtacePo group

--

5N 52021 group

T7

T8

Fig 2 Systemic arterial resistances and PVR (IU), LVSWI and RVSWI (g/m/m2). Mean -+ SEM *p < 0 05 as compared with TO values **p < 0 05 as compared with T3 values ~p < 0 0 5 as compared with placebo group.

650

NATHAN ET AL Clrdllclndex

Heert rite

HR 120

it tO0

80

SO

TO

T1

1"2

T3

T4

T5

T8

T7

%

T8

' T1

3"lml

--

P l a c e b o group

~

.r'~

. . T4

. . . T5 .

47

T8

' T8

Time e N ~;2021 group

P l a c e b o group

Systolic I r ~ e x

--

Tranepulmonary

SI

BN 5 2 0 2 1 group

gradient

Grltdlent

5Q 15

/t~

40: 3rJ

2C Ig

TO

I

I

[

T1

T8

T4

I

Tg

T6

T7

O

T8

TO

Time

--

P l a c e b o group

~

~

i

i

T1

T3

T4

i

i

t

i

T5

re

T7

TB

Time BN 5 2 0 2 1 group

--

P l a c e b o group

a significant difference between groups (Figs 1 through 3). SvO2 did not vary during the study (not shown). The volumes of colloid, crystalloid, and blood infused during surgery and during the 24 postoperatwe hours exhibited no statistical differences between groups (Table 3). Analysis of covariance showed no effect of age on PVR, SVR, LVSWI, and RVSWI. The diuresls during surgery was not different (1339 ___303 mL v 694 +_ 158 mL,p > 0.05, c~ = 0.05, 13 = 0.38, for BN 52021 and placebo group, respectively). The use of lnotropes and/or vasodilatory agents was the same at T3 between the two groups (Table 2). One patient of the placebo group experienced acute pulmonary artery hypertension atter cross-clamp release associated with venmcular failure. One patient of the BN52021 group developed postoperative low cardiac output reqmring the use of partial extracorporeal circulation (Biomedicus; Medtronic, France). Severe ventrlcular fibrillation occurred in another patient and was successfully

--

BN 52O21 group

Fig 3. CI (L/rain/m2), SI (mL/ m2/b), HR (BPM), and transpulmonary gradient (mmHg). Mean +- SEM *p < 0.05 as compared with TO values. **p < 0.05 as compared with T3 values. ap < 0 05 as compared with placebo group,

treated w]th lidocaine infusion. Two patients of the placebo group had to be reoperated before the 24 th postoperative hour because of excessive bleeding of surgical origin. DISCUSSION

Low amounts of paf are observed in blood of normal patients, 26 but the physiolog]c significance is not known. Subnanomolar doses of pal injected into animals reduced pulmonary hypertension. 25 A gradient of paf between the left atrmm and the pulmonary or the radial artery has been observed in atherosclerotlc patients scheduled to undergo coronary artery bypass grafting. 2° The positive correlation between pal amounts in the left atrium blood and the pulmonary artery pressure suggests a role of paf in regulating pulmonary pressure. 2° On the other hand, left atrxal infusion of low doses of paf in intact lamb increases cardiac output without affecting pulmonary artery pressure, e7 He-

Table 2. Pharmacologic Support at the End of CPB (at T3, Hourly Doses [mg/h]) and During the 24 Hours Postoperatively (Cumulative Doses [mg]) 48 Hours

T3

Group

BN52021

Placebo

BN52021

Placebo

Dobutamme

(n = 6/10)

(n = 2/8)

(n = 4/10)

(n = 3/8)

(36 3 -+ 10.0)

(11 9 + 9 4)

(450-1600-750-2000)

(300-250-975)

(n = 2/10)

(n = 1/8)

(n = 3/10)

(n = 1/8)

(0 8-0 7)

(0.4)

(5-5 5-26)

Ddttazem

(n = 3/10)

(n = 1/8)

(n = 5/10)

N)troglycenn

(2-2-6) (n = 0/10)

(3) (n = 2/8)

(20 0 + 8 2) (n = 1/10)

(36 7 - 11.9) (n = 4 / 8 * )

(0 75-2)

(8)

(24-37 5-28-32)

Epinephrine

n, number of patients Mean ± SEM or mdlvtdual values p <005

(1 3) (n = 6/8)

PAF-RECEPTOR ANTAGONIST AND CPB

651

Table 3 Blood Loss, Diuresis, and Fluid Infusion (mL) During Surgery BN 52021 Group (n = 10)

Placobo Group (n = 8)

Blood loss

1509 -~ 340

1501 -.+ 258

D)uresm

1339 -+ 303

694 -+ 158

Total v o l u m e refused

5183 -* 787

3845 -+ 570

Mean + SEM

modynamic parameters did not vary after venous infusion of BN 52021, suggesting that paf is not a key compound regulating the pulmonary arterial pressure. This result does not exclude a role of paf on human pulmonary hemodynamlcs under pathologic circumstances. The heart may produce p a l after CPB in humans, 2° and coronary vessels synthesize it during ischemia and after reperfusion in humans and animals. ~2,13,28 Moreover, the incidence of arrhythmias and mechamcal dysfunction after reperfusion is lowered by pretreating animals with palreceptor antagonists, s,1°-12In this study, BN52021 pretreatmerit exhibited no beneficial effect on left ventricle function indices and the use of inotropic and vasodilatory therapy after weaning from CPB. Furthermore, the pal-receptor antagonist was not able to prevent the occurrence of severe ventricular fibrillation or low cardiac output unresponsive to motropes m two patients. In this study, the mean increase m PAP at T3 was significant only in the placebo group where one patient experienced severe pulmonary artery hypertension. The PAP were not different between groups, suggesting that other substances that paf release are implicated. For example, other vasoactive substances released during CPB such as actwated C3a and Csa fractions or thromboxane A2 may directly increase pulmonary arterial pressure. 6,29-31 After CPB weaning, the higher transpulmonary gradient may explain the higher PVR and thus the higher RVSWI in the placebo-treated patients. The use of drugs that can vasodilate the pulmonary vascular bed might have led to these results. However, no statistical difference in their use was observed, and two patients of the placebo group experienced increased PVR despite treatment with nitroglycerin or diltiazem. Cardiac index and PAOP were similar between groups, and &fferent left ventricular functions may not explain these results. The increase in right ventricular afterload has been suggested as the cause of right heart dysfunction. 1,3 In this study, no patient showed acute right heart dysfunction because right heart function is main-

tained until a high level of pulmonary impedance occurs. 32 Nevertheless, a small right ventricular dysfunction can alter cardioclrculatory stab]hty during cardiac surgery. Volume loading and preoperative left ventricular dysfunction are predictive factors of postoperative right heart dysfunction, but they were the same between groups. 33-34 BN52021pretreated patients were older and should have experienced a worse postoperative right ventricular function. 35 Moreover, covariance analysis showed no effect of age on PVR and RVSWI. The beneficial effect of paf-receptor antagonists on right ventncular function has already been observed in a swine model of heart-lung transplantation by Qayumi et al. 36 They demonstrated that pretreating the recipient animals with the pal-receptor antagonist CV-3988 preserved lung oxygenation, CI, SI, and PVR, whereas the control animals died of right ventrieular failure associated with increased PVR and pulmonary edema. The post-CPB increase in PVR has been attributed to platelet and leukocyte trapping in the pulmonary mlerovasculature. 537 Similar events occur as well after paf infusion into animals. 3s The effects of paf-receptor antagonist on the cell intrapulmonary sequestration deserve to be investigated to test this hypothesis. After protamine infusion, no patient developed acute pulmonary hypertension. In the placebo group, the decrease in PAP 15 minutes after protamine infusion was associated with the decreases m CI and RVSWI. This hemodynamlc pattern is consistent with a Type I reaction to protamine as previously described by Horrow. 39 BN 52021treated patients did not exhibit such modifications, and hemodynamlc parameters remained stable after protamine infusion. The involvement of paf in the protamine effects is controversial. 21-23 Nevertheless, because protamine's adverse effects appear to be clinically multiform and may be related to animal species, the use of paf-receptor antagonists during protamine-induced paroxysmal pulmonary hypertension deserves to be evaluated in man BN 52021 pretreatment is associated with decreased PVR and right ventricular work immediately after the end of CPB Meanwhile, this effect is transient without consequence on left ventricular function radices. The beneficml effects of a paf-antagonist treatment deserves to be investigated after CPB when right heart dysfunction and acute pulmonary edema occur and jeopardize surgical success as during heart-lung transplantation.

REFERENCES

1 Stein KL, Brelsblatt W, Wolfe C, et al. Depression and recovery of right ventncular function after cardlopulmonary bypass. Crit Care Med 18:1197-1200, 1990 2 Calvin JE Acute right heart fadure: Pathophyslology, recognition, and pharmacological management. J Cardiothor Vasc Anesth 5 507-513, 1991 3 Mangano DT: Bwentncular funcUon after myocardml revasculanzat~on m humans: Detenoratlon and recovery patterns during the first 24 hours. Anesthesiology62 571-577, 1985 4. Davies GC, Sobel M, Salzman EW Elevated plasma fibrinopeptide A and thromboxane B2 levels during car&opulmonary bypass Circulation 61:808-814, 1980 5 Chenoweth DE, Cooper SW, Hugh TE, et al: Complement

actwation during cardiopulmonarybypass Evidence for generation of C3a and C5a anaphylatoxms. N Engl J Med 304.497-503, 1981 6. Peterson MB, Huttemeler PC, Zapol WM, et al Thromboxane mediates acute pulmonary hypertension in sheep extracorporeal perfuslon. Am J Physio1243:H471-479, 1982 7 Tsao PS. Lefer AM" Car&oprotectwe actions of the specific thromboxane receptor antagonist (+)-S145 Na following coronary occlusion and reperfuslon m the rat Res Commun Chem Pathol Pharmaco170:205-2ll, 1990 8. Mickelson JK. Simpson PJ, Lucchesl BR: Myocardial dysfunction and coronary vasoconstriction induced by platelet-actwatmg factor m the post-infarcted rabbit isolated heart J Mol Cell Cardiol 20:547-561, 1988

652

9 Stahl GL, Terashita Z, Lefer AM' Role of platelet activating factor m propagation of cardiac damage during myocardial ischemia. J Pharmacol Exp Ther 244:898-904, 1988 10 Wainwright CL, Parratt JR, Blgaud M. The effects of PAF antagonists on ischaemia and reperfuslon arrhythmlas and lschaemia-Induced platelet aggregation. Blomed Blochlm Acta 47'$224$227, 1988 11. Koltal M, Tosaki A, Hosford D, et al' Ginkgolide B protects isolated hearts against arrhythmlas induced by lschemia but not reperfuslon. Eur J Pharmacol 164:293-302, 1989 12. Montrucchlo G, Alloat~ G, Tetta C, et al" Release of platelet-actlvating factor from ischemlc-reperfused rabbit heart Am J Physio1256:H1236-1246. 1989 13. Maruyama M, Farber NE, Vercellotti GM, et al: Evidence for a role of platelet activating factor In the pathogenesis of irreversible but not reversible myocardial injury after reperfuslon in dogs. Am Heart J 120.510-520, 1990 14. BenvenisteJ Paf-acether, anetherphospholipldwithbJological activity, in Karnovsky ML, Leaf A, Bohs LC (eds)' Biological Membranes New-York, Alan R. Liss, 1988, pp 75-85 15 Pinckard RN, Ludwig JC. MeManus LM: Platelet-actlvating factors, in: Gallin JI, Goldsteln IM, Snyderman R (eds)' Inflammation Basic Principles and Clinical Correlates. New York, Raven Press, 1988, pp 139-167 16. Janero DR, Burghardt C- Production and release of plateletactivating factor by the injured heart-muscle cell (cardlomyocyte). Res Commun Chem Pathol Pharmaco167:205-218, 1990 17. Feuersteln G, Goldstein RE: Effect ofpafon the cardiovascular system, in Snyder F (ed). Platelet-Actwatlng Factor and Other Lipid Mediators. New York, Plenum Press, 1987, pp 403-424 18. Stahl GL. Lefer AM: Mechanisms of platelet-actwating factor-induced cardiac depression m the isolated perfused rat heart Circ Shock 23:165-167, 1987 19. Alloattl G, Montrucchio G, Camussi G: Prostacyclin inhibits the platelet-dependent effects of platelet-actwatlng factor in the rabbit isolated heart. J Cardiovasc Pharmacol 15.745-751, 1990 20. Nathan N, Denizot Y, Felss P, et al Variation of blood paf-acether levels during coronary surgery. J Cardiothor Vasc Anesth 6:692-696, 1992 21. Nathan N, Alssa J, Benveniste J, et al' Role of paf-acether in protamme-xnduced thrombocytopenIa in rabbits. Eur J Anaesth 9.347-352, 1992 22 Nathan N, Demzot Y, Feiss P, et al' Variations in blood platelet-actwatlng-factor levels after protamme reversal of heparin in humans Acta Anaesthesiol Scand 36:264-269. 1992 23. Habazettl H, Conzen PF, Vollmar B, et al Pulmonary hypertension after heparm-protamine" Roles of left-sided infusion, histamine, and platelet-actwating factor Anesth Analg 71:637-644, 1990 24. Korth R, Benvenlste J' BN 52021 displaces [3HI paf-acether from, and inhibits its binding to intact human platelets. Eur J Pharmacol 142.331-341, 1987

NATHAN ET AL

25 Imai T, Vercellotl GM, Moldow CF, et al. Pulmonary hypertension and edema induced by platelet-activating factor in isolated, perfused rat lungs are blocked by BN52021. J Lab Chn Med 111:211-217, 1988 26. Caramelo C, Fernandez-Gallardo S, Marm-Cao D, et al. Presence of platelet-actlvatlng factor in blood from humans and experimental animals Its absence in anephric lndwiduals. Biochem Blophys Res Commun 120:789-796, 1984 27. Accurso FJ, Abman SH, Wilkening RB, et al: Fetal pulmonary vasodIlatmn after exogenous platelet-actwating factor. J Appl Physlo170.778-787, 1991 28. Montrucchio G, Camussi G, Tetta C, et al: Intravascular release of platelet-actwating factor during atrial pacing. Lancet 2'293, 1986 29. Knudsen F, Andersen LW' Immunological aspects of cardlopulmonary bypass. J Cardiothor Anesth 4.245-258, 1990 30 Ito BR, Roth DM, .Chenoweth DE, et al. Thromboxane is produced in response to intracoronary Infusions of complement C5a in pigs. Cyclooxygenase blockade does not reduce the myocardial ischemla and leukocyte accumulation. Circ Res 65-1220-1232, 1989 31 Smallboue BW, Taylor NE, Mc Donald JWD Effects of L-652,731, a platelet-activatlng factor (PAF) receptor antagonist, on PAF- and complement-induced pulmonary hypertension in sheep. J Pharmacol Exp Ther 242:1035-1040, 1987 32. Piene tt. Pulmonary arterial impedance and right ventrlcular function. Physiol Rev 66'606-652, 1986 33 Boldt J, Kling D, Moosdorf R, et al. Influence of acute volume loading on right ventncular function after cardlopulmonary bypass Crit Care Med 17'518-522, 1989 34. Boldt J, Zickmann B, Herold C, et al. Right ventncular function in patients with reduced left ventricular function undergoing myocardial revascularization. J Cardlothor Vasc Anesth 6'2428, 1992 35. Boldt J, Zlckmann B, Thlel A, et al' Age and right ventricular function during cardiac surgery J Cardlothor Vasc Anesth 6.29-32, 1992 36. Qayuml AK, Jamleson WRE, Poostizadeh A' Effects of platelet-actwating factor antagonist CV-3988 m preservation of heart and lung for transplantation. Ann Thorac Surg 52:1026-1032, 1991 37. Howard R J, Cram C, Franzini DA, et al' Effects of cardlopulmonary bypass on pulmonary leukostasis and complement activation. Arch Surg 123'1496-1501. 1988 38. McManus LM, Plnckard RN. Kinetics of acetyl glyceryl ether phosphorylcholine (AGEPC)-lnduced acute lung alterations in the rabbit. Am J Pathol 121:55-68, 1985 39. Horrow JC: Protamine: A review of its toxicity. Anesth Analg 64:348-361, 1985