- Email: [email protected]
Serum concentrations of prostacyclin and thromboxane in children before, during, and after cardiopulmonary bypass
J THoRAc CARDlOVASC SURG 92:73-78, 1986
Serum concentrations of prostacyclin and thromboxane in children before, during, and after cardiopulmonary bypass Twenty-six consecutive pediatric patients lUldergoing reparative procedures necessitating cardiopulmonary bypass were prospectively studied to determine changes in serum levels of 6-ket~prostaglandin F la and tbromboxane B; Cardiac lesions included acyanotic lesions (five patients), obstructive lesions (10 patients), and right-t~left shunts (11 patients). There was a significant (p < 0.05) increase in 6-kero-prostaglandin F la from preoperative levels to levels measured at the time of arterial and venous cannula insertion. This concentration was maintained throughout cardiopulmonary bypass and remained significantly elevated (p <0.001) in the recovery room, but returned to preoperative levels by the morning after the operation. Preoperative levels of thromboxane Hz varied widely and were not significantly different from intraoperative levels. The postoperative levels of thromboxane Hz, however, were significantly different (p <0.05) from the intraoperative levels. In the pediatric age group undergoing cardiopulmonary bypass, 6-ket~prostaglandin F la and thromboxane H2 change during bypass but do not significantly differ when preoperative levels are compared to postoperative values.
William H. Fleming, M.D., Lynne B. Sarafian, R.N., M.N., M. Patricia Leuschen, Ph.D., Myrna C. Newland, M.D., Erin M. Kennedy, B.S., John D. Kugler, M.D., James W. Chapin, M.D., Barbara J. Hurlbert, M.D., David L. Bolam, M.D., and Robert M. Nelson, Jr., M.D., Omaha, Neb.
PostaCyClin and thromboxane A z are known to affect both hemostasis and vasomotor tone. Prostacyclin is a potent inhibitor of platelet aggregation, can disaggregate platelets, and also acts as a potent vasodilator. Conversely, thromboxane A z is a potent platelet proaggregant and vasoconstrictor.!" Certain diseases and behaviors have been shown to affect the serum concentrations of these two substances." Recent reports indicate changes in serum concentration for prostacyclin and thromboxane A z during cardiopulmonary bypass (CPB) in adult patients." CPB is frequently used in the repair of a spectrum of congenital heart lesions in pediatric patients. Since excessive bleeding or embolic From the Sections of Thoracic Cardiac Surgery, Neonatology, Pediatric Cardiology, and the Department of Anesthesiology, University of Nebraska Medical Center, Omaha, Neb. Received for publication July 10, 1985. Accepted for publication Oct. 2, 1985. Address for reprints: William H. Fleming, M.D., Section of Thoracic Cardiac Surgery, University of Nebraska Medical Center, 42nd and Dewey, Omaha, Neb. 68105.
phenomena affect the morbidity and mortality,'? 11 knowledge of the prostacyclin/thromboxane A z balance during and at the end of CPB may prove useful clinically. This study evaluates changes in serum concentrations of 6-keto-prostaglandin F'a(6-keto-PGF l a ) , the metabolite of prostacyclin, and thromboxane Bz, the metabolite of thromboxane A z, in pediatric patients undergoing CPB. Patients Twenty-six patients, 14 boys and 12 girls (age range 13 days to 16 years, median 3 years) who underwent cardiac operations requiring CPB, were studied prospectively. Informed consent was obtained from the legal guardian of each patient studied. The only patients not eligiblefor the study were neonates, for whom even the 1 rn1 of plasma for preoperative determination was considered medicallycontraindicated, and a 4-year-old child of Jehovah's Witness parents, for whom non-medically necessary blood drawing was considered unacceptable. The types of operative repairs are listed in Table I. 73
The Journal of
74
Fleming et al.
Thoracic and Cardiovascular Surgery
1200
c==J
1000
II
=
6-KETO-PGF1~
= TxBZ
800
..., .!: co
600
c.
400
200
Preop
Anesthesia
Immed Pre CPB
CPB Smin
cn
CPS
CPS
lhr
Zhr
3hr
Protamine
RecRm
Postop
Fig. 1. The concentrations (mean ± SD) of 6-keto-PGF 'a and thromboxane B2 (TxB]) in plasma of 26 patients before, during, and after cardiopulmonary bypass (CPB). RecRrn. Recovery room.
Table I. Operative repair of 26 children undergoing cardiopulmonary bypass Repair ASD secundum Valve replacement ± conoplasty ± VSD ASD secundum + cor triatriatum Tetralogy of Fallot Aortic valvotomy VSD ± ASD VSD + resection of subvalvular AS Pulmonary artery debanding
No. of patients 3 6 I
5 3 6 I I
Legend: ASD. Atrial septal defect. VSD, Ventricular septal defect. AS. Aortic stenosis.
Methods CPB technique. A membrane oxygenator (SciMed Life Systems, Inc., Minneapolis, Minn.) of the appropriate size for calculated flow rate was used. CPB prime solution consisted of electrolytes (Plasma-Lyte 148), albumin, mannitol, dextrose, sodium bicarbonate (to attain a pH >7.45), packed red blood cells (to maintain
a hematocrit value between 20% and 30% during bypass), and heparin (to maintain an activated clotting time >400 seconds). Sampling and assay technique. Blood samples were drawn via venipuncture with the other routine preoperative laboratory tests so that each patient could serve as his/her own control. On the day of operation a sample was drawn via the arterial line after the patient had been anesthetized, immediately after placement of the arterialtine. A third sample was drawn just before CPB was begun, after arterial and venous cannulas were placed. The fourth sample was drawn within 5 minutes after the initiation of CPB. Additional samples were drawn at 1 hour intervals during CPB. Within 15 to 45 minutes after protamine was given, another sample was drawn, with a further sample drawn upon arrival in the recovery room. The final sample was drawn from the arterial line the morning after the operation, in conjunction with other blood sampling for laboratory testing. All blood samples were collected in siliconized polypropylene test tubes coated with a solution of 4.5 mEq ethylenediaminetetraacetic acid containing a prostaglandin synthetase inhibitor, indomethacin. The serum
Volume 92
Prostacyclin and thromboxane
Number 1 July, 1986
75
FREClUENCY DISTRIBUTION OF 6. KETO PGFl ALPHA LEVELS IN CHILDREN UNDERGOING CARDIOPULMONARY BYPASS
25
Immed Pre CPS
Pre op
20 15 Z
w
10
lS...
5
LL
1 25
:r 0 0
a: w
m ::IE
20
Z
15
CPS 1 hr
::)
10 5
o
100
200
300
400
500
600
700
0
100
200
300
400
500
pg/ml PLASMA
Fig. 2. Frequency distribution of 6-keto-PGF la levels in children undergoing CPB.
Table Il, Comparison of 6-keto-PGFJa and thromboxane B 2 (TxB 2J in pediatric patients undergoing CPB
Preoperauve Anesthesia Before CPB CPB (5 min) CPB (I hr) CPB (2 hr) CPB (3 hr) Protamine Recovery room Postoperative
Mean levels of 6-keto-PGF'a Ipg/ml)
Mean levels of TxB2
40 45 159 172 227 134 316 248 137 50
345 113 114 302 311 630 311 237 98 67
(pgfml]
Statistical significance compared to preoperative value 6-keto-PGF,•
NS* <0.001 <0.001 <0.001 0.038 0.033 <0.001 <0.001
NS
I
Statistical significance compared to postoperative value
TxB2
6-keto-PGF'a
NS NS NS NS NS NS NS NS NS
0.0085 <0.001 <0.001 0.041 0.049 <0.001 <0.001 <0.001
NS
I
TxB 2
NS 0.039 0.013 0.009 0.0085 0.03 <0.001 0.003
NS
'NS = p> 0.05.
was isolated, immediately frozen, and
stored at
-800 C. The stable metabolites of two specific prostaglandins were measured: 6-keto-PGF 1a , a stable metabolite representing 60% to 80% of actual prostacyclin levels, and tbromboxane B2, the stable metabolite of thromboxane A2• These metabolites were isolated and separated on Bond eluent columns. Specific antisera to both thromboxane B2 and 6-keto-PGF la were purchased in conjunction with iodine 125 radioimmunoassay kits (New England Nuclear, Boston, Mass.). The radioactivity was quantitated in a gamma counter. Results were obtained by standard constructed dose-response curves and data were normalized to picograrns of prostaglandin per milliliter of serum. Data were analyzed with the Krus-
kal-Wallis analysis by ranks as well as frequency distribution. Results 6-Keto-PGF ta • Preoperative serum levels of 6-ketoPGF ,ao which reflect prostacyclin levels, ranged from 6 to 155 pgjml serum; 17 of 24 patients had levels of 50 (pgjml or less, and 20 of 24 had levels of 100 pgjml or less (mean 39 pgjml ± 36 standard deviation [SD], median 29 pgjml), which indicates a positively skewed population (Fig. 1). At the first intraoperative measurement, with the patient anesthetized and the arterial line in place, the serum concentration of 6-keto PGF 'a was less than 100 pgjml (range 14 to 88 pgjml, mean 45 pgjml, median 40 pgjml) in all 26 patients, a nonsignif-
The Journal of
76
Fleming et al.
Thoracic and Cardiovascular Surgery
FREQUENCY DISTRIBUTION OF THROMBOXANE B2 LEVELS IN CHILDREN UNDERGOING CARDIOPULMONARY BYPASS
25
Pre op
20
Immed Pre CPS
15
z w
a:
10
0
-'
J:
5
LL
1 25
0
0
a: w lD :::;
2
CPS 1hr
20
Post op
15 10 5
o
100
200
300
400
500
600
700
0
100
200
300
400
500
pg/ml PLASMA
Fig. 3. Frequency distribution of thromboxane 8 2 levels in children undergoing CPR
icant correlation (p > 0.1) with preoperative levels. The preoperative and postoperative levels of 6-keto-PGF'a were not significantly different (p = 0.079) either from one another or from the levels immediately after placement of the arterial line with the patient anesthetized (Fig. 2). The sample immediately before CPB with arterial and venous cannulas in place showed a much greater variation in serum concentrations (range 5 to 490 pgjml, mean 159 ± 148 pgjml [SD], median 70 pgj ml). Analysis of a frequency distribution of those data indicated a biphasic population with a significant group remaining stable, i.e., with levels of 50 pgjml or less, whereas in another group the concentration of 6keto-PGF 'a increased to 200 pgjml and above (Fig. 2). All patients had significantly higher serum levels of 6-keto-PGF 'aduring CPB (Table II) than either before or after the operation (Fig. 1). There was a wide range of values, however (Fig. 1). Thus the range for the 16 patients on CPB at the end of 1 hour was 14 to 700 pgjml, for the six patients on CPB at the end of 2 hours, 14 to 265 pgjml, and for the three patients on CPB at the end of 3 hours, 47 to 460 pgjml. The serum levelsremained elevated while the children were on CPB but decreased immediately after discontinuation of CPB before protamine administration. By the morning after the operation, there was no significant difference (p = 0.079) between preoperative and postoperative levels (Fig. 2). Thromboxane B2• Preoperative values for thrombo-
xane B2 showed a wide range (5 to 1,950 pgjml, mean 345 pgjml ± 515 [SD], median 63 pgjml). The positive skewness was due to five patients with very high levels (above 650 pgjml serum). The preoperative variability made comparisons with intraoperative values difficult, and nonparametric analysis using the KruskalWallis analysis by ranks showed no significant difference (p > 0.1) between the preoperative and intraoperative serum concentrations. The range was less remarkable in the anesthetized patient with levels drawn immediately after placement of the arterial line (range 8 to 520 pgjml serum, mean 113 pgjml, median 82 pgjml). The range, mean, and median increased considerably immediately after institution of CPB (range 13 to 1,200 pgjml, mean 302 pgjml, median 160 pgjml [Fig. 3]). This pattern continued for patients regardless of CPB duration. As with the 6-keto-PGF 1a levels, thromboxane B2 began to decrease and the range began to narrow by the time the patients were in the recovery room (range 28 to 225 pgjml serum, mean 98 pg/ml, median 88 pg/ml). Recovery room levels were moving toward the postoperative values (Fig. 3). Although there was no significant difference (p > 0.1) between the preoperative and postoperative mean values of thromboxane B2, there were significant differences between postoperative values and intraoperative values. When compared to the postoperative levels, the mean thromboxane B2 concentration was significantly elevated at the time of placement of the arterial line in the anesthetized patient (p = 0.039). The levels remained elevated in a pattern similar to that seen with
Volume 92 Number 1 July, 1986
6-keto-PGF 1a ; the highest mean values were attained in those patients remaining on CPB for 2 hours or longer. Although the levels in the recovery room began to decrease toward baseline, the mean value was still significantly higher (p < 0.(01) than postoperative levels.
Discussion Pericardium, epicardium, and atrial tissue, as well as coronary vasculature, affect the generation of prostacyclin":" and, to a lesser extent, thromboxane A; Both substances are known to affect vasomotor tone and platelet aggregation. 1·6 In addition, prostacyclin is thought by various investigators to be protective of ischemic myocardium'> IZ and hypertrophied hearts." The role ofthromboxane A z, if any, in the cardiac region has not been defined. In this study of pediatric patients, there was an increase in 6-keto-PGF 1a , but not thromboxane B z, after arterial and venous cannulas were placed. The increase is consistent with the results of a similar study of adult patients reported by Ylikorkala, Saarela, and Viinikka.' Manipulation of the pericardium and atrial structures is a likely source of 6keto-PGF la measured during this time. Changes in 6-keto-PGF l a and thromboxane B, during CPR have been documented in adults-v"!' but have not been previously reported in the pediatric age group undergoing CPB. The data from this study of pediatric patients showed an increase of 6-keto-PGF 1a and a wide range in thromboxane B, during CPB. Levels in adult studieswere not significantly different from those in our study (p > 0.1). One limitation of our study may be that the preoperative thromboxane B2 sample was drawn via venipuncture, which may account for the wide variation in thromboxane B; However, other samples that were drawn via an arterial line showed a consistently wide range until after the patient was off CPB. There was a consistent decrease in variability postoperatively and on the first postoperative day thromboxane B, levels were similar to those reported in adults before CPB. Davies, Sobel, and Salzman? noted no difference in the levels of thromboxane B, in mixed venous samples compared to blood drawn from the left ventricle. Numerous reports show that prostaglandins are important to the proper functioning of the cardiovascular system. 1,3, 4, 6 Exogenous prostaglandin I, has been used in animal experiments and in human patients in an attempt to treat various diseases and in the CPB circuit for various theoretical benefits. 10, II, 15·Z0 However, optimal levels of prostacyclin and thromboxane A z, or the ratio of prostacyclin to thromboxane A z, have not been
Prostacyclin and thromboxane
77
defined, nor have usual levels of 6-keto-PGF la or thromboxane B, been reported in groups of children with or without congenital heart disease. Prostacyclin is elevated in response to myocardial ischemia and is protective of ischemic and hypertrophied hearts." 15, 16,ZO,ZI The effect of thromboxane B, on the heart is poorly understood. Although thromboxane B, was found to be elevated in the coronary sinus blood in patients with unstable angina, it is not known if that is a cause of ischemia or an effect." To selectively raise prostacyclin by exogenous means without a concomitant rise in thromboxane Bz may produce an as yet unrecognized undesirable effect. Further differentiations of normal for these substances for various age groups and various disease states must be correlated with clinical effects in order to benefit the cardiac surgical patient. We acknowledge the valuable laboratory assistance provided by Joan Spencer and the assistance provided by Linda Stokes, Respiratory Therapist, and Lynn Gilson, Anesthesia Technologist, in the preparation of this paper. REFERENCES Dusting GJ, Moncada S, Vane JR: Prostaglandins, their intermediates and precursors. Cardiovascular actions and regulatory roles in normal and abnormal circulatory systems. Prog Cardiovasc Dis 21:405, 1979 2 Ylikorkala 0, Saarela E, Viinikka L: Increased prostacyclin and thromboxaneproduction in man during cardiopulmonary bypass. J THORAC CARDlOYASC SURG 82:245, 1981 3 Vane JR: Prostaglandins and the cardiovascular system. Br Heart J 49:405, 1981 4 Watkins WD, Peterson MB, Kong DL, Kono K, Buckley MJ, Levine FH, Philbin OM: Thromboxaneand prostacyclin changes during cardiopulmonary bypass with and without pulsatile flow. J THORAC CARDlOyASC SURG 84:250, 1982 5 Kobinia GS, La Raia PJ, Peterson MB, D'Ambra MN, Watkins WD, Austen WG, Buckley MJ: Cardiac prostacyclin kinetics during cardiopulmonary bypass. J THORAC CARDIOyASC SURG 88:965, 1984 6 Jacobsen DC: Prostaglandins and cardiovascular disease-A review. Surgery 93:564, 1983 7 Nadler JL, Velasco JS, Horton R: Cigarette smoking inhibits prostacyclin formation. Lancet 1:1248, 1983 8 Newman WH, Frankis MB, Halushka PV: Increased myocardial release of prostacyclin in dogs with heart failure. J Cardiovasc Pharmacol 5: 194, 1983 9 Davies GC, Sobel M, Salzman EW: E!evated plasma fibrinopeptide A and thromboxaneB2 levels during cardiopulmonary bypass. Circulation 61:808, 1980 10 Faichney A, Davidson KG, Wheatley OJ, Davidson MB, Walker 10: Prostacyclin in cardiopulmonary bypassoperations. J THORAC CARDIOyASC SURG 84:601, 1982 11 Malpass TW, Amory DW, Harker LA, Ivey TO, WiI-
78
12
13
14
15
16
17
The Journal of Thoracic and Cardiovascular
Fleming et al.
Iiams DB: The effect of prostacyclin infusion on platelet hemostatic function in patients undergoing cardiopulmonary bypass. J THoRAc CARDIOVASC SURG 87:550, 1984 Nolan RD, Dusting GJ, Jakubowski J, Martin TJ: The pericardium as a source of prostacyclin in the dog, ox and rat. Prostaglandins 24:887, 1982 Dusting GJ, Nolan RD, Woodman QL, Martin TJ: Prostacyclin produced by the pericardium and its influence on coronary vascular tone. Am J Cardiol 52:28A, 1983 Mehta J, Mehta P: Prostacyclin and thromboxane B2 production by human cardiac atrial tissue. Am Heart J 109:1,1985 Melin JA, Becker LC: Salvage of ischemic myocardium by prostacyclin during experimental myocardial infarction. J Am Coll Cardiol 2:279, 1983 Yui Y, Nakajima H, Kawai C, Murakami T: Prostacyclin therapy in patients with congestive heart failure. Am J Cardiol 50:320, 1982 Zusman RM, Rubin RH, Cato AE, Cocchetto DM, Crow JW, Tolkoff-Rubin N: Hemodialysis using prostacyclin
Surgery
18
19
20
21
instead of heparin as the sole antithrombotic agent. N Engl J Med 304:934, 1981 Aren C, Feddersen K, Radegran K: Effects ofprostacyclin infusion on platelet activation and postoperative blood loss in coronary bypass. Ann Thorac Surg 36:49, 1983 DiSesa VJ, Huval W, Lelcuk S, Jonas R, Maddi R, Lee-Son S, Shemin RJ, Collins 11, Hechtman HB, Cohn LH: Disadvantages of prostacyclin infusionduring cardiopulmonary bypass. A double blind study of 50 patients having coronary revascularization. Ann Thorac Surg 38:514,1984 Povzhitkov M, Haendchen RV, Meerbaum S, Fishbein MC, Shell W., Corday E: Prostaglandin E) coronary venous retroperfusion in acute myocardial ischemia. Effects on regional left ventricular function and infarct size. J Am Coll Cardiol 3:439, 1984 Hirsh PD, Hillis D, Campbell WB, Firth BG, Willerson JT: Release of prostaglandins and thromboxane into the coronary circulation in patients with ischemic heart disease. N Engl J Med 304:685, 1981
Serum concentrations of prostacyclin and thromboxane in children before, during, and after cardiopulmonary bypass Twenty-six consecutive pediatric patients lUldergoing reparative procedures necessitating cardiopulmonary bypass were prospectively studied to determine changes in serum levels of 6-ket~prostaglandin F la and tbromboxane B; Cardiac lesions included acyanotic lesions (five patients), obstructive lesions (10 patients), and right-t~left shunts (11 patients). There was a significant (p < 0.05) increase in 6-kero-prostaglandin F la from preoperative levels to levels measured at the time of arterial and venous cannula insertion. This concentration was maintained throughout cardiopulmonary bypass and remained significantly elevated (p <0.001) in the recovery room, but returned to preoperative levels by the morning after the operation. Preoperative levels of thromboxane Hz varied widely and were not significantly different from intraoperative levels. The postoperative levels of thromboxane Hz, however, were significantly different (p <0.05) from the intraoperative levels. In the pediatric age group undergoing cardiopulmonary bypass, 6-ket~prostaglandin F la and thromboxane H2 change during bypass but do not significantly differ when preoperative levels are compared to postoperative values.
William H. Fleming, M.D., Lynne B. Sarafian, R.N., M.N., M. Patricia Leuschen, Ph.D., Myrna C. Newland, M.D., Erin M. Kennedy, B.S., John D. Kugler, M.D., James W. Chapin, M.D., Barbara J. Hurlbert, M.D., David L. Bolam, M.D., and Robert M. Nelson, Jr., M.D., Omaha, Neb.
PostaCyClin and thromboxane A z are known to affect both hemostasis and vasomotor tone. Prostacyclin is a potent inhibitor of platelet aggregation, can disaggregate platelets, and also acts as a potent vasodilator. Conversely, thromboxane A z is a potent platelet proaggregant and vasoconstrictor.!" Certain diseases and behaviors have been shown to affect the serum concentrations of these two substances." Recent reports indicate changes in serum concentration for prostacyclin and thromboxane A z during cardiopulmonary bypass (CPB) in adult patients." CPB is frequently used in the repair of a spectrum of congenital heart lesions in pediatric patients. Since excessive bleeding or embolic From the Sections of Thoracic Cardiac Surgery, Neonatology, Pediatric Cardiology, and the Department of Anesthesiology, University of Nebraska Medical Center, Omaha, Neb. Received for publication July 10, 1985. Accepted for publication Oct. 2, 1985. Address for reprints: William H. Fleming, M.D., Section of Thoracic Cardiac Surgery, University of Nebraska Medical Center, 42nd and Dewey, Omaha, Neb. 68105.
phenomena affect the morbidity and mortality,'? 11 knowledge of the prostacyclin/thromboxane A z balance during and at the end of CPB may prove useful clinically. This study evaluates changes in serum concentrations of 6-keto-prostaglandin F'a(6-keto-PGF l a ) , the metabolite of prostacyclin, and thromboxane Bz, the metabolite of thromboxane A z, in pediatric patients undergoing CPB. Patients Twenty-six patients, 14 boys and 12 girls (age range 13 days to 16 years, median 3 years) who underwent cardiac operations requiring CPB, were studied prospectively. Informed consent was obtained from the legal guardian of each patient studied. The only patients not eligiblefor the study were neonates, for whom even the 1 rn1 of plasma for preoperative determination was considered medicallycontraindicated, and a 4-year-old child of Jehovah's Witness parents, for whom non-medically necessary blood drawing was considered unacceptable. The types of operative repairs are listed in Table I. 73
The Journal of
74
Fleming et al.
Thoracic and Cardiovascular Surgery
1200
c==J
1000
II
=
6-KETO-PGF1~
= TxBZ
800
..., .!: co
600
c.
400
200
Preop
Anesthesia
Immed Pre CPB
CPB Smin
cn
CPS
CPS
lhr
Zhr
3hr
Protamine
RecRm
Postop
Fig. 1. The concentrations (mean ± SD) of 6-keto-PGF 'a and thromboxane B2 (TxB]) in plasma of 26 patients before, during, and after cardiopulmonary bypass (CPB). RecRrn. Recovery room.
Table I. Operative repair of 26 children undergoing cardiopulmonary bypass Repair ASD secundum Valve replacement ± conoplasty ± VSD ASD secundum + cor triatriatum Tetralogy of Fallot Aortic valvotomy VSD ± ASD VSD + resection of subvalvular AS Pulmonary artery debanding
No. of patients 3 6 I
5 3 6 I I
Legend: ASD. Atrial septal defect. VSD, Ventricular septal defect. AS. Aortic stenosis.
Methods CPB technique. A membrane oxygenator (SciMed Life Systems, Inc., Minneapolis, Minn.) of the appropriate size for calculated flow rate was used. CPB prime solution consisted of electrolytes (Plasma-Lyte 148), albumin, mannitol, dextrose, sodium bicarbonate (to attain a pH >7.45), packed red blood cells (to maintain
a hematocrit value between 20% and 30% during bypass), and heparin (to maintain an activated clotting time >400 seconds). Sampling and assay technique. Blood samples were drawn via venipuncture with the other routine preoperative laboratory tests so that each patient could serve as his/her own control. On the day of operation a sample was drawn via the arterial line after the patient had been anesthetized, immediately after placement of the arterialtine. A third sample was drawn just before CPB was begun, after arterial and venous cannulas were placed. The fourth sample was drawn within 5 minutes after the initiation of CPB. Additional samples were drawn at 1 hour intervals during CPB. Within 15 to 45 minutes after protamine was given, another sample was drawn, with a further sample drawn upon arrival in the recovery room. The final sample was drawn from the arterial line the morning after the operation, in conjunction with other blood sampling for laboratory testing. All blood samples were collected in siliconized polypropylene test tubes coated with a solution of 4.5 mEq ethylenediaminetetraacetic acid containing a prostaglandin synthetase inhibitor, indomethacin. The serum
Volume 92
Prostacyclin and thromboxane
Number 1 July, 1986
75
FREClUENCY DISTRIBUTION OF 6. KETO PGFl ALPHA LEVELS IN CHILDREN UNDERGOING CARDIOPULMONARY BYPASS
25
Immed Pre CPS
Pre op
20 15 Z
w
10
lS...
5
LL
1 25
:r 0 0
a: w
m ::IE
20
Z
15
CPS 1 hr
::)
10 5
o
100
200
300
400
500
600
700
0
100
200
300
400
500
pg/ml PLASMA
Fig. 2. Frequency distribution of 6-keto-PGF la levels in children undergoing CPB.
Table Il, Comparison of 6-keto-PGFJa and thromboxane B 2 (TxB 2J in pediatric patients undergoing CPB
Preoperauve Anesthesia Before CPB CPB (5 min) CPB (I hr) CPB (2 hr) CPB (3 hr) Protamine Recovery room Postoperative
Mean levels of 6-keto-PGF'a Ipg/ml)
Mean levels of TxB2
40 45 159 172 227 134 316 248 137 50
345 113 114 302 311 630 311 237 98 67
(pgfml]
Statistical significance compared to preoperative value 6-keto-PGF,•
NS* <0.001 <0.001 <0.001 0.038 0.033 <0.001 <0.001
NS
I
Statistical significance compared to postoperative value
TxB2
6-keto-PGF'a
NS NS NS NS NS NS NS NS NS
0.0085 <0.001 <0.001 0.041 0.049 <0.001 <0.001 <0.001
NS
I
TxB 2
NS 0.039 0.013 0.009 0.0085 0.03 <0.001 0.003
NS
'NS = p> 0.05.
was isolated, immediately frozen, and
stored at
-800 C. The stable metabolites of two specific prostaglandins were measured: 6-keto-PGF 1a , a stable metabolite representing 60% to 80% of actual prostacyclin levels, and tbromboxane B2, the stable metabolite of thromboxane A2• These metabolites were isolated and separated on Bond eluent columns. Specific antisera to both thromboxane B2 and 6-keto-PGF la were purchased in conjunction with iodine 125 radioimmunoassay kits (New England Nuclear, Boston, Mass.). The radioactivity was quantitated in a gamma counter. Results were obtained by standard constructed dose-response curves and data were normalized to picograrns of prostaglandin per milliliter of serum. Data were analyzed with the Krus-
kal-Wallis analysis by ranks as well as frequency distribution. Results 6-Keto-PGF ta • Preoperative serum levels of 6-ketoPGF ,ao which reflect prostacyclin levels, ranged from 6 to 155 pgjml serum; 17 of 24 patients had levels of 50 (pgjml or less, and 20 of 24 had levels of 100 pgjml or less (mean 39 pgjml ± 36 standard deviation [SD], median 29 pgjml), which indicates a positively skewed population (Fig. 1). At the first intraoperative measurement, with the patient anesthetized and the arterial line in place, the serum concentration of 6-keto PGF 'a was less than 100 pgjml (range 14 to 88 pgjml, mean 45 pgjml, median 40 pgjml) in all 26 patients, a nonsignif-
The Journal of
76
Fleming et al.
Thoracic and Cardiovascular Surgery
FREQUENCY DISTRIBUTION OF THROMBOXANE B2 LEVELS IN CHILDREN UNDERGOING CARDIOPULMONARY BYPASS
25
Pre op
20
Immed Pre CPS
15
z w
a:
10
0
-'
J:
5
LL
1 25
0
0
a: w lD :::;
2
CPS 1hr
20
Post op
15 10 5
o
100
200
300
400
500
600
700
0
100
200
300
400
500
pg/ml PLASMA
Fig. 3. Frequency distribution of thromboxane 8 2 levels in children undergoing CPR
icant correlation (p > 0.1) with preoperative levels. The preoperative and postoperative levels of 6-keto-PGF'a were not significantly different (p = 0.079) either from one another or from the levels immediately after placement of the arterial line with the patient anesthetized (Fig. 2). The sample immediately before CPB with arterial and venous cannulas in place showed a much greater variation in serum concentrations (range 5 to 490 pgjml, mean 159 ± 148 pgjml [SD], median 70 pgj ml). Analysis of a frequency distribution of those data indicated a biphasic population with a significant group remaining stable, i.e., with levels of 50 pgjml or less, whereas in another group the concentration of 6keto-PGF 'a increased to 200 pgjml and above (Fig. 2). All patients had significantly higher serum levels of 6-keto-PGF 'aduring CPB (Table II) than either before or after the operation (Fig. 1). There was a wide range of values, however (Fig. 1). Thus the range for the 16 patients on CPB at the end of 1 hour was 14 to 700 pgjml, for the six patients on CPB at the end of 2 hours, 14 to 265 pgjml, and for the three patients on CPB at the end of 3 hours, 47 to 460 pgjml. The serum levelsremained elevated while the children were on CPB but decreased immediately after discontinuation of CPB before protamine administration. By the morning after the operation, there was no significant difference (p = 0.079) between preoperative and postoperative levels (Fig. 2). Thromboxane B2• Preoperative values for thrombo-
xane B2 showed a wide range (5 to 1,950 pgjml, mean 345 pgjml ± 515 [SD], median 63 pgjml). The positive skewness was due to five patients with very high levels (above 650 pgjml serum). The preoperative variability made comparisons with intraoperative values difficult, and nonparametric analysis using the KruskalWallis analysis by ranks showed no significant difference (p > 0.1) between the preoperative and intraoperative serum concentrations. The range was less remarkable in the anesthetized patient with levels drawn immediately after placement of the arterial line (range 8 to 520 pgjml serum, mean 113 pgjml, median 82 pgjml). The range, mean, and median increased considerably immediately after institution of CPB (range 13 to 1,200 pgjml, mean 302 pgjml, median 160 pgjml [Fig. 3]). This pattern continued for patients regardless of CPB duration. As with the 6-keto-PGF 1a levels, thromboxane B2 began to decrease and the range began to narrow by the time the patients were in the recovery room (range 28 to 225 pgjml serum, mean 98 pg/ml, median 88 pg/ml). Recovery room levels were moving toward the postoperative values (Fig. 3). Although there was no significant difference (p > 0.1) between the preoperative and postoperative mean values of thromboxane B2, there were significant differences between postoperative values and intraoperative values. When compared to the postoperative levels, the mean thromboxane B2 concentration was significantly elevated at the time of placement of the arterial line in the anesthetized patient (p = 0.039). The levels remained elevated in a pattern similar to that seen with
Volume 92 Number 1 July, 1986
6-keto-PGF 1a ; the highest mean values were attained in those patients remaining on CPB for 2 hours or longer. Although the levels in the recovery room began to decrease toward baseline, the mean value was still significantly higher (p < 0.(01) than postoperative levels.
Discussion Pericardium, epicardium, and atrial tissue, as well as coronary vasculature, affect the generation of prostacyclin":" and, to a lesser extent, thromboxane A; Both substances are known to affect vasomotor tone and platelet aggregation. 1·6 In addition, prostacyclin is thought by various investigators to be protective of ischemic myocardium'> IZ and hypertrophied hearts." The role ofthromboxane A z, if any, in the cardiac region has not been defined. In this study of pediatric patients, there was an increase in 6-keto-PGF 1a , but not thromboxane B z, after arterial and venous cannulas were placed. The increase is consistent with the results of a similar study of adult patients reported by Ylikorkala, Saarela, and Viinikka.' Manipulation of the pericardium and atrial structures is a likely source of 6keto-PGF la measured during this time. Changes in 6-keto-PGF l a and thromboxane B, during CPR have been documented in adults-v"!' but have not been previously reported in the pediatric age group undergoing CPB. The data from this study of pediatric patients showed an increase of 6-keto-PGF 1a and a wide range in thromboxane B, during CPB. Levels in adult studieswere not significantly different from those in our study (p > 0.1). One limitation of our study may be that the preoperative thromboxane B2 sample was drawn via venipuncture, which may account for the wide variation in thromboxane B; However, other samples that were drawn via an arterial line showed a consistently wide range until after the patient was off CPB. There was a consistent decrease in variability postoperatively and on the first postoperative day thromboxane B, levels were similar to those reported in adults before CPB. Davies, Sobel, and Salzman? noted no difference in the levels of thromboxane B, in mixed venous samples compared to blood drawn from the left ventricle. Numerous reports show that prostaglandins are important to the proper functioning of the cardiovascular system. 1,3, 4, 6 Exogenous prostaglandin I, has been used in animal experiments and in human patients in an attempt to treat various diseases and in the CPB circuit for various theoretical benefits. 10, II, 15·Z0 However, optimal levels of prostacyclin and thromboxane A z, or the ratio of prostacyclin to thromboxane A z, have not been
Prostacyclin and thromboxane
77
defined, nor have usual levels of 6-keto-PGF la or thromboxane B, been reported in groups of children with or without congenital heart disease. Prostacyclin is elevated in response to myocardial ischemia and is protective of ischemic and hypertrophied hearts." 15, 16,ZO,ZI The effect of thromboxane B, on the heart is poorly understood. Although thromboxane B, was found to be elevated in the coronary sinus blood in patients with unstable angina, it is not known if that is a cause of ischemia or an effect." To selectively raise prostacyclin by exogenous means without a concomitant rise in thromboxane Bz may produce an as yet unrecognized undesirable effect. Further differentiations of normal for these substances for various age groups and various disease states must be correlated with clinical effects in order to benefit the cardiac surgical patient. We acknowledge the valuable laboratory assistance provided by Joan Spencer and the assistance provided by Linda Stokes, Respiratory Therapist, and Lynn Gilson, Anesthesia Technologist, in the preparation of this paper. REFERENCES Dusting GJ, Moncada S, Vane JR: Prostaglandins, their intermediates and precursors. Cardiovascular actions and regulatory roles in normal and abnormal circulatory systems. Prog Cardiovasc Dis 21:405, 1979 2 Ylikorkala 0, Saarela E, Viinikka L: Increased prostacyclin and thromboxaneproduction in man during cardiopulmonary bypass. J THORAC CARDlOYASC SURG 82:245, 1981 3 Vane JR: Prostaglandins and the cardiovascular system. Br Heart J 49:405, 1981 4 Watkins WD, Peterson MB, Kong DL, Kono K, Buckley MJ, Levine FH, Philbin OM: Thromboxaneand prostacyclin changes during cardiopulmonary bypass with and without pulsatile flow. J THORAC CARDlOyASC SURG 84:250, 1982 5 Kobinia GS, La Raia PJ, Peterson MB, D'Ambra MN, Watkins WD, Austen WG, Buckley MJ: Cardiac prostacyclin kinetics during cardiopulmonary bypass. J THORAC CARDIOyASC SURG 88:965, 1984 6 Jacobsen DC: Prostaglandins and cardiovascular disease-A review. Surgery 93:564, 1983 7 Nadler JL, Velasco JS, Horton R: Cigarette smoking inhibits prostacyclin formation. Lancet 1:1248, 1983 8 Newman WH, Frankis MB, Halushka PV: Increased myocardial release of prostacyclin in dogs with heart failure. J Cardiovasc Pharmacol 5: 194, 1983 9 Davies GC, Sobel M, Salzman EW: E!evated plasma fibrinopeptide A and thromboxaneB2 levels during cardiopulmonary bypass. Circulation 61:808, 1980 10 Faichney A, Davidson KG, Wheatley OJ, Davidson MB, Walker 10: Prostacyclin in cardiopulmonary bypassoperations. J THORAC CARDIOyASC SURG 84:601, 1982 11 Malpass TW, Amory DW, Harker LA, Ivey TO, WiI-
78
12
13
14
15
16
17
The Journal of Thoracic and Cardiovascular
Fleming et al.
Iiams DB: The effect of prostacyclin infusion on platelet hemostatic function in patients undergoing cardiopulmonary bypass. J THoRAc CARDIOVASC SURG 87:550, 1984 Nolan RD, Dusting GJ, Jakubowski J, Martin TJ: The pericardium as a source of prostacyclin in the dog, ox and rat. Prostaglandins 24:887, 1982 Dusting GJ, Nolan RD, Woodman QL, Martin TJ: Prostacyclin produced by the pericardium and its influence on coronary vascular tone. Am J Cardiol 52:28A, 1983 Mehta J, Mehta P: Prostacyclin and thromboxane B2 production by human cardiac atrial tissue. Am Heart J 109:1,1985 Melin JA, Becker LC: Salvage of ischemic myocardium by prostacyclin during experimental myocardial infarction. J Am Coll Cardiol 2:279, 1983 Yui Y, Nakajima H, Kawai C, Murakami T: Prostacyclin therapy in patients with congestive heart failure. Am J Cardiol 50:320, 1982 Zusman RM, Rubin RH, Cato AE, Cocchetto DM, Crow JW, Tolkoff-Rubin N: Hemodialysis using prostacyclin
Surgery
18
19
20
21
instead of heparin as the sole antithrombotic agent. N Engl J Med 304:934, 1981 Aren C, Feddersen K, Radegran K: Effects ofprostacyclin infusion on platelet activation and postoperative blood loss in coronary bypass. Ann Thorac Surg 36:49, 1983 DiSesa VJ, Huval W, Lelcuk S, Jonas R, Maddi R, Lee-Son S, Shemin RJ, Collins 11, Hechtman HB, Cohn LH: Disadvantages of prostacyclin infusionduring cardiopulmonary bypass. A double blind study of 50 patients having coronary revascularization. Ann Thorac Surg 38:514,1984 Povzhitkov M, Haendchen RV, Meerbaum S, Fishbein MC, Shell W., Corday E: Prostaglandin E) coronary venous retroperfusion in acute myocardial ischemia. Effects on regional left ventricular function and infarct size. J Am Coll Cardiol 3:439, 1984 Hirsh PD, Hillis D, Campbell WB, Firth BG, Willerson JT: Release of prostaglandins and thromboxane into the coronary circulation in patients with ischemic heart disease. N Engl J Med 304:685, 1981