The patients were premedicated with morphine and scopolamine and anesthetized with dehydrobenzperidol, fentanyl, and nitrous

Transcription

1 Correlation of Hypotensive Effect of Plasma Protein Fraction with Prekallikrein Activator Activity: A Clinical Study in Patients Having Open-Heart Surgery Jussi Heinonen, M.D., Kaija Peltola, M.D., Jaakko-Juhani Himberg, M.D., and Hannu Suomela, Ph.D. ABSTRACT We studied the hemodynamic responses to the administration of 20 ml of three protein solutions having different prekallikrein activator (PKA) activities in patients undergoing cardiac operation. PKA levels of the two plasma protein fraction (PPF) solutions studied were 193% and 6Oy0, respectively, while that of the albumin solution was S 3%. A significant correlation was found between the PKA activity of the test solutions and the decrease in systemic vascular resistance produced by the solutions during cardiopulmonary bypass and in the postoperative period. Vasodilatation was considerably more marked in the postoperative phase than during cardiopulmonary bypass. Since cardiac output increased simultaneously with the decrease in vascular resistance in the postoperative period, the hypotension observed did not fully reflect the degree of vasodilatation. In some patients, substantial vasodilatation was observed also after the administration of albumin solution with low PKA level. We conclude that although the PPF-induced vasodilatation correlates with the PKA activity of the solution, occasional hypotensive reactions are probably due to mechanisms other than those linked with increased PKA level. Hypotensive reactions to rapid infusion of plasma protein fraction (PPF) in surgical patients were documented in Australia [l, 21 and in the United States [3, 41 early in the 1970s. From the Cardiothoracic Anesthesia Group, Department of Anesthesia, Helsinki University Central Hospital, and Finnish Red Cross Blood Transfusion Service, Helsinki, Finland. Presented at the Joint Symposium on Standardization of Albumin, Plasma Substitutes and Plasmaphoresis, Geneva, Switzerland, Nov 10-12, Accepted for publication Apr 14, Address reprint requests to Dr. Heinonen, Department of Anesthesia, Helsinki University Central Hospital, Haartmaninkatu 4, SF Helsinki 29, Finland. Subsequently, PPF-induced hypotension was observed also during large-volume plasma exchange [51. In 1978, Alving and co-workers [6] presented evidence that prekallikrein activator (PKA) (Hageman-factor fragments) in PPF is responsible for the hypotensive reactions reported in surgical patients. At the same time, other investigators ascribed the vasodilatation observed during infusion of PPF in dogs, entirely to acetate buffer present in PPF [7]. During a period of a few months, we observed numerous incidents of severe hypotensive reactions to rapid administration of a PPF solution that contained no acetate. Therefore, we decided to study the hemodynamic effects of two lots of PPF having markedly different PKA activities and those of an albumin solution with a negligible PKA activity. Material and Methods Hemodynamic responses to the administration of a small amount of PPF solution were studied in 10 patients undergoing cardiac operation (Group 1). The two PPF solutions with different PKA levels were given to each patient in two phases: (1) during cardiopulmonary bypass and (2) during recovery from anesthesia. The hemodynamic effects of a 4% albumin solution were studied in 10 additional patients undergoing cardiac operation (Group 2). The mean age of the patients in Group 1 was 52.3 years (range, 31 to 62 years) and in Group 2, 48.7 years (range, 30 to 60 years); the mean weights were 73.0 f 3.4 kg (+ standard error of the mean) and 81.3 f 3.4 kg, respectively. Most of the patients underwent myocardial revascularization. The patients were premedicated with morphine and scopolamine and anesthetized with dehydrobenzperidol, fentanyl, and nitrous by The Society of Thoracic Surgeons

2 245 Heinonen et al: Plasma Protein Fraction Correlated with Prekallikrein Activator oxide. Muscle relaxation was provided by pancuronium bromide. The prime of the heart-lung machine consisted of 1,000 ml of gelatin solution (Haemaccel), 1,000 ml of 0.9% saline solution, 500 ml of 1.4% sodium bicarbonate, and 1,000 ml of blood. During cardiopulmonary bypass the estimated flow was 2.2 to 3.0 liters per minute per square meter of body surface area, and it was kept constant during the study period. In 3 patients in Group 1, sodium nitroprusside was infused at a constant rate (0.43 to 1.36 pg per kilogram of body weight per minute) into the oxygenator. Postoperatively, partial pressure of arterial carbon dioxide was maintained within normal limits, and partial pressure of arterial oxygen was held at a level of more than 90 mm Hg by adjusting the minute volume and oxygen concentration delivered by the Servo ventilator. Sodium nitroprusside (0.25 to 1.60 pglkglmin) was infused intravenously at a constant rate in 2 patients in Group 1 and in 2 patients in Group 2; dopamine (1.36 to 3.30 pglkglmin) was infused intravenously at a constant rate in 3 patients in Group l and in 2 patients in Group 2. No other drugs were given 30 minutes before or during the study period. The test PPF solutions were given in blind fashion and in random order. When the systemic arterial pressure had been stabilized during cardiopulmonary bypass, 20 ml of the one test solution was injected rapidly into the arterial end of the disc oxygenator. The other test solution was not given until arterial pressure had returned to the control value. During this phase of the study, rectal temperature was 32.8 k 0.3"C and toe temperature, 30.7 f 0.4"C. When arterial pressure, heart rate, and filling pressures had remained stable for at least 30 minutes postoperatively, 20 ml of the one PPF solution was injected rapidly intravenously, and the other test solution was given only after hemodynamic values had returned to the control level. In this phase of the study, rectal temperature was 38.0 & 0.3"C and toe temperature, 33.9 f 0.6"C. The. patients in Group 2 were given 20 ml of albumin solution during bypass and during recovery from anesthesia. Arterial pressure was measured through a cannula inserted into the radial artery, and a Swan-Ganz thermodilution catheter allowed measurement of pulmonary arterial pressure, pulmonary capillary wedge pressure, and central venous pressure. Intravascular pressures were recorded as electronic means at endexpiration, and mid-thorax served as the zero reference point. Cardiac output was measured with the aid of a cardiac output computer (Finnomedical model 804). Cardiac index, stroke volume index, and systemic and pulmonary vascular resistances were calculated using standard formulas. The two PPF solutions were made of Cohn fraction V and the albumin solution of Cohn fraction VI, the final stages of preparation after dissolving the protein powder being similar. The protein content of the solutions was 39.7 to 40.8 gm per liter. The electrophoretically determined albumin fraction of the PPF solutions was 95% and that of the albumin solution, >97%. Acetate was not used during the processing. The PKA activity was measured according to a modification of the method described by Snape and co-workers [8]. Fifty microliters of diluted sample was incubated with 200 pl of prekallikrein substrate for one hour at 37 C. Fifty microliters of this solution was transferred into plastic cuvettes containing 500 p1 of a 0.8 mm solution of synthetic substrate S-2302 (Kabi). The increase in absorbancy at 405 nm and 37 C was recorded for 30 minutes. Dilutions of the sample and prekallikrein substrate were selected so that a change in absorbancy to time was linear during the measurement. The absence of kallikrein in prekallikrein substrate and of other esterolytic activities in the sample capable of splitting the substrate S-2302 was ascertained by using blanks. The calibration curve was made using the 100% PKA standard [6] obtained from the Bureau of Biologics (Department of Health and Human Services) in Bethesda, MD. The PKA activities of the two PPF solutions were 193% (192.6 f 3.1%; n = 16) and 60% (59.5 k 1.0%; n = 24), respectively; the PKA activity of the albumin solution was negligible (S3%). PKA activities also were measured at the Bureau of Biologics using [3H] TAME assay; the results were 215%, 5570, and 1%, respectively.

3 ~ ~ 246 The Annals of Thoracic Surgery Vol 33 No 3 March 1982 Table 1. Control Hemodynamic Values before Administration of Two Plasma Protein Fraction Solutions with Different Prekallikrein Activator Levels and before Administration of Albumin Solution" Group 1 (n = 10) Group 2 (n = 10) Hemodynamic Variables Before PPF-193 Before PPF-60 Before Albumin CARDIOPULMONARY BYPASS MAP (mm Hg) 70.2 f f f 4.7 POSTOPERATIVE PHASE MAP (mm Hg) MPAP (mm Hg) PCWP (mm Hg) CVP (mm Hg) HR (beatlmin) SI (mllm') CI (L/min/m*) SVR (dyne sec cmp5) PVR (dyne sec cmp5) _ f f f f f f ,117 f f f f f k f f f ,287 f f C f k C f f f ,148 f f 30 adata shown as mean f standard error of the mean. PPF = plasma protein fraction; MAP = mean arterial pressure; MPAP = mean pulmonary arterial pressure; P M = pulmonary capillary wedge pressure; CVP = central venous pressure; HR = heart rate; SI = stroke index; CI = cardiac index; SVR = systemic vascular resistance; PVR = pulmonary vascular resistance. Results The control hemodynamic values obtained before the administration of the two PPF solutions and albumin solution did not differ significantly from each other (Table 1). During cardiopulmonary bypass, PPF with 193% PKA activity (PPF-193) produced a significant decrease in perfusion pressure, whereas the mean change after PPF with 60% PKA activity (PPF-60) and after albumin was minimal (Table 2). Four patients showed no decrease in arterial pressure following PPF-60, whereas 1 patient reacted exceptionally with a 30% decrease from the 93 mm Hg perfusion pressure. In general, arterial pressure began to decrease about 25 seconds after the administration of PPF, and the hypotensive period was significantly longer after PPF-193 than after PPF-60 (189 f 28 versus 62 _+ 15 seconds, respectively; p < 0.01). Since flow was maintained constant during cardiopulmonary bypass, the decrease in arterial pressure represents a decrease in systemic vascular resistance. Postoperatively, both PPF solutions produced a significant decrease in arterial pressure, but PPF-193 was followed by a considerably more marked hypotension (see Table 2; Fig 1). Hypotension began to manifest itself 29 f 3 seconds following injection of PPF-193 and 45 f 3 seconds after PPF-60 (p < 0.001), lasting 201 f 24 and 102 f 19 seconds, respectively (p < 0.01). Albumin also produced hypotension in 2 of 10 patients who experienced a 16 and 17% decrease in arterial pressure respectively. At the time of the lowest arterial pressure, the changes in filling pressures and pulmonary arterial pressure were small and not significantly different after the administration of the three test solutions (see Table 2). Systemic and pulmonary vascular resistances decreased significantly after the administration of both PPF solutions, but the decrease in systemic vascular resistance was considerably greater after PPF- 193 (see Fig 1). Since cardiac flow increased simultaneously with the decrease in vascular resistances, the changes in systemic and pul-

4 247 Heinonen et al: Plasma Protein Fraction Correlated with Prekallikrein Activator Table 2. Mean Percentage Changes in Hernodynamic Values after Administration of 20 ml of Two Plasma Protein Fraction Solutions with Different Prekallikrein Activator Activity, and of Albumin Solutiona Hernodynamic Variables Group 1 Group 2 PPF-193 PPF-60 Albumin CARDIOPULMONARY BYPASS MAP f 3.4'3' -4.7 f f 2.2 POSTOPERATIVE PHASE MAP MPAP PCWP CVP HR SI CI SVR PVR f 2.1d3' +1.1 k f f f O.gb +8.8 k f f 3.8"'" k f k f f k f k f k f f * f k f f f _+ 6.6 adata shown as mean k standard error of the mean. < 0.05, < 0.01, "p < 0.001: significant difference between PPF-193 and PPF-60; paired t test. "p < 0.05, 'p < 0.001: significant difference between PPF-193 and albumin; Student t test. Abbreviations same as in Table 1. Change MAP CI SVR (%I 1 0 PPF-193 Fig 1. Mean percentage changes (k standard error of the mean) in mean arterial pressure (MAP), cardiac index (CI), and systemic vascular resistance (SVR) after the administration of huo plasma protein fraction (PPF) solutions with different prekallikrein activator activities (293% and 60%) and of albumin solution (ALB) in patients recovering from anesthesia after cardiac operation. monary arterial pressures were smaller than those in resistances, pulmonary arterial pressure remaining unchanged. A significant correlation was found between the PKA activity of the protein solutions and the decrease in systemic vascular resistance both during cardiopulmonary bypass and in the postoperative period (Fig 2). The hemodynamic changes after PPF solutions were not associated with flushing, increase in recorded airway pressures, or any other findings deviating from normal in these patients. Comment Acute vasodilatation provoked by PPF [2, 41 suggested involvement of the kinin system. However, a product having bradykinin content equal to that in normal plasma was associated with several reports of hypotension in the United States. Investigations by the Food and Drug Administration revealed that the implicated lots of PPF had appreciable PKA activity, whereas the lots not reported to be associated with adverse reactions exhibited low PKA ac-

5 248 The Annals of Thoracic Surgery Vol 33 No 3 March 1982 SVA lichangel -504 A SVAfwhange t PKA activitpj -30 y-012x-228 ' -401 r= PKA aclivityw B Fig 2. Relationship betiween the prekallikrein activator (PKA) activity in the protein solutions and the changes in systemic vascular resistance (SVR) observed after the administration of these solutions in cardiac surgical patients in two phases of the study. A is the postoperative phase and B, the cardiopulmonary bypass phase. tivity [6]. According to these authors, PKA would be expected to induce a stronger hypotensive response than exogenous bradykinin, since PKA, by acting on prekallikrein, could continue to generate bradykinin in excess of that being inactivated during a single pass through the lungs. Olinger and associates [7] ascribed the PPF-induced vasodilatation in dogs entirely to acetate buffer and consequently expressed some doubt about the role of PKA in PPF-induced hypotension, since the solutions studied by Allving and co-workers [61 contained acetate. In our patients undergoing cardiac operation, the small amounts of PPF solutions devoid of an acetate but Containing marked PKA activity induced a substantial degree of vasodilatation. Furthermore, the degree of vasodilatation correlated well with the PKA activity of the test solutions. We have also studied the effect of rapid administration of a crystalloid solution containing acetate on perfusion pressure during cardiopulmonary bypass in patients undergoing cardiac operation. Although the volume given and the acetate content of our solution were equal to those used by Olinger and coworkers [7] in dog experiments, we could find only minimal changes in perfusion pressure (increase of 0.2 k 3.4% in 6 patients).* It therefore seems likely that the differing results of Olinger and colleagues are due to species differences [91. It is, however, possible that the hemodynamic effects of PPF containing PKA may be accentuated by acetate. The moderate decrease in systemic vascular resistance observed in some of our patients after administration of the albumin solution is obviously due to causes other than PKA-induced bradykinin effect. It should be noted, however, that occasional lots of albumin solution may have a significant PKA activity [61. In 1971 and 1972, it was reported that reactions to PPF occurred during cardiopulmonary bypass [l, 31. Since bradykinin cannot be inactivated in the lungs of patients on cardiopulmonary bypass, such patients have been expected to have an increased susceptibility to PKA. However, the majority of the reactions reported by Alving and associates [61 occurred in patients who were not undergoing bypass procedures. We, too, have observed numerous hypotensive reactions in patients undergoing noncardiac operation, and in the present study, the PPF-induced vasodilatation was more prominent in the postoperative phase than during cardiopulmonary bypass. These findings and the results of bradykinin measurement during cardiopulmonary bypas9 [lo] suggest that pulmonary circulation is not essential for the elimination of bradykinin. Our results, like those of others [2], show that in patients with intact circulation, the PPFinduced fall in blood pressure does not reflect fully the decrease in vascular resistance as cardiac output is increased. The poor cardiac com- *Heinonen J, Peltola K: Unpublished data, 1980.

6 249 Heinonen et al: Plasma Protein Fraction Correlated with Prekallikrein Activator pensation, due to severe cardiac disease or inadequate preload or both, may be the principal reason why the PPF-induced reactions are seen most often in critically ill patients. Increasing awareness of the vasodilatory effect of PPF solutions should prevent situations in which PPF is infused rapidly into hypovolemic patients. Whenever a PPF solution is used, close monitoring of the blood pressure of the patients is mandatory. Until PPF solution devoid of hypotensive factor is available, the administration of albumin should be considered to be a safer alternative in critically ill patients. Interestingly, however, the hemodynamic effects of PPF-60 in many of our cardiac patients were in fact advantageous, as the unloading of the heart was associated with a significant increase in cardiac output and only a small decrease in arterial pressure. The study protocol was approved by the Ethical Committee of the hospital. We thank Drs. Barbara Alving and B. L. Mason at the Plasma Derivatives Branch of the Bureau of Biologics for providing PKA reference and carrying out PKA assay. References 1. Harrison GA, McCulloch CH, Robinson M, et al: Hypotensive effects of stable plasma protein solution (SPPS): a preliminary communication. Med J Aust 2:1040, Torda TA, Harrison GA, McCulloch CH, et al: Circulatory effects of stable plasma protein solution (SPPS). Med J Aust 1:798, Bland JHL, Laver MB, Lowenstein E: Hypotension due to 5 per cent plasma protein fractions. N Engl J Med 286:109, Bland JHL, Laver MB, Lowenstein E: Vasodilator effect of commercial 5% plasma protein fraction solutions. JAMA 224:1721, Isbister JP, Biggs JC: Reactions to rapid infusion of stable plasma protein solution during large volume plasma exchange. Anaesth Intensive Care 4: 105, Alving BM, Hojima Y, Pisano JJ, et al: Hypotension associated with prekallikrein activator (Hageman-factor fragments) in plasma protein fraction. N Engl J Med 29956, Olinger GN, Werner PH, Bonchek LI, Boerboom LE: Vasodilator effects of the sodium acetate in pooled protein fraction. Ann Surg 190:305, Snape TJ, Griffith D, Vallet L, Wesley ED: The assay of prekallikrein activator in human blood products. Dev Biol Stand 44:115, Collins JA: Discussion of Olinger et a1 [7] 10. Ellison N, Behar M, MacVaugh H 111, Marshall BE: Bradykinin, plasma protein fraction, and hypotension. Ann Thorac Surg 29:15, 1980