The Effect of Methylprednisolone on Complement Activation during Cardiopulmonary Bypass

____________ _ A prospective randomized trial in 80 patients having heart surgery was undertaken to test the hypothesis that steroids (ST) (methylprednisolone, 30 mg/kg) would prevent or reduce complement activation during cardiopulmonary bypass (CPB) with membrane (MEM) or bubble (BUB) oxygenators. Group I patients had MEM and ST; Group II, MEM without ST; Group III, BUB and ST; Group IV, BUB without ST. Complement activation was assessed by C3a radioimmunoassay determinations obtained before, during and after CPB. When Group I was compared to Group II, there was a significant increase in C3a generation in the presence of ST. Also, when Groups I and III combined both with ST were compared to Groups II and IV both without ST a significant increase in C3a with ST was observed. In contrast, comparing MEM to BUB (Group I vs. III or II vs. IV) no significant difference in C3a was seen. Thus the hypothesisthat ST would prevent or reduce complement activation was not confirmed and ST appears to actually increase complement activation. Also, MEM oxygenators were not associated with less activation of C3a than BUB oxygenators. Direct communications to: LeRoy H. Ferries, B.S., C.C.P., Marshfield Clinic, 1000 N. Oak Avenue, Marshfield. WI 54449 Supported in part by the Marshfield Clinic, Marshfield Medical Foundation, and Upjohn Diagnostics. Presented at AmSECT's 22nd International Conference, May 21-23, 1984, Las Vegas, NY. Volume 16, Number 3, Fall 1984 Introduction ____________ _ Complement activation occurs during cardiopulmonary bypass (CPB) because complement, a plasma protein, comes in contact with foreign surfaces of the pump oxygenator and the blood-gas interface. The degradation products, or split products, of complement activation include the anaphylatoxins, C3a and C5a. These anaphylatoxins are thought responsible for the occasionally observed damaging physiologic effects of CPB, including lung and kidney dysfunction as well as post-operative bleeding problems. One current theory is that these degradation products cause a whole body inflammatory reaction of varying severity. The severity of damage varies with duration of CPB, age of the patient, and peak level of anaphylatoxin reached. • If this theory is true, it would be of clinical importance to eliminate or minimize C3a and C5a production. 2 A prospective randomized double blind trial in 80 adult patients having heart surgery with CPB was done to test the hypothesis that steroids, with their anti-inflammatory effect, 3.4 would prevent or reduce complement activation. Additionally, the trial compared membrane (MEM) to bubble (BUB) oxygenators to test the hypothesis that the membrane oxygenator would be associated with less complement activation. A new, highly sensitive and reproducible radioimmunoassay for C3a was used for these studies. Materials and Methods. _______ _ Informed ConsentThe Institutional Review Board gave approval of the study. All patients studied gave informed consent prior to their being The Journal of Extra-Corporeal Technology 83 TABLE 1 Group Comparisons with Case Mix Group I Group II Group III Group IV Age 60.4 ± 11.0 60.4 ± 12.0 57.5 ± 9.5 57.6 ± 9.7 Sex M = 70% (14) M = 85% (17) M = 80% (16) M = 80% (16) Weight 79.8 ± 114.0 81.2 ± 11.3 82.6 ± 9.3 82.4 ± 16.6 Bypass Time 93.1 ± 39.0 127.7 ± 47.2 130.6 ± 73.5 118.3 ± 45.7 X-Clamp Time 52.6 ± 23.4 71.7 ± 31.0 71.5 ± 34.1 66.0 ± 31.4 24 Hour Chest Drainage 603 ± 326 702 ± 491 660 ± 222 981 ± 702 Surgeon A= 13, 8=5, C=2 A= 10, 8=9, C= I A= 12, 8=8, C=O A=9, 8=11, C=O CASE MIX CABGs 90% (18) 70% (14) 85% (17) 90o/c ( 15) Re-op. CABGs 0 20% (4) 0 15% (3) Single Valve 0 0 0 5% (I) Double Valve 5% (I) 5% (I) 0 5% (I) Valve & CABGs 0 5% (I) 10% (2) 0 ASD 5% (I) 0 5% (I) 0 A comparison of common variables demonstrated no significant difference between the four groups. entered into the randomization process. Study Design and Randomization Process Eighty adult patients having heart surgery with CPB were studied. Half were to receive steroids (ST) (methylprednisolone, 30 mg/kg) and half no ST. Half were to have MEM and half BUB oxygenators. Group I patients (n = 20) were to get MEM and ST; Group II (n = 20), MEM without ST; Group III (n = 20), BUB and ST; Group IV (n = 20), BUB without ST. The randomization was planned so as to create four groups of 20 patients by random number allocation using a standard table of random numbers. 5 After the allocation to the four groups was determined the total number sequence from the first patient to the 80th patient was given to the pharmacy. The pharmacy then dispensed the drug or placebo and designated the appropriate oxygenator according to the patient assignment. Drugs were provided in coded syringes. Neither surgeon, perfusionist nor laboratory technician knew which patient was in which group. By necessity the perfusionist and the surgeon knew which oxygenator was being used but the assignment was done by the pharmacy department. PatientsThe exclusions to participating in the study were, 1) refusal to consent, 2) emergency operations, and 3) Friday operations (no C3a determinations were done that day). Also, only the first two patients on the operating schedule for each 84 The Journal of Extra-Corporeal Technology day were asked to participate due to the time constraints of blood collection. Types of operations in the 80 study patients were, coronary artery bypass graft (CABG) 64, redo CABG 7, single valve replacement 1, double valve replacement 3, valve replacement and CABG 3, and atrial septal defect 2. (Table 1) There were no deaths. C3a DeterminationsBlood samples for C3a were drawn, 1) pre-bypass, 2) two minutes, 3) 10 minutes, 4) 20 minutes, and 5) 30 minutes on bypass. Thereafter they were drawn at 30 minute intervals, plus at the end of bypass and one hour post-bypass. The samples were drawn from the patient's arterial line preand post-CPB and from the oxygenator arterial sampling port during CPB. blood was anticoagulated with 7 mM EDT A and the plasma separated within 10 minutes and frozen on dry ice. All plasma samples were stored at 85°C until assayed. Samples from each individual were analyzed in one run. C3a was determined by radioimmunoassay. All reagents were prepared from a commercially available kit. a I 125 C3a des Arg was mixed with patient's sample and a standardized quantity of rabbit Anti-C3a des Arg. The reaction mixture was incubated for 30 minutes at room temperature. A standardized amount of goat anti-rabbit antisera was added and incubation continued for an additional "Upjohn Diagnostics, Kalamazoo, Ml 49001 Volume 16, Number 3, Fall 1984 30 minutes at room temperature. Two milliliters of isotonic saline were added and the precipitate separated by centrifugation at 2,000 x g for 10 minutes. All samples were counted in a gamma counter such that the 0 ng (Bo) standard yielded 10,000 counts. Counts per minute in each standard were corrected for nonspecific binding and the per cent bound calculated as a ratio of the Bo standard. The results were plotted on log-logit paper as a function of C3a des Arg concentration. The C3a concentration in each sample was determined after correction for nonspecific binding. Materials The bubble oxygenator used was the Shiley S-100Ab and the membrane, the Travenol (TMQ®)c oxygenator. The methylprednisolone,d or placebo, was added to the bypass prime and circulated prior to initiation of CPB. Bypass Technique Anticoagulation was managed with an activated clotting time (ACT) protocol. An initial loading dose of porcine mucosal heparine (2 mg/kg) was given and then adjusted to maintain an ACT of 460-520 seconds. Blood flows during CPB were maintained at 70 cc/kg, not to exceed 6.0 L!min. Moderate systemic hypothermia of 28°C was maintained for each patient. A crystalloid cardioplegia solution at 4°C was perfused into the aortic root to achieve and maintain cardiac standstill and myocardial hypothermia. In addition, topical cooling with Ringer's lactate slush was used. The extracorporeal circuit was composed of Tygon PVC and Travenol pumphead PVC tubing with an unfiltered Travenol cardiotomy reservoir and an Extracorporeal lntersept 20 u filter/bubble trap. r Bypass systems were primed with 2,500 cc. Normosol-R,g 50 gm. Mannitol,h 40 mg heparin, and 44.6 mEq sodium bicarbonate. i Whole blood or packed cells were added to maintain a patient/pump hematocrit of approximately 24%. At the end of CPB heparin was neutralized with protamine sulfate) bShiley Inc., Irvine, CA 92714 cTravenol Laboratories, Inc., Deerfield, IL 60015 dUpjohn Company, Kalamazoo. MI 49001 cElkins-Sinn, Inc., Cherry Hill, NJ 08034 Extracorporeal, Inc., King of Prussia, PA 19406 gAbbott Laboratories, North Chicago, IL 60064 hlnvenex, Chagrin Falls, OH 44022 iBristol Laboratories, Syracuse, NY 13201 jEli Lilly and Company, Indianapolis, IN 46285 Volume 16, Number 3, Fall 1984 Statistical Analyses ________ _ Statistical comparisons were made with a package developed for microcomputers. k Comparisons of more than two populations were performed by standard analysis of variance, while two population comparisons were made by the student-t test. Significance was defined as p < 0. 05. Results _____________ _ Table 1 compares the four groups of patients studied in regard to age, sex, weight, bypass time, cross-clamp time, volume of 24 hour chest drainage, surgeon and case mix. None of these variables differed significantly in the four groups. Table 1 is an overall measure of how well the randomization process worked.