A prospective study of the risk of an immediate adverse reaction to protamine sulfate during cardiopulmonary bypass surgery John M. Weiler, MD, Melanie A. Gellhaus, BA, RN, James G. Carter, MD, Ronald L. Meng, MD, Patricia M. Benson, BSN, RN, Rachel A. Hottel, BA, RN, Kay B. Schillig, RN, Arthur B. Vegh, MD, and William R. Clarke, PhD Iowa City, Iowa Protamine sugate administration may cause life-threatening reactions. We prospectively examined the incidence of immediate adverse reaction after protamine in 243 patients who underwent cardiopulmonary bypass surgery. Twenty-six patients (10.7%) had reactions, and 1.6% had a precipitous drop in blood pressure immediately after protamine administration. Risk factors were previous exposure to protamine, diabetes, history of receiving protamine-containing insulin, and possibly vasectomy. However, neither a positive skin test nor a positive IgE ELJSA for antiprotamine antibody predicted that a patient would have a reaction. C4a levels were increased in patients who had reactions as compared with age-, sex-, and cardiac disease-matched patients who did not have reactions, suggesting a role for complement in some reactions. Immediate adverse reactions to protamine are very common, and alternative therapies are urgently needed to eliminate the use of protamine. (J ALLERGYCLIN IMMUNOL1990;85:713-9.)
Heparin and protamine sulfate areadministeredduring cardiopulmonary bypassto control coagulationbut also may be a major causeof morbidity and mortality in patients undergoing theseprocedures.lS7Thesetwo drugs are also often administered during cardiac catheterization7and occasionally during other cardiothoracic and vascular surgical procedures,8-”and after dialysis’* and leukapheresis.I3 Previous studies indicatethat a variety of immediateadversereactionsmay occur after protamine administration, including respiratory (wheezing), cardiovascular(hypotension, decreasedcardiac contractility, bradycardia, cardiac arrest, ventricular fibrillation, or hypertension), skin From the Departmentsof Medicine, Anesthesia, Surgery and Preventive Mzedicine,the University of Iowa, and the Iowa City VeteransAdministration Medical Center, Iowa City, Iowa. Supported by the National Institutes of Health Grant AI22350, United States Public Health Service, and by a Merit Review award from the VeteransAdministration. Presentedin part at the Forty-fifth Annual Meeting of the American Academy of Allergy and Immunology, SanAntonio, Texas, Feb. 24-March 1, 1989. Received for publication May 30, 1989. Revised Oct. 17, 1989. Accepted for publication Nov. 7, 1989. Reprint requests:John M. Weiler, MD, Division of Allergy/Immunology, Departmentof Internal Medicine, SW34EUniversity Hospital, University of Iowa, Iowa City, IA 52242. l/1/18012
31
C3a,C4a, C5a: The anaphylatoxm cleavage frag-
(flushing, urticaria, and angioedema), and hematologic (bleeding) reactions.‘.7More than 290,000 openheart operations and 690,000 cardiac catheterizations were performed in the United Statesin 1985, the last yearfor which thesedataareavailable.14Nevertheless, no prospectivestudiesexist to documentthe incidence of immediate adverse reactions after protamine administration during cardiopulmonary bypass surgery. We previously reported a death in a patient who received protamine during surgery when other causes for a fatal outcome were excluded.’ Furthermore, we havebeenawarein our institution of many less serious cases in which patients apparently reacted to protamine that was administeredafter heparin during surgery or cardiaccatheterization.*Theseseriousadverse reactions may be expected to increase in number as more cardiopulmonary bypasssurgery proceduresare performed annually. This study was designed (1) to examine prospectively the incidence of an immediate adverse reaction after administration of heparin and protamine during cardiopulmonary bypass and (2) to determine the role that type I hypersensitivity and 713
714
Weiler
et al.
J. ALLERGY
complement activation may play in causing these reactions. MATERIAL AND METHODS Patient population All patients aged 18 years or older who were first observed at the University of Iowa Hospitals and Clinics cardiothoracic vascular surgery clinic for cardiopulmonary bypass surgery, which would be expected to require heparin and protamine, were eligible for this study. At the outset, we anticipated that it would be necessaryto access200 to 250 patients to be able to skin test approximately 100. The study was endedwhen we enrolled 248 patients, 85 of whom were skin testedbefore surgery. Patientswho were not skin testedeither refused skin testing, were receiving a drug that suppresseda positive skin test (for example, antihistamines), or were admitted directly to the surgical suite without first being admitted to the ward before surgery. Patients who had a previous seriousreaction to protamine were empirically pretreatedwith prednisoneto attempt to prevent a reaction and were not enrolled. Study design Patients were interviewed in the clinic or on the ward unit by a study nurse to determine past history that might suggest increased risk of a protamine reaction. We specifically asked about diabetes history and exposure to protamine-containing insulin,‘, I’. Is.I6 history of vasectomy that hasbeenreportedto result in antiprotamineantibodies,” history of fish allergy,‘*, I9 previous cardiac catheterization that required protamine,’ and previous surgical procedures that required protamine. Patients who gave consent were skin testedand had plasma obtained for ELISA testing. Staff who provided anesthesiafor patients undergoing cardiopulmonary bypass surgery were instructed by one of the investigators (J. G. C.) about the suspectednature of protamine reactions. The anesthesiologistrecordedcardiac parametersduring the procedureon a special form designed for this study; theseparameterswere monitored moreclosely just before, during, and for 5 minutes after protamine was administered. Plasmawas obtained from the central line 5 minutes after protamine administration. Historical information about each patient was obtained from the initial interview and from review of charts after dischargeof each patient. Protamine
administration
and dosage
Protamine was obtained from Eli Lilly & Co. (Indianapolis, Ind.) and was administered slowly during 3 to 7 minutes when the patient was hemodynamically stableafter bypasslines had beenremoved.The averageprotaminedose was about 300 mg, basedon the activated clotting time. In somepatients additional protamine was administered after this initial dose. However, there were no further immediate adversereactions after this additional protamine had been administered.
Determination
CLIN. IMMUNOL. APRIL 1990
of adverse reactions
We graded reactions on an objective basis. A mild reaction required a 20 to 29 mm drop, a moderatereaction required a 30 to 49 mm drop, and a severereaction required at least a 50 mm drop (or total vascular collapse) in systolic or diastolic blood pressurewithin 5 minutes after protamine administration. This 5-minute period was basedon previous experiences, indicating that all reactions occur within this time with catastrophicreactions usually occurring within 1 minute. Other reactions that were recordedin separatecategories were wheezing and flushing, urticaria, or angioedema. Blood drawing
and preservation
Plasma was obtained by collecting blood into vacuum tubes containing ethylenediaminetetraacetate.The tubes were centrifuged, and the plasma was frozen at -75” C. The first blood specimenwasdrawn on the day of admission. The second blood specimen was obtained 5 minutes after protamine administration during surgery. Skin testing
protocol
Skin testing was performed with protamine (Eli Lilly & Co.) diluted to 0.1 mg/ml, 0.01 mg/ml, and 0.001 mg/ml in ALBAY (Hollister-Stier Laboratories, Spokane, Wash.) diluent for allergenic extract. First, histamine and diluent controls were applied by intradermal injection. Then protamine, at 0.1 mg/ml, was applied by prick test; if that test was negative and the controls were positive, the subjectwas testedintradermally with the 0.001 mg/ml followed by the 0.01 mg/ml and the 0.1 mg/ml solutions. Skin tests were read 15 minutes after application, and a positive test was a wheal that was at least 4 by 4 mm larger than the negative control wheal with some erythema. No adverse reactions were noted with skin testing, nor were late skin testreactions observed. Surgeons were informed of positive skin tests, although in no casewas any special medicine administered to mitigate a reaction nor was there any other change in therapy. IgE ELISA testing
procedure
We testedfor the presenceof IgE antibody in specimen one with a commercially available ELISA to detect antiprotamine IgE in plasma(AlerCHEK, Inc., Portland, Me.). Briefly, the assaywasperformedin microtiter wells to which protamine had been preimmobilized to the polystyrene walls; control wells were used to which protamine had not been immobilized. Plasmawas then addedto either a well containing protamineor a well that did not contain protamine and incubated for 30 minutes at room temperature,and the wells were washed. Then, affinity-purified, antihuman IgE horseradish-peroxidaseconjugate was added to each well, incubation was continued for 30 minutes at room temperature, and the wells were washed again. Finally, the developing solution containing dimethylsulfoxide, 3,3’,5,5’tetramethylbenzidine, and peroxide solution was added to
VOLUME NUMBER
Adverse
85 4
reactions
to protamine
sulfate
715
each well, and the presenceof color was examined. All ELBA testing was done in batches after the surgery had been completed during the same week to assure quality control.
it was possible to determine accurately whether the patient had received protamine previously; for the remaining patients, it could not be determinedto a certainty that the patients had previously received prot-
Nonreactor group
amine.
For each patient who had an immediateadversereaction, an age-, sex- and cardiac disease-matchednonreactorpatient was selected.This nonreactorpopulation was used for the complement studies and for studies that examined whether adversereactions to protamine were dose related.
Twenty-six of the 243 patientshad adversereactions to protamine (10.7%). Eleven reactions (4.5%) were mild, nine (3.7%) were moderate, and four reactions (1.6%) were severe.One patient with a moderatere-
Detection of C3a, C4a. and C5a fragments of compk3ment Plasma obtained from specimentwo was examined for the presenceof C3a, C4a, and C5a with a commercially available assay*O following manufacturer’s instructions (AmershamCorp., Arlington Heights, Ill.).
Statisticall tests Fisher’s exact test was used to test for differences in the proportions of patientswith previous exposureto protamine, presence 0-Fdiabetes, previous exposure to protaminecontaining Insulin, the presenceof a positive skin test to protamine, and between patients who had immediate adverse reactions after protamine administration and patients who did not have reactions. Student’s paired t tests were usedto comparemeanC3a, C4a, and C5a levels in patients exhibiting an immediate adverse reaction after protamine administration with age-, sex-, andcardiacdisease-matched nonreactor patients. One-tailed tests were used, since we hypothesizedthat eachvariable (for example, previous protamine exposure, history of diabetes mellitus, history of protamine-containinginsulin, positive skin test, andpositive IgE ELBA test) would increase but not decreasethe risk of an immediate adversereaction to protamine administration. The 5% level of significance was used in all comparisons. One of the authors (W. R. C.) served as statistical consultant.
RESULT!5 Two hundred forty-eight consecutive patients undergoing cardiopulmonary bypass surgery were enrolled in this study. Two hundred forty-three patients received protamine and had charts sufficiently complete to determine whether or not there had been an immediate adversereaction to protamine administration and the nature of the reaction. The remaining five patients did not receive pro&nine (two patients) or had records that were insufficient to determine whether a reaction had occurred (three patients). The averageageof the patientswas 6 1 years, and the range was 22 to 88 years. There were 59 female and 184 male patients. Nine male patients had vasectomy.No patient had a history of fish allergy. For 201 patients
action had audible wheezing, and another patient in this group had flushing of the skin. One patient had
wheezing only, and one patient had flushing of the skin only. Patients who had reactions were treated with volume replacement and inotropic agents, if this was needed. None of the patients required return to bypass, and all reactions subsided. Patients who had
mild, moderate, and severereactions received an averageof 310 mg (4.2 mg/kg), 276 mg (3.9 mg/kg), and 296 mg (4 mg/kg) of protamine, respectively. Age-, sex-, and cardiac diseased-matched patients who did not have reactions received an average of
335 mg (4.8 mg/kg), 328 mg (3.6 mg/kg), and 300 mg (3.8 mg/kg) of protamine, respectively. The differencesamong these groups were not significant. History of previous exposure to protamine (Table I) There were 2 1 reactions in 171 patients who had previously been exposed, and no reactions in any of the 30 who had not been preexposedto protamine (p < 0.05). It was not possible to determine accurately whether the patient had previously received protamine for five of 26 patients who had reactions. In thesecases,procedures(for example, cardiac catheterizations) were done in other institutions from which adequaterecords could not be obtained or the medication record could not be verified. Next, we examinedwhether a history of diabeteswould increase the risk of a reaction. Fifty-two of the 243 patients were diabetics, 25 had previously received protaminecontaining insulin, 14 had been treated with diet and oral agents,and 13 patients had beentreatedwith diet alone. Ten of the 52 diabetics (19.2%) had a reaction to protamine, whereas 16 of the 191 nondiabetics (8.4%) had an immediate adverse reaction (p < 0.05). Six of 25 diabetics (24%) who had beentreated with pro&mine-containing insulin had a reaction when protamine was administered during surgery as compared with four reactions in the 27 diabetics (14.8%) who had not been treated with protamine-containing insulin (p = 0.3 13). All the diabetics who had reactions had received protamine previously in insulin,
716 Weiler et al.
TABLE I. Characteristics protamine administration
J. ALLERGY
of 26 patients
who had immediate
adverse
reactions
CLIN. IMMUNOL. APRIL 1990
to
Presence of characteristic
History Previous protamine exposure when exposure could be determined Diagnosis of diabetesmellius Receiving protamine-containing insulin Presenceof a positive intradermal skin test to protamine Presenceof positive antiprotamine IgE ELISA
Yes
No
Reactions/No. of patients (%I
Reactions/No. of patients (%I
p Value
21/171 (12.3)
O/30 (0)
0.028
10152(19.2) 6125 (24)
16/ 191 (8.4) 201216 (9.3)
0.028 0.037
l/3 (33)
lo/82 (12.2)
0.344
3/ 12 (25)
171115(13)
0.286
at cardiac catheterization, or at previous surgery. There were too few male patients who had vasectomy to make firm conclusions about the role vasectomy played; however, it is interesting that two of the nine patients (22%) who had vasectomyhad reactions (one mild and one severereaction). Skin test data Of the 85 patients who were skin tested (Table I), only three had positive tests (3.5%), and one of the three patients had a reaction. Ten patients who had negative skin tests to protamine had reactions (12.2%). None of 10 diabetics who had received protamine-containing insulin had a positive skin test. One diet-controlled diabetic had a positive skin test, and there were two positive skin tests in 65 nondiabetic patients. The protamine concentrationat which all three patients had positive skin tests was 0.1 mg/ml. Two of the three patients with positive skin tests did have prior exposure to protamine, and the third patient probably had prior exposureto protamine at the time of catheterization but was placed in the category of patients who had unknown exposure because we could not verify the exposure. This patient also had vasectomy. Prevalence of antiprotamine IgE antibody by ELISA Twelve patientshad IgE directed againstprotamine, but none of these tests was strongly positive (Table I), and three of the 12 patients had reactions to protamine (25%). In contrast, 17 of 115 with a negative test (14.8%) had a reaction (p = 0.29). Furthermore, there was no associationbetween a positive skin test and a positive IgE ELBA test, since there were no positive IgE ELISA testsin the patients with positive
skin testsas comparedwith sevenpositive IgE ELBA tests in the 58 patients with negative skin tests (p = 0.68). Complement activation after protamine administration (Table II) C3a levels were significantly elevatedin all patient groups over levels expected in normal plasma, although there was no statistically significant increase in patients who had reactions as compared with the nonreactor patients. C4a levels were increased in all patient groups and were statistically significantly increased in the reactor group as compared with the nonreactorgroup. C5a levels remained normal in the nonreactorgroups and in the group with mild reactions but were mildly increasedin the groups with moderate and severereactions. DISCUSSION Immediate adverse reactions after pro&mine administration are common and may be very serious. Twenty-six of 243 patients (10.7%) we followed prospectively during cardiopulmonary bypasshad an immediate reaction manifested in 24 (9.9%) by a significant drop in blood pressure.Four patients (1.6%) had a precipitous drop in blood pressure.No previous large study prospectively examined the incidence of an immediate adverse reaction to protamine during cardiopulmonary bypass. The criteria for an immediate adversereaction in this study were objective and could easily be determined by the anesthesiologist. Unfortunately, it is difficult to diagnose more mild casesof anaphylacticand anaphylactoidreactionsduring surgery becauseit is not possible to assessthe more subtlecriteria for thesereactions, suchasanxiety and feeling of impending doom, dyspnea,hoarseness,
VOLUME NUMBER
TABLE
Adverse reactions to protamine sulfate
85 4
II. C3a, C4a, and C5a levels (nanograms/milliliter
of plasma] obtained 5 minutes after
protamine administration in patients who suffered an immediate adverse with age-, sex-, and cardiac disease-matched nonreactor patients
All patients p Value Mild
p Value Moderate p Value Severe p Value Normal 1evelP
Patient*
reaction
as compared
C5a
C4a
C3a Group
Patient
Nonreactor*
7031 k 5228 5137 + 2639 0.17 4321 2 2058 3824 r 1787 0.59 8283 t 3552 5367 k 2058 0.11 11250 -c 9098 7745 2 3444 0.50 152 -r- 69
717
Nonreactor
Patient
Nonreactor
12413 f. 10362 6534 5 3681 0.03
11.4 2 14.2 4.8 r 3.3 0.06
10247 -t 7276 0.14 12517 5 9038 0.15 17130 + 17978 0.36 155 2
3.7 k 1.4 3.8 3- 1.9 0.84 12.8 k 12.7 6.6 r 5.1 0.30 26.5 + 21.1 4.3 c 1.8 0.08 5.4 2 6.6
6023 -c 3577 6207 k 4371 8175 +- 3276 33
*There were nine patients in the group with mild reactions(and nine matchednonreactors),six in the group with moderatereactions (and six matchednonreactors),and four patients in the group with severereactions (and four matchednonreactors).
weakness, palmar itching, generalized pruritus, abdominal pain, and urgency of urination. We examined the role that previous exposure to protamine played in causing an immediate reaction and found that all the reactions occurred in patients who had previously been exposedto protamine (p < 0.05; Table I). There was a statistically significant relationship between a history of diabetesand an immediatereaction after protamine administration (Table I) consistent with previous studies.‘-x ‘, ‘I. 15.I6 We also demonstrated a statistically significant increase in reactions in patients who had received protamine-containing insulin as compared with the remainder of the patients (Table I). We cannot differentiate from these data whether diabetesalone is a risk factor or whether it is necessaryfor a diabetic to have a history of treatmentwith protamine-containing insulin for the risk of an immediate adversereaction to protamine to be increased. The major mechanisms suggestedto explain immediate reactions to protamine’.’ have been (1) IgEmediated anaphylaxis to protamine, (2) complementmediated ;maphylactoid reactions causedby heparinprotamine complexes or protamine-antiprotamine complexes, (3) direct toxicity (for example, myocardial depressionor direct mast cell degranulation, and (4) idiosyncratic. We chose in this study to examine the role of the first two of these mechanismsbecause they appearto be the most likely.‘-* Previous studies suggest that some patients who have immediate adversereactions to protamine may have positive skin tests.21However, the mostreactions occurredin patientswith a negativeskin test, andthere was no statistical significance to a positive skin test. Skin testing at a higher concentration (1 mg/ ml) in normal volunteers resulted in an unacceptably high
false positivity rate of >SO% (unpublished observations). Therefore, skin testing doesnot predict an immediate adverse reaction to protamine, and reliance on the skin test will miss most patients who will have an immediate adversereaction to protamine. We also testedfor the presenceof antiprotamine IgE antibody with a commercially available ELBA (Table I) and observed that the predictive value of a positive IgE ELISA test was 25% and the predictive value of a negative test was 85%. Furthermore, there was no relationship between a positive IgE ELBA test and a positive skin test to protamine; it appearsthat these tests were measuring different things. We used this particular ELISA test because it was commercially available. Apparently, neither test singly nor both tests togetherare sufficiently sensitive to identify most patients at significant risk for a protamine reaction. A recent study by Weiss et a1.22suggeststhat some patients with adverse reactions to protamine do have antiprotamine IgE antibodies; however, their study useda RAST assayfor IgE againstprotamine and was retrospective. It remainsto be proved in a prospective study that patients can be identified who will have adversereactionsafter protamine administration.22We conclude that the ELISA assaywe examined has no role in predicting which patients who would have reactions after protamine administration. Finally, we considered the role that complement activation might play in these reactions (Table II). Complement is activated by the membranesused in the bypass machines.23-28 Previous studies have also examined extensively the capacity of heparin and/or protamine to regulate multiple points in the complement cascade29”’and to activate complement in serum,32in vivo in an animal mode13’and in patients.33-37 Our data agree, in general, with these pre-
718
Weiler
et al.
viously reported data; we found that C3a and C4a levels were uniformly elevated in patients who had reactions and in patients matched for age, sex, and cardiac diseasewho did not have reactions. We found a statistically significant increase in C4a in patients who had reactions as compared with the nonreactor patients. Furthermore, patients with moderateand severereactionshad modestelevationsin C5a. Our study and theseprevious studies all suggesta role for complement in some patients who suffer an immediate adversereaction to protamine, although that role remains difficult to define. Complement activation occurs universally in all patients who undergo bypass surgery. In patients who have reactions to protamine, pro&mine-heparin complexes, or protamineantiprotamine complexes may causeeven more complement activation and may be responsible for some of the adversereactions that are observed. The data presentedhere, combined with previous data, suggestthat reactions to protamine are common and can be severe. Although no patient in this study died, it is possible that the patients who had catastrophic falls in blood pressure would have died if cardiovascular status had been unstable. Unfortunately, our data suggest that many reactions cannot be predicted. Protamine will continue to be used to block the anticoagulant activity of heparin after cardiopulmonary bypass surgery despite its risks becauseof the need to control bleeding at the end of surgery. A less compelling casecan be madefor protamine in cardiac catheterization. In those proceduresit may be best to allow the anticoagulant activity of the heparin to reverse itself rather than to risk a reaction or to sensitize the patient so that a reaction will occur after cardiopulmonary bypass.We concludethat thereis an urgent need for protamine substitutes so that the anticoagulant effects of heparin can be reversedsafely without the risk of serious life-threatening reactions. WethankDr. Hal B. Richersonfor thoughtfulreviewof this manuscript. REFERENCES
1. Sharath MD, Metzger WJ, Richerson HB, et al. Protamineinduced fatal anaphylaxis. J Thorac Cardiovasc Surg 1985; 90:86-90. 2. Weiler JM, Freiman P, Sharath MD, et al. Serious adverse
reactions to protamine sulfate: are alternatives needed?J ALLERGY CLIN IDOL 1985375297-303. 3. Olinger GN, Becker RM, Bonchek LI. Noncardiogenic pulmonary edemaand peripheral vascular collapse following cardiopulmonary bypass: rare protamine reaction? Ann Thorac Surg 1980;29:20-5. 4. Lowenstein E, JohnstonWE, Lappas DG, et al. Catastrophic pulmonary vasoconstrictionassociatedwith protaminereversal of heparin. Anesthesiology 1983;59:470-3. 5. Holland CL, Singh AK, McMaster PRB, Fang W. Adverse
J. ALLERGY
CLIN. IMMUNOL. APRIL 1990
reactions to protamine sulfate following cardiac surgery. Clin Cardiol 1984;7:157-62. 6. Colman RW. Humoral mediatorsof catastrophicreactions associated with protamine neutralization. Anesthesiology 1987; 66:595-6. 7. Stewart WJ, McSweeney SM, Kellett MA, Faxon DP, Ryan
8. 9.
10. Il. 12.
TJ. Increasedrisk of severeprotaminereactionsin NPH insulindependentdiabetics undergoing cardiac catheterization. Circulation 1984;70:788-92. Jackson DR. Sustainedhypotension secondary to protamine sulfate. Angiology 1970;21:295-8. Moorthy SS, Pond W, Rowland RG. Severecirculatory shock following protamine (an anaphylactic reaction). Anesth Analg 1980;39:77-8. Vontz IX, Puestow EC, Cahill DJ. Anaphylactic shock following protamine administration. Am Surg 1982;48:549-51. Cobb CA, Fung DL. Shock due to protamine hypersensitivity. Surg Nemo1 1982;17:245-6. AndersenJM, JohnsonTA. Hypertensionassociatedwith protamine sulfate administration. Am J Hosp Pharm 1981;38:
701-3. 13. Herzig GP, Root RK, Graw RG. Granulocyte collection by
continuous-flow filtration leukapheresis.Blood 1972;39:55467.
14. Data from the National Health Survey; Detailed diagnosisand proceduresfor patients discharged from short-stay hospitals; United States-1985. Hyattsville, Md.: National Center for Health Statistics; US Dept of Health and Human Services (PHS). Vital and health statistics; series 13; no. 90). 15. Chung F, Miles J. Cardiac arrest following protamine administration. Can Anaesth Sot J 1984;31:314-8. 16. Horrow JC. Rotamine: a review of its toxicity. Anesth Analg 1985;64:348-61. 17. WatsonRA, Ansbacher R, Barry M, Deshon GE, Agee RE. Allergic reaction to protamine: a late complication of elective vasectomy?Urology 1983;22:493-5. 18. Caplan SN, Be&man EM. Protamine sulfate and fish allergy [Letter]. N Engl J Med. 1976;295:172. 19. Knape JTA, Schuller JL, De Haan P, De Jong AP, Bovill JG. An anaphylactic reaction to protamine in a patient allergic to fish. Anesthesiology 1981;55:324-5. 20. WagnerJL, Hugli TE. Radioimmunoassayfor anaphylatoxins: a sensitive method for determining complement activation products in biological fluids. Anal Biochem 1984;136:75-88. 21. Lakin JD, Blocker TJ, Strong DM, Yocum MW. Anaphylaxis to protaminesulfate mediatedby a complement-dependentIgG antibody. J ALLERGYCLANIMMUNOL 1978;61:102-7. 22. WeissME, Nyhan D, Peng Z, et al. Association of protamine IgE and IgG antibodies with life-threatening reactions to intravenousprotamine. N Engl J Med 1989;320:886-92. 23. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW. Complement activation during cardiopulmonary bypass. N Engl J Med 1981;304:497-503. 24. Wysocki H, Ponizynski A, Wierusz-Wysocka B, Brocki Z, Czarnecki R. Cardiopulmonary bypass-inducedcomplement activation: evidence for C5a participation. Int J Artif Organs 1982;5:305-8. 25. HammerschmidtDE, Stroncek DF, Bowers TK, et al. Complement activation and neutropenia occurring during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1981;81:370-7. 26. Kirklin JK, Westaby S, Blackstone EH, Kirklin JW, Chenoweth DE, Pacific0AD. Complementandthe damagingeffects of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1983; 86:845-57. 27. Gardinali M, Circardi M, Agostoni A, Hugli TE. Complement
activation in extmcotporeal circulation: physiological and
VOLUME NUMBER
Adverse
85 4
pathologic,%1 implications. Path01Immunopathol Res 1986;5: 352-70. 28. Tamiya T, Yamasaki M, Maeo Y, Yamashiro T, Ogoshi S, Fujimoto :S. Complement activation in cardiopulmonary bypass, with special reference to anaphylatoxin production in membrane and bubble oxygenators. Ann Thorac Surg 1988; 46~47-57. 29. Raepple E., Hill H, Loos M. Mode of interaction of different polyanions with the first (Cl, Cl), the second (C2) and the fourth (C4.)componentof complement-I. Effect on fluid phase Cl and on Cl bound to EA or EAC4. Immunochemistry 1976;13:251-5. 30. Weiler JM, Yurt RW, FearonDT, Austen KF. Modulation of the formation of the amplification convertaseof complement, C3b,Bb, by native and commercial heparin. J Exp Med 1978;147:409-21. 31. Weiler JM. Polyions regulatethe alternative amplification pathway of complement. Immunopharmacology1983;6:245-55. 32. Rent R, !Ertel N, Eisenstein R, Gewurz H. Complement activation by interaction of polyanions and polycations. I. Heparin-protamineinducedconsumptionof complement.J Immunol 1975;114:120-4.
reactions
to protamine
sulfate
33. Fehr J, Rohr H. In vivo complementactivation by polyanionpolycation complexes: evidence that CSa is generated intravascularly during heparin-protamineinteraction. Clin Immunol Immunopathol 1983;29:7-14. 34. Best N, Sinosich MJ, Teisner B, GrudzinskasJG, Fisher M. Complement activation during cardiopulmonary bypass by hepa.rin-protamineinteraction. Br J Anaesth 1984;56:339-43. 35. Cavarocchi NC, Schaff HV, Orszulak TA, Homburger HA, Schnell WA, Pluth JR. Evidence for complementactivation by protamine-heparin interaction after cardiopulmonary bypass. Surgery 1985;98:525-30. 36. Kirklin JK, ChenowethDE, Naftel DC, et al. Effects of protamine administration after cardiopulmonary bypass on complement, blood elements, and the hemodynamic state. Ann Thorac Surg 1986;41:193-9. 37. Morel DR. Zap01WM, ThomasSJ, et al. C5a andthromboxane generation associated with pulmonary vaso- and bronchoconstriction during protamine reversal of heparin. Anesthesiology 1987;66:597-604,
Evaluation of tremor and efficacy of oral procaterol in adult patients with asthma David G. Tinkelman, MD, Richard DeJong, MD, Cheryl Lutz, MT, and Dennis L. Spangler, MD Atluntu, Ga. Tremor, the most common side effect of ordinary doses of &agonists, was evaluated in a quantitative manner and compared to improvement in pulmonary function in respect to time of onset, duration, peak effect, and tachyphylaxis with procaterol in adults. Forty-jive adult patients with reversible obstructive airway disease were studied during administration of three different dosages of procaterol, 25 pg escalating to 100 pg, 50 t.~g, and 100 kg twice daily. Tremor was quantitatively evaluated with a Grass model accelerometer. The patients were evaluated during 8 hours approximately every 2 weeks for 8 weeks. Daily diaries of symptoms and side effects were also maintained. The study demonstrated no direct relationship between tremor and pulmonary function as far as time of onset, peak effect, or duration. Although tachyphylaxis was demonstrated, much of the tolerance developed to the tremor was explained by an increase in baseline tremor. Procaterol was demonstrated to be an effective bronchodilator, and tolerance to tremor was demonstrated to develop with time but is highly variable. There was not a direct relationship between tremor and pulmonary function, indicating possible differences between lung and peripheral muscle &receptors. It appeared that the best way to minimize the problem of tremor is to begin with a lower dose and increase the dose gradually during a period of weeks. (J ALLERGYCLIN IMMUNOL1990;85:719-28.)
From the Atlanta Allergy Clinic, Atlanta, Ga. Supported in part by a grant from Parke-Davis Pharmaceuticals ResearchDivision, Warner-LambertCo., Ann Arbor, Mich. Received for publication Feb. 13, 1989. Revised Oct. 31, 1989. Accepted for publication Nov. 3, 1989. Reprint requests:David G. Tinkelman, MD, Atlanta Allergy Clinic, 6667 Vernon WoodsDrive, Suite A-30, Atlanta, GA 30328.
Abbreviations used
EFz5.,5~: Forcedexpiratory flow between25% and 75% of the FVC PFT: Pulmonary function test ANOVA: Analysis of variance b.i.d.: Twice daily I
111/18013
719
Heparin and protamine sulfate areadministeredduring cardiopulmonary bypassto control coagulationbut also may be a major causeof morbidity and mortality in patients undergoing theseprocedures.lS7Thesetwo drugs are also often administered during cardiac catheterization7and occasionally during other cardiothoracic and vascular surgical procedures,8-”and after dialysis’* and leukapheresis.I3 Previous studies indicatethat a variety of immediateadversereactionsmay occur after protamine administration, including respiratory (wheezing), cardiovascular(hypotension, decreasedcardiac contractility, bradycardia, cardiac arrest, ventricular fibrillation, or hypertension), skin From the Departmentsof Medicine, Anesthesia, Surgery and Preventive Mzedicine,the University of Iowa, and the Iowa City VeteransAdministration Medical Center, Iowa City, Iowa. Supported by the National Institutes of Health Grant AI22350, United States Public Health Service, and by a Merit Review award from the VeteransAdministration. Presentedin part at the Forty-fifth Annual Meeting of the American Academy of Allergy and Immunology, SanAntonio, Texas, Feb. 24-March 1, 1989. Received for publication May 30, 1989. Revised Oct. 17, 1989. Accepted for publication Nov. 7, 1989. Reprint requests:John M. Weiler, MD, Division of Allergy/Immunology, Departmentof Internal Medicine, SW34EUniversity Hospital, University of Iowa, Iowa City, IA 52242. l/1/18012
31
C3a,C4a, C5a: The anaphylatoxm cleavage frag-
(flushing, urticaria, and angioedema), and hematologic (bleeding) reactions.‘.7More than 290,000 openheart operations and 690,000 cardiac catheterizations were performed in the United Statesin 1985, the last yearfor which thesedataareavailable.14Nevertheless, no prospectivestudiesexist to documentthe incidence of immediate adverse reactions after protamine administration during cardiopulmonary bypass surgery. We previously reported a death in a patient who received protamine during surgery when other causes for a fatal outcome were excluded.’ Furthermore, we havebeenawarein our institution of many less serious cases in which patients apparently reacted to protamine that was administeredafter heparin during surgery or cardiaccatheterization.*Theseseriousadverse reactions may be expected to increase in number as more cardiopulmonary bypasssurgery proceduresare performed annually. This study was designed (1) to examine prospectively the incidence of an immediate adverse reaction after administration of heparin and protamine during cardiopulmonary bypass and (2) to determine the role that type I hypersensitivity and 713
714
Weiler
et al.
J. ALLERGY
complement activation may play in causing these reactions. MATERIAL AND METHODS Patient population All patients aged 18 years or older who were first observed at the University of Iowa Hospitals and Clinics cardiothoracic vascular surgery clinic for cardiopulmonary bypass surgery, which would be expected to require heparin and protamine, were eligible for this study. At the outset, we anticipated that it would be necessaryto access200 to 250 patients to be able to skin test approximately 100. The study was endedwhen we enrolled 248 patients, 85 of whom were skin testedbefore surgery. Patientswho were not skin testedeither refused skin testing, were receiving a drug that suppresseda positive skin test (for example, antihistamines), or were admitted directly to the surgical suite without first being admitted to the ward before surgery. Patients who had a previous seriousreaction to protamine were empirically pretreatedwith prednisoneto attempt to prevent a reaction and were not enrolled. Study design Patients were interviewed in the clinic or on the ward unit by a study nurse to determine past history that might suggest increased risk of a protamine reaction. We specifically asked about diabetes history and exposure to protamine-containing insulin,‘, I’. Is.I6 history of vasectomy that hasbeenreportedto result in antiprotamineantibodies,” history of fish allergy,‘*, I9 previous cardiac catheterization that required protamine,’ and previous surgical procedures that required protamine. Patients who gave consent were skin testedand had plasma obtained for ELISA testing. Staff who provided anesthesiafor patients undergoing cardiopulmonary bypass surgery were instructed by one of the investigators (J. G. C.) about the suspectednature of protamine reactions. The anesthesiologistrecordedcardiac parametersduring the procedureon a special form designed for this study; theseparameterswere monitored moreclosely just before, during, and for 5 minutes after protamine was administered. Plasmawas obtained from the central line 5 minutes after protamine administration. Historical information about each patient was obtained from the initial interview and from review of charts after dischargeof each patient. Protamine
administration
and dosage
Protamine was obtained from Eli Lilly & Co. (Indianapolis, Ind.) and was administered slowly during 3 to 7 minutes when the patient was hemodynamically stableafter bypasslines had beenremoved.The averageprotaminedose was about 300 mg, basedon the activated clotting time. In somepatients additional protamine was administered after this initial dose. However, there were no further immediate adversereactions after this additional protamine had been administered.
Determination
CLIN. IMMUNOL. APRIL 1990
of adverse reactions
We graded reactions on an objective basis. A mild reaction required a 20 to 29 mm drop, a moderatereaction required a 30 to 49 mm drop, and a severereaction required at least a 50 mm drop (or total vascular collapse) in systolic or diastolic blood pressurewithin 5 minutes after protamine administration. This 5-minute period was basedon previous experiences, indicating that all reactions occur within this time with catastrophicreactions usually occurring within 1 minute. Other reactions that were recordedin separatecategories were wheezing and flushing, urticaria, or angioedema. Blood drawing
and preservation
Plasma was obtained by collecting blood into vacuum tubes containing ethylenediaminetetraacetate.The tubes were centrifuged, and the plasma was frozen at -75” C. The first blood specimenwasdrawn on the day of admission. The second blood specimen was obtained 5 minutes after protamine administration during surgery. Skin testing
protocol
Skin testing was performed with protamine (Eli Lilly & Co.) diluted to 0.1 mg/ml, 0.01 mg/ml, and 0.001 mg/ml in ALBAY (Hollister-Stier Laboratories, Spokane, Wash.) diluent for allergenic extract. First, histamine and diluent controls were applied by intradermal injection. Then protamine, at 0.1 mg/ml, was applied by prick test; if that test was negative and the controls were positive, the subjectwas testedintradermally with the 0.001 mg/ml followed by the 0.01 mg/ml and the 0.1 mg/ml solutions. Skin tests were read 15 minutes after application, and a positive test was a wheal that was at least 4 by 4 mm larger than the negative control wheal with some erythema. No adverse reactions were noted with skin testing, nor were late skin testreactions observed. Surgeons were informed of positive skin tests, although in no casewas any special medicine administered to mitigate a reaction nor was there any other change in therapy. IgE ELISA testing
procedure
We testedfor the presenceof IgE antibody in specimen one with a commercially available ELISA to detect antiprotamine IgE in plasma(AlerCHEK, Inc., Portland, Me.). Briefly, the assaywasperformedin microtiter wells to which protamine had been preimmobilized to the polystyrene walls; control wells were used to which protamine had not been immobilized. Plasmawas then addedto either a well containing protamineor a well that did not contain protamine and incubated for 30 minutes at room temperature,and the wells were washed. Then, affinity-purified, antihuman IgE horseradish-peroxidaseconjugate was added to each well, incubation was continued for 30 minutes at room temperature, and the wells were washed again. Finally, the developing solution containing dimethylsulfoxide, 3,3’,5,5’tetramethylbenzidine, and peroxide solution was added to
VOLUME NUMBER
Adverse
85 4
reactions
to protamine
sulfate
715
each well, and the presenceof color was examined. All ELBA testing was done in batches after the surgery had been completed during the same week to assure quality control.
it was possible to determine accurately whether the patient had received protamine previously; for the remaining patients, it could not be determinedto a certainty that the patients had previously received prot-
Nonreactor group
amine.
For each patient who had an immediateadversereaction, an age-, sex- and cardiac disease-matchednonreactorpatient was selected.This nonreactorpopulation was used for the complement studies and for studies that examined whether adversereactions to protamine were dose related.
Twenty-six of the 243 patientshad adversereactions to protamine (10.7%). Eleven reactions (4.5%) were mild, nine (3.7%) were moderate, and four reactions (1.6%) were severe.One patient with a moderatere-
Detection of C3a, C4a. and C5a fragments of compk3ment Plasma obtained from specimentwo was examined for the presenceof C3a, C4a, and C5a with a commercially available assay*O following manufacturer’s instructions (AmershamCorp., Arlington Heights, Ill.).
Statisticall tests Fisher’s exact test was used to test for differences in the proportions of patientswith previous exposureto protamine, presence 0-Fdiabetes, previous exposure to protaminecontaining Insulin, the presenceof a positive skin test to protamine, and between patients who had immediate adverse reactions after protamine administration and patients who did not have reactions. Student’s paired t tests were usedto comparemeanC3a, C4a, and C5a levels in patients exhibiting an immediate adverse reaction after protamine administration with age-, sex-, andcardiacdisease-matched nonreactor patients. One-tailed tests were used, since we hypothesizedthat eachvariable (for example, previous protamine exposure, history of diabetes mellitus, history of protamine-containinginsulin, positive skin test, andpositive IgE ELBA test) would increase but not decreasethe risk of an immediate adversereaction to protamine administration. The 5% level of significance was used in all comparisons. One of the authors (W. R. C.) served as statistical consultant.
RESULT!5 Two hundred forty-eight consecutive patients undergoing cardiopulmonary bypass surgery were enrolled in this study. Two hundred forty-three patients received protamine and had charts sufficiently complete to determine whether or not there had been an immediate adversereaction to protamine administration and the nature of the reaction. The remaining five patients did not receive pro&nine (two patients) or had records that were insufficient to determine whether a reaction had occurred (three patients). The averageageof the patientswas 6 1 years, and the range was 22 to 88 years. There were 59 female and 184 male patients. Nine male patients had vasectomy.No patient had a history of fish allergy. For 201 patients
action had audible wheezing, and another patient in this group had flushing of the skin. One patient had
wheezing only, and one patient had flushing of the skin only. Patients who had reactions were treated with volume replacement and inotropic agents, if this was needed. None of the patients required return to bypass, and all reactions subsided. Patients who had
mild, moderate, and severereactions received an averageof 310 mg (4.2 mg/kg), 276 mg (3.9 mg/kg), and 296 mg (4 mg/kg) of protamine, respectively. Age-, sex-, and cardiac diseased-matched patients who did not have reactions received an average of
335 mg (4.8 mg/kg), 328 mg (3.6 mg/kg), and 300 mg (3.8 mg/kg) of protamine, respectively. The differencesamong these groups were not significant. History of previous exposure to protamine (Table I) There were 2 1 reactions in 171 patients who had previously been exposed, and no reactions in any of the 30 who had not been preexposedto protamine (p < 0.05). It was not possible to determine accurately whether the patient had previously received protamine for five of 26 patients who had reactions. In thesecases,procedures(for example, cardiac catheterizations) were done in other institutions from which adequaterecords could not be obtained or the medication record could not be verified. Next, we examinedwhether a history of diabeteswould increase the risk of a reaction. Fifty-two of the 243 patients were diabetics, 25 had previously received protaminecontaining insulin, 14 had been treated with diet and oral agents,and 13 patients had beentreatedwith diet alone. Ten of the 52 diabetics (19.2%) had a reaction to protamine, whereas 16 of the 191 nondiabetics (8.4%) had an immediate adverse reaction (p < 0.05). Six of 25 diabetics (24%) who had beentreated with pro&mine-containing insulin had a reaction when protamine was administered during surgery as compared with four reactions in the 27 diabetics (14.8%) who had not been treated with protamine-containing insulin (p = 0.3 13). All the diabetics who had reactions had received protamine previously in insulin,
716 Weiler et al.
TABLE I. Characteristics protamine administration
J. ALLERGY
of 26 patients
who had immediate
adverse
reactions
CLIN. IMMUNOL. APRIL 1990
to
Presence of characteristic
History Previous protamine exposure when exposure could be determined Diagnosis of diabetesmellius Receiving protamine-containing insulin Presenceof a positive intradermal skin test to protamine Presenceof positive antiprotamine IgE ELISA
Yes
No
Reactions/No. of patients (%I
Reactions/No. of patients (%I
p Value
21/171 (12.3)
O/30 (0)
0.028
10152(19.2) 6125 (24)
16/ 191 (8.4) 201216 (9.3)
0.028 0.037
l/3 (33)
lo/82 (12.2)
0.344
3/ 12 (25)
171115(13)
0.286
at cardiac catheterization, or at previous surgery. There were too few male patients who had vasectomy to make firm conclusions about the role vasectomy played; however, it is interesting that two of the nine patients (22%) who had vasectomyhad reactions (one mild and one severereaction). Skin test data Of the 85 patients who were skin tested (Table I), only three had positive tests (3.5%), and one of the three patients had a reaction. Ten patients who had negative skin tests to protamine had reactions (12.2%). None of 10 diabetics who had received protamine-containing insulin had a positive skin test. One diet-controlled diabetic had a positive skin test, and there were two positive skin tests in 65 nondiabetic patients. The protamine concentrationat which all three patients had positive skin tests was 0.1 mg/ml. Two of the three patients with positive skin tests did have prior exposure to protamine, and the third patient probably had prior exposureto protamine at the time of catheterization but was placed in the category of patients who had unknown exposure because we could not verify the exposure. This patient also had vasectomy. Prevalence of antiprotamine IgE antibody by ELISA Twelve patientshad IgE directed againstprotamine, but none of these tests was strongly positive (Table I), and three of the 12 patients had reactions to protamine (25%). In contrast, 17 of 115 with a negative test (14.8%) had a reaction (p = 0.29). Furthermore, there was no associationbetween a positive skin test and a positive IgE ELBA test, since there were no positive IgE ELISA testsin the patients with positive
skin testsas comparedwith sevenpositive IgE ELBA tests in the 58 patients with negative skin tests (p = 0.68). Complement activation after protamine administration (Table II) C3a levels were significantly elevatedin all patient groups over levels expected in normal plasma, although there was no statistically significant increase in patients who had reactions as compared with the nonreactor patients. C4a levels were increased in all patient groups and were statistically significantly increased in the reactor group as compared with the nonreactorgroup. C5a levels remained normal in the nonreactorgroups and in the group with mild reactions but were mildly increasedin the groups with moderate and severereactions. DISCUSSION Immediate adverse reactions after pro&mine administration are common and may be very serious. Twenty-six of 243 patients (10.7%) we followed prospectively during cardiopulmonary bypasshad an immediate reaction manifested in 24 (9.9%) by a significant drop in blood pressure.Four patients (1.6%) had a precipitous drop in blood pressure.No previous large study prospectively examined the incidence of an immediate adverse reaction to protamine during cardiopulmonary bypass. The criteria for an immediate adversereaction in this study were objective and could easily be determined by the anesthesiologist. Unfortunately, it is difficult to diagnose more mild casesof anaphylacticand anaphylactoidreactionsduring surgery becauseit is not possible to assessthe more subtlecriteria for thesereactions, suchasanxiety and feeling of impending doom, dyspnea,hoarseness,
VOLUME NUMBER
TABLE
Adverse reactions to protamine sulfate
85 4
II. C3a, C4a, and C5a levels (nanograms/milliliter
of plasma] obtained 5 minutes after
protamine administration in patients who suffered an immediate adverse with age-, sex-, and cardiac disease-matched nonreactor patients
All patients p Value Mild
p Value Moderate p Value Severe p Value Normal 1evelP
Patient*
reaction
as compared
C5a
C4a
C3a Group
Patient
Nonreactor*
7031 k 5228 5137 + 2639 0.17 4321 2 2058 3824 r 1787 0.59 8283 t 3552 5367 k 2058 0.11 11250 -c 9098 7745 2 3444 0.50 152 -r- 69
717
Nonreactor
Patient
Nonreactor
12413 f. 10362 6534 5 3681 0.03
11.4 2 14.2 4.8 r 3.3 0.06
10247 -t 7276 0.14 12517 5 9038 0.15 17130 + 17978 0.36 155 2
3.7 k 1.4 3.8 3- 1.9 0.84 12.8 k 12.7 6.6 r 5.1 0.30 26.5 + 21.1 4.3 c 1.8 0.08 5.4 2 6.6
6023 -c 3577 6207 k 4371 8175 +- 3276 33
*There were nine patients in the group with mild reactions(and nine matchednonreactors),six in the group with moderatereactions (and six matchednonreactors),and four patients in the group with severereactions (and four matchednonreactors).
weakness, palmar itching, generalized pruritus, abdominal pain, and urgency of urination. We examined the role that previous exposure to protamine played in causing an immediate reaction and found that all the reactions occurred in patients who had previously been exposedto protamine (p < 0.05; Table I). There was a statistically significant relationship between a history of diabetesand an immediatereaction after protamine administration (Table I) consistent with previous studies.‘-x ‘, ‘I. 15.I6 We also demonstrated a statistically significant increase in reactions in patients who had received protamine-containing insulin as compared with the remainder of the patients (Table I). We cannot differentiate from these data whether diabetesalone is a risk factor or whether it is necessaryfor a diabetic to have a history of treatmentwith protamine-containing insulin for the risk of an immediate adversereaction to protamine to be increased. The major mechanisms suggestedto explain immediate reactions to protamine’.’ have been (1) IgEmediated anaphylaxis to protamine, (2) complementmediated ;maphylactoid reactions causedby heparinprotamine complexes or protamine-antiprotamine complexes, (3) direct toxicity (for example, myocardial depressionor direct mast cell degranulation, and (4) idiosyncratic. We chose in this study to examine the role of the first two of these mechanismsbecause they appearto be the most likely.‘-* Previous studies suggest that some patients who have immediate adversereactions to protamine may have positive skin tests.21However, the mostreactions occurredin patientswith a negativeskin test, andthere was no statistical significance to a positive skin test. Skin testing at a higher concentration (1 mg/ ml) in normal volunteers resulted in an unacceptably high
false positivity rate of >SO% (unpublished observations). Therefore, skin testing doesnot predict an immediate adverse reaction to protamine, and reliance on the skin test will miss most patients who will have an immediate adversereaction to protamine. We also testedfor the presenceof antiprotamine IgE antibody with a commercially available ELBA (Table I) and observed that the predictive value of a positive IgE ELISA test was 25% and the predictive value of a negative test was 85%. Furthermore, there was no relationship between a positive IgE ELBA test and a positive skin test to protamine; it appearsthat these tests were measuring different things. We used this particular ELISA test because it was commercially available. Apparently, neither test singly nor both tests togetherare sufficiently sensitive to identify most patients at significant risk for a protamine reaction. A recent study by Weiss et a1.22suggeststhat some patients with adverse reactions to protamine do have antiprotamine IgE antibodies; however, their study useda RAST assayfor IgE againstprotamine and was retrospective. It remainsto be proved in a prospective study that patients can be identified who will have adversereactionsafter protamine administration.22We conclude that the ELISA assaywe examined has no role in predicting which patients who would have reactions after protamine administration. Finally, we considered the role that complement activation might play in these reactions (Table II). Complement is activated by the membranesused in the bypass machines.23-28 Previous studies have also examined extensively the capacity of heparin and/or protamine to regulate multiple points in the complement cascade29”’and to activate complement in serum,32in vivo in an animal mode13’and in patients.33-37 Our data agree, in general, with these pre-
718
Weiler
et al.
viously reported data; we found that C3a and C4a levels were uniformly elevated in patients who had reactions and in patients matched for age, sex, and cardiac diseasewho did not have reactions. We found a statistically significant increase in C4a in patients who had reactions as compared with the nonreactor patients. Furthermore, patients with moderateand severereactionshad modestelevationsin C5a. Our study and theseprevious studies all suggesta role for complement in some patients who suffer an immediate adversereaction to protamine, although that role remains difficult to define. Complement activation occurs universally in all patients who undergo bypass surgery. In patients who have reactions to protamine, pro&mine-heparin complexes, or protamineantiprotamine complexes may causeeven more complement activation and may be responsible for some of the adversereactions that are observed. The data presentedhere, combined with previous data, suggestthat reactions to protamine are common and can be severe. Although no patient in this study died, it is possible that the patients who had catastrophic falls in blood pressure would have died if cardiovascular status had been unstable. Unfortunately, our data suggest that many reactions cannot be predicted. Protamine will continue to be used to block the anticoagulant activity of heparin after cardiopulmonary bypass surgery despite its risks becauseof the need to control bleeding at the end of surgery. A less compelling casecan be madefor protamine in cardiac catheterization. In those proceduresit may be best to allow the anticoagulant activity of the heparin to reverse itself rather than to risk a reaction or to sensitize the patient so that a reaction will occur after cardiopulmonary bypass.We concludethat thereis an urgent need for protamine substitutes so that the anticoagulant effects of heparin can be reversedsafely without the risk of serious life-threatening reactions. WethankDr. Hal B. Richersonfor thoughtfulreviewof this manuscript. REFERENCES
1. Sharath MD, Metzger WJ, Richerson HB, et al. Protamineinduced fatal anaphylaxis. J Thorac Cardiovasc Surg 1985; 90:86-90. 2. Weiler JM, Freiman P, Sharath MD, et al. Serious adverse
reactions to protamine sulfate: are alternatives needed?J ALLERGY CLIN IDOL 1985375297-303. 3. Olinger GN, Becker RM, Bonchek LI. Noncardiogenic pulmonary edemaand peripheral vascular collapse following cardiopulmonary bypass: rare protamine reaction? Ann Thorac Surg 1980;29:20-5. 4. Lowenstein E, JohnstonWE, Lappas DG, et al. Catastrophic pulmonary vasoconstrictionassociatedwith protaminereversal of heparin. Anesthesiology 1983;59:470-3. 5. Holland CL, Singh AK, McMaster PRB, Fang W. Adverse
J. ALLERGY
CLIN. IMMUNOL. APRIL 1990
reactions to protamine sulfate following cardiac surgery. Clin Cardiol 1984;7:157-62. 6. Colman RW. Humoral mediatorsof catastrophicreactions associated with protamine neutralization. Anesthesiology 1987; 66:595-6. 7. Stewart WJ, McSweeney SM, Kellett MA, Faxon DP, Ryan
8. 9.
10. Il. 12.
TJ. Increasedrisk of severeprotaminereactionsin NPH insulindependentdiabetics undergoing cardiac catheterization. Circulation 1984;70:788-92. Jackson DR. Sustainedhypotension secondary to protamine sulfate. Angiology 1970;21:295-8. Moorthy SS, Pond W, Rowland RG. Severecirculatory shock following protamine (an anaphylactic reaction). Anesth Analg 1980;39:77-8. Vontz IX, Puestow EC, Cahill DJ. Anaphylactic shock following protamine administration. Am Surg 1982;48:549-51. Cobb CA, Fung DL. Shock due to protamine hypersensitivity. Surg Nemo1 1982;17:245-6. AndersenJM, JohnsonTA. Hypertensionassociatedwith protamine sulfate administration. Am J Hosp Pharm 1981;38:
701-3. 13. Herzig GP, Root RK, Graw RG. Granulocyte collection by
continuous-flow filtration leukapheresis.Blood 1972;39:55467.
14. Data from the National Health Survey; Detailed diagnosisand proceduresfor patients discharged from short-stay hospitals; United States-1985. Hyattsville, Md.: National Center for Health Statistics; US Dept of Health and Human Services (PHS). Vital and health statistics; series 13; no. 90). 15. Chung F, Miles J. Cardiac arrest following protamine administration. Can Anaesth Sot J 1984;31:314-8. 16. Horrow JC. Rotamine: a review of its toxicity. Anesth Analg 1985;64:348-61. 17. WatsonRA, Ansbacher R, Barry M, Deshon GE, Agee RE. Allergic reaction to protamine: a late complication of elective vasectomy?Urology 1983;22:493-5. 18. Caplan SN, Be&man EM. Protamine sulfate and fish allergy [Letter]. N Engl J Med. 1976;295:172. 19. Knape JTA, Schuller JL, De Haan P, De Jong AP, Bovill JG. An anaphylactic reaction to protamine in a patient allergic to fish. Anesthesiology 1981;55:324-5. 20. WagnerJL, Hugli TE. Radioimmunoassayfor anaphylatoxins: a sensitive method for determining complement activation products in biological fluids. Anal Biochem 1984;136:75-88. 21. Lakin JD, Blocker TJ, Strong DM, Yocum MW. Anaphylaxis to protaminesulfate mediatedby a complement-dependentIgG antibody. J ALLERGYCLANIMMUNOL 1978;61:102-7. 22. WeissME, Nyhan D, Peng Z, et al. Association of protamine IgE and IgG antibodies with life-threatening reactions to intravenousprotamine. N Engl J Med 1989;320:886-92. 23. Chenoweth DE, Cooper SW, Hugli TE, Stewart RW, Blackstone EH, Kirklin JW. Complement activation during cardiopulmonary bypass. N Engl J Med 1981;304:497-503. 24. Wysocki H, Ponizynski A, Wierusz-Wysocka B, Brocki Z, Czarnecki R. Cardiopulmonary bypass-inducedcomplement activation: evidence for C5a participation. Int J Artif Organs 1982;5:305-8. 25. HammerschmidtDE, Stroncek DF, Bowers TK, et al. Complement activation and neutropenia occurring during cardiopulmonary bypass. J Thorac Cardiovasc Surg 1981;81:370-7. 26. Kirklin JK, Westaby S, Blackstone EH, Kirklin JW, Chenoweth DE, Pacific0AD. Complementandthe damagingeffects of cardiopulmonary bypass. J Thorac Cardiovasc Surg 1983; 86:845-57. 27. Gardinali M, Circardi M, Agostoni A, Hugli TE. Complement
activation in extmcotporeal circulation: physiological and
VOLUME NUMBER
Adverse
85 4
pathologic,%1 implications. Path01Immunopathol Res 1986;5: 352-70. 28. Tamiya T, Yamasaki M, Maeo Y, Yamashiro T, Ogoshi S, Fujimoto :S. Complement activation in cardiopulmonary bypass, with special reference to anaphylatoxin production in membrane and bubble oxygenators. Ann Thorac Surg 1988; 46~47-57. 29. Raepple E., Hill H, Loos M. Mode of interaction of different polyanions with the first (Cl, Cl), the second (C2) and the fourth (C4.)componentof complement-I. Effect on fluid phase Cl and on Cl bound to EA or EAC4. Immunochemistry 1976;13:251-5. 30. Weiler JM, Yurt RW, FearonDT, Austen KF. Modulation of the formation of the amplification convertaseof complement, C3b,Bb, by native and commercial heparin. J Exp Med 1978;147:409-21. 31. Weiler JM. Polyions regulatethe alternative amplification pathway of complement. Immunopharmacology1983;6:245-55. 32. Rent R, !Ertel N, Eisenstein R, Gewurz H. Complement activation by interaction of polyanions and polycations. I. Heparin-protamineinducedconsumptionof complement.J Immunol 1975;114:120-4.
reactions
to protamine
sulfate
33. Fehr J, Rohr H. In vivo complementactivation by polyanionpolycation complexes: evidence that CSa is generated intravascularly during heparin-protamineinteraction. Clin Immunol Immunopathol 1983;29:7-14. 34. Best N, Sinosich MJ, Teisner B, GrudzinskasJG, Fisher M. Complement activation during cardiopulmonary bypass by hepa.rin-protamineinteraction. Br J Anaesth 1984;56:339-43. 35. Cavarocchi NC, Schaff HV, Orszulak TA, Homburger HA, Schnell WA, Pluth JR. Evidence for complementactivation by protamine-heparin interaction after cardiopulmonary bypass. Surgery 1985;98:525-30. 36. Kirklin JK, ChenowethDE, Naftel DC, et al. Effects of protamine administration after cardiopulmonary bypass on complement, blood elements, and the hemodynamic state. Ann Thorac Surg 1986;41:193-9. 37. Morel DR. Zap01WM, ThomasSJ, et al. C5a andthromboxane generation associated with pulmonary vaso- and bronchoconstriction during protamine reversal of heparin. Anesthesiology 1987;66:597-604,
Evaluation of tremor and efficacy of oral procaterol in adult patients with asthma David G. Tinkelman, MD, Richard DeJong, MD, Cheryl Lutz, MT, and Dennis L. Spangler, MD Atluntu, Ga. Tremor, the most common side effect of ordinary doses of &agonists, was evaluated in a quantitative manner and compared to improvement in pulmonary function in respect to time of onset, duration, peak effect, and tachyphylaxis with procaterol in adults. Forty-jive adult patients with reversible obstructive airway disease were studied during administration of three different dosages of procaterol, 25 pg escalating to 100 pg, 50 t.~g, and 100 kg twice daily. Tremor was quantitatively evaluated with a Grass model accelerometer. The patients were evaluated during 8 hours approximately every 2 weeks for 8 weeks. Daily diaries of symptoms and side effects were also maintained. The study demonstrated no direct relationship between tremor and pulmonary function as far as time of onset, peak effect, or duration. Although tachyphylaxis was demonstrated, much of the tolerance developed to the tremor was explained by an increase in baseline tremor. Procaterol was demonstrated to be an effective bronchodilator, and tolerance to tremor was demonstrated to develop with time but is highly variable. There was not a direct relationship between tremor and pulmonary function, indicating possible differences between lung and peripheral muscle &receptors. It appeared that the best way to minimize the problem of tremor is to begin with a lower dose and increase the dose gradually during a period of weeks. (J ALLERGYCLIN IMMUNOL1990;85:719-28.)
From the Atlanta Allergy Clinic, Atlanta, Ga. Supported in part by a grant from Parke-Davis Pharmaceuticals ResearchDivision, Warner-LambertCo., Ann Arbor, Mich. Received for publication Feb. 13, 1989. Revised Oct. 31, 1989. Accepted for publication Nov. 3, 1989. Reprint requests:David G. Tinkelman, MD, Atlanta Allergy Clinic, 6667 Vernon WoodsDrive, Suite A-30, Atlanta, GA 30328.
Abbreviations used
EFz5.,5~: Forcedexpiratory flow between25% and 75% of the FVC PFT: Pulmonary function test ANOVA: Analysis of variance b.i.d.: Twice daily I
111/18013
719
