Prevalence of Heparin/Platelet Factor 4 Antibodies Before and After Cardiac Surgery

Transcription

1 Prevalence of Heparin/Platelet Factor 4 Antibodies Before and After Cardiac Surgery Brendan M. Everett, MD, Robert Yeh, MD, Shi Yin Foo, MD, PhD, David Criss, MD, Elizabeth M. Van Cott, MD, Michael Laposata, MD, PhD, Edwin G. Avery, MD, William D. Hoffman, MD, Jennifer Walker, MD, David Torchiana, MD, and Ik-Kyung Jang, MD, PhD Departments of Medicine, Cardiology Division, Pathology, Anesthesia and Critical Care, Cardiac Anesthesia Division, and Surgery, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts Background. The clinical significance of heparin/ platelet factor 4 (PF4) antibodies in subjects undergoing cardiac surgery has not been systematically studied. We prospectively investigated whether the presence of heparin/pf4 antibodies would predict clinical thrombosis in this population. Methods. In 299 patients scheduled for cardiac surgery between October 2003 and March 2005, the heparin/pf4 antibodies and platelet count were measured immediately prior to, and 5 days after, surgery. The patients were followed up at 30 days for thrombotic complications. Results. The prevalence of the heparin/pf4 antibodies was 4.3% (13 of 299) prior to surgery and increased more than fivefold to 22.4% (62 of 277) postoperatively (p < ). Thromboembolic events occurred in 8.8% of patients with negative antibody and in 6.3% of patients with positive antibody (p 0.77). Of the 62 patients with positive heparin/pf4 antibodies postoperatively, 22 (35.5%) were treated with a nonheparin anticoagulant. There was a trend toward higher rates of thromboembolic events in subjects who were thrombocytopenic compared with those who were not (17.1% and 6.7%, respectively, p 0.06), regardless of antibody status. Two out of 8 patients (25%) with both thrombocytopenia and a positive antibody (clinical heparin-induced thrombocytopenia [HIT]) suffered a thromboembolic event, compared with 17 of 222 (7.7%) without clinical HIT (p 0.13). Conclusions. The high prevalence of antibodies to the heparin/pf4 complex after cardiac surgery and the low rate of thromboembolic complications in this population suggest that the antibody alone does not confer an increased risk of thrombotic complications. Monitoring for thrombocytopenia is recommended. (Ann Thorac Surg 2007;83:592 7) 2007 by The Society of Thoracic Surgeons Accepted for publication Sept 11, Address correspondence to Dr Jang, Cardiology Division, Gray/Bigelow 800, Massachusetts General Hospital, 55 Fruit Street, Boston, MA 02114; ijang@partners.org. Heparin-induced thrombocytopenia (HIT) is a procoagulant state that is increasingly recognized as a devastating complication of heparin administration. When administered, heparin has a high affinity for platelet factor 4 (PF4), which is found in platelet -granules and the endothelium [1]. Antibodies to this complex cause platelet activation through the platelets Fc IIa receptors [2 4]. The subsequent release of procoagulant microparticles promote thrombin generation and tissue factor production, leading to an elevated risk of arterial and venous thrombosis that extends for days to weeks beyond the discontinuation of heparin therapy [2, 4 6]. The spectrum of HIT extends from the presence of antibodies to the heparin/pf4 complex without thrombocytopenia, to the presence of those antibodies with thrombocytopenia (usually defined as a platelet count 100,000 or a fall in the platelet count by 50%), to thrombocytopenia with thrombosis [7]. The prevalence of heparin/pf4 antibodies is known to rise dramatically after cardiopulmonary bypass, but it is less well understood whether these antibodies are associated with clinical thromboembolic events [2, 8 12]. We hypothesized that the presence of antibodies would be associated with an increased risk of thromboembolic complications postoperatively. Patients and Methods Between October 2003 and February 2005, we enrolled 299 patients presenting for cardiac surgery into the study. The protocol was approved by the Institutional Review Board (IRB) and written informed consent was obtained from all patients. Men and nonpregnant women over the age of 18 who were scheduled to undergo cardiac surgery and could provide written informed consent were enrolled. Subjects with a documented bleeding or thrombophilic disorder, history of HIT, documented allergy to heparin, or a high risk of bleeding were excluded. Because we excluded subjects whose heparin/pf4 antibody Dr Laposato discloses a financial relationship with GlaxoSmithKline by The Society of Thoracic Surgeons /07/$32.00 Published by Elsevier Inc doi: /j.athoracsur

2 Ann Thorac Surg EVERETT ET AL 2007;83:592 7 HEPARIN/PF 4 ANTIBODIES AND CARDIAC SURGERY 593 Table 1. Patient Demographics Characteristic Postoperative Heparin/PF4 Antibody Positive (n 62) Negative (n 215) p Male 43 (69%) 154 (72%) 0.75 Age (yr) a BMI (kg/m 2 ) a Diabetes 18 (29%) 51 (24%) 0.41 High cholesterol 43 (69%) 151 (70%) 0.87 Hypertension 43 (69%) 154 (72%) 0.75 CAD 52 (84%) 159 (74%) 0.13 CHF 17 (27%) 48 (22%) 0.40 Atrial fibrillation 10 (16%) 31 (14%) 0.69 PAD 10 (16%) 35 (16%) 1.0 History of thrombosis 4 (6%) 9 (4%) 0.50 Type of surgery CABG 36 (58%) 114 (53%) 0.56 Combined CABG and valve surgery 14 (23%) 36 (17%) 0.35 Valve surgery 11 (18%) 52 (24%) 0.31 Other 1 (2%) 13 (6%) 0.20 Platelet count (K/mm 3 ) a Prior heparin exposure 62 (100%) 213 (99%) 1.0 Heparin dose during surgery (units) b 43,000 20,500 38,500 16, Length of stay (days) b a Mean SD (p values are from a 2-tailed t test). b Median interquartile ranges (p values are from a Wilcoxon 2-sample test). The p values for dichotomous variables are from the Fisher exact test. BMI body mass index; CABG coronary artery bypass grafting; CAD coronary artery disease; CHF congestive heart failure; PAD peripheral arterial disease; PF4 platelet factor 4. test results would not be available prior to the planned operation, we could not enroll consecutive patients. Baseline platelet counts and heparin/pf4 antibodies (Asserachrom HPIA ELISA [heparin-induced platelet activation/enzyme-linked immunosorbent assay], Diagnostica Stago, Parsippany, NJ) were measured at baseline within 24 hours prior to surgery and were repeated five days later or at discharge, whichever was earlier. The assay was performed according to manufacturer s directions, with positive and negative controls included in each run. Individual patient samples were tested in duplicate and their mean titers were assessed by comparison with a reference control titer for the day. All tests were performed at the clinical laboratory of our institution. At the request of our IRB, the results of both tests were available to the physicians caring for the subjects, who could then treat the subjects in accordance with their clinical judgment. Subjects were followed until discharge and were contacted 30 days after surgery by a member of the study team blinded to the results of the antibody assay who then administered a standardized questionnaire that focused on clinical evidence of arterial and venous thromboembolism. Fentanyl, midazolam, and inhaled anesthetics were used for anesthesia. Systemic temperatures varied from 25 C to normothermic on cardiopulmonary bypass. The standard dose of porcine heparin was 350 IU/kg, followed by an infusion to maintain the activated clotting time greater than 450 seconds. A total of 13 patients received alternative anticoagulation intraoperatively. Of these, 11 (seven preoperative antibody positive and four preoperative antibody negative) received alprostadil (0.5 mcg/ minute initially, titrated upward as blood pressure tolerated) and one (preoperative antibody negative) received antithrombin III in addition to standard dose heparin. Fig 1. Prevalence of heparin/platelet factor 4 antibodies before (pre) and after (post) cardiac surgery. The prevalence of heparin/platelet factor 4 antibodies increased more than fivefold after cardiac surgery.

3 594 EVERETT ET AL Ann Thorac Surg HEPARIN/PF 4 ANTIBODIES AND CARDIAC SURGERY 2007;83:592 7 Fig 2. Frequency of thromboembolic events according to the presence or absence of antibodies to heparin/platelet factor 4 (A); frequency of thromboembolic events according to the presence or absence of postoperative thrombocytopenia (defined as a platelet count 100,000 or a fall of 50% from preoperative levels), regardless of the presence or absence of the antibody to heparin/ platelet factor 4 (B); and frequency of thromboembolic events according to the presence or absence of clinical heparin-induced thrombocytopenia (HIT; defined as a positive postoperative heparin/platelet factor 4 antibody test with thrombocytopenia) (C). One patient with antibodies to heparin/pf4 preoperatively received bivalrudin and no heparin. Clinical HIT was defined by the combination of positive antibody with either a platelet count below 100,000 or a fall in the platelet count by greater than 50% from preoperative levels. Thrombotic events included symptomatic cerebrovascular events, acute coronary syndromes requiring repeat catheterizations or emergent coronary bypass procedures due to unexpected coronary artery occlusion, peripheral thrombosis including superficial and deep venous thrombosis, and arterial thrombosis with limb ischemia. Possible thromboembolic events both during the subject s hospitalization and at the 30-day follow-up questionnaire were reviewed and adjudicated by three study physicians blinded to the subject s antibody status. Because the exposure of interest was the presence or absence of the heparin/pf4 antibody postoperatively, we decided a priori to analyze all endpoints according to postoperative antibody status. The only exception was the comparison of the frequency of the heparin/pf4 antibody preoperatively and postoperatively. The study sponsors were not involved in the design of the study, in the collection, in the analysis or interpretation of the data, in the drafting of the manuscript, or in the decision to submit the manuscript for publication. The investigators had free access to all the data included in this report. Statistical Comparisons Means and standard deviations (SD) are reported for normally distributed data and are compared using 2-tailed t tests. Non-normally distributed data are reported as medians and interquartile ranges (IQR), and comparisons were made with the Wilcoxon 2-sample test. Heparin/PF4 antibody tests preoperatively and postoperatively and platelet counts preoperatively and postoperatively were compared with the McNemar and single sample paired t tests, respectively. Frequencies and dichotomous variables were compared with the Fisher exact test. A p value of 0.05 or less (two-tailed) was required for statistical significance. All computations were performed with SAS software, version 9.1 (SAS Institute, Cary, NC). Results Baseline Characteristics A total of 299 subjects presenting for cardiac surgery were enrolled. Baseline characteristics of the patients are listed in Table 1. There was no significant difference

4 Ann Thorac Surg EVERETT ET AL 2007;83:592 7 HEPARIN/PF 4 ANTIBODIES AND CARDIAC SURGERY 595 between the groups except that the patients with detectable heparin/pf4 antibodies postoperatively received higher doses of heparin intraoperatively (median interquartile range [IQR]: 43,000 20,500 units vs 38,500 16,500 units; Wilcoxon 2-sample; p 0.03) and had a longer length of stay (median IQR: vs 9 8 days; p ). Heparin/PF4 Antibodies, HIT, and Thromboembolic Events At enrollment, 4.3% (13 out of 299) patients had detectable heparin/pf4 antibodies. Postoperatively, 62 of the 277 (22.4%) patients for whom a second test was performed developed a positive heparin/pf4 antibody (McNemar; p ; Fig 1). Of the 230 patients with postoperative antibody tests and complete 30-day follow-up (47 were lost to follow-up), 19 experienced a thromboembolic event or death. In total, there were four deaths, six strokes or transient ischemic attacks, five deep venous thrombosis-pulmonary embolisms, and four peripheral arterial occlusions. There were no differences in the rate of thromboembolism or death between the two groups, with an event rate of 6.3% (3 of 48) in the antibody positive group and 8.8% (16 of 182) in the antibody negative group (Fisher exact test; p 0.77; Fig 2A). Preoperatively, platelet counts were similar in the patients with and without the heparin/pf4 antibody (230,935 84,478/mm 3 and 227,098 76,450/mm 3, respectively; 2-tailed t test; p 0.73). The platelet count of both groups of subjects fell significantly from their preoperative baseline, but the magnitude of the fall in platelet counts was similar between the two groups (2-tailed t test; p 0.14). The mean postoperative platelet count was similar in those subjects with the heparin/pf4 antibody (202, ,052/mm 3 ) and those without the antibody (182,312 77,978/mm 3 ; 2-tailed t test; p 0.14). Fifty-one patients (17%) met the criteria for thrombocytopenia (defined as platelet count 100,000/mm 3 or fall from preoperative platelet count of 50%). Seven of 41 subjects with thrombocytopenia for whom we have follow-up data suffered a thromboembolic event (17.1%), while 14 patients of 208 (6.7%) without thrombocytopenia suffered an event (Fisher exact test; p 0.06; Fig 2B). Among those patients with both a platelet count and a second heparin/pf4 antibody test, those with a positive antibody test postoperatively were no more likely to be thrombocytopenic than those with a negative test (11 of 62 [17.7%] and 36 of 215 [16.7%], respectively; Fisher exact test; p 0.85). The frequency of the positive antibody test was similar in those with and without postoperative thrombocytopenia (11 of 47 [23.4%] vs 51 of 230 [22.2%], respectively; Fisher exact test; p 0.85). Clinical HIT, defined by a positive heparin/pf4 antibody and a platelet count less than 100,000 or a drop of greater than 50% from preoperative levels, was seen in 11 of 277 subjects (4.0%). Of the 230 for whom a postoperative heparin/pf4 antibody test and 30-day follow-up were available, two of eight patients (25.0%) with clinical HIT suffered a thromboembolic event or death, while 17 of 222 (7.7%) without clinical HIT had an event (Fisher exact test; p 0.13; Fig 2C). Management of Patients With Positive Antibodies Because patients antibody test results were made available to their attending physicians, many patients with the heparin/pf4 antibody were treated with nonheparin anticoagulants. Of the 62 patients with positive heparin/pf4 antibodies postoperatively, 22 (35.5%) were treated with argatroban, the most commonly used nonheparin anticoagulant at our institution. Two of the three postoperative antibody positive patients who suffered a thromboembolic event were treated with argatroban, and the third was treated with fondaparinux. Of the 215 patients who were heparin/pf4 antibody negative postoperatively, 18 were treated with argatroban either because their heparin/pf4 antibody test was reported as negative but borderline or because of physician preference. Sixteen of those 215 subjects without the antibody suffered a thromboembolic event or death, and four of those 16 were treated with argatroban. A majority of subjects (170 out of 277, or 61.4%) were discharged home on warfarin or other anticoagulant, and the rates of warfarin use at discharge were essentially the same in antibody negative (61.4%) and antibody positive (61.3%) patients. Only three patients had a positive antibody test as their sole indication for anticoagulation. Of the 144 patients discharged on warfarin for whom we have follow-up data, 13 (9.0%) suffered a thromboembolic event, while there were eight thromboembolic events among the 105 patients (7.6%) not discharged on warfarin (Fisher exact text; p 0.82). Comment We prospectively evaluated the prevalence of heparin/ PF4 antibodies and HIT in the population presenting for cardiac surgery in a single academic medical center. We demonstrated that the prevalence of the heparin/pf4 antibody increases fivefold, from 4.4% to 22.4% after surgery. Rates of thrombosis and(or) death were low, with similar rates in patients with (6.3%) and without (8.8%) the heparin/pf4 antibody. While two of the eight patients with clinical HIT (25%) suffered a thromboembolic event compared with 7.7% of those without clinical HIT (17 of 222), these differences were not statistically significant. Earlier reports suggested a higher prevalence of the antibody, but a low prevalence of thromboembolic complications (Table 2). In a study of 111 patients undergoing cardiac surgery, Bauer and colleagues [13] reported the prevalence of the antibody to be 19% prior to cardiopulmonary bypass and 51% on the fifth postoperative day. Rates of thrombosis were low (only two of 111 patients). In smaller studies, each of 51 patients, Trossaert and colleagues [8] and Visentin and colleagues [10] found a preoperative and postoperative heparin/pf4 antibody prevalence of 27% and 61%, respectively. The reason for the high prevalence of antibodies was perhaps because the earlier generation ELISAs were less specific, or be-

5 596 EVERETT ET AL Ann Thorac Surg HEPARIN/PF 4 ANTIBODIES AND CARDIAC SURGERY 2007;83:592 7 Table 2. Compilation of Prior Studies of the Heparin/Platelet Factor 4 Antibody in Patients Undergoing Cardiac Surgery Positive Heparin/Platelet Factor 4 Antibody Test Thromboembolic Events Study N Preoperative Postoperative Ab Pos Ab Neg Visentin et al 1996 [10] 51 11/51 (22%) 27/44 (61%) 0 0 Bauer et al 1997 [13] /111 (19%) 57/111 (51%) 0 2 (4%) Trossaert et al 1998 [8] 51 2/51 (4%) 14/51 (27%) a a Pouplard et al 1999 [11] 328 3/328 (1%) 83/328 (25%) 1 (1%) b Francis et al 2003 [12] 207 6/207 (3%) 87/207 (42%) 2 (2%) b Bennett-Guerrero et al 2005 [14] /466 (13%) c a a Current study (4%) 62/277 (22%) 3/48 (6%) 16/182 (9%) a The frequency of thromboembolic events was not assessed in this study. b The frequency of thromboembolic events in the heparin/platelet factor 4 antibody negative subjects was not reported. c The frequency of the heparin/platelet factor 4 antibody was not measured postoperatively. Ab antibody; Pos positive; Neg negative. cause in-hospital practice has changed and patients are not maintained on continuous heparin infusions for as long as they once were. Recently, Francis and colleagues [12] showed similar results to ours in 207 patients undergoing first-time coronary artery bypass grafting. Preoperatively, 2.9% were heparin/pf4 antibody positive and 42% were antibody positive by postoperative day seven. Two patients in that study were diagnosed with thromboembolic complications and HIT. In a larger study of 328 patients undergoing cardiopulmonary bypass, Pouplard and colleagues [11] noted that 83 (25.2%) had a positive antibody test between postoperative days seven and 10. The authors reported that six subjects met their criteria for HIT, but only one suffered a thrombotic event. A recently published report by Bennett-Guerrero and colleagues [14] found a slightly higher rate of positive heparin/pf4 antibody tests preoperatively and a higher rate of major complications (defined as in-hospital death or hospital stay longer than 10 days) in the patients with a positive test. While we found no difference in the frequency of our primary endpoint of thromboembolic complications between the subjects with and without a positive heparin/pf4 antibody test, we also noted a longer hospital stay in patients with a positive test postoperatively (median stay IQR: vs 9 8 days, respectively; Wilcoxon 2-sample; p ). In order to address the possibility that our results are similar to those of Bennett-Guerrero and colleagues [14], we conducted a post hoc analysis in which we included a hospital stay longer than 10 days as part of our endpoint. We found that among subjects for whom we have complete follow-up data, more subjects with the antibody reached this composite endpoint (77.1%, or 37 of 48) than those without the antibody (46.7% or 85 of 182; Fisher exact test; p ). This finding is similar to that of Bennett-Guerrero and colleagues, and suggests that while we were unable to detect an increased risk of thromboembolic complications among subjects with a positive heparin/pf4 antibody test, those subjects are more likely than those without the antibody to have a hospital stay that is longer than 10 days. Because of the nature of our study, we are unable to determine if the presence of the heparin/pf4 antibody is merely associated with severity of illness, and therefore with length of hospital stay, or if the presence of the antibody itself leads to greater morbidity and therefore an increased number of hospital days. There are a number of possible explanations for the low incidence of thromboembolic complications in our study. A positive antibody test, in the absence of thrombocytopenia, may not confer an increased risk of thromboembolism. Alternatively, because noninvasive tests were not routinely performed to detect clinically silent thrombi, only clinically significant thromboembolic events were pursued by the physicians caring for the study subjects. The results of the heparin/pf4 antibody test were reported to the physicians caring for patients enrolled in the study. Because of the well-founded concern for the morbidity and mortality of untreated HIT, many patients were managed aggressively for suspected HIT, even in the absence of a thrombocytopenia. Subjects may also have been less likely to suffer thromboembolic events because of frequent warfarin use. A large number of subjects (170 out of 277, or 61.4%) were discharged on warfarin; 140 (81.8%) for valve repair or replacement surgery, atrial fibrillation, or the combination, and 30 (18.2%) for other reasons. Seventeen patients with a positive heparin/pf4 antibody were discharged on warfarin, but only three had the heparin/pf4 antibody as their sole indication for anticoagulation. Of the remaining 14 with the antibody, eight also had atrial fibrillation, three underwent valve replacement or repair, two had cerebrovascular disease, and one had atrial fibrillation and Waldenstrom macroglobulinemia. In addition, while 95% of the postoperative antibody tests were performed on postoperative day four or later, only 10% were on postoperative day six or later, and it is possible that testing more subjects beyond five postoperative days would have detected additional patients who developed the heparin/pf4 antibody. Finally, the ELISA used to detect the heparin/pf4 antibodies is very sensitive to the presence of the antibodies but does not describe the functional significance of the antibodies.

6 Ann Thorac Surg EVERETT ET AL 2007;83:592 7 HEPARIN/PF 4 ANTIBODIES AND CARDIAC SURGERY 597 In one study performed in this population, the serotonin release assay (a measure of in vitro activation by the heparin/pf4 antibodies) was positive in only approximately 10% of patients with the antibodies as detected by ELISA [11]. Conversely, in a different published report [13], 14 of 19 (74%) patients with a positive serotonin release assay had a positive ELISA antibody test as well. Because HIT is a serious condition with a 30-day risk of thrombosis of anywhere from 38% to 76%, making the diagnosis prior to a thromboembolic complication is critical [15]. While the ELISA is very sensitive for the heparin/pf4 antibodies, the presence of the antibody in the absence of thrombocytopenia does not, in this population, appear to be associated with an increased risk of thrombosis [8, 11 13, 16]. These data do not support the routine testing for heparin/pf4 antibodies in patients who have recently had cardiac surgery, but do reinforce recently published recommendations regarding the frequency of monitoring platelet counts in patients who are at risk for developing the HIT syndrome [17]. This study was funded by a research grant from GlaxoSmith- Kline. We thank Dena DeJoseph, Josephine Laffan, and Iris McNulty for their help in administering and providing database management for this study. We would also like to thank David Dorer, PhD, of MGH Biostatistics for his assistance in reviewing the statistical methodology of the manuscript. References 1. Jang IK, Hursting MJ. When heparins promote thrombosis: review of heparin-induced thrombocytopenia. Circulation 2005;111: Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia and cardiac surgery. Ann Thorac Surg 2003;76: Chong BH, Fawaz I, Chesterman CN, Berndt MC. Heparininduced thrombocytopenia: mechanism of interaction of the heparin-dependent antibody with platelets. Br J Haematol 1989;73: Warkentin TE, Hayward CP, Boshkov LK, et al. Sera from patients with heparin-induced thrombocytopenia generate platelet-derived microparticles with procoagulant activity: an explanation for the thrombotic complications of heparin-induced thrombocytopenia. Blood 1994;84: Warkentin TE, Kelton JG. A 14-year study of heparininduced thrombocytopenia. Am J Med 1996;101: Arepally GM, Mayer IM. Antibodies from patients with heparin-induced thrombocytopenia stimulate monocytic cells to express tissue factor and secrete interleukin-8. Blood 2001;98: Warkentin TE. Heparin-induced thrombocytopenia: pathogenesis and management. Br J Haematol 2003;121: Trossaert M, Gaillard A, Commin PL, Amiral J, Vissac AM, Fressinaud E. High incidence of anti-heparin/platelet factor 4 antibodies after cardiopulmonary bypass surgery. Br J Haematol 1998;101: Warkentin TE, Sheppard JA, Horsewood P, Simpson PJ, Moore JC, Kelton JG. Impact of the patient population on the risk for heparin-induced thrombocytopenia. Blood 2000;96: Visentin GP, Malik M, Cyganiak KA, Aster RH. Patients treated with unfractionated heparin during open heart surgery are at high risk to form antibodies reactive with heparin: platelet factor 4 complexes. J Lab Clin Med 1996; 128: Pouplard C, May MA, Iochmann S, et al. Antibodies to platelet factor 4-heparin after cardiopulmonary bypass in patients anticoagulated with unfractionated heparin or a low-molecular-weight heparin: clinical implications for heparin-induced thrombocytopenia. Circulation 1999;99: Francis JL, Palmer GJ 3rd, Moroose R, Drexler A. Comparison of bovine and porcine heparin in heparin antibody formation after cardiac surgery. Ann Thorac Surg 2003;75: Bauer TL, Arepally G, Konkle BA, et al. Prevalence of heparin-associated antibodies without thrombosis in patients undergoing cardiopulmonary bypass surgery. Circulation 1997;95: Bennett-Guerrero E, Slaughter TF, White WD, et al. Preoperative anti-pf4/heparin antibody level predicts adverse outcome after cardiac surgery. J Thorac Cardiovasc Surg 2005;130: Hirsh J, Heddle N, Kelton JG. Treatment of heparin-induced thrombocytopenia: a critical review. Arch Intern Med 2004; 164: Pouplard C, May MA, Regina S, et al. Changes in platelet count after cardiac surgery can effectively predict the development of pathogenic heparin-dependent antibodies. Br J Haematol 2005;128: Warkentin TE, Greinacher A. Heparin-induced thrombocytopenia: recognition, treatment, and prevention: the Seventh ACCP Conference on Antithrombotic and Thrombolytic Therapy. Chest 2004;126:311S 37S.