Several clinical studies have demonstrated that the

Anti-Platelet Factor 4/Heparin Antibodies An Independent Predictor of 30-Day Myocardial Infarction After Acute Coronary Ischemic Syndromes R. Taylor Williams, MD; Lakshmi V. Damaraju, PhD; Mary Ann Mascelli, PhD; Elliot S. Barnathan, MD; Robert M. Califf, MD; Maarten L. Simoons, MD; Efthymios N. Deliargyris, MD; David C. Sane, MD Background We postulated that antibodies to platelet factor 4/heparin complex might contribute to recurrent ischemic events in patients with acute coronary syndrome. Methods and Results We analyzed serum from patients enrolled in the placebo/unfractionated heparin arm of the GUSTO IV-ACS trial who had high likelihood of prior heparin exposure. We selected 109 patients without thrombocytopenia with the 30-day primary end point (death, myocardial infarction [MI], or revascularization) and 109 age-, gender-, and race-matched controls who did not achieve the primary end point. An ELISA for anti-platelet factor 4/heparin antibodies was performed using 48-hour serum samples. The analyses were done by blinded investigators, and the results were correlated with clinical outcomes. Twenty-three of 218 patients (10.6%) had anti-pf4/heparin antibodies. Patients with anti-pf4/heparin antibodies were more likely to have death or MI (30.4% versus 11.3%, P 0.011) or MI (21.7% versus 6.2%, P 0.008) than patients who were negative for the antibody. After multiple logistic regression analysis, anti-pf4/heparin antibodies remained a predictor of 30-day death or MI (odds ratio, 4.0; 95% CI, 1.4 to 11.3; P 0.0093) and MI (odds ratio, 4.6; 95% CI, 1.4 to 15.0; P 0.0108). The antibody was not associated with the composite end point (death, MI, or revascularization) or with death or revascularization alone. Conclusions Antibodies to the platelet factor 4/heparin complex are a novel, independent predictor of myocardial infarction at 30 days in patients presenting with acute coronary ischemic syndromes. This finding may explain the previous association between thrombocytopenia and adverse events in patients with acute coronary syndrome and may have important implications for the choice of anticoagulant regimens. (Circulation. 2003;107:2307-2312.) Key Words: heparin thrombosis antibodies platelets Several clinical studies have demonstrated that the development of thrombocytopenia in patients with acute coronary ischemic syndromes is associated with higher rates of major adverse clinical events, including hemorrhage and thrombotic events such as death and myocardial infarction (MI). 1 6 Although the association between thrombocytopenia and hemorrhage is apparent, the increased risk of MI and death in this group requires additional study. When multivariate analyses are performed that account for known causes of thrombocytopenia (such as intra-aortic balloon pump and coronary artery bypass grafting), comorbidities, and other factors that are known to predispose to higher adverse event rates, thrombocytopenia remains an independent prognostic indicator for poor outcomes. 2 4 Thus, it does not seem that the presence of thrombocytopenia is simply a predictive marker for identifying a sicker group of patients. Antibodies to the platelet factor 4/heparin complex are linked to the pathogenesis of heparin-induced thrombocytopenia type II (HIT II) and the thrombotic complications associated with this syndrome. 7 Because the antibodies may induce thrombosis by activating the vascular endothelium as well as platelets, 7 we postulated that the presence of these antibodies would increase the risk of thrombotic events, even in nonthrombocytopenic patients. In this study, we have examined this hypothesis using the sera from 218 patients with a high likelihood of prior heparin exposure who were enrolled in the GUSTO IV-ACS trial. 8 Our findings demonstrate that anti-pf4/heparin antibodies are an independent predictor of 30-day MI in this patient population. Received December 31, 2002; revision received February 20, 2003; accepted February 26, 2003. From the Section of Cardiology (R.T.W., E.N.D., D.C.S.), Wake Forest University School of Medicine, Winston-Salem, NC; Centocor (L.V.D., M.A.M., E.S.B.), Malvern, Pa; Duke University Medical Center (R.M.C.), Durham NC; and Thoraxcenter University Hospital (M.L.S.), Rotterdam, the Netherlands. Drs Damaraju, Mascelli, and Barnathan are employees of Centocor, the manufacturer of abciximab, which was evaluated in GUSTO IV ACS. Drs Califf, Simoons, and Sane have received research funding from Centocor. Dr Califf has also received research grants and contracts from Eli Lilly, Schering Plough, Millennium, and Merck. Correspondence to David C. Sane, MD, Wake Forest University School of Medicine, Section of Cardiology, Medical Center Boulevard, Winston-Salem, NC 27157-1045. E-mail dsane@wfubmc.edu 2003 American Heart Association, Inc. Circulation is available at http://www.circulationaha.org DOI: 10.1161/01.CIR.0000066696.57519.AF 2307

2308 Circulation May 13, 2003 TABLE 1. Demographics and Cardiovascular History TABLE 2. Clinical Laboratory Values Methods (n 195) P * Age, y 64.6 10.5 65.4 10.9 0.735 Weight, kg 74.5 16.5 78.2 14.1 0.241 White, % 100 97.4 1.000 Diabetes, % 30.4 26.2 0.626 Hypercholesterolemia, % 43.4 50.8 0.660 Hypertension, % 69.6 62.1 0.649 Current smokers, % 8.7 17.4 0.383 CABG history, % 39.1 23.6 0.128 PCI history, % 26.1 28.2 1.000 Prior MI, % 78.3 85.1 0.370 Prior CHF, % 26.1 12.3 0.102 CABG indicates coronary artery bypass grafting; PCI, percutaneous coronary intervention; and CHF, chronic heart failure. *P values for continuous variables were determined using the t test and for categorical values using Fisher s exact test. Patient Population GUSTO IV-ACS was a randomized, placebo-controlled, multicenter trial designed to assess the safety and efficacy of 24- or 48-hour infusions of abciximab in patients with non ST-segment elevation acute coronary syndrome who were not undergoing early revascularization. 8 Patients were randomly assigned to 1 of 3 groups, placebo (n 2598), 24-hour abciximab infusion (n 2590), or 48- hour abciximab infusion (n 2612). Blood samples for creatine kinase and creatine kinase MB enzyme levels were drawn at baseline and 8, 16, 36, and 48 hours. The primary end point of the study was the 30-day composite end point of death or MI. In the present study, patients were selected for high likelihood of previous heparin exposure based on prior MI, percutaneous coronary intervention, or coronary artery bypass grafting. All patients were selected from the placebo arm (no abciximab therapy) and from the unfractionated heparin study group. Patients with thrombocytopenia (defined as a platelet count 100 000/mm 3 with a decline of 25% from baseline) were excluded. Out of 121 patients meeting these criteria, 12 did not have stored serum samples. Thus, we studied 109 patients with major adverse cardiac events and selected 109 age-, gender-, and race-matched controls from the same group who did not TABLE 3. Patient Medications Randomization to 48 Hours (n 195) (n 195) Hemoglobin, g/dl 13.0 1.2 13.6 1.5 Urea nitrogen,* mg/dl 24.0 10.5 22.7 14.0 Serum creatinine, mg/dl 1.29 0.95 1.04 0.60 Baseline platelet count, mm 3 214 565 61 536 224 736 64 673 Nadir platelet count, mm 3 191 122 50 529 192 959 53 953 *n 6 for antibody positive and n 48 for antibody negative. have adverse outcomes. Serum samples were stored in 1-mL aliquots at 70 C until the assay was performed. ELISA for Anti-PF4 (Bound to Polyvinylsulfonate) All assays were performed in duplicate in a blinded fashion using serum samples from the 48-hour time point (except for 1 sample from 36 hours). The GTI-PF4 assay (Genetic Testing Institute) detects IgG, IgA, and IgM against PF 4 bound to polyvinylsulfonate and is considered a standard assay for the detection of heparindependent antibodies. 9 Statistical Analysis The relationships between anti-pf4/heparin antibodies and clinical events were tabulated, and the probability value was calculated based on the 2 or Fisher s exact test for each variable. The probability values for the continuous variables (demographics) were obtained using the t test. A multiple logistic regression analysis was performed using the independent variables to predict clinical events. The nonsignificant independent variables were dropped, and a final model was determined. Results Twenty-three of 218 patients (10.6%) had a positive anti-pf4/ heparin antibody assay. The ELISA absorbance (405 nm) was 0.510 0.137 in the anti-pf4/heparin positive group and 0.245 0.064 in the anti-pf4/heparin negative patients. The patient demographics, risk factors, and clinical characteristics were similar in the 2 groups (Table 1). There was no significant difference in the time interval between prior MI and study enrollment in the antibody-positive and -negative groups (not shown). 48 Hours to 7 Days (n 195) Aspirin (%) 22 (95.7) 193 (99.0) 23 (100.0) 180 (92.3) Clopidogrel (%) 0 (0) 2 (1.0) 0 (0) 10 (5.1) Ticlopidine (%) 1 (4.3) 6 (3.1) 2 (8.7) 18 (9.2) UFH (%) 23 (100.0) 194 (99.5) 18 (78.3) 157 (80.5) OAC (%) 0 (0) 0 (0) 0 (0) 5 (2.6) ACE inhibitor (%) 7 (30.4) 71 (36.4) 8 (34.8) 87 (44.6) -Blocker (%) 19 (82.6) 156 (80.0) 19 (82.6) 154 (79.0) CCB (%) 4 (17.4) 61 (31.3) 5 (21.7) 62 (31.8) IV NTG (%) 11 (47.8) 117 (60.0) 6 (26.1) 43 (22.1) LA nitrates (%) 14 (60.9) 103 (52.8) 18 (78.3) 130 (66.7) UFH indicates unfractionated heparin; OAC, oral anticoagulants; CCB, calcium channel blocker; IN NTG, intravenous nitroglycerin; and LA, long-acting.

Williams et al Anti-Platelet Factor 4/Heparin Antibodies 2309 TABLE 4. Thirty-Day Outcomes TABLE 5. Multivariable Regression Models (Full Model) (n 195) P * Death 3 (13.0) 12 (6.2) 0.217 MI 5 (21.7) 12 (6.2) 0.008 Death or MI 7 (30.4) 22 (11.3) 0.011 Death, MI, or 12 (52.2) 97 (49.7) 0.826 Revascularization Revascularization 9 (39.1) 83 (42.6) 0.753 *P values were determined using the 2 test. Revascularization includes percutaneous coronary intervention and coronary artery bypass grafting. Table 2 shows pertinent laboratory data from the 2 groups. The baseline and nadir platelet count did not differ between the antibody-positive and -negative groups. Patients with antibody positivity had a slightly higher serum creatinine (1.29 0.95 mg/dl) compared with antibody-negative patients (1.04 0.60 mg/dl). Table 3 show the medications taken by the antibodypositive and -negative groups. There was similar use of antiplatelet and anticoagulant therapy between randomization and 48 hours and between 48 hours to 7 days. The overall GUSTO IV-ACS trial results have been previously published. 8 In the present study, patients with anti- PF4/heparin were more likely to have 30-day death or MI (30.4% versus 11.3%, P 0.011) or MI (21.7% versus 6.2%, P 0.008) than patients who were negative for the antibody (Table 4). After multiple logistic regression analysis was performed, anti-pf4/heparin remained a predictor of 30-day death or MI (odds ratio, 4.0; 95% CI, 1.4 to 11.3; P 0.0093) and MI (odds ratio, 4.6; 95% CI, 1.4 to 15.0; P 0.0108), as shown in Tables 5 and 6. Figure 1 shows the Kaplan-Meier curves for the occurrence of MI (Figure 1A) and death or MI (Figure 1B). Although there was a 2-fold higher rate of death in patients with the antibody (13.0% versus 6.2%), this difference did not reach statistical significance (P 0.217). There was no difference in the occurrence of revascularization (39.1% versus 42.6%; P 0.753) or in the composite end point of death, MI, or revascularization (52.2% versus 49.7%, P 0.826). At 30 days, there were only 2 strokes (1 intracranial hemorrhage and 1 nonhemorrhagic cerebral infarct), both occurring in the antibody-negative group. There was a single case of pulmonary embolism in the antibody-negative group at day 7. At 1 year of follow-up, there were 5 deaths (21.7%) in the antibody-positive group compared with 24 deaths (12.3%) in the antibody-negative group (P 0.203). Discussion The major finding of our study is that patients who have antibodies to the platelet factor 4-heparin complex, even in the absence of thrombocytopenia, have a significantly higher rate of 30-day MI than those who lack these antibodies. Because we analyzed 48-hour samples from patients with prior heparin exposure, our study focused on patients who had anti-pf4/heparin antibodies before study enrollment. The 30-day primary end point (death or MI) was significantly Independent Variable P Odds Ratio 95% CI Model A: death Antibody positive/negative 0.3165 2.121 (0.487 to 9.235) Weight 0.3641 0.980 (0.939 to 1.023) Diabetes 0.3697 1.726 (0.524 to 5.689) Hypertension 0.4680 1.579 (0.460 to 5.428) Previous MI 0.3853 0.564 (0.155 to 2.054) Age 0.0158 1.082 (1.015 to 1.153) Model B: MI Antibody positive/negative 0.0038 6.538 (1.833 to 23.317) Weight 0.0919 1.035 (0.994 to 1.078) Diabetes 0.5458 0.682 (0.196 to 2.364) Hypertension 0.2376 0.528 (0.183 to 1.524) Previous MI 0.4728 1.811 (0.358 to 9.167) Age 0.0176 1.069 (1.012 to 1.130) Model C: death or MI Antibody positive/negative 0.0064 4.456 (1.522 to 13.046) Weight 0.2779 1.018 (0.986 to 1.051) Diabetes 0.9257 0.956 (0.375 to 2.439) Hypertension 0.7320 0.862 (0.368 to 2.017) Previous MI 0.9726 1.020 (0.333 to 3.121) Age 0.0016 1.074 (1.027 to 1.123) Model D: revascularization Antibody positive/negative 0.9048 0.946 (0.379 to 2.361) Weight 0.2600 1.012 (0.991 to 1.034) Diabetes 0.0902 0.567 (0.294 to 1.093) Hypertension 0.1281 0.643 (0.364 to 1.136) Previous MI 0.7006 1.164 (0.536 to 2.531) Age 0.1669 0.981 (0.955 to 1.008) Model E: death or MI or revascularization Antibody positive/negative 0.6456 1.232 (0.507 to 2.992) Weight 0.1591 1.015 (0.994 to 1.036) Diabetes 0.0869 0.576 (0.307 to 1.083) Hypertension 0.2381 0.714 (0.407 to 1.250) Previous MI 0.9392 0.971 (0.458 to 2.059) Age 0.6160 1.007 (0.981 to 1.034) higher in patients in this report (11.3% or 30.4%) compared with the overall study population (8.0% to 9.1%), based largely on our selection of patients with prior MI, percutaneous coronary intervention, or bypass surgery. There are 2 forms of HIT. The first is the nonimmune, milder form with a fall in platelet count 1 to 4 days after the start of heparin administration. This form often resolves even though heparin administration is continued. Type II HIT, in contrast, occurs in 1% to 3% of patients who receive unfractionated heparin. 10 Antibodies from patients with HIT II recognize a complex of platelet factor 4 and heparin or PF4 bound to endothelial cells. 11 13 Antibodies that recognize PF4 alone have also been identified. 13 Seroconversion occurs 5 to 10 days after the initiation of heparin therapy except in

2310 Circulation May 13, 2003 TABLE 6. Multivariable Regression Models (Reduced Model) Independent Variable P Odds Ratio 95% CI Model A: MI Antibody positive/negative 0.0108 4.623 (1.424 to 15.002) Age 0.0637 1.050 (0.997 to 1.105) Model B: death or MI Antibody positive/negative 0.0093 3.978 (1.405 to 11.264) Age 0.0029 1.066 (1.022 to 1.112) Model C: death Age 0.0068 1.085 (1.023 to 1.151) patients with prior heparin therapy. These patients may have a more rapid onset of thrombocytopenia. 9 Although thrombocytopenia is a defining characteristic of HIT II, the platelet count is often not severely depressed, with a median platelet count of 60 000/mm 3. 14 In many patients with anti-pf4/heparin antibodies, the platelet count falls but remains within the normal range. Thrombotic complications can occur in patients with normal platelet counts 13 and thrombosis is the principal clinical manifestation regardless of the degree of thrombocytopenia. 15 The spectrum of thrombocytopenia in the HIT II syndrome may be related to the variability of responses to platelets to heparin-dependent IgG antibodies. 16 19 In addition to activating platelets, these antibodies can also induce a thrombotic response on the endothelial surface. 10,11 It is possible that some subclasses of antibodies react better with platelets whereas others preferentially target the endothelium, with thrombosis being the clinical manifestation in either case. Our study specifically involved patients who had acute coronary syndromes to examine the clinical significance of anti-pf4/ antibodies in the absence of thrombocytopenia. In the absence of this clinical marker, these patients might escape suspicion for being at increased risk of thrombotic events. In HIT II, the risk of thrombosis is 5% to 10% per day for the first 2 days after heparin is discontinued 20 and may persist for weeks thereafter. In a retrospective 14-year single-center study of HIT, Warkentin and Kelton 21 characterized 127 patients with HIT II confirmed using the 14 C-serotoninrelease assay. A subset of 62 patients who initially presented with isolated thrombocytopenia had a remarkable 52.8% rate of thrombosis at 30 days. 21 Most events were venous, with only 1 of these 62 patients (1.6%) having had a MI. 21 From this and other studies, the major thrombotic manifestation of Kaplan-Meier curves are shown for the occurrence of MI (A) and death or MI (B). Antibody-positive patients are indicated by dashed lines.

Williams et al Anti-Platelet Factor 4/Heparin Antibodies 2311 HIT II seems to be venous thrombosis, with a relatively low risk of MI. However, the baseline medical condition being treated before the onset of HIT II or anti-pf4 seroconversion likely plays a major role in determining the thrombotic manifestations. Patients being anticoagulated for hip surgery, for example, have a high rate of venous thromboembolism as clinical manifestation of HIT II. 15 The association between anti-pf4/heparin and 30-day MI was likely accentuated in our study because our patient population presented with acute coronary syndromes. Persistent thrombin generation from the culprit lesion for up to 30 days after the acute event, 22,23 along with antibody activation of platelets and endothelial cells in the vicinity of the plaque, could lead to recurrent thrombosis within the coronary artery. Our results confirm a previous study that examined the effect of anti-pf4/heparin in an acute coronary syndrome population. Mattioli et al 24 measured anti-pf4/heparin titers in 124 consecutive patients with unstable angina. Thirty-eight patients (30%) had a positive anti-pf4/heparin ELISA. At 1-year follow-up, the combined end point (death, MI, recurrent angina, urgent revascularization, stroke) was 66% in patients with anti-pf4 antibodies and 44% in the antibodynegative group (P 0.01). Notably, these 2 groups did not have significantly different platelet counts, confirming that the presence of anti-pf4 antibodies cannot always be discerned by the development of thrombocytopenia. Our study is also consistent with a recent report of reduced vein graft patency in 18 patients who developed HIT in the post coronary artery bypass grafting period. Patients with HIT had 68% saphenous vein graft occlusions within 6 months of surgery versus only 20% occlusion for controls, P 0.001. 25 The introduction of direct thrombin inhibitors has provided an attractive alternative to heparin for managing patients with acute coronary syndromes. Bivalirudin has been approved for patients with unstable angina undergoing percutaneous coronary interventions. 26 Lepirudin 27 and argatroban 28 have been approved by the Food and Drug Administration for treatment of HIT-related thrombosis. A recent meta-analysis has shown that there is a 15% reduction in death or MI when patients with acute coronary syndromes are treated with direct thrombin inhibitors compared with heparin. 29 The superiority may be attributable to better penetration into thrombi and inhibition of thrombus-bound thrombin. 30 The lack of generation of anti-pf4/heparin antibodies could be another mechanism for the benefit of these agents. Although alternatives are presently available for treating patients with HIT when it is recognized, our study raises the concern that a group of patients with acute coronary syndrome and anti-pf4/heparin antibodies may evade clinical detection. Additional studies will be necessary to determine if patients should be screened for the antibody and whether the substitution of direct thrombin inhibitors for heparin in this subset of patients with acute coronary syndromes will be especially beneficial. Study Limitations There are several limitations to our study. The number of patients with antibodies to PF4/heparin is small, despite selecting for patients with prior heparin exposure. Therefore, to assess the application of our report to the general acute coronary syndrome population, these findings will need to be duplicated in a larger cohort of patients. It is likely that some additional patients who were treated with heparin during the course of this study developed anti-pf4/heparin antibodies that contributed to 30-day events but whose seroconversion was not detected because of blood sample collection at the 48-hour time point. On the other hand, the earlier detection of the antibody may allow for alternative treatment, especially using direct thrombin inhibitors during the hospital course and agents such as clopidogrel for outpatient therapy. Conclusions In patients with an acute coronary syndrome who have probable prior heparin exposure, the presence of antibodies to PF4/heparin is associated with a higher rate of 30-day MI. This finding has potential implications for risk stratification, antibody screening, and the choice of anticoagulant therapy. References 1. Bovill EG, Terrin ML, Stump DC, et al for the TIMI Investigators. Hemorrhagic events during therapy with recombinant tissue-type plasminogen activator, heparin, and aspirin for acute myocardial infarction: results of the Thrombolysis in Myocardial Infarction (TIMI) Phase II trial. Ann Intern Med. 1991;115:256 265. 2. Harrington RA, Sane DC, Califf RM, et al for the Thrombolysis and Angioplasty in Myocardial Infarction Study Group. Clinical importance of thrombocytopenia occurring in the hospital phase after administration of thrombolytic therapy for acute myocardial infarction: the Thrombolysis and Angioplasty in Myocardial Infarction Study Group. J Am Coll Cardiol. 1994;23:891 898. 3. Berkowitz SD, Sane DC, Sigmon KN, et al. Occurrence and clinical significance of thrombocytopenia in a population undergoing high-risk percutaneous coronary revascularization: Evaluation of the c7e3 for the Prevention of Ischemic Complications (EPIC) Study Group. J Am Coll Cardiol. 1998;32:311 319. 4. McClure MW, Berkowitz SD, Sparapani R, et al for the PURSUIT Investigators. Clinical significance of thrombocytopenia during non-st elevation acute coronary syndromes: the platelet glycoprotein IIb/IIIa in unstable angina receptor suppression using integrilin therapy (Pursuit) trial experience. Circulation. 1999;99:2892 2900. 5. Sane DC, Damaraju LV, Topol EJ, et al. Occurrence and clinical significance of pseudothrombocytopenia during abciximab therapy. J Am Coll Cardiol. 2000;36:75 83. 6. Eikelboom JW, Anand SS, Mehta SR, et al. Prognostic significance of thrombocytopenia during hirudin and heparin therapy in acute coronary syndrome without ST elevation: Organization to Assess Strategies for Ischemic Syndromes (OASIS-2) study. Circulation. 2001;103:643 650. 7. Warkentin TE, Greinacher A. Heparin-Induced Thrombocytopenia. New York: Marcel Dekker; 2002. 8. Simoons ML. GUSTO IV-ACS Investigators: effect of glycoprotein IIb/IIIa receptor blocker abciximab on outcome in patients with acute coronary syndromes without early coronary revascularisation. The GUSTO IV-ACS randomised trial. Lancet. 2001;357:1915 1919. 9. Warkentin TE, Kelton JG. Temporal aspects of heparin-induced thrombocytopenia. N Engl J Med. 2001;344:1286 1292. 10. Warkentin TE, Chong BH, Greinacher A. Heparin-induced thrombocytopenia: towards consensus. Thromb Haemost. 1998;79:1 7. 11. Visentin GP, Ford SE, Scott JP, et al. Antibodies from patients with heparin-induced thrombocytopenia/thrombosis are specific for platelet factor 4 complexed with heparin or bound to endothelial cells. J Clin Invest. 1994;93:81 88. 12. Kelton JG, Smith JW, Warkentin TE, et al. Immunoglobulin G from patients with heparin-induced thrombocytopenia binds to a complex of heparin and platelet factor 4. Blood. 1994;83:3232 3239. 13. Greinacher A, Potzsch B, Amiral J, et al. Heparin-associated thrombocytopenia: isolation of the antibody and characterization of a multimo-

2312 Circulation May 13, 2003 lecular PF4-heparin complex as the major antigen. Thromb Haemost. 1994;72:783 784. 14. Warkentin TE. Clinical presentation of heparin-induced thrombocytopenia. Semin Hematol. 1998;35(suppl 5):9 16. 15. Warkentin TE, Levine MN, Hirsh J, et al. Heparin-induced thrombocytopenia in patients treated with low-molecular-weight heparin or unfractionated heparin. N Engl J Med. 1995;332:1330 1335. 16. Warkentin TE, Hayward CPM, Smith CA, et al. Determinants of donor platelet variability when testing for heparin-induced thrombocytopenia. J Lab Clin Med. 1992;120:371 379. 17. Chong BH, Pilgrim RL, Cooley MA, et al. Increased expression of platelet IgG Fc receptors in immune heparin-induced thrombocytopenia. Blood. 1993;81:988 993. 18. Chong BH, Burgess J, Ismail F. The clinical usefulness of the platelet aggregation test for the diagnosis of heparin-induced thrombocytopenia. Thromb Haemost. 1993;69:344 350. 19. Isenhart CE, Brandt JT. Platelet aggregation studies for the diagnosis of heparin-induced thrombocytopenia. Am J Clin Pathol. 1993;99:324 330. 20. Greinacher A, Volpel H, Janssens U, et al for the HIT investigators. Recombinant hirudin (lepirudin) provides safe and effective anticoagulation in patients with heparin-induced thrombocytopenia: a prospective study. Circulation. 1999;99:73 80. 21. Warkentin TE, Kelton JG. A 14-year study of heparin-induced thrombocytopenia. Am J Med. 1996;101:502 507. 22. Merlini PA, Ardissino D, Rosenberg RD, et al. In vivo thrombin generation and activity during and after intravenous infusion of heparin or recombinant hirudin in patients with unstable angina pectoris. Arterioscler Thromb Vasc Biol. 2000;20:2162 2166. 23. Belle EV, Lablanche J-M, Bauters C, et al. Coronary angioscopic findings in the infarct-related vessel within 1 month of acute myocardial infarction. Circulation. 1998;97:26 33. 24. Mattioli AV, Bonetti L, Sternieri S, et al. Heparin-induced thrombocytopenia in patients treated with unfractionated heparin: prevalence of thrombosis in a 1 year follow-up. Ital Heart J. 2000;1:39 42. 25. Liu JC, Lewis BE, Steen LH, et al. Patency of coronary artery bypass grafts in patients with heparin-induced thrombocytopenia. Am J Cardiol. 2002;89:979 981. 26. Antman EM, Braunwald E. A second look at bivalirudin. Am Heart J. 2001;142:929 931. 27. Greinacher A, Eichler P, Lubenow N, et al. Heparin-induced thrombocytopenia and thromboembolic complications: meta-analysis of 2 prospective trials to assess the value of parenteral treatment with lepirudin and its therapeutic aptt range. Blood. 2000;96:846 851. 28. Lewis BE, Walenga JM, Wallis DE. Anticoagulation with Novastan (Argatroban) in patients with heparin-induced thrombocytopenia and heparin-induced thrombocytopenia and thrombosis syndrome. Semin Thromb Hemost. 1997;23:197 202. 29. The Direct Thrombin Inhibitor Trialists Collaborative Group. Direct thrombin inhibitors in acute coronary syndromes: principal results of a meta-analysis based on individual patients data. Lancet. 2002;359:294 302. 30. Weitz JI, Hudoba M, Massel D, et al. Clot-bound thrombin is protected from inhibition by heparin-antithrombin III but is susceptible to inactivation by antithrombin III-independent inhibitors. J Clin Invest. 1990;86: 385 391.