Evaluation of diagnostic tests and argatroban or lepirudin therapy in patients with suspected heparin-induced thrombocytopenia.

STUDY OBJECTIVES

To evaluate diagnostic tests for heparin-induced thrombocytopenia (HIT), a serious drug reaction that can occur in patients receiving heparin, and to evaluate treatment with direct thrombin inhibitors-the only initial drug therapy that decreases the risk of thromboembolism associated with immune-mediated HIT.

DESIGN

Retrospective cohort study.

SETTING

University teaching hospital.

SUBJECTS

Patients with HIT treated with argatroban or lepirudin between January 1, 2000, and December 31, 2003.

MEASUREMENTS AND MAIN RESULTS

Patients were assessed for dosage and duration of argatroban or lepirudin therapy, HIT diagnostic tests, and clinically significant adverse events. Thirty-four patients received argatroban, 42 received lepirudin. Mean+/-SD doses of argatroban and lepirudin were 1.2+/-0.9 microg/kg/minute and 0.09+/-0.11 mg/kg/hour, respectively; both were 40% lower than the recommended doses. Mean duration of therapy was 10+/-9 days. Supratherapeutic activated partial thromboplastin times were observed in 30 (39%), 10 (13%) , and six (8%) of 76 patients on days 1, 2, and 3, respectively (p<0.0001). Clinically significant bleeding occurred in 6% of patients receiving argatroban and in 5% of those receiving lepirudin (p=0.99); all patients had an activated partial thromboplastin time of longer than 100 seconds. Although platelet-aggregation tests were ordered in 55 (72%) of 76 patients, they were not useful in 16 (29%) because of equivocal or contradictory results.

CONCLUSION

Due to supratherapeutic activated partial thromboplastin times, our patients often required doses of argatroban and lepirudin lower than those usually recommended. Thus, direct thrombin inhibitors should be started at low initial doses and titrated to target activated partial thromboplastin times to achieve appropriate efficacy and to avoid increasing the risk of bleeding. Platelet-aggregation tests were least useful for evaluating HIT. Appropriate diagnostic strategies should be used to avoid unnecessary drug use.