Replacement of unfractionated heparin by low-molecular-weight heparin for postorthopedic surgery antithrombotic prophylaxis lowers the overall risk of symptomatic thrombosis because of a lower frequency of heparin-induced thrombocytopenia

Heparin-induced thrombocytopenia (HIT) is a prothrombotic immune-mediated adverse drug reaction that usually begins 5 or more days after starting heparin.¹  One randomized clinical trial reported a significant difference in HIT between unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) in postoperative orthopedic patients.²  To corroborate that LMWH is safer than UFH in this patient population, we performed a “before-after” prospective cohort study in which 231 consecutive patients over a 9-month period received UFH (Liquemin [Roche, Basel, Switzerland], 5000 IU 3 times/day subcutaneously) following hip or knee replacement. During the subsequent 12-month period, 271 patients received LMWH (enoxaparin [Clexane; Sanofi-Aventis, Frankfurt, Germany], 40 mg/d subcutaneously). Both anticoagulants were started the night before surgery. The primary study end point was HIT, defined a priori as a positive test for HIT antibodies (platelet activation assay³ ) plus a 30% or greater fall in the platelet count between postoperative days 4 to 16, and/or symptomatic, objectively documented thrombosis (venous or arterial). We predefined thrombotic events occurring between days 5 to 16 as potentially HIT related. Routine testing for HIT antibodies was also performed in most patients at end of study (day 16). Stored sera were also tested posthoc for anti-PF4/polyanion antibodies (GTI, Brookfield, WI).⁴  Clinical events were adjudicated by 2 investigators by consensus. The study was approved by the ethics committee of the University of Greifswald. The patients we report overlap somewhat with those included in a previous publication⁵  reflecting the participating surgeons' perspective; however, that report included an additional 130 nonstudy patients, used a different definition of HIT (> 50% platelet count fall), and did not report systematic serologic investigations.

We found a higher frequency of HIT in patients who received UFH (5.2%; 95% confidence interval [CI], 2.7%-8.9%), compared with LMWH (0%; 95% CI, 0.0%-1.4%); P < .001 (Table 1). In addition to the 6 patients who had thrombosis associated with HIT, 5 patients had thrombosis without HIT (3 in the UFH group, 2 in the LMWH group). Thus, total thrombosis (both HIT and non-HIT) was significantly more frequent in UFH-treated patients (Table 1). Two other findings deserve comment. First, among the 431 patients undergoing serologic testing who did not develop HIT manifesting as a platelet count fall, there was a strong association between symptomatic thrombosis after day 4 and a positive platelet activation test for HIT antibodies: 3 (50.0%) of 6 patients with thrombosis tested antibody positive, compared with 26 (6.1%) of 425 patients without thrombosis (odds ratio [OR], 15.3 [95% CI, 2.94-25.23]; P = .005). This supports previous suggestions⁶  that formation of platelet-activating HIT antibodies can be associated with thrombosis even in the absence of a significant platelet count fall. Second, among the 87.7% of study patients who underwent serologic testing for HIT antibodies, there was approximately a 60% lower seroconversion rate with LMWH compared with UFH (55.6% and 64.7% by the platelet activation assay and immunoassay, respectively). Nevertheless, despite this moderate reduction in frequency of HIT antibody formation, there was complete avoidance of clinical HIT using LMWH (0 versus 12 cases). This underscores the importance of considering HIT to be a largely preventable adverse drug reaction, at least in postorthopedic surgery patients.