BACKGROUND

Both unfractionated heparin (UFH) and low-molecular-weight heparin (LMWH) are approved for venous thromboembolism (VTE) prophylaxis. Which agent is superior remains controversial.

OBJECTIVE

To compare the effectiveness, complications, and costs of UFH and LMWH as VTE prophylaxis for hospitalized medical patients.

DESIGN

Retrospective cohort.

SETTING

Three hundred thirty-three acute care facilities in 2004-2005.

PATIENTS

Adults with 4 common medical diagnoses considered to carry moderate-to-high risk of VTE. Excluded were patients on warfarin or with hospital stays of ≤ 2 days. VTE prophylaxis was assessed from billing data.

INTERVENTION

None.

MEASUREMENTS

VTE, major bleeding or heparin-induced thrombocytopenia, mortality, and cost.

RESULTS

Of 32,104 patients who received prophylaxis, 55% received LMWH and the remainder received UFH. The hospital where the patient obtained care was the strongest predictor of receiving LMWH. VTE was observed in 163 (0.51%) patients; complications, followed by stopping therapy, were rare (<0.2%). In analysis adjusted for the propensity for UFH and other covariates, patients treated with UFH had an odds ratio for VTE of 1.04 (95% confidence interval [CI] 0.76 to 1.43) compared to LMWH. In a grouped treatment model, the odds of VTE with UFH was 1.14 (95% CI 0.72 to 1.81). Adjusted odds of bleeding with UFH compared to LMWH were 1.64 (95% CI 0.50 to 5.33), adjusted odds of complications followed by stopping prophylaxis were 2.84 (95% CI 1.43 to 45.66), and adjusted cost ratio was 0.97 (95% CI 0.90 to 1.05).

CONCLUSIONS

For VTE prophylaxis, the effectiveness and cost of LMWH and UFH are similar, but LMWH is associated with fewer complications.