Low-molecular-weight heparin versus compression stockings for thromboprophylaxis after knee arthroscopy: a randomized trial.

Background: Knee arthroscopy, the most common orthopedic operation worldwide, carries a definite risk for deep venous thrombosis; however, postsurgical thromboprophylaxis is not routinely recommended.

Objective: To evaluate whether low-molecular-weight heparin (LMWH) better prevents deep venous thrombosis and does not cause more complications than graduated compression stockings in adults having knee arthroscopy.

Design: Assessor-blind, randomized, controlled trial.

Setting: The Department of Knee Surgery, Abano Terme Clinic, Abano Terme (knee surgery, random assignment, and bleeding event survey), and the Unit of Angiology, University Hospital of Padua, Padua (efficacy outcomes evaluation, follow-up, data management, and analysis), Italy.

Patients: 1761 consecutive patients undergoing knee arthroscopy between March 2002 and January 2006.

Intervention: Patients were randomly assigned to wear full-length graduated compression stocking for 7 days (660 patients) or to receive a once-daily subcutaneous injection of LMWH (nadroparin, 3800 anti-Xa IU) for 7 days (657 patients) or 14 days (444 patients). The data and safety monitoring board prematurely stopped the 14-day heparin group after the second interim analysis.

Measurements: Combined incidence of asymptomatic proximal deep venous thrombosis, symptomatic venous thromboembolism, and all-cause mortality (primary efficacy end point) and combined incidence of major and clinically relevant bleeding events (primary safety end point). All patients had bilateral whole-leg ultrasonography at the end of the allocated prophylactic regimen or earlier if indicated. All patients with normal findings were followed for 3 months, and none was lost to follow-up.

Results: The 3-month cumulative incidence of asymptomatic proximal deep venous thrombosis, symptomatic venous thromboembolism, and all-cause mortality was 3.2% (21 of 660 patients) in the stockings group, 0.9% (6 of 657 patients) in the 7-day LMWH group (absolute difference, 2.3 percentage points [95% CI, 0.7 to 4.0 percentage points]; P = 0.005), and 0.9% (4 of 444 patients) in the prematurely stopped 14-day LMWH group. The cumulative incidence of major or clinically relevant bleeding events was 0.3% (2 of 660 patients) in the stockings group, 0.9% (6 of 657 patients) in the 7-day LMWH group (absolute difference, -0.6 percentage point [CI, -1.5 to 0.2 percentage points]), and 0.5% (2 of 444 patients) in the 14-day LMWH group.

Limitations: The study was not double-blind or double-dummy. Almost half of the events making up the composite outcome measure were distal deep venous thromboses. Stockings were used instead of placebo because of local prophylaxis policies.

Conclusion: In patients undergoing knee arthroscopy, prophylactic LMWH for 1 week reduced a composite end point of asymptomatic proximal deep venous thrombosis, symptomatic venous thromboembolism, and all-cause mortality more than did graduated compression stockings.