Background: The optimal duration of thromboprophylaxis after major orthopedic surgery is unclear.
Purpose: To compare the benefits and harms of prolonged versus standard-duration thromboprophylaxis after major orthopedic surgery in adults.
Data sources: Cochrane Central Register of Controlled Trials and Scopus from 1980 to July 2011 and MEDLINE from 1980 through November 2011, without language restrictions.
Study selection: Randomized trials reporting thromboembolic or bleeding outcomes that compared prolonged (≥21 days) with standard-duration (7 to 10 days) thromboprophylaxis.
Data abstraction: Two independent reviewers abstracted data and rated study quality and strength of evidence.
Data synthesis: Eight randomized, controlled trials (3 good-quality and 5 fair-quality) met the inclusion criteria. High-strength evidence showed that compared with standard-duration therapy, prolonged prophylaxis resulted in fewer cases of pulmonary embolism (PE) (5 trials; odds ratio [OR], 0.14 [95% CI, 0.04 to 0.47]; absolute risk reduction [ARR], 0.8%), asymptomatic deep venous thrombosis (DVT) (4 trials; relative risk [RR], 0.48 [CI, 0.31 to 0.75]; ARR, 5.8%), symptomatic DVT (4 trials; OR, 0.36 [CI, 0.16 to 0.81]; ARR, 1.5%), and proximal DVT (6 trials; RR, 0.29 [CI, 0.16 to 0.52]; ARR, 7.1%). Moderate-strength evidence showed fewer symptomatic objectively confirmed episodes of venous thromboembolism (4 trials; RR, 0.38 [CI, 0.19 to 0.77]; ARR, 5.7%), nonfatal PE (4 trials; OR, 0.13 [CI, 0.03 to 0.54]; ARR, 0.7%), and DVT (7 trials; RR, 0.37 [CI, 0.21 to 0.64]; ARR, 12.1%) with prolonged prophylaxis. High-strength evidence showed more minor bleeding events with prolonged prophylaxis (OR, 2.44 [CI, 1.41 to 4.20]; absolute risk increase, 6.3%), and insufficient evidence from 1 trial on hip fracture surgery suggested more surgical-site bleeding events (OR, 7.55 [CI, 1.51 to 37.64]) with prolonged prophylaxis.
Limitations: Data relevant to knee replacement or hip fracture surgery were scant and insufficient. Most trials had few events; the strength of evidence ratings that were used may not adequately capture uncertainty in such situations.
Conclusion: Prolonged prophylaxis decreases the risk for venous thromboembolism, PE, and DVT while increasing the risk for minor bleeding in patients undergoing total hip replacement.