July 2013

Pharmacologic and Mechanical Strategies for Preventing Venous Thromboembolism After Bariatric SurgeryA Systematic Review and Meta-analysis

Author Affiliations
  • 1Department of Medicine, The Johns Hopkins School of Medicine, The Johns Hopkins University, Baltimore, Maryland
  • 2Health Policy and Management, The Johns Hopkins Bloomberg University School of Public Health, The Johns Hopkins University, Baltimore, Maryland
  • 3Department of Surgery, The Johns Hopkins School of Medicine, The Johns Hopkins University, Baltimore, Maryland

Copyright 2013 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.

JAMA Surg. 2013;148(7):675-686. doi:10.1001/jamasurg.2013.72

We sought to assess the comparative effectiveness and safety of pharmacologic and mechanical strategies to prevent venous thromboembolism (VTE) in patients undergoing bariatric surgery. We searched (through August 2012) for primary studies that had at least 2 different interventions. Of 30 902 citations, we identified 8 studies of pharmacologic strategies and 5 studies of filter placement. No studies randomized patients to receive different interventions. One study suggested that low-molecular-weight heparin is more efficacious than unfractionated heparin in preventing VTE (0.25% vs 0.68%, P < .001), with no significant difference in bleeding. One study suggested that prolonged therapy (after discharge) with enoxaparin sodium may prevent VTE better than inpatient treatment only. There was insufficient evidence supporting the hypothesis that filters reduce the risk of pulmonary embolism, with a point estimate suggesting increased rates with filters (pooled relative risk [RR], 1.21 95% CI, 0.57-2.56). There was low-grade evidence that filters are associated with higher mortality (pooled RR, 4.30 95% CI, 1.60-11.54) and higher deep vein thrombosis rates (2.94 1.35-6.38). There was insufficient evidence to support that augmented subcutaneous enoxaparin doses (>40 mg daily or 30 mg twice daily) are more efficacious than standard dosing, with a trend toward increased bleeding. Of note, for both filters and augmented pharmacologic dosing strategies, patients at highest risk for VTE were more likely to receive more intensive interventions, limiting our ability to attribute outcomes to prophylactic strategies used.