Venous thromboembolism (VTE) is a common complication among patients in the intensive care unit. While anticoagulation remains standard therapy, vena caval filters are an important alternative when anticoagulation is contraindicated. To determine the safety and efficacy of vena caval filters in the treatment of VTE, a comprehensive review of the English-language medical literature was performed. Except for one randomized controlled trial, the literature supporting the use of vena caval filters consists almost exclusively of case series, which in many instances are limited by incomplete and short follow-up. While case series suggest that filters function effectively in the prevention of pulmonary embolism (2%-4% symptomatic pulmonary embolism [PE], fatal PE < 2%), recent higher quality studies indicate that filters may not provide significant additional protection to that provided by anticoagulation alone. Furthermore, filters are associated with a 2- fold increase in the incidence of recurrent DVT. Until randomized comparative studies are available, the safety and efficacy of all the available devices should be considered to be roughly equivalent. Since filters do not inhibit continued clot formation, all filter patients should receive anticoagulation for durations appropriate for their thrombotic disorder. Although extended anticoagulation may prevent thrombotic complications associated with filter placement, this strategy has yet to be experimentally tested. While many additional indications for vena caval filter use have been proposed (VTE in cancer patients, PE prophylaxis in trauma patients, etc), well-designed clinical trials demonstrating their efficacy in these situations are lacking. Further development of temporary/retrievable filters, which offer the potential to avoid the long-term complications of permanent filters, should be a research priority. Until additional data are available, vena caval filters should generally be restricted to patients with VTE who cannot receive anticoagulation.