Thromboembolic disease in spinal surgery: a systematic review.

STUDY DESIGN Systematic review of the literature and analysis of pooled data. OBJECTIVES To better understand the incidence of thromboembolic disease in postoperative spinal patients, and to establish a starting point for defining appropriate postoperative prophylaxis protocols. SUMMARY OF BACKGROUND DATA The risk of thromboembolic disease is well studied for some orthopedic procedures. However, the incidence of postoperative thromboembolic disease is less well-defined in patients who have had spinal surgery. METHODS The MEDLINE database was queried using the search terms deep venous thrombosis or DVT, pulmonary embolus, thromboembolic disease, and spinal or spine surgery. Abstracts of all identified articles were reviewed. Detailed information from eligible articles was extracted. Data were compiled and analyzed by simple summation methods when possible to stratify rates of DVT and/or pulmonary embolus for a given prophylaxis protocol, screening method, and type of spinal surgery. RESULTS Twenty-five articles were eligible for full review. DVT risk ranged from 0.3% to 31%, varying between patient populations and methods of surveillance. Pooling data from the 25 studies, the overall rate of DVT was 2.1%. DVT rate was influenced by prophylaxis method: no prophylaxis, 2.7%; compression stockings (CS), 2.7%; pneumatic sequential compression device (PSCD), 4.6%; PSCD and CS, 1.3%; chemical anticoagulants, 0.6%; and inferior vena cava filters with/without another method of prophylaxis, 22%. DVT rate was also influenced by the method of diagnosis, ranging from 1% to 12.3%. CONCLUSION As risk of DVT after routine elective spinal surgery is fairly low, it seems reasonable to use CS with PSCD as a primary method of prophylaxis. There is insufficient evidence to support or refute the use of chemical anticoagulants in routine elective spinal surgery. In addition, there is insufficient evidence to suggest that screening patients undergoing elective spinal surgery with ultrasound or venogram is routinely warranted.