Independent Predictors of 30-Day Perioperative Deep Vein Thrombosis in 1346 Consecutive Patients After Spine Surgery.

Background Deep vein thrombosis (DVT) is a morbid postsurgical complication. Identifying the perioperative DVT risk profile will improve surgeons' ability to assess patients for surgical candidacy. In addition, these data will help to identify patients who would benefit from DVT chemoprophylaxis. Methods We evaluated all medical records for 1346 consecutive patients who underwent spinal surgery at Duke University for incidence of DVT within 30 days of surgery and documented all demographic, preoperative, operative, and postoperative variables. DVT treatment and long-term outcomes were also documented. Associations between postoperative DVT and individual risk factors in all patients were determined using adjusted logistic regression analysis. Patients were stratified into emergent and elective groups and a similar analysis was performed. Results Overall, 15 patients (1.1%) had a DVT in the 30 days after surgery, 7 patients (0.6%) after elective surgery and 8 patients (4.2%) after emergent surgery ( P  = 0.03). Overall, multivariate logistic regression determined that previous DVT, postoperative urinary tract infection, and creatinine level >2.0 mg/dL were identified as positive predictors. When stratified by emergent surgery, we found packed red blood cell transfusion, surgical blood loss >2.0 L, and deep surgical site infection to be independently associated with increased risk of postoperative DVT. When stratified by elective surgery, we found that coronary artery disease and atrial fibrillation were associated with increased risk of DVT. No patients died in the 30-day perioperative period and 5 (33.3%) patients died within 1 year. Conclusions This study identifies patient factors predictive of postoperative DVT. Postoperative DVT prophylaxis may be warranted for patients undergoing emergent spine surgery because these patients have significantly higher risk of developing postoperative DVT.