Introduction A prospective cohort study was conducted to assess the incidence of venous thromboembolism (VTE) in 300 patients undergoing elective spinal surgery. Although this subject has been extensively studied in patients undergoing joint replacement surgery, the true incidence of VTE related to spinal surgery remains unknown. We also sought to assess the efficacy of various forms of VTE prophylaxis, another area which has not been adequately studied.

Methods Bilateral lower extremity venous duplex scans were performed pre-operatively, within one week postoperatively and at 4 to 6 weeks post-operatively. Information was collected regarding age, gender, body mass index, type and duration of surgical procedure, intra-operative blood loss and blood transfusion, pre- and post-operative mobility and other risk factors for VTE. All patients received vigorous mechanical prophylaxis with 56% of patients receiving pharmacological prophylaxis (unfractionated or low molecular weight heparin) according to surgeon preference.

Results The overall incidence of post-operative deep vein thrombosis (DVT) was 3.4%. The incidence in those receiving pharmacological prophylaxis was 1.2% versus 6.3% in those who received mechanical prophylaxis alone (p< 0.05). Just under half of the post-operative DVTs (1.4%) occurred after hospital discharge with none of these patients receiving heparin. There were two cases of pulmonary embolism (0.7%), both occurring during hospitalization, and there were no deaths. There was no difference in blood loss or transfusion rates between patients receiving or not receiving pharmacological prophylaxis and there were no serious complications related to blood loss. There was a surprisingly high incidence of abnormalities detected on pre-operative scanning, with DVT occurring in 3.0% and superficial thrombophlebitis in 24.8%.

Discussion The incidence of DVT in patients after spinal surgery was 3.4% and was significantly lower (1.2 %) in patients who received pharmacological prophylaxis. Given that the rates of bleeding-related complications were not different between the two groups, we conclude that it is both efficacious and safe to use pharmacological prophylaxis in combination with mechanical prophylaxis routinely in patients undergoing spinal surgery. A 27.8% incidence of pre-operative venous abnormalities may reflect pre-operative immobility due to pain in this group of patients, and justifies the use of ultrasound scanning as an important pre-operative screening tool. In addition, the high incidence of late-onset DVT justifies the need for follow-up scanning several weeks after discharge. This study is the amongst the largest of its type to date, however a larger, randomized and controlled trial is now necessary to confirm these results and allow the formulation of unambiguous guidelines for spinal surgery and VTE prophylaxis.