Is DVT prophylaxis overemphasized? A randomized prospective study

This study was designed to prospectively evaluate a previously published prognostic index for predicting deep venous thrombosis (DVT) in general surgical patients with conventional prophylaxis. Patients undergoing procedures of at least 1 hr duration (abdominal, thoracic, head and neck, inguinal) requiring general or spinal anesthetic were prospectively randomized into the following groups: Group 1, sequential pneumatic compression devices during surgery and 2 days postoperatively; Group 2, subcutaneous heparin (5000 U q 12 hr) starting 1 hr before surgery and for 7 days postop; Group 3, control group. All patients underwent duplex evaluation of bilateral lower extremity deep venous systems preoperatively and on postoperative Days 1, 3, and 30. In addition, a previously developed predictive DVT incidence indicator, the prognostic index (PI), was calculated for each patient. A total of 137 patients were entered into the study with 29 removed for patient/staff reasons. There were no differences in PI among the three groups at the 0.05 level (ANOVA). The distribution of risk factors for DVT including increased age, body size, hemoglobin (Hb), and colorectal procedures were distributed evenly among the groups. Additional factors such as diabetes, COPD, PVD, immobilization, and cancer were also evenly distributed among the groups. The PI predicted a 20% incidence of DVT. For Groups 1 (n = 25), 2 (n = 38), and 3 (n = 45) no DVTs were detected over the 30 days of study. During the study period, 8 DVTs were detected by duplex evaluation in general surgical patients not in the study (1.5%). In conclusion, in a prospective randomized study using sequential pneumatic compression devices, subcutaneous heparin or no prophylaxis in matched general surgical patients at moderate to high risk for thromboembolism, no DVTs occurred for up to 30 days. Furthermore, neither a PI nor other factors associated with DVT accurately predicted the incidence of DVT in this patient population.