Variables associated with postoperative deep venous thrombosis: a prospective study of 411 gynecology patients and creation of a prognostic model.

Deep venous thrombosis is a major complication following gynecologic surgery. Assessing a patient's risk of developing deep venous thrombosis is important for patient selection and in choosing appropriate prophylactic methods. Four hundred eleven patients undergoing major gynecologic surgery were evaluated prospectively. All known variables associated with deep venous thrombosis were recorded. Deep venous thrombosis was diagnosed by 125I fibrinogen leg counting of all patients. Univariate analysis of all variables identified the following to be significantly related (P less than .05) to postoperative deep venous thrombosis: a prior history of deep venous thrombosis, leg edema or venous stasis changes, venous varicosities, degree of preoperative ambulation, type of surgery, nonwhite race, recurrent malignancy, prior pelvic radiation therapy, age above 45 years, excessive body weight, intraoperative blood loss, and duration of anesthesia. A stepwise logistic regression analysis of these variables was performed. The following preoperative prognostic factors remained significant: type of surgery, age, leg edema, nonwhite patients, severity of venous varicosities, prior radiation therapy, and prior history of deep venous thrombosis. Duration of anesthesia was also important when intraoperative factors were considered in the analysis. Using these factors, a prognostic model was created and tested. The model resulted in a degree of concordance of 0.82 and allows one to evaluate the risks of postoperative deep venous thrombosis for an individual patient.