A Predictive Index for Postoperative Deep Vein Thrombosis in Thoracic Surgery Patients

In a single-centre prospective trial 200 consecutive patients undergoing thoracic surgery were randomised to receive one of two prophylactic regimes against deep vein thrombosis (DVT). These were 5000 units of subcutaneous heparin twice a day, alone or combined with the wearing of graded compression stockings. The diagnosis of DVT was made clinically and with 131I labelled fibrinogen. Six DVTs developed in the stocking group and 11 in the non-stocking group. The results suggest that the use of stockings reduces the incidence of DVT when added to herparin but the difference is not statistically significant. To obtain a predictive index for the development of DVT, discriminant analysis was applied to the control and stocking groups separately and combined. Five simple clinical variables gave a true positive prediction rate, for the combined group, of 94% and a false positive prediction rate of 26%.