Surgery

PATIENTS treated by a general surgical service were divided into three groups: those with a high risk o! venous thromboembolism in whom prophylactic anticoagulant therapy was possible, those at high risk in whom anticoagulants were contraindicated, and those at low risk. The f')llowing condttrons placed patients in the high risk category: obeeity, anticipated bed rest of more than three days, visceral malignancy, surgical dissection within the pelvis, and a history of previous thromboembolism or varicose veins. Anticoagulants were contraindicated by a history of peptic ulcer, a bleeding tendency, recent injury to central nervous system or major visceral injury, hremoptysis, previous central nervous system heemorrhage, gastrointestinal or renal carcinoma, gastrointestinal bleeding or portal hypertenslon. Warfarin was the drug used, and dosage was controHed by one-stage prothrombin time determinations. Treatment was started in the recovery room in most patients, or preoperattvely in patients already bedridden. It was continued until the patient was fully ambulant or until discharged from hospital. Patients at high risk included 336 who were treated with warfarin. The total incidence of thromboembolic episodes in this group was 1'7%, compared to a 1% incidence among low risk, untreated patients, and a 10% incidence in high risk patients for whom anticoagulants were contraindicated. Some treated patients (8'6%) sul'J'ered bleeding episodes while under treatment, compared with 1'4% of untreated patients. Six patients died, out of a total of 29 in the group that bled, but in only one was the fatal outcome thought to have been caused by the anticoagulants. In practically all patients needing transfusion for bleeding episodes during anticoagulation, the prothrombin time was found to be less than 10%.