The optimal administration regimens of low molecular weight heparins (LMWHs) have not yet been established. The aim of this study was to compare the efficacy and safety of 2500 and 5000 XaI units of the LMWH dalteparin in patients undergoing elective general surgery for malignant and benign abdominal disease. Prophylaxis was started in the evening before surgery and given once-daily every evening thereafter. The study was designed as a prospective, randomized, double-blind, multicentre trial. Some 66.4 per cent of patients were operated on for a malignant disorder. The primary endpoint was deep vein thrombosis (DVT) detected with the fibrinogen uptake test. Bleeding complications were recorded and classified. Analysis was made both on an intention to treat basis and in patients given correct prophylaxis (86.3 per cent). A total of 2097 patients were randomized and 27 excluded after randomization. A technically correct fibrinogen uptake test was obtained in 1957 patients. The incidence of DVT was significantly lower in patients given 5000 XaI units, this being true for both correct prophylaxis (6.8 versus 13.1 per cent, P < 0.001), on an intention to treat basis (6.6 versus 12.7 per cent, P < 0.001), and in patients with malignant disease (8.5 versus 14.9 per cent, P < 0.001). Sixty-seven patients (3.2 per cent) died within 30 days with no difference between the groups. There were two cases of fatal pulmonary embolism. The frequency of bleeding complications in the whole series was higher in patients randomized to 5000 XaI units (4.7 versus 2.7 per cent, P = 0.02), although this was not the case in those operated on for malignant disease (4.6 versus 3.6 per cent, P not significant). Dalteparin in the dose of 5000 XaI units started in the evening before surgery has a good thromboprophylactic effect in high-risk general surgery at the cost of a small bleeding risk. In patients with malignant disease there was no increased risk of bleeding. The overall frequency of fatal pulmonary embolism with dalteparin is extremely low, even in this high-risk group of patients.

The less pronounced perioperative bowel ischaemia in TPEG patients indicates an advantage of the TPEG technique. Splanchnic ischaemia was not related to postoperative fever, nor the IL-6 or TNF response.

The infection rate was similar in the two groups. Thus, oral administration of ciprofloxacin is an attractive, cost-effective and safe alternative to prophylaxis in vascular patients capable of taking oral medication on the day of surgery.

No abstract