Outcome analysis of over 1000 episodes of limb ischaemia treated by peripheral thrombolysis

Background: In the past 10 years, 13 hospitals with an interest in peripheral arterial thrombolysis have collected information on 1064 episodes of thrombolysis used to treat some 933 patients. The aim of the database was to identify clinical factors at the time of admission to hospital and relate them to the patients' outcome. Methods: After verification of the database, a regression analysis was done in three stages: univariate analysis, logistic regression including the initial success of thrombolysis, and logistic regression analysis excluding initial success. The endpoints were amputation-free survival (AFS), survival but with amputation, and death. Statistical analysis was performed with SPSS software (SPSS, Chicago, Illinois, USA). Results: The overall AFS was 72·9 per cent; 12·6 per cent of events resulted in amputation and 12·9 per cent of patients died. Univariate analysis showed that patients receiving warfarin had a survival advantage, whereas those with diabetes and ischaemia that caused sensory loss were significantly more likely to require amputation. Logistic regression analysis confirmed the above effects of diabetes and ischaemia. Major haemorrhage occurred in 8 per cent of events, but there was a significantly increased risk if the patient had a history of peripheral vascular disease (P = 0·007) or presented with graft thrombosis (P = 0·001). Stroke occurred in 2·4 per cent of events; a greater incidence occurred in patients with a duration of ischaemia of less than 8 h. Conclusion: Further detailed analysis of this large and unique database should help clinicians to refine the indications for peripheral thrombolysis in the management of acute leg ischaemia. © 2001 British Journal of Surgery Society Ltd