Association between age and survival following major amputation. The Scottish Vascular Audit Group

To determine whether age is associated with survival following major amputation and whether this association is independent or simply reflects selection bias in amputation level.

Computer linkage of routine discharge and death data on the 2759 patients undergoing major amputation in Scotland between 1989 and 1993 for peripheral arterial disease.

Cox's proportional hazards model and multivariate logistic regression analysis using death as the outcome variable and age, sex, urgency, amputation level and recent arterial reconstructive surgery as predictor variables.

Proximal amputation was more common in older patients. Survival was associated with both age (p < 0.001) and amputation level (p < 0.001). Age was an independent predictor of death at 30 days (p < 0.0001), 6 months (p < 0.001), 12 months (p < 0.0001) and 2 years (p < 0.0001) postoperation.

Survival following amputation was poor, with only half the patients alive at 2 years. Above-knee amputation was associated with poorer survival, presumably due to the presence of more severe and widespread disease, and was undertaken more commonly in older patients. However, age remained a predictor of survival after adjustment for amputation level. Higher early mortality suggest that a worse prognosis in elderly patients cannot be attributed wholly to actuarial considerations.