ALBUMIN EXCRETION AND VASCULAR DEATHS IN NIDDM

Non-insulin-dependent diabetes mellitus (NIDDM) is associated with premature mortality, generally thought to be exaggerated in patients with microalbuminuria. This prospective 8-year follow-up study aimed to determine outcome, mortality and cause of death in NIDDM patients with abnormal urinary albumin excretion compared to those with normal albumin excretion. We recruited 153 NIDDM patients with abnormal urinary albumin excretion and 153 control subjects with albumin excretion within the normal non-diabetic range, matched for age, sex and duration of diabetes, from three University hospital diabetic clinics in Newcastle upon Tyne. The outcome measures were status at follow-up, mortality and cause of death. Subjects with abnormal albumin excretion had a significantly higher 8-year mortality than matched control subjects (Odds Ratio 1.47, p = 0.02; 108 vs 66 per 1000 person years follow-up, p < 0.001). This difference was seen at all levels of abnormal albumin excretion, from just outside the normal range (10.6-29.9 mu g/min: 104 vs 61 per 1000 person years follow-up, p < 0.001) to more conventional definitions of microalbuminuria (greater than or equal to 30 mu g/min: 111 vs 71 per 1000 person years follow-up, p < 0.01). Those with abnormal albumin excretion had an excess of vascular deaths compared to matched control subjects (Odds Ratio 1.70, p = 0.009), again at different levels of albumin excretion (10.6-29.9 mu g/min p < 0.01, 30-150 mu g/min p < 0.05). On multivariate analysis, age, initial ischaemic heart disease and initial albumin excretion rates were independent predictors of death from all causes. Even a minor elevation of albumin excretion above the normal non-diabetic range is associated with excess mortality from vascular causes in NIDDM.