The need for early predictors of diabetic nephropathy risk: is albumin excretion rate sufficient?
Initial studies showing an approximately 80% rate of progression from microalbuminuria (MA) to proteinuria in type 1 diabetic patients led to the broad acceptance of MA as a useful clinical predictor of increased diabetic nephropathy (DN) risk. Some MA patients, however, have quite advanced renal structural changes, and MA may, in these cases, be a marker rather than a predictor of DN. More recent studies have observed only about a 30-45% risk of progression of MA to proteinuria over 10 years, while about 30% of type 1 diabetic patients with MA became normoalbuminuric and the rest remained microalbuminuric. The finding that some MA patients have only mild diabetic renal lesions is consistent with the lower than originally estimated risk of progression from MA to proteinuria and with the notion that some MA patients revert to normoalbuminuria. To increase the complexity of the scenario, some normoalbuminuric long-standing type 1 diabetic patients have well-established DN lesions and approximately 40% of all patients destined to progress to proteinuria are normoalbuminuric at initial screening, despite many years of diabetes. A similar picture is emerging in type 2 diabetic patients, although fewer studies have been conducted. Thus, the predictive precision for MA to progress to overt nephropathy over the subsequent decade or so is considerably less than originally described. It is unclear whether this is due to changes in the natural history of DN resulting from improved glycemia and blood pressure control, or whether there were overestimates of risk in the original studies due to the small sample sizes, post hoc analyses, and variable MA definitions. Albumin excretion rate (AER) remains the best available noninvasive predictor of DN risk and should be regularly measured according to established guidelines. However, AER may be unable to define patients who are safe from or at risk of DN with an accuracy that is adequate for optimal clinical decision making or for the design of certain clinical trials. Investigations into new risk markers or into the combined use of several currently available predictive parameters are needed.