The principal end point in the evaluation of treatment in incipient and overt diabetic nephropathy is rate of decline in glomerular filtration rate (GFR). Therefore, information on reproducibility of GFR measurements is essential in the planning and evaluation of clinical trials. We studied reproducibility of GFR measurements in insulin-dependent and non-insulin-dependent diabetes mellitus patients using, respectively, a constant-infusion technique with urine collection and labeled iothalamate as a tracer marker and a single-shot procedure using Cr-EDTA, measuring the GFR from the decline in plasma level after bolus injection. The coefficient of variance in the insulin-dependent patients was from 7.5% to 8.8% with repeated measurements. In longitudinal studies with several measurements the mean coefficient of variances varied between 7.4% and 3.4%. In the non-insulin-dependent patients the coefficient of variances between two tests were 7.0% and 5.3% for normoalbuminuric and microalbuminuric patients, respectively. In cross-sectional studies as well as in longitudinal studies, it has been consistently shown that GFR is well preserved and at a supranormal level in patients with normoalbuminuria and microalbuminuria. A decline in GFR appears to start around the transition from microalbuminuria to overt diabetic renal disease, although more detailed studies are needed to support this finding. With regard to intervention trials, several studies document that microalbuminuria can be reduced by effective antihypertensive treatment, particularly with angiotensin-converting enzyme inhibitors, also in patients with normal or close to normal blood pressure. Preliminary results from long-term studies suggest that reduction in microalbuminuria in these patients is associated with preservation of GFR and, thus, apparently renoprotection. In patients with overt renal disease, it has been consistently shown that antihypertensive treatment reduces albuminuria as well as the rate of decline in GFR. This is also observed with combined treatment regimens, for instance beta blockers or angiotensin-converting enzyme inhibitors combined with diuretics, or the three types of drugs in combination.