How often is NIDDM complicated with non-diabetic renal disease? An analysis of renal biopsies and the literature.

According to extensive autopsy studies, non-diabetic renal disease seems to be rare in diabetes mellitus, but recent publications suggest a significant prevalence of non-diabetic renal disease in non-insulin-dependent diabetic (NIDDM) patients, especially in the absence of retinopathy. The purpose of this study was to evaluate the prevalence of non-diabetic renal disease in NIDDM patients in renal biopsies from clinical practice, in patients suspected of having non-diabetic renal disease. In addition we systematically reviewed the literature. Biopsies were evaluated at the University Department of Pathology, Aarhus, Denmark, but had been collected at several departments of nephrology. In total 33 consecutive biopsies were available from 1988-1995 (mean age of patients: 62 years (range 39-75) (mean known diabetes duration 8 years (range 1-25); the main clinical reason for a biopsy was proteinuria. Renal function changes ranged from slight elevation of serum creatinine to uraemia. In addition 9 original papers, including our own material 580 patients were examined. On the basis of careful morphological evaluation according to international criteria, no patient exhibited an unequivocal sign of non-diabetic glomerular disease. Two patients had strongly but not completely convincing evidence of glomerulonephritis. One patient had some evidence of glomerulonephritis. These 3 patients also exhibited diabetic lesions. One patient with end-stage renal disease showed evidence of interstitial nephropathy without glomerular lesions. Thus, in 4 patients evidence of non-diabetic lesions was found. In the remaining 29 patients typical diffuse (n = 9) or nodular (n = 20) diabetic lesions were found. Twenty patients showed evidence of diabetic retinopathy. One of the patients with evidence of non-diabetic renal disease had simplex retinopathy. In the literature a considerable bias exists towards including patients with non-diabetic renal disease. In non-biased materials with proteinuria the prevalence of non-diabetic renal disease is very similar to our series. In microalbuminuric patients non-diabetic renal disease seems to be very rare. It can be concluded that in our material non-diabetic renal disease is uncommon in NIDDM patients, even if a clinician has suggested renal disease of other origin. A considerable bias towards including non-diabetic renal disease in NIDDM patients exists in the literature. The indication for biopsy should be evaluated carefully, and biopsy should by no means be routinely performed in NIDDM patients with proteinuria.