To clarify the diagnostic relevance of urinary type IV collagen (IV-C) and laminin in diabetic nephropathy, the excretion of these basement membrane proteins were determined by enzyme immunoassay in 172 non-insulin-dependent diabetic patients with different grades of nephropathy and 64 non-diabetic control subjects, and were evaluated in comparison with those of urinary albumin, N-acetyl-β- d-glucosaminidase (NAG) and α 1-microglobulin (α 1MG). These excretions were also compared between a group of non-diabetic renal disease (NDRD) patients ( n = 24) and a subgroup of the diabetic patients studied ( n = 76), whose urinary albumin excretion (UAE) varied within the ranges of micro-and macroalbuminuria. Of the diabetic patients studied, 49.7%, 53.4% and 32.4% had raised urinary albumin, NAG and α 1MG excretion, respectively. In these patients, 54% and 53% exceeded the upper limit of normal for urinary IV-C and laminin. The level of IV-C and laminin excretion and the prevalence of their abnormal excretion showed a trend to increase with increasing grade of nephropathy, as assessed by UAE. In the normoalbuminuric [UAE < 20 mg/g creatinine (Cr)] stage, 28.3% and 26.3% patients had raised urinary IV-C and laminin excretion, respectively. In this stage, the excretion values for IV-C and laminin also rose significantly even when the UAE was ≤ 10 mg/g Cr ( P < 0.05 and P < 0.005, respectively). There was a close linear relationship between IV-C and laminin excretion ( r = 0.73, P < 0.0001), together with their significant relationships with albumin, NAG and α 1MG excretion. The relationship of urinary IV-C and laminin with urinary NAG and α 1MG excretion remained significant even in normoalbuminuric patients. The normoalbuminuric patients with raised NAG and/or α 1MG excretion also had a higher prevalence of raised IV-C and laminin excretion than those with normal NAG and α 1MG excretion. The excretion values for IV-C and laminin, and the excretion ratios for IV-C albumin and laminin albumin were significantly higher in diabetic patients with evidence of incipient and clinical nephropathy than in NDRD patients, though the two patient groups had a comparable level of serum Cr and UAE. We conclude that the measurement of urinary IV-C and laminin may have potential for the evaluation of diabetic nephropathy. Furthermore, their determination might be helpful for distinguishing diabetic versus non-diabetic etiologies of altered renal function in diabetic patients.