Object: Although measurement of albuminuria which reflects glomerular damage is usually recommended, non-albumin proteinuria (NAP), which reflects both tubular and glomerular damage, is often neglected. In this study, we investigated the clinical significance of NAP in type 2 diabetes mellitus (T2DM) patients. Methods: We enrolled a total of 764 T2DM patients who measured both urine albumin to creatinine ratio (uACR) and protein to creatinine ratio (uPCR). We defined albuminuria as uACR ≥ 30 mg/g and isolated NAP (i-NAP) as uPCR ≥ 150 mg/g and uACR <30mg/g, respectively. The associations between albuminuria, NAP and several indices of glucose metabolism were investigated. Results: We classified the subjects into i-NAP (110 (14.4%)), albuminuria (205 (26.8%)) and non-proteinuria (449 (58.8%)) groups. The glucose parameters were similar between i-NAP and albuminuria groups. The i-NAP group showed lower body mass index (BMI) and lower homeostatic model assessment-insulin resistance (HOMA-IR) values. Albuminuria group had the highest values of HOMA-IR. The i-NAP group was independently associated with females (OR 1.62 (95% CI, 1.08-2.44)), BMI (OR 0.92 (95% CI, 0.87-0.98)), systolic blood pressure (OR 0.98 (95% CI, 0.96-0.99)) and low HOMA-beta (OR 1.68 (95% CI, 1.11-2.53)). The 'low' HOMA-beta value was defined based on the median value of the data. Conclusion: Compared with the albuminuria group, the isolated NAP group showed lower insulin resistance with similar glucose parameters. We postulated that the characteristics of the i-NAP group might be related to the low HOMA-beta value. We suggest that i-NAP might be associated with a decreased beta cell function in T2DM patients.