Aims: To clarify the predictive value of hematuria in patients with diabetes and nondiabetic renal disease (NDRD). Methods: The databases of Medline, Embase and the Cochrane Library were searched up to November 22, 2017. The pooled sensitivity, specificity, positive and negative likelihood ratio (PLR, NLR), diagnostic odds ratio (DOR), and area under the summary receiver operating characteristic (SROC) curve (AUC) were calculated using a bivariate mixed-effects regression model. Results: Thirty-eight articles were eligible, of which 35 articles with 4005 patients investigated hematuria. The pooled sensitivity and specificity of hematuria to predict NDRD were 0.42 (95% CI 0.35-0.49) and 0.72 (95% CI 0.64-0.79), respectively. The pooled PLR and NLR were 1.49 (95% CI 1.28-1.75) and 0.81 (95% CI 0.75-0.87), respectively. The DOR was 1.85 (95% CI 1.49-2.30). The pooled AUC was 0.59 (95% CI 0.54-0.63). For dysmorphic erythrocytes, the pooled sensitivity was 0.27 (95% CI 0.23-0.32), while the specificity was 0.94 (95% CI 0.91-0.97). There was heterogeneity among studies (p < 0.001), and no publication bias was identified. Conclusions: Type 2 diabetes patients presenting with hematuria are slightly more likely to develop NDRD. Dysmorphic erythrocytes may be more useful than microhematuria in diagnosing for NDRD in type 2 diabetes with proteinuria. (C) 2018 Elsevier B. V. All rights reserved.