Identifying Parameters to Distinguish Non-Diabetic Renal Diseases from Diabetic Nephropathy in Patients with Type 2 Diabetes Mellitus: A Meta-Analysis

BACKGROUND: Renal injuries in patients with diabetes include diabetic nephropathy (DN) and non-diabetic renal diseases (NDRD). The value of a clinical diagnosis of DN and NDRD remains inconclusive. We conducted a meta-analysis of the literature to identify predictive factors of NDRD and to compare the clinical characteristics of DN and NDRD for differential diagnosis. METHODS: We searched PubMed (1990 to January 2012), Embase (1990 to February 2009), and CNKI (1990 to January 2012) to identify studies that enrolled patients with DN and NDRD. Then, the quality of the studies was assessed, and data were extracted. The results were summarized as odds ratios (ORs) for dichotomous outcomes and weighted mean differences (WMDs) for continuous outcomes. RESULTS: Twenty-six relevant studies with 2,322 patients were included. The meta-analysis showed that the absence of diabetic retinopathy (DR) predicts NDRD (OR, 0.15; 95% confidence interval [CI], 0.09–0.26, p<0.00001). A shorter duration of diabetes mellitus (DM) also predicted NDRD (weighted mean difference, −34.67; 95% CI, −45.23–−24.11, p<0.00001). The levels of glycosylated hemoglobin (HbA1C%), blood pressure (BP), and total cholesterol were lower in patients with NDRD, whereas triglycerides and body mass index were higher. Other clinical parameters, including age, 24-h urinary protein excretion, serum creatinine, creatinine clearance, blood urea nitrogen, and glomerular filtration rate were not different between patients with NDRD and DN. CONCLUSIONS: We identified that the absence of DR, shorter duration of DM, lower HbA1C, and lower BP may help to distinguish NDRD from DN in patients with diabetes. This could assist clinicians in making a safe and sound diagnosis and lead to more effective treatments.