| Background and objective:Renal impairments in diabetic patients include diabetic nephropathy (DN) and non-diabetic renal diseases (NDRD). The correct diagnosis is important for the treatment and prognosis of these diseases. Renal biopsy is a major basis for diagnosis. But due to the low renal biopsy rate in these patients, a novel effective non-invasive clinical identification method is needed. In this study, based on the renal pathology as definite clinical diagnosis evidence, the difference of clinical features between the DN and NDRD was clearly set up. The practicability of the two pre-established clinical diagnosis models, namely NDT model and JDB model, was validated, and the differential significance of hematuria, an important clinical parameter, was further evaluated.Methods:Among all the patients with type 2 diabetes in the renal biopsy in Chinese PLA General Hospital from May 2012 to December 2014, (1) 248 patients who met the criteria were screened and were divided into DN group and NDRD group on the basis of renal pathology. A retrospective analysis of differences of the demographic data and the clinical features between the two groups was performed; the distribution of pathological types of NDRD group was established explicitly; the differential efficiency of the following three clinical parameters of NDRD was evaluated by diagnostic tests:history of diabetes is equal to or less than 5years, no diabetic retinopathy, and active urinary sediment change. (2) According to the inclusion and exclusion criteria for both NDT and JDB models,255 patients were selected and their clinical pathological features was compared and analyzed with those patients from 1993-2003 (from NDT model data sources) and those from 2004-April 2012 (from JDB model data sources). Disease evolution was clearly defined; the merits and defects of both models were evaluated, and the subgroups were analyzed; at the same time, indications were extended for generalization ability evaluation of the models. (3) 300 patients who met the inclusion criteria were enrolled, and the differential effects of hematuria under various standards and glomerular hematuriain DN and NDRD were compared. Factors affecting hematuria in patients with DN were screened using single factor and multiple factor Logistic regression analysis.Results:(1) 96 (38.71%) DN cases and 152 (61.29%) NDRD cases were diagnosed among 248 patients with renal biopsy in recent 3 years in our center. The pathological type of NDRD patients was mainly membranous nephropathy (36.18%), followed by IgA nephropathy (32.23%). Compared with DN group, NDRD group showed a shorter median time of diabetes history (36 months versus 156 months, P<0.01), a lower rate of diabetic retinopathy (7.9% versus 82.3%, P<0.01), and a lower median value of 24 hour urinary protein quantification (2.29 versus 3.86, P=0.01), with no difference in the incidence rate of nephrotic syndrome between the two groups (42.1% versus 39.6%, P=0.69). In NDRD group, the level of glycosylated hemoglobin is under better control (6.57±1.14 versus 7.02±1.78, P=0.03) and the incidence rate of active urinary sediment is higher (27.6% versus 6.3%, P=0.01). Among three clinical parameters predicting NDRD, the specificity of diabetic retinopathy (92.11%), Youden index (0.74), and the accuracy (88.31%) are the highest. (2) The incidence rate of hematuria in DN group in three study periods showed ascendant trend over time (16.7%,32.3% and 37.7% respectively), and the difference from that in NDRD group was eventually disappeared; the incidence rates of hypertension and nephrotic syndrome showed upward trends. JDB model showed higher diagnosis accuracy (89.80%), while the diagnostic specificity (98.66%) of NDT model is higher, but no difference in the overall diagnostic performance was found. There was no difference in the diagnostic performance in subgroups with different history of diabetes, blood pressure, and with/without nephrotic syndrome. With the extended indications, the diagnostic performance of JDB model is more robust. (3) All the standards of hematuria and nephrogenic hematuria, except urinary red cell accounts, which are> 2 per high-power field (/hpf), showing no statistical difference between DN and NDRD groups (P=0.09), were different between the two groups. Glomerular hematuria showed the highest differential performance for both groups (AUC=0.608), but no difference with the different standards of hematuria was found. Nephrotic syndrome is a risk factor for hematuria in patients with DN (OR=2.411, P<0.05).Conclusion:There are significant differences in the clinical features of DN and NDRD, and the pathological distribution of NDRD changes over time. Methods for the clinical differential diagnosis of both diseases, such as differential diagnosis model and specific clinical parameters, are effective means to distinguish between DN and NDRD, and the high selectivity for NDRD patients is a beneficial supplement for the indications of renal biopsy. |
PDF Full Text Request