Impact of Diabetic Lesions on Pathology, Treatment, and Outcomes of Glomerular Diseases

KEY POINTS: People with glomerular disease (GD) and comorbid diabetes have similar baseline characteristics irrespective of superimposed diabetic lesions. Immunosuppression for GD with comorbid diabetes is the same regardless of superimposed diabetic glomerular lesions. ESKD or death is more rapid in GD and comorbid diabetes only in the presence of moderate-severe diabetic glomerular lesions. BACKGROUND: We aimed to evaluate whether concomitant diabetic glomerulosclerosis (DGS) and its severity affect the treatment and outcomes of primary glomerular diseases (GDs) with comorbid diabetes. METHODS: We conducted a retrospective review of people with diabetes and GD. We searched the GD Collaborative Network for biopsies from 2008 to 2015 among persons with diabetes and any of the following diagnoses: FSGS, IgA nephropathy, minimal change disease, membranous nephropathy, or antineutrophil cytoplasmic autoantibody GN. Data were abstracted from health records and histologic diabetic glomerular class scored. The primary composite end point was ESKD or death. Multivariable Cox regression models tested whether any or the severity of diabetes histopathology affected the primary end point. RESULTS: Data from 134 cases were available for analysis (78 DGS+GD and 56 GD alone). Diabetes duration and glycemic control were similar between the two groups (P = 0.2; P = 0.09, respectively). Use of immunosuppression did not differ between the two groups (P = 0.3). The composite end point was significantly higher in DGS+GD (22.5 cases per 100 person-years [95% confidence interval (CI), 16.6 to 30.5]) versus GD alone (10.2 cases per 100 person-years [95% CI, 6.4 to 16.2]). Regression analyses demonstrated that compared with the GD-alone group, the risk for the composite end point was similar in the group with mild DGS+GD (DGS class 1, 2a) (hazard ratio, 1.15 [95% CI, 0.54 to 2.43]) while the group with severe DGS+GD (DGS class 2b, 3, 4) had a greater risk (hazard ratio, 3.60 [1.79 to 7.22]). CONCLUSIONS: Among people with diabetes and GD, mild diabetic glomerular lesions on biopsy do not affect outcomes, but moderate-severe lesions increase the risk for ESKD and death. Whether use of immunosuppression, particularly glucocorticoids, is less successful in inducing GD remission in people with moderate-severe diabetic lesions will be a focus of future study in a larger population.