Severity of Coexisting Diabetic Glomerular Lesions Affects Outcomes in Primary Glomerular Disease

Abstract

The prevalence of diabetes (including type 1 and type 2 diabetes) has been rising in the United States and the rest of the world with some variations attributed to racioethnic and socioeconomic factors. 1,2 Diabetic kidney disease is a frequent complication of diabetes, typically manifesting as the clinical syndrome of diabetic nephropathy (DNP). Diabetes is a major comorbid condition in and DNP is a leading cause of ESKD in the United States. 3 The pathophysiology of DNP has been extensively investigated, and the spectrum of histopathologic findings seen in DNP has been well-characterized. The histopathology of diabetic kidney disease is manifested in glomerular, tubulointerstitial, and vascular compartments. 4 A consensus pathologic classification system of DNP based on extent of diabetic glomerular sclerosis (DGS) has also been extensively adopted in clinical practice. 5 Patients with DGS can also develop de novo primary glomerular disease (GD). While different single-center/single health system studies have analyzed the prevalence of primary GD in diabetic patients, 6,7 there continue to be significant knowledge gaps about the effect of diabetes on the pathophysiology and outcomes of primary GD. In this issue of Kidney360, Kim et al. detail the conclusions of a retrospective review of the clinicopathologic findings and longitudinal outcomes in a cohort of patients seen within the GD Collaborative Network between the years 2008 and 2015 who had diabetes along with a primary GD (limited to the following subgroups: FSGS, IgA nephropathy, minimal change disease, membranous nephropathy, and ANCA GN). 8 After applying their inclusion and exclusion criteria on this cohort, the authors analyzed the clinicopathologic characteristics and outcomes in 134 patients with diabetes and primary GD. The presence and extent of DGS was determined by a pathologist after light microscopy and electron microscopy review except for six cases where slides were unavailable; all other pathology data points and clinical information were extracted from paper and electronic health records including nephropathology reports. They compared their findings in patients with biopsy-proven DGS of varying severities and primary GD (n=78) with patients with primary GD alone (n=56) within this cohort. With the obvious limitations of a retrospective single-center study and a small sample size, their data and analysis yield interesting and provocative conclusions that merit further investigation. In their study, the sociodemographic and the baseline clinical parameters in the two groups were similar except that the proportion of patients treated with insulin, systolic BP, and urine protein creatinine ratio was higher, and the serum albumin level was lower in the DGS+GD group compared with the patients with primary GD alone. While there were some numerical differences in the immunosuppression protocols used in the two groups, they are reported as being statistically similar. Some of the conclusions of the study make intuitive sense. For example, the severity of global glomerulosclerosis, vascular, and tubulointerstitial lesions was worse in the DGS+GD group likely indicative of the additional metabolic injury to the kidney in this group. There were no statistically significant differences in the morphologic patterns of the lesions of the primary GD in the two groups. The data of long-term outcomes in this cohort are interesting. The DGS+GD group showed more rapid progression to ESKD or death. When outcomes were stratified based on the severity of DGS, patients with mild DGS (defined as Renal Pathology Society class 1 or 2a) and primary GD showed similar outcomes as the primary GD group alone. On the other hand, patients with severe DGS (defined as Renal Pathology Society class 2b, 3, or 4) had significantly worse prognosis. These data suggest that the severity of DGS is important in the long-term prognosis of patients with DGS+GD and could affect their clinical management. It will be interesting to see whether this trend is replicated in other patient cohorts and larger datasets. Other important questions are raised by this study. A statistically nonsignificant numeric trend suggestive of longer chronicity indicated by segmental glomerular sclerosis and fibrous crescents was noted in patients with IgA nephropathy+DGS and ANCA GN+DGS, respectively. These data suggest that patients with DGS+GD are biopsied later in their disease course. Additional studies in larger patient cohorts are needed to determine whether the clinical parameters that influence the decision to biopsy diabetic patients with suspected primary GD need to be better defined. 9 Their data also suggest that the small sample size may have affected conclusions regarding the effect of GD subgroups and demographic factors on the outcomes in this cohort. In addition, the numerical trends in this study are slightly different and in some cases contradictory to data from other single-center studies investigating the effect of diabetes on GD. For example, a separate single-center study found that diabetes was associated with kidney failure and cardiovascular adverse events, but not mortality in patients with GN. 7 These interesting findings underscore the need for larger, prospective multicenter studies to investigate the effect of diabetes and coexisting DNP on clinical outcomes in primary GD.