POS-344 ASSOCIATION BETWEEN RELATIVE ESTIMATED GLOMERULAR FILTRATION RATE CHANGES AND CLINICAL OUTCOMES IN PATIENTS WITH TYPE 2 DIABETES

Abstract

Research supporting the validity of changes in estimated glomerular filtration rate (eGFR) as a surrogate endpoint for end-stage kidney disease (ESKD) or other clinical outcomes in patients with type 2 diabetes mellitus (T2DM) is limited. The primary purpose of this study was to evaluate the association between relative changes in eGFR and clinical outcomes in patients with T2DM. This was a retrospective cohort study using administrative claims data. Patients 65-89 years of age with T2DM, without initial stage V chronic kidney disease (CKD) or ESKD, enrolled in a Medicare Advantage and Prescription Drug Plan, with an index eGFR 25-89 ml/min/1.73m2 in 2008-2017 were identified. Patients with one serum creatinine test result available for calculation of eGFR and a second serum creatinine test result within 3-24 months of the first value were included. The primary exposure of interest was relative decline in eGFR of ≥40% in up to a 2 year period (landmark time). Outcomes included ESKD/kidney failure, all-cause mortality, and a composite cardiovascular (CV) event measure (non-fatal myocardial infarction, stroke, hospitalization for heart failure, or death in-hospital or within 30 days of discharge for a CV-related admission). The associations between relative eGFR change and the clinical outcomes of interest were evaluated with multivariable adjusted survival (Cox) models after a landmark time of 2 years. Age, race, region, low-income subsidy/dual eligibility status, first eGFR, comorbidities, laboratory values (e.g., hemoglobin A1c, cholesterol, urine albumin-to-creatinine ratio), medications, healthcare resource use and costs were included as variables in the models. A total of 288,170 patients were included; 90,990 patients with a first eGFR of 25-59 ml/min/1.73m2 and 197,180 with a first eGFR of 60-89 ml/min/1.73m2. The median age was 72 years, 53.1% were female, and 76.9% were white. A majority (73.9%) of patients were living in the Southern United States, with 67.9% living in urban locations, and 11.2% and 8.1% were low-income subsidy and dual eligible, respectively. Overall, 1.6% (n=4,581) of patients had a relative eGFR decline of ≥40% in the landmark time. The overall adjusted hazard ratio (HR) for ESKD/kidney failure was 4.38 (95% CI: 3.99, 4.81) in patients with a ≥40% decline as compared to those with a <40% decline. The adjusted HR for all-cause mortality was 1.98 (95% confidence interval [CI]: 1.87, 2.10) and the composite CV outcome was 1.67 (95% CI: 1.53, 1.82). The association between a ≥40% relative eGFR decline and ESKD/kidney failure was generally similar in the subgroup of patients with an initial eGFR of 60-89 ml/min/1.73m2 (HR: 4.37 [95% CI: 3.79, 5.04]) and in the subgroup with an initial eGFR of 25-59 ml/min/m2 (HR: 4.60 [95% CI: 4.06, 5.20]). Subgroup analyses by initial eGFR were consistent with the overall analysis for the CV outcome and all-cause mortality. The results for the overall and subgroup analyses are presented in Figure 1. In an older patient population with eGFR 25-89 ml/min/m2 and T2DM, a relative eGFR change of ≥40% was associated with ESKD/kidney failure, as well as with all-cause mortality and CV events, providing further support to the validity of this measure as a proxy for “hard” clinical outcomes.