HbA1c variability and the development of nephropathy in individuals with type 1 diabetes mellitus from Rwanda

Abstract

AimTo estimate the incidence and prevalence of nephropathy and investigate whether high and fluctuating HbA1c levels were associated with development of nephropathy among T1 diabetes individuals in Rwanda. MethodsFrom 2009 to 2018, 471 T1 diabetes individuals from Rwanda were assessed for nephropathy (albumin-creatinine ratio (ACR) ≥ 30 mg/g). We calculated the mean HbA1c (HbA1c-MEAN) and two measures of HbA1c variability, i.e. A): intra-individual standard deviation (HbA1c-SD), adjusted for HbA1c assessments (HbA1c-AdjSD) and coefficient of variation (HbA1c-CV) and B): (number of HbA1c variability measures > 11 mmol/mol between two measures/number of comparisons between measurements)*100. We followed individuals from first ACR-measurement (baseline) until nephropathy, death or last ACR-measurement (end-of-follow-up), and calculated HRs for developing nephropathy using Cox-regression. ResultsThe incidence and prevalence of nephropathy were 25% and 40%, respectively. All HbA1c variability measures were associated with lower HRs of developing nephropathy, i.e. individuals with HbA1c-AdjSD levels of 0.8–1.5%, 1.5–2.1% and > 4.1% had 53% (95 %CI:0.26;0.86), 55% (95 %CI:0.25;0.82) and 53% (95 %CI:0.26;0.84) lower HRs, respectively, of nephropathy compared to individuals with HbA1c-AdjSD < 0.8%. Results did not change after adjustments, though some estimates were no longer significant. Individuals with adjusted HbA1c-MEAN 9.9–11.4% and > 11.4% had HRs of 1.7 (95 %CI: 0.87;3.32) and 1.8 (95 %CI: 0.94;3.50) compared to individuals with HbA1c-MEAN of 4.9–8.5%. ConclusionThe incidence and prevalence of nephropathy was high. Higher mean HbA1c was associated with higher HRs of developing nephropathy whereas higher HbA1c variability was associated with lower HRs of nephropathy. This indicates that higher HbA1c levels rather than fluctuating HbA1c levels is a risk factor for developing nephropathy.