Age- and time-dependent increases in incident anti-glomerular basement membrane disease: a nationwide cohort study

Abstract

ABSTRACT Background Epidemiologic assessments of anti-glomerular basement membrane (GBM) disease have been challenging due to its rare occurrence. We examined changes in the incidence and outcomes from 1998 to 2018 using nationwide healthcare registries. Methods All patients with incident anti-GBM disease were identified using the International Classification of Diseases, 10th Revision code DM31.0A. Controls were matched 4:1 on birthyear and sex using exposure density sampling. Log link regression adjusted for time, age and sex was applied to model survival. Results We identified 97 patients with incident anti-GBM disease, corresponding to an incidence of 0.91 cases/million/year [standard deviation (SD) 0.6]. The incidence increased over time [1998–2004: 0.50 (SD 0.2), 2005–2011: 0.80 (SD 0.4), 2012–2018: 1.4 (SD 0.5); P = .02] and with age [0.76 (SD 0.4), 1.5 (SD 1.04) and 4.9 (SD 2.6) for patients <45, 45–75 and >75 years]. The median age was 56 years (interquartile range 46) and 51.6% were female. Dialysis was required in 58.4%, 61.9% and 62.9% of patients at day 30, 180 and 360, respectively. The 1-year kidney survival probability was 0.38 (SD 0.05) and exhibited time-dependent changes [1998–2004: 0.47 (SD 0.13), 2005–2011: 0.16 (SD 0.07), 2012–2018: 0.46 (SD 0.07); P = .035]. The 5-year mortality was 26.8% and mortality remained stable over time (P = .228). The risk of death was greater than that of the matched background population {absolute risk ratio [ARR] 5.27 [confidence interval (CI) 2.45–11.3], P < .001}, however, it was comparable to that of patients with anti-neutrophil cytoplasmic antibody–associated vasculitis (AAV) requiring renal dialysis at presentation [ARR 0.82 (CI 0.48–1.41), P = .50]. Conclusion The incidence of anti-GBM disease increased over time, possibly related to temporal demographic changes. Mortality remained high and was comparable with an age- and sex-matched cohort of dialysis-dependent AAV patients.