Development and assessment of a predictive nomogram for the progression of IgA nephropathy

Abstract

The present study is to establish a nomogram for predicting the prognosis of IgA nephropathy (IgAN). Of the 869 IgAN patients, four-fifths were randomly assigned to the development cohort and one-fifth to the validation cohort. The primary outcome was a composite event of either a ≥ 50% reduction in estimated glomerular filtration rate (eGFR), end-stage renal disease or death. The mean follow-up time was 44 months. The Cox regression model identified urinary protein excretion (1-3.5 g/d, HR 11.639, 95% CI 3.601-37.625; ≥ 3.5 g/d, HR 32.435, 95% CI 10.079-104.380), eGFR (G2, HR 5.293, 95% CI 2.011-13.932; G3, HR 15.797, 95% CI 6.584-37.905; G4, HR 34.619, 95% CI 13.887-86.301; G5, HR 217.651, 95% CI 83.807-565.248), hyperuricaemia (HR 7.031, 95% CI 4.126-11.980), mesangial proliferation (HR 36.667, 95% CI 5.098-263.711), segmental glomerulosclerosis (HR 5.122, 95% CI 3.114-8.425), tubular atrophy/interstitial fibrosis (T1, HR 33.351, 95% CI 7.831-142.044; T2, HR 213.888, 95% CI 51.048-896.182), crescents (C1, HR 3.123, 95% CI 1.771-5.510; C2, HR 7.353, 95% CI 3.590-15.062) and glomerulosclerosis (25-49%, HR 3.123, 95% CI 1.771-5.510; ≥ 50%, HR 14.384, 95% CI 8.813-23.479) for developing the nomogram. The C-index was 0.945 (95% CI 0.914-0.976) in both the development and validation cohorts, showing good agreement between the nomogram-predicted probability and actual free-of-progression probability. Thus, our nomogram could accurately predict the progression of IgAN patients.