The survival nomograms for anti-neutrophil cytoplasmic antibody-associated glomerulonephritis patients with a follow-up of more than one year

Abstract

Objective Anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis with kidney injury, manifested as ANCA-associated glomerulonephritis (AAGN), often portends a poor prognosis of renal function and life survival in long term. Methods A cohort of 339 AAGN patients were enrolled retrospectively. These patients survived and were followed up for at least 12 months after diagnosis in our centre. Multivariate Cox regression analysis and nomogram models were performed to determine the risk factors associated with renal survival and patient survival. Results The median follow-up time of all 339 patients was 65.2 (IQR 45.1, 91.3) months and the median age was 61(IQR 53, 69) years. In order to analyse the impact of the factors on renal survival, we divided the patients into 2 groups: non-dialysis group (204 patients without dialysis at the final visit) and dialysis group (135 patients with maintaining dialysis). The patients in dialysis group had lower haemoglobin level, lower eGFR level, lower platelets count, more daily urine protein, and higher Birmingham Vasculitis Activity Score (BVAS) at admission than those in non-dialysis group. Multivariate Cox regression revealed that low haemoglobin (HR=0.977, 95%CI 0.965-0.990, p<0.001), low eGFR (HR=0.957, 95%CI 0.941-0.973, p<0.001) and high proteinuria (HR=1.139, 95%CI 1.055-1.230, p=0.001) at admission were independent risk factors for developing maintaining dialysis. A nomogram was established based on the results of multivariate Cox analysis and the internal bootstrap resampling approach showed the C-index of the nomogram was 0.83. Then we divided all patients into death group (n=99) and survival group (n=240). The patients in death group had older age, more hypertension, more chronic lung disease, lower platelets count, lower serum albumin, higher BVAS and lower eGFR at admission than those in survival group. Multivariate Cox regression revealed that the status of maintaining dialysis (HR 3.51, 95% CI 1.91-6.47, p<0.001) and old age (HR 1.07, 95% CI 1.04-1.09, p<0.001) were independent risk factors for all-cause mortality. Again, a nomogram was established and the C-index was 0.74. Conclusion We analysed the independent risk factors for maintaining dialysis and all-cause mortality in AAGN patients with a follow-up of more than 12 months. The two proposed nomograms were of predictive value.