Researchers from China designed a novel model to predict long-term survival in patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV).
This model was specifically created for AAV, which is an advantage over the widely adopted revised version of the Five Factor Score (rFFS)—a system proposed by the French Vasculitis Study Group to evaluate disease severity and predict outcomes in patients with AAV and polyarteritis nodosa.
The model performed well in the study cohort as demonstrated by the optimism-corrected C-index, the integrated Brier score, and the calibration plots. Moreover, decision curve analysis suggested that the novel prediction model had higher net benefits compared with the rFFS and the Birmingham Vasculitis Activity Score system.
The analysis of risk groups (low, moderate, and high risk) demonstrated that neither rFFS nor Birmingham Vasculitis Activity Score were able to accurately predict survival. Conversely, the novel prediction model identified significant differences between the 3 risk groups.
Read more about AAV prognosis
The apparent C-index of the prediction model was 0.753 and the integrated Brier score was 0.112. After bias correction, the optimism-corrected C-index and the integrated Brier score were 0.728 and 0.109, respectively.
The novel prediction model integrates clinical organ involvement patterns, demographic data, and laboratory test results.
Initially, the researchers selected 20 parameters as candidate predictors and used the Lasso regression method to shorten the list for the final prediction model. The 6 selected predictors were age at clinical admission, chest and cardiovascular involvement, AAV subtypes, serum creatinine, and hemoglobin.
The study enrolled 653 patients with AAV. Of those, 303 had microscopic polyangiitis, 245 had granulomatosis with polyangiitis, and 105 patients had eosinophilic granulomatosis with polyangiitis. The median follow-up was 33
months.
Reference
Chen Z, Tian X, Qu J, Chen J, Yang Y, Li J. Development and internal validation of a model to predict long-term survival of ANCA associated vasculitis. Rheumatol Immunol Res. Published online April 18, 2023. doi:10.2478/rir-2023-0005
