The study examined risk factors for lung injury in 71 untreated patients with anti-neutrophil cytoplasmic antibody (AAV)-associated vasculitis, of whom 38 had lung involvement. Using logistic regression and the LASSO method, they identified four independent predictors: intermittent pattern of symptoms (OR = 8.52, 95% CI: 2.17–33.21), paresis (OR = 95% CI: 1.98–75.50), older age at onset (OR = 1.05 per year, 95% CI: 1.01–1.10) and lower hemoglobin level (OR = 0.97 per g/L, 95% CI: 0.95–0.99). They created a nomogram for the prediction of pulmonary involvement that achieved an area under the curve (AUC) of 0.88 (95% CI: 0.77–0.98) in the training set and 0.83 (95% CI: 0.64–1.00) in the validation set (distribution 7:3). The model was confirmed as clinically useful by calibration graphs and decision curve analysis. The study highlights the need for external validation in prospective multicenter cohorts for personalized risk monitoring.