Research on multi-marker risk prediction model for RA-ILD based on the real world

Objective To explore the laboratory biomarkers which are of value for early diagnosis of rheumatoid arthritis associated interstitial lung disease（RA-ILD）, and to establish a risk prediction model for RA-ILD.Methods 80 patients with RA-ILD diagnosed in West China Hospital of Sichuan University from January 2014 to October 2021 were selected as disease group (RA-ILD group), 89 patients with RA as disease control group (RA group) and 100 healthy subjects as healthy control group (HC group). The basic clinical information and laboratory examination results of the patients were collected, and the differences of various indexes among the three groups were analyzed. The RA-ILD risk prediction model was constructed by Logistic regression, and the effect of the prediction model was evaluated by the subject working characteristic curve (ROC).Results Compared with HC group, the ratio of neutrophils to lymphocytes, the absolute value of monocytes and the level of lactate dehydrogenase (LDH) in both RA-ILD group and RA group were significantly higher, and the levels of these three indexes in RA-ILD group were significantly higher than those in RA group. The prediction model established by Logistic regression was logit (p) =-4.280+0.013*LDH+1.561*MON+0.084*NLR. The area under the working curve of the subjects (AUC) was 0.780, and the sensitivity and specificity were 85.5% and 68.0%, respectively.Conclusions The absolute value of monocytes, the ratio of neutrophils to lymphocytes and lactate dehydrogenase are important non-invasive biomarkers for the early diagnosis of RA-ILD. The model established based on these biomarkers is effective in predicting RA-ILD.