This research aims to study the correlation between serum immune factors and post-infectious inflammatory response syndrome (PIIRS) in immunocompetent cryptococcal meningitis (CM), and explore whether serum immune factors could be used to predict the development of PIIRS.
A cohort of 30 patients with PIIRS and 87 patients without PIIRS was selected from 347 CM patients. We analyzed the general clinical information and immunological indexes (cytokines, complement, immunoglobulin, inflammation, related cytological and biochemical indexes). Spearman correlation analysis and principal component analysis were used to explore the effects of the variables on PIIRS. Additionally, the variables were identified by a random forest-based classifier for predicting the development of PIIRS. The clinical value of predictors was verified by survival analysis.
Compared with patients without PIIRS, patients with PIIRS had lower baseline serum interleukin-6 (IL-6, P = 0.006), immunoglobulin M (IgM, P = 0.004), and a higher baseline neutrophil ratio (P <0.001). The baseline neutrophil ratio (r = 0.359, P = 0.001), IgM (r = −0.272, P = 0.025), and IL-6 (r = −0.259, P = 0.027) were significantly correlated with PIIRS. Combining principal component analysis and random forest results, neutrophil ratio, neutrophil count, IgM, IL-6, and D-dimer were useful predictors. The accuracy of random forest prediction was 75.00%, AUC, and sensitivity were 0.76 and 70%, respectively. Further survival analysis of the time from treatment to PIIRS revealed that the development of PIIRS was associated with IgM (more than 98 days of treatment) and neutrophil ratio/count.
Baseline neutrophils ratio, neutrophil count, IgM, IL-6, and D-dimer may be clinically useful predictors of PIIRS in HIV-negative immunocompetent CM patients.