COX Regression Analysis and Mortality Risk Prediction Model of 85 Adult Patients with Secondary Hemophagocytic Lymphohistiocytosis

Aims/Background Secondary hemophagocytic lymphohistiocytosis (sHLH) is a rare, rapidly progressive and highly lethal disease. This retrospective cohort study aims to analyze the factors influencing the mortality risk in adult patients with sHLH, which are instrumental to improving our understanding of the high mortality risks associated with sHLH. Methods This study included 85 patients diagnosed with sHLH who were admitted and treated in the Department of Emergency, Peking University People's Hospital between April 2015 and July 2023. Participants were classified based on prognosis into two groups: the death group and the survival group. We collected demographic data, routine blood tests, comprehensive biochemical profiles, coagulation analyses, serum ferritin levels, natural killer (NK) cell counts, soluble interleukin-2 receptor (sCD25) levels, and potential etiological factors upon admission. The mortality risk factors influencing the prognosis of sHLH were analyzed with univariate and multivariate COX regression. Additionally, a mortality risk prediction model was established, and its accuracy was validated and optimized using the concordance index (C-index), time-dependent receiver operating characteristic (ROC) curve, calibration curves and clinical decision curve analysis (DCA). Results A total of 85 patients were included in this study, the male-to-female ratio is 1:1.4. The median age at diagnosis of sHLH was 56.00 (33.00-69.00) years. Clinical symptoms were atypical, with fever being the most prevalent symptom (81 cases, 95.3%), followed by disturbance of consciousness (10 cases, 11.8%). Univariate COX analysis and Multivariate COX regression analysis revealed that age (hazard ratio (HR) [95% confidence interval (CI)], 1.098 [1.025-1.177], p = 0.008), Alanine transaminase (ALT) (HR [95% CI], 1.016 [1.001-1.031], p = 0.034), Aspartate transaminase (AST) (HR [95% CI], 1.005 [1.001-1.008], p = 0.004), and Troponin I (TNI) levels (HR [95% CI], 1.196 [1.011-1.414], p = 0.037) were independent risk factors affecting prognosis. Specifically, sHLH patients aged >= 63.5 years (sensitivity 82.8%, specificity 85.7%), with AST levels >= 111 U/L (sensitivity 82.8%, specificity 82.1%), ALT >= 41 U/L (sensitivity 58.6%, specificity 64.3%) and TNI levels >= 2.15 ng/mL (sensitivity 62.1%, specificity 100%), faced a higher risk of mortality. We established a mortality risk prediction model for sHLH patients, which yielded a C-index of 0.848 (0.773-0.901), indicating strong agreement between predicted and observed outcomes. The ROC curves of the 28-day, 60-day, and 90-day mortality risk prediction model for sHLH patients were drawn, and the results showed that the 28-day, 60-day, and 90-day area under the curve (AUC) were 0.900 (0.829-0.971), 0.940 (0.882-0.998), and 0.930 (0.874-0.986), respectively. The predictive effect of the prediction model is satisfactory. Additionally, the clinical decision curve analysis for 28, 60 and 90 days in sHLH patients indicated that the net benefit of the nomogram model was higher than that line of extremes models (treat all and treat none). Conclusion Patients with sHLH have frequently atypical clinical presentation, with early death risk and notably elevated mortality rate. Independent risk factors influencing mortality risk in sHLH patients include age >= 63.5 years, AST >= 111 U/L, ALT >= 41 U/L, and TNI >= 2.15 ng/mL. With high accuracy and efficacy, the risk prediction model constructed can facilitate timely identification of sHLH patients at elevated risk of mortality, which is critical for optimizing clinical interventions.