A novel prognostic nomogram for elderly patients with severe fever with thrombocytopenia syndrome.

BACKGROUND: Severe fever with thrombocytopenia syndrome (SFTS) is a life-threatening infectious disease triggered by a novel bunyavirus (SFTSV). The elderly population is vulnerable to infectious diseases. This study aimed to explore the clinical features and outcomes of elderly patients with SFTS and establish an effective prognostic nomogram for them. METHODS: Data on demographics, comorbidities, clinical manifestations, laboratory parameters, and outcomes of patients with SFTS were collected. Independent predictors for in-hospital mortality identified by multivariate logistic regression were used to construct the predictive model. RESULTS: 253 patients with SFTS were retrospectively enrolled, including 142 (56.1%) elderly patients and 111 (43.9%) non-elderly patients. Compared with non-elderly patients, elderly patients had higher serum levels of laboratory parameters indicating liver, kidney, pancreas, heart, coagulation system injury, and higher viral load. The in-hospital mortality rate of elderly patients was significantly higher than that of non-elderly patients. The univariate and multivariate binary logistic regression analyses demonstrated that neurological manifestations, creatinine, CKMB, and prothrombin time were proven to be independent predictors for in-hospital mortality of elderly patients with SFTS, which were adopted as parameters of the nomogram. The nomogram showed a good calibration and discrimination, with an area under the receiver operating characteristic curve of 0.849 (95% CI 0.774-0.923). Decision curve analysis confirmed the clinical utility of the predictive model. CONCLUSIONS: Elderly patients are more susceptible to adverse outcomes, and the nomogram can help clinicians predict in-hospital mortality of elderly patients with SFTS, which may facilitate optimal therapeutic strategies to improve their prognosis.