[Clinical features and risk factors for secondary hemophagocytic lymphohistiocytosis in elderly patients with severe SARS-CoV-2 infection: a multicenter retrospective cohort study]
Yunxin Deng,Shasha Lu,Guofang Zhang,Wenqing Sun,Yufeng Chu,Mei Meng,Yunliang Cui,Pibao Li
DOI: https://doi.org/10.3760/cma.j.cn121430-20230510-00158
Abstract:Objective: To explore the incidence of secondary hemophagocytic lymphohistiocytosis (sHLH) in elderly patients with severe SARS-CoV-2 infection, and to analyze and summarize its clinical features and risk factors for early identification of high-risk groups. Methods: A retrospective cohort study was conducted. From January to May 2020, No. 960 Hospital of People's Liberation Army, the Second Hospital Affiliated to Cheeloo College of Medicine of Shandong Province, the First Rehabilitation Hospital of Shandong Province, the Public Health Clinical Center Affiliated to Shandong University, and Ruijin Hospital Affiliated to Shanghai Jiao Tong University School of Medicine received 248 patients over 60 years old who were diagnosed with severe SARS-CoV-2 infection during their assistance to Hubei or support for diagnosis and treatment of SARS-CoV-2 infection in Shandong Province. The clinical data of patients were collected. According to the hemophagocytic lymphohistiocytosis diagnosis scoring (HScore) criteria, the patients were divided into sHLH group (HScore > 169) and non-sHLH group (HScore < 98). The demographic data, clinical features, laboratory results, the proportion of organ failure and 60-day mortality of patients were collected and compared between the two groups. The risk factors of sHLH and 60-day death were evaluated through binary multivariate Logistic regression analysis in elderly patients with severe SARS-CoV-2 infection. The receiver operator characteristic curve (ROC curve) was plotted to analyze the diagnostic value of indicators only or combined for sHLH. Results: Among 248 elderly patients with severe SARS-CoV-2 infection, 82 patients with incomplete data and untraceable clinical outcomes, and 35 patients with HScore of 98-169 were excluded. Finally, 131 patients were enrolled in the final follow-up and statistics, including 25 patients in the sHLH group and 106 patients in the non-sHLH group. Compared with the non-sHLH group, plasma albumin (ALB), hemoglobin (Hb), lymphocyte count (LYM), platelet count (PLT), fibrinogen (Fib) and prealbumin (PAB) in the sHLH group were significantly reduced, while alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), MB isoenzyme of creatine kinase (CK-MB), serum creatinine (SCr), C-reactive protein (CRP), D-dimer, ferritin (Fer), lactate dehydrogenase (LDH), procalcitonin (PCT), cardiac troponin I (cTnI), triglycerides (TG), interleukin-6 (IL-6), total bilirubin (TBil) were significantly higher. The fever and fatigue in the sHLH group were more severe than those in the non-sHLH group, and the patients in the sHLH group had higher rates of shock, acute kidney injury, liver dysfunction, and cardiac injury than the non-sHLH group. The 60-day mortality of patient in the sHLH group was significantly higher than that in the non-sHLH group [84.0% (21/25) vs. 40.6% (43/106), P < 0.01]. Binary multivariate Logistic regression analysis showed that high Fer [odds ratio (OR) = 0.997, 95% confidence interval (95%CI) was 0.996-0.998], D-dimer (OR = 0.960, 95%CI was 0.944-0.977), LDH (OR = 0.998, 95%CI was 0.997-0.999) and TG (OR = 0.706, 95%CI was 0.579-0.860) were independent risk factors for sHLH in elderly patients with severe SARS-CoV-2 infection (all P < 0.01), while elevated Fer (OR = 1.001, 95%CI was 1.001-1.002), LDH (OR = 1.004, 95%CI was 1.002-1.005) and D-dimer (OR = 1.036, 95%CI was 1.018-1.055) were independent risk factors for 60-day death of patients (all P < 0.01). The death risk of the sHLH patients was 7.692 times higher than that of the non-sHLH patients (OR = 7.692, 95%CI was 2.466-23.987, P = 0.000). ROC curve analysis showed that a three-composite-index composed of LDH, D-dimer and TG had good diagnostic value for sHLH in elderly patients with severe SARS-CoV-2 infection [area under the ROC curve (AUC) = 0.920, 95%CI was 0.866-0.973, P = 0.000]. Conclusions: Elderly patients with severe SARS-CoV-2 infection complicated by sHLH tend to be critically ill and have refractory status and worse prognosis. High Fer, LDH, D-dimer and TG are independent risk factors for sHLH, and are highly suggestive of poor outcome. The comprehensive index composed of LDH, D-dimer and TG has good diagnostic value, and can be used as an early screening tool for sHLH in elderly patients with severe SARS-CoV-2 infection.