Red blood cell distribution width and platelet distribution width may aid in predicting the prognosis of influenza patients with lower respiratory tract infection

Teng Fei; Guo Ruiwen; Liang Menglin; Mei Xue

doi:10.3969/j.issn.1002-1949.2024.04.003

Abstract：Objective To investigate whether blood cell counts and related parameters can predict the prognosis of influenza patients with lower respiratory tract infection.Methods A retrospective study was conducted on the patients of influenza combined with lower respiratory tract infection admitted to or kept in the Emergency Medicine Department,Fever Clinic or Respiratory Department of Beijing Chaoyang Hospital,Capital Medical University from 2013 to 2023.The patients were divided into survival group and death group based on 28-day death as the study endpoint,and the differences of blood cell counts and related parameters between the two groups were compared and analyzed.Independent predictors of death were also screened by Logistic regression,and the prediction model was constructed and evaluated.Results A total of 417 patients were enrolled in the study,including 361 patients(86.6%)in the 28-day survival group and 56 patients(13.4%)in the 28-day death group.The lymphocyte count and platelet count in the death group were significantly lower than those in the survival group.The red blood cell distribution width,the platelet distribution width,the mean platelet volume and the large platelet ratio in the death group were higher than those in the survival group(P<0.05).Multivariate Logistic regression analysis screened independent correlated variables,and found when lymphocyte count≤0.68×109/L,or platelet count≤147×109/L,or red blood cell distribution width>12.7%,or platelet distribution width>13.9 fL,the patients had an increased risk of 28-day mortality.The AUC of the predictive model established by the four indicators was 0.734(95%CI 0.664-0.804),sensitivity was 58.9%,specificity was 80.4%,positive predictive value was 32.0%,negative predictive value was 92.7%,positive likelihood ratio was 3.01,and negative likelihood ratio was 0.51.The concordance index of the model was 0.735,and Hosmer-Lemeshow goodness of fit test showed a good effect(P =0.406).Conclusions Lymphocyte count,platelet count,red blood cell distribution width and platelet distribution width can be used as early indicators to predict the 28-day mortality risk in the patients with influenza-associated lower respiratory tract infection.The mortality risk probability corresponding to clinical measurements can be calculated according to the nomogram.