Development of a nomogram for predicting poor prognosis in elderly patients with acute respiratory distress syn-drome treated with mechanical ventilation

SUN Feng; RU Jiang-li; CHANG Qing; LIN Zhi-qiang; ZHANG Meng-meng; ZHANG Hai-tao; WEI Dong

doi:10.13820/j.cnki.gdyx.20242913

Abstract：Objective To analyze the factors influencing poor prognosis in elderly patients with acute respiratory distress syndrome(ARDS)treated with mechanical ventilation and to construct a nomogram model for prognosis prediction based on these factors,followed by internal and external validation.Methods A total of 160 elderly patients with ARDS who received mechanical ventilation between February 2020 and September 2022 were enrolled in the study as the model-ing cohort.Univariate and multivariate Cox proportional hazards models were used to identify independent factors influen-cing poor prognosis in these patients.The internal data was used to validate the clinical performance of the nomogram mod-el.An additional 69 elderly patients who received mechanical ventilation for ARDS between October 2022 and February 2024 were included as the validation cohort for external validation,with the sample size for the validation cohort being 30％of the total sample.Results There were no significant differences in the age,gender,and other baseline character-istics between the two groups(P＞0.05).Compared to the survivors,patients in the death group had higher neutrophil-to-lymphocyte ratio(NLR),acute physiologyand chronic health evaluation(APACHE Ⅱ)score,and sequential organ failureassessment(SOFA)score(P＜0.05).NLR,APACHE Ⅱ score,and SOFA score were independent factors influ-encing the prognosis of elderly patients with ARDS treated with mechanical ventilation(P＜0.05).The time-calibration curve of the nomogram showed that the C-index for predicting poor prognosis at 1,2,and 3 months were 0.829(0.822-0.869),0.854(0.832-0.876),and 0.862(0.844-0.887),respectively.The time-decision curve demonstrated that the nomogram provided significant clinical net benefit.The receiver operating characteristic(ROC)curve for the vali-dation cohort showed an area under the ROC curve(AUC)of 0.949(95％CI:0.897-1.000).The Hosmer-Leme-show(H-L)goodness-of-fit test showed x2=3.198,P=0.921(P＞0.05),indicating that the prediction model had good discriminative ability and consistency.Conclusion The nomogram based on NLR,APACHE Ⅱ score,and SO-FA score can effectively assess the prognosis of elderly patients with ARDS treated with mechanical ventilation,and it helps implement more aggressive treatment measures for these patients.