Analysis of factors influencing poor clinical outcomes in patients with acute hypoxemic respiratory failure after Venturi mask oxygen therapy and construction of a nomogram model

JIANG Long; FANG Hui; XIA Yunqing; HU Ai; LI Bisheng; WANG Lingling; WANG Rui; WANG Bin

doi:10.3969/j.issn.1674-3806.2025.10.13

Abstract：Objective To analyze the factors influencing poor clinical outcomes in patients with acute hypoxemic respiratory failure(AHRF)after Venturi mask oxygen therapy and to construct a nomogram model.Methods A retrospective analysis was conducted on the clinical data of 203 patients with AHRF who were admitted to Anqing Hospital of Chinese Medicine from November 2021 to April 2024.All these patients received Venturi mask oxygen therapy.According to the clinical outcomes of the patients after the therapy,they were divided into poor outcome group(57 patients)and good outcome group(146 patients).Multivariate logistic regression was used to analyze the factors influencing the clinical outcomes of patients with AHRF after Venturi mask oxygen therapy,and a nomogram model for predicting poor clinical outcomes of patients with AHRF after Venturi mask oxygen therapy was constructed based on the screened indicators.The predictive efficiency and application value of the constructed nomogram model were evaluated through receiver operating characteristic(ROC)curve,Hosmer-Lemeshow test,calibration curve and decision curve.Results In the 203 patients,57 patients(28.08%)hadpoor clinical outcomes.Amongthe57 patientswithpoor clinical outcomes,45 patientswere converted to mechanical ventilation,and 10 patients had deteriorated conditions leading to rescue,and 2 patients died.The results of multivariate logistic regression analysis showed that older age[OR(95%CI)=1.754(1.040-2.958)],higher Acute Physiology and Chronic Health EvaluationⅡ(APACHEⅡ)scores[OR(95%CI)=2.373(1.535-3.669)],fever[OR(95%CI)=2.274(1.170-4.419)]and concurrent multiple organ dysfunction syndrome(MODS)[OR(95%CI)=1.635(1.093-2.446)]were independent risk factors for poor clinical outcomes in patients with AHRF after Venturi mask oxygen therapy(P＜0.05).A high respiratory rate-oxygenation(ROX)index at 1 hour after treatment[OR(95%CI)=0.556(0.364-0.849)]was an independent protective factor for promoting good clinical outcomes in the patients(P＜0.05).A nomogram model for predicting poor clinical outcomes in patients with AHRF after Venturi mask oxygen therapy was constructed on basis of age,fever,concurrent MODS,APACHEⅡscores and ROX index at 1 hour after treatment.The results of ROC curve analysis showed that the nomogram model had good predictive efficiency[AUC(95%CI)=0.925(0.880-0.957),P＜0.001],with a sensitivity of 92.98%and a specificity of 85.62%.The results of Hosmer-Lemeshow test showed that there was no statistically significant difference between the probability of the patients'poor clinical outcomes predicted by the nomogram model and the actual probability(χ2=0.859,P=0.354).The calibration curve showed that the predicted probability of the nomogram model had a good consistency with the actual probability.The decision curve indicated that the nomogram model could be applied to the vast majority of clinical decision-making settings.Conclusion Age,fever,concurrent MODS,APACHEⅡscores,and the ROX index at 1 hour after treatment are the factors influencing the clinical outcomes of AHRF patients after Venturi mask oxygen therapy.The nomogram model constructed on basis of these indicators is helpful for rapid clinical decision-making,improves the patients'prognosis,and has good clinical application value.