Prognostic value of the combination of Charlson comorbidity index,lactate/albumin ratio and sequential organ failure assessment in the patients with acute respiratory distress syndrome

Xi Zhenchuang; Wang Rui; Li Meiling; Mao Wenjie; Cao Wen

doi:10.3969/j.issn.1002-1949.2025.05.007

Abstract：Objective To assess the prognostic value of Charlson comorbidity index(CCI),lactate/albumin ratio(L/A)and sequential organ failure assessment(SOFA)in the patients with acute respiratory distress syndrome(ARDS).Methods A total of 173 patients with ARDS admitted to Intensive Care Unit of the Second Hospital of Lanzhou University from April 2018 to February 2024 were retrospectively analyzed.Data were collected,including genders,ages,underlying diseases,the causes of ARDS,blood lactate(LAC),serum albumin,procalcitonin(PCT),blood gas analysis,acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)score,Murray lung injury score,SOFA score,ICU stay,mechanical ventilation time,28-day mortality.At the same time,the L/A and CCI scores were calculated.The patients were divided into survival group(n=108)and death group(n=65)depending on their clinical outcomes after 28-day treatment.Univariate and multivariate Logistic regression analysis were used to screen independent prognostic risk factors for ARDS,and a nomogram prediction model was established.Receiver operating characteristic(ROC)curves and calibration curves were used to evaluate the discriminability and calibration of the nomogram model.Decision curve analysis(DCA)was used to evaluate the clinical applicability of the model.Results 173 patients with ARDS were enrolled in this study.The mean age was(64.17±12.02)years,65 patients died and the fatality rate was 37.6％.The age,underlying diseases,APACHE Ⅱ,SOFA,L/A,CCI score and LAC level in death group were significantly higher than those in survival group(all P＜0.05).Regarding the length of ICU stay,time of mechanical ventilation,Murray lung injury score and PCT level between the survival group and the death group,the difference was not statistically significant(P≥0.05).Multivariate Logistic regression analysis showed that age(OR=1.088,95％CI 1.041-1.138),SOFA score(OR=1.296,95％CI 1.074-1.564),APACHEⅡ score(OR=1.141,95％CI 1.021-1.274),L/A(OR=1.097,95％CI 1.023-1.177)and CCI score(OR=1.700,95％CI 1.281-2.255)were independent risk factors for 28-day prognosis of patients with ARDS(all P＜0.05).The five independent variables were used to construct a nomogram model for predicting the prognosis of ARDS patients through the R language(R4.2.1)rms package.The test results showed that the area under the ROC curve(AUC)of the nomogram was 0.890(95％CI 0.834-0.933),with a sensitivity of 67.69％and a specificity of 95.37％,which were significantly higher than CCI score(Z=3.787,P＜0.001),L/A score(Z=5.354,P＜0.001)and SOFA score(Z=4.393,P＜0.001),suggesting that the nomogram has better discrimination ability in predicting outcomes of ARDS patients.The calibration curve showed that the predicted probability of the model was well correlated with the actual probability(Hosmer-Lemeshow test:x2=7.742,P=0.459).Clinical decision curve analysis showed that the nomogram has a good clinical net benefit.Conclusions The nomogram constructed by CCI and L/A combined with SOFA score has good predictive performance and clinical practicability for 28-day prognosis of ARDS patients.