Construction and Evaluation of the Model Based on Serum Anion Gap,Albumin and APACHE Ⅱ Score to Predict the Prognosis of Patients with Severe Pulmonary Tuberculosis
Objective:To analysis the value of prediction model based on serum anion gap(AG),albumin(Alb)and acute physiology and chronic health evaluation Ⅱ(APACHE Ⅱ)score in patients with severe pulmonary tuberculosis(PTB).Methods:60 severe PTB patients admitted to our hospital from October 2020 to January 2023 were included in severe PTB group,patients were divided into death group(21 cases)and survival group(39 cases)according to the in-hospital survival status,and 60 healthy subjects in our hospital during the same period were selected as control group.The levels of serum Alb and AG were detected,and APACHE Ⅱscore was calculated.The prognostic influence factors of severe PTB patients were analyzed by multivariate logistic regression model and the prediction model was constructed based on these influence factors.The predictive value of serum AG,Alb,APACHE Ⅱ score and prediction model for the prognosis of severe PTB patients were analyzed by receiver operating characteristic(ROC)curve.Results:Compared with control group,the serum AG level was increased and the Alb level was decreased in severe PTB group(P＜0.05).The in-hospital mortality of 60 severe PTB patients was 35.00％(21/60).Multivariate logistic regression model analysis showed that,the increase of AG and APACHE Ⅱ score were independent risk factors for the prognosis of severe PTB patients,and the increase of Alb was an independent protective factor(P＜0.05).The prognostic prediction model equation of severe PTB patients was Ln(P/1-P)=-0.173+0.105×AG-0.057×Alb+0.057×APACHE Ⅱ score,Hosmer-Lemeshow test(P＞0.05).ROC curve analysis showed that,the area under the curve(AUC)of the prediction model for predicting the prognosis of severe PTB patients was 0.859,which was greater than the independent prediction of serum AG,Alb and APACHE Ⅱ score.Conclusion:Serum AG,Alb and APACHE Ⅱ score were independently associate with the prognosis of severe PTB patients,the prediction model base on serum AG,Alb and APACHE Ⅱ score had a high predictive value for the prognosis of severe PTB patients.
Severe pulmonary tuberculosisAnion gapAlbuminAPACHE Ⅱ scorePrognosisPrediction model
NETL
NSTL
万方数据
