Nomogram To Predict The Risk Of Acute Kidney Injury In Patients With Diabetic Ketoacidosis

Objective: This study aimed to investigate the risk factors for acute kidney injury(AKI)in patients with diabetic ketoacidosis(DKA)in the intensive care unit(ICU)and to develop a nomogram to predict the risk of AKI based on this.Materials and Methods: A total of 760 DKA patients met the inclusion and exclusion criteria in the Medical Information Mart for Intensive Care III(MIMIC-III)database.These 760 patients were randomly divided into the training set(70%,n = 532)and the validation set(30%,n = 228).And then,clinical characteristics of the data set were collected and utilized to establish a nomogram for the prediction of AKI during ICU stay,including vital signs,comorbidities,complications,laboratory tests,therapeutic measures,length of hospital stay,and survival status.The candidate predictors were determined using the Lasso algorithm(LASSO)regression model.Meanwhile,a multivariate logistic regression analysis was performed based on variables derived from LASSO regression,in which variables with P < 0.1 were included in the final model.Then,a nomogram was constructed applying these significant risk predictors based on a multivariate logistic regression model.The discriminatory ability of the model was determined by illustrating a receiver operating curve(ROC)and calculating the area under the curve(AUC).The sensitivity,specificity,positive predictive value(PPV),and negative predictive value(NPV)was also calculated.Moreover,the calibration plot and HosmerLemeshow goodness-of-fit test(HL test)were conducted to evaluate the performance of our newly bullied nomogram.Decision curve analysis(DCA)was performed to evaluate the clinical net benefit.Results:A total of 760 patients were finally included in this study,and 314 patients developed AKI during ICU stay,with an incidence rate of 41%;the results of multivariate logistics regression model showed that diabetes classification(T2DM),combined microangiopathy,history of congestive heart failure(CHF),history of hypertension,lower diastolic blood pressure(DBP),lower urine volume,higher Glasgow Coma Scale(GCS)and higher respiratory rate(RR)were independent risk factors for AKI in DKA patients during ICU stay(P < 0.05),and a nomogram was constructed based on this.The AUC in this nomogram modeling dataset was 0.747(95% CI,0.706 – 0.789),and the AUC in the validation dataset was 0.712(95% CI,0.642 – 0.782);the nomogram showed good calibration according to the calibration graph and HL test(P > 0.05);The sensitivity,specificity,PPV,and NPV were 83.8,55.9,58.8,82.1% in train dataset,and 66.3,69.7,57.0,77.3% in validation set.The DCA results showed that when the prediction probability thresholds of the modeling and validation cohorts were set at 17% – 100% and 23% – 71%,the net benefit ranged from 0 – 31% and 0 – 20%,respectively,and the smaller the threshold,the net benefit,which indicated that the model had good clinical usability.Conclusions: The nomogram predicted model for predicting AKI in patients with DKA was constructed.This predicted model can help clinical physicians to identify the patients with high risk earlier and prevent the occurrence of AKI and intervene timely to improve prognosis.