Establishment Of Early Warning Model Of Acute Kidney Injury In Patients With Severe Multiple Injuries

Objective:By analyzing the independent risk factors of Acute Kidney Injury in patients with severe multiple injuries,an early warning model of AKI in patients with severe multiple injuries is intended to be established,so as to reduce the incidence and mortality of AKI in patients with severe multiple injuries and provide help for early detection,prevention and improvement of the prognosis of patients with AKI.Methods:Patients with severe multiple injuries who met the inclusion and exclusion criteria admitted to the Intensive Care Unit（ICU）of Cangzhou People’s Hospital from January 2018 to December 2022 were collected as the research objects.Patients with severe multiple injuries were divided into AKI group and non-AKI（N-AKI）group according to whether they developed AKI within 7 days of ICU admission（including day 7）.SPSS26.0 was used for data analysis,and general clinical data of the two groups were compared:The results included gender,age,time of admission and discharge,length of stay in ICU,transfer out of the department or discharge,time of arrival after Injury,cause and location of injury,past history,Injury Severity score（ISS）,APACHEⅡscore（Acute Physiology and Chronic Health Evaluation,acute physiology and chronic health score）,whether mechanical ventilation was performed within 7 days after admission.Laboratory indicators:Including blood routine,C-reactive protein（CRP）,liver function,Lactate（Lac）,blood Potassium（K）,blood natrium（Na）and blood calcium（Ca）,UREA nitrogen（UREA）,Activated Partial Thromboplastin Time（APTT）,procalcitonin（PCT）,D-Dimer（D-D）and other indicators,and the independent risk factors of AKI in patients with severe multiple injuries were obtained by multivariate binary Logistic regression analysis,and the early warning model of AKI in patients with severe multiple injuries was built.Receiver Operating Characteristic curve（ROC curve）was plotted and The Area Under The Curve（AUC）was used to evaluate the prediction effect of this model.Hosmer-Lemeshow goodness of fit test was used to evaluate the calibration degree of the model.This study suggests that P<0.05 was statistically significant.Results:1.A total of 288 patients were included in this study,including 202males（70.1%）and 86 females（29.9%）,with a male to female ratio of 1:0.43.The mean age was（49.55±16.74）years,and the median age was 51.5 years.The top three reasons for severe multiple injuries were traffic accidents in233 cases（80.9%）,falling from high places in 27 cases（9.4%）and smashing heavy objects in 15 cases（5.2%）.The most common injury sites were limb injury in 131 cases（45.4%）,craniocerebral injury in 68 cases（23.6%）and abdominal injury in 44 cases（15.3%）,followed by chest injury in 27 cases（9.3%）,head and face injury in 9 cases（3.1%）and spine injury in 9 cases（3.1%）.The clinical outcomes of patients were 110 cases（38.2%）transferred to orthopedics,38 cases（13.2%）transferred to brain surgery,29 cases（10.1%）transferred to hepatobiliary surgery/gastrointestinal surgery,25 cases（8.7%）transferred to thoracic surgery,43 cases（14.9%）discharged automatically,40cases（13.9%）died,and 3 cases（1%）transferred to other departments.2.The median time for patients to see a doctor（the time from the accident to the hospital for treatment）was 1h;The average APACHEⅡscore was（20.89±9.39）.The average ISS score was（31.42±9.63）.3.AKI occurred in 63 patients with severe multiple injuries,with an incidence of 21.9%.4.Univariate analysis showed that shock,hyperlactemia,mechanical ventilation,age,PH,actual bicarbonate,albumin,K,Ca,APACHEⅡscore,AP-TT were associated with AKI in patients with severe multiple injuries（P<0.05）.5.The results of multivariate Logistic regression analysis showed shock,hyperlactemia,mechanical ventilation,high level APACHEⅡscore,age>50was an independent risk factor for AKI in patients with severe multiple injuries（P<0.05）.6.Successful construction of severe multiple injuries combined wit-h AKI risk warning model:Logistic（P）=-5.173+1.206χ₁+0.969χ₂+1.18-5χ₃+1.321χ₄+1.067χ₅,the occurrence of shock was recorded asχ₁,t-he o-ccurrence of hyperlactemia was recorded asχ₂,mechanical ventil-ation was recorded asχ₃,high level APACHEⅡscore was recorded asχ₄,age>50 isχ₅;The model was tested and verified.Hosmer-Leme show goodness of fit test P=0.871,indicating that the model was well fitted（P>0.05）;The AUC of the model was 0.839（standard error SE w-as 0.029,P<0.001,95%CI:0.782-0.895）,the Yoden index was 0.560,and t-he optimal cut-off value was 0.192.The sensitivity and specificity of t-he model were 77.78%and 78.22%.Conclusions:1.The patients with severe multiple injuries were mainly middle-aged people,and the top three main injury sites were limb injury,craniocerebral injury and abdominal injury.2.The incidence of AKI in patients with severe multiple injuries was21.9%.3.Shock,hyperlactemia,mechanical ventilation,high level APACHEⅡscore,age>50 is an independent risk factor for AKI in patients with severe multiple injuries,and a risk warning model for AKI in patients with severe multiple injuries was successfully constructed.4.The early warning model of AKI in patients with severe multiple injuries was tested,indicating that the model has good resolution and calibration,and can better evaluate the risk of AKI in patients with severe multiple injuries,providing a basis for early clinical identification and intervention.