Analysis Of Risk Factors,and Biomarkers Predicting Stroke Associated Acute Kidney Injury

Objective To investigate the incidence,risk factors and predictive biomarkers of acute kidney injury(AKI)and 28-day mortality among patients with stroke in the intensive care unit(ICU).It is beneficial for clinical staff to early identify,prevent and treat such high-risk groups,which may better understand the epidemiology and clinical characteristics of AKI and 28-day mortality in stroke patients.Materials and Methods This was a single-center,retrospective cohort study that included adult stroke patients admitted to the ICU of Zhongnan Hospital of Wuhan University in the past five years from January 2013 to January 2018.All the selected cases were diagnosed as intracerebral hemorrhage or ischemic stroke based on the imaging examination.Patients with any of the following were excluded:time between stroke diagnosis and ICU admission was more than 7 days;hospital stay was less than48 hours;without serum creatinine data within 24 hours after admission;baseline estimated glomerular filtration rate(e GFR)<15 m L/min/1.73 m~2;history of end stage kidney disease(ESKD)on maintenance dialysis,or kidney transplantation;pregnancy;or other missing data.Case data were collected from the basic information,past medical history and related auxiliary examinations.The patients were divided into AKI group and non-AKI group within 7 days after ICU admission,and SPSS 20.0statistical software was used to compare the two groups.The primary endpoint was the incidence of AKI,and the secondary endpoint was 28-day mortality.The definition of AKI was based on the Kidney Disease Improving Global Outcomes(KIDGO)criteria.In addition,the risk factors of AKI were assessed by Logistic regression,receiver operating characteristic(ROC)curves were used to analyze predictive biomarkers of AKI and 28-day mortality,and the risk factors of 28-day mortality were evaluated by Cox regression and Kaplan-Meier survival curve.Result(1)In this study,479 patients were initially included and 381 patients were finally enrolled according to the exclusion criteria.Among them,115 were AKI patients,266 were non-AKI patients,and the incidence of AKI was 30.18%.(2)Logistic regression analyses revealed that the following features at ICU admission significantly increased the risk of developing AKI even after other adjustments:an increased National Institutes of Health Stroke Scale(NIHSS)score(odds ratio[OR]=1.136,95%confidence interal[CI]:1.074～1.202,p<0.001)and Acute Physiology and Chronic Health Evaluation(APACHE)II score(OR=1.107,95%CI:1.004～1.220,p=0.042),hypertension(OR=2.346,95%CI:1.244～4.426,p=0.008),use of loop diuretics(OR=1.961,95%CI:1.060～3.625,p=0.032)and higher serum cystatin C(s Cys C;OR=8.156,95%CI:2.353～28.272,p=0.001).(3)The area under the ROC curves for predicting AKI using s Cys C was 0.772(95%CI:0.726～0.813,p<0.001)with a sensitivity of 70.43%and a specificity of 71.05%,which was slightly better than that of other biomarkers.(4)In this study,113 patients died 28 days after ICU admission,while 268patients survived,and the 28-day mortality was 29.7%.The Cox regression analyse showed that risk factors of 28-day mortality after ICU admission:a history of prior stroke(hazard ratio[HR]=1.768,95%CI:1.122～2.786,p=0.014),a higher NIHSS score(HR=1.133,95%CI:1.087～1.180,p<0.001),a lower Glasgow Coma Scale(GCS)score(HR=0.884,95%CI:0.793～0.985,p=0.026),a higher APACHE II score(HR=1.177,95%CI:1.110～1.249,p<0.001)and s Cys C≥0.93mg/L(HR=1.844,95%CI:1.217～2.796,p=0.004).(5)The ROC curve indicated that the area under curve(AUC)of s Cys C was0.723(95%CI:0.675～0.767)for predicting 28-day mortality,and the optimal cutoff value was 0.93 mg/L,with a sensitivity and a specificity of 66.37 and 69.03%,respectively.However,there was no statistical difference between s Cys C and Scr in predicting 28-day mortality.(6)The Kaplan-Meier survival curve showed that the cumulative survival rate at28 days was greater in patients with below this threshold compared with s Cys C≥0.93mg/L(Log rank,p<0.001).Conclusions In this study,we explored the incidence,risk factors and predictive biomarkers of AKI and 28-day mortality among stroke patients in the ICU.The incidence of AKI was 30.18%and the 28-day mortality was 29.7%in this population.The risk factors of AKI after ICU admission:an increased NIHSS and APACHE II score,hypertension,use of loop diuretics;the risk factors of 28-day mortality:a higher NIHSS score and APACHE II score,a lower GCS score and a history of prior stroke.In addition,the levels of s Cys C at ICU admission were important biomarkers for predicting AKI and 28-day mortality.