Effectiveness Of A Risk Prediction Score For Kidney Failure Or Mortality In Post-traumatic Rhabdomyolysis

Background:With the development of society, trauma has become the leading cause of mortality and disability, meantime, it is the first cause which is responsible for the death of people aged1to44. Trauma is a common one among many factors cuasing rhabdomyolysis. The objective of this study was to determine the model that McMahon generated effectively for post-traumatic rhabdomyolysis (PTRML)Design:A retrospective study. Patients:We enrolled consecutive adult trauma patients(≥18yrs) from June1,2012to April30,2014from Emergency center of the second affiliated hospital of Zhejiang University, with creatine phosphokinase (CPK) levels in excess of500U/L within72hours after blunt injury. Exclusion criteria:To reduce confounding factors, data from patients who preexisted end-stage renal disease, transferred from an outside facility who were receiving RRT for at least24hours, suffered with epilepsy (including traumatic epilepsy), acute myocardial infarction, arrhythmia(any type) and myositis were excluded from the study. Method:Continuous normally or near normally distributed variables are reported as means±tandard deviations (SD), as odds ratio±95%confidence interval (CI), compared by Student’s t-test. Non-normally distributed continuous data are expressed as medians with inter-quartile ranges (IQR) appropriately and compared by Mann Whitney U test. Categorical variables described as percentages and compared using Chi-square test. We calculated MC score for each patient with admission score and extreme score in72hours. Logistic regression analysis was used to assess the association of variables with poor prognosis. Outcome:The primary outcome:poor prognosis (CRRT or death), the secondary outcome:acute kidney injury (AKI).Result:During the study period,1,115patients were admitted to the EC of417patients (37.40%) admitted with inclusion criteria. The poor prognosis and AKI among the417patients are45(10.97%) and189(45.32%, see) respectively. Systolic blood pressure (SBP), diastolic blood pressure (DBP), ISS were all with statistical significance between RML or not. McMahon score failed to predict the poor prognosis of TRML with CK>500U/L, the area under the ROC is0.532, P=0.485,95%CI0.445-0.619and0.581(P=0.074,95%CI0.495-0.668). We assessed the association of variables with poor prognosis using ROC, with area under the curve is0.881(P=0.000,95%CI,0.822-0.962).Conclusion:Currently, we can use the MC score to predict the prognosis of PTRML with CK>5000U/L.But this score can not use to predict the poor prognosis for patients with500U/L<CK<5000U/L.