Development And Validation Of A Prediction Model For Death In Patients With Intracerebral Hemorrhage

Objectives Intracerebral hemorrhage(ICH)is a serious neurological disease with a high rate of death and disability.In the past few decades,some risk factors related to the risk of death from spontaneous intracerebral hemorrhage have been discovered,then established many models to predict mortality after cerebral hemorrhage,but most of the models have not been widely accepted and used.So far,suitable prognosis models that can be widely applied for clinical prognosis have not been studied clearly to date.This study aims to establish and verify a predictive model of spontaneous cerebral hemorrhage death based on admission characteristics,which is used to predict the 2-day death of patients with cerebral hemorrhage.Methods This study is a retrospective cohort study.During 2015～2017,the records of a cohort of 397 patients with clinically diagnosed cerebral hemorrhage were collected for model development.Minimum absolute contraction and the selection operator(lasso)regression model were used to determine factors that most consistently and correctly predicted death after cerebral hemorrhage.Discrimination and calibration were used to evaluate the performance of the resulting nomogram.After internal validation,the nomogram was further assessed during 2017～2018 using a different cohort of 200 consecutive subjects.This study uses Stata 13.0 and R 3.6.3 for data analysis.Results Univariate analysis results showed that Glasgow Coma Scale,history of anticoagulant drug use,hematoma volume,primary intraventricular hemorrhage,brainstem hemorrhage,systolic blood pressure,diastolic blood pressure,pulse pressure,mean arterial pressure,white blood cell count,neutrophil count,Lymphocyte count,monocyte count,platelet to lymphocyte ratio,D-dimer level,total protein level,albumin level,alanine aminotransferase level,plasma glucose level were related to the 2-day death of patients with spontaneous intracerebral hemorrhage;The nomogram included four predictors from the lasso regression analysis: Glasgow Coma Scale,hematoma location,hematoma volume,and primary intraventricular hemorrhage.Internal verification showed that the model had good discrimination,(the area under the curve is 0.928),and good calibration is good(unreliability(U)statistic,P = 0.774).The nomogram still showed good discrimination(area under the curve=0.909)and good calibration is good(unreliability(U)statistic,P =0.844)in the verification cohort data.Decision curve analysis showed that the prediction nomogram was clinically useful.Conclusions GCS score,hematoma location,hematoma volume,and primary intraventricular hemorrhage are important predictors of cerebral hemorrhage death.This prediction model has excellent performance and can be used for early,simple,and accurate prediction of early death following cerebral hemorrhage.Figure 5;Table 3;Reference 113...