Prognostic Factors For Stroke Associated With Non-valvular Atrial Fibrillation

As a common severe ischemic cerebrovascular disease,cardioembolic stroke,especially caused by nonvalvular atrial fibrillation(NAVF),has a high mortality and disability.Several prognostic scales have been proposed for mortality at hospital discharge in patients with ischemic stroke.However,there is not any outcome predictive model for stroke associated with NAVF.In order to reduce the mortality and recurrence rate in the light of individualized treatment strategy,and further develop a tailored treatment protocol for patients with acute stroke caused by NAVF,we performed a retrospective cohort study with a multi-center setting in Shaanxi province.Initially,we investigated the 30-day mortality of stroke associated with NAVF.Then,independent prognostic factors were identified in a multivariate logistic regression model.In the end,a predictive score for 30-day outcome in stroke associated with NVAF was developed,namely GPS-GF score.It can help clinicians evaluate those patients more accurately so that tailored treatments can be implemented to reduce the mortality of stroke associated with NVAF.Epilepsy is one of the most common complications of severe cerebrovascular disease.Epileptic seizures caused by stoke not only worsen the condition,but also affect the quality of life,sometimes even can be life-threatening.In previous studies,cardioembolic cause was indicated to be a major risk factor for epileptic seizures secondary to stroke.Therefore,we reviewed the consecutive patients with stroke associated with NVAF,and investigate the risk factors for early e pileptic seizures secondary to stroke associated with NVAF,in order to administ er early targeted interventions to prevent early epileptic seizures.Part 1 Development of the predictive model for 30-day mortality in stroke associated with non-valvular atrial fibrillationObjective: To develop and validate a predictive score for 30-day mortality in stroke associated with non-valvular atrial fibrillation(NVAF).Methods: From 2008 to 2015,consecutive patients with stroke associated with NVAF in 14 hospitals were reviewed.The survival outcome was assessed 30 days after stroke onset.All the patients were randomly divided into derivation group and validation group.Independent risk factors were identified in the multivariate regression model.Points were assigned based on their corresponding ?-coefficients in the logistic regression model.The receiver operating characteristic curve(ROC)and H-L analyses were used to evaluate the performance of the predictive model we made.Result: A total of 1077 patients with stroke associated with NVAF were enrolled in the study.There were 200 patients(18.6%)died within 30 days after stroke onset.From all the participants,718 patients were included into the derivation group.Glasgow Coma Scale at admission ? 8(OR 5.596,95%CI 3.300 – 9.489,P < 0.01),the complication of Pneumonia(OR 3.936,95%CI 2.101 – 7.371,P < 0.01),midline shift ? 1 cm(OR 6.859,95%CI 3.965 – 11.863,P < 0.01),random glucose level ?150 mg / dl at admission(OR 2.691,95%CI 1.599 – 4.531,P < 0.01)and female(OR 2.207,95%CI 1.302 – 3.741,P < 0.01)were identified as independent risk factors for 30-day mortality.An eight-point predictive model was built based on ?-coefficients in the multivariate regression model,which was referred to as GPS-GF score.The ROC analysis showed that GPS-GF score produced an excellent prediction of 30-day mortality in patients with stroke associated with NVAF(derivation group: AUC = 0.909,95% CI 0.885 – 0.929;validation group: AUC = 0.922,95% CI 0.889 – 0.947).Compared to i Score(AUC = 0.805,95% CI 0.760 – 0.845),GPS-GF has a significantly higher predictive accuracy(P < 0.01).Conclusion: The GPS-GF score can accurately predict the 30-day mortality of stroke associated with NVAF.Part 2 Incidence and risk factors for early epileptic seizures secondary to stroke associated with non-valvular atrial fibrillationObjective: To investigate the Incidence and risk factors of early epileptic seizures secondary to stroke associated with non-valvular atrial fibrillation(NVAF).Methods: From 2008 to 2015,consecutive patients with stroke associated with NVAF in 14 hospitals were reviewed.The incidence of early epileptic seizures within one week after stroke onset was recorded.Univariate comparisons and multivariate logistic regression analysis were used to identify the independent risk factors for early epileptic seizures secondary to stroke associated with NVAF.Results: A total of 1077 patients with stroke associated with NVAF were enrolled in the study.540 subjects were male,and 537 were female.The median age of the study is 73(ranging from 34 – 94).In total,21 patients(1.95%)had epileptic seizures within one week after stroke onset.Multivariate logistic regression analysis showed that NIHSS score on admission(95% CI 1.642-9.701,P < 0.01)and previous history of stroke(95% CI 2.631-15.593,P < 0.01)were independent risk factors for early epileptic seizures secondary to stroke associated with NVAF.Conclusion: Severe post-stroke neurological deficits and previous history of stroke are independent risk factors for early epileptic seizures secondary to stroke associated with NVAF.