Rates and Risk Factors of Acute Ischemic Stroke Mortality [A Multi-Center Prospective Cohort Study]

ABSTRACT


INTRODUCTION
Stroke is the second commonest cause of mortality and a significant cause of morbidity [disability] worldwide regardless the recent improvement and advances in the management options [1] .Stroke [ischemic or hemorrhagic] is the commonest known cerebrovascular disease.It could be induced by modifiable and nonmodifiable risk factors.The commonest nonmodifiable risk factors are age and race.On the other side, the modifiable risk factors include elevated blood pressure [hypertension] or blood sugars [diabetes mellitus], dyslipidemia, obesity, physical inactivity and cardiovascular disorders .Other nontraditional risk factors are linked to ischemic stroke [e.g., inflammatory bowel disease [6] , blood pressure and nocturnal heart rate variability, endothelial function and recurrent cerebrovascular accidents [7] .However, there are heterogeneity in results and paucity of data.
The stroke-related comorbidities are frequently reported in the first week after the stroke.The mortality risk is higher during hospital admission and remains high during the whole duration of hospital stay.Intensive care unit [ICU] admission is increasing and this is especially for aged populations [8] .ICU admission is extremely critical for the management of critically-ill stroke patients.The ICU admission criteria include dysfunction/ failure of single or multiple organs, low score of Glasgow Coma Scale, need for advanced and continuous monitoring and postoperative observation [9] .
The neurocritical care aimed principally directed to prevention, early identification and treatment of secondary brain insults [e.g., cerebral edema, intracranial hypertension, cerebral circulation impairment, hypo-or hyper-tension, hypoxia, hypercapnia and hyperthermia].Thus, it is critically important to recognize the potential modifiable risk factors in critically ill patients with stroke.Thus, this could improve the overall patient outcome [9] .This study aimed to determine the incidence and risk factors associated with in-hospital ischemic stroke mortality to improve the patient survival [in acute phase of ischemic stroke].

PATIENTS AND METHODS
In this prospective cohort study, 128 with acute ischemic stroke were included.The diagnosis was based on the clinical and radiological findings.They were selected from the emergency department [ED] and stroke-treatment unit, department of neurology, Al-Azhar University hospitals.They were selected from those admitted between the first of December 2022 to the end of May 2023.

Statistical analysis of data:
The qualitative data were presented as frequencies and percentages.Otherwise, the quantitative data were presented by their means ± standard deviation [SD] when normally distributed.Median and interquartile range [IQR] were used for preservation of nonnormally distributed data.Patients were then categorized into alive and dead.Simple and multivariate analysis was performed to estimate the mortality associated risk factors.P value < 0.05 was set as the value of significant differences.The association between groups were calculated by independent samples student "t" test or Chi square test for quantitative and qualitative data respectively.All statistical analyses were achieved by the statistical package for social sciences [SPSS], version 20 [IBM®, Armonk, USA].

RESULTS
The present study included 128 patients with acute ischemic stroke.When searching factors associated with mortality, we found significant association between mortality and each of atrial fibrillation, older age, progression of stroke, hemorrhagic transformation, edema, respiratory and GIT complications [Table 2].
Table [3] showed significant association between mortality and each of NIHSS, mRS, APACHE-II and GCS at admission.There was a significant increase of all scores in died than alive patients.
The logistic regression model showed that patients who had respiratory complications were 6.4 times more likely to die compared to patients without respiratory complications [post stroke pneumonia] and adjusted to demographic and comorbid factors [adjusted OR = 6.4,p = 0.04].Additionally, GIT complications [GIT bleeding] were 7.9 time more associated with death among stroke patients during the inhospital period adjusted to demographic and comorbid factors OR = 7.9, p = 0.03].Moreover, AF patients were 11.8 times more likely to have death during the in-hospital period compared to non-AF patients adjusted to other variables in the model [Table 4, Figure 1].
The area under the ROC curve is significantly higher than the area of chance, which is 0.5, concluding that the model has a significant prediction ability [Table 5].

DISCUSSION
This study aimed to determinate the incidence of in-hospital mortality rate of ischemic stroke patients in addition to risk factors and complications associated with mortality.The study included 128 patients, and in-hospital mortality was reported among 11 patients [8.6%].There was no significant difference between anterior and posterior circulation ischemic stroke as regard mortality rate.This is supported by the study Jalali et al. [10] who showed that the origin of emboli or thrombus were not an outcome predictor in ischemic strokes.The most important predictor was baseline NIHSS.In addition, Zürcher et al. [11] showed that there were specific differences in clinical presentation, etiology, and arterial pathology between posterior and anterior circulation strokes.But did not affect the clinical outcome.
The in-hospital mortality rate was 8.6% with significant older age and higher coma score in died than alive patients.Kortazar-Zubizarreta et al. [12] revealed in-hospital mortality rate was 7.13%.Viderman et al. [9] in-hospital mortality rate of ischemic stroke was 36%.The mean age was 63 years, 45% were male and the mean GCS at baseline was 10.3.low mean GCS and involvement of recurrent stroke patients in this study explain the cause of high mortality rate.
In the hospital-based follow-up study by Alhazzani et al. [13] , the in-hospital mortality was comparable to ours [9.7%], with an average length of stay of 10.67 days and a median of 7 days.
In logistic regression, we showed significant association between atrial fibrillation and in hospital mortality, and this explained by that cardiac causes of death are common in acute stage of ischemic stroke.In addition, ischemic stroke in patients with AF was also associated with higher incidence of acute kidney injury, bleeding and infectious complications and severe disability, as reported by Alkhouli et al. [14] .
In a study of Hannawi et al. [15] the case fatality rate was significantly associated with increased age and mortality was higher in ischemic stroke patients with AF than in those without AF [13.0%vs 7.3%; P <0.001].AF was a significant predictor of in-hospital mortality [OR 1.30; 95% CI 1.28-1.31;P <.001].
In logistic regression we showed significant association between pneumonia, GIT bleeding and in hospital mortality.In majority of SAP [Strokeassociated pneumonia], the causative factor was hospital-acquired infection.Keller et al. [16] included parapneumonic effusion, necrotizing pneumonia, empyema, pneumatocele and lung abscess formation, as risk factors for mortality in addition to pneumonia.However, Kortazar-Zubizarreta et al. [12] reported that, post-stroke pneumonia was not associated with the intensive care unit mortality but was associated with a 1.6-fold increase in ICU length of stay with an increase in 1-year mortality rate.However, this study is different than the current one.They focused on critically-ill acute ischemic stroke patients requiring invasive MV "whatever the reason".
In our study, there was a significant association between high NIHSS and in hospital mortality.This is supported by Kortazar-Zubizarreta et al. [12]   , Liu et al. [17] and Heuschmann et al. [18] .In addition, the significant older age associated with in-hospital mortality is supported by Liu et al. [16] and Heuschmann et al. [18] who showed the same association in females only.
In conclusion, higher age, high stroke severity, coma, hypertension, atrial fibrillation on admission were found to be independent predictors for in-hospital mortality after ischemic stroke in routine clinical care.Development of hemorrhagic transformation, cerebral edema, pneumonia, GIT bleeding during admission appears to be associated with poor outcome.However, the small number of cases and lower rate of mortality are limiting steps for generalization of results.Future largescale studies are recommended.

Financial and non-financial relations and activities of interest: None
Every patient was assessed on the clinical and radiological basis.The clinical assessment included socioeconomic data [e.g., age, sex, residence, marital status, occupation], clinical data [e.g., vascular risk factor, body temperature, blood pressure, pulse, and respiration rate] and state of consciousness [Glasgow coma scale, NIHSS score, and APACHEII score].
They were followed up during hospitalization.Females represented 53.1%.Venous thrombosis was reported among 14 patients [12 of them were females].However, arterial thrombosis was reported in 114 patients with nearly equal distribution of males and females.The majority of them were married [68.8%] [Table 1].Death was reported among 11 patients [8.6%].

Table [ 1
]: Demographic data of the studied patients Table[2]: Demographics, comorbid conditions and hospital events in relationship with mortality

Table [ 3
]: Scores of strokes at admission in relation to mortality

Table [ 4
]: Logistic regression for mortality prediction using complications, comorbidities, and demographic data Figure [1]: ROC curve for prediction of mortality among patients with strokeTable [5]: ROC curve characteristics of the logistic regression model