Clinical Outcomes of Acute Ischemic Stroke in Sudanese ICUs & HDUs: A Multicenter Study

Raja Mustafa ¹,   Ashraf Yousif ^2*  Mohammed Tayfor ³

Abstract

Doi : https://doi.org/10.5281/zenodo.17790830 

Acute ischemic stroke (AIS) requiring critical care carries high mortality, yet outcome data from low-resource settings remain scarce. We quantified short-term outcomes and determinants among AIS patients admitted to intensive care units (ICUs) and high-dependency units (HDUs) across major hospitals in Khartoum, Sudan.

Methods: Prospective, multicenter observational study across five hospitals (January–June 2023). Adults (≥18 years) with CT-confirmed AIS admitted to ICU/HDU were consecutively enrolled. Exposures included demographics, comorbidities, prior cerebrovascular disease, early enteral nutrition (EEN ≤48 h), complications, and CT territory. Primary outcome was in-hospital mortality; secondary outcomes were hospital length of stay (LOS) and complication profile. Associations were assessed using χ²/Fisher’s exact tests and t-tests (α=0.05).

Results: Seventy-two patients were included (61.1% female; mean age ~65 years). Comorbidities were frequent: hypertension 62.5%, diabetes 44.4%, cardiac disease 30.6%, and renal disease 31.9%; 26.4% had prior stroke. EEN was provided to 84.7%. Median LOS was 7 days (IQR not computed); mean 9.74±12.04 days; 87.5% were discharged or died within ≤14 days. In-hospital mortality was 63.9% (46/72). Complications were common: stroke-associated pneumonia 41.7%, mechanical ventilation 38.9%, coma 23.6%, UTI 18.1%, pressure ulcers 15.3%, malaria 13.9%. On univariate analysis, renal disease was the only factor significantly associated with mortality (death in 82.6% with renal disease vs 55.1% without; χ²=5.133, p=0.023). Sex, age group, diabetes, hypertension, cardiac disease, smoking, prior stroke/TIA, and EEN showed no significant association with mortality. Prior stroke predicted longer hospitalization (mean 15.37 vs 7.72 days; p=0.016); no other covariates, including EEN, significantly affected LOS. CT most often showed MCA-territory involvement; infarct territory was not associated with mortality. No patient received thrombolysis or thrombectomy.

Conclusions: AIS requiring ICU/HDU care in this Sudanese cohort had very high mortality (64%) and

a heavy complication burden. Renal disease identified a particularly high-risk subgroup for death, while prior stroke prolonged LOS among survivors. Widespread EEN use showed no detectable association with survival or LOS. Findings underscore urgent system-level needs earlier presentation, availability of reperfusion therapy, dedicated stroke units and meticulous critical-care practices (infection prevention, renal optimization, nutrition protocols) to improve outcomes in resource-limited settings.