Work

Diagnostic and therapeutic gaps in recurrent stroke management: real-world insights from a Ghanaian hospital-based registry

Year: 2026

Type: article

Abstract: BACKGROUND: Recurrent strokes account for up to 1 out of 4 strokes in Low-and-Middle-Income-Countries (LMICs), culminating in worsening of outcomes. International guidelines recommend subtype-specific etiological workup to help tailor strategies for prevention of recurrence after an index stroke. However, huge gaps in secondary stroke prevention persist in LMICs. OBJECTIVES: We aimed to assess the extent of diagnostic evaluation (for risk-factor and underlying etiological identification) and treatment approaches among recurrent stroke patients in a Ghanaian tertiary hospital. METHODS: This was a cross-sectional study among recurrent stroke cases admitted to Komfo Anokye Teaching Hospital between 2023 and 2025. The data was analyzed using SPSS Version 28. RESULTS: Recurrent strokes accounted for 17.8% of all stroke admissions over 2 years, with ischemic strokes comprising 76.2% of these recurrent events. Hypertension was the most dominant modifiable risk factor among both ischemic and hemorrhagic subtypes, 97.0% and 92.1% respectively, (p = 0.297). Only a minority of ischemic stroke patients (12.6%) had sufficient diagnostic evaluation to allow definitive etiological classification according to the TOAST framework. Only 7.9% of hemorrhagic strokes underwent Computed tomography angiography (CTA)/ Magnetic resonance angiography (MRA) to evaluate for structural causes. Most ischemic stroke patients were prescribed a statin (93.7%) and an antiplatelet, while anticoagulant therapy was used only in a minority of patients (3.91%). Carotid endarterectomy or carotid stenting was not performed for any of the ischemic stroke patients. CONCLUSION: Our findings highlight significant gaps in recurrent stroke evaluation and management in the Ghanaian setting. Standardized algorithms for secondary stroke prevention and treatment are urgently needed to mitigate recurrence.

Source: BMC Neurology

DOI: 10.1186/s12883-026-05013-1