ABSTRACT
BACKGROUND: Stroke registries are pivotal to the monitoring and improvement of the quality of stroke care. We report data from the initial phase of a nationally representative hospital-based stroke registry in Sri Lanka. METHODS: Based on an observational cohort design, all consecutive patients aged ≥18 years with a diagnosis of stroke or transient ischemic attack (TIA) presenting to six tertiary-care hospitals in the Western, Eastern, Southern, Northern and Central provinces of Sri Lanka within 14 days of onset were recruited. RESULTS: During a period of 14 months, 5893 patients with a stroke/TIA (58.8% men; mean age 65.22 years, SD=13.28) were entered into the database; 69.8% (n=4111) had an ischaemic stroke (IS); 20.9% (n=1233) had a haemorrhagic stroke (HS); 7.2% (424) had a TIA; and 2.1% (125) had a venous stroke. While IS were more common among women (71.7% vs 68.4%; p=0.006), HS were more common among men (22.3% vs 19.0%; p=0.003). Hemiparesis (86.2% vs 83.2%; p=0.011), headache (29% vs 11.6%; p<0.001), seizures (5.9% vs 4.2%; p=0.013), sphincter dysfunction (11.8% vs 7.7%; p<0.001) and hypertension (72% vs 67.3%, p=0.002) were more common in HS while dysphasia (63.2% vs 50.0%; p<0.001), ataxia (9.6% vs 7.3%; p=0.014), sensory disturbances (10.8% vs 6.0%; p<0.001) and diabetes mellitus (37.9% vs 28.1%; p<0.001) were more common in IS. Thrombolysis was administered in only 2.1% while only 14.6% had access to stroke units. The mean door-to-needle time was 100.2 (SD=46.0) minutes. The hospital mortality rate was 8.3%. Recurrent strokes within three months were more common in HS than IS (3.2% vs 1.6%; p=0.007). CONCLUSIONS: While stroke characteristics in Sri Lanka are similar to that of other populations, the rate of thrombolysis and access to stroke units are inadequate. The stroke registry provides useful data for the appraisal and improvement of stroke services.
