Prehospital blood pressure in acute stroke.

Abstract

Background: Blood pressure (BP) is a potential target for early stroke treatment and triage, yet our knowledge about prehospital BP is limited. We aimed to investigate whether prehospital BP differs by stroke type and severity in patients with suspected acute stroke and explore associations between prehospital BP and outcomes.

Methods: This single-centre retrospective study analysed consecutive patients with suspected stroke admitted to a primary hospital within 4.5 hours of symptom onset, comparing prehospital BP by discharge diagnosis, ischaemic stroke subtypes and stroke severity. Associations between prehospital BP and 90-day modified Rankin Scale and National Institutes of Health Stroke Scale (NIHSS) score changes within 24 hours of admission were analysed using logistic and ordinal regression.

Results: Of the 1165 patients included, 64% (742) had acute stroke and 36% (423) were stroke mimics. Patients who had an acute stroke had higher mean prehospital systolic BP (SBP) than mimics (162±31 mm Hg vs 152±29 mm Hg, p<0.001). Patients with intracerebral haemorrhage (ICH) had the highest SBP (173±34 mm Hg). Among ischaemic strokes, lacunar stroke had higher SBP than non-lacunar stroke (169±33 vs 158±29 mm Hg, p=0.002), and large vessel occlusion (LVO) had lower SBP than non-LVO (150±28 vs 164±30 mm Hg, p<0.001). Patients with NIHSS score >15 had lower SBP than patients with NIHSS score ≤5 (155±34 vs 165±29 mm Hg, p=0.009). In ischaemic stroke, higher SBP was associated with less chance of early neurological improvement (adjusted OR 0.95, (95% CI, 0.91 to 0.98)).

Conclusion: Prehospital BP varies by stroke subtype and severity and is associated with outcome. Blood pressure was higher in ICH than in ischaemic stroke, with lacunar infarcts showing the highest BP among ischaemic subtypes. The association between prehospital BP and poor outcomes in observational studies highlights the need for further research on early BP management across different stroke subtypes.