Differences in subacute motor recovery after intracerebral haemorrhage and ischaemic stroke: Analysis using the VISTA database cohort.

Background: Motor impairment is a significant contributor to disability after stroke, but recovery is often incomplete. Whether motor recovery differs between intracerebral haemorrhage (ICH), a subgroup of stroke with far worse outcomes, and ischaemic stroke is not clear.

Methods: We performed a retrospective observational longitudinal cohort study using individual patient-level data from the Virtual International Stroke Trials Archive (VISTA) database (ICH n=892, ischaemic stroke n=6912). Differences in motor recovery to 90-days were examined between ICH and ischaemic stroke patients with mixed effect regression models adjusted for a priori determined confounders. Motor weakness was measured by NIHSS face, arm and leg sum with secondary analyses of total NIHSS, and NIHSS language score.

Results: Recovery was observed in all NIHSS domains for both stroke types to 30-days (NIHSSb=-2.78, 95%CI -2.89,-2.68; NIHSSb=-5.74, 95%CI -5.92,-5.56; NIHSSb=-0.28 95%CI -0.31,-0.24) and 90-days (NIHSSb=-3.62, 95%CI -3.69,-3.54; NIHSSb=-7.17, 95%CI -7.30,-7.05; NIHSSb=-0.74, 95%CI -0.78,-0.71). Baseline impairment between groups was well matched with only motor impairment being slightly greater in ICH; NIHSS mean(SD)=13.0 (5.3) vs 12.3 (5.4). To 30-days the extent of recovery was not different between stroke types but recovery to 90-days was greater in ICH for motor and statistically significant for total NIHSS score (b=-0.35, 95%CI -0.71,-0.002). Ischaemic stroke survivors recovered more in NIHSS language domain.

Conclusions: Timing and extent of recovery is different between stroke types. Motor recovery in ICH is greater and occurs later. Therefore, the assumption that most recovery occurs within 30-days and proportionality of recovery should be revisited in this population.