Abstract Details

We propose an acute treatment protocol for patients with fluctuating exam secondary to salvageable micropenumbra in high risk small vessel disease (SVD)-related acute ischemic stroke (AIS).

Inclusion criteria included acute SVD-related stroke with NIHSS fluctuation ≥ 2 previously described as a sensitive indicator of neurological deterioration in SVD ischemic stroke. Patients with large vessel atheroma secondarily causing stroke were excluded. Treatment protocol consisted of albumin IV, eptifibatide IV, magnesium sulfate IV, cilostazol PO, normoglycemia, normothermia, aggressive fluid resuscitation, and targeted blood pressure parameters in neurocritical care setting. Protocol was prospectively initiated August 2020 at onset of exam fluctuation and continued until plateau NIHSS was reached. Retrospective data for subjects from June 2016 to July 2020 was collected for historical controls. Primary outcomes measures included safety and early efficacy end points. Efficacy was measured by change from maximum NIHSS to plateau NIHSS (NIHSS_diff). Wilcoxon rank-sum test was used to evaluate significance difference of NIHSS_diff in both groups. Social Science Statistics was used for data analysis.

From January 2017 - May 2021, out of 7,146 AIS patients, 30 met selection criteria. From August 2020 to May 2021, consecutive subjects received treatment protocol (n=15, baseline NIHSS 4.93, 95% CI [3.3572, 6.5028]). They were compared with historical controls (n=15, baseline NIHSS 5.87, 95% CI [2.9407, 8.7993]). There was no significant difference in baseline characteristics (p=1; U=112 at p<0.05; z-score 0). In the SVD group, no subjects had adverse events leading to early termination. Early efficacy as suggested by NIHSS_diff between groups was statistically significant (p=0.00228; U=38.5; z-score 3.04864; 80% power at p<0.05). 

To our knowledge, this is the first systematic demonstration of safety and early efficacy of multimodal intervention for acute SVD-related ischemic stroke. Larger randomized trials using concurrent controls are required to corroborate our findings.