Intro: Unclear onset time with delayed arrival is a common reason for non-treatment with tPA. The MR-WITNESS trial is studying if arrival 4.5 h from last known well, favorable clinical factors and FLAIR signal intensity ratios can safely select patients for IV tPA. We performed a retrospective analysis to compare patient outcomes between tPA patients treated within 4.5 h vs. potential MR WITNESS candidates who did not receive tPA.
Methods: Using our local GWTG-Stroke database we identified 515 patients from 1/2009 through 6/2015 who met study criteria. Patients were defined as potential candidates for MR WITNESS based on a subset of the trial inclusion criteria that were available in our dataset: they met onset time criteria (arrived >= 4.5 but <24 h from last known well, <4.5 h from symptom discovery), did not receive on-label IV tPA or thrombectomy, had NIHSS<=25, blood glucose 50-400 mg/dL and no current oral anticoagulant treatment. Because they did not get acute MR, imaging criteria could not be evaluated. Univariate and multivariable testing were performed to identify predictors of discharge home.
Results: In univariate testing compared to tPA patients, potential MR WITNESS candidates were less likely to arrive by EMS, more likely to have carotid stenosis, dyslipidemia and prior stroke/TIA, and had lower median NIHSS. Despite lower NIHSS, they were less likely to ambulate independently at discharge or be discharged home. On multivariable analysis, only potential MR Witness eligibility (aOR=0.37 (0.23,0.59); p-value=0.001) and lower NIHSS (aOR=0.86 per point (0.83,0.90); p-value<0.001) were independently associated with discharge home.
Conclusion: Patients who may be potential candidates for extended window IV tPA per MR WITNESS protocol have poor outcomes when untreated compared to patients who present early and are treated with IV tPA. Further research is warranted to determine if tPA treatment in this cohort can improve outcomes.