Robotic telestroke model takes longer but is as safe as standard care

Patients receiving telestroke care may wait as much as 18 minutes longer for treatment, but safety outcomes may be as good as standard, vascular neurologist stroke alert care.

Published in the March issue of Telemedicine and e-Health, these findings could influence how care is provided to patients in areas where a vascular neurologist may not be immediately available, such as rural areas. Cumara B. O’Carroll, MD, MPH, from the neurology department at the Mayo Clinic in Phoenix, and colleagues retrospectively assessed care at the primary stroke center between 2009 and 2012. The Mayo Clinic’s robotic telepresence stroke alert assessments began in 2009.

They compared 98 eligible telestroke patients to 98 patients who received standard care over the same period. While the primary outcome of interest was length of time to either decision to treat or downgrade, O’Carroll et al also looked at how long alert-to-needle times were for those treated, morbidity, mortality and discharge rates.

Time from alert activation to disposition of either treatment or downgrade was 8.6 minutes longer in patients assessed with the robotic telestroke method. More patients also were found to have acute stroke in the telestroke group compared with the standard group, 77 percent vs. 60 percent, respectively. Controlling for this factor reduced the difference between the two groups to 7.7 minutes. Mean time from alert to either treatment or downgrade was 8.9 minutes longer for patients experiencing symptoms of acute stroke as opposed to those who did not have acute stroke in the telestroke group. Among patients receiving standard care, this difference was 3.5 minutes.

More patients in the telestroke group required intravenous thrombolysis (23 percent vs. 13 percent). Among acute ischemic strokes, more patients were treated with thrombolysis if they were in the telestroke group as opposed to the standard group (37 percent vs. 24 percent, respectively). For patients in the telestroke group, 74 percent received treatment in under three hours. Comparatively, among the standard group, 92 percent had alert-to-needle windows of less than three hours.

However, they found that hospital stays for either group were on average four days. Among patients without acute stroke, 91 percent were discharged home by telestroke as opposed to 74 percent among standard care. Acute stroke home discharge rates were 27 percent and 46 percent, respectively, for the telestroke and standard care groups. Because of the higher rate of acute ischemic strokes among patients tended by telestroke, fewer overall were discharged home compared to standard care (42 percent vs. 57 percent, respectively).

Mortality rates were slightly higher among patients hospitalized by the telestroke intervention (8 percent vs. 6 percent), although O’Carroll et al noted that this was within an acceptable range. Rates of thrombolysis protocol violations and intracranial hemorrhagic complications were low and not significantly different between both groups.

O’Carroll et al wrote that in part, these numbers may be driven by the inexperience of residents in the telestroke protocol in the early days of its inception. This, they suggested, may have skewed the data somewhat and further analysis is under way to determine if care has improved since. “Outcomes from robotic telepresence acute stroke supervisory assessments were good and approached the gold standard,” they wrote.

“Robotic telepresence may be preferable in situations where no stroke specialist is available in house, especially for middle of the night and weekend staffing of residents, when a 9-minute robot-associated delay is likely better than the delay associated with the supervising physician driving to the hospital.”