Abstract SNACC-7

Return to Poster Listing

During endovascular stroke treatment, controlling for baseline stroke severity with ASPECTS eliminates the advantage of sedation over general anesthesia

Kearns M, Nogueira R, Gershon R, Whalin M
Emory University, Atlanta, GA, USA

Several retrospective studies of endovascular stroke therapy have found a correlation between sedation and good neurologic outcome. The major limitation of these studies is that the choice of anesthesia was not randomized so there could be a selection bias for severely injured patients to receive general anesthesia. Most studies to date have used the NIH stroke scale (NIHSS) as a variable in their regression models to control for stroke severity. Review of data from the first two years of our stroke program suggests that the Alberta Stroke Program Early CT Score (ASPECTS), an imaging measure of stroke severity, is more strongly associated with neurologic outcome than was NIHSS. Furthermore, inclusion of ASPECTS in our regression model eliminates the outcome benefit of sedation compared to general anesthesia.

We studied 216 consecutive patients treated for anterior circulation strokes, of whom 133 received general anesthesia and 83 received sedation with dexmedetomidine (DEX). Good outcome was defined as a modified Rankin score of two or less 90 days after the stroke. After comparing hemodynamic, demographic, and angiographic variables in a univariate analysis we constructed a binary regression model with age, successful reperfusion, NIHSS, symptomatic hemorrhage, and DEX sedation. In this model sedation with DEX carried an odds ratio for good outcome of 2.13 (95% CI of 1.09 to 4.18, p=0.027). For the subset of 186 patients for whom ASPECT scores were available we found that ASPECTS > 7 carried an OR for good outcome of 4.59 (95% CI of 2.1 to 10.03, p<0.001). With ASPECTS > 7 in the model, the OR for DEX sedation dropped to 1.48 and was no longer statistically significant (95% CI of 0.70 to 3.12, p=0.306). We also observed a correlation between lowest recorded intraprocedural blood pressure and outcome (see Figure).

In conclusion, adding ASPECTS to our risk model appeared to eliminate any outcome difference between general anesthesia and dexmedetomidine sedation. Our initial set of 186 patients may have been underpowered, so we are seeking to validate our results in an updated population of over 500 endovascular stroke cases. Future prospective trials of anesthesia type should include ASPECTS in addition to NIHSS to control for stroke severity.

  • SNACC-7 Image 1

Back to Top