Letter by Dexter and Hindman Regarding Article, “Anesthesia Technique and Outcomes of Mechanical Thrombectomy in Patients With Acute Ischemic Stroke”
To the Editor:
Bekelis et al1 used instrumental variable analysis to model the association between the use of general anesthesia or conscious sedation for mechanical thrombectomy and the incidence of hospital mortality among patients with acute ischemic stroke.
A potential limitation of the analysis is that the administrative data used lacked some patient characteristics that predict mortality (eg, National Institutes of Health stroke scale and vascular distribution of the stroke). Suppose that most studied hospitals used almost exclusively 1 type of anesthesia (ie, either 95% to 100% or 0% to 5% general anesthesia). Then, the unaccounted for patient characteristics that determine outcome would not have influenced the decisions of type of anesthesia (ie, not biased …