Objectives
Carotid endarterectomy (CEA) is usually performed under general anesthesia (GA) although some advocate regional anesthesia (RA) to reduce hemodynamic instability, allow neurologic monitoring and selective shunting. RA does not reduce risk of periprocedural stroke or death, although some series show a reduction in myocardial infarction (MI). We investigated the association of anesthesia type and periprocedural MI among patients receiving GA or RA for CEA and patients undergoing carotid artery stenting (CAS) in the Carotid Revascularization Endarterectomy versus Stenting Trial (CREST).
Methods
Between 2000 and 2008, 1151 patients underwent CEA (anesthetic type available on 1149 patients) and 1123 patients underwent CAS within 30 days of randomization in CREST. CEA patients were categorized by anesthetic type (GA vs RA). CREST defined protocol MI as chest pain or ECG change plus biomarker evidence of MI and total MI was defined as protocol MI plus biomarker-positive (biomarker +) only MI. The incidence of protocol MI and total MI were compared between patients undergoing CEA under GA, RA and those undergoing CAS. Other study endpoints were similarly compared. Differences in baseline characteristics and periprocedural events were also evaluated among the three groups. Logistic regression was used to assess group differences adjusting for age and symptomatic status.
Results
The three groups had similar demographic risk factors except for prevalence of symptomatic carotid stenosis, which was lowest in the CEA-RA group (P=.03). Of the 111 patients in the in the CEA-RA group, there were not any protocol MI’s and only 2 biomarker + only MI’s for an overall incidence of 1.8%, similar to the 1.7% overall incidence seen in patients undergoing CAS. In contrast, the combined incidence of protocol and biomarker + MI’s in the 1038 patients in the CEA-GA group was significantly higher at 3.4% (P=.04), twice the risk of protocol MI and biomarker + only MI when compared to those undergoing CAS (OR, 2.01; 95%, CI 1.14-3.54). Direct comparison of the MI incidence between CEA-RA and CEA-GA showed no statistical difference. Patients undergoing CEA under GA had lower odds of a periprocedural stroke (OR, 0.48; 95% CI 0.28-0.79) and stroke or death (OR, 0.46; 95% CI 0.27-0.76) when compared to those undergoing CAS but were not significantly different than those undergoing CEA-RA.
Conclusions
In CREST, patients undergoing CEA under RA had similar risk of periprocedural MI as those undergoing CAS while the risk for CEA-GA was twice that compared to patients undergoing CAS. Nevertheless, since periprocedural MI is one of the few variables favoring CAS over CEA, and has been associated with decreased long term survival, RA should be seriously considered for patients undergoing CEA.