Perioperative stress response to carotid endarterectomy: the impact of anesthetic modality

Abstract

Objective: Surgery for extracranial carotid artery disease has been challenged by carotid angioplasty stenting because the latter is less invasive and avoids surgical trauma. In fact, the magnitude of the perioperative stress response evoked by carotid endarterectomy (CEA) has never been evaluated. Our aim was to determine the degree of surgical trauma caused by CEA and to define differences related to the use of locoregional or general anesthesia.

Methods: We prospectively studied 113 consecutive CEAs performed on 109 patients admitted at a community institutional center. Patients were stratified for demographics and risk factors and operated on under locoregional (LA) or general anesthesia (GA) depending on both the surgeon preference and patient's compliance. Selective carotid shunting was performed for patients who manifested neurologic deficits under LA or had stump pressure values </=30 mm Hg under GA. Markers of the stress response, including cortisol, adrenocorticotropic hormone, prolactin, and C-reactive protein, were measured intraoperatively, before and after carotid artery cross-clamping (CACC), and postoperatively up to the third day after surgery. Hemodynamic variability was assessed during surgery and for 24 hours postoperatively. Operative times were also measured. Surgeons were considered as independent variables for stress response. Statistics were run by means of nonparametric tests and univariate and multivariate analysis with a linear regression model.

Results: CEA was performed under GA in 63 cases (55.8%) and under LA in 50 (44.2%). The two groups were comparable in terms of demographics and risk factors. Intraoperatively, cortisol and adrenocorticotropic hormone levels were significantly higher in the LA group (both P <.001). CACC increased the intraoperative cortisol levels in both the GA (P =.019) and the LA groups (P =.006). However, in patients who underwent carotid shunting, this effect was abolished (GA group, P =.779; LA group, P = 1.0). During the early postoperative period there was no difference between the two groups. On postoperative day 1 the stress response was abolished in both groups. Prolactin levels increased intraoperatively in both the LA and GA groups and returned within preoperative values on postoperative day 1. Prolactin levels were higher in the GA group (P =.003 intraoperatively and P <.001 postoperatively). C-reactive protein significantly increased in both GA and LA groups on postoperative days 1 and 2 and started to decrease on day 3 with no differences between the two groups at any time. Hemodynamic variability and considered risk factors including individual surgeon were not significant variables. Gender-related differences were found only in prolactin secretion. The length of surgery had an impact for procedures that lasted >120 minutes. Three patients experienced an intraoperative neurologic event and had higher post-CACC cortisol values as compared to asymptomatic patients.

Conclusions: Intraoperative surgical stress was higher under LA and was blunted by carotid shunting under both LA and GA. Within 2 hours after surgery the anesthetic modality no longer had any impact on surgical trauma. The stress response to CEA, regardless of the type of anesthesia, was abolished within 24 hours. Intraoperative stress response, namely hypercortisolemia, directly correlated with subclinical and clinical cerebral hypoperfusion/ischemia during CACC. Hence, attenuation of the stress response to CEA might decrease the incidence of cerebral ischemic events.