Outcome of Near-Infrared Spectroscopy-Guided Selective Shunting During Carotid Endarterectomy in General Anesthesia.

ABSTRACT

BACKGROUND: To analyze the outcome of near-infrared spectroscopy (NIRS)-guided selective shunting during carotid endarterectomy and the procedural outcome. METHODS: In this retrospective single-center study, patients undergoing carotid endarterectomy in general anesthesia and receiving bihemispheric NIRS as single neuromonitoring tool between January 2009 and January 2014 were included. Shunting was applied if the reduction in the NIRS values after cross-clamping on the ipsilateral side exceeded 15%. Patients with contralateral occlusion of the internal carotid artery (ICA) were excluded, as were patients operated on by surgeons performing routine shunting. All patients underwent intraoperative angiography after vessel recanalization. RESULTS: NIRS trend was available in 441 patients. Twenty-eight were excluded from this study (14 due to preference for general shunting, 13 due to contralateral ICA occlusion, and 1 due to intraoperative ICA occlusion), resulting in a final sample of 413 patients. We observed a >15% drop in NIRS values on the ipsilateral side in 29 (7%) patients. Accordingly, an intraluminal shunt was placed into the ICA. Shunting was not performed in 384 patients (<15% drop in NIRS values). Interestingly, the NIRS values on the contralateral side were significantly elevated after cross-clamping compared with baseline in the group without shunt (P < 0.0001). On the contrary, patients requiring an ICA shunt revealed a statistically significant reduction in the rSO2 on the contralateral side compared with the baseline (after ipsilateral clamping) (P = 0.047). Three patients overall suffered a stroke, all of whom were in the no-shunt group (combined stroke rate of 0.8% [3/384] with no significant intergroup difference). There was no difference in morbidity factors between the two groups. However, surgical revision after intraoperative angiography was significantly more frequent in the shunt group (17.2%, 5/29) versus the no-shunt group (6%, 23/384), (P < 0.037). CONCLUSIONS: An NIRS-guided selective shunting strategy was associated with excellent clinical outcomes and has the potential to identify patients at risk for hypoperfusion during the clamping period. However, a potentially shunt-associated higher rate of requiring local revisions (due to flaps, twisting, stenosis, and kinking) in ICA was observed. Additional studies are needed to further refine cut-off values for NIRS, indicating the need for shunting.