Near infrared spectroscopy monitoring during carotid endarterectomy: Which threshold value is critical?

Objectives. Retrospectively to verify which decreasing percentage in regional oxygen saturation (rSO(2)) identified patients with good collateralisation during carotid artery cross clamp. Materials and methods. During 594 endarterectomies under general anaesthesia the decreasing percentage from preclamp value to value detected in the first 2 min after clamping the CCA and/or ICA was calculated in real time. No temporary shunt was placed in any case. ROC analysis was performed to determine the optimal cut-off for rSO(2) decrease to identify the occurrence of neurological complications. Results. A cut-off of 11.7% was identified as optimal. Sensitivity and specificity were 75% (95% CI 71-78) and 77% (95% CI 74-80), respectively. The cut-off of 20% had a lower sensitivity (30%) and a higher specificity (98%) to identify patients with complications, with positive and negative predictive value of 37 and 98%, respectively. Conclusions. The study suggest that a relative decrease in rSO(2) of < 20% from preclamp to early cross clamp value has a high negative predictive value, i.e. if rSO(2) does non decrease more than 20%, ischemia by hypoperfusion is unlikely and a shunt should not be necessary. Moreover, a relative decrease > 20% may not always indicate intraoperative neurological complications.