Do cerebral and somatic tissue oxygen saturation measurements correlate with each other during surgery?

Intraoperative maintenance of optimal tissue oxygenation is critical; however, it is uncertain whether measurements of different tissue beds correlate with each other. Cerebral tissue oxygen saturation (SctO) measured on the forehead and somatic tissue oxygen saturation (SstO) measured on limbs, using a tissue near-infrared spectroscopy, were simultaneously recorded every 2 s in patients having spine surgery or robotic hysterectomy. Simple linear regression was used to determine the static correlation between SctO and SstO using the median values of each min for each patient. The dynamic correlation between SctO and SstO was assessed by Pearson's correlation coefficient (CC) for each non-overlapping 2-min epoch. In patients having spine surgery (n = 99), SctO and SstO (mean ± SD) were 69.8 ± 4.9% and 75.5 ± 8.7%, whereas in patients having robotic hysterectomy (n = 106), the corresponding values were 74.9 ± 6.8% and 83.7 ± 6.2%. The static correlation between SctO and SstO was inconsistent (r ranging from - 0.86 to 0.93 in spine surgery and from - 0.74 to 0.85 in robotic hysterectomy). The proportional durations with CC ≤ - 0.3 (negative correlation), - 0.3 < CC < 0.3 (poor correlation) and CC ≥ 0.3 (positive correlation) were 18.3 ± 9.6%, 52.6 ± 12.1% and 29.0 ± 9.6%, respectively, in patients having spine surgery and 19.6 ± 9.0%, 58.6 ± 13.1% and 21.8 ± 8.0%, respectively, in patients having robotic hysterectomy. There are a large discrepancy and inconsistent correlation between intraoperative SctO and SstO measurements, suggesting their non-interchangeability.