Association between regional oxygen saturation and central venous saturation in pediatric patients undergoing cardiac surgery: A prospective observational study

Abstract

AbstractBackgroundWe evaluated the correlation between regional oxygen saturation (rSO2) in the frontal and right renal dorsum (cerebral rSO2 and somatic rSO2) measured by near‐infrared spectroscopy (INVOS™ 5100C, Medtronic) and central venous oxygen saturation (ScvO2) measured with a fiber‐optic oximetry catheter (PediaSat™, Edwards Lifesciences) during surgery in order to determine whether noninvasive rSO2 could be used as an alternative to ScvO2 in pediatric cardiac surgery patients. We evaluated the correlation between regional tissue oxygen saturation (cerebral rSO2 and somatic rSO2) measured by near‐infrared spectroscopy and other patient measures with central venous oxygen saturation (ScvO2) measured with a fiber‐optic oximetry catheter to track global oxygen supply demand as a potential alternative or supplement to ScvO2.Patients and MethodsThis single‐center prospective observational study enrolled 33 children (weight < 10 kg) who underwent cardiac surgery for congenital heart disease between February 2018 and November 2021. ScvO2, cerebral rSO2, and somatic rSO2 were recorded simultaneously after anesthesia induction and central venous catheter placement. Pearson's correlation coefficient and Bland–Altman analysis were used to determine the relationship between ScvO2 and rSO2. We conducted correlation, Bland Altman, and multiple regression analyses to identify associations between rSO2, patient measures, and ScvO2 values.ResultsThe patients' median age was 11.0 (quartile 2.0–16.0) months. Their weight was 7.2 (quartile 4.5–9.2) kg. Cerebral rSO2 was significantly positively correlated with ScvO2 (r2 = 0.29, p = .002 in all patients; r2 = 0.61, p = .013 in the patients without mixing at the atrial level), whereas somatic rSO2 was not. The Bland–Altman analysis demonstrated biases [95% confidence interval; 95% CI] (lower and upper limits of agreement [95% CI]) of 0.27% [−4.26 to 4.80] (−24.79 [−32.61 to −16.96] to 25.33 [17.50 to 33.16]) between cerebral rSO2 and ScvO2 and 0.91% [−5.48 to 7.30] (−34.43 [−45.47 to −23.39] to 36.25 [25.21 to 47.29]) between somatic rSO2 and ScvO2. Preoperative brain natriuretic peptide (BNP) and SpO2 were independent variables associated with ScvO2 and cerebral and somatic rSO2.ConclusionCerebral rSO2, SpO2, and BNP were significantly correlated with ScvO2, although the cerebral rSO2 correlation was greater for lesions without atrial mixing. rSO2, BNP, and SpO2 might be used to track changes in ScvO2 but cerebral rSO2 is not sufficiently precise to replace it.