Regional cerebral oxygenation by NIRS does not correlate with central or jugular venous oxygen saturation during interventional catheterisation in children

• Published in: Acta anaesthesiologica Scandinavica Vol. 52; no. 10; pp. 1370 - 1374
• Main Authors: KNIRSCH, W., STUTZ, K., KRETSCHMAR, O., TOMASKE, M., BALMER, C., SCHMITZ, A., BERGER, F., BAUERSFELD, U., WEISS, M.
• Format: Journal Article
• Language: English
• Published: Oxford, UK Blackwell Publishing Ltd 01-11-2008; Blackwell
• Subjects: Adolescent; Anesthesia; Anesthesia. Intensive care medicine. Transfusions. Cell therapy and gene therapy; Biological and medical sciences; Brain - blood supply; Catheterization, Central Venous - methods; Child; Child, Preschool; children; Female; Humans; Infant; Jugular Veins; Linear Models; Male; Medical sciences; near-infrared spectroscopy; Oximetry - statistics & numerical data; Oxygen - blood; Oxygenation; Sensitivity and Specificity; Spectroscopy, Near-Infrared; Vena Cava, Superior; Human; Oxygen; near-infrared spectroscopy; children; Saturation; Anesthesia; Oxygenation; Child; Encephalon
• ISSN: 0001-5172; 1399-6576
• DOI: 10.1111/j.1399-6576.2008.01703.x

Purpose: To evaluate whether regional cerebral oxygenation (rSO2) by near‐infrared spectroscopy correlates with central venous (SvO2) or internal jugular (SjO2) oxygen saturation, and whether changes over time (Δ) in rSO2 (ΔrSO2) predict changes in SvO2 (ΔSvO2) and SjO2 (ΔSjO2).
Methods: The rSO2 values were measured using the INVOS 5100 cerebral oximeter in children undergoing interventional cardiac catheterization and were compared with the oxygen saturation of analysed central venous and internal jugular blood samples. Changes over time (Δ) were calculated as the difference between the values before and after catheter intervention for rSO2·(ΔrSO2), SvO2·(ΔSvO2) and SjO2·(ΔSjO2). Simple regression and Bland–Altman analysis were performed. Data are presented as median (range).
Results: Sixty patients aged 4.3 (0.2–16.0) years were investigated. A closer correlation was found between rSO2 and SvO2 (r=0.728, P<0.0001) than between rSO2 and SjO2 (r=0.665, P<0.0001). The bias between rSO2 and SvO2·(SjO2) was 0.17% (−0.60%), with limits of agreement from −15.5% to 15. 9% (−18.6–17.4%). The sensitivity/specificity for ΔrSO2 to indicate a fall in SvO2 or in SjO2 was 70.3%/65.2% and 68.6%/60.0%, respectively.
Conclusion: Neither absolute values nor changes in rSO2 using the INVOS 5100 allowed reliable estimation of SvO2 or SjO2 and their trends.