Objective: Precision is a fundamental requirement for quantitative monitoring of physiological variables. The aim of this study was to assess the precision of the tissue oxygenation index (TOI) measured using spatially resolved spectroscopy by comparing simultaneous measurements of TOI in a cohort of preterm infants.
Methods: Twenty-six infants born at a median (range) postmenstrual age of 26 weeks (23–32) were studied. Their median (range) birthweight was 778 g (504–1530) and age at study was 15 h (7–72) after birth. TOI data were collected using a dual channel NIRO 300 spectrophotometer (Hamamatsu Photonics, KK, Japan). For each infant the pair of optodes was placed on either side of the infant’s head. Simultaneous 20-minute measurements from each hemisphere were compared using Bland–Altman bias analysis.
Results: There was poor agreement between measurements in each infant. The bias/precision was −0.12% and 95% limits of agreement between −13.1 to +12.8% (see figure).
Discussion: Discrepancy between left and right hemispheres could be attributed to differences in surface geometry and tissue inhomogeneity. The validity of the TOI measurement is limited by the lack of precision. This questions the value of TOI in providing a quantitative measurement of cerebral oxygenation in the preterm brain.