Feasibility of long-term cerebral and peripheral regional tissue oxygen saturation measurements

The aim of this study was to analyse the feasibility of long-term measurements of cerebral (crSO2) and peripheral (prSO2) regional tissue oxygen saturation on the first day of life by determining the amount of artefacts and their influence on rSO2. Near infrared spectroscopy (NIRS) measurements were performed fronto-parietal left (crSO2) and on the right forearm (prSO2). Arterial oxygen saturation (SpO2) was measured by pulse oximetry on the right wrist. Three criteria (C) were defined to identify artefacts (C1: missing values, C2: rSO2 jumping >15%, C3: rSO2 ≥ SpO2). The number of artefacts as a percentage of measurement time and mean rSO2 was calculated after the introduction of each criterion. Measurements were performed in 40 neonates. The number of artefacts in crSO2 measurements was similar after introduction of C1 (7.37 ± 4.64%) and after introduction of all criteria (8.89 ± 4.59%). The number of artefacts in prSO2 measurements after introduction of C1 was 10.83 ± 4.21%, and after introduction of all criteria significantly higher with 17.78 ± 4.27%. After introduction of C1, further criteria did not significantly change rSO2: crSO2 (78.6 ± 1.3% versus 78.5 ± 1.2%) and prSO2 (83.7 ± 0.9% versus 83.5 ± 0.9%). In conclusion, long-term NIRS measurements of crSO2 and prSO2 are feasible, since most artefacts are due to missing values and therefore easy to recognize.