Comparison of cerebral blood flow, its reliability and its carbon dioxide response in neonates in three centers: a near-infrared spectrophotometry study
M. Wolf et al., Comparison of cerebral blood flow, its reliability and its carbon dioxide response in neonates in three centers: a near-infrared spectrophotometry study, PRENAT N M, 5(5), 2000, pp. 329-335
Although all near-infrared spectrophotometry (NIRS) instruments rely on the
same physical principles, the implementation varies considerably. Therefor
e, the comparability of the cerebral blood flow (CBF) measurements using NI
RS is questionable. Th ree centers decided to study the well-known physiolo
gical response of CBF to a change in pCO(2). The following questions were t
he aim of the study: how reliable are the CBF measurements for each center,
and how well do the CBF, its reliability and the CO2 response of CBF compa
re between centers? Twelve infants were enrolled. Their mean gestational ag
e was 29.2 (SD 3.5) weeks, their mean birth weight was 1366 (SD 803) g and
their mean postnatal age upon measuring was 3.3 (SD 3.7) days. The global m
ean CBF was 12.0 (SD 4.3) ml/100 g per min. In Copenhagen the mean CBF was
10.9 ml/100g per min, in Nijmegen 12.9 ml/100 g per min and in Zurich 12.2
ml/100 g per min. The global mean of the test-retest variability (TRV) of t
he CBF measurements was 34.7 (SD 18.6)%. The mean TRV was 33.1% for Copenha
gen, 37.1% for Nijmegen and 29.3% for Zurich. The mean CO2 response was 52.
1 (SD 53.6) %/kPa. The global mean of the CO2 response was 98%/kPa for Cope
nhagen, 15.5%/kPa for Nijmegen and 42.7%/kPa for Zurich. Analysis of varian
ce showed no significant difference between the centers for any of the ment
ioned parameters. The intercenter variability was 0% (upper limit 34.5%) fo
r CBF, 0% (upper limit 69.1%) for TRV and 68.7% (upper limit 353.7%) for th
e CO2 response. Further analysis of the data showed a significant positive
correlation between the CO2 response and the positive end-expiratory pressu
re. This study proved that the CBF and TRV Values were similar between thre
e clinical centers. Therefore, the effect of different NIRS instruments amo
ng centers is small. The TRV was higher than in previous studies. Among the
centers there was no statistically significant difference in CO2 response.