The non-invasive monitoring of cerebral tissue oxygenation.

The instrument drift was found to be less than 0.004 OD/hour and from measurements on glass filters of 8 optical density units, a coefficient of variation of 0.01 over the 30 second averaging time was observed. The instrument is sufficiently sensitive to enable monitoring of changes in the cerebral oxygen saturation level of haemoglobin and to enable changes in the concentration of cyt aa3 (oxidised form), to be measured with reasonable confidence. It is of the utmost importance in NIRS investigations to be certain of the specificity of the technique, and it is vital that reliable determinations of the amount of cytochrome aa3 in brain are made in addition to measurements of the extinction coefficients. This is still a matter of considerable debate, not only in the isolation and properties of the multisubunit structured membrane bound protein, which many enzymologists have been investigating for the last 50 years, (Keilin and Hartree, 1939; Brunori et al, 1981) but also in relating to in-vivo versus in vitro comparisons. An additional point for consideration is the validity of employing multiplier coefficients derived from the rat brain, as several groups have done (Wyatt et al 1986, Ferrari et al 1985,) and applying them to the human brain and to the infant brain. There may be significant differences in the activity of the enzyme, and the profound physiological effects which will arise during the end point fluorocarbon studies: the presence of fetal haemoglobin must also be considered (Carta et al., 1987). The clinical determination of the optical path length is a considerable problem (Cope etal., 1988,) and once again estimates of the pathlength correction factor made in animals and dead fetuses may not be valid for living human tissue. Results from our animal studies indicate that NIRS can be used to monitor changes in the oxygenation level of Hb and, in extreme hypoxia, changes in the level of the redox state of cyt aa3 can be reliably measured and are well within the sensitivity of the instrument. The results indicate that under small change in saturation the redox state of cyt aa3 appears to be unaltered. It may be that under normal physiological conditions the redox state of aa3 appears to be apparently unchanged under episodes of mild hypoxia of short duration.