Roger Springett

Quantitative near infrared spectroscopy measurement of cerebral hemodynamics in newborn piglets.

Severely premature infants are often at increased risk of cerebral hemorrhage and/or ischemic injury caused by immature autoregulatory control of blood flow to the brain. If blood flow is too high, the infant is at risk of hemorrhage, whereas too little blood flow can result in ischemic injury. The development of a noninvasive, bedside means of measuring cerebral hemodynamics would greatly facilitate both diagnosis and monitoring of afflicted individuals. It is to this end that we have developed a near infrared spectroscopy (NIRS) system that allows for quantitative, bedside measurement of cerebral blood flow (CBF), cerebral blood volume (CBV), and mean transit time (MTT). The technique requires an i.v. injection of the near infrared chromophore indocyanine green. Six newborn piglets, median age of 18 h (range 6–54 h), median weight of 1.75 kg (range 1.5–2.1 kg), were studied. Measurements of CBF, CBV, and MTT were made at normocapnia, hypocapnia, and hypercapnia to test the technique over a range of hemodynamic conditions. The accuracy of our new approach has been determined by direct comparison with measurements made using a previously validated computed tomography technique. Paired t tests showed no significant difference between computed tomography and NIRS measurements of CBF, CBV, and MTT, and mean biases between the two methods were −2.05 mL·min−1·100 g−1, −0.18 mL·100 g−1, and 0.43 s, respectively. The precision of NIRS CBF, CBV, and MTT measurements, as determined by repeated-measures ANOVA, was 9.71%, 13.05%, and 7.57%, respectively.

Oscillations in Cerebral Haemodynamics

The blood oxygenation level dependent (BOLD) contrast signal in functional magnetic resonance imaging (fMRI) provides highly spatially resolved deoxygenation maps of the activated cortex (Turner 1995). In common with many other activation measurement techniques, (including optical measurements) fMRI relies upon multiple repetition of the stimulus followed by signal averaging, baseline subtraction and extensive thresholding techniques to provide difference images of activation.

Use of mitochondrial inhibitors to demonstrate that cytochrome oxidase near-infrared spectroscopy can measure mitochondrial dysfunction noninvasively in the brain.

The use of near-infrared spectroscopy to measure noninvasively changes in the redox state of cerebral cytochrome oxidase in vivo is controversial. We therefore tested these measurements using a multiwavelength detector in the neonatal pig brain. Exchange transfusion with perfluorocarbons revealed that the spectrum of cytochrome oxidase in the near-infrared was identical in the neonatal pig, the adult rat, and in the purified enzyme. Under normoxic conditions, the neonatal pig brain contained 15 μmol/L deoxyhemoglobin, 29 μmol/L oxyhemoglobin, and 1.2 μmol/L oxidized cytochrome oxidase. The mitochondrial inhibitor cyanide was used to determine whether redox changes in cytochrome oxidase could be detected in the presence of the larger cerebral hemoglobin concentration. Addition of cyanide induced full reduction of cytochrome oxidase in both blooded and bloodless animals. In the blooded animals, subsequent anoxia caused large changes in hemoglobin oxygenation and concentration but did not affect the cytochrome oxidase near-infrared signal. Simultaneous blood oxygenation level-dependent magnetic resonance imaging measurements showed a good correlation with near-infrared measurements of deoxyhemoglobin concentration. Possible interference in the near-infrared measurements from light scattering changes was discounted by simultaneous measurements of the optical pathlength using the cerebral water absorbance as a standard chromophore. We conclude that, under these conditions, near-infrared spectroscopy can accurately measure changes in the cerebral cytochrome oxidase redox state.

Utility and limitations of near-infrared spectroscopy during cardiopulmonary bypass in a piglet model

Near-infrared spectroscopy assessment of cytochrome oxygenation could be a valuable technique to monitor cerebral intraneuronal oxygen delivery during cardiopulmonary bypass. However, the validity of the cytochrome signal has been questioned as it could easily be overwhelmed by the Hb signal. Five- to six-week-old control piglets (n = 5) underwent cardiopulmonary bypass at 37°C. Study animals (n = 10) received 100 mg/kg of sodium cyanide to uncouple cytochrome from Hb. Hematocrit was then decreased in steps of 5% from 35 to 5% with crystalloid hemodilution. In study piglets, the initiation of cardiopulmonary bypass was associated with oxygenated Hb increasing from 0 to 62.9 ± 25.6 μM times the differential path-length factor, and oxidized cytochrome a,a3 increasing to 1.9 ± 1.8 μM times the differential path-length factor. Cyanide caused oxygenated Hb to increase further to 132.3 ± 48.9 μM times the differential path-length factor, and oxidized cytochrome c decreased to −7.0 ± 2.6 μM times the differential path-length factor as anticipated, confirming uncoupling of electron transport. However, hemodilution subsequently resulted in linear decreases in oxidized cytochrome a,a3 (F = 8.57, p < 0.001) suggesting important cross-talk between the Hb and cytochrome signals as cytochrome is only intracellular. In control piglets, tissue oxygenation index showed a positive correlation with pump flow (r = 0.986, p = 0.013). The cytochrome signal as presently measured by near-infrared spectroscopy is highly dependent on hematocrit.

Precise measurement of cerebral blood flow in newborn piglets from the bolus passage of indocyanine green.

Indocyanine green (ICG) is a near-infrared dye that has the potential to be used as a tracer for the minimally invasive measurement of cerebral blood flow (CBF). In order to examine the technique, the arterial and cerebral concentrations of ICG were measured in newborn piglets during the bolus passage of ICG at normocapnia and two levels of mild hypercapnia. The results were analysed by applying the Fick principle in both integral and differential forms using a linear regression technique to improve the precision of calculated values of CBF. It was found that the integral method, which has been used previously, is particularly sensitive to errors in the time registration between the arterial and tissue signals whereas the differential method is less so. In addition, the differential method allows the venous outflow to be calculated which gives further information on the state of the capillary bed. CBF was 39.7 +/- 4.6 ml 100 g(-1) min(-1) at an arterial carbon dioxide tension (PaCO2) of 33.0+/-2.2 mmHg and increased to 53.7+/-9.1 and 75.4+/-15.2 ml 100 g(-1) min(-1) at a PaCO2 of 42.1 +/- 2.6 and 54.2 +/- 3.1 mmHg respectively (mean +/- SD, n = 7). There was no significant change in cerebral metabolic rate for oxygen, validating the value of blood flow to an arbitrary scaling factor. When the inspired CO2 fraction was returned to zero, calculated CBF returned to baseline with a variation of 7% of the mean, indicating that this technique is highly precise.

Oxidation and reduction of cytochrome oxidase in the neonatal brain observed by in vivo near-infrared spectroscopy

Near-infrared spectroscopy was used to determine the relationship between the redox state of mitochondrial cytochrome oxidase CuA and haemoglobin oxygenation in the isoflurane-anaesthetized neonatal pig brain. Adding 7% CO2 to the inspired gases increased the total haemoglobin concentration by 8 microM and oxidized CuA by 0.2 microM. Decreasing the inspired oxygen fraction to zero for 90 s dropped the oxyhaemoglobin concentration by 27 microM and reduced CuA by 1.8 microM. However, no change in the CuA redox state was observed until oxyhaemoglobin had decreased by more than 10 microM. The response of the CuA redox state to these stimuli was very similar following 80% replacement of the haemoglobin by a perfluorocarbon blood substitute; this demonstrates that the results in the normal haematocrit were not a spectral artefact due to the high haemoglobin/cytochrome oxidase ratio. We conclude that the large reductions in the CuA redox state during anoxia are caused by a decrease in the rate of oxygen delivery to the cytochrome oxidase oxygen binding site; the small oxidations, however, are likely to reflect the effects of metabolic changes on the redox state of CuA, rather than increases in the rate of oxygen delivery.

Measurement of cytochrome oxidase redox state by near infrared spectroscopy.

Although near infrared spectroscopy (NIRS) is primarily used to probe changes in oxyhaemoglobin (HbO2) and deoxyhaemoglobin (dHb) concentrations, it has long been realised that there is a significant oxygen-concentration dependent near infrared signal from the mitochondrial enzyme cytochrome c oxidase. In this paper we discuss the origins of this near infrared (NIR) signal, the possible factors affecting its intensity and its likely physiological and clinical significance. This paper complements our recent review on this subject1.

Oxygen dependency of cerebral oxidative phosphorylation in newborn piglets

Changes in hemoglobin oxygenation and oxidation state of the CuA centre of cytochrome oxidase were measured with full spectral near infrared spectroscopy simultaneously with phosphorus metabolites using nuclear magnetic resonance 31P spectroscopy at high time resolution (10 seconds) during transient anoxia (FiO2 = 0.0 for 105 seconds) in the newborn piglet brain. During the onset of anoxia, there was no change in either phosphocreatine (PCr) concentration or the oxidation state of the CuA centre of cytochrome oxidase until there was a substantial fall in cerebral hemoglobin oxygenation, at which point the CuA centre reduced simultaneously with the decline in PCr. At a later time during the anoxia, intracellular pH decreased rapidly, consistent with a fall in cerebral metabolic rate for O2 and reduced flux through the tricarboxylic acid cycle. The simultaneous reduction of CuA and decline in PCr can be explained in terms of the effects of the falling mitochondrial electrochemical potential. From these observations, it is concluded that, at normoxia, oxidative phosphorylation and the oxidation state of the components of the electron transport chain are independent of cerebral oxygenation and that the reduction in the CuA signal occurs when oxygen tension limits the capacity of oxidative phosphorylation to maintain the phosphorylation potential.

Nitric Oxide does not Inhibit Cerebral Cytochrome Oxidase In Vivo or in the Reactive Hyperemic Phase after Brief Anoxia in the Adult Rat

In this study, near-infrared spectroscopy was applied to examine whether cytochrome oxidase in the rat brain is inhibited by nitric oxide in vivo. During normoxia, intravenous NG-nitro-l-arginine methyl ester (l-NAME) administration significantly decreased the cerebral saturation of hemoglobin with oxygen but did not alter the cytochrome oxidase redox state. Anoxia significantly reduced the cytochrome oxidase. The time course of the recovery of the redox state during reoxygenation was not altered by l-NAME. The results suggest that in adult rats, cytochrome oxidase is not inhibited by nitric oxide, either in physiologic conditions or during reoxygenation after a brief anoxic period.

A new combined deep-body-temperature/NIRs probe for noninvasive metabolic measurements on human skeletal muscle

Temperature modulates tissue metabolism through thermodynamic phenomena, e.g. changes in enzymatic activities, pH variations, etc. In humans, temperature changes may be induced by stressful environmental conditions (e.g. immersion in cold water, etc.) or, artificially, during therapeutic procedures (extracorporeal circulation, cancer treatment using hyperthermia, etc.). Moreover, temperature distribution is not uniform in the body. In a forearm immersed in cold water at 20°C, it is possible to observe radial gradient values reaching 2°C/cm (Ducharme et al., 1991). Thus, the measurement of total tissue metabolism often results from the contribution of several subsets of tissue exposed to different temperatures, making the interpretation of the data sometimes difficult. In this context, the aim of the present study was to develop a non-in-vasive probe allowing one to simultaneously: 1) establish in a human skeletal muscle a region of uniform temperature; 2) measure the temperature; 3) measure the oxidative metabolism in the corresponding region.