Validation of Cerebral Venous Oxygenation Measured Using Near-Infrared Spectroscopy and Partial Jugular Venous Occlusion in the Newborn Lamb

Abstract

Near-infrared spectroscopy combined with partial jugular venous occlusion (JVO) offers promise for determining cerebral venous saturation (CSvO₂) in sick preterm infants, but has not been validated in the newborn brain or under conditions of hypoxaemia. We assessed the accuracy of the CSvO₂ estimate using cerebral venous oxygen saturation in superior sagittal sinus blood (SSSO₂) as the ‘gold standard’. Comparisons were made in seven newborn lambs over a wide range of arterial oxygen saturations (SaO₂) of 20% to 100%. Overall, median (range) CSvO₂ was 49.8% (10.6% to 88.5%), whereas SSSO₂ was 45.5% (4.3% to 76.6%); Bland—Altman analysis revealed a mean difference (CSvO₂—SSSO₂) of 5.1% and limits of agreement of ±27.4%. The change in cerebral blood volume (ΔCBV) induced by JVO increased with SaO₂ (P <0.05). In addition, the strength of the correlation of CSvO₂ with SSSO₂ progressively improved with increasing change in total haemoglobin concentration (ΔHbT) induced by JVO. With Bland—Altman analysis repeated for data with ΔHbT >30 μmol cm, the mean difference (CSvO₂—SSSO₂) decreased to 2.4% with limits of agreement of ±18.8%. We conclude that the accuracy of estimating CSvO₂ varies with the ΔCBV induced by JVO. Potential differences of optical properties between the head of the lamb and the human infant suggest that caution be exercised in directly applying these data to the human newborn. Nevertheless, this critical aspect of the JVO technique needs to be taken into consideration in developing an accurate measurement for sick preterm human infants.

Keywords

cerebral hypoxia cerebral venous saturation jugular venous occlusion near-infrared spectroscopy

Introduction

Abnormal cerebral haemodynamics and oxygen delivery are major aetiologic factors for cerebral injury of sick preterm infants (Meek et al, 1999b; Tsuji et al, 2000), and new methods for assessment of cerebral circulation and oxygenation are needed to aid their clinical management and improve their outcomes. Continuous wave near-infrared spectroscopy (NIRS) has been used for several assessments of cerebral circulatory function in preterm infants including cerebral blood flow (CBF), cerebral blood volume (CBV), and cerebral autoregulation (Hintz, 2001; Owen-Reece et al, 1999; Volpe, 2001c). Despite its promise, the difficulty of applying continuous wave NIRS to determine CBF and CBV renders it of limited use for routine bedside investigations. For example, CBF and CBV measurements require manipulation of arterial oxygenation, and cannot be performed in babies with severe respiratory disease or in babies who are not oxygen-dependent. However, several other NIRS-based assessments of cerebrovascular oxygenation offer great clinical promise. Estimation of cerebral venous saturation (CSvO₂), a measure which is sensitive to cerebral oxygen delivery, can be obtained using NIRS by inducing small changes in CBV by head-tilting (Skov et al, 1993; Wyatt et al, 1986), jugular venous occlusion (JVO) (Yoxall et al, 1995), and making use of ventilator-induced changes in CBV (Wolf et al, 1997).

Among these methods, JVO used in combination with NIRS (Yoxall et al, 1995) is a simple and easily applicable technique. The JVO technique has been applied in several studies of term (Buchvald et al, 1999) and preterm infants (Kissack et al, 2004a, b, 2005; Victor et al, 2006; Wardle et al, 2000; Yoxall and Weindling, 1998), after it was given limited validation in children (Yoxall et al, 1995). In this single validation study (Yoxall et al, 1995), CSvO₂ was contrasted with jugular venous oxygen saturation obtained during cardiac catheterization. However, as jugular venous blood includes contributions from extracerebral tissue, it may not accurately represent cerebral venous blood.

The lamb has been widely used in NIRS studies as a surrogate model for the human newborn brain (Volpe, 2001a), and NIRS-derived values of CBV (mL/100 g tissue) in the lamb are essentially identical to those in the human neonate (Barfield et al, 1999; Wyatt et al, 1990) and identical to the value derived from the gold-standard radiolabel technique (Cr⁵¹ red blood cells, Barfield et al, 1999). Using the newborn lamb model, we aimed to validate the estimate of CSvO₂ obtained using NIRS in combination with JVO. Validation in a newborn model is needed as critical features of cerebral oxygen metabolism change significantly between birth and childhood. In the newborn, CBF and cerebral oxygen delivery are relatively low, whereas cerebral metabolic rate for oxygen is high, (Jones et al, 1982; Richardson et al, 1989) and CSvO₂ is likely to be lower than the mature brain. Possibly, estimation of low venous saturation levels by the NIRS-JVO technique may be problematic in the immature brain, and the applicability of the technique at this age requires evaluation.

Our validation compared CSvO₂ estimates with cerebral venous oxygen saturations measured in blood samples taken from the superior sagittal sinus (SSSO₂) as the ‘gold standard’. In addition, we assessed the correspondence of CSvO₂ to SSSO₂ over a range of hypoxaemic conditions that predispose to cerebral injury.

Methods

Animal Preparation

Seven newborn lambs of Merino-Border Leicester cross were studied at 1 to 3 days after birth. All procedures were performed in accordance with guidelines established by the National Health and Medical Research Council of Australia, and were approved by the Ethics in Animal Experimentation Committee of Monash University. A nonocclusive intravenous catheter (Intracath 19 GA; Becton Dickinson, Sandy, UT, USA) was inserted into the left jugular vein of the newborn lamb for administration of maintenance fluid and anaesthetic medication. Newborn lambs were intubated and ventilated under general anaesthesia (100 mg/kg α-chloralose and 5 mg/kg ketamine hydrochloride for induction, followed by 25 to 50 mg/kg/h α-chloralose for maintenance). An intravenous catheter (Insyte-N 24 G, Becton Dickinson) filled with a saline-heparin solution (50 IU/mL) was inserted in the right axillary artery to enable monitoring of arterial pressure and extraction of arterial blood samples. For cerebral venous blood sampling, a similar catheter (Insyte-N 24 G, Becton Dickinson) was inserted into the superior sagittal sinus through a small hole (2 mm in diameter) drilled in the skull along the sagittal suture just anterior to the lamboid suture.

Physiological Measurements

Beat-to-beat arterial oxygen saturation (SpO₂) was measured with a pulse oximeter placed on the lamb's tongue (Nellcor 200; Nellcor Inc., Pleasanton, CA, USA). Arterial blood pressure was measured with a calibrated strain-gauge pressure transducer (Cobe CDX III; Cobe Laboratories, Lakewood, CO, USA) connected to a bridge amplifier (Quad Bridge Amp; ADInstruments, Sydney, Australia). Heart rate was derived from the arterial pressure pulse (Chart v4.2, ADInstruments). The signals were sampled at 400 Hz and displayed throughout the study (PowerLab/16R SP, ADInstruments).

Near-infrared Spectroscopy Measurement of Cerebral Venous Saturation

Optodes (NIRO-500; Hamamatsu Photonics K.K., Hamamatsu City, Japan) were positioned 2.5 cm apart on the right parietal-occipital area, and covered with a lightproof dressing. NIRS measurements of changes in concentrations (μmol cm) of oxyhaemoglobin (μHbO), deoxyhaemoglobin (ΔHb), and total haemoglobin (ΔHbT = ΔHbO + ΔHb) were recorded at 0.5 sec intervals. Partial jugular venous occlusions (JVOs) were performed by investigator FW by applying gentle pressure to the lamb's neck over the right jugular vein for 20 secs. Occlusions induced a small and temporary increase in HbO and HbT (Figure 1A) measured by NIRS. Cerebral venous saturation was calculated off-line using Microsoft Excel from the NIRS data obtained over the first 5 secs of the rise in HbT and HbO (Yoxall et al, 1995) using the relationship CSvO₂ = AHbO/ΔHbT.

Figure 1

Changes in cerebral HbT and HbO induced using JVO in (A) normoxaemia and (B) hypoxaemia.

Data were accepted as suitable for the calculation of CSvO₂ if there was a stable baseline before the occlusion, a rise in the concentration of both HbT and HbO during the occlusion, and a return to the baseline after release of the occlusion.

Blood Gas Analysis

After each JVO, blood samples of 500 μL were collected from superior sagittal sinus and the axillary artery in heparinised plastic syringes and analysed immediately for HbO saturation (SSSO₂ and arterial oxygen saturation (SaO₂), respectively; OSM2 Hemoximeter; Radiometer, Copenhagen, Denmark).

Induced Hypoxia

JVO was performed at least three times while SaO₂ was ≥95%, then over a range of arterial hypoxaemia (SaO₂ of 20% to 94%) induced by ventilating the lamb using 10% to 40% inspired oxygen, balance nitrogen. Arterial oxygen saturation was reduced by ∼10% at each step, monitored with pulse oximetry measurements (SpO₂) and confirmed with arterial blood gas analysis. After SpO₂ was held stable for 15 mins at each step, JVO was performed, and blood was sampled from the superior sagittal sinus for measurement of SSSO₂ (Hemoximeter OSM2, Radiometer).

Change in Cerebral Blood Volume during Jugular Venous Occlusion

A differential pathlength factor of 4.99 was used for conversion of NIRS measurements into μmol/L (Delpy et al, 1988; Duncan et al, 1995). Change in CBV (mL/100 g) was calculated from NIRS measurement of the maximum ΔHbT (μmol/L) during the JVO, using the following formula (Elwell, 1995):

Δ C B V = \frac{Δ H b T \times M W_{Hb} \times 10^{- 6}}{t H b \times 10^{- 2} \times C L V H R \times D_{t} \times 10^{- 6}}

where MW_Hb = molecular weight of haemoglobin = 64,500, tHb = concentration of haemoglobin in large vessels in g/100 mL, CLVHR = cerebral to large vessel haemotocrit ratio = 0.69, and D_t = brain tissue density in g/mL = 1.05.

Statistics

Measurements of CSvO₂ by NIRS-JVO were compared with the corresponding SSSO₂ measurements using Pearson linear regression analysis (SigmaStat; SPSS Inc., Chicago, IL, USA). To examine the accuracy of the NIRS-JVO technique with respect to the extent of ΔCSV induced by JVO, the coefficient of determination (R²) was calculated for different levels of ΔHbT. An analysis of agreement of CSvO₂ and SSSO₂ was based on the method described by Bland and Altman (1986). Measurements of ΔCRV and ΔHbT induced by JVO during normoxaemia and hypoxaemia were compared using the Mann—Whitney U test. In all statistical tests, P <0.05 was considered significant.

Results

Near-infrared spectroscopy data collected during 95 JVOs met selection criteria for the calculation of CSvO₂. Fifty-two occlusions were performed when the lambs were normoxaemic with SaO₂ ≥95%, whereas 43 occlusions were performed during induced hypoxemia when SaO₂ was 20% to 94%. Over all tests, the median (range) of CSvO₂ was 49.8% (10.6% to 88.5%), whereas SSSO₂ was 45.5% (4.3% to 76.6%). The mean difference was 5.1% and limits of agreement (2 standard deviations) were ± 27.4% (Bland and Altman, 1986).

The NIRS-determined magnitudes of ΔHbT and ΔHbO generated using JVO during hypoxia tend to be smaller than those that could be generated during normoxaemia (Figure 1). Similarly, estimates of ΔCRV generated by JVO during hypoxia was smaller than that during normoxaemia (0.52 (0.37 to 0.58) mL/100 g versus 0.58 (0.45 to 0.68) mL/100 g, P <0.05).

Values of R² for the regression of CSvO₂ on SSSO₂ were dependent on the extent of ΔHbT (μmol cm) during JVO (Figure 2). R² equalled 0.51 when all data were included in the analysis, increased as data with smaller ΔHbT were excluded, and then plateaued for ΔHbT more than 30 μmol cm. Figure 3 illustrates the significant correlation between CSvO₂ and SSSO₂ (R² = 0.76, P <0.0001) for the 87 occlusions with ΔHbT > 30 μmol cm.

Figure 2

The degree of correlation between CSvO₂ on SSSO₂ was affected by the degree of ΔHbT induced by JVO. Stronger correlation (greater R²) was obtained with larger ΔHbT, with a plateau reached for ΔHbT >30 μmol cm.

Figure 3

Correlation between CSvO₂ and SSSO₂ for JVO-induced ΔHbT >30 μmol cm.

The Bland-Altman analysis of CSvO₂ versus SSSO₂ for data with ΔHbT > 30 μmol cm is shown in Figure 4A. The mean difference (CSvO₂—SSSO₂) was 2.4% with limits of agreement (2 standard deviations) of ±18.8%. The CSvO₂—SSSO₂ difference varied with the level of SSSO₂, with CSvO₂ slightly higher than SSSO₂ at low values of oxygen saturation, and slightly lower at high values of oxygen saturation (Figure 4B). The mean difference (limits of agreement) was 9.6% (−5.5% to 24.7%) at SSSO₂ of 20%, 3.6% (−11.4% to 18.6%) at SSSO₂ of 40%, and −2.4% (−17.5% to 12.6%) at SSSO₂ of 60%. Figure 4C represents CSvO₂ data detrended by subtracting values calculated using the regression equation of Figure 4B. After this correction, the limits of agreement around the mean difference of 0 were ±14.8.

Figure 4

(A) Bland-Altman plot of the difference between CSvO₂ (NIRS) and SSSO₂ (blood gas analysis) plotted against SSSO₂, for data with ΔHbT >30 μmol cm. Mean difference = 2.4%, limits of agreement (2 standard deviations) = ± 18.8%. (B) Regression of (CSvO₂—SSSO₂) versus SSSO₂, with the 95% confidence limits for the data population shown. (C) Data detrended according to the regression equation of (B), with mean difference = 0, limits of agreement (2 standard deviations) = ±14.8%.

Discussion

Near-infrared spectroscopy in combination with partial JVO has been used to estimate CSvO₂ in human subjects in several studies, principally premature newborns who are at risk of cerebral hypoxia (Kissack et al, 2004a, b , 2005; Wardle et al, 2000; Yoxall and Weindling, 1998). Despite its wide application, the prior validation of the method consisted of a single study in children where CSvO₂ was compared with oxygen saturation of jugular venous blood (Yoxall et al, 1995). Our study is the first validation of the technique against oxygen saturation in the cerebral venous outflow. Our principal finding is the confirmation that CSvO₂ corresponds closely to sagittal sinus saturation (SSSO₂) provided that the magnitude of ΔHbT induced by JVO is adequate.

Our data clearly show that correlation between CSvO₂ and SSSO₂ progressively improved with increases in ΔHbT, until the increase in ΔHbT reached ∼30 μmol cm (Figure 2). With this degree of HbT increment, the agreement of CSvO₂ with SSSO₂ approximated the limits of agreement reported in the previous validation study (Yoxall et al, 1995). At lower ΔHbT, the relatively greater noise-to-signal ratio in the NIRS measurements, evident in Figure 1B, may cause less precision in the CSvO₂ estimate and poorer agreement with SSSO₂.

During conditions of severe hypoxaemia, the estimated increment in CBV with the JVO manoeuvre was smaller than that during normoxaemia. This may be explained by the nonlinear form of the volume-pressure relationship for the cerebral vascular compartment (Folkow and Neil, 1971). Under conditions of cerebral hypoxaemia, the cerebral circulation would be widely dilated and the cerebral venous compartment likely to be highly distended (Bennet et al, 1998; Lassen, 1966; Livera et al, 1991). When highly distended, the compliance of veins would be low and JVO would induce little increment in CBV. Small CBV increments may result in small ΔHbT, especially if the blood haematocrit is low. As our findings suggest, ΔHbT <30 μmol cm may compromise the applicability of the NIRS-JVO technique.

It is unlikely that infants being ventilated would have cerebral hypoxaemia to the extent induced in our experimental manipulation. However, hypercapnia (Wyatt et al, 1991), or hypotension (Ferrari et al, 1992), or vasoparalysis due to hypoxicischaemic injury (Meek et al, 1999a; Wyatt et al, 1990), all common conditions in sick preterm infants, could increase baseline CBV and limit the CBV increment produced by JVO. In published studies in sick preterm infants (Kissack et al, 2004a, b , 2005; Wardle et al, 2000; Yoxall and Weindling, 1998), the change of CBV or HbT induced by JVO was not reported, so the accuracy of previous CSvO₂ estimates is uncertain. The estimated increase in CBV in the validation study (Yoxall et al, 1995) was in the order of 0.25 to 0.33 mL/100 g, equivalent to ΔHbT of 42 to 55 μmol cm assuming a haemoglobin concentration of 12 g/100mL. On the basis of our findings, this ΔHbT should have ensured accurate estimate of CSvO₂ in that study. The safety of manipulating the cerebral circulation in infants at risk for intracranial haemorrhage merits careful consideration (Volpe, 2001b). Spontaneous fluctuations in CBV similar in magnitude to those induced by JVOs were observed in ventilated and nonventilated preterm infants (Yoxall et al, 1995), suggesting that an adequate CBV can be safely incremented using JVO without risk of intracranial haemorrhage.

We have identified a systematic variation in the difference between CSvO₂ and SSSO₂ that is related to the level of cerebral venous oxygen saturation (Figure 4B). CSvO₂ slightly overestimated SSSO₂ at low oxygen saturations (<20%), and underestimated SSSO₂ at high oxygen saturations (>70%). We propose that this finding arises from inhomogeneity of blood flow and oxygenation within the region of tissue illuminated by NIRS, and their variation with arterial oxygen saturation. At extremely low levels of oxygen saturation, overestimation of SSSO₂ by the NIRS-JVO technique may be explained by recruitment of vessels that are upstream of the venous compartment and that contain blood of relatively higher oxygen saturation. Wide vasodilatation and high CBF, which can be anticipated under the conditions of severe hypoxaemia, may amplify the increment in intravascular pressure arising from JVO. This, combined with lower compliance of the widely dilated venous compartment, may result in pressure arising from JVO distending capillary or small arterial vessels in addition to cerebral veins. With inclusion of ΔHbO and ΔHbT arising from the upstream compartment, CSvO₂ estimate would exceed SSSO₂. At the high end of cerebral oxygenation, underestimation of CSvO₂ by the NIRS-JVO measurement (Figure 4B) may reflect greater contribution of white matter to the NIRS signal. As previously shown, CBF in most regions of the newborn lamb's brain decreases with increase in arterial oxygenation, except in the white matter (Szymonowicz et al, 1988). Should higher cerebral oxygenation reduce CBV in the cortex, the vasculature of the white matter may become more prominently represented in the NIRS measurement. As the white matter is mostly drained by the deep cerebral venous system, venous saturation of the sagittal sinus blood which principally drains the cortex (Grant et al, 1995) may not fully represent the venous oxygenation detected by the NIRS.

Importantly, we have validated CSvO₂ estimate against cerebral venous oxygen saturation, using SSSO₂ as the gold standard. When corrected for the small systematic variation across the SSSO₂ range, the limits of agreement improved to ±14.6% (Figure 4C), almost identical to that reported previously (Yoxall et al, 1995). The close correspondence in the previous validation study (Yoxall et al, 1995) may be explained by comparable contributions of extracerebral tissue to both the jugular venous blood and to the NIRS measurement. In our study, close agreement may have occurred because extracerebral contamination of the NIRS measurement was limited. The overlying skin and skull tissue of the newborn lamb is thin (<5mm), whereas the light penetration by NIRS is approximately 1.6 cm with the interoptode distance of 2.5 cm (Brown et al, 2002).

Whether these results can be transferred directly to the human newborn is uncertain, as differences in geometry of the head of the human infant and lamb may lead to optical property differences affecting NIRS measurements. The propagation of light by NIRS in heterogeneous tissue is theoretically complex and is affected by the geometry and thickness of the layered structures such as skin, skull, cerebrospinal fluid, and grey and white matter (Boas et al, 2001; Okada and Delpy, 2003a, b ). Hence, the possibility exists for differences in NIRS measurements, and in CSvO₂ estimates between the two species. Moreover, optical properties (e.g., light scattering and absorption coefficients) have not been studied in the lamb brain, and at comparable gestational age, the lamb brain is more mature in the degree of myelination (Mallard et al, 1998; Raju, 1992). However, gross cortical anatomy is similar, as cortical thickness is 1.66 mm in the term lamb, compares closely with human infant values of 1.4 mm at 30 weeks gestation and 2 mm at term (Marin-Padilla, 1970; Rees et al, 1988). In any event, in practical terms the effect of superficial tissues on NIR light propagation is low in the neonatal head compared with the adult, because of the low scattering coefficient of the neonatal brain, the thin cerebrospinal fluid layer, and the small cranial diameter (Fukui et al, 2003; Okada, 2000).

A major technical advantage of measuring CSvO₂ with the NIRS-JVO method is that, being a ratio of two concentrations, the measurement is free of the potential errors introduced by uncertainty of the differential pathlength factor (Owen-Reece et al, 1999). A practical advantage of the NIRS-JVO method is the simplicity and ease of application. By estimating venous oxygen saturation, the measurement offers important information on the balance of cerebral oxygen transport and oxygen utilisation, in addition to the cerebral haemodynamic parameters that can be measured by NIRS (Edwards et al, 1988; Wyatt et al, 1990, 1986). Our validation lends further support to its application in sick preterm infants.

Conclusion

Near-infrared spectroscopy in combination with JVO produces reliable estimates of CSvO₂, provided there is adequate change in cerebral venous blood volume induced by the JVO manoeuvre. Potential differences of optical properties between the head of the lamb and the human infant suggest that caution be exercised in directly applying these data to the human newborn. Nevertheless, this critical aspect of the NIRS-JVO technique needs to be taken into consideration in clinical applications.

Footnotes

Acknowledgements

We acknowledge financial support from the National Health and Medical Research Council of Australia.

References

Barfield

Noma

Kukita

Cussen

Oates

Walker

(1999) Cerebral blood volume measured using near-infrared spectroscopy and radio labels in the immature lamb brain. Pediatr Res 46:50–6

Bennet

Peebles

Edwards

Rios

Hanson

(1998) The cerebral hemodynamic response to asphyxia and hypoxia in the near-term fetal sheep as measured by near infrared spectroscopy. Pediatr Res 44:951–7

Bland

Altman

(1986) Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1:307–10

Boas

Gaudette

Strangman

Cheng

Marota

Mandeville

(2001) The accuracy of near infrared spectroscopy and imaging during focal changes in cerebral hemodynamics. Neuroimage 13:76–90

Brown

Picot

Naeini

Springett

Delpy

Lee

(2002) Quantitative near infrared spectroscopy measurement of cerebral hemodynamics in newborn piglets. Pediatr Res 51:564–70

Buchvald

Kesje

Greisen

(1999) Measurement of cerebral oxyhaemoglobin saturation and jugular blood flow in term healthy newborn infants by near-infrared spectroscopy and jugular venous occlusion. Biol Neonate 75:97–103

Delpy

Cope

van der Zee

Arridge

Wray

Wyatt

(1988) Estimation of optical pathlength through tissue from direct time of flight measurement. Phys Med Biol 33:1433–42

Duncan

Meek

Clemence

Elwell

Tyszczuk

Cope

Delpy

(1995) Optical pathlength measurements on adult head, calf and forearm and the head of the newborn infant using phase resolved optical spectroscopy. Phys Med Biol 40:295–304

Edwards

Wyatt

Richardson

Delpy

Cope

Reynolds

(1988) Cotside measurement of cerebral blood flow in ill newborn infants by near infrared spectroscopy. Lancet 2:770–1

10.

Elwell

(1995) Appendix I: Conversion of Units. In: Elwell

, (ed), A Practical User's Guide to Near Infrared Spectroscopy. Hamamatsu City, Japan: Hamamatsu Photonics KK

11.

Ferrari

Wilson

Hanley

Traystman

(1992) Effects of graded hypotension on cerebral blood flow, blood volume, and mean transit time in dogs. Am J Physiol 262:H1908–14

12.

Folkow

Neil

(1971) Vascular length and radius. In: Folkow

Neil

(eds), Circulation. New York: Oxford University Press, 36–56

13.

Fukui

Ajichi

Okada

(2003) Monte Carlo prediction of near-infrared light propagation in realistic adult and neonatal head models. Appl Opt 42:2881–7

14.

Grant

Franzini

Wild

Walker

(1995) Continuous measurement of blood flow in the superior sagittal sinus of the lamb. Am J Physiol 269:R274–9

15.

Hintz

(2001) Near-infrared spectroscopy: neonatal and perinatal applications. NeoReviews 2:e22–8

16.

Jones

Jr Rosenberg

Simmons

Molteni

Koehler

Traystman

(1982) Oxygen delivery to the brain before and after birth. Science 216:324–325

17.

Kissack

Garr

Wardle

Weindling

(2004a) Cerebral fractional oxygen extraction in very low birth weight infants is high when there is low left ventricular output and hypocarbia but is unaffected by hypotension. Pediatr Res 55:400–5

18.

Kissack

Garr

Wardle

Weindling

(2004b) Postnatal changes in cerebral oxygen extraction in the preterm infant are associated with intraventricular hemorrhage and hemorrhagic parenchymal infarction but not periventricular leukomalacia. Pediatr Res 56:111–6

19.

Kissack

Garr

Wardle

Weindling

(2005) Cerebral fractional oxygen extraction is inversely correlated with oxygen delivery in the sick, newborn, preterm infant. J Cereb Blood Flow Metab 25:545–53

20.

Lassen

(1966) The luxury-perfusion syndrome and its possible relation to acute metabolic acidosis localised within the brain. Lancet 2:1113–5

21.

Livera

Spencer

Thorniley

Wickramasinghe

Rolfe

(1991) Effects of hypoxaemia and bradycardia on neonatal cerebral haemodynamics. Arch Dis Child 66:376–80

22.

Mallard

Rees

Stringer

Cock

Harding

(1998) Effects of chronic placental insufficiency on brain development in fetal sheep. Pediatr Res 43:262–270

23.

Marin-Padilla

(1970) Prenatal and early postnatal ontogenesis of the human motor cortex: a golgi study. I. The sequential development of the cortical layers. Brain Res 23:167–83

24.

Meek

Elwell

McCormick

Edwards

Town-send

Stewart

Wyatt

(1999a) Abnormal cerebral haemodynamics in perinatally asphyxiated neonates related to outcome. Arch Dis Child Fetal Neonatal Ed 81:F110–5

25.

Meek

Tyszczuk

Elwell

Wyatt

(1999b) Low cerebral blood flow is a risk factor for severe intraventricular haemorrhage. Arch Dis Child Fetal Neonatal Ed 81:F15–8

26.

Okada

Delpy

(2003a) Near-infrared light propagation in an adult head model. I. Modeling of low-level scattering in the cerebrospinal fluid layer. Appl Opt 42:2906–14

27.

Okada

Delpy

(2003b) Near-infrared light propagation in an adult head model. II. Effect of superficial tissue thickness on the sensitivity of the near-infrared spectroscopy signal. Appl Opt 42:2915–22

28.

Okada

(2000) The effect of superficial tissue of the head on spatial sensitivity profiles for near infrared spectroscopy and imaging. Optical Rev 7:375–82

29.

Owen-Reece

Smith

Elwell

Goldstone

(1999) Near infrared spectroscopy. Br J Anaesth 82:418–26

30.

Raju

(1992) Some animal models for the study of perinatal asphyxia. Biol Neonate 62:202–14

31.

Rees

Bocking

Harding

(1988) Structure of the fetal sheep brain in experimental growth retardation. J Dev Physiol 10:211–25

32.

Richardson

Carmichael

Homan

Gagnon

(1989) Cerebral oxidative metabolism in lambs during perinatal period: relationship to electrocortical state. Am J Physiol 257:R1251–7

33.

Skov

Pryds

Greisen

Lou

(1993) Estimation of cerebral venous saturation in newborn infants by near infrared spectroscopy. Pediatr Res 33:52–5

34.

Szymonowicz

Walker

Cussen

Cannata

(1988) Developmental changes in regional cerebral blood flow in fetal and newborn lambs. Am J Physiol 254:H52–8

35.

Tsuji

Saul

du Plessis

Eichenwald

Sobh

Crocker

Volpe

(2000) Cerebral intravascular oxygenation correlates with mean arterial pressure in critically ill premature infants. Pediatrics 106:625–32

36.

Victor

Marson

Appleton

Beirne

Weindling

(2006) Relationship between blood pressure, cerebral electrical activity, cerebral fractional oxygen extraction, and peripheral blood flow in very low birth weight newborn infants. Pediatr Res 59:314–9

37.

Volpe

(2001a) Hypoxic—ischaemic encephalopathy: biochemical and physiological aspects. In: Volpe

(ed), Neurology of the Newborn. Philadelphia: Saunders, 217–76

38.

Volpe

(2001b) Intracranial hemorrhage: germinal matrix-intraventricular hemorrhage of the premature infant. In: Volpe

(ed), Neurology of the Newborn. Philadephia: WB Saunders, 428–48

39.

Volpe

(2001c) Specialized Studies in the Neurological evaluation. In: Volpe

(ed), Neurology of the Newborn. Philadephia: WB Saunders, 134–77

40.

Wardle

Yoxall

Weindling

(2000) Determinants of cerebral fractional oxygen extraction using near infrared spectroscopy in preterm neonates. J Cereb Blood Flow Metab 20:272–9

41.

Wolf

Duc

Keel

Niederer

von Siebenthal

Bucher

(1997) Continuous noninvasive measurement of cerebral arterial and venous oxygen saturation at the bedside in mechanically ventilated neonates. Crit Care Med 25:1579–82

42.

Wyatt

Cope

Delpy

Richardson

Edwards

Wray

Reynolds

(1990) Quantitation of cerebral blood volume in human infants by near-infrared spectroscopy. J Appl Physiol 68:1086–91

43.

Wyatt

Cope

Delpy

Wray

Reynolds

(1986) Quantification of cerebral oxygenation and haemodynamics in sick newborn infants by near infrared spectrophotometry. Lancet 2:1063–6

44.

Wyatt

Edwards

Cope

Delpy

McCormick

Potter

Reynolds

(1991) Response of cerebral blood volume to changes in arterial carbon dioxide tension in preterm and term infants. Pediatr Res 29:553–7

45.

Yoxall

Weindling

Dawani

Peart

(1995) Measurement of cerebral venous oxyhemoglobin saturation in children by near-infrared spectroscopy and partial jugular venous occlusion. Pediatr Res 38:319–23

46.

Yoxall

Weindling

(1998) Measurement of cerebral oxygen consumption in the human neonate using near infrared spectroscopy: cerebral oxygen consumption increases with advancing gestational age. Pediatr Res 44:283–90