• No results found

Noninvasive Monitoring of Cerebral Oxygenation in Preterm Infants: Preliminary Observations


Academic year: 2020

Share "Noninvasive Monitoring of Cerebral Oxygenation in Preterm Infants: Preliminary Observations"


Loading.... (view fulltext now)

Full text






of Cerebral


in Preterm




Jane E. Brazy, MD, Darrell V. Lewis, MD, Michael H. Mitnick, PhD, and

Frans F. J#{246}bsisvander Vliet, PhD

From the Departments of Pediatrics and Physiology Duke University Medical Center

Durham, North Carolina

ABSTRACT. A noninvasive optical method for bedside

monitoring of cerebral oxygenation in small preterm in-fants was evaluated. Through differential absorbance of

near infrared light, changes in the oxidation-reduction level of cytochrome aa, in the oxygenation state of he-moglobin and in tissue blood volume were assessed in the transilluminated anterior cerebral field. Overall, cerebral oxygenated hemoglobin correlated significantly with

transcutaneous oxygen, r = .44 p < .0001; however, cor-relation was best in the absence of cardiorespiratory disease. Hypoxia with or without bradycardia led to

he-moglobin deoxygenation and a shift in cytochrome aa to a more reduced state. When hypoxic episodes came in

series or were prolonged, aa reduction occurred

simul-taneous with hemoglobin deoxygenation but its recovery

to base-line values sometimes lagged behind the return of hemoglobin oxygenation. In one infant with a large patent ductus arteriosus, even brief episodes of mild bradycardia caused precipitous reduction of cytochrome

1l3 before any shift to greater hemoglobin deoxygenation.

This response disappeared after ductal ligation. In

gen-eral, the antecedent state of cerebral oxygenation, the

severity and duration of deoxygenation, and the presence

or absence of circulatory abnormalities all influenced the

aa3 response to hypoxia. Continuous noninvasive near infrared monitoring of cerebral oxygenation can be per-formed on sick preterm infants at the bedside. Pediatrics

1985;75:217-225; bedside monitoring, cerebral

oxygena-tion, cytochrome aa3, oxidative metabolism, preterm in-fants.

Received for publication May 31, 1984; accepted Oct 12, 1984.

Reprint requests to (J.E.B.) Box 3967, Duke University Medical Center, Durham, NC 27710.

PEDIATRICS (ISSN 0031 4005). Copyright © 1985 by the American Academy of Pediatrics.

A major



for neonates


intensive care is provision of adequate nutrition



to the brain

to allow

for normal

func-tion and development. Current methods of assess-ing the adequacy of therapy involve measurements of blood pressure, glucose concentration, blood ox-ygen content, and transcutaneous oxygen at sites distant from the brain; thus, a more direct assess-ment of cerebral oxygen delivery and oxygen utili-zation is needed to evaluate therapies and avoid toxicities.

In this paper, we describe the application of a noninvasive optical method for determining

cere-bra! oxygenation

in the patient

in the intensive


nursery. The instrument used, which we have







sufficiency scope), is a noninvasive transcutaneous


for simultaneous

and continuous


of changes in the oxidation-reduction state of cy-tochrome l2Z3


the relative



oxygen-ated and deoxygenated hemoglobin in the optical field. This technique takes advantage of the semi-transparent quality of the skin and infant skull in the near infrared spectrum to allow photon trans-mission through the brain tissue. By appropriate selection of near infrared wavelengths, changes in light absorbance, which are characteristic of cyto-chrome aa3


of hemoglobin,


be used


monitor changes in the relative oxidation-reduction state of cytochrome aa, the amounts of oxygenated


hemoglobin volume in the illuminated anterior cerebral field.

Cytochrome aa3


aa3), also known as

cyto-chrome c oxidase, is the terminal member in the mitochondrial electron transport chain which

ac-counts for more than 90% of cellular oxygen

utili-zation in aerobic metabolism. This enzyme can


as the ultimate indicator of mitochondrial

oxygen sufficiency within brain tissue because it changes photon absorption characteristics, both in the visible and near infrared ranges, in parallel to the adequacy of oxygen available for its function.

Visible light has poor tissue penetration and thus, has been useful mainly in the laboratory. More

recently, J#{246}bsis1has described the characteristic

cytochrome aa3 absorption of near infrared light and its potential usefulness. With its greater pene-tration, near infrared light allows monitoring of

less favorable optical media such as intact organs,

and it may be used in less controlled environments such as intensive care units and operating rooms.

Simultaneous measurement of oxygenated

he-moglobin (tHbO2) and deoxygenated hemoglobin


within the tissue under observation provides direct information on oxygen delivery; the two to-gether, total hemoglobin, indicate relative tissue blood volume (tBV). A single instrument that gives information about the dynamics of the four

param-eters, cytochrome aa, deoxygenated hemoglobin, oxygenated hemoglobin, and tissue blood volume,

offers the opportunity to assess cerebral oxygen

delivery and oxygen utilization on a continuous

basis and to study the influence of disease states

and therapies on brain metabolism.


The near infrared technique uses three to four

diode lasers (Laser Diode Labs, mc) as sources of near infrared wavelengths between 760 and 904 nm.

This study used 775, 815, and 904 nm. The lasers

are pulsed sequentially at 1 kHz for 200 ns each

and the light from the lasers is captured in glass fiberoptic strands, which are randomly intermixed and combined to form a 3-mm bundle, 1.5 m (5 ft) in length, which, after a 90-degree bend, ends in a

1.8-cm diameter cylindrical disk to form the light

delivering optrode (Duke Department of


Photons are collected by a 5-mm pickup bun-dle in a similar receiving optrode and are conducted back to a photomultiplier in the NIROS-SCOPE

chassis to be measured. The signals are then

de-modulated, amplified, and transformed into loga-rithmic form so that changes in concentration are a linear function of the ultimate



al-gorithms derived from animal experiments, signals

are generated representing changes in the relative

absorption of cytochrome aa, deoxygenated he-moglobin, oxygenated hemoglobin, and relative blood volume in the tissue being monitored.

Appar-ent extinction coefficients for the wavelengths

em-ployed were experimentally obtained in animals. Optical signals from a cat’s head were recorded while the animal was perfused with totally anoxic red cell suspensions at different hematocrits, and conversely with totally oxygenated red cell

suspen-sions using three atmospheres of oxygen in the

hyperbaric chamber. The cytochrome aa3 contri-bution was also determined in the hyperbaric cham-ber by observing the change in optical signals when a blood-free animal preparation was rendered an-oxic. Because scattering within the cranium pre-vents a definition of the optical path length, the

results must be expressed in relative terms rather

than in absolute absorbance units. Thus, the in-strument is calibrated in units of “variation in density” (v/d), where 1 v/d represents an order of

magnitude change in the signal received. Data are displayed on a multichannel chart recorder with additional channels representing the patient’s ECG, respiration rate, and transcutaneous Po2

levels, obtained from standard bedside monitors

(Mennen Medical mc, and TCOM, Novametrics

Medical System mc).

The NIROS-SCOPE was analyzed for laser

safety before use in clinical studies. Lasers used are

of class


B, which

are not



to the skin.2 Calculations both of the peak energy per pulse (2.25 x iO J/cm2) and energy flux during

prolonged exposure (0.09 W/cm2) showed both

quantities to be below maximum permissible

expo-sures as published by the American National

Standards Institute.2 The study was approved by the Clinical Investigation Review Board and paren-tal consent was obtained. Because class III B lasers

should not be viewed directly by the eye, infants’ eyes were covered with phototherapy eye shields

(Bili-mask, Olympic Medical), and any detachment of the input optrode caused an automatic power

shut off. The site of illumination on the infant’s scalp was inspected immediately after optrode

re-moval, a few hours later, and the next day. No


or sign of injury was found in any case. For infant monitoring, the flat surface of each optrode was attached to the skin of the temple

using a double-sided adhesive circle (Novametrics)

and the sides of the optrodes were secured with additional tape to assure exclusion of light.

Oppo-site temples of the head were chosen as the sites of


was supported by a horseshoe-shaped, soft foam rubber cushion to prevent major turning

move-ments of the head, which would apply uneven pres-sure on the optrode disk and alter the base line. Small movements, although detectable on the

trac-ing as narrow simultaneous spikes in each of the optical channels,


not alter the base line. The transcutaneous oxygen monitor placement was in the preductal distribution as close as possible to the cerebral circulation. Sites used included cheek, sub-mentum, neck, and right upper thorax.

We chose to monitor serially three infants whose

courses were characteristic of preterm infants in

intensive care units. All were studied at more than 72 hours of age and were free of intraventricular

hemorrhage by ultrasound assessment. mnfant A

represented a normal healthy preterm infant of 29 weeks’ gestation. He was free of significant apnea and bradycardia and was receiving no respiratory support. Infant B, with birth weight of 890 g and

gestation of 26 weeks, had pulmonary immaturity of prematurity. In general, her inspired oxygen

varied from 21% to 28%, although it was increased

during procedures. She was ventilated at low rates to prevent apnea and bradycardia. Arterial blood

gas sampling and transcutaneous oxygen

monitor-ing demonstrated frequent spontaneous swings in

oxygenation. Baby C, twin to baby A, weighed 1,150

g at birth and developed significant respiratory distress syndrome. Study periods beginning with

the recovery phase of respiratory distress syndrome included multiple observations before and after


ligation of a large patent ductus arteriosus. Aortic Doppler studies prior to ligation

demon-strated marked reverse aortic diastolic flow.

Twenty-seven monitoring periods of 15 to 150

minutes each (average 60 minutes) were conducted

over a 2-week span. A monitoring period consisted

of a single-site placement of the transcutaneous

oxygen monitor, allowing time for warm up and calibration. Readings from the strip recorder for deoxygenated hemoglobin, oxygenated hemoglobin,

tissue blood volume, cytochrome aa and

transcu-taneous Po2 were taken at five-minute intervals for statistical comparisons. If movement artifact was present at a reading time (2% of occasions), the

entire reading was omitted from analysis. In all, data from 344 observation time points were ana-lyzed.

Correlation coefficients for measurement van-ables within babies were calculated by fitting a multivaniate analysis of variance model to the data, controlling for different measurement sequences. The error matrix was inverted, producing a corre-lation matrix in which possible scale base-line changes, due to different sequences, were partialled

out. Correlations over all babies were calculated similarly; the multivaniate analysis of variance model was of a repeated-measures form to control for base-line differences both between babies and for different sequences within babies.


Characteristic patterns of cerebral oxygenation during stable and unstable periods and the temporal course in which one event (such as apnea or hy-poxia) leads to sequential changes in other param-eters are best appreciated by directly examining the

NIROS-SCOPE recordings. Base-line values for

NIROS-SCOPE measurements are selected from

periods of stability. These are used to determine relative changes, but cannot be quantitated into absolute scales. Thus, each child serves as his or her own control within a study period and each

study period may have a different base line.

Throughout the monitoring of infant A, and dur-ing clinically stable periods with the other infants,


few fluctuations in any of the parameters (deoxygenated hemoglobin, oxygenated hemoglo-bin, tissue blood volume, cytochrome aa3, transcu-taneous P02) were noted (see Fig 1). With periodic deep breathing, especially during sleep, small cyclic

changes in deoxygenated hemoglobin, oxygenated hemoglobin, tissue blood volume, and cytochrome

aa3 are seen.

In contrast, a five-minute period of spontaneous decreased oxygenation in baby B (Fig 2) is concur-rently recorded in oxygenated hemoglobin, cyto-chrome aa3, and transcutaneous Po2. Although cerebral oxygen availability as reflected by oxygen-ated hemoglobin, improves rapidly, the cytochnome aa3, an indicator of cellular oxygenation, is much slower to recover and did not reach its base-line

state for an additional six minutes beyond the end of the tracing in Fig 2. The rapidity of aa3 reduction to decreased hemoglobin oxygenation appears to be influenced by its immediate prior state of oxygen-ation (Fig 3). When preceded by several minutes of stable oxygenation, the decline in aa3 occurs after the decrease in oxygenated hemoglobin and re-covers rapidly with the increase in oxygenated he-moglobin. With a closely following second period of decreased oxygenation, the aa3 reduction occurs with that of oxygenated hemoglobin. In this case, recovery to base-line values occurred with oxygen-ated hemoglobin recovery.

The fragility of the cerebral oxidative state with a large patent ductus arteniosus is clearly seen in Fig 4. Prior to ductal ligation (top tracings) short bradycardic episodes caused immediate reduction


oxygen-I minute

tHb V7



tHbO ‘5 V







- _


-- ----







--Cytgg3 ---

-ozid.tion$ QSZ




- --- ....---J-- -.---..___-h.._.


- 200

1cP02 - ---



Fig 1. Tracing from baby A, characteristic of stable sue blood volume (tBV), and cytochrome aa occur with monitoring period. Slight fluctuations of deoxygenated deep respirations; otherwise, little change is noted over hemoglobin (tHb), oxygenated hemoglobin (tHbO2), tis- several minutes.

I- Iminuts .

t Hb - - - - a1Yd

tHbO2 O.7d





#{176}2 -



--1-ECG . .


-- -,- --- - - -.


Fig 2. Tracing from baby B during five-minute sponta- ation is followed rapidly by a return in transcutaneous

neous episode of decreased oxygenation. Oxygenated he- Po2 to its base-line value; however, aa3 remains reduced

moglobin (tHbO2), cytochrome aa,, and transcutaneous and did not return to its base-line value for six additional

Po2 all decrease together with the increase in deoxygen- minutes. Abbreviation used is: tBV, tissue blood volume.





t By. ________________________________________________-




Cytog ---- - --- - -- - --- - ---




--reduction+ -


-- 200



- - - -- - -- - T::--



.,:, LIJ -0

---. - --1-. L.Atk.UL - I - .


-- .- - - .- TV- -w-i-’r-Y - -..-.-. -ru,.--’.T’--- -.



#{174} #{174}#{174}#{174}


#{174} #{174}#{174}#{174}#{174}


Procedures are: (1) measuring abdominal girth, (2) chang-ing diaper, (3) chest physiotherapy begins, (4) bed-tilting with chest physiotherapy, (5) baby kicking, and (6)

suc-tioning and bagging. Abbreviations used are: tHb,

deox-ygenated hemoglobin; tBV, tissue blood volume. Iminuts

-:-Fig 3. Two sequential periods of decreased oxygenation

associated with nursing care procedures in baby B. Dur-ing first episodes, reduction of cytochrome aa lags behind shift of hemoglobin to deoxygenation by one minute. Cytochrome aa3 and oxygenated hemoglobin (tHbO2)

re-turn to base-line values concurrently in both episodes.

ated hemoglobin. After ligation (bottom tracings), cytochrome aa3 remained almost unchanged during these bradycardic episodes.

Values for deoxygenated hemoglobin, oxygenated

hemoglobin, tissue blood volume, cytochrome aa,

and transcutaneous Po2 were taken from the chart

recorder at five-minute intervals and analyzed

sta-tistically to compare the overall directional trends

and general interrelationships of these variables. It is important to note, however, that events that take

place at intervals shorter than five minutes and

events in which one parameter is affected before a second are not assessed by this method of analysis.

Significant correlations were observed between

measurements made by the NIROS-SCOPE and

transcutaneous Po2 as measured by the

transcuta-neous oxygen electrode (Table). Cerebral

oxygen-ated hemoglobin correlated highly and significantly with transcutaneous Po2 in infants A and B.

Con-versely, deoxygenated hemoglobin and transcuta-neous Po2 showed a significant inverse correlation,

indicating that skin oxygen delivery and cerebral oxygen delivery were generally similar in these

in-fants. Prior to ductal ligation in infant C, there was

no correlation between transcutaneous Po2 and cerebral oxygenated hemoglobin or deoxygenated

hemoglobin, whereas after ligation, the correlation was as significant as in the other two infants.

There were marked differences in the correlation coefficients between transcutaneous P02 and

oxi-dized cytochrome aa3 among the three infants. Overall, the correlation was weak, although

statis-tically significant. The tissue blood volume showed no relationship with transcutaneous Po2 for the

combined infants. Only infant A demonstrated a weak positive relationship between these two


Correlations were observed between individual

parameters measured by the NIROS-SCOPE.

In-fants A and B and infant C after ductal ligation, showed the expected significant inverse correlation

between deoxygenated hemoglobin and oxygenated hemoglobin. However, prior to ductal ligation in

infant C, this relationship was not observed. A

significant inverse correlation between deoxygen-ated hemoglobin and oxidized cytochnome aa was noted only in infant B. Infant B was also the only

one who had repeated episodes of hypoxia and

cyanosis during the study periods. A significant

negative correlation coefficient between oxidized cytochrome aa3 and tissue blood volume was noted




tHb 0.57d




tHbO2 0.5j




tB.V. 05/



cyt gq3 :8Yd



LLi L#{149}__




200-TP02 torr100_____









::2 Q57d







iB I





L_ B




200-tort 100..


-1 a5jj








u_i L_LJ I B I

..-., 00 115

__t_ :











increaset 0.5


Cyt gg




---I00 tort


Fig 4. Before ductal ligation of baby C (top tracings),

short episodes of bradycardia cause immediate and

marked cytochrome aa reduction with only small changes in oxygenated hemoglobin (tHbO2). After ductal

ligation (bottom tracing) cytochrome aa3 remains

un-changed, despite brief shifts to hemoglobin deoxygena-tion. Abbreviations used are: tHb, deoxygenated


Infant No. of


Transcutaneous Po2 tHb


Cytochrome aa

tHbO2 tHb tBV Cytochrome tHb tHbO2 tBV


A 26 .789” .526’ .464a -.351 .605’ .195 497a

B 152 .448” #{216}#{216}9 .179#{176} -.749’ .296c .035 -.459”

C with PDA 109 .182 -.046 -.002 .321’s -.008 -.046 .009 -.074

C without PDA 57 543d 605d -.186 .074 -.020 -.021

All 344 439d -.002 163b .210c .054

* Abbreviations used are: tHbO2, oxygenated hemoglobin; tHb, deoxygenated hemoglobin; tBV, tissue blood volume; PDA, patent ductus arteriosus. Significance is indicated as follows: a, P < .05; b, P < .01; c, P < .001; d, P < .0001.

TABLE. Correlations Between Variables*

cerebral hypoxemia were to cause increase in blood volume.


The monitoring devices currently in use in neo-natal intensive cane units have markedly improved our ability to assess sick neonates. The

transcuta-neous oxygen monitor gives a continuous indication

of oxygen at that site and when combined with intermittent arterial sampling, can help assess blood oxygen content on a regular basis. Neither of

these, however, give specific information about ox-ygen delivery to the brain or cerebral oxygen

utili-zation. Nuclear magnetic resonance (NMR) and

positron emission tomography (PET), although

po-tentially useful in assessing aspects of cerebral me-tabolism and blood flow, are expensive, require pnescheduled time, and necessitate that the patient be transported away from his or her regular care

setting to the instrument. These factors consider-ably limit the usefulness of these instruments in

making moment-to-moment therapeutic decisions for treatment of sick neonates.

For effective measurement of cerebral oxygen sufficiency in a sick newborn, an instrument should be noninvasive, be adaptable to the bedside, not interfere with patient care, and give continuous rapid information. The signals should directly

as-sess brain oxygen delivery and utilization and be

sensitive to small changes. The NIROS-SCOPE appears to fulfill these basic requirements.

In baby A, the healthy infant, the correlation coefficient for oxygenated hemoglobin


tnans-cutaneous Po2 was high (0.8), indicating homoge-neity in body oxygenation. In the quiet awake state almost no fluctuations occurred in deoxygenated hemoglobin, oxygenated hemoglobin, tissue blood volume, or cytochrome aa3 (Fig 1). With sleep, periodic deep inspiration caused small changes in the parameters measured by the NIROS-SCOPE. The significance of these small increases in deoxy-genated hemoglobin, oxygenated hemoglobin,

tis-sue blood volume, and slight decrease in cytochrome

aa3 is unknown. Using an ear oximeter to study

periodic breathing, Rigatto and Brad? noted an increase in Pao2 with deep breathing. This suggests our observed increase in oxygenated hemoglobin is a real phenomenon, not movement artifact. How-ever, the expected response to increased oxygenated hemoglobin is cytochnome oxidation, which was generally not seen. We ascribe this to the inherently small secondary response of the cytochrome aa3

signal to hyperoxygenation as seen in animal

models (F. F. J#{246}bsis and M. H. Mitnick,

unpub-lished data, 1984).

During spontaneous episodes of decreased oxygen availability (decreased oxygenated hemoglobin), the cytochrome WZ3 became more reduced. The time

course, however, was variable and appeared to de-pend on three major factors: (1) the state of oxy-genation over the preceding minutes, (2) the

rapid-ity and duration of the deoxygenation, and (3) the

presence or absence of a circulatory disturbance such as a patent ductus arteniosus. If an infant experienced a long stable period of adequate oxy-genation prior to an hypoxic episode, the decrease in oxygenated hemoglobin preceded cytochnome aa3

reduction by several seconds. If the antecedent period was one of low oxygenated hemoglobin (high deoxygenated hemoglobin) or if the hypoxic epi-sodes were in rapid sequence, as in Fig 3, cyto-chrome aa3 reduction occurred simultaneous with the shift in hemoglobin to deoxygenation. Likewise, the longer the duration of deoxygenation, the greater the reduction of cytochrome aa3 and the slower the recovery of cytochrome aa3 to its base-line value. For example, with the five-minute period of relative hypoxia shown in Fig 2, oxygenated hemoglobin nose to its base-line level (end of Fig-ure) before the return of cytochnome aa, which did not occur for an additional six minutes. Studies of the oxidized and reduced forms of nicotinamide adenine dinucleotide have also shown that a pne-vious hypoxic episode influences the cerebral met-abolic response to repeated hypoxia.5


reflect the rate of anaerobic metabolism. The

oxi-dation-reduction state of cytochrome aa3 represents

a balance between oxygen supply and the flow rate

of reducing equivalents down the electron

trans-parent chain. In hypoxia, the supply of oxygen may be low relative to the flow of reducing equivalents. Furthermore, periods of hypoxia may be associated

with an increased rate of anaerobic metabolism and build up of intermediary substrates such as

pyru-vate and reducing equivalents for mitochondnial

respiration. With reoxygenation, the oxidation-re-duction state of cytochrome aa3 may remain in a relatively reduced state until the accumulated

re-ducing equivalents are oxidized. It should be

em-phasized, however, that in adult animals (cats, rats)

evidence for the hypothesized build up of

anaero-bically produced substrate supply has not been found consistently.6 A second possible explanation

is that in the preterm infant, the increase in oxy-genated hemoglobin occurs in a compartment not directly exchanging oxygen and carbon dioxide with the cells providing the cytochrome aa signal. Shunt vessels may constitute a larger fraction of the total

cerebral vascular space or, given the absence of cranial rigidity may dilate to a greater extent. Thus,

the sluggishness of cytochnome aa recovery may represent the continued presence of a low resistance pathway which limits microcirculatory flow.

The most remarkable reductions in cytochrome

aa3 occurred in baby C prior to ductal ligation with

transient reductions in heart rate, often not pro-longed enough nor low enough to trigger or sustain monitor alarms. Usually, these were not

accompa-nied by changes in transcutaneous P02 and thus,

without near infrared monitoring, the significance

of these events would have been unsuspected. Lip-man et al,7 using the transfontanel Doppler tech-nique, demonstrated that during diastole forward flow in the anterior cerebral circulation may be markedly reduced or even reversed in infants with a large patent ductus arteniosus, particularly ones with significant aortic reverse flow. Peanlman et al,8 observing the same phenomenon, postulated a “cerebral steal” by the ductus. The rapidity with

which changes in heart rate led to a shift to cyto-chrome aa3 reduction in the preligation studies in infant C (Fig 4, top) suggests that the effects of this disturbed diastolic flow are exaggerated during

bradycardia and immediately alter cerebral

oxida-tive metabolism. It is also possible that the

base-line state was one of relative cerebral hypoxia, so

any additional factor compromising oxygen delivery

abruptly reduced cytochrome aa further. For ex-ample, if the base-line measurements of oxygenated hemoglobin were already at low (hypoxic) levels, then this parameter may be unable to decrease

much further. Under these conditions, the cyto-chrome aa3 may be a more sensitive indicator of further changes in cellular availability of oxygen.

After ligation of the ductus, cytochrome aa3

re-mained stable during short or mild episodes of

bradycardia and only reduced when hemoglobin


Although oxygen is essential for cellular

respira-tion, other factors such as glucose availability, phosphate, and brain activity would all be expected to influence (UZ3 activity. Animal research using differential reflectance spectrophotometry provides us with more information. In contrast to in vitro

tissue studies, which show that cytochrome aa is not rate-limiting with respect to oxygen except at subphysiologic P02 levels, in vivo studies of cerebral

cortex have shown that cytochrome aa is not max-imally oxidized under resting conditions, but in-stead is significantly reduced.6 The usefulness of near infrared monitoring rests on this phenomenon.

The oxidation-reduction level of cytochrome aa3 is markedly affected by the presence of pulmonary

pathology9 and by other conditions. Studies of

anes-thetized cats have demonstrated that the

cyto-chrome aa3 becomes more oxidized with stimulated cortical activity, increased inspired oxygen, and increased inspired carbon dioxide.10” The

cyto-chrome becomes more reduced during hypoxia, an-oxia, and ischemia.11”2 Studies on adult human

volunteers using the reflectance spectrophotometry have confirmed WZ3 changes with variations in in-spired oxygen and carbon dioxide, and clinical stud-ies of patients during surgery have demonstrated

aa3 reduction during periods of systemic hypoten-sion.’3

Clearly, more information on neonates and

im-provement in technique are needed before

monitor-ing of cerebral oxygen sufficiency with near infrared light can be generally applied in the intensive care nursery setting. Quantitation of the changes, not just qualitative trends, needs to be achieved. The

degree of normal cytochrome reduction in neonates and the level and duration of cytochrome aa re-duction necessary to cause cerebral injury are

cur-rently unknown. However, the loss of parallel

be-havior of cytochrome aa3 and oxygenated hemoglo-bin during hypoxia and/or oligemia might well prove to be a useful threshold for defining the

potential for imminent cell injury. Neonatal animal models that can be manipulated under controlled conditions as well as observations of spontaneously occurring events in newborns such as these are

needed to gather this information and interpret clinical events. Nonetheless, these early studies of noninvasive monitoring of cerebral oxygenation


suggest that in vivo observations of cerebral

oxi-dative metabolism may help to further our

under-standing of newborn brain physiology and the changes brought about by pathologic processes and therapies.


This work was supported, in part, by Ross Laborato-ries.

We acknowledge with thanks the technical assistance

of Patty Harrington, the statistical assistance of Dr M.

Helms, and the secretarial assistance of Carol Padgett.


1. JObsis FF: Noninvasive, infrared monitoring of cerebral and

myocardial oxygen sufficiency and circulatory parameters.

Science 1977;198:1264-1267

2. American National Standard for the Safe Use of Lasers. New

York, American National Standards Institute, Inc,

publica-tion No. Z136.1, 1976

3. Rigatto H, Brady JP: Periodic breathing and apnea in pre-term infants: II Hypoxia as a primary event. Pediatrics


4. Deleted in proof

5. Mitnick MH: In vivo studies of hypoxia on cerebral mito-chondrial NADH redox state and EEG. Presented at the Annual Meeting of the Society for Neuroscience, St Louis,

Nov 5-9, 1978, vol 4, p 78

6. J#{246}bsisFF, LaManna JC: Kinetic aspects of intracellular redox reactions, in Robin ED (ed): Extrapulmonary

Mani-festations of Respiratory Disease. New York, Marcel

Dek-ker, 1978, pp 63-105

7. Lipman B, Serwer GA, Brazy JE: Abnormal cerebral he-modynamics in preterm infants with patent ductus

arterio-sus. Pediatrics 1982;69:778-781

8. Perlman J, Hill A, Volpe J: The effect of patent ductus

arteriosus on flow velocity in the anterior cerebral arteries:

Ductal steal in the premature newborn infant. J Pediatr 1981;99:767

9. Sylvia AL, Rosenthal M: The effect of age and lung pathol-ogy on cytochrome an3 redox levels in rat cerebral crotex.

Brain Res 1978;146:109-122

10. J#{246}bsis FF, Rosenthal M, LaManna JC, et a): Metabolic

activity in epileptic seizures, in Brain Work, Alfred Benzon

Symposium VIII. Copenhagen, Munksgaard, 1975

11. J#{246}bsisFF, Keizer JH, LaManna JC, et al: Reflectance

spectrophotometry ofcytochrome an3 in vivo. JAppiPhysiol


12. J#{246}bsis FF: Oxidative metabolic effects of cerebral hypoxia.

Adv Neurol 1979;26:299-318

13. Device gauges anesthetized patient’s brain 02. JAMA 1982;248:2086-2087


Institutionalizing Infant Care Review Committees, the one strategy on which the government and the American Academy of Pediatrics agree, calls for more professional input and offers no voice to parents. Far from guaranteeing equi-table and humane treatment, the committee solution leaves the definition of

infant and family interests to the vicissitudes of group dynamics among doctors,


ministers, ethicists, and representatives of handicapped groups who

bear no consequences for the final decision.

Submitted by Student




Jane E. Brazy, Darrell V. Lewis, Michael H. Mitnick and Frans F. Jöbsis vander Vliet


Noninvasive Monitoring of Cerebral Oxygenation in Preterm Infants: Preliminary


Updated Information &


including high resolution figures, can be found at:

Permissions & Licensing


entirety can be found online at:

Information about reproducing this article in parts (figures, tables) or in its






Jane E. Brazy, Darrell V. Lewis, Michael H. Mitnick and Frans F. Jöbsis vander Vliet


Noninvasive Monitoring of Cerebral Oxygenation in Preterm Infants: Preliminary


the World Wide Web at:

The online version of this article, along with updated information and services, is located on

American Academy of Pediatrics. All rights reserved. Print ISSN: 1073-0397.


Fig 2.TracingatedneousmoglobinPo2frombabyBduringfive-minutesponta-episodeof decreasedoxygenation.Oxygenatedhe-(tHbO2),cytochromeaa,,andtranscutaneousalldecreasetogetherwiththeincreaseindeoxygen-hemoglobin(tHb).Recoveryof hemoglobinoxygen-
Fig 3.Twoassociatedturnsequentialperiodsof decreasedoxygenationwithnursingcareproceduresinbabyB.Dur-ingfirstepisodes,reductionof cytochromeaa�lagsbehindshiftofhemoglobintodeoxygenationbyoneminute.Cytochromeaa3 andoxygenatedhemoglobin(tHbO2)re-tobase-linevaluesconcurrentlyinbothepisodes.
Fig 4.BeforechangesductalligationofbabyC(toptracings),shortepisodesofbradycardiacauseimmediateandmarkedcytochromeaa�reductionwithonlysmallin oxygenatedhemoglobin(tHbO2).Afterductal


Related documents

reported the first published case of an adult toxic epidermal necrolysis patient with ureteropelvic mu- cosa damage, haematuria, extensive ureteral mucosal sloughing, and acute

The aim of this research is to investigate the comfort and safety of a novel device for measuring P-A trunk stiffness in a sample of young adults.. Methods: A sample of young

Also, both diabetic groups there were a positive immunoreactivity of the photoreceptor inner segment, and this was also seen among control ani- mals treated with a

penicillin, skin testing for immediate hyper-.. sensitivity to the agent should

ings were observed in the renal tissue of the Cis group with significantly increased levels of oxidants and IL-1β, such as wide hemorrhage areas, dilated and congested blood

1. Gonzalez-Diaz, H., Vilar, S., Santana, L., Uriarte, E., 2007, &#34; Medicinal chemistry and bioinformatics--current trends in drugs discovery with networks topological

Parents, hybrids, parents against hybrids, the ability to combine plants (GCA) and genotypes, which include the unique ability of hybrids, mean squares in the analysis of

The paper is discussed for various techniques for sensor localization and various interpolation methods for variety of prediction methods used by various applications