Paediatrics and Child Health April2012

EgyptianPediatrics Yahoo Group

http://health.groups.yahoo.com/group/EgyptianPediatrics/

Paediatrics andChildHealthPaediatrics and Child Healthisthecontinuouslyupdatedreviewofpaediatricsandchildhealth(formerlyCurrent Paediatrics)

Paediatrics and Child Health isanauthoritativeandcomprehensiveresourcethatprovidesallpaediatriciansandchildhealthcarespecialistswithup-to-datereviewsonallaspectsofhospital/communitypaediatricsandneonatology,includinginvestigationsandtechnicalproceduresina4-yearcycleof48issues.Theemphasisofthejournalisontheclear,concisepresentationofinformationofdirectclinicalrelevancetobothhospitalandcommunity-basedpaediatricians.Contributorsarechosenfortheirrecognizedknowl-edgeofthesubject.ThejournalisabstractedandindexedinCurrentAwarenessinBiologicalSciences.Thelayoutofthejournal,includingthedesignandcolour,enablesfastassimilationofkeyinformation.Foreaseofreference,Paediatrics and Child Healthisavailableinprintandonlineformats.

Editor-in-ChiefPatrick Cartlidge DM FRCP FRCPCH

SeniorLecturerinChildHealthandHonoraryConsultantNeonatologist,WalesCollegeofMedicine,CardiffUniversity,Cardiff,UK

Visitourwebsiteat:www.paediatricsandchildhealthjournal.co.ukforpreviousissues,subscriptioninformationandfurtherdetails.

Paediatrics and Child Health hasaneminenteditorialboardandawidearrayofauthors,allofwhomarerecognizedexpertsintheirfield.

International Advisory BoardR Adelman (Phoenix,USA) A Bagga(NewDelhi,India)

Z Bhutta (Karachi,Pakistan) H Buller (Rotterdam,TheNetherlands)

MC Chiu (Kowloon,HongKong) M Hassan (Islamabad,Pakistan)

P Malleson (Vancouver,Canada) A Martini (Genova,Italy)

A Moosa(Saffat,Kuwait) C Morley(Carlton,Australia)

BJC Perera(Colombo,SriLanka) J Pettifor(Johannesburg,SouthAfrica)

M Uchiyama(Niigata,Japan) M van de Bor(Nijmegen,TheNetherlands)

Founding EditorRichard WilsonMB FRCP FRCPCH DCH

Associate EditorsAllan ColverMA MD FRCPCH

ProfessorofCommunityChildHealth,SirJamesSpenceInstitute,NewcastleUniversity,Newcastle,UK

Harish VyasDM FRCP FRCPCHProfessorinPICUandRespiratoryMedicine,Queen’sMedicalCentre,NottinghamUniversityHospital,UK

Doug SimkissBMedSci MBChB DCH DTMH MSc FRCP (Ed) FRCPCH FHEA

AssociateProfessorinChildHealth,WarwickMedicalSchool,Warwick,UKHonoraryConsultantPaediatrician,BirminghamCommunityHealthCareNHSTrustandSandwellandWestBirminghamNHSTrust,Birmingham,UK

Nicholas MannMD FRCP FRCPCH DCHConsultantPaediatrician,DepartmentofPaediatrics,RoyalBerkshireHospital,Reading,UK

Alistair ThomsonMA MD BChir FRCPCH FRCP DCH DRCOGConsultantPaediatrician,LeightonHospital,Crewe,UK

Colin PowellMBChB DCH MRCP FRACP FRCPCH MD

ConsultantPaediatrician,UniversityHospitalofWales,Cardiff,UK

Peter HeinzMD State Exam Med FRCPCH

ConsultantPaediatrician,Addenbrooke’sHospital,Cambridge,UK

SYMPOSIUM: NEONATOLOGY

Temperature monitoring andcontrol in the newborn babyYvonne Freer

Andrew Lyon

AbstractThe importance of keeping the newborn baby warm has been known for

centuries but worldwide in the 21st century hypothermia remains a major

contributor to neonatal mortality. Although less of a problem in high

income countries there is evidence that low temperatures have an impact

on outcome at vulnerable times, particularly in the baby born preterm. It

is clear that if we are to see further improvements in mortality and

morbidity in the most immature babies there must be careful attention

given to all aspects of basic neonatal care, including thermoregulation.

Continuous dual temperature monitoring has advantages over intermit-

tent measurements and is the method of choice in the immature and

sick newborn. There is no evidence of any differences in outcome

between radiant heaters or incubators. Whichever device is used fluid

and heat loss from evaporation due to high transepidermal water loss

remains a problem. This is best managed by increasing environmental

humidity but the optimum level of added humidity, and the length of

time that this should be applied, is still unknown.

Keywords temperature monitoring: devices and methods; temperature

support: devices and techniques

Introduction

Outcome for the newborn, and in particular the preterm, baby

improved dramatically in the latter half of the 20th century. There

were many contributory factors with the understanding of

temperature control being one of the major influences. Since the

mid 1990s there has been much slower, if any, change in overall

rates of mortality or morbidity. It is unlikely that the dramatic

changes of the past will be seen again and further improvements in

outcomewill bemore difficult to achieve and of a smaller scale. It is

increasingly important that careful attention is given to the basics

of neonatal care and that lessons from the past are not forgotten.

Thermoregulation

Heat is produced as a by-product of cell metabolism and is lost or

gained with the environment through conduction, radiation,

Yvonne Freer RGN RM RSCN BSc Midwifery PhD is Clinical Reader in the

Neonatal Intensive Care Unit at the Simpson Centre for Reproductive

Health, The Royal Infirmary of Edinburgh, Edinburgh, UK. Conflict of

interest: none.

Andrew Lyon MA MB FRCP FRCPCH is Consultant Neonatologist in the

Neonatal Intensive Care Unit at the Simpson Centre for Reproductive

Health, The Royal Infirmary of Edinburgh, Edinburgh, UK. Conflict of

interest: none.

PAEDIATRICS AND CHILD HEALTH 22:4 127

convection and evaporation. Children and adults are homeo-

thermic, maintaining a constant deep body temperature over

a wide range of ambient thermal conditions. The newborn infant,

by comparison, can only achieve control of temperature over

a narrower range of ambient conditions. The preterm infant has

even greater difficulty and the most immature behave at times as

if they are poikilothermic e their body temperature drifting up

and down with the ambient temperature. The range of ambient

temperature over which an infant can maintain body tempera-

ture, with minimal energy expenditure [thermoneutral range

(NTR)], is very narrow in the immature infant. As environmental

temperature moves outside this range, the infant adopts different

strategies to maintain normothermia.

If the environment is cooler than the body, metabolic heat

production increases. Catecholamine release stimulates the

oxidation of brown adipose tissue distributed in the neck,

between the scapulae and along the aorta (non-shivering ther-

mogenesis). The term baby can alter body posture and skin blood

flow reduces as the superficial capillaries constrict. As environ-

mental temperature falls further outside the NTR, heat produc-

tion reaches a maximum and below this point deep body

temperature falls. If the environment is hotter than the body,

heat is gained through conduction and radiation, as in the use of

skin to skin care and radiant heaters. When above the NTR,

sweating occurs in the term infant.

Heat production is delayed during adaptation to extrauterine

life, especially if there is immaturity, asphyxia, hypoxia or

maternal sedative administration. The preterm infant, particu-

larly below 28 weeks’ gestation, has lower heat production per

unit area and a more prolonged impairment of non-shivering

thermogenesis. The immature infant is further disadvantaged

because of increased evaporative heat losses e a consequence of

high transepidermal water loss (TEWL) due to passive diffusion

of water through a thin, poorly keratinized epidermis. The ability

to alter skin blood flow and change posture are also impaired in

the preterm infant as well as in the presence of illness. Sweating

is delayed in the most immature newborns by 2e3 weeks,

a result of neurological rather than glandular immaturity.

Thermoregulation and outcome

William Silverman and others showed, in a series of randomized

controlled trials, that keeping small babies warm could result in

an absolute reduction in mortality of at least 25% of that seen in

the 1950s.This improvement was seen over all gestation and

birthweight groups. The importance of humidity was recognized

and, in the 1970s, Hammarlund and Sedin published data on the

heat fluxes due to transepidermal water losses in the preterm

baby. Recommendations for optimum environmental tempera-

ture settings, based on the concept of the neutral thermal envi-

ronment, were developed and technological advances have

improved the devices used to keep small babies warm.

By the 1990s the impression was that thermoregulation of the

newborn was understood and well managed but, worldwide

today, hypothermia is still a major cause of death after birth. The

extent of this problem is such that the World Health Organization

(WHO) has published guidelines on the management of the

newborn aimed at reducing the deaths from hypothermia (http://

whqlibdoc.who.int/hq/1997/WHO_RHT_MSM_97.2.pdf).

Crown Copyright � 2011 Published by Elsevier Ltd. All rights reserved.

http://whqlibdoc.who.int/hq/1997/WHO_RHT_MSM_97.2.pdf

http://whqlibdoc.who.int/hq/1997/WHO_RHT_MSM_97.2.pdf

http://dx.doi.org/10.1016/j.paed.2011.09.002


Recent evidence has shown that poor management of temper-

ature in the baby at vulnerable times can impact on outcome.

Whether as a cause or consequence, low body temperature in

newborns lead to increased metabolism and hypoglycaemia,

reduced tissue perfusion, ischaemia and metabolic acidosis and

has been shown to be inversely related to mortality.

The effects of hyperthermia are less well understood but in

infants with moderate-to-severe hypoxic ischaemic encephalop-

athy, hyperthermia is associated with an increased risk of death

or moderate-to-severe disability. There is little data on the effects

of hyperthermia and the human preterm infant but in the preterm

animal model, hyperthermia is associated with severe lung injury

and increased inflammatory cytokine expression.

It is clear, more than 50 years after the work of Silverman, that

thermal control, particularly of the immature infant, is still an

important issue in need of further thought and study. Given the

importance of thermoregulation two questions are paramount:

how best to measure temperature and how best to maintain

thermal balance.

Temperature measurement

The normal temperature

In day-to-day care the only means of assessing the thermal state of

a baby is by measuring body temperature. Defining a ‘normal’

body temperature is difficult as this will depend on where, how,

and the time of day/nightwhenmeasured. The temperaturewithin

the tissues of the body varies with metabolic rate and there is no

such thing as a single central body temperature. Normal temper-

ature ranges for newborns have not been clearly established but

published ranges for both term and preterm infants are:

� rectal at 36.5e37.5 �C� axillary at 35.6e37 �C.It has been suggested that conditions for thermoneutrality are

met in very low birthweight infants when core temperature is

between 36.7 and 37.3 �C and the central-skin temperature

difference fluctuates less than0.2e0.3 �C/hour.An infant may expend a large amount of energy to maintain

a ‘normal’ central temperature. The preterm baby is at higher risk as

a consequence of the very narrow NTR when compared with those

bornat term.Despite anapparentnormal temperature the infantmay

well be thermally stressed and at increased risk of adverse outcome.

How, where and frequency of temperature measurement

Theprimary purpose ofmeasuring the newborn’s temperature is to

detect cold stress as fever is an unusual symptom of illness and

most often influenced by environmental factors. Treatment is often

initiated on relatively small changes in temperature so devices

used for measurement must be accurate, reliable and easy to use.

In the newborn sites of measurement are: rectum, axilla or skin,

although these all offer only an estimate of body core temperature.

Devices

Mercury in-glass thermometers were used for many years

however with concerns about accuracy, the length of time to

reach a stable point and risk of harm posed by mercury they are

no longer used in most high income countries, except as

a research tool for comparing new devices. More automated

thermometers have become available that utilize electronic,

infrared, chemical and liquid crystal technologies.


Electronic thermometers have temperature sensors inside the

tip and are covered with a sheath. They are used in monitoring or

predictivemode. In themonitormode the temperature is displayed

once a steady state has been achieved, usually about 3e5 min.

With predictive mode the temperature is ‘predicted’ by a calcula-

tion based on the rate of rise in the few seconds of use. Tympanic

thermometers have an infrared sensor that records heat radiated

from the tympanic membrane. Temporal artery thermometers

work by detecting changes in heat emitted from the superficial

temporal artery (STA). The thermometer is moved across the

forehead and as it passes over the STA there is a peak in the emitted

temperature which is captured by the sensor. Chemical/liquid

crystal thermometers work by placing a plastic strip, impregnated

with temperature sensitive chemicals/crystals, against the skin.

These change colour in response to variations in temperature.

Site and frequency

Intermittent rectal thermometry is used infrequently (the excep-

tion is during monitoring of therapeutic hypothermia where

a continuous readout is given). As well as concerns about rectal

trauma the reproducibility under clinical conditions is uncertain.

This is affected by differences in the depth of insertion, dwell

time, whether the baby has just passed stool and on the flow and

temperature of the blood returning from the lower limbs. Axillary

temperature is usually measured intermittently. There is little

associated risk but error in measurement is observed depending

on placement of the probe, adequate closure of the axillary pit,

blood flow to the axillary region and possibly activation of non-

shivering thermogenesis. Electronic probes and chemical/crystal

strips attached to the skin have little associated risk and are used

for both intermittent and continuous temperature monitoring.

Accuracy of these techniques may be affected by environmental

temperature, approximation of probe to skin and peripheral

perfusion.

Continuous monitoring using electronic probes, if placed

correctly, offer a close approximation to deep body temperature,

particularly when using the ‘zero heat flux’ principle. With this

technique a probe is placed over an area of skin from which no

heat can be lost. The skin under the probe equilibrates with the

deep body temperature as heat moves down the temperature

gradient from core to skin. In practice this can be achieved when

the baby is lying on its back on a non-conducting mattress with

a probe placed between the scapulae. Measuring temperature

intermittently gives a ‘snapshot’ of the baby’s temperature; it

tells nothing about the energy the baby may be expending to

maintain that temperature. The continuous measurement and

display of central (abdominal, axilla or zero heat flux) and

peripheral (sole of the foot) temperatures gives an early indica-

tion of thermal stress by showing a change in the centrale

peripheral difference which occurs before any alteration in

central temperature.

The preterm baby who appears to be comfortable in its

environment has a central temperature in the normal range for

whichever site is being used and a centraleperipheral tempera-

ture difference of 0.5e1 �C. An increasing centraleperipheral

temperature difference, particularly above 2 �C, is usually due to

cold stress (Figure 1), and occurs before any fall in central

temperature. A high central temperature, particularly if unstable,

along with a wide centraleperipheral gap is seen in septic babies.



Figure 1 Central temperature shown in green; peripheral temperature shown in blue. At 01.45 h the incubator door is opened, the peripheral temperature

falls dramatically with little change in the central temperature.


Variation due to device and measurement method

Accuracy of thermometers has been reviewed and according to

manufacturing standards, devices should be within �0.2 �Cacross a wide temperature range. However, many manufactures

indicate an accuracy of �0.6 �C when used in predictive mode.

Such differences could result in some intervention if the infant’s

temperature is at the margins of the normal range.

Many factors may be responsible for the failure to develop

a consensus recommendation for temperature monitoring in

neonatal care. Studies differ in thermometer, population case mix

and sample size and the external heat sources used. There are

major problems with the variation in statistical methods used

when comparing devices. Use of the correlation coefficient is

inappropriate, when comparing any device ormethod of recording

a physiological parameter, as this measures strength of a relation-

ship rather than agreement. Bland and Altman propose that by

comparing the individual differences between twomeasurements,

in the context of the mean of their combined readings, will better

assess the agreement of the device/technique. Recent studies have

compared temperature measurements comparing electronic rectal

and axillary thermometers, an infrared temporal artery thermom-

eter and an electronic axillary thermometer against an indwelling

rectal probe and an electronic and an infrared axillary skin ther-

mometer against a glass mercury thermometer using this tech-

nique in the neonatal population. The mean difference between

devices range from 0.1e0.3 �C however there are wide 95% limits

of agreement giving a variability of between 0.8 and 1.7 �C.These few small studies suggest that there is a large variation

betweenmethodsand this is of concern in clinical practice.Although

the mean difference is small, and of no consequence to clinical


decision-making, the degree of variability suggests that interven-

tionsmay be carried out unnecessarily in some infants and not at all

in others when they would be appropriate. These concerns further

strengthen the argument in favour of using continuous monitoring

of dual temperatures as a means of following trends rather than

concentrating on absolute values of intermittent measurements.

How best to maintain temperature

Simplemethods for preventing heat loss are well knowne awarm

delivery room, drying, wrapping and applying a hat, skin to skin

care or lying baby on a non-conducting mattress, breastfeeding,

warm resuscitation and transportation. Despite this knowledge

there are still reports of ‘cold’ babies in both high and low income

countries. Most trials on temperature maintenance have focused

primarily on how best to keep immature infants warm. Although

the lowest acceptable admission temperature is not known, it is

suggested that temperatures should be above 36 �C in this group.

Trials have generally looked at two approaches, barriers to heat

loss or supplemental heat application.

At birth the baby will lose heat rapidly, particularly due to

evaporation.Heat losses canbeminimized, as described above, but

many preterm babies are cold on arrival in the neonatal unit. Heat

fromradiant heaters is often insufficient to compensate for the large

losses due to evaporation. To overcome this, babies arewrapped in

polyethylene occlusive skin wraps or placed into plastic bags to

reduce TEWL. One systematic review of five studies utilizing

plasticwraps/bags and stockinet hats showed that infants less than

28 weeks gestation whowere wrapped were warmer on admission

to the NICU (WMD 0.76 �C; 95% CI 0.49, 1.03) but no such effect

was seen in the stockinet trial. However, two of the trials included




in the systematic review showed that, although admission

temperature had increased with the use of plastic wraps, over

a third of babies were still admitted with a temperature less than

36.5 �C. Another technique, the use of gel warmingmattresses, has

also been trialled. In each of these trials the warming mattress

contributed to an increase in the admission temperature of the

baby. While this is seen as an improvement, between 3 and 55%

babies were admitted to the NICU with hypothermia (temperature

less than 36.5 �C) and 1e28% of babies had an admission

temperature of more than 37.5 �C. Despite the recognized associ-

ation between low temperature and poor outcome no trial to date

has shown that measures to improve admission temperature has

resulted in lower morbidity or death before discharge.

The optimum management of the thermal environment

during ongoing neonatal care has been discussed in the litera-

ture. No study has shown any significant difference in outcome

for babies nursed using either radiant heater or incubator. It is

important that units consider seriously the management of the

thermal environment of the preterm baby but the choice of

device used is a matter of individual preference. However,

whichever method is adopted, evaporative fluid losses remain

a major challenge in the management of the preterm baby.

The skin barrier of the preterm baby is immature and there is

a high water concentration gradient between the body and the

external environment. This results in TEWL with the gradient

being very steep if the ambient water vapour pressure is low. The

more immature the baby the steeper the gradient and higher the

losses, and this is exacerbated if the skin is further damaged

during neonatal care procedures. The skin matures rapidly after

birth and, in practical terms, TEWL falls to around that of the

term baby within 10e14 days of age. The most effective method

of reducing TEWL is by an increase in environmental humidity

around the baby. In the baby under 28 weeks’ gestation this

should be maintained for at least the first 10 days of life and is

most easily achieved by the use of humidified incubators. A

plastic cover can be used with radiant heaters but it must be

remembered that there will be rapid fluid losses whenever this is

removed for any procedure. Covering the skin with a semi-

permeable/impermeable membrane, or the use of emollients,

creates a barrier reducing TEWL. Whilst these have been shown

to reduce TEWL, in some studies they have been associated with

an increased incidence of bacterial and fungal infection.

Movement of water through the skin is important in the

maturation process however the optimum level of environmental

humidity is as yet unknown as prolonged exposure to relatively

high ambient humidity delays the establishment of an effective

skin barrier structure and function.

Practice points

C Methods of measuring temperature are not interchangeable and

it is vital that staff understand the limitations of the devices they

use if unnecessary treatment changes are to be avoided.

Conclusions

The importance of keeping babies warm has been recognized for

centuries. However, even in the 21st century, our understanding of

what is a normal temperature and how best to measure it still

remains a challenge. In clinical practice decisions must be made on

which method of measurement to use, whether intermittent or

continuous monitoring is appropriate and how the data are inter-

preted and acted upon. There are, as yet, no good data to guide us in

deciding on the optimum management of the baby’s external envi-

ronment. What is clear is that this important area of care can no


longer be considered to be a ‘problem solved’. A practical approach

is to try and understand what it is that you are trying to achieve,

knowing the advantages, disadvantages and limitations of the

instruments and techniques that are available to you andnot assume

approaches for assessing a temperature are interchangeable. A

FURTHER READING

Almeida PG, Chandley J, Davis J, Harrigan RC. Use of the heated gel

mattress and its impact on admission temperature of very low birth-

weight infants. Adv Neonatal Care 2009; 9: 34e9.

Crawford DC, Hicks B, Thompson MJ. Which thermometer? Factors influ-

encing best choice for intermittent clinical temperature assessment.

J Med Eng Technol 2006; 30: 199e211.

Flenady VJ, Woodgate PG. Radiant warmers versus incubators for regu-

lating body temperature in newborn infants. Cochrane Database Syst

Rev 2003; 4: CD000435.

Hafis Ibrahim CP, Yoxall CW. Use of self-heating gel mattresses eliminates

admission hypothermia in infants born below 28 weeks gestation. Eur

J Pediatr 2010; 169: 795e9.

Hissink Muller PCE, van Berkel LH, de Beaufort AJ. Axillary and rectal

temperature measurements poorly agree in newborn infants.

Neonatology 2008; 94: 31e4.

Jirapaet V, Jirapaet K. Comparisons of tympanic membrane, abdominal

skin, auxiliary, and rectal temperature measurements in term and

preterm neonates. Nurs Health Sci 2000; 2: 1e8.

Lee G, Flannery-Bergey D, Randall-Rollins K, et al. Accuracy of temporal

artery thermometry in neonatal intensive care infants. Adv Neonatal

Care 2011; 11: 62e70.

LeslieA,WardleSP,BudgeH,MarlowN,BrocklehurstP. Randomised controlled

trial of gel warming mattresses to prevent hypothermia during the resus-

citation at birth of premature infants. In: http://www.neonatalsociety.ac.uk/

abstracts/lesliea_2007_gelwarmingmattressesresuscitation.shtml

Lyon AJ, Freer Y. Goals and options in keeping preterm babies warm. Arch

Dis Child Fetal Neonatal Ed 2011; 96: F71e4.

McCall EM, Alderdice F, Halliday HL, Jenkins JG, Vohra S. Interventions to

prevent hypothermia at birth in preterm and/or low birthweight

infants. Cochrane Database Syst Rev 2010;(3). Art. No.: CD004210.

McCarthy LK, O’Donnell CP. Warming preterm infants in the delivery room:

polyethylene bags, exothermic mattresses or both? Acta Paediatr

2011. doi:10.1111/j.1651-2227.2011.02375.x.

Rosenthal HM, Leslie A. Measuring temperature of NICU patients-A

comparison of three devices. J Neonatal Nurs 2006; 12: 125e9.

Singh A, Duckett J, Newton T, Watkinson MJ. Improving neonatal unit

admission temperatures in preterm babies: exothermic mattresses,

polythene bags or a traditional approach? J Perinatol 2010; 30: 45e9.

Simon P, Dannaway D, Bright B, et al. Thermal defense of extremely low

gestational age newborns during resuscitation: exothermic mattresses

vs polyethylene wrap. J Perinatl 2011; 31: 33e7.

Watkinson M. Temperature control of premature infants in the delivery

room. Clin Perinatol 2006; 33: 43e53.


http://www.neonatalsociety.ac.uk/abstracts/lesliea_2007_gelwarmingmattressesresuscitation.shtml

http://www.neonatalsociety.ac.uk/abstracts/lesliea_2007_gelwarmingmattressesresuscitation.shtml



Investigation, prevention andmanagement of neonatalhypoglycaemia (impairedpostnatal metabolicadaptation)Jane M Hawdon

AbstractBlood glucose levels fall in the hours after birth in all babies but for most

babies the normal process of neonatal metabolic adaptation mobilizes alter-

native fuels (eg ketone bodies) from stores so that the physiological fall in

blood glucose is tolerated. However, some babies are at risk of impaired

neonatal metabolic adaptation and for these babies it is important to prevent

hypoglycaemia, to recognize clinically significant hypoglycaemia, and to treat

it without causing unnecessary separation of mother and baby or disruption

of breast feeding. Investigations for underlying cause of hypoglycaemia

should be performed if hypoglycaemia is persistent, resistant or unexpected.

Keywords alternative fuels; blood glucose monitoring; breast feeding;

hypoglycaemia; neonatal metabolic adaptation

Introduction

Much debate surrounds neonatal hypoglycaemia in terms of the

definition of the condition, its clinical significance and its optimal

management. This is in part because there is a continuum

between the normal postnatal metabolic changes, with a physio-

logical fall in blood glucose after birth accompanied by protective

metabolic responses, and the more worrying situations where

there is delay or failure of the normal metabolic adaptation to

birth. Therefore, hypoglycaemia cannot strictly be applied as

a pathological diagnostic term and it is preferable to consider

a diagnosis of impaired metabolic adaptation. Invariably

“neonatal hypoglycaemia” is used as a shorthand term for this. It

is important to prevent potentially damaging hypoglycaemia in

vulnerable babies, but this must be balanced against the risks of

overly invasive management e separation of mother and baby,

placing at risk the establishment of breast feeding, and unnec-

essary administration of formula or intravenous glucose which in

turn impair metabolic adaptation to postnatal life.

Metabolic changes at birth

During pregnancy, the human fetus receives from its mother via

the placental circulation a supply of substrates necessary for

Jane M Hawdon MA MBBS MRCP FRCPCH PhD is Consultant neonatologist

with the University College London Hospitals NHS Foundation Trust,

London, UK. Conflict of interest: none.


growth, for the deposition of fuel stores which are essential after

birth, and for energy to meet the basal metabolic rate and

requirements for growth. When the continuous flow of nutrients

from the placenta is abruptly discontinued at birth, immediate

postnatal metabolic changes preserve fuel supplies for vital organ

function. The newborn infant must adapt to the fast-feed cycle and

to the change in major energy source, from glucose transfer across

the placenta to fat released from adipose tissue stores and ingested

withmilk feeds. After birth, plasma insulin levels fall and there are

rapid surges of catecholamine and pancreatic glucagon release.

These endocrine changes switch on the essential enzymes for

glycogenolysis (the release of glucose stored as glycogen in liver,

cardiac muscle and brain), for gluconeogenesis (glucose produc-

tion from 3-carbon precursor molecules by the liver), lipolysis

(release of fatty acids from adipose tissue stores), and ketogenesis

(the b oxidation of fatty acids by the liver). Some tissues, for

example the kidney, are obligate glucose users but others burn

fatty fuels to provide energy. Of the organs that utilize alternative

fuels to glucose, the brain is the most important in that it takes up

and oxidizes ketone bodies at higher rates than seen in adults, and

the neonatal brain uses ketone bodies more efficiently than

glucose. Lactate has also been identified as an alternative fuel.

In clinical terms, low blood glucose concentrations are

commonly found during the first postnatal days in healthy AGA

term neonates, particularly those who are breast fed. However,

these infants have high ketone body levels when blood glucose

concentrations are low, and it is likely that these alternative fuels

protect them from neurological injury.

Clinical significance of impaired metabolic adaptation

In some circumstances (see below), such as following preterm

delivery, intrauterine growth restriction, perinatal hypoxia-

ischaemia or suboptimal control of diabetes in pregnancy, there

may be impaired ketone body production and in these babies

circulating blood glucose concentrations acquire greater clinical

significance and hypoglycaemia, if present, must be diagnosed

and treated effectively.

No study has yet satisfactorily addressed the duration of

absent or reduced availability of metabolic fuels which is harmful

to the human neonate. Animal studies indicate that hours (rather

than minutes) of hypoglycaemia are required to cause injury, and

that injury is unlikely to occur if there are no abnormal clinical

signs. For babies in whom prolonged neonatal hypoglycaemia

has been associated with abnormal clinical signs (most usually

hypotonia, reduced level of consciousness or fits) adverse long-

term outcomes have been reported. There is evidence from

case reports that profound and prolonged hypoglycaemia is

associated with both transient and permanent structural changes

in the brain. Grey matter damage is most commonly reported

with the parieto-occipital regions being most affected.

Causes of impaired neonatal metabolic adaptation

Insufficient availability of glucose and alternative fuels

Preterm birth: the preterm baby has not had sufficient time in

utero to lay down glycogen and adipose tissues stores. In addi-

tion, hormonal and enzyme adaptive responses may by imma-

ture or the baby may have systemic conditions which affect

hepatic function and glucose production, eg severe infection.

� 2011 Elsevier Ltd. All rights reserved.



Intrauterine growth restriction (IUGR): it is important to use

this term rather than “small for gestational age” because not all

IUGR infants will have birthweights on a low centile. Conversely,

not all small for gestational age infants will have been subject to

placental insufficiency e they may be constitutionally small and

will not experience impaired postnatal metabolic adaptation. The

baby who has experienced IUGR has reduced stores of carbo-

hydrate and fat, these fuels were required for metabolism in fetal

life. Therefore, the IUGR baby is at risk of hypoglycaemia prior to

the successful establishment of milk feeds and may have reduced

availability of alternative fuels for cerebral metabolism.

However, it has been shown that healthy breast fed IUGR babies

can mount a ketogenic response, and that excessive formula milk

supplementation is associated with a suppressed response.

Perinatal hypoxia-ischaemia: the high metabolic requirement for

anaerobicmetabolismwill reduceendogenous fuel stores if the fetus

is exposed to significant hypoxia-ischaemia. In addition, hypoxic

liver damage will reduce the activity of the counterregulatory

metabolic responses. Although concurrent hypoglycaemia and

hypoxia-ischaemia are more damaging than either insult alone,

there is no evidence that hypoglycaemia following cessation of

a hypoxic-ischaemic insult worsens hypoxic-ischaemic injury.

Systemic conditions: any condition which increases metabolic

demands (eg hypothermia, systemic infection), or which affects

adequacy of feeding, or which affects perfusion or function of the

gut of liver places the baby at risk of impairedmetabolic adaptation.

If hypoglycaemia is diagnosed, there must be urgent investigation

for underlying conditions and appropriate management of these.

Inborn errors of metabolism and endocrine insufficiency:

these conditions are rare, but for affected individuals frequently

present in the neonatal period when nutrient intake is low. The

most common metabolic disorders presenting at this time are

defects of b oxidation of fatty acids. The most common

congenital endocrine disorders presenting with neonatal hypo-

glycaemia are defects in cortisol production.

Maternal medication: maternal beta blocker therapy has been

associated with impaired neonatal metabolic adaptation,

although passage across the placenta and in breast milk is vari-

able. This is not a contraindication to breast feeding. Often, the

baby of the mother with hypertension also has IUGR, thus

increasing the risk.

Prolonged starvation: as described above, various factors affect

the sufficiency of endogenous fuel stores at the time of birth. If

exposed to prolonged inadequacy of nutrient intake, even the

healthy well grown baby will run out of endogenous stores and

metabolic adaptation will fail.

Neonatal hyperinsulinism

If the fetal insulin levels are raised and do not fall after birth, or if

there is excessive insulin release from the neonatal pancreas, the

actions of insulin are to increase glucose uptake into cells,

suppress endogenous glucose production, and suppress release of

fat from adipose tissue stores. In these circumstances the baby is at

risk of hypoglycaemia and an absence of alternative metabolic


fuels. Clinical features are that glucose requirements to maintain

normoglycaemia are high, in excess of 8 mg/kg/min, as compared

to the 4e6 mg/kg/min usually required by neonates, and the

infant may be macrosomic if hyperinsulinism was of fetal origin.

Maternal diabetes mellitus: for babies born after diabetes in

pregnancywhich has not beenwell controlled, the postnatal fall in

blood glucose concentration is more prolonged or becomes clini-

cally significant and this is the most common cause of neonatal

hyperinsulinaemic hypoglycaemia. Fetal and neonatal hyperin-

sulinism may occur after maternal type 1 or type 2 diabetes or

diabeteswhose onset is in pregnancy, and is the result of increased

placental transfer of glucose and other nutrients stimulating

increased fetal insulin secretion. For affected babies, plasma

insulin levels usually fall to normal within 12e24 h of birth and

this form of hypoglycaemia presents early and is self-limiting.

Congenital hyperinsulinaemic hypoglycaemia (HH): although

a rare condition, this is the most common cause of recurrent and

persistent hypoglycaemia in infancy and childhood. There are

a number of underlying pathologies and understanding of the

molecular and genetic basis of these is becoming more clear. HH

is usually associated with macrosomia and high glucose

requirements. The condition may be self-limiting in the neonatal

period or extend beyond this time. As there is no protective

ketone body response to hypoglycaemia, there are usually

neurological signs and the risk of brain injury is high. Therefore,

urgent treatment is required (see below).

BeckwitheWiedemann syndrome: this condition is character-

ized by exomphalos, macroglossia, visceromegaly, earlobe

abnormalities and an increased later incidence of malignancies.

Hyperinsulinism is a common but not invariable feature which

usually resolves in the days after birth.

Other causes: transient hyperinsulinism has also been reported

in association with perinatal hypoxia-ischaemia, intrauterine

growth restriction and rhesus haemolytic disease, although the

mechanisms for this have not been determined. Maternal thia-

zide diuretic use may cause neonatal hyperinsulinism.

Iatrogenic or factitious hyperinsulinism: hyperinsulinism may

result from erroneous or malicious administration of insulin.

Although rare, these circumstances should be suspected if

hypoglycaemia is unexpected, profound or resistant to treatment.

Diagnosis of clinically significant hypoglycaemia

Much controversy and confusion has surrounded the definition

of hypoglycaemia. Factors which should be considered are the

blood glucose concentration considered to be the minimum safe

level, the duration beyond which the low blood glucose level is

considered to be harmful, the presence of clinical signs, the

group of infants studied, the consideration of alternative fuel

availability, the conditions of sampling and the assay methods.

Most of these have not been adequately addressed in scientific

studies. Therefore, a pragmatic approach based upon thresholds

for intervention has been proposed. If there are neurological

signs in association with low blood glucose levels there should be




urgent investigation for underlying cause (Table 2) and institu-

tion of treatment. For infants without clinical signs but at risk of

impaired metabolic adaptation (Table 1), intervention to raise

blood glucose should be considered if two consecutive blood

glucose levels are below 2 mmol/litre (measured using accurate

device) or a single blood glucose level is below 1 mmol/litre.

Detection and management of neonatal hypoglycaemia

Clinical scenario A

Sick baby or very preterm in NNU A

I

Any baby with neurological signs A

- Abnormal tone F

- Decreased activity A

- Lethargic/reduced level of consciousness I

- Abnormal cry

- Seizures

Babies “at risk”: E

R

c

- SGA (<2nd centile)

- Clinically wasted infants

- Infants of diabetic mothers where antenatal glucose

control suboptimal and/or if baby macrosomic

- Post mature infants if wasted

- Perinatal hypoxia-ischaemia

- Severe Rhesus disease

- Preterm infants (<37/40)

- Congenital heart disease

- Infection

- Hypothermia

- Fluid restriction

- Maternal beta blocker (eg labetalol)

What to do if “at risk” and: E

C No abnormal signs and feeding well, BG> 2.0 mmol/l N

p

What to do if “at risk” and: C

AC No abnormal signs and feeding well, BG 1.0e2.0 mmol/l

on 2 consecutive pre-feed measurements

C

T

What to do if “at risk” and: A

t

C BG< 1.0 mmol/l

Healthy term infants N

Includes LGA babies who are not: E

- Macrosomic appearance or I

w

c

- Infants of diabetic mothers

NNU, neonatal unit; EBM, expressed breast milk; BG, blood glucose; LGA, large for g

Table 1


It is well known that glucose reagent strips, commonly used in

neonatal and maternity units, are insufficiently reliable for the

diagnosis. If samples are to be sent to a distant laboratory, blood

glucose levels diminish with time, even in fluoridated tubes.

Therefore, a ward based accurate glucose analyser should be

available to allowprompt and accurate blood glucosemeasurement.

ction

im to maintain BG> 2.5 mmol/l

f hyperinsulinism suspected aim to maintain glucose >3 mmol/l

dmit to NNU

ull investigations including BG

im to maintain glucose >2.5 mmol/l

f hyperinsulinism suspected aim to maintain glucose >3 mmol/l

ncourage breastfeeding as soon as possible after birth (if appropriate)

egular BG measurements e approximately 3e4 hourly pre-feed,

ommencing before the second feed after birth

ncourage breast feeding and observe

o formula supplements if breast fed. Check pre-feed BG 3e4 h later

re-feed or sooner if abnormal signs (as above)

ontinue breast feeding.

dd EBM and/or formula supplement (initially 10 ml/kg/feed)

heck pre-feed 3e4 h later or sooner if abnormal signs

itrate volume of supplements against 3e4 hourly pre-feed BG

dmit to NNU for investigations and IV glucose, continue feeds if

olerated

o glucose monitoring

ncourage and support breast feeding

f concerns around feeding assess for abnormal signs and clinical

ellbeing. Investigate (including infection screen and BG) only if clinical

oncerns

estational age; IV, intravenous; SGA small for gestational age.




Prevention and management of neonatal hypoglycaemia

Healthy, full-term appropriate weight for gestational age

(AGA) neonates

As described above, healthy full-term AGA neonates often have

low blood glucose concentrations in the first postnatal days, but

are protected by the presence of ketone bodies and lactate as

alternative fuels. These babies do not need routine blood glucose

monitoring or formula supplementation of breast feeds. However,

staff should be alert to systemic conditions (eg neonatal infection)

which may impact upon feeding and the risk of neonatal hypo-

glycaemia, or the very rare risk that a baby who is apparently

healthy at birth may have an underlying metabolic disorder.

Appropriate investigations, including blood glucose measurement,

should be carried out for any baby who presents with abnormal

clinical signs.

Babies at risk of impaired metabolic adaptation (Table 1)

The first step in management is to identify these babies. Some

risk factors will be clear, for example the preterm baby, others

may require more detailed clinical evaluation (for example

fat and muscle wasting arising from intrauterine growth

restriction).

At-risk babies should have regular clinical monitoring to

include feeding behaviour and pre-feed blood glucose moni-

toring (approximately 4 hourly). It is imperative that any infant

with neurological signs should have urgent, accurate blood

glucose measurement. Monitoring should commence before the

second feed (ie not so soon after birth that the physiological fall

in blood glucose level causes confusion and over-treatment) and

pre-feed monitoring should be continued until the infant has had

at least two satisfactory measurements. Monitoring should be

recommenced if the infant’s clinical condition worsens or energy

intake decreases. If monitoring is by reagent strip, low levels

must be confirmed promptly by accurate measurement (see

above).

The importance of early milk feeding has been appreciated for

many years. Both breast and formula milks provide important

gluconeogenic precursors and fatty acids for b oxidation.

Therefore, all infants who are expected to tolerate enteral feeds

should be fed with milk as soon as possible after birth, and at

frequent intervals thereafter. Babies who are capable of sucking

should be offered the breast at each feed (if this is the mother’s

wish). If it is likely that babies will need supplementary formula

feeds, maternal breast milk expression should be encouraged.

The requirement for formula feeds must be titrated against the

clinical condition of the baby, blood glucose monitoring, and the

supply of maternal breast milk. In the breast fed baby, formula

intake should be kept to the minimum necessary, so as to

enhance breast feeding and avoid suppression of normal meta-

bolic adaptation.

In the at-risk baby who is establishing oral feeds there is

a potential nadir at which body stores are steadily reducing but

milk feeds have not yet started to replenish these stores. For this

reason, vulnerable babies should not be transferred to the

community at less than 48 h, and only when experienced staff

are satisfied that feeding is effective.

If a baby requires intravenous glucose from birth (for example

if extremely preterm), usually 10% dextrose at 3 ml/kg/h (5 mg


glucose/kg/min) is sufficient to prevent hypoglycaemia. If fluid

restriction is required, a central line should be inserted for

infusion of more concentrated dextrose solutions.

If low blood glucose levels persist or are associated with

clinical signs in the milk-fed infant despite the above measures,

it may be possible to increase further the volumes and/or

frequencies of feeds. If this is not possible, or if the hypo-

glycaemia is resistant to this strategy, intravenous glucose

will be required. If the infant is tolerating milk feeds these

should be neither stopped nor reduced (unless HH is suspected,

see below). The initial rate of 10% glucose infusion should be

3 ml/kg/h (5 mg/kg/min), but adjusted according to frequent

accurate blood glucose measurements. Boluses of concentrated

glucose solution should be avoided because of the risk of

rebound hypoglycaemia and cerebral oedema, and if boluses are

required (for example if there are neurological signs of hypo-

glycaemia) they should be of 10% dextrose (3e5 ml/kg), given

slowly, and always followed by an infusion. All reductions in

infusion rate should be gradual, and any interruption of infusion

should be promptly remedied.

Specific treatments

The baby born after maternal diabetes mellitus in pregnancy:

significant hypoglycaemia is very rare if control of maternal

diabetes has been good. In these circumstances and if the baby is

well and feeding effectively, there is no requirement to intervene

if a single blood glucose level is low. If early blood glucose

measurements are satisfactory, continued blood glucose moni-

toring is not required.

If significant hypoglycaemia occurs this will be in the hours

after birth and only in rare cases will this be prolonged. If the

baby has presented with abnormal clinical signs and requires

intravenous glucose, the target blood glucose level should be 3

mmol/litre and high rates of glucose delivery may be required.

A single injection of glucagon (0.03e0.1 mg/kg), which has

a temporary hyperglycaemic effect by releasing glucose from

glycogen stores, is a useful measure in the event of delay in siting

intravenous lines.

Congenital hyperinsulinaemic hypoglycaemia (HH): recogni-

tion of hyperinsulinism and early prevention and treatment of

hypoglycaemia, with advice from or referral to a specialist centre,

is essential to reduce the incidence of permanent neurological

damage which has been widely reported.

Milk feeds should be stopped, pending discussion with the

specialist centre, as in some case of HH milk feeds further

stimulate insulin release. Intravenous glucose should be

prescribed to maintain blood glucose levels above 3 mmol/litre,

and this may require siting of a central line to deliver concen-

trated glucose solutions. If hypoglycaemia is still resistant to high

glucose delivery rates, diazoxide (10e20 mg/kg/day) and

chlorthiazide (7e10 mg/kg) may be given. Some cases respond

to the calcium channel blocker, nifedipine. Glucagon (200 mg/kg

bolus i.v. or i.m. or infusion 5e10 mg/kg/h), has a temporary

glycaemic effect but its prolonged use is limited because

glucagon further stimulates insulin release. Somatostatin

analogue (octreotide) administered intravenously or subcutane-

ously at a dose of 10 mg/kg/day also suppresses insulin release.

These additional treatments should only be given after




discussion with a specialist centre, and pending transfer of the

baby to the specialist centre.

Inborn errors of metabolism and endocrine insufficiency:

where possible diagnostic samples should be taken before

correction of hypoglycaemia, but should not delay treatment

(Table 2). Emergency management is to ensure adequate blood

glucose levels are sustained, usually requiring intravenous

glucose. If cortisol deficiency is suspected, replacement dose

hydrocortisone may be given empirically pending investigation.

The specialist management of these individual conditions is

beyond the scope of this article and should be discussed with

appropriate teams.

Investigations for persistent, resistant or unexpectedneonatal hypoglycaemia

Blood glucose

Blood gas

Blood lactate

Infection screen

Liver function/blood clotting studies

Urea and electrolytes

Blood b hydroxybutyrate and/or acetoacetatea

Plasma fatty acidsa

Plasma insulina

Plasma cortisola

Plasma/urine amino acid profile

Urine organic acidsa

Blood acyl carnitines

a Samples only informative if taken at the time of hypoglycaemia.

Table 2

Practice points

C Blood glucose levels fall after birth in all babies, and most

compensate for this by mobilizing alternative fuels

C Identification of babies at risk of impaired neonatal metabolic

adaptation and prevention of hypoglycaemia is important to

avoid brain injury

C Unnecessary separation of mother and baby or formula feeding

of the baby should be avoided

C Investigations for underlying cause of hypoglycaemia should

be carried out if hypoglycaemia is persistent, resistant or

unexpected

Summary

Many babies are at risk of impaired metabolic adaptation and

clinically significant hypoglycaemia. Fortunately with prompt

recognition of risk factors and attention to adequacy of energy

intake, it is rare for babies to present with clinical signs or to

sustain brain injury as a result of hypoglycaemia. However, in

rare cases hypoglycaemia is resistant to standard management or

there is serious underlying pathology and specialist advice must

be sought.

Finally, the impact of hypoglycaemia and its treatment on the

mother and baby must be considered. The early neonatal period

is an emotionally sensitive time, and the diagnosis of hypo-

glycaemia may create or add to anxiety for the parents. Treat-

ment of the infant with intravenous glucose involves separation

of the baby and mother with a negative impact on breast feeding,

and may be perceived as invasive or painful. Formula supple-

mentation also disrupts breast feeding and appears to have

a negative effect on normal neonatal metabolic adaptation, so

should be avoided unless there is a clear clinical indication.

Emphasis should be on the early prevention of hypoglycaemia

and strategies of management that do not involve the separation

of mother and baby. A


FURTHER READING

1 Auer RN, Siesjo BK. Hypoglycaemia: brain neurochemistry and

neuropathology. Bailli�eres Clin Endocrinol Metab 1993; 7: 611e25.

2 Boluyt N, van Kempen A, Offringa M. Neurodevelopment after neonatal

hypoglycaemia: a systematic review and design of optimal future

study. Pediatrics 2006; 117: 2231e43.

3 Cornblath M, Hawdon JM, Williams AF, et al. Controversies regarding

definition of neonatal hypoglycemia: suggested operational thresh-

olds. Pediatrics 2000; 105: 1141e5.

4 de Rooy LJ, Hawdon JM. Nutritional factors that affect the postnatal

metabolic adaptation of full-term small- and large-for-gestational-age

infants. Pediatrics 2002; 109: E42.

5 Eidelman AI. Hypoglycemia and the breastfed neonate. Pediatr Clin

North Am 2001; 48: 377e87.

6 Hawdon JM. Care of the neonate. In: McCance DR, Maresh M,

Sacks DA, eds. Practical management of diabetes. Oxford: Wiley-

Blackwell, 2010.

7 Hawdon JM, Ward Platt MP, Aynsley-Green A. Patterns of metabolic

adaptation for preterm and term infants in the first neonatal week.

Arch Dis Child 1992; 67: 357e65.

8 Hay Jr WW, Raju TN, Higgins RD, Kalhan SC, Devaskar SU. Knowledge

gaps and research needs for understanding and treating neonatal

hypoglycaemia: workshop report from Eunice Kennedy Shriver

National Institute of Child Health and Human Development. J Pediatr

2009; 155: 612e7.

9 Kapoor RR, James C, Hussain K. Advances in the diagnosis and

management of hyperinsulinaemic hypoglycaemia. Nat Clin Pract

Endocrinol Metab 2009; 5: 101e12.

10 Medical Devices Agency. Extra-laboratory use of blood glucose

meters and test strips: contraindications, training and advice to the

users. Safety Notice MDA SN 9616 1996.

11 National Childbirth Trust. Hypoglycaemia of the newborn. Mod

Midwife 1997; 7: 31e3.

12 Persson B. Neonatal glucose metabolism in offspring of mothers with

varying degrees of hyperglycaemia during pregnancy. Semin Fetal

Neonatal Med 2009; 14: 106e10.

13 Rozance PJ, Hay WW. Hypoglycaemia in newborn infants: features

associated with adverse outcomes. Biol Neonate 2006; 90: 74e86.

14 Srinivasan G, Pildes RS, Cattamanchi G, Voora S, Lilien LD. Plasma

glucose values in normal neonates: a new look. J Pediatr Surg 1986;

21: 114e7.

15 Vanuucci RC, Vannucci SJ. Hypoglycaemic brain injury. Semin

Neonatal 2001; 6: 147e55.




Resuscitation of the term andpreterm infantAngela E Hayward

AbstractThe vast majority of newborn infants make the transition from intra-

uterine to extrauterine life uneventfully, however there are a significant

number who do require some assistance to make this transition. The

unique physiology at this time needs to be taken into account when

commencing resuscitation efforts. When a newborn infant requires assis-

tance to make the transition to extrauterine life this usually takes the form

of basic airway and breathing management, with more advanced airway

management and chest compressions needed in fewer cases and pharma-

cological support only needed in rare cases.

Keywords guidelines; neonatal; newborn; premature; resuscitation

Background

In 2010 the European Resuscitation Council (ERC) guidelines for

Cardiopulmonary resuscitation (CPR) were updated and pub-

lished. The guideline for resuscitation of babies at birth was

developed based on the most recent International Consensus on

CPR Science and Treatment Recommendations (CoSTR). These 5

yearly reviews have culminated in a several new guideline

recommendations for the resuscitation of newborn infants.

In the UK in 2010, there were 807,272 live births. Approxi-

mately 10% of these newborn infants will have required some

form of assistance to make the transition to extrauterine life and

less than 1% will have required more extensive resuscitation

efforts. The need for resuscitation can often be anticipated;

however there are instances when a baby is born in unexpectedly

poor condition. It is important to ensure that all personnel who

are present at the time of delivery are able to provide basic

newborn resuscitation. In cases where it is predicted the baby

may require newborn resuscitation, the appropriate skilled

personnel should be summoned to attend the delivery.

Normal newborn physiology

There are a number of unique physiological events that occur to

enable a fetus to make the transition to extrauterine life. Most

babies establish their independent breathing and circulation

within a few minutes of being born and quickly become pink.

This transition begins with lung expansion created by a large,

negative intrathoracic pressure and expiration against a partially

closed glottis i.e. generating a cry. Physiological experiments

have shown this initial inspiratory pressure is at least 20 cm of

Angela E Hayward MBBS MRCPCH is Consultant Neonatologist at the

University Hospital of Wales, Heath Park, Cardiff, UK. Conflict of

interest: none.


water and may be as high as 70 cm of water in some newborns.

This has the effect of clearing the lung liquid from the trachea

and alveoli to establish a functional residual volume. The

umbilical cord is clamped and the systemic blood pressure

increases. With regular respiration the pulmonary vascular

pressure falls allowing pulmonary perfusion to the now air filled

lungs, and hence gas exchange. There is a more gradual transi-

tion from the fetal to adult circulation with the closure of the

ductus arteriosus, foramen ovale and ductus venosus.

Physiology of newborn asphyxia

In the 1960s an animal model for neonatal hypoxia was devel-

oped allowing physiological data to be analysed; this now forms

part of the basis for resuscitation management of the newborn. In

these animal experiments, the uterus of a pregnant animal under

anaesthesia was opened, the fetal head was placed in a bag of

fluid and the fetoplacental circulation was obstructed. It was

found that at the onset of the obstruction to the fetoplacental

circulation the fetus attempted to breathe, however due to the

plastic bag the fetus was unable to aerate the lungs. This resulted

in the fetus becoming hypoxic. After a few minutes these

breathing movements would cease and the fetus would enter

a period of primary apnoea. The heart rate was maintained at

a normal level for a period of time before rapidly dropping to

approximately half of its normal level. The heart continued to

beat at this lower rate due to the less efficient anaerobic

metabolism. The blood pressure was maintained due to the

circulation being shut down to all but the most essential areas. If

the hypoxic event continued, the fetus entered a period of deep

gasping movements that were driven by the primitive spinal

centres. After a time of increasing respiratory and metabolic

acidosis the fetus would stop all gasping movements and would

enter a phase described as terminal apnoea. The heart muscle

would no longer function and the fetus would die. This whole

process would take approximately 20 min.

Clearly, total asphyxia of a human infant is rare, more

commonly there is a prolonged partial asphyxia. However, these

experiments have provided us with important information to

help us to understand the physiological process that may occur in

an asphyxiated human infant around the time of birth. A baby

that has not commenced any breathing movements following

delivery may be in primary or terminal apnoea; we need a clear

strategy to attempt to reverse this apnoeic process. See Figure 1.

Resuscitation at birth

Initial actions

Preparation: as with many things, preparation is the key to

a successful outcome. The type of preparation that should occur

will vary with the clinical situation. Resuscitation is far more

likely to be required in situations where there is known fetal

compromise, preterm delivery, vaginal breech delivery and for

multiple pregnancies. However, not all resuscitations can be

predicted so it is important personnel trained in newborn

resuscitation are ready available, with the necessary equipment

to institute initial resuscitation efforts, whilst summoning

personnel trained in advanced resuscitation techniques. Local

guidelines should indicate who is most suitable to attend the

different deliveries and what equipment should be available.



Newborn Life Support

Dry the baby

Remove any wet towels and cover

Start the clock or note the time

2010 Resuscitation Guidelines

Resuscitation Council (UK)

Reassess heart rate every 30 s

If heart rate is not detectable or slow (< 60 min-1)

consider venous access and drugs

Assess (tone), breathing and heart rate

If gasping or not breathing:

Open the airway

Give 5 inflation breaths Consider SpO2 monitoring

Re-assess If no increase in heart rate look for chest movement

If chest not moving:

Recheck head position

Consider 2-person airway control and other airway manoeuvres

Repeat inflation breaths Consider SpO2 monitoring

Look for a response

If no increase in heart rate look for chest movement

When the chest is moving:

If heart rate is not detectable or slow (< 60 min-1)

Start chest compressions 3 compressions to each breath

Birth

30 s

60 s

AT

ALL

STAGES

ASK:

DO

YOU

NEED

HELP?

Acceptable pre-ductal SpO2

2 min 60% 3 min 70% 4 min 80% 5 min 85% 10 min 90%

Figure 1 Newborn Life Support Algorithm 2010. With kind permission of the Resuscitation Council (UK).


PAEDIATRICS AND CHILD HEALTH 22:4 137 � 2011 Elsevier Ltd. All rights reserved.



When there is a planned high-risk delivery, there needs to be

adequate communication between the multi-disciplinary teams

caring for the mother and neonate. This allows for the environ-

ment and equipment to be prepared. If time allows, the team

responsible for the neonatal care should introduce themselves to

the parents, revisit and outline the plan and invite any questions.

Cord clamping: in the uncompromised term infant there is

evidence that suggests delaying of cord clamping for at least 1

min is beneficial. This practice results in the baby having

increased iron stores at 3 months but may result in an increased

need for phototherapy in the immediate neonatal period. In the

preterm baby delayed cord clamping of at least 30 s has been

associated with a suggestion of an increase in postnatal blood

pressure, less need for postnatal blood transfusion, decreased

rates of intraventricular haemorrhage and reduced late onset

sepsis. It is also associated with increased rates of jaundice. For

babies requiring active resuscitation and stabilization there is not

sufficient evidence to suggest delayed cord clamping is beneficial

and commencing resuscitation should remain the priority.

The environment and temperature control: all babies are born

naked and wet and thus have the potential to become cold.

Preterm babies are at particular risk of hypothermia. It is known

that cold stress lowers arterial oxygen tension, increases meta-

bolic acidosis and inhibits surfactant production. The delivery

room should be kept warm and draught free. Following delivery

the term baby should be dried and then wrapped in a warm dry

towel. Alternatively, the baby can be dried and then placed skin

to skin with the mother, covering the baby with a dry towel. If

the term baby is judged to be compromised, the baby should be

dried, wrapped and placed on a warm flat surface directly under

a radiant heater.

For babies delivering at 28 weeks gestation and below, the

delivery room should be kept at 26 �C. Drying and wrapping the

preterm baby is often not sufficient to maintain body tempera-

ture. Placing the wet baby directly into a polyethylene bag or

wrapping in a polyethylene sheet (food or medical grade) up to

the neck and then placing the baby directly under a radiant

heater has been shown to be effective at maintaining tempera-

ture. Resuscitation and stabilization can occur with the plastic

cover in place. The baby should then be kept wrapped in poly-

ethylene until admission to the neonatal unit and the tempera-

ture checked.

Assessment: APGAR scores are not calculated to guide resusci-

tation efforts. The initial assessment for a newborn baby consists

of assessing the heart rate, respiratory effort, colour and tone. The

heart rate should be assessed using a stethoscope, palpating the

base of the umbilical cord is not as reliable. Colour is a poor guide

to oxygenation. Accuracy in assessing heart rate and oxygen

saturations are improvedwith pulse oximetry. Studies have shown

that if a neonatal pulse oximeter probe is attached preductally and

then connected to the pulse oximeter, reliable results will be

available in approximately 90 s. Pulse oximetry has the additional

benefit of providing a continuous measurement of heart rate and

oxygen saturations. An improving heart rate will indicate that

resuscitation efforts are being successful. If the heart rate is not

improving further resuscitation efforts may be required.


Airway and breathing support

Neutral head position: a baby is in primary or terminal apnoea if

there are no signs of respiratory effort at birth. If a baby is

gasping, provided the airway is patent, the baby will aerate the

lungs and subsequently develop regular respiratory movements.

To maintain a patent airway the baby should be placed on a flat

surface with the head in a neutral position. This may be difficult

to achieve in a floppy baby due to the baby’s relatively large

occiput. A rolled towel or padding of not more than 2 cm may be

placed under the baby’s shoulders to assist with neutral posi-

tioning of the airway. A patent airway may also be provided by

using jaw thrust or the insertion of an appropriate sized

oropharyngeal airway, endotracheal tube or laryngeal mask.

Airway suctioning and meconium: obstruction of the airway

may be caused by particulate meconium, vernix, blood clots and

mucus. However, most newborn babies do not need to have their

airway suctioned to remove clear secretions. Suctioning of the

airway has been shown to lower oxygen tension and induces

bradycardia.

In circumstances of meconium stained liquor, the practice of

suctioning the mouth and nose of the baby on the perineum has

been shown to be ineffective at preventing meconium aspiration

syndrome. In addition, routine intubation and suctioning of the

trachea of the vigorous baby has also been shown to be inef-

fective at preventing meconium aspiration syndrome. In the case

of the non-vigorous baby, the available evidence does not

support or refute routine suctioning of the trachea. This is an

area where further randomized clinical trials would be helpful to

inform clinical practice.

Positive pressure ventilation: if the newborn baby is showing

inadequate or no spontaneous respiratory efforts, the lungs need

to be inflated. In the term newborn, giving five inflation breaths

each lasting 2e3 s at a pressure of 30 cm of water should result in

the establishment of the functional residual volume. Occasion-

ally, higher pressures may be required. The effectiveness of these

inflations should be judged by an increase in the heart rate. If

there is no improvement in the heart rate, chest wall movement

should be examined and face-mask seal technique should be

checked to ensure there are no leaks. Once the functional

residual volume has been established the lungs should be

ventilated at a rate of 30e40 breaths/min.

Inflation and subsequent ventilation breaths are given by face

mask in the majority of cases. The face mask may be attached to

a bag/valve system or T-piece system. The benefits of the

bag/valve system are that it does not require a gas supply and is

portable. However, its limitations include the inability to give set

pressures or continuous positive airway pressure (CPAP). The

bag/valve system comes with a pressure limiting valve usually

set to 40 cm of water; if the bag is squeezed vigorously pressures

far in excess of this may be generated. The benefits of the T-piece

system are varying inflation times may be given and the peak

inspiratory pressure (PIP) and peak expiratory end pressure

(PEEP) may be set to the desired level. The T-piece is limited

by the fact that in order to operate the system it requires

a compressed gas supply.

Preterm term lungs are damaged by large volume inflations

following birth. Animal studies have shown the benefit of



0

Ox

yg

en

sa

tura

tio

n (

%)

Minutes after birth

a

1 2 3 104 5 6 7 8 9

100

90

80

70

60

50

40

30

20

10

0

0

Ox

yg

en

sa

tura

tio

n (

%)

Minutes after birth

b

1 2 3 104 5 6 7 8 9

100

90

80

70

60

50

40

30

20

10

0

0

Ox

yg

en

sa

tura

tio

n (

%)

Minutes after birth

c

1 2 3 104 5 6 7 8 9

100

90

80

70

60

50

40

30

20

10

0

3rd 10th 25th

75th 90th 97th

50th

Figure 2 Oxygen saturation percentiles for newborn infants with no medical

intervention after birth. (a) Third, 10th, 25th, 50th, 75th, 90th and 97th Spo2



maintaining a PEEP to prevent this damage, improve lung

compliance and gas exchange. However, excessive PEEP may

reduce pulmonary blood flow and cause pneumothorax. It is

thought that initial inflation pressures of 20e25 cm of water are

adequate in preterm deliveries. Very obvious chest wall move-

ment in preterm infants undergoing positive pressure ventilation

may indicate excessive tidal volumes and pressures should be

adjusted accordingly. A preterm infant who is spontaneously

breathing may benefit from CPAP in the delivery room. The

COIN and SUPPORT trials have reported that there is no signif-

icant reduction in death or bronchopulmonary dysplasia in

infants treated with either early CPAP or intubation and surfac-

tant in the delivery room. They also reported preterm infants

exposed to CPAP following delivery had fewer days of ventilation

and oxygen requirement.

Air versus oxygen: in term newborn infants who require posi-

tive pressure ventilation, air has been shown to be as effective as

100% oxygen. There is now compelling evidence showing that

hypoxic tissues when exposed to high concentrations of oxygen

come to additional harm from oxygen free radicals and antioxi-

dants. Hyperoxaemia has been shown to be damaging to the

brain and other organs at the cellular level, particularly after

a hypoxic event. Pulse oximetry studies have provided us with

data for an uncompromised population of newborns making the

transition from intrauterine to extrauterine life. Babies born at

term and at sea level have saturations of approximately 60% in

utero and this rises to more than 90% by 10 min of age. When

born at higher altitude or by Caesarean section, these values are

found to be lower. It therefore seems reasonable to start resus-

citating in room air and any need for supplementary oxygen to be

guided by the pulse oximeter readings and heart rate.

In preterm infants resuscitating in air or 100% oxygen results

in hypoxaemia or hyperoxaemia respectively. There is not yet

sufficient data to indicate what oxygen concentration should be

used when commencing resuscitation in preterm infants. It

seems reasonable to aim for saturations resembling those of

healthy term infants and adjust the oxygen blender whilst being

guided by the pulse oximeter readings and heart rate. There are

now reference ranges for oxygen saturations in the first few

minutes following delivery for both term and preterm infants.

See Figure 2.

Endotracheal intubation: endotracheal intubation is of benefit

when face-mask ventilation has proved ineffective, tracheal

suction is required, ventilation is prolonged or there are special

circumstances e.g. congenital anomalies or administration of

surfactant. The timing of endotracheal intubation may depend on

operator skill and experience. Once the endotracheal tube has

percentiles for term infants at more than 37 weeks of gestation with no

medical intervention after birth. (b) Third, 10th, 25th, 50th, 75th, 90th and

97th Spo2 percentiles for preterm infants at 32e36 weeks of gestation with

nomedical intervention after birth. (c) Third, 10th, 25th, 50th, 75th, 90th and

97th Spo2 percentiles for preterm infants at less than 32 weeks of gestation

with no medical intervention after birth.

Dawson JA, Kamlin CO, VentoMet al. Defining the reference range for oxygen

saturation for infants after birth. Pediatrics 2010; 125(6): e1340ee1347withkind permission of the American Academy of Pediatrics.




been inserted, its position in the trachea needs to be confirmed.

There is an increasing body of evidence indicating detection of

exhaled carbon dioxide confirms tracheal intubation in neonates

with an adequate cardiac output faster than clinical assessment

alone. Clinical assessment includes seeing the tip of the endo-

tracheal tube pass through the vocal cords, prompt increase in

heart rate, equal chest wall movement, equal breath sounds

bilaterally and condensation (misting) in the endotracheal tube.

Oesophageal intubation should be suspected if the carbon

dioxide detection device fails to detect carbon dioxide. These

devices are suitable for use in term and preterm infants, but their

use has not been studied sufficiently in situations of circulatory

arrest where poor or absent pulmonary blood flow may prevent

carbon dioxide detection. False positives have also been docu-

mented in colorimetric devices when they have been contami-

nated with adrenaline, surfactant and atropine. Once the

endotracheal tube has been confirmed to be in the trachea, it is

important to ensure that it is secured at the correct depth. See

Table 1.

Laryngeal masks: in the newborn setting, a laryngeal mask that

fits over the laryngeal inlet has been used successfully in babies

requiring positive pressure ventilation with a gestation greater

than 34 weeks and weight more than 2000 g. They have also

been used successfully when face-mask ventilation and endo-

tracheal tube placement have proved unsuccessful. There is very

limited data for their use in smaller preterm infants. There is no

or insufficient evidence for the use of the laryngeal mask in the

setting of meconium stained liquor, during chest compressions or

during the administration of tracheal medications.

Circulation support

Chest compressions: chest compressions to provide circulatory

support are only effective if the lungs have been successfully

inflated. In the presence of a low heart rate, effective ventilation is

judged by the observation of chest wall movement. If the heart rate

remains below 60 beats per minute and there has been chest wall

movement, compressions should be commenced. There is no

scientific data that supports a given ratio of ventilations and

compressions in the neonatal population. Currently, the recom-

mended compression to ventilation ratio is 3:1. Chest compressions

Tracheal tube lengths by gestation and weight

ETT length at lips (cm) Gestation (weeks) Weight (kg)

5.5 23e24 0.5e0.6

6.0 25e26 0.7e0.8

6.5 27e29 1.9e1.0

7.0 30e23 1.1e1.4

7.5 33e34 1.5e1.8

8.0 35e37 1.9e2.4

8.5 38e40 2.5e3.1

9.0 41e43 3.2e4.2

Kempley ST, Moreiras JW, Petrone FL. Endotracheal tube length for neonatal

intubation. Resuscitation 2008; 77: 369e373.

Table 1


are more effectively and efficiently provided when the two thumb

encircling method is used rather than the two finger method. The

chest should be compressed one-third of its anterioreposterior

diameter above the xiphisternum and just below the nipple line. In

1 min the aim should be to deliver 90 compressions and 30 venti-

lations. The chest wall should be allowed to return to its relaxed

position in between compressions to allow the heart to refill

passively after the compression. Compressions that are delivered

effectively will generate a pulsation seen on a pulse oximeter. The

heart rate should be assessed every 30 s and the chest compressions

should be discontinued once the heart is beating spontaneously

above 60 beats/min.

Medication and fluid: the use of medication or fluid is very

rarely required in neonatal resuscitation. Bradycardia is usually

secondary to hypoxia, therefore ensuring adequate ventilation

and chest compressions should result in the reversal of the

bradycardia. If the heart rate remains below 60 and the above

interventions are being performed correctly, the use of medica-

tion may be required.

Adrenaline e there is insufficient neonatal data to indicate

the most appropriate dose and route for adrenaline usage and the

data we do have is based on case series and case reports. From

the limited information available, the dose via the tracheal route

would need to be higher than that by the intravenous route to

obtain a similar clinical effect. When extrapolating data from

animal and paediatric studies, high doses of adrenaline given

intravenously, may cause cardiac and neurological dysfunction.

The current recommendations state that adrenaline should be

given intravenously via an umbilical venous catheter at a dose of

0.01e0.03 mg/kg. If the intravenous route is not available

tracheal adrenaline may be considered using a higher dose of

0.05e0.1 mg/kg.

Sodium bicarbonate e the use of sodium bicarbonate is

controversial and current guidance suggests that it is only

appropriate to use in prolonged resuscitation when other

methods have failed. The dose suggested for sodium bicarbonate

is 1e2 mmol/kg intravenously. The animal studies in the 1960s

showed that giving a mixture of alkali and glucose when the fetal

animal was known to be in terminal apnoea resulted in the return

of gasping and an increase in heart rate in the absence of any

other resuscitation efforts. In a number of studies sodium

bicarbonate has been shown to have a number of potential side

effects including depression of myocardial function, paradoxical

intracellular acidosis and decreased cerebral blood flow.

Fluid e it is rare that fluid is given during a neonatal resus-

citation and again there is limited neonatal data. It would seem

sensible to give volume when there is known or suspected blood

loss. In this situation it would be more appropriate to give

emergency blood to improve intravascular volume. If there is

a delay in obtaining suitable blood, isotonic crystalloid is

advised. A bolus of 10 ml/kg fluid should be given in the first

instance and its effect assessed prior to any further volume

administration.

Ongoing care following resuscitation

Newborns who have responded to resuscitation efforts remain at

risk for further deterioration. It is essential for them to have



Practice points

C The need for resuscitation is not always predictable.

C Effective airway management is vital.

C Pulse oximetry readings should guide the need for supple-

mentary oxygen.

C Where there is evidence of hypoxiceischaemic encephalop-

athy, cooling should be considered.

C Ongoing clinical research is required to further our knowledge.


further observation and clinical care, usually within the neonatal

unit. After resuscitation, it is important to consider whether the

newborn meets the criteria for induced hypothermia (cooling).

There are several, good quality trials showing that infants born at

a gestation greater than 36 weeks, suffering from moderate to

severe hypoxiceischaemic encephalopathy, who are cooled,

have significantly reduced death and neuro-disability at 18

months. Cooling should be commenced as soon as possible after

resuscitation is completed; the current evidence suggests that

cooling is unlikely to be beneficial if delayed for greater than 6 h.

In centres where there are no facilities for cooling, the local

cooling centre should be contacted and arrangements for trans-

port made. Whilst awaiting transport, passive cooling should be

commenced.

Failure to respond to resuscitation

If a baby fails to respond to good quality resuscitation efforts and

the heart rate continues to be undetectable after 10 min, it is

appropriate to consider stopping resuscitation efforts. The deci-

sion to continue the resuscitation attempts may be affected by

a number of considerations, including, the newborns gestation,

the underlying aetiology and the parents expressed wishes. This

is a very difficult clinical situation for the resuscitation team and

experienced senior advice should aid decision making. A

FURTHER READING

Cramer K, Wiebe N, Hartling L, Crumley E, Vohra S. Heat loss prevention:

a systematic review of occlusive skin wrap for premature neonates.

J Perinatol 2005; 25: 763e9.

Davis PG, Tan A, O’Donnell CP, Schulze A. Resuscitation of newborn

infants with 100% oxygen or air: a systematic review and meta-anal-

ysis. Lancet 2004; 364: 1329e33.

Dawson JA, Kamlin CO, Vento M, et al. Defining the reference range for

oxygen saturation for infants after birth. Pediatrics 2010; 125:

e1340e7.

Finer NN, Waldemar AC, Walsh MC, et al. Early CPAP versus surfactant in

extremely preterm infants. N Engl J Med 2010; 362: 1970e9.

Hosono S, Inami I, Fujita H, Minato M, Takahashi S, Mugishima H. A role of

end-tidal carbon dioxide monitoring for assessment of tracheal intu-

bations in very low birth weight infants during neonatal resuscitation

at birth. J Perinat Med 2009; 37: 79e84.


Kempley ST, Moreiras JW, Petrone FL. Endotracheal tube length for

neonatal intubation. Resuscitation 2008; 77: 369e73.

McDonald SJ, Middleton P. Effect of timing of umbilical cord clamping of

term infants on maternal and neonatal outcomes. Cochrane Database

Syst Rev 2008. CD004074.

Milner AD, Lagercrantz H. Adaptation at birth. In: Greenough A, Milner AD,

eds. Neonatal respiratory disorders. Arnold, 2003; 59e66.

Morley CJ, Davis PG, Doyle LW, Brion LP, Hascoet JM, Carlin JB. Nasal CPAP

or intubation at birth for very preterm infants. N Engl J Med 2008; 358:

700e8.

Rabe H, Reynolds G, Diaz-Rossello J. A systematic review and meta-

analysis of a brief delay in clamping the umbilical cord of preterm

infants. Neonatology 2008; 93: 138e44.

Richmond S, ed. Newborn life support: resuscitation at birth. London:

Resuscitation Council (UK), 2011.

Richmond S, Wyllie J. European Resuscitation Council guidelines for

resuscitation 2010 section 7: resuscitation of babies at birth. Resus-

citation 2010; 81: 1389e99.

Vain NE, Szyld EG, Prudent LM, Wisewell TE, Aguilar AM, Vivas NI.

Oropharyngeal and nasopharyngeal suctioning of meconium-stained

neonates before delivery of their shoulders: multicentre, randomised

controlled trial. Lancet 2004; 364: 597e602.

Wiswell TE, Gannon CM, Jacob J, et al. Delivery room management of the

apparently vigorous meconium stained neonate: results of a multi-

center, international collaborative trial. Pediatrics 2000; 105: 1e7.

Wyllie J, Perlman JM, Kattwinkel J, et al. Part 11: neonatal resuscitation

2010 International Consensus on Cardiopulmonary Resuscitation and

Emergency Cardiovascular Care Science with treatment recommenda-

tions. Resuscitation 2010; 81S: e260e87.




Understanding blood gases/acidebase balanceNitin Goel

Jennifer Calvert

AbstractAcidebase balance is regulated by intracellular & extracellular buffers and

by the renal and respiratory systems. Normal pH is necessary for the

optimal function of cellular enzymes and metabolism. Disorders of acide

base balance can interfere with these physiological mechanisms leading

to acidosis or alkalosis and can be potentially life threatening. Blood

gas analysis is a routine procedure performed in the neonatal unit and

combined with non-invasive monitoring, aids in the assessment and

management of ventilation and oxygenation and provides an insight

into the metabolic status of the patient. The following discussion details

the basic terminology and pathophysiology of acidebase balance and the

main disorders. It aims to provide a logical and systematic approach to

the understanding and interpretation of blood gases in the newborn

period. The application of these concepts, together with relevant history

and examination, will help the clinician assess the medical condition,

make therapeutic decisions and evaluate the effectiveness of any inter-

vention provided.

Keywords acidebase balance; acidosis; alkalosis; anion gap; base

deficit; blood gas analysis; pH

Introduction & terminology

Acidebase balance is the complex physiological process, which

acts to maintain a stable extracellular pH within the body. It is

regulated by intracellular & extracellular buffers and by the renal

and respiratory systems. Any derangement in this balance can

interfere with physiological processes and can be potentially life

threatening. An understanding of acidebase balance is required

for the interpretation of blood gases, to assess both the respira-

tory and metabolic status of patients and thereby enable their

effective clinical management.

Normal pH is maintained between 7.35 and 7.45, which

creates an optimal environment for cellular metabolism. The pH

is inversely related to the concentration of Hþ ions.

pH a 1=Hþ

Nitin Goel MBBS MD MRCPCH is a Neonatal Registrar at the Neonatal

Intensive Care Unit in the University Hospital of Wales, Cardiff, UK.

Conflict of interest: none.

Jennifer Calvert BA BM BCh MRCP(UK) MRCPCH is a Consultant Neonatologist

at the Neonatal Intensive Care Unit, University Hospital of Wales,

Cardiff, UK. Conflict of interest: none.


An acid (HA) is a substance that donates Hþ ions (e.g. carbonic

acid). In contrast, a base (A�) accepts Hþ ions, (e.g. hydroxyl

ions, ammonia) and in solution combines with the acid to

neutralize it. An acid can dissociate into Hþ and a conjugate

base.

HA4Hþ þA�

Equilibrium is maintained based on the above equation.

Thus addition of acid (HA) increases Hþ and A� and shifts the

equation towards the right. During normal metabolism, Hþ ions

are constantly being produced and neutralized to maintain pH

homeostasis. Neonates produce higher levels of Hþ due to their

rapid growth and metabolism and therefore maintaining balance

can be challenging in newborn period.

Normal acidebase regulation

The process of maintaining pH balance during normal metabo-

lism involves buffer systems and compensatory mechanisms in

the respiratory and renal systems.

Buffer systems

Buffers are substances that attenuate the change in pH when

acid/base levels increase. On addition of acid, they bind to

any extra Hþ ions and prevent decline in pH. Similarly when

base is added, the buffers prevent a rise in pH by releasing Hþ

ions. The best buffers are weak acids and bases and work best

when they are 50% dissociated. The pH at which this happens

is called pK and is close to 7.40 for some buffers. The

HendersoneHasselbalch equation expresses the relationship

between pH, pK and concentrations of an acid and its conju-

gate base.

pH¼ pKþ log½A��=½HA�

Extracellular buffers: the bicarbonate system is the principal

buffer in the extracellular fluid (ECF) and is based on the rela-

tionship between carbon dioxide (CO2) and bicarbonate

(HCO3�), where the former combined with water acts as an acid

(carbonic acid H2CO3) and the latter as base.

Hþ þHCO�3 4H2CO34CO2 þH2O

The pK for this buffer is 6.1. For bicarbonate buffer, the

HendersoneHasselbalch equation is:

pH¼ 6:1þ log½HCO�3 �=½CO2�

Mathematical manipulation of the above equation produces the

following relationship,

½Hþ� ¼ 24� pCO2=½HCO�3 �

which emphasizes that Hþ ion concentration and hence pH is

determined by the ratio of pCO2 and HCO3� concentration, and

not their absolute values.

When Hþ ions are added to the system, the equation shifts to

right and pH is maintained at the expense of HCO3� ions, referred




to as ‘Base Deficit’. There is also an increase in dissolved CO2

levels (as H2CO3), which can be clinically estimated by measuring

the partial pressure of CO2 (pCO2). Thus with addition of Hþ ions,

the pH decreases with a decrease in base and an increase in CO2

levels. The lungs then excrete the excess CO2. With addition of

base, there is a decrease in CO2 and the lungs then reduce CO2

excretion. In this way the bicarbonate buffer system works as an

open system and plays an important role in pH homeostasis.

Intracellular buffers: these are non-bicarbonate buffers and

include various proteins and organic phosphates. The proteins

consist of acid histidine, with a side chain, which accepts Hþ ions

in exchange for intracellular potassium (Kþ) and sodium (Naþ)ions. In acute metabolic acidosis, hyperkalaemia can develop due

to the exchange of Kþ for Hþ.Phosphate can bind up to three Hþ ions and in its mono- and

di-hydrogen forms acts as an effective buffer in the urine.

H2PO1�4 4Hþ þHPO2�

4

Bone is also an important buffer and releases base on dissolution,

so can buffer an acid load, but at the expense of bone density.

During bone formation, it also consumes base thus buffering any

excess.

Compensatory mechanisms

Although buffers represent the first line of defence against pH

changes, they cannot maintain acidebase balance in disease

states for prolonged periods of time or with sudden significant

alterations of Hþ ion production. Instead, compensatory physi-

ological changes by the renal and respiratory systems are

employed. In a primary metabolic disorder, the respiratory

system provides the compensation, whereas in a primary respi-

ratory disorder, the regulation is by the renal system. Respiratory

responses occur more rapidly (minutesehours) than renal

mechanisms, which take about 3e4 days, with renal base

excretion more rapid than acid excretion. These compensatory

mechanisms must be followed by corrective measures to

normalize the acidebase balance, by treating the primary cause

of the imbalance.

Respiratory compensation: the respiratory system modifies pH

by balancing the production of Hþ with excretion of CO2.

During normal metabolism CO2 is generated, which is a weak

acid. Any increase in physical activity leads to an increase in

metabolism and thus an increase in pCO2. The lungs respond

by increasing ventilation and excreting excess CO2, thus

maintaining a normal pCO2 (4.5e6 kPa). Conversely, hypo-

ventilation causes CO2 retention and thus an increase in pCO2.

The resulting increase in Hþ ions directly stimulates chemore-

ceptors in the brain causing an increase in respiratory rate.

Thus changes in alveolar ventilation can alter pH and vice

versa.

Renal compensation: the kidneys prevent loss of HCO3� in the

urine and maintain plasma levels by excreting acid and gener-

ating new bicarbonate. They can thus respond to acidebase

imbalance by acidifying or alkalinizing the urine. This is

accomplished by:


(1) Reabsorption of filtered HCO3, which takes place in the

proximal tubules (85%) and in the thick ascending loop of

Henle (15%).

Normally large amounts of bicarbonate enter the proximal

tubules (PT) daily and if this bicarbonate is not reclaimed by the

nephrons, severe acidosis can result. In the proximal tubular

cells, CO2 derived from cell metabolism or diffusion through the

tubular lumen, combines with water to form carbonic acid. This

dissociates into Hþ ions and bicarbonate via carbonic anhydrase.

The bicarbonate is transported back to the circulation, while the

Hþ ions are secreted into the tubular lumen, where they combine

with the filtered bicarbonate to form H2O and CO2. The CO2

diffuses back in the PT cells to repeat the cycle. The net effect is

that for each Hþ ion secreted, one HCO3� is retained, so that

bicarbonate reserves are continuously regenerated.

Factors causing an increase in Hþ ion secretion and thus

increased bicarbonate reabsorption include increased filtered

bicarbonate, volume depletion due to any cause and resulting

activation of renineangiotensin system, increased plasma pCO2

and hypokalaemia. Conversely Hþ ions secretion and thus

bicarbonate reabsorption is decreased in conditions with reduced

filtered bicarbonate, expansion of ECF volume and decreased

plasma pCO2. Hyperparathyroidism and disease states such as

proximal renal tubular acidosis (RTA), cystinosis, or neph-

rotoxins can also affect proximal tubules and limit bicarbonate

reabsorption.

Newborn infants and in particular preterm babies have

a lower glomerular filtration rate, immature tubular function and

limited capacity to retain bicarbonate and are therefore predis-

posed to metabolic acidosis.

(2) Excretion of Hþ ions which takes place at the distal tubules

and the collecting duct, thus acidifying the urine. The prin-

cipal buffers at these sites are phosphate and ammonia.

In normal conditions large amounts of phosphate ions are

present in the tubular fluid, which combine with Hþ ions,

forming titratable acid, thus reducing urinary pH. However,

phosphate buffering capacity is limited as there is no mechanism

for increasing urinary phosphate excretion in response to acide

base status.

Ammonia is generated in the cells of the proximal tubules,

diffuses into the tubular fluid and combines with the intra-

luminal Hþ ions to form ammonium ion, which cannot diffuse

back into the tubular cells, thus making ammonia an effective

buffer.

These two processes reduce free Hþ in the tubular fluid,

thereby increasing Hþ excretion into the urine and allowing the

generation of new bicarbonate in the cells, which can then

enter the plasma to replenish depleted levels. The major

regulator of Hþ secretion in the distal tubule is aldosterone with

other influencing factors being pCO2 and the sodium concen-

tration delivered to these segments. Sodium is reabsorbed in

exchange for either potassium or Hþ ions, under the influence

of aldosterone. These mechanisms may be impaired by intrinsic

defects in the tubules causing primary distal renal tubular

acidosis (RTA), or by various insults including nephrocalci-

nosis, vitamin D intoxication or Amphotericin B administra-

tion, which produce secondary distal RTA. Patients with distal

RTA cannot acidify their urine and have a urine pH more than

5.5, despite acidosis.




Disturbances of acidebase balance

Abnormalities in blood pH, due to an increase in Hþ ions above

normal, is called ‘acidaemia’ (pH less than 7.35), and due to

a decrease is termed ‘alkalaemia’ (pH more than 7.45). The

clinical process, which causes the acid or alkali to accumulate, is

called ‘acidosis or alkalosis’, respectively.

As shown in Figure 1, acidosis is caused by conditions

resulting in either a reduction in HCO3� or an increase in pCO2,

leading to an increase in Hþ ions and decreased pH. Alkalosis is

caused when the primary disturbance causes either an increase

in HCO3� or a decrease in pCO2, leading to a decrease in Hþ ions

and an increased pH.

Metabolic acidosis

This results from an alteration in the balance between production

and excretion of acid; by increased exogenous intake or endoge-

nous production of Hþ ions, inadequate excretion, or by excessive

loss of bicarbonate in urine or stools (Table 1). Premature infants

less than 32 weeks gestation, frequently manifest a proximal or

distal RTA. In proximal RTA, there is limited secretion of Hþ ions

and incomplete bicarbonate reabsorption. Urine pH remains less

than 5, but becomes alkaline after a bicarbonate infusion, even

without normal serum bicarbonate levels. In distal RTA, the distal

tubules cannot secrete Hþ ions and thus the urine pH remains

alkaline (more than 7), rarely falling below 5.5.

Carbohydrate, fat and protein metabolism in the body

generate about 2e3 mEq/kg/day Hþ ions. Normally the CO2,

resulting from complete oxidation of carbohydrates and fats is

removed by the lungs. However anaerobic metabolism, as in

tissue hypoxia, produces lactic acid from glucose metabolism

and ketoacids from triglycerides, leading to acidosis.

pH < 7.

pH > 7.

CO + H O ↔ H CO

↑PCO

↓PCO

Respiratory acidosis

Respiratory compensation

Hyperventilation

↓PCO

Hypoventilation

↑PCO

Respiratory alkalosis

Figure 1 Acidebase regulation: interplay of bicarbon


Systemic acidosis stimulates the respiratory centre directly,

the rate of breathing is increased and CO2 is excreted. The

acidosis also stimulates the kidneys to increase Hþ ion excre-

tion, accompanied by bicarbonate reabsorption. In renal insuf-

ficiency, the ability of kidneys to generate ammonia and secrete

Hþ ions is limited, leading to acidosis. In unstable neonates,

respiratory compensation is limited and because of tubular

immaturity, the acidosis worsens rapidly if the underlying cause

is not treated.

Anion gap: an important tool in evaluating the cause of meta-

bolic acidosis is the ‘anion gap’, the difference in the measure-

ment of the most abundant serum cation (Naþ) and the sum of

two most abundant serum anions (HCO3� and chloride, Cl�).

½Naþ� � ð½Cl�� þ ½HCO�3 �Þ

This gap also represents the difference between unmeasured

anions (phosphate, sulphate, proteins, acids e.g. lactate, ketoa-

cids) and unmeasured cations (potassium, magnesium, calcium).

It should not be interpreted in isolation but in conjunction with

other laboratory abnormalities and the clinical history. The

normal anion gap for neonates is 5e15 mEq/litre.

An elevated anion gap represents an increase in unmeasured

anions (Figure 2) and can result from overproduction or under

excretion of acids. Normal anion gap acidosis results from the net

loss of bicarbonate. In these cases Cl� reabsorption is increased

and it becomes the major anion accompanying Naþ and so the

sum of anions in plasma remains normal. Thus, normal anion

gap acidosis is also referred to as hyperchloraemic metabolic

acidosis.

35

45

↔ H+ + HCO –

↓HCO –

↑HCO –

Metabolic compensation

↑Bicarbonate

reabsorption

↑HCO –

↓Bicarbonate

reabsorption

↓HCO –

Metabolic acidosis

Metabolic alkalosis

ate buffer, respiratory and renal systems.



Acidebase disorders, blood gas findings and common causes in neonates

Disorder Blood gas analysis (normal range) Causes

pH

(7.30e7.45)

pCO2

(4.5e6 kPa)

HCO3L

(19e24 mmol/l)

BE

(L3 to D3)

Metabolic acidosis

Uncompensated Y Normal Y Y Increased anion gap (more than 16 mEq/l)

Hypoxaemia/lactic acidosis: sepsis, shock, respiratory or cardiac disorders, anaemia, intraventricular haemorrhage,

perinatal asphyxia, necrotizing enterocolitis

Renal failure

Inborn errors of metabolism

Total parenteral nutrition

Normal anion gap (8e16 mEq/l)

Prematurity: hyperchloraemic acidosis

Renal tubular acidosis: proximal/distal

Gastrointestinal bicarbonate losses: ileostomy, diarrhoea

Compensated Low Y Y Y

Normal

Metabolic alkalosis

Uncompensated [ Normal [ [ Decreased urinary chloride (<10 mEq/l)

Gastric losses: vomiting, pyloric stenosis, excess naso-gastric aspirates

Diuretics

Chloride losing diarrhoea

Hypokalaemia

Increased urinary chloride (>20 mEq/l)

Hyperaldosteronism

Adrenal hyperplasia

Excess alkali administration

Compensated High [ [ [

Normal


Uncompensated Y [ Normal Normal Respiratory abnormalities: respiratory distress syndrome, chronic lung disease, pneumothorax, meconium aspi-

ration, transient tachypnoea of newborn, pneumonia, pulmonary hypoplasia, congenital lung malformations

Central nervous system depression: hypoxic ischaemic encephalopathy, excess opioids, raised intracranial pres-

sure, central hypoventilation, meningitis, malformations

Neuro-muscular disorders: congenital myopathies, neuropathies, spinal and neuro-muscular junction disorders

Upper airway obstruction: Pierre-Robin sequence, choanal atresia, laryngeal oedema/spasm/mass etc.

Iatrogenic: inadequate ventilator settings in mechanically ventilated patient

Compensated Low [ [ [

Normal


Uncompensated [ Y Normal Normal Increased sensitivity of respiratory centre: hypoxia due to any cause

Medications: Caffeine

Extra pulmonary CO2 losses: ECMO, dialysis

Iatrogenic: over-ventilation of mechanically ventilated patient

Compensated High Y Y Y

Normal

Table 1

SYMPOSIUM:NEONATOLO

GY

PAEDIATRICSANDCHILD

HEALTH22:4

145

�2011Elsevie

rLtd

.Allrig

hts

reserve

d.


Normalplasma

UC

UC, unmeasured cations; UA, unmeasured anions.

The anion gap

UA

HCO

Na

UCUA

UC

UA

Cl

HCO

Cl

NaHCO

Cl

Na

Acidosis(no gap)

Acidosis(increased gap)

Figure 2 Anion gap.


Metabolic alkalosis

This results from increased bicarbonate and/or excessive loss of

Hþ ions. It is uncommon in the neonatal period. Causes are

related to increased renal reabsorption of HCO3, loss of Hþ ions

or increase addition of bicarbonate (Table 1).

The buffers try to minimize the changes, but bicarbonate and

pH rise, respiration is depressed, and there is an increase in

pCO2. Respiratory compensation is limited by increasing

hypoxia, so cannot normalize the pH. The kidneys respond

to this by increasing base excretion, with urine pH increasing to

8.5e9.0. The alkalosis can worsen if there is co-existing ECF

contraction and hypokalaemia, as it conversely increases bicar-

bonate reabsorption. This can only be corrected by treating the

underlying disorder.

Hypochloraemia and hypokalaemia are usually present, due

to increased urinary losses. Measurement of urinary chloride can

help differentiate the causes of metabolic alkalosis (Table 1). If

urine chloride levels are less than10 mEq/litre, the underlying

cause is generally volume depletion from extra-renal losses, with

loss of Naþ, Kþ and chloride. These cases are responsive to

sodium chloride. The use of diuretics in neonates can lead to

increased fluid and Naþ losses in the kidneys, stimulating Naþ

reabsorption in exchange for Hþ ions, thus leading to bicarbonate

reabsorption and metabolic alkalosis. If metabolic alkalosis is

secondary to excessive mineralocorticoid activity or potassium

depletion, the urine chloride is more than20 mEq/litre, and is

resistant to sodium chloride treatment.


This occurs due to inadequate pulmonary excretion of

CO2 leading to increases in pCO2 and H2CO3, with a resulting

rise in Hþ ions. This occurs both acutely and in a chronic

form, in conditions affecting the respiratory or neurological

systems (Table 1). The rise in pCO2 is initially buffered by


non-bicarbonate buffers, protein & phosphate. If the rise is

sustained, as in preterm babies with chronic lung disease, the

kidneys are stimulated to excrete Hþ ions and to generate &

reabsorb bicarbonate. This causes plasma bicarbonate levels to

increase above normal and the pH returns to normal. This is

the compensated phase of respiratory acidosis and occurs over

days.


This occurs with excessive pulmonary losses of CO2 and result-

ing fall in pCO2. This occurs with hyperventilation due to any

cause (Table 1). It is often iatrogenic, related to mechanical

ventilation. A rapid decrease in pCO2 has been associated with

periventricular leukomalacia and intraventricular haemorrhage,

so timely intervention is critical.

With decreased pCO2, pH rises and a rapid buffering occurs

with release of Hþ ions to decrease the plasma bicarbonate.

There is also increased renal excretion of HCO3�. This results

in a decrease in plasma bicarbonate and pH normalizes. Final

correction is achieved by treatment of the underlying

disorder.

Mixed disorders

In certain conditions, more than one disturbance can co-exist.

This should be suspected if the compensatory response falls

outside the expected range. For example, in respiratory

distress syndrome or pneumonia with sepsis, respiratory

acidosis (due to ventilatory failure) and metabolic acidosis

(due to lactic acidosis) often co-exist. The respiratory disease

prevents the compensatory fall of pCO2 and the metabolic

component prevents compensatory rise of plasma bicar-

bonate, resulting in a greater fall in pH. Similarly in chronic

lung disease with the use of loop diuretics, respiratory

acidosis and metabolic alkalosis can result. Thus the plasma

bicarbonate and pH are higher than expected. Patients with

hepatic failure can have metabolic acidosis and respiratory

alkalosis, with a greater than usual drop in plasma bicar-

bonate & pCO2 and little change in pH.

Implications of acidebase disorders

The effects of pH changes at a cellular level are poorly

understood. A low pH can reduce myocardial contractility and

impair catecholamine action, increasing the risk of

arrhythmia. The metabolic activity of proteins is pH depen-

dent and any changes may adversely affect enzyme activity.

An increase in Hþ ions can also cause disturbances in ion

transport within the kidneys. With acidosis, a decrease in

carbohydrate tolerance is observed and with alkalosis, an

increase in neuro-muscular irritability can occur, either in

a latent form or manifesting as tetany.

Invasive & non-invasive blood gas analysis in the neonatal unit

Blood gas analysis is routinely performed in the neonatal unit.

In conjunction with non-invasive monitoring, it enables

clinicians to appropriately assess & monitor the respiratory

status and modify ventilation strategy accordingly. It can also

provide information on metabolic status, acidebase



Guide for blood gas values in neonates

Analyte Normal reference ranges

(arterial sample)

pH 7.30e7.45

PaCO2 (kPa) 4.5e6.0

PaO2 (kPa) 6.0e8.0

HCO3� (mmol/l) 19e24

BE (mmol/l) �3 to þ3

Table 2


imbalance and whether any respiratory or renal compensation

is taking place.

Blood gas values vary depending on the site of the sample,

i.e., arterial, capillary or venous; arterial samples are the most

informative. The technique of sampling is equally important; the

sample site should be warm if capillary, the sample itself should

be free flowing, un-diluted with no air bubbles and processed in

a timely manner (less than 15 min). Arterial gases provide

information about pulmonary gas exchange, while central

venous samples give information regarding the acidebase status

of tissues in conditions of severe hypoperfusion. If the sample is

taken from an arterial line running heparinized saline solution,

there is a risk of dilution with erroneously low pCO2 and bicar-

bonate values. Central venous pH is lower than arterial pH by

approximately 0.03 units and venous pCO2 is higher by 0.8 kPa.

These differences increase in hypoventilation and circulatory

failure, with a pH difference up to 0.1 units and pCO2 difference

of up to 3.2 kPa.

Capillary blood samples are commonly used in the neonatal

unit for blood gas estimation. The capillary values for pH and

pCO2 are usually within 1 kPa of the corresponding arterial

values. However, they have their limitations and are less reli-

able for babies with hypotension, poor perfusion or cold

peripheries. Capillary blood samples also cannot reliably

monitor oxygenation status or predict the degree of hypo-

xaemia. In these settings, an arterial blood gas is more useful,

although invasive.

Non-invasive monitoring using pulse oximetry to monitor

oxygen saturation in blood (SpO2) and transcutaneous moni-

toring are useful adjuncts to blood gas measurements. Pulse

oximeters work on the principle that oxygenated and deoxy-

genated haemoglobin absorb different wavelengths of light. The

oximeter provides a measure of the oxygen saturation of pulsatile

arterial blood compared with that from non-pulsatile venous

blood. It can be unreliable in hypoperfusion or with movement

artefacts. Transcutaneous electrodes measure oxygen (TcPO2)

and CO2 pressures (TcPCO2). They rely on diffusion from vaso-

dilated vessels in heated skin, so are particularly useful in the

newborn period when the skin is thin, but can be unreliable in

hypoperfusion. Transcutaneous levels usually match arterial

blood levels closely, thus careful application can be used to

monitor trends and may allow the frequency of blood gas

sampling to be reduced. Continuous end tidal CO2 monitors can

also be useful for monitoring CO2 levels in infants with stable

ventilation.

Clinical interpretation of blood gases

Blood gas analyzers measure pH, pCO2, PO2 and HCO3� (Table

2). They measure ‘actual’ HCO3� in the blood sample from

which they calculate ‘actual’ BE. Normally all the bicarbonate in

blood is produced by the ‘metabolic’ system, i.e., liver and

kidneys. However hypercapnia increases H2CO3 dissociation into

bicarbonate. ‘Standardised’ figures therefore calculate bicar-

bonate derived from CO2 and subtract this from the actual

measurements to reflect metabolic function. Thus in patients

with respiratory problems, it is advisable to use the ‘standard-

ized’ HCO3� and BE. Normal ranges vary slightly with gestation

& postnatal age and the desired values of these parameters for


any specific medical condition can vary with clinical practice, for

example with approaches such as “permissive hypercarbia” or

“gentle ventilation”. Understanding that pH is maintained by the

ratio of HCO3�/pCO2, a patients’ acidebase status can be readily

ascertained from a blood gas.

The following steps can be used as a guide for blood gas

interpretation (see Table 1):

1. Is there acidaemia or alkalaemia, i.e., pH less than 7.30 or pH

more than 7.45?

2. Is it primarily metabolic, i.e., HCO3� less than 19 or more

than 24 mmol/litre & BE less than �3 or more than þ3? OR

Is it primarily respiratory, i.e., pCO2 less than 4.5 or more

than 6 kPa?

3. Is there any compensation?

4. Is there a mixed disorder present, i.e., values outside the

normal compensation?

Blood gases should always be interpreted in conjunction with

information from a detailed history and thorough clinical

examination, the type of sample and non-invasive monitoring.

The prudent use of blood gas analysis in conjunction with

continuous monitoring allows optimal assessment of the patient

and prompt intervention when required, the response to which

can then be monitored and the blood gas repeated after an

appropriate time period to ensure clinical improvement.

Management should always be directed at the underlying cause

and an understanding of the processes involved in acidebase

balance aids this interpretation. A

FURTHER READING

1 Greenbaum Larry A. Chapter 52.7. Acidebase balance. In:

Kleigman RM, Behrman RE, Jenson HB, Stanton BF, eds. Nelson text-

book of paediatrics. 18th Edn. WB Saunders, 2007.

2 Quigley R, Baum M. Neonatal acid base balance and disturbances.

Semin Perinatol 2004 Apr; 28: 97e102.

3 Adelman RD, Solhaug MJ. Chapter 52. Hydrogen ion. In: Behrman RE,

Kleigman RM, Jenson HB, eds. Nelson textbook of paediatrics. 16th

Edn. WB Saunders, 2000.

4 Modi N. Chapter 39. In: Rennie JM, Roberton NRC, eds. Textbook of

neonatology. 3rd Edn. Churchill Livingstone, 1999.

5 Cloherty JP, Eichen EC, Stark AR, eds. Manual of neonatal care.

6th Edn. Lippincott Williams & Wilkins, 2008.

6 Woodrow P. Essential principles: blood gas analysis. Nurs Crit Care

2010 MayeJun; 15: 152e6.



Practice points

C A stable pH is essential for optimal cellular metabolism and

can be challenging in the newborn period

C Acidebase balance is regulated by buffers, the respiratory &

renal systems

C When the compensatory response falls outside the expected

value, a mixed acidebase disorder is likely

C Blood gases can be used to monitor acidebase balance

C Blood gases should always be interpreted in conjunction with

information from the clinical history & examination and non-

invasive monitoring


7 Lorenz JM, Kleinman LI, Markarian K, Oliver M, Fernandez J.

Serum anion gap in the differential diagnosis of metabolic

acidosis in critically ill newborns. J Pediatr 1999 Dec; 135:

751e5.

8 Brouilette RT, Waxman DH. Evaluation of the newborn’s blood gas

status. Clin Chem 1997 Jan; 43: 215e21.

9 Edwards SL. Pathophysiology of acid base balance: the theory practice

relationship. Intensive Crit Care Nurs 2008; 24: 28e40.

10 Williams AJ. ABC of oxygen: assessing and interpreting

arterial blood gases and acidebase balance. BMJ 1998 Oct 31;

317: 1213e6.

11 Foxall F. Arterial blood gas analysis: an easy learning guide. 1st Edn.

London: M&K Update Ltd, 2008.

12 Hennessey IAM, Japp AG. Arterial blood gases made easy. 1st Edn.

Edinburgh: Churchill Livingstone, 2007.

PAEDIATRICS AND CHILD HEALTH 22:4 148 � 2011 Elsevier Ltd. All rights reserved.



Recognition andmanagementof neonatal seizuresVaishali Patel

Amit Kandhari

Shobha Cherian

AbstractSeizures in the neonate are frequent and often the only sign of neurological

dysfunction. They can be caused by a variety of conditions ranging from the

benign and self-limiting to life threatening disorders. Several unresolved

issues remain concerning when to initiate treatment, duration of therapy

and what anticonvulsant medications should be used. Recent insights

into the pathophysiology may provide the foundation for better treatment.

Keywords neonatal; seizures; treatment

Introduction

Seizures occur more frequently in the neonatal period than at any

other time during the human lifespan. The overall incidence is 1e3

per 1000 live births. The incidence in high-risk premature infants

maybeashigh as 57e132per 1000 live births. 80%of seizures occur

in the first week of life and are often the first sign of neurological

dysfunction. Rapid diagnosis of the underlying cause is important in

order to institute specific therapy. Although the present treatment of

neonatal seizures is often unsatisfactory, considerable progress has

been made in understanding the pathogenesis of seizures and the

response by the neonate to anticonvulsant therapy.

Pathophysiology

Seizures in both term and preterm infants differ considerably

from those in older children and adults both in frequency and

clinical presentation. Understanding the probable mechanisms

for the genesis of seizures within the immature central nervous

system will lead to an understanding of its varied clinical

presentation and the conundrums associated with treatment.

Biochemical basis of neonatal seizures

A seizure is a sudden, excessive, synchronous electrical discharge

of a group of neurons within the central nervous system. This

Vaishali Patel MBBS MD MRCPCH is Paediatric Registrar in the Neonatal

Intensive care unit at the University Hospital of Wales, Cardiff, UK.


Amit Kandhari MBBS MRCPCH is Paediatric Registrar in the Department of

Paediatrics at the Princess of Wales Hospital, Bridgend, UK.


Shobha Cherian MB BS MD MRCP(UK) MRCPCH is Consultant Neonatologist in

the Neonatal Intensive care unit, University Hospital of Wales, Cardiff, UK.



electrical discharge is due to depolarization of neurons, resulting

from an influx of sodium ions. Negative potential across neuronal

cells ismaintained by anATP (adenosine triphosphate) dependant

Naþ (sodium)e(Kþ) potassium pump. Depolarization can result

from one of four mechanisms:

1 Decreased energy production and failure of ATP dependant

NaþeKþ pump e.g. following hypoxia-ischaemia and hypog-

lycaemia

2 Excessive release of the excitatory neurotransmitter gluta-

mate and reduced (energy dependant) uptake into cells e.g.

following hypoxia-ischaemia

3 Deficiency of inhibitory neurotransmitters: gamma-amino

butyric acid (GABA) is the predominant inhibitory neuro-

transmitter in the brain. A deficiency of pyridoxine,

a cofactor for GABA synthesis, will lead to reduced levels of

GABA and consequently to seizures

4 Hypocalcaemia and hypomagnesaemia also cause seizures

as both calcium and magnesium inhibit Naþ movement

across neuronal cells.

Neurodevelopmental basis for neonatal seizures

Anatomical: neonates rarely develop easily recognizable tonic-

clonic seizures. Motor phenomena are often asynchronous and

not well propagated. Subtle seizures presenting with sucking,

occulomotor phenomenon and apnoea are more frequent. This

may be because myelination, dendritic outgrowth and formation

of synaptic junctions in the cerebral cortex are relatively

incomplete, while the development of subcortical and limbic

structures are relatively advanced.

Physiological: glutamate is the major excitatory neurotransmitter

in the central nervous system, while GABA is the major inhibitory

neurotransmitter. Glutamate receptors are located at synapses, on

non-synaptic sites, on neurons and on glia. There are three types

of ionotropic (i.e. linked to calcium and sodium ion channels)

glutamate receptors; the NMDA receptor (N-methyl-D-aspartate),

the AMPA receptor (alpha-amino-3 hydroxy-5-methyl-4-iso-

xazolepropionic acid) and the kainite receptor. Experimental

evidence has shown that in the neonatal brain, both NMDA and

AMPA receptors are over expressed and their subunit composition

renders them susceptible to enhanced excitability. NMDA

receptor antagonist drugs (ketamine, meperidine) suppress

seizures, however they cause deep sedation and induce apoptotic

cell death in the immature brain. AMPA receptor antagonists

(topiramate) appear to be potentially highly effective against

neonatal seizures.

To compound the relative excess of ‘excitability’ of the perinatal

brain, GABA receptor expression is low in early life. In addition,

GABA receptor activation produces excitation rather than inhibi-

tion of neurotransmission. This paradoxical action of GABA in the

neonate is due to age related differences in chloride homeostasis.

Chloride transport is a function of two membrane pumps. In the

neonate NaþeKþeCl� co-transporter (NKCC1) imports large

amounts of chloride into the neuron. Chloride levels within

the neuron remain high because of relative under expression of

KþeCl� co-transporter, (KCC2) which is a Kþ exporter. When the

chloride permeable GABA receptors are activated, chloride flows

out of the cell depolarizing it. As a result GABA activation is

excitatory rather than inhibitory. This explains why in the




neonatal period, GABA agonists e.g. barbiturates and benzodiaz-

epines, are relatively ineffective. With maturation, NKCC1

expression diminishes and KCC2 expression increases. GABA

activation then causes chloride to flow into the cell and hyperpo-

larize it (Figure 1). Maturation of these chloride co-transporters

occurs in a caudal to rostral direction with maturation of spinal

cord and brain stem receptors occurring before that of the cerebral

cortex. This explains why treatment with GABA agonists often

results in suppression of motor manifestations with persisting

electrical seizures. Early clinical trials have shown the NKCC1

inhibitor, bumetanide, to be effective against neonatal seizures.

Clinical features

There are four major types of neonatal seizures. Their incidence,

clinical manifestations and EEG (electroencephalograph) corre-

lates are shown in Table 1. Seizures must be differentiated from

three types of non-convulsive movements: jitteriness, benign

sleep myoclonus and hyperekplexia.

Jitteriness: jitteriness is tremulousness. It is not accompanied by

ocular, orobuccal or autonomic phenomenon and can be elimi-

nated by gentle passive flexion of the affected limb. Jitteriness

may occur with hypoglycaemia, hypocalcaemia, drug with-

drawal and hypoxic-ischaemic encephalopathy.

Benign sleep myoclonus: is characterized by bilateral, repetitive,

myoclonic movements involving the upper or lower limbs that

occur only during sleep. Myoclonus may be provoked by gentle

rocking of the mattress that stops abruptly on arousal of the infant.

The episodes can last for several minutes. The interictal EEG is

either normal or shows minor non-specific changes. It resolves

within about 2 months and the neurological outcome is normal.

Figure 1 Transmission of electrical impulses at synapses in the neonatal perio

neuron (left panel) by activating NMDA and AMPA receptors. GABA release (righ

in the immature CNS. NKCC1: sodium-chloride co-transporter, KCC2: potassiu

alpha-amino-3 hydroxy-5-methyl-4-isoxazolepropionic acid, Naþ: sodium, Ca2


Hyperekplexia: also known as ‘startle disease’, is characterized

by an exaggerated startle response and sustained tonic spasm to

handling and to unexpected auditory, visual stimuli. Nocturnal

myoclonus and generalized hypertonia may occur which can

interfere with bathing, diaper change and feeding. Forced truncal

flexion can terminate an episode. The EEG is invariably normal.

It is an autosomal dominant disorder caused by increased

excitability of reticular neurons in brain stem. Hyperekplexia

disappears spontaneously by about 2 years of age.

Aetiology

The main causes of seizures are listed in Table 2. Asphyxia is the

commonest cause and accounts for up to 40% of all neonatal

seizures. Other frequent causes are cerebral arteriovenous

infarction (20%), intracranial haemorrhage (12e20%), infection

(3e20%), hypoglycaemia and hypocalcaemia (3e19%),

congenital cerebral anomaly (5e10%). The aetiology remains

unknown in 10e13% cases.

Epilepsy syndromes: are of unclear aetiology and unlike the

non-convulsive movements detailed earlier have documented

EEG abnormalities.

Benign idiopathic neonatal seizure (fifth day fits) e seizures

begin by day 3e5 and may last for 2 weeks. The diagnosis is one

of exclusion. The interictal neurological examination and EEG

are normal. The cause remains unknown, however low CSF zinc

deficiency has been demonstrated in a few cases. The develop-

mental outcome is favourable.

Benign familial neonatal convulsions e this rare condition

has an autosomal dominant inheritance involving mutations in

voltage-gated potassium channel genes. Seizures occur in

an otherwise healthy neonate on day 2e3 of life and may last for

d. Presynaptic release of glutamate results in excitation of post synaptic

t panel) results in hyperpolarization in the mature CNS and depolarization

m-chloride co-transporter, NMDA: N-methyl-D-aspartate, AMPA:þ: calcium, Cl�: chloride.



Incidence, clinical characteristics and EEG correlates in various forms of neonatal seizures

Clinical seizure Incidence Clinical manifestation EEG activity

Subtle 50%

- More common in premature infants

- Orobuccal: sucking, chewing, lip smacking,

hiccups

- Ocular: blinking, staring, horizontal devia-

tion of eyes

- Limbs: cycling, rowing

- Autonomic: alteration in heart rate, blood

pressure, apnoea, colour change

þ/�

Clonic 25e30% - Repetitive jerking

- Focal, multifocal or generalized (rare)

þ

Tonic 5% - Sustained posturing of limbs/trunk

- Deviation of head/eyes

- Focal, generalized

- Generalized may mimic decerebrate or

decorticate posturing

Focal: þGeneralized: �

Myoclonic 15e20% - Rapid isolated jerks

- Common in flexor group of muscles

- Focal, multifocal, generalized, axial

Focal and multifocal: �Generalized:þ

Table 1


1e6 months. There is often a family history of neonatal seizures

and development is normal. However, secondary epilepsy may

occur in 10e15%.

Early myoclonic encephalopathy e is characterized by severe

recurrent myoclonic and focal clonic seizures, which then prog-

resses to tonic spasms. Onset is in the first weeks of life, however

intrauterine onset has been documented. The underlying cause

may be an inborn error of metabolism e.g. nonketotic hyper-

glycinemia but in 50% cases the cause is unknown. The EEG

shows burst suppression that is enhanced by sleep and persists

beyond 1 year of age. Prognosis is poor and seizures are resistant

to treatment needing multiple anticonvulsant drugs.

Early infantile epileptic encephalopathy (Ohtahara

syndrome) e seizures begin in the first 3 months of life as tonic

spasms that progress to myoclonic spasms. Seizures are resistant

to treatment and are often accompanied by severe encephalop-

athy. The EEG shows burst suppression like early myoclonic

encephalopathy, but is not altered by sleep or waking. The

prognosis of this syndrome is also poor with death or severe

psychomotor retardation occurring in the first few years of life.

Management of neonatal seizures

There are several fundamental controversies regarding the

diagnosis and treatment of neonatal seizures:

1 Recognition of seizures

2 Why treat?

3 When to treat?

4 What is appropriate treatment?

5 How long to treat?

Recognition of seizures

Neonatal seizures are difficult to recognize, as there are often no

clinical manifestations of electrographic seizures; a phenomenon


called electro clinical dissociation. This is believed to occur as

connectivity within the nervous system is not fully developed

and myelination is incomplete. Infants may show no signs or

very subtle signs of electrical seizures. As up to 80% of EEG

documented seizures are not accompanied by clinically observ-

able seizures, full channel EEGs are essential for diagnosis and

for assessing efficacy of treatment. Video-EEG monitoring, where

continuous EEG monitoring is accompanied by contempora-

neous video recording documenting suspicious clinical behav-

iours, is considered the neurophysiological ‘gold standard’.

Several subtle seizures, tonic seizures and myoclonic jerks have

no EEG correlates and are believed to be generated at a deep

subcortical level.

In viewof this diagnostic complexity, some argue that diagnosis

of neonatal seizures should not be based on clinical observation

alone. As EEGs are difficult to obtain round the clock on the

neonatal unit, initial diagnosis and treatment is based on clinical

observation. Several centres are now using amplitude integrated

EEG (aEEG) as a readily available bedside tool. This device uses

a single or dual channel EEG and acute variations in spectral width

to detect seizures. Several reports show that aEEG detects

approximately 75% of seizures detected by conventional EEGs.

The complete list of investigations to be performed would

vary with aetiology but should include blood glucose, serum

electrolytes, calcium, magnesium, full blood count, C reactive

protein, blood gas analysis, blood culture and lumbar puncture.

Urine toxicology, TORCH screen and a metabolic screen should

be performed if indicated. Neuroimaging in the form of a cranial

ultrasound scan, magnetic resonance imaging or computed

tomography are often indicated.

Why treat?

The impact of convulsions on the immature brain has long been

debated. Although the immature brain is more prone to seizures,



Aetiology of neonatal seizures

Hypoxic Ischaemic encephalopathy

Cerebro-vascular

Arterial and venous stroke

Sinus thrombosis

Intracranial haemorrhage

Intraventricular/periventricular

Subarachnoid

Subdural/epidural

Trauma

Birth trauma

Non accidental injury

Intracranial infection

Bacterial meningitis (Group B Strep, E. coli, Listeria)

Viral encephalitis (Herpes simplex, Enterovirus)

Intrauterine TORCH infection

Malformations of cerebral development

Polymicrogyria

Pachygyria

Lissencephaly

Neurocutaneous syndromes

Tuberous sclerosis, incontinentia pigmenti

Electrolyte and metabolic abnormalities

Hypoglycaemia

Hypocalcaemia

Hypomagnesaemia

Hyponatraemia, Hypernatraemia

Neonatal drug withdrawal

Kernicterus

Inborn errors of metabolism

Amino acid, organic acid, urea cycle disorders

Mitochondrial and peroxisomal disorders

Pyridoxine dependency

Inadvertent injections of local anaesthetics during delivery

Epilepsy syndromes

Benign idiopathic neonatal convulsions (fifth day fits)

Benign familial neonatal convulsions

Early myoclonic encephalopathy

Early infantile epileptic encephalopathy (Ohtahara syndrome)

Table 2


it is more resistant to post seizure damage than the mature brain.

However, evidence from several animal models and a few studies

on human infants has shown that prolonged and recurrent

seizures have widespread effects, which are deleterious to the

developing brain.

The proposed mechanisms for brain injury are shown in

Figure 2. Animal studies have shown that seizures result in

reduced density of dendritic spines in hippocampal pyramidal

neurons, delayed neuronal loss, decreased neurogenesis,

synaptic reorganization and changes in hippocampal plasticity.

Magnetic resonance spectroscopy studies on human infants with

seizures have shown disturbances in cerebral metabolism and

adverse long-term neurological sequelae. This suggests that seizures

might cause or exacerbate cerebral injury by increasing cerebral


metabolic demands above that of energy supply. Infants with

hypoxia-ischaemia and clinical seizures have been shown to have

a significantly worse outcome than those without seizures, inde-

pendent of the severity of hypoxia-ischaemia.Also, post-hoc analysis

of data from theCool-Caphypothermia trial revealed that theabsence

of seizures was an independent predictor for better outcomes.

Unfortunately there is insufficient evidence from randomized

controlled trials to either support or refute the use of anticon-

vulsants for the treatment of neonatal seizures. The question of

whether aggressive treatment of seizures, most of which occur

against the background of pre-existing brain injury, confers

benefit requires further clinical investigation.

When to treat?

Poorly controlled and prolonged seizures are associated with poor

neurodevelopmental outcome, however the severity of the

underlying disease may account for both the poor seizure control

and outcome. Several questions remain unanswered. How long

must a seizure last before it leads to brain injury? Does the degree

of CNS injury vary with type of seizure, particularly those without

documented EEG changes? Does treatment with anticonvulsants

alter developmental outcome when the underlying disorder is

controlled for? Should electrical seizures be treated or should only

clinical seizures be treated? There are no randomized trials that

answer all these questions. Some feel it is reasonable to treat

frequent and prolonged seizures especially if they are associated

with cardiorespiratory compromise. As there is no agreed defini-

tion of ‘prolonged’ seizures, many neonatologists take a pragmatic

approach and treat seizures by the ‘rule of 3’ i.e. treat if there are

more than three seizures per hour or if any one seizure lasts more

than 3 min. Others feel that all seizures, both electrical and clin-

ical, should be treated to prevent further brain injury.

What is appropriate therapy?

The evaluation and initial management of the infant with clinical

seizures should not await EEG confirmation. The following basic

principles should be followed:

1 Support airway, breathing and circulation

2 Check blood glucose and secure vascular access. If hypo-

glycaemia is present in a convulsing infant, give 2 ml/kg of

10% dextrose intravenously and start maintenance dextrose

solution to achieve normal blood glucose levels.

3 Investigate and treat the underlying cause

4 Balance the benefits of controlling some or all of the seizures

with anticonvulsant drugs against risks of potential side

effects from the medications.

Anticonvulsant therapy:

Phenobarbitone e is a GABA agonist. It controls approximately

70% of clinical seizures and 50% of electrical seizures. This is

probably because many GABA receptors are excitatory and have

immature chloride channels. (See Pathogenesis) However, it

continues to be the drug of choice in neonates as it has been well

studied, has a longhalf-life (2e4days) andenters theCSF rapidly. It is

given as a loading dose of 20 mg/kg (which may be repeated if the

initial dose is ineffective) andachieves therapeutic levels (20e40mg/

litre) in the serum within a short time.

Phenytoin e acts by reducing electrical conductance in

neurons by stabilizing sodium channels. Both phenytoin and



Prolonged/ repeated seizures affects

Respiratory system Cardiovascular system Energy metabolism Neuro-transmitters

Brain injury

Po2 Pco2 BP Glycolysis ADPRe-uptakeof EAA

Releaseof EAA

EAABrain glucose

Haemorrhage

CBF CBFLactate+ H+

ATP/

Figure 2 Mechanisms for the development of brain injury following prolonged or repeated seizures. PO2: oxygen pressure, PCO2: carbon dioxide pressure,

BP: blood pressure, ATP: adenosine triphosphate, ADP: adenosine diphosphate, EAA: excitatory amino acid, Hþ: hydrogen ion, CBF: cerebral blood flow.


phenobarbitone are equally but incompletely effective in

achieving complete control of clinical and electrographic

seizures. Their combination achieves control of 85% of clinical

seizures and up to 80% of electrical seizures. Phenytoin should

be given in the dose of 20 mg/kg intravenously slowly, under

cardiac monitoring, as it can cause hypotension and arrhythmias

especially in the presence of myocardial damage accompanying

hypoxia-ischaemia. Skin rashes have also been reported.

Benzodiazepines e are GABA agonists and are used to control

seizures where the combination of phenobarbitone and phenytoin

has been ineffective. Both clonazepam and midazolam are widely

used. Clonazepam has a long half-life (24e48 h) and is given as

a bolus dose of 100 mg/kg. It causes increased respiratory and oral

secretions that may interfere with respiratory function. Midazolam

has a shorter half-life (approximately 6 h in sick and premature

infants) and is administered as a bolus dose of 200 mg/kg followed

by infusion of 60e300 mg/kg/h. It has been reported to cause

myoclonic jerks and dystonic posturing in premature infants.

Lorazepam has been used in neonates in the dose of 100 mcg/kg

given intravenously and has duration of action of 6e24 h.

Lidocaine e suppresses seizures by suppressing sodium entry

into the neuron. It has been shown to decrease seizure burden in

up to 60e75% infants who have not responded to phenobarbi-

tone and benzodiazepines. A loading dose of 2 mg/kg is followed

by an infusion of 6 mg/kg/h for 6 h, 4 mg/kg/h for 12 h and then

2 mg/kg/h for 12 h. It has a narrow therapeutic range and can

induce seizures in high doses. As it can induce cardiac arrhyth-

mias and hypotension, it should not be given with phenytoin and

must be administered under continuous cardiac monitoring.

Levetiracetam (Keppra) e is a commonly prescribed anti-

convulsant medication in older children and adults. Its anticon-

vulsant action is not well understood but is believed to impede


nerve conduction across synapses. Pharmacokinetic and safety

data in neonates is lacking, however it has been tried with some

success in seizures resistant to other medications.

Topiramate e has multiple proposed mechanisms of action. It

acts as a glutamate antagonist by blocking AMPA receptors, is

a Naþ channel blocker and has been shown to be neuroprotective

following hypoxia-ischaemia in animal models. It has not yet

been studied for safety, dosing or efficacy in neonates, however,

a recent retrospective, cohort study reported good results in six

newborn infants with seizures refractory to standard agents.

Bumetanide e is commonly used as a diuretic. It inhibits the

NKCC1 co-transporter, creating a Cl� gradient more like the adult

neuron. In animal models this switches the GABA equilibrium

potential from excitatory to inhibitory. Although it has been

shown to be a promising therapy in the laboratory, there are no

reported clinical trials.

Other drugs:

Paraldehydeewas used as an effective adjunct anticonvulsant.

It has a short half-life and is eliminated by the lungs and liver and is

not affected by altered renal function. However, as it must be

administered per rectally, is reported to cause pulmonary oedema,

hepatic necrosis and hypotension, it is no longer widely used.

Sodium valproate e has been used to treat intractable

seizures. Due to serious concerns regarding hyperammonaemia

and hepatotoxicity its value is uncertain.

Carbamazepine e has been reported to be effective in the

treatment of neonatal seizures. However it must be administered

orally and blood levels are very variable.

Pyridoxine e diagnosis of pyridoxine dependent seizures is

suspected when an infant develops multifocal clonic seizures

resistant to conventional anticonvulsants soon after birth. 50e100




mg of pyridoxine should be administered intravenously under EEG

monitoring. Seizure activity stops within minutes and the EEG

normalizes.

How long to treat?

Practice points

C The neonatal brain is more prone to seizures than the mature

brain due to an over expression of glutamate receptors (NMDA,

AMPA) and under expression of GABA receptors. In addition

GABA receptors are excitatory in the neonate

C Electrical seizures may occur without any clinical manifesta-

tion, a phenomenon known as electro clinical dissociation

C The need to treat electrical seizures is controversial

C Phenobarbitone remains the mainstay of therapy despite

being ineffective in a significant proportion

C Levetiracetam, topiramate and bumetanide may have a role in

the future

Once seizures are controlled, should maintenance therapy be

administered and if so for how long? Once again, there is no

consensus on this issue. Antiepileptic drugs have a deleterious

effect on the developing brain. Animal studies have demon-

strated that systemic therapy with phenobarbitone, benzodiaze-

pines, phenytoin and valproate increase apoptotic neuronal

death. Combination therapy produced greater adverse effects.

Abnormal cognitive development has been documented in

infants and children that have received phenobarbitone.

The risk of developing recurrent seizures once seizure

control is achieved and anticonvulsants discontinued is under

10% in infants with normal EEG background activity. In infants

with both abnormal neurological examination and EEG back-

ground activity, it may be as high as 50%. It would therefore be

prudent to administer maintenance therapy (phenobarbitone 3

e5 mg/kg) only if the neurological examination or EEG back-

ground activity is abnormal. Once neurological examination is

normal, phenobarbitone should be withdrawn. In most cases

this can be done before discharge from the neonatal unit. If

neurological examination remains abnormal, obtain an EEG and

discontinue phenobarbitone if there is no electrical seizure

activity.

Prognosis

The long-term neurodevelopmental outcome in infants with

seizures varies with aetiology, gestational age, seizure type and

interictal EEG. The prognosis following benign familial seizures

is excellent while 30e50% of infants with hypoxia-ischaemia,

hypoglycaemia and meningitis have abnormal developmental

outcome. Nearly all infants with CNS malformations have a poor

outcome. When seizures occur with normal EEG background

activity, the outcome is good, while those with low voltage EEG,

burst suppression or electrocerebral silence are associated with

neurodevelopmental deficits in more than 90% cases. Only 20%

premature infants with a birth weight less than 1500 g and

seizures have a normal outcome as compared to 60% of term

infants. Intractable seizures, generalized myoclonic and tonic

seizures are often associated with poor outcome. A


FURTHER READING

1 Lawrence R, Inder T. Neonatal status epilepticus. Semin Pediatr

Neurol 2010; 17: 163e8.

2 Glass HC, Glidden D, Jeremy RJ, Barkovich J, Ferriero DM, Miller SP.

Clinical neonatal seizures are independently associated with

outcome in infants at risk for hypoxic-ischemic brain injury. J Pediatr

2009; 155: 318e23.

3 Levene M. The clinical conundrum of neonatal seizures. Arch Dis

Child Fetal Neonatal Ed 2002; 86: 75e7.

4 Boylan GB, Rennie JM, Chorley G, et al. Second-line anticonvulsant

treatment of neonatal seizures: a video-EEG monitoring study.

Neurology 2004; 62: 486e8.

5 Volpe JJ. Neonatal seizures. In: Neurology of newborn. 5th Edn. W. B.

Saunders, 2009; 203e244.

6 Silverstein FS, Ferriero DM. Off-label use of antiepileptic drugs for the

treatment of neonatal seizures. Pediatr Neurol 2008; 39: 77e9.

7 Dzhala VI, Talos DM, Sdrulla DA, et al. NKCC1 transporter facilitates

seizures in the developing brain. Nat Med 2005; 11: 1205e13.

8 Booth D, Evans DJ. Anticonvulsants for neonates with seizures.

Cochrane Database Syst Rev 2004; 18: CD004218.

9 Bassan H, Bental Y, Shany E, et al. Neonatal seizures: dilemmas in

workup and management. Pediatr Neurol 2008; 38: 415e21.

10 El-Dib M, Chang T, Tsuchida TN, Clancy RR. Amplitude-integrated elec-

troencephalography in neonates. Pediatr Neurol 2009; 41: 315e26.

11 Jensen FE. Neonatal seizures: an update on mechanisms and

management. Clin Perinatol 2009; 36: 881e900.




The role of brain MRIscanning in the newbornPia Wintermark

AbstractVery sick newborns are at high risk for brain injuries and adverse neuro-

developmental outcomes. Accurate diagnosis of these injuries and

adequate prognostication of future outcome is one of the most difficult

tasks confronting caregivers in neonatal intensive care units. In the

past several years, there have been tremendous advancements in the

development of magnetic resonance imaging (MRI) technologies devoted

to studying the newborn brain. MRI is now slowly becoming the new stan-

dard of care for evaluating the exact nature and extent of brain injuries in

sick newborns, as well as reliably predicting the future prognosis of these

newborns. In addition, it is giving unique insights about how brain

injuries develop in these patients and how they further impact brain

maturation. This will probably help in the future to refine therapeutic

strategies offered to these patients, and to evaluate the efficiency of

such changes. In this article, we thus review some of the current and

future roles of brain MRI scanning in the newborn.

Keywords brain injuries; magnetic resonance imaging; newborn brain

Diseases leading to brain injuries in newborns

Hypoxic-ischaemic encephalopathy

Stroke

Cerebral sinovenous thrombosis

Prematurity

Congenital cardiopathy

Infection: Cytomegalovirus, Enterovirus, Parechovirus, Herpes

Simplex Virus, .

Introduction

Very sick newborns are at high risk for brain injuries and adverse

neurodevelopmental outcomes. A major issue confronting care-

givers who work with these children is to provide the most

accurate prognosis about their future to their families. The major

challenge of researchers in the neonatal neurology field is to find

innovative treatments to prevent or repair brain injuries in these

newborns. Nowadays, advances in modern neuroimaging are

continuously ongoing, allowing us to improve our understanding

of how neonatal brain injuries develop and how they impact on

further brain development. Among the available neuroimaging

techniques for these patients, magnetic resonance imaging (MRI)

is proving more and more to be useful, even if more expensive

and more challenging to obtain compared to head ultrasounds.

Brain MRI is slowly becoming the new standard of care for

evaluating the exact nature and extent of brain injuries in sick

newborns, as it provides good spatial resolution and thus accu-

rate anatomical details that cannot be obtained by any other

imaging modality. Furthermore, more and more studies are

Abbreviations: DTI, diffusion-tensor imaging; DWI, diffusion-weighted

imaging; MRI, magnetic resonance imaging; NICU, neonatal intensive care

unit; SWI, susceptibility-weighted imaging.

Pia Wintermark MD is an Assistant Professor in the Department of

Pediatrics in the Division of Newborn Medicine, Montreal Children’s

Hospital, McGill University, Montreal, Canada, an associate Member in

the Department of Neurology and Neurosurgery at McGill University and

an Associate Member of the Integrated Program in Neurosciences at

McGill University. Conflict of interest: none.


demonstrating the role of MRI as a reliable predictor of the future

prognosis of newborns with brain injuries, as well as its potential

to give further insights into maturation, destruction and repair

processes occurring concurrently in a newborn brain. In this

article, we review some of the current and future roles of brain

MRI scanning in the newborn.

The newborn brain

The human brain begins forming very early in prenatal life (just

three-four weeks after conception), but continues to develop

years postnatally. Especially very active and complex processes

such as elaboration of dendritic and axonal ramifications,

establishment of synaptic contacts, selective elimination of

neuronal processes and synapses, proliferation and differentia-

tion of glia, and myelination, start during gestation, but continue

for several years after birth. This plasticity explains why the

newborn brain is more vulnerable to insults. Brain injuries in the

newborn occur thus against these active developmental events.

After the acute damage, neuronal circuits pursue develop-

mental processes with a significant cell loss, leading to disruption

of normal developmental processes. This can cause dramatic

deterioration of subsequent brain development and brain func-

tions. The key issue is to understand better the molecular and

cellular mechanisms of these processes and their timing. This is

of utmost importance for developing new strategies to improve

prevention and repair of brain injuries in the newborn.

Diseases leading to brain injuries in newborns

Many different diseases can lead to brain injuries in newborns.

This review does not present an exhaustive list, but rather

discusses those most frequently encountered in the neonatal

intensive care unit (NICU) (Table 1).

Newborns with hypoxic-ischaemic encephalopathy displayed

heterogeneous brain injuries. Patterns of these injuries have been

associated with varying clinical presentations and different

neurodevelopmental outcomes. They have been classically

Neonatal hypoglycemia

Other metabolic diseases: inborn errors of metabolism, .

Malformation of brain development: focal cortical dysplasia,

hemimegalencephaly, .

Anomalies of cerebral vasculature

Phakomatoses: tuberous sclerosis, neurofibromatosis, .

Brain tumours

Perinatal brain trauma

Table 1




described as basal ganglia injury pattern, boundary zone injury

pattern (watershed injury pattern) and total cortical injury

pattern, according to the injured parts of the brain. Perinatal

stroke (Figure 1) is another not-uncommon clinical entity

affecting the newborn with impact on long-term neurological

outcome. It is characterized by focal infarction of the brain

parenchyma, which is most often ischaemic in nature in

newborns rather than hemorrhagic. Newborns are also at higher

risk to develop cerebral sinovenous thrombosis, which is often

complicated by intraventricular haemorrhage associated with

unilateral thalamic haemorrhage or bilateral white matter

involvement.

Premature newborns are prone to germinal matrix haemor-

rhage, intraventricular haemorrhage and periventricular hae-

morrhagic infarction, but also white matter and grey matter

injuries. They may display a large spectrum of white matter and

grey matter injuries, including white matter signal abnormality,

decreased white matter volume, cystic abnormalities in white

matter, ventricular dilatation, decreased myelination in the

posterior limb of the internal capsule, thinning of the corpus

callosum, grey matter signal abnormality, simpler gyral pattern,

increased subarachnoid space and cerebellar injuries. Newborns

with congenital cardiopathy are another population of newborns

at high risk of developing brain injuries. Even before cardio-

vascular surgery, these newborns have been shown to have

delayed third trimester brain growth, impaired white matter

maturation, reduced N-acetyl-aspartate and increased lactate,

suggesting an early onset of impaired brain growth and

development.

Among congenital infections leading to brain injuries,

congenital Cytomegalovirus infection is probably one of the most

devastating. Depending on the timing of the infection, it can

cause different types of brain injuries, i.e. ventriculomegaly,

subependymal cysts, intraventricular septa, calcifications,

cortical migrational disturbances, cerebellar hypoplasia and

temporal white matter injuries. White matter changes resembling

periventricular leukomalacia have been seen in cases of neonatal

meningoencephalitis with Enterovirus and Parechovirus, and an

infection by these viruses should be excluded when scattered

white matter injuries are present in term newborns without clear

explanations. Neonatal Herpes Simplex Virus Type 2 typically

causes multifocal brain injuries that are mostly located to

temporal lobes, brainstem and cerebellum. Bacterial meningitis

Figure 1 Stroke in a term newborn. MRI performed at 42 4/7 weeks of correcte

image, (c) Axial apparent diffusion coefficient (ADC) map, and (d) Axial diffus

territory.


may also lead to focal (single or multiple) diffuse and/or

hemorrhagic infarcts, with associated meningeal enhancement.

Symptomatic neonatal hypoglycemia is associated with

involvement of parietal and occipital cortex, subcortical white

matter, posterior limb of internal capsule, basal ganglia, and

thalami. Other metabolic diseases, especially inborn errors of

metabolism, are often very difficult to diagnose for the physician. In

these cases, brain MRI may play a useful role in diagnosis, as there

are few neuroradiological features for differentiating these errors.

Imaging appearance of inborn errors of metabolism includes white

matter injuries (true leukodystrophy or Wallerian degeneration),

grey matter injuries and involvement of basal nuclei.

Malformations of brain development, anomalies of cerebral

vasculature, phakomatoses (Figure 2), brain tumours and peri-

natal brain trauma are other potential diseases associated with

brain injuries in newborns. For each of these entities, brain MRI

plays an important role for diagnosis, treatment planning and

prognosis.

MRI sequences available for brain newborn imaging

A dedicated imaging protocol should be developed for scanning

the newborn brain. Standardized protocols are now available

through literature. The typical imaging protocol should include

all the essential MRI sequences to evaluate brain injuries in the

newborn, including high spatial resolution T1- and T2-weighted

imaging, diffusion-weighted imaging (DWI) and spectroscopy, as

well as fluid attenuated inversion recovery (FLAIR) imaging, MR

angiography, MR venography and susceptibility-weighted

imaging (SWI) if required. It may also include newer MRI

adjuncts, such as diffusion-tensor imaging (DTI) and diffusion

tractography, functional MRI and perfusion-weighted imaging.

Conventional T1- and T2-weighted imaging are the most

widely used sequences, specifically differentiating fat from water.

Perinatal lesions are typically at their most obvious on conven-

tional imaging between 1 and 2 weeks from birth, explaining

why this timing is most often chosen to perform an MRI in

newborns with hypoxic-ischaemic encephalopathy to define the

extent of the brain injuries and give a prognosis. DWI explores

micromovements of water molecules and can detect changes in

water diffusion associated with cellular dysfunction. It can

differentiate between cytotoxic and vasogenic oedema in cases of

early brain hypoxic-ischaemic infarcts. DWI is very useful for the

early identification of ischaemic tissue in the neonatal brain

d age (5 days of life). (a) Axial T1-weighted image, (b) Axial T2-weighted

ion-weighted image showed the infarction within the left cerebral artery



Figure 2 Tuberous sclerosis in a term newborn. MRI performed at 42 3/7 weeks of corrected age. (a) Sagittal T1-weighted image showed multiple

subependymal hamartomas (thin arrows) along the wall of the lateral ventricle. (b) Sagittal T1-weighted image showed a mass, presumably a giant cell

tumour (thick arrow) located at the level of the foramen of Monro. (c) Axial FLAIR image showed the cortical tubers (curved arrows) as hyperintense

signals. (d) Axial FLAIR image showed the white matter tracts (black arrowheads) extending from the tubers to the ventricular surface.


(Figure 1) but may underestimate the final extent of injuries,

particularly basal ganglia and thalamic lesions. DWI enables

quantitative measurements of apparent diffusion coefficient

(ADC) values in brain tissue. Spectroscopy allows exploration of

the molecular composition of the tissue and monitor biochemical

changes over time. Proton spectroscopy especially permits the

study of metabolites which can be altered in newborns devel-

oping brain injuries, such as lactates (product of anaerobic

glycolysis), N-acetyl-aspartate (NAA) (neuronal marker), gluta-

mine/GABA (neurotransmitter), creatine (energy metabolism),

choline (cell membrane marker) and myo-inositol (glial cell

marker). Early elevation of lactate and later reduction of N-

acetyl-aspartate have for example been demonstrated in cases of

newborns with hypoxic-ischaemic brain injuries. Metabolic data

from proton MR spectroscopy might also be a useful adjunct to

more conventional sequences, especially in helping to diagnose

an inborn error of metabolism (e.g. neonatal maple syrup urine

disease with specific peak at 0.9 ppm representing branched

chain amino acids and ketoacids).

Fluid attenuated inversion recovery (FLAIR) images are

T2-weighted images with the cerebrospinal fluid signal sup-

pressed. This imaging technique is more sensitive than T1- and

T2-weighted imaging for detecting pathologies specifically

located periventricular or subcortical within the brain paren-

chyma, such as for example tubers in tuberous sclerosis

(Figure 2). MR angiography might be added to the imaging

protocol for newborns, in order to look at the anatomic varia-

tions of the neonatal circle of Willis and to explore possible

vascular-related abnormalities, such as arteriovenous malfor-

mations. MR venography should be added to study possible

neonatal cerebral sinovenous thrombosis. SWI, another MR

imaging technique, accentuates the paramagnetic properties of

blood products such as deoxyhemoglobin, intracellular meth-

aemoglobin and haemosiderin. It is particularly well suited for

detecting intravascular venous deoxygenated blood as well as

extravascular blood products, especially for visualizing even

very small normal or abnormal veins or parenchymatous

haemorrhage.

More advanced MR-based neuroimaging approaches are now

also becoming available for newborns. DTI, functional connec-

tivity MRI (fcMRI), volumetric MR analysis, and surface based

morphometry (SBM) are now providing insight into structural

and functional brain maturation and the impact of brain injuries

on this development. Perfusion-weighted imaging, and more


specifically arterial spin labeling (ASL), now enables direct

noninvasive imaging of cerebral perfusion, and this may have

several implications for better understanding how brain injuries

develop in newborns, as abnormal brain perfusion is a key

mechanism in many of them.

Brain MRI for clinical purposes

Brain MRI currently has two main clinical applications in the

newborn. The main role of brain MRI is to define the extent of

brain injuries and to give important clues about the cause and

timing of an insult by the combination of the different MRI

sequences. As mentioned above, conventional imaging can

detect patterns of injuries that provide valuable information

about prognosis. The addition of DWI and spectroscopy might

provide guidance as to the timing of the event or help to diagnose

the aetiology of some brain injuries. All this information cannot

be obtained by any other neuroimaging modality in the newborn.

The second important clinical application of brain MRI in the

newborn is to provide valuable information about the long-term

prognosis. For example, abnormal findings on MRI at term

equivalent in very preterm infants have been shown to strongly

predict adverse neurodevelopmental outcomes at 2 years of age,

permitting the use of MRI at term equivalent in risk stratification

for these infants. Similarly, the pattern of perinatal lesions in

newborns with hypoxic-ischaemic encephalopathy or stroke

provides valuable information about prognosis in these patients.

Brain MRI for research purposes

MRI techniques offer great potential for further understanding how

brain injuries develop in the newborn brain and how they can be

prevented or repaired. The pattern of injuries seen in the newborn

is unique due to the combination of effective loss of brain tissue

and remodelling of subsequent brain development. The exact

pathogenesis of these lesions in the developing brain is still not

well understood. It is considered to be multifactorial, with impli-

cations of several prenatal, perinatal but also postnatal factors. As

mentioned previously, some of the most recent MRI sequences

adjuncts provide important insights into the trajectory of brain

development and the impact of injuries on this developmental

trajectory. Further results of such ongoing studies should improve

our knowledge and offer us important clues on how to develop

specific and efficient treatments in these newborns. For these

reasons, these more advanced neuroimaging techniques will




probably soon be part of the regular imaging protocol for

newborns, as they already are in adult MRI protocols.

As MR imaging is an excellent predictor of outcome following

perinatal brain injuries, it also has an huge potential to be used as

a surrogate, short-term outcome measure in clinical studies eval-

uating new interventional trials designed to reduce injuries in the

developing brain and improve neurodevelopmental outcome. It

will help in delineating which infants have the most to gain from

these newer therapeutic strategies, but may also act as an early

biomarker to gauge response to these new interventions.

Safety of brain MRI scanning in the newborn

MRI is a noninvasive and nonionizing neuroimaging technique

and does not involve harmful radiation. Thus, when available, it

should be preferred over computed tomography (CT) scan in

newborns, especially as it provides additional detailed imaging of

the brain parenchyma. Obtaining a brain MRI in this population

of sick children might appear challenging, especially when the

NICU is located at distance from the MRI suite. However, brain

MRI in newborns can be safely and easily obtained with

a minimum of requirements and training. A specialized team

should be dedicated to these scans. This team should include

neonatologists, neonatology fellows or neonatal nurse practi-

tioners, NICU nurses and respiratory therapists who care exclu-

sively for the newborn from NICU departure to return, as well as

neuroradiologists and MRI technicians who know how to oper-

ate, acquire and read neonatal neuroimaging. Time inside the

MRI should be balanced between the haemodynamic stability of

the newborn being imaged and the need to obtain optimal data

for analysis, diagnosis, and prognosis, as well as the time

available on the machine for such exam. As mentioned above,

specific imaging protocol should be developed for imaging of

these patients. A dedicated neonatal imaging coil or the best

available coil adapted for this type of imaging should be chosen

in order to optimize the signal-to-noise ratio.

Most of newborns do not need to be sedated for a brain MRI.

Neonates should be placed in a MRI-compatible isolette

or wrapped with one or two thin blankets and placed on

a MRI-compatible pillow containing small polystyrene spheres.

Once the neonate is placed on the pillow, the air in the

MRI-compatible pillow will be removed by suction to mould the

shape of the pillow to the infant’s head and body and further

reduce motion artifacts. If possible, feeding should be adminis-

tered after wrapping the newborn as described above, and some

time should be available to let the baby fall asleep naturally. Ears

should be covered with earmuffs to reduce noise exposure. All

metal objects should be removed form the newborn before

entering the MRI suite. Supportive therapies, including

mechanical ventilation, vasoactive infusions, antiepileptic treat-

ments and sedation, should be maintained throughout the exam

per current clinical practice. Additional sedation should be

administered only if deemed clinically necessary and should be

rarely required. Of note, most of the neonatal treatments, such as

hypothermia, mechanical ventilation and pressor support, can be

continued in the MRI with special precautions. Ventilation by

high-frequency oscillatory ventilation and/or administration of

nitric oxide might be the only treatments that cannot be admin-

istered in some of the MRI suites.


Conclusions

In conclusion, there have been tremendous advancements in the

development of MRI technologies devoted to the newborn brain.

Brain MRI should be the gold standard for neonates who have

encephalopathy or suspected brain injuries in order to clearly

define the extent of the injuries. It should be used to identify the

newborns, who are most at risk for subsequent neuro-

developmental disability and who may benefit from early inter-

vention services. But also, as often as possible, it should be used

to improve our knowledge of these brain injuries to further refine

our therapeutic strategies in these patients and to evaluate the

efficiency of such changes. A

FURTHER READING

1 MRI step-by-step, interactive course on magnetic resonance imaging:

http://www.imaios.com/en/e-Courses/e-MRI.

2 Ajayi-Obe M, Saeed N, Cowan FM, Rutherford MA, Edwards AD.

Reduced development of cerebral cortex in extremely preterm infants.

Lancet 2000; 356: 1162e3.

3 Barkovich AJ. MR imaging of the neonatal brain. Neuroimaging Clin N

Am 2006; 16: 117e35.

4 Barkovich AJ. An approach to MRI of metabolic disorders in children.

J Neuroradiol 2007; 34: 75e88.

5 Berfelo FJ, Kersbergen KJ, van Ommen CH, et al. Neonatal cerebral sino-

venous thrombosis fromsymptom tooutcome.Stroke2010;41:1382e8.

6 Burns CM, Rutherford MA, Boardman JP, Cowan FM. Patterns of cere-

bral injury and neurodevelopmental outcomes after symptomatic

neonatal hypoglycemia. Pediatrics 2008; 122: 65e74.

7 Counsell SJ, Tranter SL, Rutherford MA. Magnetic resonance imaging of

brain injury in the high-risk term infant.Semin Perinatol 2010; 34: 67e78.

8 De Vries LS, Groenendaall F. Patterns of neonatal hypoxic-ischaemic

brain injury. Neuroradiology 2010; 52: 555e66.

9 Govaert P, Ramenghi L, Taar R, De Vries L, DeVeber G. Diagnosis of

perinatal stroke I: definitions, differential diagnosis and registration.

Acta Paediatr 2009; 98: 1556e67.

10 Hagmann CF, De Vita E, Bainbridge A, et al. T2 at MR imaging is an

objective quantitative measure of cerebral white matter signal

intensity abnormality in preterm infants at term-equivalent age.

Radiology 2009; 252: 209e17.

11 Inder TE, Wells SJ, Mogridge NB, Spencer C, Volpe JJ. Defining the

nature of the cerebral abnormalities in the premature infant: a

qualitative magnetic resonance imaging study. J Pediatr 2003; 143:

171e9.

12 Inder TE, Warfield SK, Wang H, Huppi PS, Volpe JJ. Abnormal cerebral

structures present at term in premature infants. Pediatrics 2005; 115:

286e94.

13 Kate R, Atkinson D, Brant-Zawadzki M. Fluid-attenuated inversion

recovery (FLAIR): clinical prospectus of current and future applica-

tions. Top Magn Reson Imaging 1996; 8: 389e96.

14 Kersbergen KJ, Groeneendaal F, Benders MJ, de Vries LS. Neonatal

cerebral sinovenous thrombosis: neuroimaging and long-term follow-

up. J Child Neurol 2011; 26: 1111e20.

15 Kirton A, DeVeber G. Advances in perinatal ischemic stroke. Pediatr

Neurol 2009; 40: 205e14.

16 Lawrence RK, Inder TE. Anatomic changes and imaging in assessing

brain injury in the term infant. Clin Perinatol 2008; 35: 679e93.


http://www.imaios.com/en/e-Courses/e-MRI


Practice points

C There have been tremendous advancements in the develop-

ment of magnetic resonance imaging (MRI) technologies

devoted to the newborn brain.

C MRI is a noninvasive and nonionizing neuroimaging technique.

When available, it should be preferred over computed

tomography scan in newborns, especially as it provides addi-

tional detailed imaging of the brain parenchyma.

C Brain MRI in newborns can be safely and easily obtained with

a minimum of requirements and training. Most of newborns do

not need to be sedated for a brain MRI. Most of the neonatal

treatments, such as hypothermia, mechanical ventilation and

pressor support, can be continued in the MRI with special

precautions.

C A dedicated imaging protocol should be developed for scan-

ning the newborn brain. The typical imaging protocol should

include all the essential MRI sequences to evaluate brain

injuries in the newborn, including high spatial resolution T1-

and T2-weighted imaging, diffusion-weighted imaging (DWI)

and spectroscopy, as well as Fluid Attenuated Inversion

Recovery (FLAIR) imaging, MR angiography, MR venography

and susceptibility-weighted imaging (SWI) if required. It may

also include newer MRI adjuncts, such as diffusion-tensor

imaging (DTI) and diffusion tractography, functional MRI and

perfusion-weighted imaging.

C Brain MRI should be the gold standard for neonates who have

encephalopathy or suspected brain injuries in order to clearly

define the extent of brain injuries. It should be used to identify

the newborns, who are most at risk for subsequent neuro-

developmental disability and who may benefit from early

intervention services.

C Brain MRI should be used to improve our knowledge of these

brain injuries, to further refine our therapeutic strategies in

these patients and to evaluate the efficiency of such changes.


17 Limperopoulos C. Extreme prematurity, cerebellar injury, and autism.

Semin Pediatr Neurol 2010; 17: 25e9.

18 Lodygensky GA, Vasung L, Sizonenko SV, H€uppi PS. Neuroimaging of

cortical development and brain connectivity in human newborns and

animal models. J Anat 2010; 217: 418e28.

19 Malamateniou C, Adams ME, Srinivasan L, et al. The anatomic vari-

ations of the circle of Willis in preterm-at-term and term-born infants:

an MR angiography study at 3T. AJNR Am J Neuroradiol 2009; 30:

1955e62.

20 Mathur AM, Neil JJ, McKinstry RC, Inder TE. Transport, monitoring, and

successful brain MR imaging in unsedated neonates. Pediatr Radiol

2008; 38: 260e4.

21 Mathur AM, Neil JJ, Inder TE. Understanding brain injury and neuro-

developmental disabilities in the preterm infant: the evolving role

of advanced magnetic resonance imaging. Semin Perinatol 2010; 34:

57e66.

22 Miller SP, Ramaswamy V, Michelson D, et al. Patterns of brain injury in

term neonatal encephalopathy. J Pediatr 2005; 146: 453e60.

23 Miller SP, Ferriero DM, Leonard C, et al. Early brain injury in premature

newborns detected with magnetic resonance imaging is associated

with adverse early neurodevelopmental outcome. J Pediatr 2005;

147: 609e16.

24 Miller SP, McQuillen PS, Hamrick S, et al. Abnormal brain develop-

ment in newborns with congenital heart disease. N Engl J Med 2007;

357: 1928e38.

25 Mukherjee P, McKinstry RC. Diffusion tensor imaging and tractog-

raphy of human brain development. Neuroimaging Clin N Am 2006;

16: 19e43.

26 Owen M, Shevell M, Majnemer A, Limperopoulos C. Abnormal brain

structure and function in newborns with complex congenital heart

defects before open heart surgery: a review of the evidence. J Child

Neurol 2011; 26: 743e55.

27 Rutherford M, Biarge MM, Allsop J, Counsell S, Cowan F. MRI of

perinatal brain injury. Pediatr Radiol 2010; 40: 819e33.

28 Seghier ML, H€uppi PS. The role of functional magnetic resonance

imaging in the study of brain development, injury, and recovery in the

newborn. Semin Perinatol 2010; 34: 79e86.

29 Tong KA, Ashwal S, Obenaus A, Nickerson JP, Kido D, Haacke EM.

Susceptibility-weighted MR imaging: a review of clinical applications

in children. AJNR Am J Neuroradiol 2008; 29: 9e17.

30 Verboon-Maciolek MA, Groenendaal F, Hahn CD, et al. Human

parechovirus causes encephalitis with white matter injury in

neonates. Ann Neurol 2008; 64: 266e73.

31 Volpe JJ. Neonatal encephalitis and white matter injury: more than

just inflammation? Ann Neurol 2008; 64: 232e6.

32 Volpe JJ. Brain injury in premature infants: a complex amalgam of

destructive and developmental disturbances. Lancet Neurol 2009; 8:

110e24.

33 Volpe JJ. The encephalopathy of prematurityebrain injury and

impaired brain development inextricably intertwined. Semin Pediatr

Neurol 2009; 16: 167e78.

34 Woodward LJ, Anderson PJ, Austin NC, Howard K, Inder TE. Neonatal

MRI to predict neurodevelopmental outcomes in preterm infants.

N Engl J Med 2006; 355: 685e94.


35 Wintermark P, Labrecque M, Warfield SK, DeHart S, Hansen A. Can

induced hypothermia be assured during brain MRI in neonates with

hypoxic-ischemic encephalopathy? Pediatr Radiol 2010; 40: 1950e4.

36 Wintermark P, Hansen A, Gregas MC, et al. Brain perfusion in

asphyxiated newborns treated with therapeutic hypothermia. AJNR

Am J Neuroradiol in press.

37 Xu D, Vigneron D. Magnetic resonance spectroscopy imaging of the

newborn brain e a technical review. Semin Perinatol 2010; 34: 20e7.

Acknowledgements

The author thanks Aaron Johnstone and Therese Perreault for their

thorough review of the manuscript.



OCCASIONAL REVIEW

Cleft lip and palate: currentmanagementTim Goodacre

Marc C Swan

Figure 1 Left complete unilateral cleft of lip, alveolus, hard and soft

palate.

AbstractCleft lip and palate are the most common presenting congenital condi-

tions of the face and cranial bones. This article describes current under-

standing of the aetiology and presentation of the deformity and

management of the child from prenatal diagnosis until maturity. Principle

concerns include correction of the physical defect with the best possible

functional and cosmetic outcome, optimal speech correction, satisfactory

feeding and hearing, and dental and orthodontic health. The value of

comprehensive management of all aspects of care within a multidisci-

plinary team including clinical psychology support for child and family

is discussed.

Keywords alveolar bone graft; cleft lip; cleft palate; naso-alveolar

moulding; orthognathic; pharyngoplasty; postoperative emergencies;

submucous

Definitions

Cleft lip (CL) is defined as a congenital abnormality of the

primary palate (i.e. anterior to the incisive foramen). It may be

complete, incomplete or microform, unilateral or bilateral, and

may involve a palatal cleft (CL � P) (see Figure 1).

A cleft palate (CP) is a congenital abnormality of the

secondary palate and may be complete or incomplete, unilateral

or bilateral, or submucous.

CL � P is epidemiologically and aetiologically distinct from

isolated CP.

Epidemiology

The overall incidence of orofacial clefting is approximately one in

700 live births, amounting to approximately 1000 new cases per

annum in the UK. However, the incidence varies with ethnicity,

geography and the nature of the cleft itself.

In the context of CL � P, the incidence is approximately

0.3 per 1000 in African American populations, 1.0 per 1000 in

Caucasian populations, and 2.1 per 1000 in Japanese

Tim Goodacre BSc FRCS is a Senior Clinical Lecturer at the Nuffield

Department of Surgery, and a Consultant Plastic Surgeon, Spires Cleft

Centre, Children’s Hospital Oxford & Nuffield Department of Surgery,

Oxfordshire, UK. Conflict of interest: none.

Marc C Swan MRCS is a Specialist Registrar in Plastic & Reconstructive,

Surgery at the Oxford & Wessex Deanery, Spires Cleft Centre Children’s

Hospital Oxford & Nuffield Department of Surgery, Oxfordshire, UK.



populations. The incidence of isolated CP is racially homoge-

neous at approximately 0.5 per 1000 live births.

Unilateral clefts are nine times as common as bilateral clefts,

and occur twice as frequently on the left than the right. The ratio

of left:right:bilateral clefts is 6:3:1. Males are predominantly

affected by CL � P (M:F 2:1) whereas females are more

commonly affected by isolated CP.

Aetiology

Over 300 syndromes are associatedwith orofacial clefting andmost

occur as an isolated abnormalitye the so-called non-syndromic CL

� P. Isolated CP is more likely to be syndromic than CL � P.

The cause of isolated clefting is multifactorial involving

a complex influence of environmental and genetic factors. There

is a predisposition for familial clustering. In one Danish study the

concordance rate for CL � P was 60% in monozygotic twins and

10% in dizygotic twins. There is a doseeresponse relationship

between maternal periconception smoking and orofacial clefting.

Maternal alcohol consumption is also associated with an

increased risk of isolated CP. Other maternal risk factors include

diabetes, nutritional factors (e.g. vitamin A, folic acid), and

anticonvulsant medication.

Genetics

Inheritance may be chromosomal, Mendelian or sporadic

(Table 1).

With respect to non-syndromic clefts, the risk of unaffected

parents with one child with CL � P having a second affected

child is 4%, while with two affected children, this risk increases

to 9%. If one parent has a CL � P, the risk of having an affected

child is 4%, which increases to 17% for a second affected child.

A total of 35% of CL � P patients and 54% of isolated CP

patients are associated with another anomaly, although less than

3% of these is due to a single gene disorder.

Numerous ‘candidate’ genes/loci have been proposed on the

basis of linkage and/or association studies and include TGF-a

TGF-b-3 MSX-1 and IRF-6.

A recent longitudinal population based study from Norway

demonstrated that the risk of recurrence of an isolated cleft in

first degree relatives does not seem to be related to the

anatomical severity of the defect. Furthermore, the relative risk



Genetic associations with orofacial clefting

CL � P CP

Chromosomal Trisomy 13 or 21

Single gene Van der Woude (Chromosome 1, AD) EEC (ectrodactyly,

ectodermal hyperplasia and CL � P) Syndrome

(Chromosome 3, AD)

Treacher Collins Syndrome (Chromosome 5, AD) Stickler

Syndrome (Chromosome 12, AD) Velocardiofacial Syndrome

(Chromosome 22, AD) Opitz G/BBB Syndrome (AD)

Sporadic Pierre Robin Sequence

CL � P, cleft lip and palate; CP, cleft palate; AD, autosomal dominant.

Table 1

OCCASIONAL REVIEW

of cleft recurrence in first degree relatives was 32 for any cleft lip

and 56 for isolated cleft palate e thus indicating that genetics

contribute more to cleft palate alone than to cleft lip. There was

a low (three-fold) crossover risk between the incidence of cleft lip

and isolated cleft palate in families, which implies that genes

such as MSX-1 and IRF-6 may participate in all forms of oral

clefting. Several major groups are now investigating the genetic

basis of clefting with genome wide association studies alongside

single family investigation and tissue bank establishment.

However, it is unlikely that this work will influence the

management of the majority of cases for more than a decade.

Antenatal diagnosis

Since first reported in the prenatal diagnosis of facial clefting,

most centres performing 20-week foetal anomaly ultrasound

scanning now include observation of the facial elements as

a routine. Detection of cleft lip and alveolus (gum) is around

70% cases in the best series, but the sensitivity is generally in the

order of 20%, although there is high specificity. Missed cases will

inevitably occur due to foetal movement and adverse position

during the scan. Isolated CP is particularly difficult to diagnose

on account of the acoustic shadow created by the facial bones.

Currently up to 25% of cleft lips (with or without CP) are diag-

nosed antenatally. The addition of three-dimensional (and now

four-dimensional) ultrasound methods gives better quality

a b

Figure 2 Intra-uterine magnetic resonance imaging scan (sagittal view) demon

(red arrow) in (b) is a pathognomic sign of cleft palate.


pictures for parents’ benefit, but do not significantly improve the

ability to predict foetal palate status. To date, only foetal

magnetic resonance imaging offers a realistic means of predicting

important additional information about the palate, which has

a bearing upon the future child’s feeding, speech, and facial

growth capacity (Figure 2).

Prenatal diagnosis has been described as a ‘mixed blessing’.

Psychological studies of parents indicate the appreciation of

preparatory knowledge, but an increased anxiety level during the

remaining pregnancy. Diagnosis, therefore, carries with it

a considerable obligation for parental support and counselling e

now offered routinely by most of the newly configured cleft

teams. Training for ultrasonographers involved in the first

moments of detection has been shown to be beneficial. Outcome

following prenatal diagnosis of clefting across UK maternity units

is unknown, but carefully organized support for parents has

avoided the high levels of termination of pregnancy for isolated

clefting that have been reported elsewhere.

Cleft types

Most clefts fall within the fusion lines of the fronto-nasal process

and lateral maxillary elements and the midline of the palatal

shelves in the mouth once posterior to the incisive foramen.

Clefts in other lines are rare and were classified by Paul Tessier

as craniofacial clefts no. 0e14. They are not the subject of this

strating normal (a) and cleft (b) palates. The absent palatal stripe



OCCASIONAL REVIEW

review, and are almost always best managed by referral to

a specialist paediatric craniofacial team.

‘Typical’ clefts may involve all or part of one or both lip philtral

columns, alveolar (gum) bone, hard palate or soft palate. The cleft

can also be complete, incomplete, or a forme fruste involving

muscle dehiscence only. In the lip, these latter ‘near misses’ can

produce asymmetry of smile and nasal shape and thus are of

cosmetic importance. In the soft palate, a forme fruste is of even

greater importance, presenting as a submucous cleft palate.

The approximate distribution of the major cleft types is as

follows:

� cleft lip and palate 46%

� isolated cleft palate 33%

� isolated cleft lip 22%

A total of 86% of bilateral cleft lips and 68% of unilateral cleft

lips are associated with a cleft palate deformity.

Submucous cleft palate

Submucous cleft palate (SMCP) is sufficiently important to merit

further mention. It may present ‘overtly’ with a classical triad of

signs:

� notched hard palate posterior margin (however small)

� bifid uvula

� lucency of midline of palate (the ‘zona pellucida’ e due to

muscular diastasis) (Figure 3).

Even the most marginal of hard palate notches is a hard sign

of possible SMCP, in contrast to the bifidity of the uvula, which is

common in those with no other signs of muscle dehiscence, and

may be of no clinical consequence.

‘Occult’ presentation of SMCP presents with the speech and

swallowing difficulties of SMCP, but none of the classical triad of

signs. It can be confirmed by observing a characteristic ‘grooved’

surface on the dorsum of the soft palate during nasendoscopy.

SMCP presents a management conundrum. Most e detected

by neonatal examination or as a consequence of early feeding

difficulties e will progress to severe speech dysfunction if left

untreated. For these children, the only effective treatment is

surgical muscle repositioning, with subsequent specialist speech

and language therapy together with surgical pharyngoplasty if

required. However, a certain number may develop perfectly

normal speech and feeding, suggesting that careful monitoring of

early babbling patterns and speech development by highly

Figure 3 The midline zona pellucida of a submucous cleft palate.


specialized speech and language therapists is an acceptable early

management plan for these cases.

Diagnosis of SMCP is often delayed. Awareness of the pre-

senting features should enable detection during all neonatal

screening examinations if the mouth is examined correctly.

Casual slipping of a finger into the mouth is not adequate.

Suspicion should always be raised by neonates who fail to suck

with good pressure, and in older toddlers whose speech develops

with cleft type characteristics (see Speech section).

Bilateral clefts

Bilateral cleft lip presents a difficult problem when associated

with alveolar þ/� more posterior clefting. The bony element

(the ‘premaxilla’) may be unrestrained by the normal ring of lip

muscle, and protrude in a much distorted and upwardly rotated

position (Figure 4). Pre-surgical orthopaedics (using a dental

plate, and sometimes lip strapping) is often helpful in such cases

and may make the primary surgery a much easier (and, there-

fore, successful) procedure.

Early care

Pre-surgical orthopaedics and naso-alveolar moulding

There is a long history of the use of dental devices to assist cleft lip

and palate management. Prospective randomized trial data now

shows that there is no benefit to infant feeding with the use of such

treatment. The term ‘feeding plate’ is, therefore, now defunct.

Moulding of the dental arch form with orthopaedic devices is

more controversial, and the subject of a large multicentre trial

still accruing data. Contradictory data exist for whether such

bony manipulation improves outcome, affects growth, or makes

surgery more straightforward. Improved surgical ability is likely

to be the most consistent and valuable effect, especially if it

produces more satisfactory cosmetic outcomes.

A variety of appliances have been described, which if ‘active’

may contain springs or other mechanisms to gently oppose the

gum ridges. No UK unit currently uses the more severe Latham

device, which requires pin fixation to the jaw arch line.

Naso-alveolar moulding (NAM) involves adding an extension

to the orthopaedic device, to exert moulding pressure on the

distorted nasal margin. The principle is similar to that espoused

for ear cartilage moulding in early months (‘ear buddies’). It

requires much additional work (and cost) from the orthodontic

team, and adds considerably to the burden of care for the new

parents. No UK unit currently offers this service routinely, but

the best series in the United States have impressive results, which

would be expected to lead to better cosmetic outcomes. It is also

becoming an established method in China and south east Asia,

and is likely to become the subject of careful benefit analysis in

the UK in the near future.

Postoperative nasal splints

Along similar lines, post lip repair nostril splinting (using

conformers) is thought by many to be beneficial. Use of

progressively enlarging conformers to shape the nostril aperture

and lift the slumped rim has a good evidence base, but again

requires considerable commitment from the parents in conjunc-

tion with good nurse specialist support. The use of such splints is

routine now across south east Asia.



Figure 4 Pre-operative views of bilateral cleft lip and palate (a) and (b) with post-operative image (c).

OCCASIONAL REVIEW

Specialist nursing care

One of the indisputable advances in UK cleft care over the past 15

years has been the widespread development of highly skilled

nurse specialists to offer support to expectant parents, peri-natal

care, complex feeding advice, home visits and peri-operative

care. The success of such specialists has improved continuity

of care from the more centralized teams, and raised standards

from previously ad hoc support structures. Arguably, this

development has had a far greater impact on outcomes than any

specific surgical methodological change.

Pierre Robin sequence

The sequence of cleft palate associated with micrognathia, glos-

soptosis, and respiratory difficulty, was described by Robin in

1923. The incidence is approximately one in 14,000 births. Not

indicative of a specific syndrome, the spectrum of severity ranges

greatly. The most severe forms exhibit wide clefts with gross

maxillary shelf and muscle hypoplasia, the tongue prolapsing as


a ‘ball valve’ into the posterior nasopharynx, a minute jaw, and

major problems with airway maintenance. Management of Pierre

Robin sequence is concentrated on establishing a secure airway

at all times and satisfactory oral feeding (Figure 5).

Current best practice is invariably to use a nasopharyngeal

airway for the mainstay of airway protection. The widespread

adoption of this method has obviated the need for more invasive

techniques of the past, such as glossopexy (fixation of the tongue

tip to the lip/jaw) or the Burston frame (a prone positioning

frame to allow forward head projection). Very early surgical

distraction osteogenesis of the hypoplastic mandibular arch has

been advocated by several current authorities in the United

States. However, the method has found few devotees elsewhere,

where surgical enthusiasm is more reasonably balanced by wise,

less invasive, medical support. The most cogent argument for

such mandibular distraction is to reduce the period of time that

a child might require a tracheostomy for the most extreme severe

cases of Pierre Robin sequence.



a

b

Figure 5 Child with Pierre Robin sequence a with intra-oral view of wide

cleft palate (b). The child is successfully managed at home with a naso-

pharyngeal airway and nasogastric feeding tube in situ.

OCCASIONAL REVIEW

Feeding support for such children can be problematic.

Nasogastric supplementation may be required, but every effort is

made to prevent the child becoming overly dependent upon

nasogastric feeds in preference to using normal oral sucking.

Postoperative emergencies

Primary surgery: The most commonly encountered problems

after primary cleft surgery are airway obstruction and bleeding.

Airway obstruction usually follows narrowing of the nostril

apertures in lip/anterior palate closure, in the child who is still an

obligate nasal breather. Use of the nostril conforming splint,

together with perhaps a nasogastric tube, can further obstruct the

nasal airway. Relief of such obstruction is usually easy with gentle

exterior suction and use of a nasopharyngeal airway if required.


Airway obstruction following posterior palate repair can be

more difficult. If anticipated, a nasopharyngeal airway can be left

in situ at the end of the procedure, together with placement of

a temporary tongue suture to assist with positioning. If a naso-

pharyngeal airway requires placement on the ward post-

operatively, it should be passed with the utmost care to avoid

suture line disruption with inevitable additional bleeding and

functional consequences for the repair.

Significant postoperative bleeding is a surgical emergency. Some

bleeding in thefirst 12h is expected, although the surgeonshouldbe

alerted if it is fresh or brisk. Later bleeding can be reactionary or

secondary in nature. Either way, the child will require some seda-

tion (usuallymorphine 0.1mg/kg), and a topical adrenaline soaked

gauze swab is used to apply pressure onto the roof of the mouth.

The most common bleeding site is one of the lateral releasing inci-

sions used to close the cleft palate and such digital pressure can

establish control remarkably quickly. Unnecessary staff and/or

parents should be removed from the roomand very great care given

to any suction (avoided if at all possible). The child should be

prepared for urgent return to theatre, although often ward control

obviates the need for any more active surgical intervention.

Secondary surgery: Secondary procedures include phar-

yngoplasty, alveolar bone grafting and osteotomyeorthognathic

surgery. Bleeding is the most common later emergency. It should

be managed with all usual supportive measures (ABC, intrave-

nous fluids and cross matching if appropriate) and early return to

theatre. Airway obstruction can frequently follow the less phys-

iological forms of pharyngoplasty.

Current treatment protocols

Timing of cleft repair

The late 20th century UK-based controversy surrounding the value

of neonatal cleft lip repair has now almost disappeared; the

purported benefit of improved lip scarring from foetal wound

healing patterns is disproven.Work on the impact of early repair as

opposed to more conventional timing of lip closure at 3 months is

now beginning to improve understanding of early neural

development. Further work to investigate the nature of early

mothereinfant interactions and the disruption caused by facial

disfigurement is in progress and may influence surgical timing in

the coming decade. Most UK centres (and similarly almost all

world centres) now undertake first surgery once feeding patterns

have been established and birth weight regained.

The main difference in timing protocol in major centres is

found in the sequence in which the lip and palatal elements are

operated upon. The more frequent pattern used is lip and ante-

rior palate as a primary procedure around 2e3 months, with soft

palate closure once the airway is more secure e from 4 to 12

months. The opposing view (common in France e sometimes

termed the ‘Malek sequence’) aims to avoid any early surgical

interference with the hard palate growth centres, and repairs the

lip þ/� soft palate at around 2e3 months, followed by delayed

hard palate closure at times varying from 6 to 60 months. Those

centres adopting palate repairs later than 12 months frequently

use hard palate cover plates to reduce abnormal airflow and

permit better speech development than would occur in the

presence of an oronasal fistula.



OCCASIONAL REVIEW

The dilemma of the mutually opposing benefits of early and

later hard palate repair on palatal growth versus speech devel-

opment remains one of the most controversial and difficult

aspects of cleft management. Robust evidence accounting

adequately for all variables and cleft types is lacking, and many

opinions accept that it is the quality of overall surgical tissue

handling rather than defined technique or timing that has most

influence on long-term outcome for growth and speech.

Cleft lip repair

Lip repairs adopt a form of lengthening of the greater segment

margin, the rotation advancement being the most popular. Almost

all authorities now include some form of primary nasal tip correc-

tion in the primary lip repair, results usually improving upon the

status quo if nothing is done. Radical muscle repositioning is also

widely adopted, and many surgeons now use a subperiosteal

dissection of the muscle away from the maxilla, in order to mini-

mize deep seated scar tissue and offer the potential to generatemore

bone from theunder surface of the periosteum. Subtle corrections of

the lip scarwith the interposition of small flaps above thewhite roll,

and within the dry vermilion mucosa as well as the avoidance of

cuts around the base of the lateral nostril margin are all advances

that improve long-term outcome (Figure 6).

Cleft palate repair

The most significant advance in palate repair over the past

15 years has been the widespread adoption of the radical levator

palate muscle repositioning procedure described by Brian Som-

merlad. The outcome of this procedure appears to have no

adverse growth effect and produces the very best speech

a

Figure 6 Pre-operative (a) and post-operative (b) appearan


capability with lowest incidence of velopharyngeal incompe-

tence. It also has potential benefits on Eustachian tube function.

The technique involves closure of the nasal mucosa, followed by

transposition of the medial insertion of the levator muscles by

90� so that the two mobilized ends can be sutured together to

form a new, extensible muscle ‘sling’, which is capable of velo-

pharyngeal closure (Figure 7).

An alternative procedure adopted widely around the world is

the Furlow palatoplasty, with double opposing Z-plasties to

facilitate the muscle correction as espoused by Sommerlad.

The remaining development in palate surgery is the trend

towards minimizing lateral releasing incisions (as with the Von

Langenbeck procedure, now in use for over 100 years) by radical

undermining of the palatal flaps. Self-inflating tissue expanders

might hold some improvement in this respect in the coming 10years.

Foetal surgery

The advent of improved antenatal diagnosis of intrauterine

pathology has made foetal surgery a feasible option, however the

standard ‘open’ techniques are associated with significant

morbidity and mortality. Thus, the development of less invasive

feto-endoscopic techniques appears encouraging, and has been

demonstrated to be effective in vivo using cleft animalmodels. The

major advantage is the ‘holy grail’ of scarless wound healing,

which has been reported at mid-gestation and would have clear

functional and aesthetic advantages. Consensus criteria exist as to

which congenital malformations are considered appropriate for

intrauterine surgery. At present such surgery is purely experi-

mental with respect to orofacial clefting and there is little prospect

of clinical trials commencing in the foreseeable future.

b

ces of a left unilateral cleft lip and palate.



a b

Figure 7 Sommerlad radical muscle-positioning technique for soft palate repair: pre-operative view (a) and intra-operative view (b). (Reproduced with

permission from Plastic & Reconstructive Surgery).

OCCASIONAL REVIEW

Speech

Cleft palate (overt or submucous) carries an inevitable potential

for speech to develop abnormally due to the position of the soft

palate musculature. Of the five known soft palate muscles, the

palatoglossus and palatopharyngeus are principally involved in

swallowing. The tensor and levator veli palatini muscles are

speech motors, acting to extend and lift the soft palate in a ‘knee’

shaped valvular action, which closes the posterior nasopharynx

from the oropharynx. This action is essential if air pressure is to

be raised in the mouth e a necessary component of normal

speech in most languages.

Raised oral pressure is needed particularly for fricative sounds

(such as ‘s’, ‘f ’, ‘sh’) and plosives (‘p’, ‘m’, ‘b’). Failure of this

action results in the speech pattern described as velopharyngeal

incompetence (VPI) the component parts of which are hyper-

nasality, nasal emission and nasal turbulence or resonance.

When young children have incompetent palatal musculature,

somewill develop very substantial compensatorymisarticulations

in order to attempt normal speech. These can lead to severely

compromised intelligibility, and in some places have resulted in

erroneous association with learning disorders.

Recent evidence supports the adoption of therapeutic inter-

vention at an early stage (the babbling phase) to counter adverse

speech development. Once normal speech is developing, it is

essential that specialized speech and language therapists are

involved to monitor and guide speech development. Significant

abnormalities in palatal function can then be identified at an early

stage, and investigated to ascertain whether secondary speech

surgery will offer any benefit to the child.


Audiology

Most children with cleft palate would develop glue ear without

intervention. This relates to the abnormal positioning of the

stylopharyngeus muscle. Some authorities, therefore, advocate

early grommet insertion by way of prophylaxis for this condition,

to improve hearing, and to prevent chronic secretory otitis media

and worse (e.g. cholesteatoma). However, there is no consensus

for early middle ear management for cleft children, other than

agreement that careful audiological prolonged assessment is

mandatory, with appropriate intervention (either grommet

insertion or use of hearing aids) as required. It is possible that the

newer, more radical, palate muscle repositioning techniques will

lead to ‘normalization’ of middle ear Eustachian ventilation and

reduce the need for external ventilating grommets. However, no

published evidence currently exists to support this view.

Secondary speech surgery

Although primary cleft palate repair techniques have improved

early outcomes, a substantial number of children will remain with

deficient palatal function despite optimal speech therapy. Those

demonstrated to have VPI are considered for either palatal muscle

re-repair (along the lines described by Sommerlad for primary

repair) or pharyngoplasty (an operation performed on the pharyn-

geal wall in order to improve closure of the velopharyngeal orifice).

Pharyngeal flap

The oldest, and to some extent most consistently reliable, means

of improving velopharyngeal closure is by elevating a myomu-

cosal flap from the posterior pharyngeal wall and attaching it to



OCCASIONAL REVIEW

the posterior soft palate. The static flap acts as a broad ‘wafer’ of

tissue tethering the palate and also obstructing the widest portion

of the airway.

The pharyngeal flap e of which there are numerous anatomic

variations e is also the most obstructive, and can produce

significant sleep apnoea as well as other adverse sequelae, such

as severe snoring, mucous obstruction, and long-term airflow

obstruction (implicated rarely in right heart failure).

Pharyngoplasty procedures

In an attempt to reduce such symptoms, the use of lateral

pharyngeal wall or tonsillar pillar tissue to constrict the

nasopharyngeal valve was developed. Lateral pharyngeal wall

flaps set high in the adenoidal area act as a form of ‘speed

bump’ to bring the posterior pharyngeal wall closer to the soft

palate and assist closure of the valve (the ‘Hynes’ phar-

yngoplasty procedure). Posterior tonsillar pillar flaps, set lower,

act as a form of dynamic sphincter, and tend to be somewhat

more obstructive to airflow (the ‘Orticochoea’ pharyngoplasty

procedure).

Posterior pharyngeal wall augmentation

a

In an attempt to avoid all adverse obstructive consequences of

pharyngoplasty, some authorities create the ‘speed bump’ effect

on the posterior pharyngeal wall entirely by placing a piece of

material (cartilage or alloplastic) beneath the pharyngeal

mucosa.

Such secondary speech procedures are now only performed

with full pre- and postoperative support by a specialized

speech and language therapy team. Additional, non-surgical,

methods for the infrequent severe and resistant cases include

the use of prosthetic devices such as ‘speech bulbs’ attached to

permanently worn dental plates, and biofeedback therapy

techniques.

Alveolar bone grafting

b

Figure 8 Secondary bone grafting of the cleft alveolus: pre-operative

(a) and immediate post-operative (b) views.

Primary cleft repair does not correct the bony deformity in the

gum ridge, although recent work on primary gingivoper-

iosteoplasty is an attempt to address this. Since the 1970s, it has

been understood that the bone defect is best filled with cancel-

lous (marrow) bone graft in the secondary dentition phase,

before the eruption of the secondary canine tooth, which usually

lies adjacent to the cleft gum (Figure 8).

The timing of this procedure is determined by the specialist

orthodontist and relates to dental maturity. It is usually between

the ages of 7 and 11 years and involves reopening the cleft bony

line and packing the mucosally lined cavity with chips of graft

harvested from the iliac crest or tibial plateau. This procedure

also offers the best opportunity to close any residual oronasal

fistula in the anterior palatal region. It is a highly effective

operation, and enables the subsequent orthodontic management

of what are frequently much distorted teeth into a well-corrected

arch form.

The necessity for such bone grafting is well established.

Careful paediatric dental health is a mandatory part of early cleft

care, and later specialist restorative dental expertise enables all

but the most resistant cases to obtain a high level of dental health

and appearance.


Orthognathic surgery

Despite the best outcomes available, a certain proportion of

children with complete cleft lip and palate will develop

secondary maxillary hypoplasia following primary repair. The

characteristic ‘dish face’ appearance is often the subject of severe

teasing and self-consciousness, and is best addressed by

a combination of psychological support with orthognathic (jaw

moving) surgical correction in the late teenage years.

Orthognathic surgery involves moving the maxilla forwards by

means of controlled bone cuts (‘osteotomies’) in patterns

described by Ren�e Le Fort in the 19th century. Maxillary

advancement may be combined with surgical correction of the

mandible by sliding it backwards again using bone cuts. Recent

use of bone distraction osteogenesis has improved the long-term

outcome of such procedures. Orthognathic procedures are

amongst themost effective in thewhole gamut of cleft surgery, and

should be considered even in less severe cases of hypoplasia if the

surgery can be undertaken by a skilled maxillofacial surgeon.

The cause of mid-facial growth failure remains controversial.

The adverse effect of primary surgical intervention is indisput-

able (as shown by careful analysis of adult unrepaired clefts in

Sri Lanka over many years) but the relative impact of various

forms of primary management is still unclear. It would appear

that, despite the best efforts to minimize adverse sequelae of

surgery, some children are ‘poor growers’ with inherent hypo-

plasia, and are destined to require significant secondary surgery

regardless of their primary treatment protocol.



OCCASIONAL REVIEW

Clinical psychology

Practice points

Throughout this description of interventional medical care for

cleft lip and palate children, it will be evident that the goal of all

care should be a well-rounded and healthy child able to achieve

his/her full potential in life. They should be able to enjoy

a normal childhood as little disrupted by treatment or adverse

consequences of the cleft as possible.

The input from skilled specialist psychologists to cleft manage-

ment from prenatal diagnosis to adulthood is invaluable, and

essential as a guide to achieve holistic care from all teammembers.

Since the Clinical Standards Advisory Group (CSAG) report and

reorganization, a full psychology service has become a mandatory

part of the team structure. It is probable that the number of

unnecessary or ill-advised secondary procedures is reduced by such

input, as well as demonstrable improvements in individual and

family wellbeing and dynamics. Future developments in the area of

psychological input will focus on the potential value of early inter-

vention in families at high risk of adverse psychological conse-

quences when older. Much evidence is accumulating of the

particular value of very early (before 12 weeks) psychological

support, with a possible ‘sensitive period’ in development influ-

encing subsequent cognitive ability becoming clear.

C Cleft lip and palate are the most commonly encountered
CSAG
anomalies of the craniofacial region

C The genetic basis of non-syndromic clefting is complex and

poorly understood. Environmental factors are involved in

a proportion of cases

C Antenatal diagnosis of lip clefts can be expected in about two-

thirds of cases. Palatal clefts cannot be diagnosed before birth

other than with magnetic resonance scanning. Prenatal diagnosis

carries with it a responsibility for rapid access to specialist coun-

selling and advice. Outcome following such support is very good

C The spectrum of cleft types is considerable. Incomplete and

forme fruste conditions include submucous clefting of the

palate, which is frequently missed in postnatal examinations

C The triad of notched posterior hard palate, bifid uvula, and

midline zona pellucida are diagnostic of submucous cleft

palate, which should be referred for specialist opinion

C The role of pre-surgical orthopaedics to improve surgical repair

of wider cleft lips remains controversial. It has no beneficial

effect on feeding, but many surgeons value the outcome

C There is little (if any) role for surgical intervention in early severe

Pierre Robin sequence. Nasopharyngeal airway management is

highly successful in support during the early months

C Cleft care has been centralized into less than 10 units in the UK.

This has enabled coordinated nursing, surgical, and other

specialist service provision, with a rising of standards since

instigated. All newly diagnosed or suspected clefts and related

conditions should be referred to these teamsas early as possible

C Optimal surgical technique remains elusive. However, radical

palatal muscle repositioning has improved speech outcomes

considerably. Mid-facial growth disturbance remains the most

common long-term adverse outcome of surgical intervention

C Comprehensive and ‘holistic’ cleft care involves mandatory

clinical psychology support of children and their families.

Unnecessary surgical interventions can be reduced, and long-

term global health outcomes are improved

Perhaps the single most effective change in cleft care standards in

the UK has been brought about by reorganization and service

centralization stimulated by the 1996 report from the CSAG e

now a defunct organization. Service provision was reduced from

57 centres in the early 1990s to nine centres (some ‘twin site’) in

the UK from 2006. This centralization has been accompanied by

more careful financial investment by regional commissioners,

and has brought UK clinical outcomes to a level that is arguably

among the best in the world. A

FURTHER READING

Ardinger HH, Buetow KH, Bell GI, Bardach J, VanDemark DR, Murray JC.

Association of genetic variation of the transforming growth factoralpha

gene with cleft lip and palate. Am J Hum Genet 1989; 45: 348e53.

Boyne PJ, Sands NR. Secondary bone grafting of residual alveolar and

palatal clefts. J Oral Surg 1972; 30: 87e92.

Calnan J. Submucous cleft palate. Br J Plast Surg 1954; 6: 264e82.

Curtis E, Fraser FC, Warburton D. Congenital cleft lip and palate: risk

factors for counseling. Am J Dis Child 1961; 102: 853e7.

Descamps MJL, Golding S, Sibley J, McIntyre A, Alvey C, Goodacre T. MRI

for definitive in utero diagnosis of cleft palate: a useful adjunct to

antenatal care. Cleft Palate Craniofac J 2010; 47: 578e85.

Jugessur A, Murray JC. Orofacial clefting: recent insights into a complex

trait. Curr Opin Genet Dev 2005; 15: 270e8.

Kobus KF. Cleft palate repair with the use of osmotic expanders:

a preliminary report. J Plast Reconstr Aesthet Surg 2007; 60: 414e21.

Lees M. Familial risks of oral clefts. BMJ 2008; 336: 399.

Masarei AG,WadeA,MarsM, SommerladBC, Sell D.A randomized control trial

investigating the effect of presurgical orthopedics on feeding in infants

with cleft lip and/or palate. Cleft Palate Craniofac J 2007; 44: 182e93.

Pfeifer TM, Grayson BH, Cutting CB. Nasoalveolar molding and gingivo-

periosteoplasty versus alveolar bone graft: an outcome analysis of

costs in the treatment of unilateral cleft alveolus. Cleft Palate Cra-

niofac J 2002; 39: 26e9.


Rice DPC. Craniofacial anomalies: from development to molecular path-

ogenesis. Curr Mol Med 2005; 5: 699e722.

Rollnick BR, Pruzansky S. Genetic services at a center for craniofacial

anomalies. Cleft Palate J 1981; 18: 304e13.

Sivertsen A, Wilcox AJ, Skjaerven R, et al. Familial risk of oral clefts by

morphological type and severity: population based cohort study of

first degree relatives. BMJ 2008; 336: 432e4.

Sommerlad BC, Fenn C, Harland K, et al. Submucous cleft palate:

a grading system and review of 40 consecutive submucous cleft palate

repairs. Cleft Palate Craniofac J 2004; 41: 114e23.

Sommerlad BC. A technique for cleft palate repair. Plast Reconstr Surg

2003; 112: 1542e8.

Tessier P. Anatomical classification facial, cranio-facial and laterofacial

clefts. J Maxillofac Surg 1976; 4: 69e92.

Yazdy MM, Honein MA, Rasmussen SA, Frias JL. Priorities for future public

health research inorofacial clefts.Cleft Palate Craniofac J2007;44:351e7.

Zucchero TM, Cooper ME, Maher BS, et al. Interferon regulatory factor 6

(IRF6) gene variants and the risk of isolated cleft lip or palate. N Engl J

Med 2004; 351: 769e80.



PERSONAL PRACTICE

ConjugatedhyperbilirubinaemiaJulie Brent

Mansoor Ahmed

What is conjugated hyperbilirubinaemia?

Neonatal jaundice is common in the first 1e2 weeks after birth

and usually resolves spontaneously. In the majority of cases, it is

physiological. Prolonged neonatal jaundice is jaundice lasting for

more than 2 weeks (more than 3 weeks for preterm babies). In an

otherwise healthy and asymptomatic breast fed neonate, this

may reflect breast milk jaundice (unconjugated hyper-

bilirubinaemia). A direct or conjugated bilirubin of more than 1.0

mg/dL if total bilirubin is less than 5 mg/dL or more than 20% if

total bilirubin is more than 5 mg/dL is considered as abnormal

(pathological) at any time and constitutes conjugated hyper-

bilirubinaemia. This personal practice review focuses on the

initial diagnostic approach and management of conjugated

hyperbilirubinaemia during the first few weeks after birth.

How common is conjugated hyperbilirubinaemia?

Neonatal cholestasis effects one in 2500 live births. Due to its

relatively low incidence, it is infrequently seen by most providers

of medical care to infants. Extra hepatic biliary atresia is a rare

disorder with an incidence of approximately 1:15,000 live births

and comprises of approximately 1/3 cases of neonatal

cholestasis. Alpha 1 antitrypsin deficiency is the cause in 5e15%

and other inherited forms of cholestasis occur in 10e20% of

cases. Inborn errors of metabolism and congenital infection

(TORCH) cause 20% and 5% cases respectively. Incidence of

idiopathic neonatal hepatitis has decreased significantly (now

stands at 10e15%) due to improved diagnostic methodologies

and better understanding of genetics of bile acid metabolism.

Bilirubin metabolism and pathophysiology

Conjugated bilirubin accumulates in the blood when there is

impaired bile formation by the hepatocytes or from obstruction of

bile flow through intra or extra hepatic biliary tree (cholestasis).

This leads of accumulation of biliary substances (bilirubin, bile

acids and cholesterol) in the liver, blood and extra hepatic tissues.

The pathway for bilirubin metabolism is shown in Figure 1.

In the liver, water insoluble unconjugated bilirubin is taken

up by hepatocytes at the sinusoidal membrane and conjugated to

Julie Brent MBBS MRCPCH is an ST-5 Paediatrics, at Queen’s Hospital,

Burton Upon Trent UK. Conflicts of interest: None.

Mansoor Ahmed MBBS FRCP FRCPCH is Consultant Paediatrician at Queen’s

Hospital, Burton Upon Trent, UK. Conflicts of interest: None.


become water soluble. Bile production depends upon an active

transport of bile acids (and other substances) into the biliary

canaliculi. Major transporters at the basolateral membrane are

Naþ Taurocholate Cotransporting Polypeptide (NTCP) and

Organic Anion Transporting Proteins (OATP). Bile secretion at

the canalicular membrane is facilitated by Bile Salt Export Pump

(BSEP) and Multidrug Resistant Proteins (MRP2). The role of

these transporters in cholestatic diseases is being increasingly

recognized.

Causes of conjugated hyperbilirubinaemia

There are numerous Intrahepatic and extra hepatic causes of

neonatal conjugated hyperbilirubinaemia (Table 1).

Common causes include biliary atresia, inherited/metabolic

forms of cholestasis idiopathic neonatal cholestasis, and the

multifactorial cholestasis seen in ex-preterm infants requiring

total parenteral nutrition and/or neonatal surgery.

What are the key presenting features?

Neonatal conjugated hyperbilirubinaemia usually presents as

prolonged jaundice in a well infant. Pale acholic stools (cardinal

feature of cholestasis) and dark urine are important pointers in

history but not always recognized by parents. In an unwell

neonate, presentation may include bleeding due to coagulopathy

unresponsive to vitamin K. Conjugated hyperbilirubinaemia may

also present as a feature of systemic conditions in a more unwell

neonate who may have sepsis, shock, seizure, irritability, heart

failure, hypopituitarism or metabolic disorder (such as gal-

actosaemia or tyrosinaemia). Important features in the history

include obstetric history particularly looking for intrauterine

infections or cholestasis of pregnancy, consanguinity, family

history of cholestasis or liver disease, early neonatal course

including resuscitation at birth, neonatal intensive care admis-

sion and details of parenteral/enteral nutrition.

Focussed physical examination

Anthropometric assessment (including weight, length and head

circumference measurement) should be undertaken. In intra-

uterine infections, intrauterine growth restriction and skin rash

may be seen. Stigma of syndromic disorders with congenital

anomalies, facial dysmorphic features, oedema, ascites and

evidence of congenital heart disease should also be carefully

looked for. Abdominal examination often reveals hepatomegaly

and less commonly splenomegaly. A choledochal cyst may be felt

as a mass in the right upper quadrant of abdomen. Formal ocular

assessment (posterior embryotoxin in Alagille syndrome, optic

nerve hypoplasia in panhypopituitarism, chorioretinitis in

congenital infections and cataract in congenital infections/gal-

actosaemia) and cardiac evaluation may also be required.

Diagnostic workup

All babies with prolonged jaundice should have a split bilirubin

(total and conjugated bilirubin) measurement taken. All babies

with conjugated hyperbilirubinaemia should be promptly

referred to a paediatrician for initial investigations.



Haemoglobin

Haem

Unconjugated Bilirubin

Conjugated Bilirubinbilirubin diglucuronide

(Water soluble)

Globin

Uridine diphosphoglucuronicacid glucuronyl transferase

Bile

Bile salts

Colonic bacteria

Urobilinogen

Reabsorbed

Urobilin

Stercobilinogen

Stercobilin

Old red blood cells breakdown

spleen

Albumin

Hepatocytes

Kidneyy

Bilirubin-albumin complex(Water insoluble)

Figure 1 Schematic diagram of bilirubin pathway.

PERSONAL PRACTICE

Initial investigations

Taking into account wide range of differential diagnosis (Table 1),

a structured approach to investigate a neonate with cholestasis

should identify conditions or complications (such as coagulop-

athy, neoplastic disorders, acute liver failure, hypoglycaemia,

sepsis, metabolic disorders like galactosaemia and pan-

hypopituitarism) requiring immediate treatment. Once these

disorders have been excluded, the most important differential is to

look for biliary atresia. A paediatric hepatologist should be con-

sulted early as prompt referral for surgery to manage biliary atresia

before 60 days improves prognosis.

Initial investigations in neonates with conjugated hyper-

bilirubinaemia are listed in Table 2. The serum transaminases

(ALT, AST) are sensitive markers for hepatocellular injury but

are non-specific. Alkaline phosphatase is also non-specific


(found in liver, bone and kidney) and is likely to be raised in

biliary obstruction. Gamma glutamyl transpeptidase (GGT), an

enzyme in bile duct epithelium, is a sensitive marker of biliary

obstruction and is raised in most cholestatic disorders. However,

it may be normal or low in progressive familial intrahepatic

cholestasis and disorders of bile acid metabolism. Alpha 1 anti-

trypsin deficiency can be difficult to distinguish from extra

hepatic biliary atresia on clinical and histological findings. Both

alpha 1 antitrypsin level as well as phenotype should be assessed

(serum levels may be falsely normal or raised, as it is an acute

phase protein). An abdominal ultrasound is useful in identifying

choledochal cysts, gall stones, sludge in the biliary tree or gall-

bladder. A small or absent gallbladder is suggestive but not

diagnostic of biliary atresia and the triangular cord sign (echo-

genic area at porta hepatis) is thought to be specific for biliary



Causes of neonatal conjugated hyperbilirubinaemia

Extra hepatic

Hepatic bile duct

anomalies

Biliary atresia, choledochal cyst, cholelithiasis, inspissated bile secretion, spontaneous perforation of the

bile duct, neonatal sclerosing cholangitis

Intrahepatic

Intrahepatic bile

duct anomalies

Intrahepatic biliary hypoplasia (Alagille’s syndrome), caroli’s disease (dilated intrahepatic bile ducts),

congenital hepatic fibrosis

Idiopathic neonatal hepatitis

Infections Viral (TORCH, HIV, echovirus, adenovirus, coxsackie virus, human herpes virus-6, hepatitis B & C,

parvovirus B19, varicella zoster), Bacterial (sepsis, urinary tract infection, syphilis, tuberculosis, listeriosis)

Parasitic (toxoplasmosis, malaria)

Metabolic disorders Alpha1-antitrypsin deficiency, galactosaemia, glycogen storage disorder type IV, cystic fibrosis, neonatal

haemochromatosis, tyrosinaemia, inborn errors of bile acid metabolism, dubin-johnson and rotor syndrome,

hereditary fructosaemia, niemann-pick type C, gaucher’s disease, progressive familial intrahepatic cholestasis,

aagenaes syndrome, wolman’s disease, peroxisomal disorders (zellweger’s syndrome), carbohydrate deficient

glycoprotein syndrome

Chromosomal disorders Down’s syndrome, trisomy 13 and 18, turner’s syndrome

Endocrinopathies Hypothyroidism, hypopituitarism,

Toxic Parenteral nutrition, foetal alcohol syndrome, drugs

Vascular Budd-chiari syndrome, neonatal asphyxia, congestive heart failure, multiple haemangiomata

Neoplastic Neonatal leukaemia, histiocytosis X, neuroblastoma, Hepatoblastoma, erythrophagocytic lymphohistiocytosis

Other Neonatal lupus erythematosus

Table 1

PERSONAL PRACTICE

atresia. A normal gallbladder makes biliary atresia unlikely but

does not exclude it.

Subsequent investigations

Further investigation to establish the cause of conjugated

hyperbilirubinaemia should be tailored according to history,

examination, and initial laboratory results. Second/third

line investigations should ideally be performed in a tertiary

institution under close guidance and supervision of paediatric

hepatologist. Hepatobiliary scintigraphy, endoscopic retrograde

cholangiopancreatography, magnetic resonance cholangio-

Initial investigations for neonates with conjugated hyperbiliru

First line inve

Immediate investigations Total and conjugated bilirubin, blood group

blood glucose, sodium, potassium, urea, cre

for reducing substances.

If acutely unwell add Plasma ammonia, lactate, pyruvate, acid-bas

samples of plasma and urine for future anal

Liver function AST, ALT, alkaline phosphatase, gamma GT,

Infection Blood cultures, c-reactive protein, urine cult

hepatitis A, B, and C serology

Metabolic/storage Cystic fibrosis genetics or sweat test, galact

phenotype, plasma and urine amino acids, u

iron and ferritin

Endocrine Thyroid function tests, cortisol (preferably a

Miscellaneous Ultrasound scan of abdomen after 4 h fast l

Table 2


pancreatography and liver biopsy may be required in selected

cases. In biliary atresia, typical liver biopsy findings include bile

duct proliferation, bile plugs in small bile ducts, portal tract

oedema and fibrosis. Liver biopsy is also useful for other specific

conditions such as alpha 1 antitrypsin deficiency and some

storage disorders. White cell enzyme analysis for glycogen and

lysosomal storage disorder, muscle biopsy for mitochondrial

cytopathy, bone marrow aspiration for storage disorders,

karyotype, ferritin and transferrin saturation, cerebrospinal fluid

examination for protein & lactate, MRI head and skin biopsy for

fibroblast culture are rarely required.

binaemia

stigations

and coomb’s test, full blood count, blood film and reticulocyte count,

atinine, bicarbonate, calcium, phosphate, coagulation screen, urine

e (blood gas), urinary pH, protein and ketones, chest X-ray, freeze

ysis.

albumin, cholesterol, triglycerides

ure and CMV, serology (IgM to Toxoplasma, Rubella, CMV, Herpes),

ose-1-phosphate uridyl transferase, alpha 1 antitrypsin level and

rine organic acids (succinyl acetone), carnitine and acyl carnitine,

fter 4 h fast)

ooking for gallbladder and choledochal cyst



PERSONAL PRACTICE

Initial steps in the treatment of conjugated hyperbilirubinaemia

This involves diagnosing conditions amenable to specific medical

therapy (sepsis, galactosaemia, hypothyroidism, and hypopitu-

itarism) or early surgical intervention (biliary atresia, chol-

edochal cyst). In the majority of the other conditions, medical

management is mainly supportive; optimizing growth and

nutrition, and treating complications such as pruritis, portal

hypertension and liver failure.

Nutritional management

Practice points

In neonatal cholestasis, long chain fatty acids are not well

absorbed leading to malnutrition and fat-soluble vitamin defi-

ciency. Medium chain triglycerides (MCT) have better absorption

as they are relatively water soluble. Infants who are formula fed

should immediately be changed to a MCT based hydrolyzed

formula. Similarly, breast fed infants should also be temporarily

switched to lactose free (e.g. MCT based hydrolyzed) formula till

the investigation results exclude galactosaemia. In the mean

time, mother should be advised to continue to express breast

milk to prevent subsequent lactation failure. Caloric content

should be increased to 120e150% of the recommended intake.

Occasionally, parenteral nutrition may be required.
C Neonatal cholestasis is an uncommon serious disorder which
requires urgent diagnostic workup
Medication
C Neonates with prolonged jaundice should have total and direct

serum bilirubin measurement and if found to have conjugated

hyperbilirubinaemia, should be promptly referred for further

investigation

C First line investigations look for immediately treatable condi-

tions and complications

C Biliary atresia is the most common cause of conjugated

hyperbilirubinaemia and early detection/referral for surgery

before 60 days significantly improves its prognosis

C Supportive management for cholestasis includes optimizing

nutrition and replacing fat soluble vitamins

Fat soluble vitamin supplements should be added in doses 2e4

times the recommended daily allowance and should continue for

at least 3 months after resolution of jaundice as there is delay

before normal bile flow is re-established. Either multivitamin

preparation or individual vitamins (Vitamin K 1e2 mg per day,

Vitamin E 100 mg per day, alphacalcidol 30e50 nanogram/kg

per day and vitamin A 5000 international units per day) should

be prescribed.

Ursodeoxycholic acid is a hydrophilic bile acid which replaces

hydrophobic bile acids in the bile pool and stimulates bile flow. It

has been shown to improve biochemical measures of cholestasis

and pruritis. Initial dose is 20 mg/kg per day in divided doses.

It should be discontinued when cholestasis has resolved.


Occasionally, other medication (such as rifampicin, Pheno-

barbital or cholestyramine) may be required to treat pruritis from

cholestasis. A

FURTHER READING

1 Venigalla S, Gourley RG. Neonatal cholestasis. Semin Perinatol 2004; 28:

348e55.

2 De Bruyne R, Van Biervliet S, Vande Velde S, Van Winckel M. Clinical

practice; neonatal cholestasis. Eur J Pediatr 2011; 170: 279e84.

3 McKiernan PJ. Neonatal cholestasis. Semin Neonatol 2002; 7: 153e65.

4 Roberts E. Neonatal hepatitis syndrome. Semin Neonatol 2003; 8:

357e74.

5 Kelly DA, Davenport M. Current management of biliary atresia. Arch Dis

Child 2007; 92: 1132e5.

6 Moyer V, Freese DK, Whitington PF, et al. Guideline for the evaluation

of cholestatic jaundice in infants: recommendations of the North

American Society for Paediatric Gastroenterology, Hepatology and

Nutrition. JPGN 2004; 39: 115e28.



SELF-ASSESSMENT

PAE

Self-assessment

Part A

An eight-year-old girl is referred to the paediatric unit with

(b) Codeine

(c) Ibuprofen

(d) Piroxicam
a 7-week history of persistent swelling and redness of her
right knee. There is no history of trauma or recent foreign

travel; she has been feeling weak, but has not had a fever or

weight loss or other constitutional symptoms. On further

questioning she has had pain reoccurring every month for

the past 4 months and lasting about a week each time.

There is no family history of note. General examination is

unremarkable. She walks with an antalgic gait, and her

right knee is red and swollen with restricted movements

and a small joint effusion. There is no leg length discrep-

ancy. Initial investigations include:

Haemoglobin 13.6 g/dl (12.0e15.5) C3 Normal range

White cell count 8.4 � 109/L (4.5e13.0) C4 Normal range

Platelets 278 � 109/L (150e400)

C reactive protein 1.5 mg/L (<3.0) Smooth muscle abs Positive

Phosphate 1.51 mmol/L (1.00e2.00) Mitochondrial abs Negative

Alkaline phosphatase 771 U/L (130e900) Antinuclear factor Negative

Calcium 2.53 mmol/L (2.20e2.60) Parietal cell abs Negative

IgG 8.8 g/L (5.4e16.1) Urinary HMA/VMA Normal range

IgA 1.10 g/L (0.4e2.4)

IgM 0.7 g/L (0.5e1.8)

Questions

1. What is the most likely diagnosis? Select one answer

(a) Septic arthritis

(b) Juvenile idiopathic arthritis e oligoarticular

(c) Traumatic injury

(d) Haemophilia

(e) Juvenile idiopathic arthritis e polyarticular

(f) Tuberculous osteomyelitis

(g) Leukaemia

(h) Hypermobility

2. What medical management should be instigated first?

Select one answer

(a) Paracetamol

Rebecca Balfour MB BCh is a Speciality Trainee at the Child and

Adolescent Health Directorate, Hywel Dda Local Health Board,

Bronglais Hospital, Aberystwyth, Wales, UK. Conflict of interest: none.

Simon Fountain-Polley MB BCh, MRCPCH is a Consultant at the Child and

Adolescent Health Directorate, Hywel Dda Local Health Board,

Bronglais Hospital, Aberystwyth, Wales, UK. Conflict of

interest: none.

DIATRICS AND CHILD HEALTH 22:4 173

(e) Intra-articular corticosteroid

(f) Methotrexate

3. If initial management is unsuccessful select the next best

management option:

(a) Paracetamol

(b) Codeine

(c) Ibuprofen

(d) Piroxicam

(e) Intra-articular corticosteroid

(f) Methotrexate

Part B

Questions

For the following statements choose the most appropriate

answer from the list of conditions:

A. Juvenile idiopathic arthritis e oligoarticular

B. Juvenile idiopathic arthritis e extended oligoarticular

C. Juvenile idiopathic arthritis e polyarticular, RF negative

D. Juvenile idiopathic arthritis e polyarticular, RF positive

E. Juvenile idiopathic arthritis e psoriatic

F. Juvenile idiopathic arthritis e enthesitis related

G. Juvenile idiopathic arthritis e systemic

H. Juvenile idiopathic arthritis e undifferentiated

1. A 14-year-old boy attends for repeat joint injection of an

inflamed right knee. He has a raised red rash over his

knees, which his mother says has been present only

over the last 2 weeks.

2. A 5-year-old girl is examined prior to a planned repeat

joint injection for an arthritic right knee noted on

a recent follow-up clinic appointment. Her father

mentions she has had pain on eating, especially apples.

As well as a restricted swollen right knee she has

restriction of movement in her left knee and both ankles

associated with swelling.



SELF-ASSESSMENT

3. A 15-year-old boy complains of pain in his ankles when

walking and in the mornings. He was initially seen by

the orthopaedic team who put him in plaster casts for

both legs which relieved his symptoms. There is

a family history of ulcerative colitis.

4. A 4-year-old girl with 8 weeks of painful restriction of

movement in a swollen left knee e the only positive

finding. No evidence for malignancy is found. The

family commutes between the UK and Gibraltar. Her

father has psoriasis.

Part A

Answers

1. b e JIA e oligoarticular

Juvenile Idiopathic arthritis is a diagnosis of exclusion

when no other cause can be found. It is diagnosed in chil-

dren under 16 years of age with arthritis in one or more

joint persisting for more than 6 weeks. They present with

joint swelling, tenderness and signs of inflammation. This

can be further divided depending upon the number of joints

involved at presentation and the presence of any other

features including psoriasis, HLA-B27 positivity, Rheuma-

toid factor positivity or systemic features. Oligoarthritis is

when there are four or less joints involved at presentation.

2. d e Piroxicam

Treatment of Juvenile Idiopathic Arthritis (JIA):

The treatment of JIA should be multi-disciplinary,

involving paediatricians, paediatric rheumatologists, phys-

iotherapists, occupational therapists, podiatrists, orthotists,

specialist nurses, psychologists, family support groups,

ophthalmologists and dentists. Fortunately there are many

children with chronic arthritis who have remission and this

should be the main aim of treatment. Further treatment

aims should be to control pain, preserving joint ranges of

movement and function, and to facilitate normal growth

and psychological development. It is important that the

child can continue to participate in school and physical

education activities to prevent adverse psychological

effects. Pharmacological treatment should start with the

safest measures and escalate depending upon disease

progression and the response to medications.

Non-steroidal anti-inflammatory medications (NSAIDs)

are recommended as the first line treatment because rapidly

controlling inflammation can reduce permanent sequelae.

There are several different options available and there is no

convincing evidence that one NSAID is superior to another.

Piroxicam is recommended because it is administered on

a once-daily basis, and is available as a melt, which hypo-

thetically increases the likelihood of concordance with

treatment. With NSAIDs gastro-protective medication may

need to be considered. Clinical response to these medica-

tions can be variable and can take as long as 4 weeks to see

any improvement in symptoms.

3. e e Intra-articular corticosteroid

Intra-articular glucocorticoid injections are indicated in

oligoarthritis or polyarticular arthritis when one or a few


joints have not responded to adequate anti-inflammatory

treatment. Repeat injections can be required. The proce-

dure can be performed under local anesthetic in children

over 7 years, but general anaesthetic may be needed if the

hip joint is involved or several joints are being injected. Side

effects from injections include subcutaneous atrophy,

cutaneous depigmentation, and increased pain for 24e48 h

post-injection and theoretically septic arthritis. Avascular

necrosis after hip injections has been recognized. Systemic

side effects are rare with the use of triamcinolone hex-

acetonide (Lederspan).

Methotrexate is the second line therapy for children with

arthritis who show inadequate improvement with first line

medications. This can be used in conjunction with intra-

articular steroid injections. Methotrexate is recommended

because of the rapid onset of action and acceptable side

effects. Initial treatment can be a once weekly oral medica-

tion, with a starting dose of 10e15 mg/m2. Subcutaneous

injections may need to be given if the response is inadequate

or there is associated nausea and vomiting. Subcutaneous

injections have been shown to increase bioavailability by

10e12%. In most children a response would be expected

within the first 3 months although this can take up to 9e12

months. Current debate revolves around the use of SC

methotrexate from the start of disease modifying antirheu-

matic medication, as there is some evidence that it may lead

to quicker, more effective symptom resolution.

The best time to discontinue methotrexate is unclear, but

most Paediatric Rheumatologists will begin dose reduction

following 1e2 years of disease control, although up to half

of these children will have flare-ups and this is more

common if the age of onset of symptoms is less than 5

years. Side effects of methotrexate include nausea, vomit-

ing, mouth ulcers, reduced appetite, alopecia, transient rise

in liver enzymes and leucopenia. Folic acid has been shown

in adults to reduce side effects and debate continues as to

whether it should be given routinely to children while on

methotrexate. Malignancy is rare following methotrexate

but Non-Hodgkin’s lymphoma has been reported.

Other medications that can be considered are systemic

corticosteroids, sulfasalazine, or the increasing number of

available biologic agents.

Part B

Answers

1. E e Juvenile idiopathic arthritis e psoriatic

Classification of juvenile idiopathic arthritis:

The ILAR (International League of Associations for

Rheumatology) proposed a classification of Juvenile Idio-

pathic Arthritis with a subsequent revision. The children

included will be under 16 years of age and have had

arthritis in one or more joints persisting for more than 6

weeks, and for which no other cause is found. JIA is

a diagnosis of exclusion. The full classification has

a number of distinct inclusion and exclusion criteria but the

following summarises each group:



SELF-ASSESSMENT

To determine classification the following exclusion

criteria are assessed:

(a) Psoriasis in patient/first degree relative

(b) Arthritis in HLA B27 positive male with onset after 6

years of age

(c) Ankylosing spondylitis, enthesitis-related arthritis,

sacro-ilitiitis with inflammatory bowel disease, Reiter

syndrome, acute anterior uveitis in a first-degree relative

(d) Presence of IgM rheumatoid factor on at least two

occasions more than 3 months apart

(e) Presence of systemic arthritis

Psoriatic arthritis is classified in children with arthritis

and psoriasis or arthritis plus two of the following: dacty-

litis, nail pitting/onycholysis or psoriasis in a first-degree

relative. The most severe form is arthritis mutilans, which

is severely deforming. Prognosis is dependent on which

subtype is present.

Exclusions e b, c, d, e.

2. Be Juvenile idiopathic arthritise extended oligoarticular

Oligoarticular arthritis can be divided further into

persistent and extended. If persistent then four or less joints

are affected throughout the disease process. Extended

means that during the first 6 months four or less joints are

involved but with further joint involvement after 6 months.

These children are at high risk of associated uveitis and

ANA positivity may have prognostic significance. Oligoar-

thritis tends to be more common than polyarthritis,

affecting girls more than boys, with a peak in early child-

hood and a good prognosis; uveitis may affect up to a third

of children and requires regular ophthalmology reviews.

Exclusions e a, b, c, d, e.

3. F e Juvenile idiopathic arthritis e enthesitis related.

Children with enthesis related arthritis either have

enthesitis and arthritis or arthritis plus two of the following:

Sacro-iliac joint involvement, HLA-B27 antigen positivity,

onset at more than 6 years in a male, acute anterior uveitis

or a first degree relative with HLA-B27 related disease.

These children can have associated inflammatory bowel

disease, erythema nodosum and pyoderma gangrenosum.

There is usually a good prognosis for peripheral joint

involvement but permanent changes in the hips and spine

occur frequently.

Exclusions e a, d, e.


4. H e Juvenile idiopathic arthritis e undifferentiated

The classification of undifferentiated arthritis is used for

any JIA, which doesn’t fit the criteria for any of the above

categories. It is also used if the arthritis fulfils the criteria for

two or more of the categories.

There are two further categories in JIA; these are poly-

arthritis and systemic disease. In polyarthritis there are five

or more joints involved during the first 6 months from the

onset of symptoms. This can then be further divided

depending on whether rheumatoid factor (RF) is positive

(exclusions e a, b, c, d, e) or negative (exclusions e a, b,

c, e). Those who are RF positive are more likely to be older at

the time of presentation, have rheumatoid nodules and

articular erosions. If they are RF positive they tend to have

aworse prognosis with symptoms continuing into adulthood

and have similar findings to adult rheumatoid arthritis.

Children with systemic disease can present at any age

with arthritis and fever for at least 3 days with one or more of

the following: evanescent erythematous rash, lymphade-

nopathy, hepatospelnomegaly and serositis. There is

a similar male to female ratio; antibodies and uveitis are

rarely present. The arthritis is usually oligoarticular at the

start but progresses to polyarticular. It most commonly

involves the knees, wrists and ankles. It can also involve the

cervical spine, hips, temporomandibular joint and the small

joints of hands. There can be extra-articular complications of

pericarditis, secondary amyloidosis and pulmonary intersti-

tial disease. Around half of the children will recover almost

completely with the other half having progressive involve-

ment of more and more joints and subsequent disability.

Exclusions e a, b, c, d.

FURTHER READING

Brough R, Cleary G. When does a knee “need” a “joint” assess-

ment? Arch Dis Child Educ Pract Ed 2007; 92: 44e49.

Cassidy JT, Petty RE, Laxer RM, Lindsley CB. Textbook of Paediatric

Rheumatology. Philadelphia Elsevier Saunders, 2005.

McCann LJ, Wedderburn LR, Hasson N. Juvenile Idiopathic Arthritis.

Arch Dis Child Educ Pract Ed 2006; 91: ep29eep36.

Szer S L, Kimura Y, Malleson PN, Southwood TR. Arthritis in Chil-

dren and Adolescents Juvenile Idiopathic Arthritis. Oxford

University Press, 2006.



Documents

Paediatrics and Child Health April2012