Jennifer Paratz Yvonne Burns

The control of cerebral circulation and intracranial dynamics differs markedly in the pre-term and full term neonate from that in the adult. Immaturity can combine with several clinical conditions and iatrogenic factors to predispose the neonate to cerebral lesions. which may subsequently increase morbidity. As physiotherapists play an increasingly important role in neonatology. it is importantto appreciate the immaturityofthe nervous system and to recognise the risk factors for such conditions as peri-intraventricular haemorrhage and periventricular leucomalacia. This paper describes intracranial dynamics in the pre-term and full term infant and iliustrC'tes how these factors may interact with clinical conditions to cause cerebral lesions. Studies which examine the effect of respiratory physiotherapy on intracranial dynamics are reviewed and suggestions made for further research. [Paratz JD and Burns VR: Intracranial dynamics in pre-term infants and neonates: implications for physiotherapists. Australian Journal of Physiotherapy 39: 171-178]

Key words: Infant, newborn; Neonatology; Respiratory therapy JO Paratz MPhty. MAPA is a fUll-time post graduate student in the Department of Physiotherapy. The University of Queensland.

YR Burns MPhty. PhD. MAPA is Professor and Head. Department of Physiotherapy.The University of Queensland. Correspondence: Jennifer Paratz. Department of PhySiotherapy,The University ofOueensland. Old 4072.

ORIGINAL ARTICLE

Intracranial dynamics in pre-term infants and neonates: implications for physiotherapists

As survival rates for pre-term infants increase, cerebral haemorrhagic and hypoxic

ischaemic lesions continue as a major cause of morbidity in neonates (Volpe 1989). Should cerebral lesions occur, infants with parenchymal involvement, ie Grade IV peri-intraventricular haemorrhage (P-IVH) and especially periventricular leucomalacia (PVL), are at a significant risk of poor neurodevelopmental outcome.

The most consistent factors associated with these cerebral lesions have been pre-term birth (less than 35 weeks gestation), respiratory distress syndrome (RDS), asphyxia, and

. cardiocirculatory disturbances (Lou 1988a). A number of clinical risk factors are implicated and it has been suggested (Greisen et al1985, Perlman and Volpe 1983) that respiratory physiotherapy may be a potential cause ofP-IVH. This paper aims to explore that proposition.

As pathogenesis ofP-IVH and PVL is presently believed to be due to fluctuations in arterial blood pressure, arterial blood gases and especially abmptdecreases or increases in cerebral blood flow (Lou 1988a, Volpe 1989), ids conceivable that some respiratory physiotherapy techniques may predispose towards cerebral lesions. However, many interactive factors combine to cause these lesions and it is difficult to isolate physiotherapy asa single factor, when

the severity of the clinical condition will dictate the extent of intervention. In order to examine the effects that

physiotherapy may have on these patients, it is necessary to discuss intracranial dynamics in the pre-term and full term infant. It is also important to list the factors that may predispose to cerebral lesions, to illustrate how these may affect cerebral blood flow, and then review the reports of research studies of the effect of physiotherapy on the cerebral circulation in this patient group.

Intracranial dynamics Maintenance of cerebral oxygenation and metabolism is achieved by a complex process of regulation of cerebral blood flow (CBF). The mechanisms controlling CBF in the mature brain (Harper and Jennett 1990) are represented in Figure 1. Many of these control mechanisms are immature or absent in the premature infant, leading to irregular blood flow and inadequate perfusion pressure.

Autoregulation An important aspect of intra cranial physiology is the concept of autoregulation. Under normal conditions, CBF is maintained independently of systemic blood pressure, within certain limits, by the adjustment of the diameters and resistance of cerebral arteries and

ORIGINAL ARTICLE

:". ::" : :-.~:':':'.: ' .. ::: :;~~. :'." ';':~':':' :~:;./.' ~-;.:: :';::.;>: ."," .. ,

'. : : - ',' : :. - " :-: ':. - ~

O:nCEN (PaC~2) " :+ •... >.'~riIo~(l'aCo,) . pa~f!Bt ECF [lJ:+,I, . lp.()z .... ttP~02J agellesll;le.+ tlll+J .'

t t PaCOz ,... Autoregulation ,.... CBF

Abo~·"

'=-CBF' . .... '. ..... ." ....; ~.....; .'

~v·'· . .", '". .":",

MYOGENIC····· METABO:UC CA'" =- Releas~()f 1lI!Urritrarismitier .'. Adenosine ,... Vasodilation· ::'V=::r "'Si

.............•...•. ~,". ······~.···>ii,·,; ................ ~

'CPp -.... Release ofmetahoIites

...> ...... ,.:J . . ' ..... .

··'···.M~· Figure 1. Factors involved in control of cerebral blood flow in the mature brain.

From Page 111 arterioles. Postulated limits for neonates are listed in Table 1. Should autoregulation be immature or not functioning, any alteration in blood pressure could subject the brain to ischaemia or hyperperfusion. Autoregulation appears to be functioning in the full term infant; however it can be abolished by hypoxaernia, hypercarbia and acidosis (Lou .etal 1988b).

Depending on birth weight; gestation and clinical condition, autoregulation may hot be functioning in the pre-'term infant. Jorch and Jorch (1987) maintain that pre-term infants who are less thah

1500g at birth and ofless than 30 weeks gestation, do not reach the range of autoregulation.

Mean arterial blood pressure (MAP), systolic arterial blood pressure (SAP) and diastolic blood pressure (DAP) are all lower in the pre-term infant than in the full term infant. Versmold et al (1981) have provided average values and upper and lower limits for MAP, SAP and DAP in infants weighing 750g and 1000g (fable 1). .

It can be seen ftomthis data that the average MAP in this age group does not reach the autoregulatory range. This suggests that there is a passive dependency on changes in arterial

{;~.:; ji~~N~~t![>·~L;d~· .~;." "'"':.<;':' -,">" :'O'".f'--,.--; .,' .~.\ .. ;.-:-•. ~ ••••. ::-.-.): •• ...... . .. ,,~. ccc'

-- :::-'-':.';< -.:-:-~'. :~(> ,:·2<:-; -~-".: ',,<:".' __ ' __ ';-_<:-~~'_:'_ ~~~:_i\:~: "

blood pressure. Pre-term infants of less than lOOOg at birth, and ofless than 30 weeks gestation, therefore are subject to constant fluctuations in CBF. The relative size of the brain to body

weight also affects cerebral perfusion. The brain in the very pre-term infant is 20 per cent of their total body weight and requires a large proportion of the cardiac output (Cooke 1989). If there is any fall in cardiac output, CBF will fall, whether or not the infant is able toautoregulate.

Neurogenic mechanisms As demonstrated in Figure 1, central neurogenic mechanisms play an

ORIGINAL ARTICLE

·\:e ...... :;::mi,.c:-::.:rl"J"'....:::::t'£ " "'own in Pft_~iS. "" , " , , .. ", ' ." ~ ; <" ," ,

, , ,

'Intracr~pressure

!~~:!~II* "4 -5.8mtriH~,

important part in controlling CBF. At various stages of foetal life, brain stem nuclei are not fully developed (Nakamura et al1987). It is therefore possible that the autoregulatory role of central neurons in very pre-term infants may not be completely functional. As the exact role of each nucleus in the control of CBF in the mature brain is not fully understood, it is difficult to interpret the effects of this immaturity on the premature hrain. A major difference which may be of relevance clinically is that the locus ceruleus, one of the largest noradrenaline containing neurons in the brain is, in the pre-term infant, equally sensitive to noxious and non­noxious stimuli (Nakamura et alI987). As development continues, the locus ceruleus neurons become less sensitive to innocuous, somatosensory stimuli such as touch, while simultaneously becoming more sensitive to noxi.ous stimuli. This may partially explam why very pre-term infants react to e,:en gentle handling with hypoxaelllia and

=I~ (StrassbOj1,:g ~~' at 1988)

increased blood pressure.

The sympathetic nervous system (SNS) is not fully developed in the pre­term infant (Rogers et aI1980), thus the SNS cannot control increases in blood pressure and the force of any such increase is likely to be transmitted straight to the vascular bed. It appears, however (Monin et al 1990), that in full term infants, the SNS is closely involved in autoregulation of CBF flow. The parasympathetic nervous system is less mature than the SNS, especially in pre-term infants, so alterations in blood pressure below the autoregulatory limits (ie a decreasein blood pressure) may result in -ischaemia.

Arterial blood gases In the mature brain, the partial pressure of arterial carbon dioxide _ (p.CO) has pow:erfuleffects on CBF~ In the pre-term mfant, after the first 24 hours, CBF increases with rising p.C02

(van Bel et al 1988).

Cerebral circulation in the foetus is critically dependent on p.Oz' with . decreases of only 7 -llmmHg caUSIng increases in CBF.

Hyperoxia appears to have a powerful effect on CBF, with the administration of 100 per cent oxygen resulting in a 33 per cent decrease in CBF c~mpared with a 12 per cent decrease which occurs in adults (Donnegan 1986).

Intracranial volume-pressure relationships An understanding of intracranial volume-pressure relationships is also important in understanding the effects of respiratory physiotherapy on cerebral dynamics. Events that increase the volume in the cranial cavity may all lead to an increase in intracranial pressure (ICP), especially w:ith la~k of autoregulation. These relatIonships are illustrated in Figure2b (Shapiro and Marmarou 1982).

The pressure-volume curve is much steeper in the infant (Figure 2a). This indicates that for a much smaller change in volume, a higher ICP will result.

Normal ICP is much lower in the infant, with these proposed values listed in Table 1. During the first few days after birth, ICP decreas~s in some cases, due to loss-of bram water, to asub-atInospheric ICP (Welch 1980). After this, ICP begins a steady increase. It was hypothesised by Welch (1980) that the decrease in ICP ~ue to a decrease in brain water immedIately after birth produces pressure gradients sufficient to cause capillary rupture. In a recent review, Volpe (1989) discounted this theory and suggested that the timing of the two incidents differs.

Cerebral perfusion pressure A further important concept in understanding intracranial dynamics is cerebral perfusion pressure (CPP). The calculation of CPP is used clinically to evaluate the level of tissue oxygenation and is defined as the difference between MAP and ICP. It is best expressed by the formula:

ORIGINAl ARTIClE

,.' '," in p~ri·,i.".t.ventri~~,j.~,~aemo.r~M. and 1l~,~¥~~tricUI~f ,,~omalai~';{;:, ,',

• > ".iA:rci~::DIST_rnJi.ME Therapy'

t P1!lt:ra:~;~ Fluctl.1r~tion in IT!?

_(~I;,·· .,. Pneup;1,dthoru ... J in C~F.Y

,smtri!:I~:~inCB~::i,;~(·~!:,

From Page 113 CPP = MAP - ICP where MAP = DAP + 113 (SAP-DAP) (Rossanda and Vecchi 1978)

Cerebral perfusion is an indicator of the level of tissue perfusion and, in turn, autoregulation responds to CPP. Critical levels of perfusion pressure in the neonate have not yet been fully decided. Obviously they will be lower than in adults, as MAP is lower. Raju et al (1982) claim that CPP in healthy pre-term and term infants can be calculated from the following:

Mean CPP (mmHg) in infants during first week

= Gestational Age in weeks Mean CPP (mmHg) up to three weeks after birth

Effeets of RDS. "

:I,I:=lit+C~t • Flu.ctuatingABP ,,'':ii:i''\' ;', •

\\\: .:: .. ~:,. .;.: .. ":":'.:'~\:<:

i\· " :',', . • 'dl~ Y;;:~r' :·:Acid~:~etalbolic ?r

Respirilt?ry)'>mw~1~f;l~: ;"

t:,5~"

= Conceptional Age + 3 Mean CPP (nunHg) up to five weeks after birth = Conceptional Age + 5

As normal values ofICP are so low in this popuiation,arise in ICP is sometimes not considered a threat to cerebral dynamics. However, as only small increases in volume are needed for an increase in ICP, together with the fact that MAP is very low, perfusion pressure is likely to be at ischaemic levels. CPP is often threatened in the pre-term infant, as the blood pressure is low, rather than because the ICP is too high. Miall­Allen et al (1987) have noted that in infants of less than 31 weeks gestation, a MAP ofless than 30mmHg is associated with severe haemorrhage, ischaemic cerebral lesions or death.

The Cushing response is a sympathetically mediated rise in blood pressure that occurs in response to a rise inICP, in order to ensure an adequate CPP. Although the value of this response is debatable, as it is thought to occur when neurological deterioration is imminent, it is interesting to note that it appears to be only partially developed in pre-term infants., perhaps due to immaturity of brain stem nuclei and efferent sympathetic conduction.

Blood brain barrier and cerebral vascular structures As well as the more dynamic functions, there are various anatomical structures which are immature and may contribute to cerebral lesions.

The blood brain barrier does not appear to be fully developed in foetal

ORIGINAL ARTICLE

AV (ml)

Figure 2. The pressure-volume curves shown were generated by injecting ami withdrawing CSf from (a) a normal infant (left) ami (b) a teenager (right). The slope of the infant's pressure­volume curve is steeper than that of the teenager, resulting in less ability to buffer increments of volume. (Reprinted with kind permission from Shapiro and Marmarou 1982.)

and perinatal life Goo 1987). Noxious stimuli causing a peripheral release of catecholamines are therefore able to affect cerebral arteries, producing a significant increase in CBF.

Important differences occur in the cerebral vascular supply of the pre­term infant. The germinal matrix in the ventricular and subventricular zones is a site of proliferation of neuronal and glial precursors. The large irregular vessels supplying this tissue are not distinguishable as arteries, veins and capillaries. These vessels are larger in pre-term infants, with a larger portion of CBF going through the germinal matrix, making the area more vulnerable to haemorrhage. Van Bel et al (1987) have described fluctuations of blood flow in the anterior cerebral arteries, vessels which are especially important for blood supply to the germinal matrix.

In summary, a mllnbet of immaturities exist in the pre-tem and

full term infant. It is evident therefore that many factors involved in the control of CBF are immature in the pre-term and full term infant. This exposes the neonate to the possibility of a cerebral lesion.

Risk factors for cerebral lesions As well as the immaturity of control mechanisms, a number of clinical conditions and medical interventions may occur which can predispose the infant to cerebral lesions.

The most common factors associated with the conditions ofP-IVH and PVL are gestational age, respiratory distress syndrome (RDS), asphyxia and cardiocirculatory disturbances (Van Bel et al 1987). Other risk factors have been hypothesised, and these are often associated with, or a direct result of, these factors (Lou 1988a). Table 2 lists the many risk factors for these two conditions and indicates how they may affect cerebral dynamics.

Many factors related to the

immaturity of the nervous system, for example the structure of the germinal matrix, irregular blood flow velocity, impairment of autoregulation, lack. of the blood brain barrier, unstable blood pressure and features caused by RDS, all predispose both full and pre-term infants towards cerebral lesions. It is difficult to isolate factors, as they will interact. For example, the administration of artificial surfactant for RDS will affect CBF, but the same infant will also have had high PaCO~ and low PaOZ' frequent endotracheal suctioning, altered electrolytes and fluctuations in blood pressure. The common feature of cerebral lesions appears to be widespread arterial dilatation, hypoxaemia andlor hypercarbia with irregular CBF.

Preventative measures introduced by some neonatal units include intubating the infant while in the delivery room, hyperventilating to keep the P a CO low, and using muscle paralysis to ~eep the blood pressure and arterial blood gases under control during the critical stage when the incidence ofP-IVH is high (Ment 1988). In addition, physiotherapists shorildexamine their treatment methods to decide whether any components of the techniques used could cause or aggravate potential risks forP-IVH.

Effect of respi ratory physiotherapy on intracranial dynamics Physiotherapy is an important component of management in the acutely ill pre-term and full term infant, in the active treatment of such conditions as aspiration syndromes, RDS, pneumonia and atelectasis and the prevention of iatrogenic complications of ventilation. Without adequate removal of secretions and aeration of the lung, hypercarbia and hypoxaemia can result, causing alterations in CBF and thus putting the infant at risk for cerebral lesions. As very premature infants have a poor tolerance for any form of stimrilation, physiotherapy must be carefully planned.

There are very few studies which

From Page document the effect of respiratory physiotherapy on intracranial dynamics in this particular population. Hence it is important to review other studies, not directly measuring such parameters as ICP and CBF velocity.

Early research reports appeared to show that physiotherapy decreased p.o z (Holloway et al 1969). With improved treatment methods, subsequent studies (Curran and Kachoyeanos 1979, Dall' Alba and Burns 1990, Finer and Boyd 1978, Tudehope and Bagley 1980) have found that cupping, that is percussion with a neonatal resuscitation face mask, appears to be an effective form of treatment. These authors indicated that PaOZ increased during most physiotherapy interventions. This effect may be due to shorter treatment periods and pre-oxygenation before treatment.

Cupping is not in itself a noxious stimulus. It must be remembered, however, that very premature infants may react adversely to non-noxious tactile stimulation.

Changes in intrathoracic pressure have been documented during cupping and vibration, which could have significant effects on ICP (Hillman 1988).

Greisen et al (1985) measured adrenaline, noradrenaline and MAP in 13 pre-term ventilated infants just before and just after respiratory physiotherapy, consisting of vibration and endotracheal suctioning. After the procedure, both adrenaline and noradrenaline were significantly higher than baseline levels. The adrenaline response was reduced by phenobarbitone. MAP was not affected by the procedure and did not correlate with the rise in catecholamines. The lack of a blood brain barrier in this population would mean that the catecholamines could affect CBF.

In a study designed to investigate the efficacy of respiratory physiotherapy, ~aval et al (1987) divided their subjects mto two groups. The treatment group received postural drainage, percussion and suction, while the control group

ORIGINAl ARTIClE

were given suction only. Five of 10 in the treatment group suffered a Grade III or Grade IV P-IVH, whilst subjects in the control group demonstrated Grade I or Grade II P-IVH. These findings appear at first to support the proposition that respiratory physiotherapy causes P-IVH. On closer examination of the methodology, it is apparent that treatment was given to the infants at a mean postnatal age of 4.9 hours (range 1-12 hours). The infants were also in the early stage of RDS. In most neonatal units today, pre-term infants would not be treated with active physiotherapy at this early stage (except in the case of meconium aspiration) but would receive positioning and suctioning only. The actual treatment regime, as described by Raval et al (1987), consisted of percussion in both the head up and head down position. This is a very ~ggressive intervention strategy for mfants of this postnatal age. Since the infant would be very unlikely to have autoregulation, the CBF would have undergone a dramatic decrease and increase over few minutes, an event very likely to cause P-IVH.

Positioning, both nursing postures ~nd drainage for physiotherapy, are an Important part of physiotherapy. The handling and increased activity level associated with physiotherapy int~rvention could affect oxygenation and cerebral dynamics. However, alterations in posture are necessary for a variety of reasons related to respiratory function, neurological maturation, pressure area care and cranial moulding.

Prone is an ideal position for the pre­term infant in terms of reflex activity, haemodynamics, energy expenditure, respiratory function and decreased risk of aspiration (Masterson et aI1987). Reservations about head rotation causing trapping of CSF and subsequently increasing ICP have been expressed (Goldberg 1988). Because of the high pressure volume index (PVI) and low arterial blood pressure, a small increase in volume, perhaps caused by trapping of CSF during head rotation, may decrease CPP. Modifying the

position to a one quarter turn from prone can assist in keeping the neck central. Side to side postures are also used and these are preferable if elevated ICP is likely to be a problem.

Postural drainage should be carefully considered in a neonate with risk factors for cerebral lesions. If the head down position is used, this could increase ICP and CBF. Thus head down positions are rarely advocated for extremely premature infants.

Manual hyperinflation (MHI) is a technique used occasionally with full term neonates. The technique aims to give an increased minute volume in order to provide oxygenation before suction and expand atelectatic alveoli.

MHI has the potential to decrease cardiac output through the application of increased positive pressure. This is offset by a sympathetic compensatory response. Restoration of cardiac output, however, results only if the SNS is capable of this response. Evidence exists that sympathetic innervation and control of cardiac contractility and rhythm are abnormal in the neonate (Rogers et aI1980). A decrease in cardiac output can decrease CBF, causing either ischaemia or widespread cerebral vasodilatation, exposing the brain to potential haemorrhage. The fact that manual hyperinflation is sometimes performed with 100 per cent oxygen could also be dangerous, as this could reduce CBF to ischaemic levels. It is also dangerous to hyperventilate infants to a P CO of <15mmHg. Greisen et al (1987) found significant neurodevelopmental deficits in infants who had been severely hypocarbic. As the technique also carries a high risk of pneumothoraces, it has little place in the care of pre­term infants.

Suction, whether endotracheal, nasopharyngeal or oral, has inherent hazards and can cause hypoxaemia, arrythmias, distress and alter haemodynamic and intracranial variables. Suction, is however, a necessary technique required to ensure patency of the endotracheal tube.

Vaughan et al (1978) suggest that pneumothorax occurs frequently during suction, due to perforation of

the segmental bronchi by the suction catheter. Pneumothorax can precipitate changes in CBF which can lead to a P-IVH. Simbruner et al (1981) indicate that the hypoxaemia which occurs during suction can cause necrosing enterocolitis, persistent patent ductus arterious and intra pulmonary hypoperfusion.

These effects can also lead to cerebral vasodilatation, increased CBF, increased ICP and perhaps decreased CPP. A study reported by Perlman and Volpe (1983) demonstrated a marked increase in CBF velocity in the anterior cerebral arteries, arterial blood pressure and ICP, events which could predispose to P-IVH. Fanconi and Duc (1987) suggest that muscle paralysis may minimise significant increases in ICP, leading to dangerous decreases in CPP during suctioning, but this management is controversial and can lead to other problems. The use of in-line suction adaptors to

help prevent hypoxaemia and disconnection during suction has become frequent practice in neonatal intensive care (Gunderson et al 1986). These modifications may reduce some of the risk associated with the procedure but require further study.

Thorough assessment of the haemodynamic as well as the respiratory system, can yield important clues to potential adverse effects on intracranial dynamics and should be a major precursor to any physiotherapeutic management in high risk neonates.

Although a frequent occurrence in pre-term infants, bradycardias can trigger and involve associated changes in CBF and ICP Neither absolute heart rate values nor trends in heart rate prior to bradycardia appear to predict these occurrences (Gorski et al 1990). It has been found, however, that there appears to be a marginal change in transcutaneous oxygen(T· PO over .. ... ~ ~

five minutes prior to bradycardia. If the infant is touched when they are already compromised, that is when low T cP02 present, this can potentiate bradycardia. Clearance of the airway, however, is often required when TlO:z is low. .

ORIGINAL ARTICLE

Assessment of the infant's blood pressure is necessary, since if blood pressure is too low, it will not be in the autoregulatory range, and CPP may be at risk. Hypotension also indicates that the cerebral vessels are widely dilated and any sudden increase in blood pressure at this stage could precipitate a haemorrhage.

Based upon reports in the available literature, and knowledge of the techniques employed, it does appear that physiotherapy has the potential to cause alterations in cerebral venous pressure, CBF velocity, and CPP. The fact that the peak incidence of P-IVH is usually in the first 24 hours, (Dolfin et a11983) and active physiotherapy is not usually commenced during this period, does undermine the proposal that physiotherapy is responsible for this particular cerebral lesion. The results of existing studies of the

effect of physiotherapy on intracranial dynamics are not conclusive and further work is indicated.

With the increased availability of non-invasive monitoring, such as cerebral doppler ultrasound Gorch et al 1986), transcutaneous monitoring of blood gases and fontanometers for monitoring ofICP (Rochefort et al 1987), physiotherapists should be better able to assess patients who are at risk for cerebral variations in blood flow and to gain direct feedback on the effect of their techniques.

Acknowledgements The graphics in this article were partially funded by a grant from the Physiotherapy Research Foundation.

References CookeRWI,DubowitzLMS,EyreJandWhitelaw

A (1989): Neurological abnormalities. In Harvey D, Cooke RWIand LevittGA (Eds): The Baby Under 1000g. London: Wright, pp. 22.6-28.8.

Curran CL and Kachoyeanos MK (1979): The effect on neonates of two methods of chest physical therapy. Mothercraft Nursmg4: 309-313. .

Dall'Alba PT and Burns YR (1990): The relationship between arterial· blood· gases and removal of airway secretions in .neonates. Physiotherapy Theory and Practice 6: 107~116.

Oolfin T,Skidmore MB and Fong DK (1983): Incidence, severity and timing of

subependymal and intraventricular haemorrhages in preterm infants born in a perinatal unit as detected by real time ultrasound. Pediatrics 7: 541-544.

Donegan J (1986): Anesthesia for pediatric neurosurgery. In CottrellJE and T umdorfH (Eds): Anesthesiaand Neurosurgery. StLouis: CVMosbyCo., pp. 173-182.

Fanconi S and Duc G (1987): Intratracheal suctioningin sickpreterm infants: Prevention of intracranial hypertension and cerebral hypoperfusion by muscle paralysis. Pediatrics 79: 538-543.

Finer NN and Boyd (1978): Chest physiotherapy in the neonate: A controlled study. Pediatrics 61: 282-285.

Goldberg R, Joshi A, Moscoso P and Castillo T (1983): The effect of head position on intracranial pressure in the neonate. Critical Care Medicine II: 428-430.

Gorski PA, Huntington L and Lewkowicz DJ (1990): Handlingpreterm infantsinhospitals. Clinics in Perinatology 17: 103-112.

Greisen G, Frederiksen PS, Hertel J and Christensen NJ (1985): Catecholamine response to chest physiotherapy and endotracheal suctioning in preterm infants. Acta Paediatrica Scandinavia 74: 525-529.

Greisen G, Munck H and Lou H (1987): Severe hypocarbia in preterm infants and neurodevelopmental deficit. Acta Paediatrica Scandinavia 76: 401-404.

Gunderson LP, McPhee AJ, and Donovan EF (1986): Partially ventilated endotracheal suction: use in newborns with respiratory distress syndrome. American JournalofDisease in Childhood 140: 462-465.

Harper AM and Jennett S (1990): Cerebral Blood Flow and Metabolism. Great Britain: Manchester University Press.

Hillman K (1987): Intrathoracic pressure fluctuations and periventricular haemorrhage in the newborn. Australian Paediatric Journal 23: 343-346.

Holloway R, Adams EB, Desai SD and Thambian AK (1969): Effects of chest physiotherapy on blood gases of neonates treated by intermittent positive pressure respiration. Thorax 24:421-426.

Joo F (1987): Current aspects of the development of the blood-brain barrier. International Journal of Developmental Neuroscience 5: 369-372.

Jorch G, Pfannschmidt J and Rabe H(l986): Investigationofcerebral circulation in preterm and te.rm neonates by pulsed doppler ultrasonography. Basics, Technique, Reference Values. Monattschr Kinderhezlkd 134; 804-807.

Jorch Gand Jorch N (1987): Failure of autoregulation of cerebral blood flow in neonates studied by pulsed Doppler ultrasound of the internal carotid artery. EuropelinJournal ofPediattla 146: 468-472.

from Page 117 LouHC (1988a):Perinatal hypoxia-ischaernicbrain

d~~ge ,an~ ~traventricular haemorrhage. Bat/here s Chmcal Obstetrics and Gynaecology 2: 213-220.

Lou HC (1988b): The lost autoregulatory hypothesis. Brain and Development 10: 143-146.

MastersonJ, Zucker C and Schulze (1987): Prone and supine positioning effects on energy expenditure and behaviour of low birth weight neonates. Pediatrics 80: 689-692.

Ment LR, Ehrenkranz RAand Duncan CC (1988): Intraventricular haemorrhage of the preterm neonate: Prevention studies. Seminars in Perinatology 12: 359-372.

Miall-Allen VM, de Vries LS, Dubowitz LMS and Whitelaw AGL (1989): Blood pressure ~uctuations and intraventricular haemorrhage m the preterm infant of less than 31 weeks gestation. Pediatrics 83: 657-661.

Monin P, FeilletF, HascoetJMand VertP (1990): Effect of sympathetic nervous system on cerebral blood flow in the newborn piglet. Biology of the Neonate 58: 192-199.

Nakamura S, Kimura F and Sakaguchi T (1987): ~ostuatal developmental of electrical activity m the locus ceruleus.JournalofNeurophysiology 58: 510-524.

PerlmanJMandVolpeJJ (1983): Suctioningin the preterm infant: effects of cerebral blood flow

ORIGINAL ARTICLE

velocity, intracranial pressure, and arterial blood pressure. Peditttrics 72: 329-334.

Raju TNK, Doshi UV and Vidyasagar D (1982): Cerebral perfusion pressure studies in healthy preterm and term newborn infants. Journal of Pediatrics 100: 139-142.

Raval I?, Yeh TF, Mora A, Cuevas P, Pyati S and Pildes ~S (1987~: Chest physiotherapy in pretermmfantsWlthRDS in the first 24 hours of life. Journal of Perina to logy vii :301-304.

Rochefort MJ, Rolfe P and WIlkinson AR (1987): New fontanometerforcontinuous estimation of intracranial pressure in the newborn. Archives of Diseases in Childhood 62: 152-155.

RogersMC, Nugent SK and Traystman RJ (1980): Control of cerebral circulation in the neonate and infant. Critical Care Medicine 8: 570-574.

Rossanda M and Vecchi G (1979): Determination of cerebral autoregulatory status and P CO

. 1: a 2 responsIveneSS. nternational Anaesthesiology Clinics 17: 425-438.

Shapiro KandMarmarouA(1982): Mechanisms of intracranial hypertension in children. In McLaurin RL (Ed.): Pediatric Neurosurgery: Surgery of the Developing Nervous System. New York: Grune and Stratton pp.255-264.

Simbruner G, Coradello H, Fodor M, Havelec L, Lubec G and Pollafe A (1981): Effect of tracheal suction on oxygenation, circulation and lung mechanics in newborn infants. Archives of Diseases in Childhood 56: 326-330.

T udehop~ DI and B~glo/ C (1980): Techniques of phYSIOtherapy m mtubated babies with the respiratory distress syndrome. Australian PaediatricJournaI16:226-228.

Van Bel F, van de Bor M, Stijnen T, BaanJ and Ruys JH (1987): Aetiological role of cerebral blood-~ow alterations in development and extenSIOn of peri-intraventricular hemorrhage. DevelopmentalMedicineand Child Neurology 29: 601-614.

Van Bel F, van de Bor M, Baan J and Ruys JH (1988): The influence of abnormal blood gases on cerebral blood flow velocity in the preterm newborn. Neuropediatrics 19: 27-32.

Vaughan RS, MenkeJA, and Giacoia GP (1978): Pneumothorax: A complication of endotrachealsuctioning.JournalofPaeditttrics 92: 633-634.

Versmold HT, KittermanJA, Phibbs RH, Gregory GA and Tooley WH (1981): Aortic blood pressure during the first 12 hours of life in infants with birth weight 610 to 4,220 grants. Pediatrics 67: 607-613.

Volpe JJ (1989): Intraventricular haemorrhage in the premature infant -current concepts. Part I. Annals of Neurology 25: 3-11.

Welch K (1980): The intracranial pressure in infants. Journal of Neurosurgery 52: 693-699.