SPECIALREPORT - BMJ

Archives ofDisease in Childhood 1992; 67: 1221-1227

SPECIAL REPORT

Development of audit measures and guidelines forgood practice in the management of neonatalrespiratory distress syndrome

Report of a Joint Working Group of the British Association of Perinatal Medicine and theResearch Unit of the Royal College of Physicians

Neonatal respiratory distress syndrome (RDS)is a condition of increasing respiratory distress,commencing at, or shortly after, birth andincreasing in severity until progressive resolu-tion occurs among the survivors, usuallybetween the 2nd to 4th day. It is due, at least inpart, to insufficiency of pulmonary surfactantand is mainly confined to preterm infants withimperfect cardiopulmonary adaptation toextrauterine life. In extremely immature infantssevere under development of respiratory archi-tecture contributes to the pathophysiology ofthe condition. The syndrome is manifest byrespiratory distress (cyanosis, tachypnoea,grunting, and recession) and respiratory failureis diagnosed by blood gas analysis. Diagnosis isusually confirmed by an x ray film (groundglass appearance and air bronchograms), butthese radiological features are not patho-gnomonic of RDS.

Correspondence to:Professor M I Levene,University Department ofPaediatrics and ChildHealth, D Floor,Clarendon Wing,The General Infirmaryat Leeds,Belmont Grove,Leeds LS2 9NS.

Aims ofthe working groupAudit of clinical practice requires an assess-ment of how personal practice measures upto certain defined standards. It is of course

necessary for these standards (or audit mea-sures) to be agreed beforehand by those con-ducting the audit. Agreement on appropriateaudit measures may be reached locally, but forcertain conditions such as neonatal RDS thereare advantages in using audit measures whichcan be applied nationally. This is a soundbasis for producing comparative data and itshould also theoretically lead to uniformimprovements in standards nationally.A Joint Working Group of the Research

Unit of the Royal College of Physicians andthe British Association of Perinatal Medicinedevised the present guidelines. Members ofthe working group are listed at the end of thepaper. The working group met on the 22ndNovember 1990. Background papers were cir-culated before the meeting so that there wassufficient time for discussion and for develop-ment of the proposed audit measures.Statements of good practice are, whereverpossible, supported by references to pub-lished trials of good design or meta-analyses.There were areas recognised where furtherresearch is required to clarify certain areas.

These are listed in appendix A. Points of

clinical practice suitable for audit are listed atthe end of this paper (appendix B). In someinstances the working group agreed that acourse of action was appropriate, even if therewas no published evidence to support it. Thesestatements are followed by a (C) for consen-sus. The purpose of this document is to de-velop audit measures for the prevention andmanagement of RDS. The audit measures rec-ommended reflect 'good practice' based on aconsensus view of the working group. It wasnot the aim of the working group to produce adetailed account of the 'best way' of managingor preventing RDS. There are too many areasof uncertainty and often little evidence to sug-gest that one approach to management is betterthan another.These guidelines are recommended for good

practice in all obstetric units and neonatalunits offering intensive care to babies withRDS.

Clinical guidelines are published as theymay help doctors by providing an analyticalframework for the evaluation and treatment ofsome common clinical problems. They arenot intended to replace a doctor's clinicaljudgment, and are not necessarily the onlyway in which a particular condition can bemanaged. They do, however, provide a frame-work within which audit and review of clinicalpractice can take place. The guidelines reflectthe views of the individual clinicians whoattended the workshops.

Guidelines for prenatal managementThe vast majority of babies who develop RDSdo so because they are born preterm. Pretermdelivery occurs either because of the unantici-pated and unplanned onset of preterm labour,or as the result of an obstetric decision tointervene, either because the mother is con-sidered to be at risk, or more commonly,because of the belief that the risks to the fetusof remaining in utero are greater than the risksof being born too soon.

Every obstetric unit must have a clear pro-tocol for the management of preterm labour.There have been no randomised controlledtrials comparing antenatal with neonatal refer-ral to centres that can provide intensive peri-natal care, but the working group's view is

1221

on March 20, 2022 by guest. P

rotected by copyright.http://adc.bm

j.com/

Arch D

is Child: first published as 10.1136/adc.67.10_S

pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

12oint Working Group of the British Association ofPerinatal Medicine and the Research Unit of the Royal College ofPhysicians

that delivery should occur in a hospital wherethere are obstetricians experienced in manag-ing the problems of preterm labour and deliv-ery and where staff are experienced in neona-tal resuscitation (C).

All obstetric units should have specificguidelines for antenatal transfers of womenwith pregnancies at above average risk. Theassessment of the extent of the risk to thefetus of remaining in utero requires the appro-priate application of antenatal investigativetechniques and this may necessitate in uterotransfer to tertiary referral units. On occasion,in utero transfer may be deemed to be unsafe,and therefore every district general hospitalshould have facilities available for the safemanagement of the labour, for safe delivery ofthe baby, and for adequate resuscitation andcare of the infant until the baby can be trans-ferred to a more appropriate centre. Eachregional health authority should designateregional or subregional units which can pro-vide expertise in the management of pretermlabour, transfer of the infant, and neonatalintensive care.

CERVICAL CERCLAGEIn some specific circumstances cervical cer-clage has been shown to reduce the risk ofvery preterm delivery. Women who have hadthree or more second trimester miscarriagesand/or preterm deliveries without obvious rea-son are particularly likely to benefit from cer-vical cerclage in a subsequent pregnancy.2This form of treatment has not, however,been shown to reduce the risk of RDS.2

ASYMPTOMATIC BACTERIURIAMeta-analysis of the 12 well controlled clini-cal trials indicate that antibiotic treatment ofasymptomatic bacteriuria reduces the risk ofpreterm delivery by about 40%.3 There are nopublished data demonstrating that this alsoreduces the risk of RDS. Routine screeningfor asymptomatic bacteriuria at a stage inpregnancy (that is 18-21 weeks) where appro-priate antibiotic treatment will be likely toprevent the infant from being born severelypreterm may be beneficial in the prevention ofRDS although this is not proved.

[-MIMETICSAn overview of 15 well controlled trials of5-mimetics in preterm labour demonstratesunequivocally that delivery can be delayed,and that this delay is reflected in a reductionin the incidence of preterm delivery and lowbirth weight.4 Unfortunately, there is no evi-dence that this is translated into a beneficialeffect in reducing the severity of neonatal res-piratory problems,4 although a moderate, andclinically important beneficial effect on theincidence of RDS cannot be ruled out. P-mimetic treatment in preterm labour seemsmost likely to be beneficial when time gainedbefore delivery can allow other effective mea-sures to be implemented. These include trans-

fer of the mother to a centre with appropriatefacilities for intensive care of the mother andbaby and/or the administration of antenatalsteroids (see below). The working groupconsiders that the evidence is such that P-mimetics be considered in preterm labour toallow other effective management to beundertaken. In view of rare serious side effectssuch as pulmonary oedema, contraindicationsinclude women with cardiac disease, hyper-thyroidism, or diabetes.

CORTICOSTEROIDSOverviews of the results of 12 controlled trialsprovide clear evidence that corticosteroidadministration to women at increased risk ofearly delivery significantly reduces the risk ofRDS by about 50%.5 The overviews clarifythat the reduction in the odds of developingthe syndrome is independent of gender andgestational age.5 The greatest benefit inrespect of RDS is seen when the time intervalbetween the start of treatment and delivery ismore than 24 hours and less than seven days.Nevertheless, some babies born before or afterthis optimal time interval still appear to derivebenefit. The place, if any, for examination ofamniotic fluid for an lecithin:sphingomyelin(L:S) ratio or equivalent test of surfactantproduction is unclear.The working group recommends that either

betamethasone or dexamethasone be giventwice a day for two days by intramuscularinjection to all women where the baby is likelyto be delivered before 32 weeks of gestation.High risk pregnancies beyond 32 weeks ofgestation may also benefit from antenatal cor-ticosteroids. We suggest that the same criteriabe used where there is prolonged rupture ofthe membranes, but in these cases the clini-cians must be particularly vigilant for any evi-dence of infection in both the mother and/orfetus. If infection is strongly suspected, anti-biotic treatment should be started rapidly anddelivery expedited.

Contraindications to antenatal cortico-steroids are maternal thyrotoxicosis, cardio-myopathy, and active maternal infection orchorioamnionitis. The effect of steroids on thefetuses of diabetic mothers is unclear, but wedo not believe there is a definite contra-indication in this particular case providingcare is taken to correct the effect of the corti-costeroid on the mother's blood glucose.

MODE OF DELIVERYDelivery by caesarean section may be associ-ated with a higher risk of RDS than vaginaldelivery.6 Where there is a choice in the modeof delivery, the working group believes thatin the absence of more reliable data fromproperly controlled trials, caesarean deliveryshould be the exception rather than the rule,because of the known risk to the mother (C).This should apply both to the fetus presentingby the breech as well as the fetus with cephalicpresentation. Where caesarean section is nec-essary, lower segment section may prove very

1222



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

Development of audit measures and guidelines for good practice in the management of neonatal respiratory distress syndrome

traumatic to the small infant and in such cir-cumstances ready recourse to upper uterinesegment incision should be considered, andthe implications of this discussed with themother beforehand.

Guidelines for resuscitationIt is most important that any baby at risk ofRDS should be born in a hospital that has thefacilities and expertise to resuscitate and sup-port the infant until transfer can be arrangedto a referral centre (C). All relevant medicaland midwifery staff should have a period offormal and practical training in neonatalresuscitation and this level of training shouldbe regularly maintained. This includes instruc-tion on endotracheal intubation for those witha regular responsibility for resuscitation aspart of their normal duty. Others should betrained in the safe use of a bag and mask.With reference to RDS, it is particularlyimportant to train staff in the resuscitation ofinfants of below 30 weeks' gestation. A writ-ten resuscitation protocol should be given toall relevant personnel.

Important goals for effective resuscitationinclude the need to avoid the infant becomingcold,78 the need to give active respiratory sup-port to an infant who is not breathing,9 10 tooxygenate the infant adequately,'1 and tomaintain an appropriate blood glucose con-centration.'2 13

ROUTINE INTUBATIONUnfortunately, there is a lack of fundamentaldata from physiological studies with which toformulate further guidelines for effectiveresuscitation of infants at risk of RDS.Although it is clear that the infant's airwayshould be stabilised from the first breath, fur-ther research needs to be done to establish thebest method for doing this. There is no goodevidence that babies (including very pre-mature ones) who are breathing effectively atbirth benefit from routine intubation andpositive pressure ventilation. Intubation shouldbe performed if the infant is cyanosed or mak-ing inadequate respiratory efforts.

If an infant requires intubation, the inspir-atory pressure should be high enough to appro-priately move the chest wall and these pressuresmay need to be as high as 30 cm H20 or more(C). In general, the lowest pressure which willeffectively move the chest should be chosen.

EARLY CONTINUOUS POSITIVE AIRWAYPRESSURE (CPAP)The role of early CPAP by either nasal prongor face mask immediately after resuscitation ofpremature infants at risk of RDS remains con-troversial and further controlled clinical trialsare required to resolve this.

Guidelines for ventilatory managementAll infants at risk of RDS should be closelymonitored for clinical and blood gas evidence

of respiratory failure. Each unit that under-takes long term (>24 hour) care of theseinfants should have at least one consultantpaediatrician with an up to date knowledge ofthe principles of mechanical ventilation andwho should be responsible for providing aclear respiratory management protocol for allthe staff working on that unit. Particularattention must be paid to the provision ofmodern neonatal ventilators (see below) andefficient humidification systems.

VENTILATION TECHNIQUESThere is no consensus view on the ventilatormanagement of RDS and two basic methodsexist which can be referred to as slow rate andfast rate. The slow rate method aims to startbabies on the ventilator at rates of 30-40breaths per minute (bpm) with peak inspira-tory pressures (PIP) of 20-25 cm H20 andinitial inspiratory to expiratory (I:E) ratio ofup to 1:1. If the baby shows deteriorating lungdisease and changes in the ventilator settingsare thought necessary, either increase of theI:E ratio (making the inspiratory time longerthan the expiratory) or increasing PIP may beappropriate. High inspired oxygen concentra-tions are used if required and the PIP shouldbe the lowest setting compatible with satisfac-tory blood gases.'4The fast rate method commences ventila-

tion at about 60 bpm and may increase to 120bpm if the baby is breathing at a faster ratethan the ventilator. The expiratory time shouldexceed the inspiratory time to prevent in-advertent alveolar overdistension, and inspira-tory times should be limited to a maximum of0 5 sec throughout the duration of mechanicalventilation.'5 There is no evidence to suggestthat the use of ventilation rates above 150bpm or oscillatory ventilation techniques areof benefit (C).

It is essential to use appropriate modernmechanical ventilators which are designed foruse in preterm infants. The role for patienttriggered ventilators is as yet not clear.

THE INFANT'S RESPIRATORY ACTIVITY'Fighting the ventilator' or breathing out ofphase with the mechanical ventilator is agreedto be a risk factor for a variety of complica-tions such as pneumothorax and intraven-tricular haemorrhage (C). It is probablyimpossible to recognise every baby with thispattern of respiratory interaction by simpletechniques, but it is apparent in many babiesby either close observation of the chest wall orlooking for excessive beat-to-beat variability(>10%) on examination of the blood pressuretrace measured directly from an indwellingarterial catheter/cannula (C). Hypovolaemiamust also be considered as the cause of exces-sive beat-to-beat variability in the trace.There is no proved policy for the manage-

ment of this problem, but the baby's need forventilation should be reviewed. If continuedmechanical ventilation is considered neces-sary, then sedation may reduce the babies res-

1223



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

1Joint Working Group of the British Association of Perinatal Medicine and the Research Unit of the Royal College ofPhysicians

piratory activity or a non-depolarising muscleblocker should be used. Alternatively the ven-tilator rate can be increased to 'capture' therespiratory rate of the baby to achieve syn-chrony between the baby's breathing and theventilator. If a non-depolarising muscle relax-ing agent is used, the ventilatory requirementsof the baby may change and this must beanticipated.

BLOOD GAS MONITORINGThere are two main reasons for monitoringblood gases. Firstly, as a guide to the need for,and the level of, ventilatory support andsecondly to minimise the risk of retinopathy ofprematurity. Unfortunately, there are noagreed guidelines as to what arterial oxygentension (Pao2) levels are safe in the preventionof this condition in very premature infants. Itis therefore impossible to state 'normal' limitsfor pH or blood gas variables. The frequentmonitoring of blood gases is, however, essen-tial during the acute stages of RDS to assessthe need for or effect of respiratory support.This is most reliably achieved through um-bilical artery catheterisation or by indwellingperipheral arterial cannulae. Monitoring ofoxygen by arterial sampling from an indwellingarterial catheter is the 'gold standard' mea-surement and continuous monitoring by acatheter tip oxygen sensor is optimal. Alter-natively, non-invasive methods such as the useof transcutaneous oxygen or carbon dioxidetension monitors or pulse oximetry should beavailable to detect trends, but should only beused in conjunction with blood sampling, bal-ancing the known hazards associated with pro-longed intravascular monitoring against theprobable (but ill defined) increased risks ofretinopathy of prematurity in very pretermbabies subjected to high preductal arterial oxy-gen pressures in the first weeks of life. Pulseoximetry may be a useful guide to oxygenationduring neonatal transport, but appropriatelevels of arterial oxygen saturation (Sao2) havenot yet been agreed. Acceptable levels for Sao2of 85-93% have been proposed, but errors inthe technique of measurement are potentiallygreat and oximetry cannot be used as the onlyform of monitoring arterial oxygen levels in theearly phases of RDS.There is agreement on the following blood

gas values (C):pH: Avoid arterial pH levels <7-2. When

cellular metabolic function is likely to be com-promised aim to maintain pH at 7-25 orabove.

Pao2: The recommended range is 6-10kPa. The lower acceptable limit of Pao2 in aninfant with RDS may be lower than this pro-vided oxygen delivery to the tissues is ade-quate as judged by haematocrit, peripheralperfusion, and base excess.

Paco2: More important than the Paco2level is the pH and in general terms if this ismaintained above 7-25 then the Paco2 isprobably acceptable. Unless there is a specificreason for inducing hypocarbia the lower limitof Paco2 should be maintained above 5 kPA.

PHYSIOTHERAPYThere is no evidence to support the practice ofroutine physiotherapy and suction especially inthe first 24 hours of life in an infant with RDSand these procedures may be hazardous (C).

ENDOTRACHEAL TUBESThere is no definite evidence that shoulderedendotracheal tubes are associated with eithergreater or fewer complications than straighttubes.

Guidelines for the use of surfactanttreatmentThe use of surfactant treatment in RDS hasbeen shown to reduce the risk of death fromthe disease. Meta-analyses of 34 randomisedcontrolled trials of surfactant replacementcomprising over 6000 babies show significantreductions in the risk of neonatal death bothwhen the surfactant is given early (prophy-lactically) and when used for established RDS(rescue treatment).16 These analyses showconclusively that surfactant treatment alsoreduces the risk of pneumothorax.

It is not yet clear whether prophylacticadministration has any advantage over rescuetreatment and this issue is currently beingaddressed in randomised controlled trials.Pending the results of these studies, the work-ing group recommends that surfactant shouldbe given as rescue treatment because this isless expensive. There is currently a majorstudy being conducted on costing of surfac-tant throughout the UK and more accurateinformation will shortly become available.

Further research must be undertaken toestablish whether synthetic or natural forms ofsurfactant are more effective, and theirefficacy in infants below 26 weeks' gestation.17A further unresolved question is whether sur-factant is cost effective in infants over 31weeks' gestation particularly when given pro-phylactically.18 19 The answers to these ques-tions will be obtained by further well designedcontrolled studies and participation of indi-vidual units in these trials is to be encouraged.

Surfactant should be given by practitionerswho are aware of the short and long term risksof the treatment (related to rapid changes inboth respiratory and cardiovascular function)so that rapid changes in physiological statecan be detected and acted upon in a neonatalintensive care environment.

Complications of RDS and guidelines fortheir preventionThe presence of RDS increases the risk ofcomplications in other systems includingbrain (periventricular haemorrhage and/orleucomalacia), bowel (necrotising entero-colitis) and circulation (hypotension, patentductus arteriosus). The working groupbelieves that the risk of at least some of thesecomplications may be reduced by ensuringgentle delivery, rapid and effective resusci-tation, avoidance of hypothermia, hypo-

1224



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

Development of audit measures and guidelines for good practice in the management of neonatal respiratory distress syndrome

glycaemia, and hypoxaemia (C). Acidosisshould be investigated in order to identifyunderlying causes which may need treatmentin their own rights. Severe acidosis should becorrected. A bolus injection of sodium bicar-bonate may cause rapid changes in serumosmolality and is to be avoided. Slow correc-tion, over an hour or two, is usually preferable.

CARDIOVASCULAR STABILISATIONIt is essential to monitor blood pressure sothat hypotension can be promptly recognised,its cause assessed, and appropriate treatmentoffered. Facilities for intravascular blood pres-sure monitoring should be available, but non-invasive blood pressure measurement usingthe Doppler technique may give a reliableestimate of systolic pressure. Further studiesof the normal range of blood pressure in verypremature infants are needed, but at the pre-sent time the working group agrees that amean arterial blood pressure equivalent to thegestational age in weeks is adequate as a mini-mum value (C).Hypotension should be treated initially with

colloid or blood if there is the possibility ofhypovolaemia (C). The effectiveness ofinotrope infusion in the preterm newborn hasnot been proved, but a starting dose ofdopamine 10 ,ug/kg/min may be needed in thepreterm neonate as they are relatively resistantto this form of treatment. Its use must beavoided in the presence of hypovolaemia.

BRAINPreterm cerebral injury (periventricular haem-orrhage or leucomalacia) is to some extentrelated to perturbations of cerebral haemo-dynamics.20 In particular, prevention of theinfant fighting the ventilator may reduce therisk of periventricular haemorrhage,21 but fur-ther studies are necessary to assess the effectsof sedation and muscle relaxation both on thereduction in frequency of intracranial lesionsand their effects on the cerebral circulation.

DELAYED CLOSURE OF THE DUCTUSARTERIOSUSWhen the infant is requiring mechanical venti-lation for RDS and there is evidence of apatent ductus arteriosus with significant left toright shunt (diagnosed either clinically or withultrasound), indomethacin should be givenand preferably before congestive cardiac fail-ure occurs.22 23 Treatment should also be con-sidered for non-ventilated infants requiringoxygen who show signs of a large left to rightshunt diagnosed clinically or with ultrasound.Indomethacin given intravenously may be

associated with reduction in blood flow tobrain,24 25 kidneys and bowel,26 27 but thisdoes not appear to cause clinical problems.The optimal dosage schedule is not clear andthis requires further clinical studies. Theworking group believes that either three dosesof 0-1 mg/kg every 12 hours or 0-2 mg/kg/dayfor three days is effective (C). A platelet

count of <100x109/l, significant coagulationimpairment, and renal dysfunction are impor-tant (but not necessarily absolute) contra-indications to indomethacin treatment. Ifclosure of a significant ductus arteriosus hasnot been achieved medically surgical ligationby an experienced surgeon should be con-sidered (C).

NECROTISING ENTEROCOLITISThere are few objective data to suggest thatthe risk of necrotising enterocolitis can bereduced by alterations in the management ofRDS. The working group agrees that the riskof necrotising enterocolitis can be minimisedby avoidance, or rapid treatment, of shockand by the practice of general measures tomaintain homoeostasis (C).Adequate nutrition is an important part of

the management of RDS (C). Facilities fortotal parenteral nutrition must be available,but minimal enteral feeding should be con-sidered in infants with stable or improvingRDS (C). There is also no evidence thatuncomplicated umbilical catheterisation in-creases the risk of necrotising enterocolitis.

Chronic lung diseaseThis has been defined as the requirement ofsupplementary oxygen after 28 days frombirth.28 It has been shown that an additionaloxygen requirement in a prematurely borninfant after 36 weeks' postmenstrual age is abetter predictor of severe pulmonary out-come.29 The majority of these infants will havebronchopulmonary dysplasia. Steroids reducethe duration of mechanical ventilation inpreterm infants and should be given to thosewho have not improved by two weeks afterbirth.3034 This benefit outweighs any potentialside effects (C). The dose and duration ofsteroid treatment are unclear, but it shouldinitially be given as a short course and con-tinued if a definite response is seen (C).Dosage reduction can then be titrated againstclinical progress. The role of steroid treatmentin non-ventilator dependent babies who remainin oxygen is not established. The place of eitherearly steroid administration or vitamin A pro-phylaxis for this condition is unproved.There is no evidence that other treatments

for bronchopulmonary dysplasia includingbronchodilator treatment and diuretics have adefinite beneficial effect. Diuretics are indi-cated for episodes of cardiac failure associatedwith bronchopulmonary dysplasia (C), buttheir effectiveness as long term treatment isuncertain. If diuretics are used, considerationshould be given to the use of a calcium spar-ing regimen. Antibiotics should only be usedfor active infection (C).Treatment of babies with chronic lung

disease requires facilities for long term moni-toring of oxygenation state (C). Non-invasiveoxygen saturation monitoring is probablymost accurate, but at present firm recom-mendations for the optimal range of satu-ration cannot be agreed.

1225



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

1Joint Working Group of the British Association of Perinatal Medicine and the Research Unit of the Royal College ofPhysicians

Appendix A: Areas for further researchThe working group identified a number ofareas where further controlled clinical studiesor more basic research at a physiological levelmay considerably enhance our management ofthe infant with RDS. The major areas aresummarised here.

(1) Does cervical cerclage in women at highrisk of preterm labour reduce the overall riskof RDS in their infants?

(2) Does routine screening for asymp-tomatic bacteriuria at 18-21 weeks andappropriate antibiotic treatment reduce theoverall risk of RDS in infants of thesewomen?

(3) Does supression of preterm labour withP-mimetics result in a reduction of risk ofRDS?

(4) Does examination of the amniotic fluidfor L:S ratio (or equivalent test of surfactantproduction) help in the prevention of RDS inpreterm infants?

(5) Are antenatal steroids of benefit inreducing RDS and its complications in infantsover 32 weeks of gestation?

(6) Is there a benefit of antenatal corticos-teroids in diabetic women with pretermlabour compared with the potential risks tomother and infant?

(7) What are the physiological effects of dif-ferent methods of resuscitation on the imma-ture infant and do these methods help toreduce the severity of RDS?

(8) What is the best method for the earlyphysiological stabilisation of the airway andlungs in preterm infants at risk of RDS?

(9) What is the role of early CPAP immedi-ately after resuscitation of preterm infants anddoes this reduce the risk of severity of RDS?

(10) What is the role of patient triggeredventilators in the management of preterminfants with RDS?

(11) In infants who 'fight the ventilator',does matching the ventilator rate to that of thebaby effectively resolve this condition? Is thisa feasible method of clinical management inpreterm infants?

(12) What are safe levels of Sao2 in infantswith RDS?

(13) What are the effects of different meth-ods of ventilatory support on cerebral oxy-genation and haemodynamics?

(14) Are straight endotracheal tubes ofbenefit compared with shouldered tubes inpreterm ventilated infants?

(15) Is prophylactic surfactant therapy inpreterm infants at risk ofRDS cost-effective?

(16) What is the most effective and safe sur-factant for use in preterm infants?

(17) Is surfactant cost effective in infantsover 31 weeks of gestation and below 26weeks of gestation?

(18) What is the normal range of bloodpressure measurements in preterm infantsover a wide range of gestational ages?

(19) Is inotrope infusion of benefit inhypotensive infants with RDS? What is theoptimal dosage regimen?

(20) What is the effect of sedation and/ormuscle relaxation on the cerebral circulation?

(21) What is the optimal dosage regimenfor indomethacin use in the management ofpatent ductus arteriosus in preterm infantswith RDS?

(22) What is the optimal dosage regimen ofcorticosteroids in the management of infantswith chronic lung disease after RDS?

(23) What is the indication for corti-costeroid use in non-ventilator dependentbabies who require prolonged oxygen treat-ment after RDS?

(24) What is the role of early corticosteroidtreatment in the management of infants withRDS who are at risk of chronic lung disease?

(25) What is the role of other treatmentssuch as vitamin A, diuretics, and broncho-dilators in the prevention or management ofchronic lung disease?

(26) What are the safe levels of Sao2 ininfants with chronic lung disease?

Appendix B: Proposed audit measuresOBSTETRIC UNITS

(1) Does the obstetric unit have a clearwritten protocol for the management ofpreterm labour?

(2) Does the obstetric unit have writtenguidelines for deciding on in utero transfer ofat-risk pregnancies?

(3) Does the obstetric unit have a writtenpolicy for evaluating every woman in pretermlabour for risk/benefit of (a) n-mimetictocolytic treatment and (b) antenatal cortico-steroids?

(4) Does the obstetric unit have a policy todiscuss with the mother requiring caesareansection the implications for upper uterinesegment caesarean section should this provenecessary?

(5) Is a formal and practical resuscitationtraining programme for the prematurely borninfant in place in every labour ward? Howoften are staff retrained?

(6) Should a baby be born preterm and un-expectedly in a unit unable to offer long termneonatal intensive care, does the unit haveappropriate facilities and staff skilled in look-ing after that baby until a team arrives fromthe hospital of referral? How many staff in anyunit giving care to premature infants havereceived English National Board training inneonatal intensive care?

(7) Was the body temperature of a preterminfant above 36°C on arrival in the neonatalunit?

NEONATAL INTENSIVE CARE UNITS(8) Does every neonatal unit undertaking

intensive care have one consultant responsiblefor establishing a protocol for neonatalmechanical ventilation? Is there a writtenpolicy?

(9) Are the mechanical ventilators used forinfants with RDS designed for this use? Dothe staff understand the principles whichgovern the functioning of ventilators?

(10) Is there a written policy of acceptable

1226



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

Development of audit measures and guidelines for good practice in the management of neonatal respiratory distress syndrome 1227

limits for blood gas and pH variables forpreterm babies with RDS? Is there a writtenpolicy for acceptable Sao2 ranges?

(11) If exogenous surfactant is being used,have the staff on the newborn intensive careunit been instructed about the rapid respira-tory and cardiovascular changes induced bysome of these agents? Is the unit appropriatelystaffed and equipped to monitor thesechanges and respond accordingly?

(12) Does the unit have a written policygoverning the use of indwelling arterialcatheters including their supervision, so thatischaemic complications can be promptlyrecognised and treated?

(13) Are facilities for dispensing and moni-toring a total perenteral nutrition regimenavailable for infants with severe RDS?

(14) If an infant with RDS is ventilatordependent, with radiological evidence ofchronic lung disease and dependent on addi-tional oxygen at 2 weeks, are steroids con-sidered?

Background papers* Prenatal care to reduce the risk of neonatalidiopathic respiratory distress syndrome:Dr Adrian Grant.

* Resuscitation of the premature infant: DrDavid Field.

* Surfactant therapy: Dr Henry Halliday.* The complications of RDS and their pre-vention: Dr Janet Rennie.

* Chronic lung disease: Dr Rosamond Jones.

The background papers can be obtainedfrom the Publications Department of theRoyal College of Physicians, 11 St AndrewsPlace, London NW1 4LE. The cost is C5 tocover photocopying and postage.

Members of the working group: Malcolm Levene (co-chairman, professor of paediatrics), Malcolm Chiswick (co-chairman, consultant paediatrician), David Field (consultantpaediatrician), Stewart Forsyth (consultant paediatrician).Harold Gamsu (consultant neonatologist) Adrian Grant(Director, Perinatal Trials Service) Anne Greenough (consul-tant paediatrician), Henry Halliday (consultant neonatologist),Judith Hetherington (clinical nurse adviser), Edmund Hey(consultant paediatrician), Anthony Hopkins (Director,Research Unit of the Royal College of Physicians), RosamondJones (consultant paediatrician), David Lloyd (consultantneonatologist), Una MacFadyen (consultant paediatrician),Anthony Milner (professor of paediatrics), Janet Rennie (con-sultant paediatrician), Osmund Reynolds (professor of neo-natal paediatrics), Michael Silverman (consultant paedia-trician), Paul Vinall (consultant obstetrician), MichaelWeindling (consultant paediatrician).

1 Royal College of Physicians. Medical care of the newborn inEngland and Wales. A report of the Royal College ofPhysicians. London: Royal College of Physicians, 1988.

2 MRC/RCOG Working Party on Cervical Cerclage. Interimreport of the Medical Research Council/Royal College ofObstetricians and Gynaecologists multicentre trial of cer-vical cerclage. BrJ7 Obstet Gynaecol 1988;95:437-45.

3 Smaill F. Antibiotics versus no treatment for asymptomaticbacteriuria in pregnancy. In: Chalmers I, ed. Oxforddatabase of perinatal trials. Version 1.2, Disk issue 4,August 1990, Record 3170.

4 King JF, Grant A, Keirse MJNC, Chalmers I. Beta-mimetics in preterm labour: an overview of the random-ized controlled trials. Br J Obstet Gynaecol 1988; 95:211-22.

5 Crowley P, Chalmers I, Keirse MJNC. The effects of corti-

costeroid administration before preterm delivery: anoverview of the evidence from controlled trials. Br JfObstet Gynaecol 1990;96:11-26.

6 Fedrick J, Butler NR. Certain causes of neonatal death. I.Hyaline membranes. Biol Neonate 1970;15: 229-55.

7 Day RL, Callguiri L, Kamenski C, Erhlich F. Body tem-perature and survival of preterm infants. Pediatrics1964;34:171-81.

8 Hey E. Thermal neutrality. Br Med Bull 1975;31:69-74.9 Dawes GS, Mott JC. The vascular tone of the foetal lung.

JPhysiol (Lond) 1962;164:465-77.10 Adamson K, Behrman R, Dawes GS, James LS, Koford C.

Resuscitation by positive pressure ventilation and trishydroxymethylaminomethane of Rhesus monkeysasphyxiated at birth. J Pediatr 1964;65:807-18.

11 Strang LB, MacLeish MH. Ventilatory failure and right toleft shunt in newborn infants with respiratory distress.Pediatrics 1961;28:17-27.

12 Dawes GS, Jacobson HN, Mott JC, Shelly HI, Stafford A.The treatment of asphyxiated mature foetal lambs andRhesus monkeys with intravenous glucose and sodiumbicarbonate. J Physiol (Lond) 1963;169: 167-84.

13 Adamson K, Behrman R, Dawes GS, Dawkins MJR,James LS, Ross BB. The treatment of acidosis with alkaliand glucose during asphyxia in foetal Rhesus monkeys.Jf Physiol (Lond) 1963;169:679-89.

14 Reynolds EOR. Ventilator therapy. In: Thibault DW,Gregory GA, eds. Neonatal respiratory care. Menlo Park:Addison Wesley, 1979:217-36.

15 Greenough A, Milner AD. High frequency ventilation inthe neonatal period. EurJ Pediatr 1987:146:446-9.

16 Soll RF. Overviews of surfactant trials. In:Chalmers I, ed.Oxford database of perinatal trials. Version 1.2, disk issue5, February 1991, records 5206, 5207, 5252, 5253.

17 Morley CJ. Surfactant treatment for premature babies-areview of clinical trials. Arch Dis Child 1991 ;66:445-50.

18 Tubman TRJ, Halliday HC, Normand C. Cost of surfac-tant replacement treatment for severe neonatal respira-tory distress syndrome: a randomised controlled trial.BMJ. 1990;301:842-5.

19 Mugford M, Piercy J, Chalmers I. Cost implications ofdifferent approaches to the prevention of respiratory dis-tress syndrome. Arch Dis Child 199 1;66:757-64.

20 Perlman JM, McMenamin JB, Volpe nJ. Fluctuating cere-bral blood flow velocity in respiratory distress syndrome.Relation to the development of intraventricular hemor-rhage. NEnglJMed 1983;309:204-9.

21 Perlman JM, Goodman S, Kreusser KL, Volpe JJ.Reduction in intraventricular hemorrhage by eliminationof fluctuating cerebral blood flow velocity in preterminfants with respiratory distress syndrome. N Engl Jf Med1985;312:1353-7.

22 Merritt TA, Harris JP, Roglemann K, et al. Early closure ofthe patent ductus in very low birthweight infants: a con-trolled trial. J Pediatr 198 1;99:281-6.

23 Mahoney L, Carnero V, Brett C, Haymann MA, ClymanRI. Prophylactic indomethacin therapy for patent ductusarteriousus in very low birthweight infants. N Engl Jf Med1982;306:506-10.

24 Evans DH, Levene MI, Archer LNJ. Effect ofindomethacin on cerebral blood flow velocity in prema-ture infants. Dev Med Child Neurol 1987;29:776-82.

25 Pryds 0, Greisen G, Johansen KH. Indomethacin andcerebral blood flow in premature infants for patent duc-tus arteriosus. EurJPediatr 1987;147:315-6.

26 van Bel F. Zoeren DV, Scipper J, Guit GL, Baan J. Effectof indomethacin on superior mesenteric artery bloodflow velocity in preterm infants. J Pediatr 1990;116:965-70.

27 Wong SN, Lo RN, Hui PW. Abnormal renal and splanch-nic arterial Doppler path in premature babies with symp-tomtic PDA. J7 Ultrasound Med 1990;9: 125-30.

28 Avery ME, Tooley WH, Keller JB, et al. Is chronic lungdisease in low birth weight infants preventable? A surveyof eight centers. Pediatrics 1987;79:26-30.

29 Shennan AT, Dunn MS, Ohlsson A, Lennox K, HoskinsEM. Abnormal pulmonary outcome in premature infants:prediction from oxygen requirement in the neonatal peri-od. Pediatrics 1988;82:527-32.

30 Mammel MC, Green TP, Johnson DE, Thompson TR.Controlled trial of dexamethasone therapy ininfants with bronchopulmonary dysplasia. Lancet 1983;i: 1356-8.

31 Avery GB, Fletcher AB, Kaplan M, Brudno DS.Controlled trial of dexamethasone in respirator-depen-dent infants with bronchopulmonary dysplasia. Pediatrics1985;75:106-1 1.

32 Cumming Jj, D'Eugenio DB, Gross SJ. A controlled trial ofdexamethasone in preterm infants at high risk forbronchopuhmonary dysplasia. N EngI J Med 1989;320:1505-10.

33 Harkavy KL, Scanlon JW, Chowdhry PK, Grylack LJ.Dexamethasone therapy for chronic lung disease in venti-lator- and oxygen-dependent infants: a controlled trial.JPediatr 1989;115:979-83.

34 Collaborative dexamnethasone trial group. Collaborativerandomised trial of dexamethasone in neonatal chroniclung disease: an international placebo-controlled trial.Pediatrics (in press).



j.com/

Arch D


pec_No.1221 on 1 O

ctober 1992. Dow

nloaded from

http://adc.bmj.com/

Documents

SPECIALREPORT - BMJ