key pointsnewborn screening on pulmonary health in cystic fibrosis has yet to be proven. ......

Key points

• Lung disease remains the major cause of morbidity and mortality in cystic fibrosis

• Despite clear improvements in outcomes such as nutrition, the benefit of early diagnosis by newborn screening on pulmonary health in cystic fibrosis has yet to be proven.

• Early lung disease is challenging to detect and monitor in young children with cystic fibrosis, but there is an urgent requirement to define useful outcome measures in this population, particularly as new treatments emerge.

• Many techniques are available, but difficulties with measurement or interpretation must be recognised when appraising past studies, considering their use in practice or as endpoints in clinical trials.

• There is no superior monitoring test at present, but emerging evidence from longitudinal studies in children with cystic fibrosis will help define which are most useful.

Monitoring early lung diseasein cystic fibrosis: where arewe now?

Educational aimsN To understand which techniques are available to monitor early lung disease in cystic

fibrosis, recognise difficulties inherent to each measure, and describe the use of testsin clinical practice or interventional trials.

SummaryIt is a particular challenge to detect and monitor lung disease in children youngerthan 6 years old. Various methods exist that can reveal abnormality in youngchildren, but each technique presents difficulties on performing the test orinterpreting results. Most children with cystic fibrosis are now diagnosed shortlyafter birth, but it is still unclear how to evaluate new therapies, or at what age toinitiate them. This review summarises current options for monitoring earlydisease, limitations of individual techniques and how evidence to date influencestheir use in research and clinical practice.

Introduction

Significant advances have been made incystic fibrosis (CF) care in the last decade.Earlier diagnosis, care in specialist centresand more intensive therapy have contributedto an increased median age of survival from23 years in 2002 [1] to 43.5 years in 2012 [2]in the UK. Newborn screening has enabledintervention shortly after birth, and considerable

improvement has been seen in outcomes such

as nutritional status [3]. The impact on pulmo-

nary health however is less striking [4]. Pro-

gressive lung damage may be delayed but

inevitably still occurs and remains the major

cause of morbidity and mortality in CF.

Children with CF often appear well duringthe early years, but there is compel-ling evidence that pulmonary inflammation,infection and structural changes take hold

Statement of InterestNone declared.

ERS 2014

HERMES syllabus link:modules B.15.1

Breathe | March 2014 | Volume 10 | No 1 35DOI: 10.1183/20734735.010813

Julie A. Duncan,

Paul Aurora

Portex Unit of RespiratoryPhysiology and Medicine,UCL Institute of ChildHealth, London, UK

Julie A. Duncan: Portex Unitof Respiratory Physiologyand Medicine, UCL Instituteof Child Health, 30Guildford Street, LondonWC1N 1EH, UK

[email protected]

alarmingly soon after diagnosis [5–7]. What ismost concerning is that there are often noeasily detectable markers of disease andby the time symptoms develop, irreversibledamage may have occurred. Early infectionand inflammation also seem to impact onfuture health, with evidence that lung func-tion does not return to baseline after anexacerbation, even in infancy [8].

Earlier studies of monitoring measuresare based on children diagnosed clinicallywith CF. Prompt diagnosis in newbornscreened infants now offers a unique oppor-tunity to determine those who get earlierdisease, identifying candidates for moreaggressive treatment or interventional trials.This becomes increasingly important astreatments targeting the basic defect in thecystic fibrosis transmembrane conductanceregulator (CFTR) protein are developed.Which markers are most informative remainsan area of contention and investigation ininfants and young children involves practical,ethical and regulatory difficulties [9]. Thisreview focuses on the most promisingmethods to detect lung disease in childrenwith CF, and what must be considered whenevaluating their use (table 1).

Infant lung functiontechniques

Infant lung function testing is a noninvasiveoption when monitoring CF lung diseasethroughout infancy. A number of methodsand measures are available. The two that givethe most useful information in early CFdisease are the raised volume rapid thoraco-abdominal compression (RVRTC) technique(fig. 1) and full body plethysmography, bothnormally performed under sedation. In theRVRTC technique [10], the lungs are inflatedtowards total lung capacity with three to fivebreaths until an inspiratory pause is produced.

A vest wrapped around the infant’s chest isquickly inflated and forces expiration toresidual volume [11], giving measures of forcedexpiratory flow and volume. The forced expiredvolume in 0.5 seconds (FEV0.5) is normallyreported as the lungs may empty in infantsbefore 1 second has passed.

Lung volumes are measured in a plethys-mograph when an infant breathes against aclosed shutter [12]. Changes in pressure andvolume inside the closed system are used tocalculate unknown volumes of the lungs.Functional residual capacity (FRC), the volumeof gas left in the lungs after a tidal expiration,is the most commonly reported measure.

Commercial devices are available for themeasurement of infant lung function, andstandards for testing are available for both theRVRTC technique [13] and plethysmography [12].Important considerations in any measurementof lung function are feasibility, reproducibilityand choice of reference data to interpret results.In a recent 10-centre study of infants with CF,acceptability rates were relatively high overall, at72% for RVRTC measures and 89% forplethysmographic FRC [14]. However, despitemeticulous training and quality control, feas-ibility varied significantly between sites depend-ing on experience. The reproducibility of thesetests is summarised in table 2.

Data on longer-term reproducibility arescarce as repeated test occasions, and there-fore sedation, of the same infant would berequired. The need for sedation also meansthat reference data from healthy infants arelimited. The Jones reference equations arefrequently used for RVRTC values, but arederived from only 155 infants measured acrosstwo centres using noncommercial equipment[16]. These equations have been shownto grossly overestimate abnormality whenapplied to data measured on a differentcommercially made device [17]. Equations forFRCpleth have recently been published from 153infants [18], but again should be consideredequipment specific. Use of contemporaneoushealthy controls or equipment specific equa-tions in studies is essential for meaningfulinterpretation of data [19] and important toappraise in research reports.

Studies of infant lung function inchildren with CF

Two main groups have described the evolutionof lung function, alongside other measures,

Table 1. Monitoring methods in earlydisease

N Clinical / scoring systemsN Lung functionN ImagingN Bronchoscopy and lavageN Cough swab/sputum

Monitoring early lung disease in cystic fibrosis

Breathe | March 2014 | Volume 10 | No 136

in infants with CF: the AREST-CF (AustralianRespiratory Early Surveillance in CF) groupand the LCFC (London Cystic FibrosisCollaboration). They have both found thatmeasures from RVRTC and plethysmographydistinguish patients with CF as a group fromhealthy infants [5, 14, 20–23], but there isconsiderable overlap in individuals. The age atwhich lung function becomes abnormal andits evolution in infancy varies between groups.In the AREST-CF cohort, lung function wasinitially reported as normal in the first sixmonths of life [20], but was found to beabnormal before this age in a subsequentstudy [24]. A significant decline of lungfunction in infancy was found despite earlydiagnosis and treatment, with FEV0.5 2 z-scores below the mean by one year. By 2 yearsthis deficit was even more abnormal at -4 z-scores, despite most infants being asympto-matic of disease. In the newborn screenedinfants of the LCFC cohort, lung function wasabnormal by 3 months of age [5]. In contrast tothe AREST-CF findings, lung function eitherimproved or remained stable at 1- and 2-yearfollowup visits [25, 26]. Importantly thisdiffered from earlier reports in clinicallydiagnosed infants with CF, for whom abnor-mal lung function progressed after diagnosis[22], suggesting that earlier instigation oftreatment in newborn screened infants mayhelp to preserve lung function in infancy. It isunclear why reports differ so greatly betweenthese two similar cohorts of children. Differenttreatment protocols could have an effect. Itmay also be explained in part by the use ofreference data to interpret results. The AREST-CF group used historical control data derivedfrom children tested on different equipment tocalculate z-scores; whereas the London groupstudied healthy control subjects tested underthe same conditions and derived equipmentspecific reference data [27].

Regardless of mode of diagnosis, abnor-mal infant lung function tests correlate withstructural lung changes on computed tomo-graphy (CT) [28] and inflammation detectedby bronchoalveolar lavage (BAL) [24, 29]. Agreater decline in lung function is seen inthose infected with respiratory pathogens[24, 30, 31], providing evidence for aggressivetreatment of infection in infancy.

An important question is whether infantlung function measures can predict those inolder children, and identify an ‘‘at risk’’ groupthat can be targeted for intervention.

In clinically diagnosed infants with CF,abnormal results predicted deficits in lungfunction later in infancy, at preschool age andat school age [22, 25, 30, 32, 33]. Data innewborn screened infants are scarce, withone study reporting lung function in infancyand pre-school spirometry [31]. In this study,no clear longitudinal association was seen,and there were significant inter-individualdifferences in tracking of lung function.

Infant lung function measures have beenused in interventional studies, but numbersof participants are often small. In 11 infants,lung function was shown to improve afterantibiotic treatment of a respiratory exacer-bation [8], showing potential to detect changewith treatment. RVRTC measures havebeen used in the Infant Study of InhaledHypertonic Saline in Cystic Fibrosis (ISIS)trial, with FEV0.5 showing a small improve-ment with treatment [34]. Large numbers ofinfants are likely to be needed if infant lungfunction is to be used as a meaningfuloutcome measure. Data from a recent multi-centre observational study estimated that 150infants would be needed in each treatmentarm to sufficiently power an interventionaltrial to detect a significant treatment effect[35]. Other considerations when using infantlung function techniques are the need forsedation, time-consuming procedures, intensetraining of staff and expensive equipment [27].It currently remains a measure confined tospecialist centres, and use in clinical practice

Figure 1The raised volume technique in a sedated infant. Figure reproduced with permissionfrom J. Stocks.


Breathe | March 2014 | Volume 10 | No 1 37

or as an outcome measure is limited forthese reasons.

Pre-school spirometry

With regards to lung function, the timebetween ages 2 and 5 years are often referredto as ‘‘silent’’; in which infant lung functiontesting is no longer feasible and childrencannot reliably perform conventional spiro-metry. However, in an age-appropriate,relaxed and enthusiastic environment, areproducible forced expiratory manoeuvrecan be obtained in a preschool child.Training with games using bubbles or whis-tles and computerised incentives duringtesting can help achieve maximal peak flowand vital capacity (fig. 2) [36].

Many commercial devices are available tomeasure pre-school spirometry. Guidelinesfor testing and modified quality controlcriteria are available [37, 38] and are sum-marised in table 3.

Quality control should include real-timeinspection of flow–volume and volume–timetraces. A full expiration may be reachedbefore one second in some children, soFEV0.5 and FEV0.75 should be recorded inaddition to FEV1 [38]. The largest FEVt andFVC should be reported even if not takenfrom the same manoeuvre, and forcedexpiratory flows are taken from the man-oeuvre with the largest sum of FEV0.5 andFVC [37]. In an experienced laboratory,acceptable spirometry can be obtained inover 75% of pre-school children, and successrates increase with age [39, 40]. Between-testreproducibility is good for FEV0.5, FEV1 andFVC (intraclass correlation coefficient (ICC)0.85–0.92) and moderate for FEF25–75 (ICC0.60) [39].

Current software programs for commer-cial devices use a combination of referenceequations to cover age ranges. This may leadto over- or under-diagnosis of abnormality oncertain test occasions [41] with obviousclinical consequences. The recently publishedGlobal Lung Function Intitiative ‘‘all age’’equations are based on measurements ofover 4000 children, span from 3 to 95 yearsand consider ethnic group [42]. They covertransition from child to adulthood moresmoothly to aid accurate interpretation ofresults. The range (or standard deviation) ofnormality varies greatly with age, especially inyounger children, and standard z-scoresshould be reported rather than percent pre-dicted values [35] to avoid misinterpretation.

Previous studies of spirometry inchildren with CF

Spirometric values are on average lower inpreschool children with CF when comparedwith controls, but only a small proportion (9–36%) have an FEVt below the normal range[32, 36, 40, 43–45] as disease is often mild.FEVt has also been reported as normal whenother markers indicate disease [40, 46] and istherefore relatively insensitive to abnormalityin this age group. Forced expiratory flows aremore reduced than volumes [32, 44] and maybe a more useful measure. When comparedwith other measures of lung function at pre-school age, a multicentre study showed thatspirometry was more sensitive in detectingabnormality in CF than inductance plethys-mography or forced oscillation [39].

With regards to predicting lung functionin later life, most data are available inclinically diagnosed children. Two studieshave reported longitudinal data from infancy[24, 30]. Abnormalities appear to track from

Table 2. Reproducibility of infant lung function measures

Measure Within test CV# 1 month between test ICC#

RVRTC FEVt" 3.4–4.1% 0.93

RVRTC FEF" 7.8–8.9% 0.85

FRCpleth" 2.3% 0.93

#: Values are expressed as coefficients of variation (CV), the extent of variability in relation to the mean of the population; and intraclass correlationcoefficient (ICC), the level of agreement between tests; ": FEVt: forced expired volume in t seconds, where t 5 0.5, 0.75 or 1 second. FEF: forcedexpiratory flow; FRCpleth: plethysmographic functional residual capacity. Data are summarised from [15].



pre-school to school age, and worsen withage [33]. Data in new-born screened childrenare awaited.

The familiarity of clinicians with para-meters reported in spirometry makes itan attractive measure for use in practice.If performed accurately in an experiencedcentre, spirometry is repeatable and maybe used to monitor progress over time orwith intervention. However its use is limitedby insensitivity to very early changes inCF lung disease, as large numbers wouldbe required to detect any significant treat-ment effect.

Inert gas washout

Inert gas washout was first described in the1940s and examines gas mixing in the lung[47]. In young children, multiple- (rather thansingle-) breath washout (MBW) is commonlyused as it can be performed during tidalbreathing when an infant is sedated, or pre-schooler distracted by watching television.The technique involves recording the declinein concentration (washout) of an inert tracergas from the lungs. Tracer gases must havelow solubility in the blood and not participatein gas exchange, and those commonly usedare sulphur hexafluoride (SF6), helium andresident nitrogen. With SF6 or helium, theinert gas is washed into the lungs, i.e.,inspired until an equilibrium concentrationis reached (fig. 3). The gas supply is thendisconnected at end-expiration and the wash-out recorded until an end concentration ofthe inert gas is reached, commonly 1/40th ofthe starting concentration. Nitrogen washoutis performed in a similar way, but no wash-inperiod is required as resident nitrogen can bewashed out by inspiring 100% oxygen. Amouthpiece is used in older subjects, but inyoung children a mask sealed with putty isapplied to the face to ensure no leak of gasduring the washout and minimise dead spaceof the mask.

Functional residual capacity can be calcu-lated from the washout (fig. 4). The initialconcentration of tracer gas is known at thebeginning (Cinit) and end (Cend) of the wash-out, as is the total volume of gas expired.FRC can therefore be calculated using:FRC 5 volume of gas expired/((Cinit) - (Cend).

FRC measured by MBW includes only thevolume that equilibrates with tracer gas

during tidal breathing (i.e. not trapped gas)and may be a smaller volume than thatcalculated by plethysmography, which mea-sures total thoracic gas [48]. In diseasedairways, inflammation, mucous plugging andairway wall damage will alter gas mixing andit becomes inhomogeneous [49]. The lungclearance index (LCI) is a simple measurecommonly reported from washout recordingsto describe overall ventilation homogeneity.It reflects the number of times the FRC mustbe ‘‘turned over’’ to bring the tracer gas toreach the set end-concentration [11], and isan index of this turnover volume dividedby the measured FRC. In disease, the numberof FRC turnovers to clear the tracer gas,and therefore LCI, are increased: LCI 5

cumulative expired volume/FRC.

Other measures that can be reported fromthe washout include progression analysis ofthe alveolar plateau (phase III slope) ofexpired breaths. This gives more informationabout the relative contributions of conductingor acinar airways to ventilation inhomogen-eity, but the feasibility of these measures islimited in preschool children, and particularlyinfants, due to difficulty in analysis.

Guidelines for technique and qualitycontrol in MBW are available [37, 50]. A fastgas analyser is required and the goldstandard at present for measurement is massspectrometry. Use therefore is limited to

Figure 2Examples of computerised incentives used in spirometry. a) Blowing out candlesincentive to encourage maximal peak expiratory flow. b) Bowling game to encourageexpiration to forced vital capacity. Image reproduced with permission from CareFusionCorporation.



specialist centres, but validation of commer-cial devices in children is currently underway.It is highly feasible, with success rates of over80% in sedated infants and preschool chil-dren [21, 40]. There is greater within- andbetween-test variability of LCI in infants [51],but within-test coefficient of variation in pre-schoolers is only 5.2% [40]. LCI is independ-ent of age in older children and adults,making it a useful longitudinal measure;however, the upper limit of normal is higherin infants and pre-school children. Referenceequations in children from 497 subjects havebeen published using SF6 measured by massspectrometry [52] but are tracer gas anddevice specific. Comparisons to healthy con-trols are again essential in studies usingother testing methods.

Previous studies of MBW in childrenwith CF

In the LCFC cohort, Lum et al [21] reportedraised volume tests and LCI in infantsdiagnosed with CF by newborn screeningand controls up to 1 year of age. LCI wasabnormal in 56% of infants with CF. Therewas correlation with RVRTC and FRCpleth

measures in some but not all of these infants[21]. RVRTC measures and LCI seemed to becomplementary in detecting early disease;both were abnormal in 41%, and eachmeasure identified a further 15% with abnor-mal results. By 1 year LCI remained stable,but abnormal LCI at 3 months did not predictthose with abnormal LCI at 1 year [25].Longitudinal data to later childhood innewborn screened cohorts are awaited.LCI is abnormal in infants infected withPseudomonas aeruginosa and correlates withinflammatory markers from BAL fluid [53].Only weak associations have been foundbetween LCI and structural changes on CTin infancy [54].

Reports in pre-school children contrast tothose from infancy, but most reports are inthose clinically diagnosed with CF. At thisage, LCI clearly distinguishes those with CFfrom healthy controls [40] and in this report,73% of children with CF had abnormal LCI,47% abnormal plethysmography and only13% abnormal spirometry. LCI appears assensitive as computed tomography whendetecting abnormality at school age [55].Correlation between these two measures inpreschool children is yet to be reported.Importantly, increased LCI at preschool agehas been shown to predict abnormal spiro-metry in school-age children with CF [43] andin a group of patients studied from 6 to20 years, LCI was an earlier predictor for theprogression of lung disease than FEV1 [56].

Table 3. Quality control in pre-school spirometry, adapted from [37]

Start of test Within-test End of test Repeatability

Rapid rise to peakexpiratory flow

Back extrapolated volume, 12.5% FVC or 80 ml(whichever is greater)

Smooth expiration, noglottic closure or cough

No leak or obstruction ofmouthpiece

No early inspiration

Plateau reached on volumetime curve

Cessation of flow must notoccur at 10% peak flow (butmay be able to report FEVt)

2 manoeuvres within 0.1 L or 10%highest value for FVC and FEVt

1 manoeuvre may be reported ifacceptable but this should be clearly

stated

Data are adapted from [37].

Figure 3A pre-school child during the wash-in phase of multiple breath washout. Imagecopyright P. Aurora.



LCI has detected a treatment effect intrials of dornase alpha [57], inhaled hypertonicsaline [58] and ivacaftor [59] in children withotherwise normal lung function. Disease ininfants may be too subtle for MBW to detector predict abnormality but, in preschoolchildren, it has great potential as an outcomemeasure in trials and can inform clinicalpractice. It is still necessary to show thatregular monitoring with LCI results inimproved outcome [60] and it is currentlylimited to centres with facilities to performthe test. Its use can become more widespreadwith validation of commercial equipment.

Other measures ofpreschool lung function

Plethysmography is commonly used to meas-ure lung volumes and airway resistance inadults, but involves breathing against aclosed shutter which younger children finddifficult to tolerate. An alternative plethysmo-graphic measure is specific airway resistance(sRaw)(fig. 5), which is the product of airwayresistance and FRC, and can be measuredduring tidal breathing [61]. It has beenmeasured successfully in children from2 years of age [62]. Preschool subjectsbreathe through a mouthpiece attached to apneumotachograph, wearing a nose-clip withtheir hands on their cheeks. This can bedemonstrated to the child in steps beforerecording to improve success rates, which arereported to be between 77 and 81% [40, 45].Airflow and plethysmographic pressurechange are recorded and specific resistancecan be calculated.

As sRaw increases with higher FRC orairway resistance, it is a useful measure inchildren with CF, as both can be raised in

obstructive lung disease or hyperinflation [15].Reference equations have been recentlypublished using measurements from nearly3000 children as part of the Asthma UKinitiative [63]. Significant methodological dif-ferences between centres when measuringsRaw were noted in this study, and consensusfor appropriate breathing frequency and otherquality control factors are yet to be decided,with guidelines for standard operating proce-dures urgently needed. Two main studies havereported sRaw measures in children with CF[40, 45]. In both studies sRaw was significantlyhigher than healthy controls. sRaw was persis-tently abnormal over a 4-year period andidentified those with abnormal lung functionearlier than spirometry [45]. A positive correla-tion has been seen with lung clearance index[40]. sRaw has potential to monitor early CFlung disease, but standard operating proce-dures are crucial if this is to be developed.

Other available measures of pre-schoollung function are the forced oscillation (FOT)and the interrupter technique. They are alsoboth tidal breathing measures. Forced oscil-lation measures respiratory mechanics, andinvolves the application of an external soundwave to the respiratory system. The resultingpressure and air flow response is thenanalysed and respiratory resistance (Rrs),reactance (Xrs), and impedance calculated.There are conflicting reports of FOT in youngchildren with CF, with some studies showingabnormal Rrs and Xrs [64, 65] and othersnormal values [39, 45]. No relationship wasfound in these studies between FOT mea-sures and respiratory infection.

The interrupter technique involves meas-uring pressure at the mouth over a series ofbrief airway occlusions during tidal breathing.It assumes that alveolar pressure equilibrateswith mouth pressure during the occlusions.

5.04.54.03.53.02.52.01.51.00.50.0

1.0

0.5

0.5

0.0

-0.5

-1.0

-1.5

Flow

L·s

-1

Con

cent

ratio

n %

Figure 4Example of a washout recording. The black trace is flow and green trace SF6 concentration.



Interrupter resistance (Rint) is the ratio ofpressure at the mouth measured after anocclusion and flow just before it [66]. Moststudies in children with CF have shown thatRint does distinguish those from healthycontrols and remains stable over time,despite other parameters indicating worsen-ing disease [45, 67]. Further reports will helpdefine the role of these alternative tests inmonitoring early CF lung disease.

Imaging

Progressive structural changes, particularlybronchiectasis, are the hallmark of lungdisease in adults with CF [68]. These changescan also be apparent in children, andhistological specimens have shown bronch-iectasis and mucus plugging as early as4 months of age [69]. CT is more sensitivethan chest radiography in detecting CF lungdisease in children [70] and advances inimaging techniques provide the opportunityto image structural changes in ever increas-ing detail (fig. 6).

As young children are unable to breath-hold sufficiently, scans are normally per-formed under general anaesthesia. Scoringof abnormality differs with lung volume, sostandardisation of lung inflation duringimaging is particularly important. Imagestaken in inspiration and expiration in

newborn screened children (aged 1–5 years)have been compared, with bronchiectasisreported in 30% of inspiratory scans, butonly 9% of expiratory scans, and 6% in tidalbreathing [71]. Recent studies in newbornscreened infants use strict ventilation proto-cols before imaging, and administer a‘‘breath-hold’’ (usually at 25 cm H2O) duringthe inspiratory scan [7, 72]. Concerns overradiation exposure in such young childrenhave led to the development of low-dosescanning techniques [73], and ‘‘limited slice’’protocols taking 3–6 images at standardanatomical positions in inspiration andrepeated in expiration have been used [74].However, due to the irregular distribution oflung disease in CF, limited slice protocolsmay neither capture all areas of structuralabnormality [75] nor provide adequate resolu-tion to quantify early disease [28].

Various scoring systems are available toquantify structural abnormalities. ModifiedCT scores have been used in recent studies ofnew-born screened infants with CF, in theAREST-CF and LCFC cohorts [7, 72] but thevalidity of scores in infants and youngchildren needs clarification.

Previous studies of CT in childrenwith CF

Structural changes that can be observed onCT in children less than 5 years of age with CFinclude bronchial wall thickening, air trappingand bronchiectasis, but changes are oftenmild [70]. Abnormalities have been detectedby CT in newborn screened infants shortlyafter diagnosis in the AREST-CF cohort [7]. At3 months of age 18.6% had bronchial dilata-tion, 45% bronchial wall thickening and66.7% gas trapping, giving a total of 80.7%of infants with an abnormally scored CT.Most children had no clinical symptoms ofdisease. The proportion of children withbronchiectasis increased with age from 8.5%in the first year of life to 56% in 4–6 year olds[76]. Once present, bronchiectasis was seento persist in nearly 75% of children andbecame more severe in over half [77], despiteearly diagnosis and treatment. In 27%,some improvement was seen, suggesting thatsome changes reversed despite beingdescribed as bronchiectatic. Progression ofbronchiectasis was associated with bacterialinfection, particularly with P. aeruginosa [76],and inflammation detected on BAL fluid [77].

Figure 5A pre-school child performing tidal breathing manoeuvres for the measurement of sRaw

inside a plethysmograph. Image copyright P. Aurora.



The marker most associated with the pres-ence, extent and progression of bronchiectasiswas neutrophil elastase [7, 77]. It is importantto note that scans in these reports were scoredby one individual, and the reproducibility ofthis scoring method is therefore unknown.

In contrast to these findings, the LCFCcohort studies found CT to be abnormal by1 year of age but changes were very mild [72].Using a similar scoring system, most scanswere scored as normal [78]. Only poor-to-moderate inter- and intra- observer agreementwas reported despite two experienced seniorpaediatric radiologists undergoing thoroughtraining in the score. This brings into questionthe accuracy of using current scoring systemsin this young population (with such subtledisease) that to date have only been validatedin older children. Automated quantitativescoring systems have potential to be a morereproducible scoring measure, but also requirevalidation.

Abnormalities in early childhood appearto predict lung disease progression, withabnormal CT score associated with signific-antly worse lung function in later life [79].CT scores have been shown to improveafter treatment for an exacerbation [80],dornase alpha [81] and inhaled tobramycin[82] and its use as an outcome measurein young children is promising. Howeverit is difficult to justify regular surveillanceCT in such a young population who alreadyhave a high lifetime radiation exposureand are surviving longer, particularly until

robust scoring systems for CT imagingare developed. Magnetic resonance imaging(MRI) has been shown to detect changesin adults with CF [83, 84] but is limitedat present when studying subtle earlysmall airway changes in younger children.Hyperpolarised helium MRI is more sens-itive for the detection of lung abnormalitiesin CF [85], but use is limited by cost andavailability. As functional and structuraltechniques develop the use of MRI in thispopulation may increase.

BAL

As young children may be unable to expect-orate sputum, BAL is a useful alternative todetect infection and quantify inflammatorybiomarkers in CF, particularly when per-formed alongside other measures of earlydisease. An ERS taskforce statement waspublished in 2000 giving guidelines forobtaining, storing and processing lavage fluid[86]. Centres researching the role of BAL inyoung children with CF have developedprotocols to standardise test procedures,but there is still potential for technique tovary. It is recommended that lavage isperformed in the most affected lobe ifdetected by prior imaging, or the right middlelobe in the presence of diffuse disease.However, as CF lung disease is ‘‘patchy’’,lavaging one lobe does have the potential tomiss areas of disease [87]. There is noconsensus on the number or volume of

a)

d) e) f)

b) c)

Figure 6Examples of CT images in CF showing a–c) bronchial wall thickening/dilatation and atelectasis; and d–f) airtrapping.



aliquots that should be taken at present.There may be differences in volumes anddilution of fluid retrieved from each patient,and values of biomarkers vary in sequentialaliquots. During analysis, individual orpooled aliquots can be studied, but are likelyto yield different results [87]. There are manypotential assays available which vary betweenlaboratories, and data on reproducibility ofmarkers are scarce. The European CysticFibrosis Society Clinical Trial NetworkStandardisation Committee have pub-lished a review on BAL markers andreport neutrophil count or percentage, neu-trophil elastase, interlukin-6 and interlukin-8as the most useful currently in CF [87].Limitations of this procedure in youngchildren also include its invasive nature andneed for general anaesthesia. Minor adverseevents such as fever are common butotherwise the procedure is reported asrelatively safe [88].

Previous studies of BAL in childrenwith CF

Inflammation and infection have been iden-tified at diagnosis in newborn screenedinfants [7, 89] and persists to the preschoolyears [76]. There is emerging evidence thatinflammation may occur before infection inCF [7] and inflammation (absolute cell andneutrophil count) has been detected ininfants at 4 months with negative microbio-logy [89]. Many studies report increasedinflammation with infection [7, 90–92]. Ina recent longitudinal study of newbornscreened infants with CF [7], raised freeneutrophil elastase activity at 3 months ofage was the strongest predictor for bronch-iectasis diagnosed by CT, with the odds ofpersistent bronchiectasis increased seven-fold at 12 months and 4-fold at 3 years [93].There is a correlation between increasedinflammation and abnormal infant lungfunction tests [24].

Markers in BAL have been studied in trialsof tobramycin inhaled solution [94] andrecombinant human DNase [95], but no clearchange in any one marker was found. BAL isalso a useful clinical tool when infection issuspected with otherwise negative microbio-logy. However, in a trial of BAL-directedtherapy compared with standard treatmentin children infected with P. aeruginosa (fromdiagnosis to 5 years of age), no difference inthe rate of pseudomonas eradication or CTscore between the two approaches was found[96]. BAL remains useful in the early stages oftrials to describe pathophysiological pro-cesses in CF [87] and free neutrophil elastaseactivity detected in infants appears animportant predictor for more severe lungdisease.

Conclusion

The evidence for early inflammation andinfection in young children with cystic fibrosiscontinues to grow, as does our understand-ing of the early disease processes occurringin the lungs of children with CF. In infancy,it appears that lung function testing andLCI may have a complementary role inidentifying abnormality, and further reportswill help define whether they can be usedto predict future disease in children diag-nosed by newborn screening. The role ofimaging as an outcome measure in youngchildren will become clearer as techniquesadvance and more robust scoring systemsare developed. In preschool children, MBWis a feasible test sensitive to pulmonarydisease in CF, and its role in interventionalstudies and clinical care will increase ascommercial devices are validated in thisage group. The opportunity to identifyand treat disease in children before per-manent lung damage occurs remains anexciting possibility, and further research inthis area will bring us closer to this goal.

REFERENCES

1. The Cystic Fibrosis Trust UK Cystic Fibrosis RegistryAnnual data report September 2004.

2. The Cystic Fibrosis Trust UK Cystic Fibrosis RegistryAnnual data report September 2012.

3. Farrell PM, Kosorok MR, Laxova A, et al.Nutritional benefits of neonatal screening forcystic fibrosis. New Eng J Med 1997; 337:963–969.

EducationalquestionsWhich of the followingstatements are true?1. Regarding infantlung function

a) Guidelines forstandards of testingare available for boththe RVRTC techniqueand plethysmographyb) Reference dataare derived fromtesting healthy con-trol subjectsc) Published ref-erence equations canbe applied to any testdeviced) After appropriatetraining in techniqueand quality control,feasibility varies littlebetween centrese) In infants with CFa greater rate ofdecline is seen inthose infected withrespiratory pathogens

2. Lung functionmeasures obtainedduring tidal breathinginclude

a) Specific airwayresistanceb) Respiratoryimpedancec) Multiple breathwashoutd) Forced vitalcapacitye) Interrupterresistance

3. Regarding imagingin children with CF

a) A robustlyvalidated CT scoringsystem is availablefor use in infantsb) Scoring ofabnormality varieslittle when images



4. Farrell PM, Li Z, Kosorok MR, et al.Bronchopulmonary disease in children with cysticfibrosis after early or delayed diagnosis. Am J RespirCrit Care Med 2003; 168: 1100–1108.

5. Hoo AF, Thia LP, Nguyen TT, et al. Lung function isabnormal in 3-month-old infants with cystic fibrosisdiagnosed by newborn screening. Thorax 2012; 67:874–881.

6. Khan TZ, Wagener JS, Bost T, et al. Early pulmonaryinflammation in infants with cystic fibrosis.Am J Respir Crit Care Med 1995; 151: 1075–1082.

7. Sly PD, Brennan S, Gangell C, et al. Lung disease atdiagnosis in infants with cystic fibrosis detected bynewborn screening. Am J Respir Crit Care Med 2009;180: 146–152.

8. Pittman JE, Johnson RC, Davis SD. Improvement inpulmonary function following antibiotics in infantswith cystic fibrosis. Pediatr Pulmonol 2012; 47:441–446.

9. Davies JC. Cystic fibrosis: bridging the treatment gapin early childhood. Lancet Respir Med 2013; 1:433–434.

10. Turner DJ, Stick SM, Lesouef PN. A new technique togenerate and assess forced expiration from raisedlung volume in infants. Am J Respir Crit Care Med1995; 151: 1441–1450.

11. Stocks J, Lum S. Pulmonary function tests in infantsand preschool children. In: Wilmott RW, Boat TF,Bush A, et al. eds. Disorders of the respiratory tract inchildren. 8th edn. Philadelphia, Elsevier, 2012; pp.169–210.

12. Stocks J, Godfrey S, Beardsmore C, et al.Plethysmographic measurements of lung volumeand airway resistance. ERS/ATS Task Force onStandards for Infant Respiratory Function Testing.European Respiratory Society/American ThoracicSociety. Eur Respir J 2001; 17: 302–312.

13. ATS/ERS statement. raised volume forced expirationsin infants: guidelines for current practice. Am J RespirCrit Care Med 2005; 172:11, 1463–71.

14. Davis SD, Rosenfeld M, Kerby GS, Brumback L,Kloster MH, Acton JD, et al. Multicenter evaluation ofinfant lung function tests as cystic fibrosis clinicaltrial endpoints. Am J Respir Crit Care Med 2010; 182:1387–1397.

15. Rosenfeld M, Allen J, Arets BH, et al. An officialAmerican Thoracic Society workshop report: optimallung function tests for monitoring cystic fibrosis,bronchopulmonary dysplasia, and recurrent wheez-ing in children less than 6 years of age. Ann AmThorac Soc 2013; 10: S1–S11.

16. Jones M, Castile R, Davis S, et al. Forced expiratoryflows and volumes in infants. Normative data andlung growth. Am J Respir Crit Care Med 2000; 161:353–359.

17. Lum S, Hoo AF, Hulskamp G, et al. Potentialmisinterpretation of infant lung function unlessprospective healthy controls are studied. PedPulmonol 2010; 45: 906–913.

18. Nguyen TT, Hoo AF, Lum S, et al. New referenceequations to improve interpretation of infant lungfunction. Ped Pulmonol 2013; 48: 370–380.

19. Stocks J, Modi N, Tepper R. Need for healthy controlsubjects when assessing lung function in infants withrespiratory disease. Am J Respir Crit Care Med 2010;182: 1340–1342.

20. Linnane BM, Hall GL, Nolan G, et al. Lung functionin infants with cystic fibrosis diagnosed by newbornscreening. Am J Respir Crit Care Med 2008; 178:1238–1244.

21. Lum S, Gustafsson P, Ljungberg H, et al. Earlydetection of cystic fibrosis lung disease: multiple-breath washout versus raised volume tests. Thorax2007; 62: 341–347.

22. Ranganathan SC, Stocks J, Dezateux C, et al. Theevolution of airway function in early childhoodfollowing clinical diagnosis of cystic fibrosis.Am J Respir Crit Care Med 2004; 169: 928–933.

23. Ranganathan SC, Bush A, Dezateux C, et al. Relativeability of full and partial forced expiratory maneuversto identify diminished airway function in infants withcystic fibrosis. Am J Respir Crit Care Med 2002; 166:1350–1357.

24. Pillarisetti N, Williamson E, Linnane B, et al.Infection, inflammation, and lung function decline ininfants with cystic fibrosis. Am J Respir Crit Care Med2011; 184: 75–81.

25. Nguyen TT, Thia LP, Hoo AF, et al. Evolution of lungfunction during the first year of life in newbornscreened cystic fibrosis infants. Thorax 2013 [In press:DOI: 10.1136/thoraxjnl-2013-204023].

26. Thia LP, Hoo AF, Brennan L, et al. Stable lungfunction is maintained over 2 years in newbornscreened (NBS) CF infants. Eur Respir J 2013; 42:Suppl. 57, 1072s.

27. Stocks J, Thia LP, Sonnappa S. Evaluation and use ofchildhood lung function tests in cystic fibrosis. CurrOp Pulmon Med 2012; 18: 602–608.

28. Martinez TM, Llapur CJ, Williams TH, et al. High-resolution computed tomography imaging of airwaydisease in infants with cystic fibrosis. Am J Respir CritCare Med 2005; 172: 1133–1138.

29. Peterson-Carmichael SL, Harris WT, Goel R, et al.Association of lower airway inflammation withphysiologic findings in young children with cysticfibrosis. Pediatr Pulmonol 2009; 44: 503–511.

30. Kozlowska WJ, Bush A, Wade A, et al. Lung functionfrom infancy to the preschool years after clinicaldiagnosis of cystic fibrosis. Am J Respir Crit Care Med2008; 178: 42–49.

31. Brumback LC, Davis SD, Kerby GS, et al. Lungfunction from infancy to preschool in a cohort ofchildren with cystic fibrosis. Eur Respir J 2013; 41:60–66.

32. Marostica PJ, Weist AD, Eigen H, et al. Spirometry in3- to 6-year-old children with cystic fibrosis.Am J Respir Crit Care Med 2002; 166: 67–71.

33. Harrison AN, Regelmann WE, Zirbes JM, et al.Longitudinal assessment of lung function frominfancy to childhood in patients with cystic fibrosis.Pediatr Pulmonol 2009; 44: 330–339.

34. Rosenfeld M, Ratjen F, Brumback L, et al. Inhaledhypertonic saline in infants and children youngerthan 6 years with cystic fibrosis: the ISIS randomizedcontrolled trial. JAMA 2012; 307: 2269–2277.

35. Stick S, Tiddens H, Aurora P, et al. Early interventionstudies in infants and preschool children with cysticfibrosis: are we ready? Eur Respir J 2013; 42: 527–538.

36. Kozlowska WJ, Aurora P. Spirometry in the pre-schoolage group. Paediatr Respir Rev 2005; 6: 267–272.

37. Beydon N, Davis SD, Lombardi E, et al. An officialAmerican Thoracic Society/European RespiratorySociety statement: pulmonary function testing inpreschool children. Am J Respir Crit Care Med 2007;175: 1304–1345.

38. Aurora P, Stocks J, Oliver C, et al. Quality control forspirometry in preschool children with and withoutlung disease. Am J Respir Crit Care Med 2004; 169:1152–1159.

39. Kerby GS, Rosenfeld M, Ren CL, et al. Lung functiondistinguishes preschool children with CF fromhealthy controls in a multi-center setting. PediatrPulmonol 2012; 47: 597–605.

40. Aurora P, Bush A, Gustafsson P, et al. Multiple-breath washout as a marker of lung disease inpreschool children with cystic fibrosis. Am J RespirCrit Care Med 2005; 171: 249–256.

are taken at differentlung volumesc) Bronchiectasis onCT imaging has beenreported at diagnosisin infantsd) Once detected,bronchiectasis ispersistent in mostchildrene) Hyperpolarisedhelium MRI is lesssensitive thanconventional MRItechniques indetecting abnormality

4. Bronchoalveolarlavage in childrenwith CF

a) Should be per-formed in the rightmiddle lobe whendisease is diffuseb) Has been shownto improve rates ofPseudomonasaeruginosa eradica-tion when treatmentis directed by lavageresultsc) Can predict adecreased risk of thedevelopment ofbronchiectasis if freeneutrophil elastase isdetected at 3 monthsof aged) Yields similarvalues of inflammat-ory markers in indi-vidual and pooledaliquots of lavagefluide) Results in fre-quent major adverseevents

5. With respect to inertgas washout

a) Single breathwashout is a usefulmeasure in pre-school childrenb) Published ref-erence equations are



41. Stanojevic S, Wade A, Stocks J. Reference values forlung function: past, present and future. Eur Respir J2010; 36: 12–19.

42. Quanjer PH, Stanojevic S, Cole TJ, et al. Multi-ethnicreference values for spirometry for the 3-95-yr agerange: the global lung function 2012 equations. EurRespir J 2012; 40: 1324–1343.

43. Aurora P, Stanojevic S, Wade A, et al. Lung clearanceindex at 4 years predicts subsequent lung function inchildren with cystic fibrosis. Am J Respir Crit Care Med2011; 183: 752–758.

44. Vilozni D, Bentur L, Efrati O, et al. Spirometry in earlychildhood in cystic fibrosis patients. Chest 2007; 131:356–361.

45. Nielsen KG, Pressler T, Klug B, et al. Serial lungfunction and responsiveness in cystic fibrosis duringearly childhood. Am J Respir Crit Care Med 2004; 169:1209–1216.

46. Tiddens HA, Donaldson SH, Rosenfeld M, et al.Cystic fibrosis lung disease starts in the smallairways: can we treat it more effectively? PediatrPulmonol 2010; 45: 107–117.

47. Darling RC, Cournand A, Mansfield JS, et al. Studieson the intrapulmonary mixture of gases. 1. Nitrogenelimination from blood and body tissues during highoxygen breathing. J Clin Invest 1940; 19: 591–597.

48. Gappa M, Ranganathan SC, Stocks J. Lung functiontesting in infants with cystic fibrosis: lessons fromthe past and future directions. Pediatr Pulmonol 2001;32: 228–245.

49. Horsley A. Lung clearance index in the assessment ofairways disease. Respir Med 2009; 103: 793–799.

50. Robinson PD, Latzin P, Verbanck S, et al. Consensusstatement for inert gas washout measurement usingmultiple- and single- breath tests. Eur Respir J 2013;41: 507–522.

51. Sinhal S, Galati J, Baldwin DN, et al. Reproducibilityof multiple breath washout indices in the unsedatedpreterm neonate. Pediatr Pulmonol 2010; 45: 62–70.

52. Lum S, Stocks J, Stanojevic S, et al. Age and heightdependence of lung clearance index and functionalresidual capacity. Eur Respir J 2013; 41: 1371–1377.

53. Belessis Y, Dixon B, Hawkins G, et al. Early cysticfibrosis lung disease detected by bronchoalveolarlavage and lung clearance index. Am J Respir Crit CareMed 2012; 185: 862–873.

54. Hall GL, Logie KM, Parsons F, et al. Air trapping onchest CT is associated with worse ventilationdistribution in infants with cystic fibrosis diagnosedfollowing newborn screening. PloS One 2011; 6:e23932.

55. Owens CM, Aurora P, Stanojevic S, et al. LungClearance Index and HRCT are complementarymarkers of lung abnormalities in young children withCF. Thorax 2011; 66: 481–488.

56. Kraemer R, Blum A, Schibler A, et al. Ventilationinhomogeneities in relation to standard lung func-tion in patients with cystic fibrosis. Am J Respir CritCare Med 2005; 171: 371–378.

57. Amin R, Subbarao P, Lou W, et al. The effect ofdornase alfa on ventilation inhomogeneity in patientswith cystic fibrosis. Eur Respir J 2011; 37: 806–812.

58. Amin R, Subbarao P, Jabar A, et al. Hypertonic salineimproves the LCI in paediatric patients with CF withnormal lung function. Thorax 2010; 65: 379–383.

59. Ratjen F, Sheridan H, Lee P-S, et al. Lung clearanceindex as an endpoint in a multicenter randomizedcontrol trial of ivacaftor in subjects with cysticfibrosis who have mild lung disease. B35Pathogenesis and clinical issues in cystic fibrosis.American Thoracic Society InternationalConference Abstracts: American Thoracic Society.p. A2819–A.

60. Aurora P. Multiple-breath inert gas washout test andearly cystic fibrosis lung disease. Thorax 2010; 65:373–374.

61. Dab I, Alexander F. A simplified approach to themeasurement of specific airway resistance. PediatrRes 1976; 10: 998–999.

62. Klug B, Bisgaard H. Specific airway resistance,interrupter resistance, and respiratory impedance inhealthy children aged 2-7 years. Pediatr Pulmonol1998; 25: 322–331.

63. Kirkby J, Stanojevic S, Welsh L, et al. Referenceequations for specific airway resistance in children:the Asthma UK initiative. Eur Respir J 2010; 36:622–629.

64. Solymar L, Aronsson PH, Sixt R. The forcedoscillation technique in children with respiratorydisease. Pediatr Pulmonol 1985; 1: 256–261.

65. Cogswell JJ, Hull D, Milner AD, et al. Lung function inchildhood. III. Measurement of airflow resistance inhealthy children. Br J Dis Chest 1975; 69: 177–187.

66. Beydon N. Interrupter resistance: What’s feasible?Paed Respir Rev 2006; 7, Supplement; 1: S5–S7.

67. Terheggen-Lagro SW, Arets HG, van der Laag J, et al.Radiological and functional changes over 3 years inyoung children with cystic fibrosis. Eur Respir J 2007;30: 279–285.

68. Hansell DM, Strickland B. High-resolution computedtomography in pulmonary cystic fibrosis. Br J Radiol1989; 62: 1–5.

69. Bedrossian CW, Greenberg SD, Singer DB, et al. Thelung in cystic fibrosis. A quantitative study includingprevalence of pathologic findings among differentage groups. Human Pathol 1976; 7: 195–204.

70. Helbich TH, Heinz-Peer G, Eichler I, et al. Cysticfibrosis: CT assessment of lung involvement inchildren and adults. Radiol 1999; 213: 537–544.

71. Mott LS, Graniel KG, Park J, et al. Assessment of earlybronchiectasis in young children with cystic fibrosisis dependent on lung volume. Chest 2013; 144:1193–1198.

72. Thia LP, Calder A, Stocks J, et al. Is chest CT useful innewborn screened infants with cystic fibrosis at 1year of age? Thorax 2013; [In press: DOI: 10.1136/thoraxjnl-2013-204176].

73. Brody AS, Tiddens HA, Castile RG, et al. Computedtomography in the evaluation of cystic fibrosis lungdisease. Am J Respir Crit Care Med 2005; 172:1246–1252.

74. Linnane B, Robinson P, Ranganathan S, et al. Role ofhigh-resolution computed tomography in the detec-tion of early cystic fibrosis lung disease. PaediatrRespir Rev 2008; 9: 168–174.

75. de Jong PA, Nakano Y, Lequin MH, et al. Dosereduction for CT in children with cystic fibrosis: is itfeasible to reduce the number of images per scan?Pediatr Radiol 2006; 36: 50–53.

76. Stick SM, Brennan S, Murray C, et al. Bronchiectasisin infants and preschool children diagnosed withcystic fibrosis after newborn screening. J Pediatr2009; 155: 623–628.

77. Mott LS, Park J, Murray CP, et al. Progression of earlystructural lung disease in young children with cysticfibrosis assessed using CT. Thorax 2012; 67: 509–516.

78. Thia L, Calder A, Owens C, et al. High resolutioncomputed tomography (HRCT) in one year old CFnewborn screened (NBS) infants: not a usefuloutcome measure. Pediatr Pulmonol 2012; 47: 350.

79. Sanders DB, Li Z, Brody AS, Farrell PM. Chestcomputed tomography scores of severity are asso-ciated with future lung disease progression inchildren with cystic fibrosis. Am J Respir Crit Care Med2011; 184: 816–821.

80. Davis SD, Fordham LA, Brody AS, et al. Computedtomography reflects lower airway inflammation and

available for theinterpretation of LCIc) Commercialdevices are validatedin young childrend) Abnormal LCIdetected at diagnosisof CF persiststhroughout infancy inmost childrene) MBW is moresensitive than spiro-metry in the detec-tion of abnormalityin pre-schoolchildren with CF



tracks changes in early cystic fibrosis. Am J Respir CritCare Med 2007; 175: 943–950.

81. Robinson TE, Goris ML, Zhu HJ, et al. Dornase alfareduces air trapping in children with mild cysticfibrosis lung disease: a quantitative analysis. Chest2005; 128: 2327–2335.

82. Nasr SZ, Gordon D, Sakmar E, et al. High resolutioncomputerized tomography of the chest andpulmonary function testing in evaluating the effect oftobramycin solution for inhalation in cystic fibrosispatients. Pediatr Pulmonol 2006; 41:1129–1137.

83. McMahon CJ, Dodd JD, Hill C, et al. Hyperpolarized3helium magnetic resonance ventilation imaging ofthe lung in cystic fibrosis: comparison with highresolution CT and spirometry. Europ Radiol 2006; 16:2483–2490.

84. Puderbach M, Eichinger M, Haeselbarth J, et al.Assessment of morphological MRI for pulmonarychanges in cystic fibrosis (CF) patients: comparisonto thin-section CT and chest x-ray. Invest Radiol 2007;42: 715–725.

85. Fain S, Schiebler ML, McCormack DG, et al. Imagingof lung function using hyperpolarized helium-3magnetic resonance imaging: Review of current andemerging translational methods and applications.JMRI 2010; 32: 1398–1408.

86. de Blic J, Midulla F, Barbato A, et al. Bronchoalveolarlavage in children. ERS Task Force on broncho-alveolar lavage in children. Eur Respir J 2000; 15:217–231.

87. Fayon M, Kent L, Bui S, et al. Clinimetric properties ofbroncho-alveolar lavage inflammatory markers incystic fibrosis. Eur Respir J 2014; [In press: DOI:10.1183/09031936.00017713].

88. Wainwright CE, Grimwood K, Carlin JB, et al. Safetyof bronchoalveolar lavage in young children withcystic fibrosis. Pediatr Pulmonol 2008; 43: 965–972.

89. Thursfield R, Bush A, Alton E, et al. S82 AirwayInflammation is Present by 4 Months in CF InfantsDiagnosed on Newborn Screening. Thorax 2012; 67:Suppl. 2, A40–A1.

90. Gangell C, Gard S, Douglas T, et al. Inflammatoryresponses to individual microorganisms in the lungsof children with cystic fibrosis. Clin Infect Dis 2011; 53:425–432.

91. Armstrong DS, Grimwood K, Carlin JB, et al. Lowerairway inflammation in infants and young childrenwith cystic fibrosis. Am J Respir Crit Care Med 1997;156: 1197–1204.

92. Rosenfeld M, Gibson RL, McNamara S, et al. Earlypulmonary infection, inflammation, and clinical out-comes in infants with cystic fibrosis. Pediatr Pulmonol2001; 32: 356–366.

93. Sly PD, Gangell CL, Chen L, et al. Risk factors forbronchiectasis in children with cystic fibrosis. NEJM2013; 368: 1963–1970.

94. Noah TL, Ivins SS, Abode KA, et al. Inhaled versussystemic antibiotics and airway inflammation inchildren with cystic fibrosis and Pseudomonas.Pediatr Pulmonol 2010; 45: 281–290.

95. Paul K, Rietschel E, Ballmann M, et al. Effect oftreatment with dornase alpha on airway inflam-mation in patients with cystic fibrosis. Am J RespirCrit Care Med 2004; 169: 719–725.

96. Wainwright CE, Vidmar S, Armstrong DS, et al. Effectof bronchoalveolar lavage-directed therapy onPseudomonas aeruginosa infection and structurallung injury in children with cysticfibrosis: a randomized trial. JAMA 2011; 306:163–171.

Answers toeducationalquestions1. a, b, e2. a, b, c, e3. c, d4. a5. b, e



key pointsnewborn screening on pulmonary health in cystic fibrosis has yet to be proven. ......

Documents