Pediatric Pulmonology 44:80–85 (2009)

Chest Radiograph Thoracic Areas and Lung Volumes inInfants Developing Bronchopulmonary Dysplasia

Caroline May, MRCPCH, Michael Prendergast, MRCPI, Saba Salman, BSc,Gerrard F. Rafferty, PhD, and Anne Greenough, MD*

Summary. Objectives: To determine whether chest radiograph (CXR) thoracic areas and lung

volumes differed between infants who did and did not develop BPD and according to the severity of

BPD developed. Working Hypothesis: Infants developing BPD, particularly if moderate or severe,

would have low CXR thoracic areas and lung volumes in the perinatal period. Study Design:

Prospective study. Patient-Subject Selection: 53 infants with a median gestational age of 28 (range

24–32) weeks. Methodology: CXR thoracic areas were calculated using a Picture Archiving and

Communicating System (PACS) and lung volume assessed by measurement of functional residual

capacity (FRC) in the first 72 hr after birth. BPD was diagnosed if the infants were oxygen

dependent beyond 28 days, mild BPD in infants no longer oxygen dependent at 36 weeks post-

menstrual age (PMA) and moderate/severe BPD in infants who required supplementary oxygen

with or without respiratory support at 36 weeks PMA. Results: Thirty two infants developed BPD,

21 had moderate/severe BPD. The median CXR thoracic areas were higher (P<0.0001) and

FRCs were lower (P< 0.0001) in the BPD compared to no BPD infants. The median CXR thoracic

areas of the moderate/severe group (P<0.001) and the mild group (P<0.05) were greater than

that of the no BPD group and the median FRC of the moderate/severe BPD group was lower than

the no BPD group (<0.001) and the mild BPD group (P<0.05). Conclusion: These results highlight

that in the perinatal period infants developing BPD, particularly if moderate/severe, have low

functional lung volumes and may have gas trapping, which likely reflects ventilation inhomogeneity.

Pediatr Pulmonol. 2009; 44:80–85. � 2008 Wiley-Liss, Inc.

Key words: prematurity, lung growth, lung volumes.

INTRODUCTION

Bronchopulmonary dysplasia (BPD) is a commonadverse outcome of very premature birth, affecting 40%of infants born before 29 weeks of gestation.1 Affectedinfants may be oxygen dependent for many months andrequire supplementary oxygen at home.2 They can requirefrequent readmissions in the first 2 years after birth3 andsuffer troublesome symptoms and lung function abnor-malities even as adolescents and young adults.4–6 Nowa-days, infants who are chronically oxygen dependent aredescribed as suffering from ‘‘new’’ BPD. The fewpathological reports of the condition highlight dilationof the distal gas exchange structures and decreasedalveolarization with minimal small airway injury andless prominent fibrosis than seen in infants who died ofclassical BPD.7,8 Similarly, in ventilated and oxygenexposed, prematurely born baboons, the major lungabnormalities seen were dilated alveolar ducts, decreasedalveolarization and a decreased total lung surface area.9

As a consequence, it has been proposed that ‘‘new’’ BPDmay be a maldevelopment sequence resulting from theinterference/interruption of normal development signal-ing for terminal maturation and alveolarization of thelungs.10 It would seem likely, therefore, that infants

developing BPD, particularly if it was moderate or severe,would have low lung volumes in the perinatal period.

We have previously demonstrated that infants whosubsequently develop BPD compared to those who didnot had significantly lower lung volumes, as assessed bymeasurement of functional residual capacity (FRC) usinga helium gas dilution technique.11 Such a technique,although demonstrating infants with respiratory distresssyndrome have low lung volumes and their response tosuch treatment strategies as addition of positive endexpiratory pressure,12 may under-record lung volumes if

MRC-Asthma Centre, Division of Asthma, Allergy and Lung Biology,

King’s College London, London, UK.

*Correspondence to: Anne Greenough, MD, Regional Neonatal Intensive

Care Centre, 4th Floor Golden Jubilee Wing, King’s College Hospital,

Denmark Hill, London SE5 9RS, UK. E-mail: anne.greenough@kcl.ac.uk

Received 30 April 2008; Revised 21 August 2008; Accepted 21 August

2008.

DOI 10.1002/ppul.20952

Published online 9 December 2008 in Wiley InterScience

(www.interscience.wiley.com).

Grant sponsor: Charles Wolfson Charitable Trust.

� 2008 Wiley-Liss, Inc.

there is an elevated respiratory resistance, as suboptimalgas diffusion occurs in lung units with long time constants.Assessment of lung area by calculation of chest radio-graph (CXR) thoracic areas avoids such a problem. Hence,an aim of this study was to test the hypothesis that infantsdeveloping BPD would have low CXR thoracic areas.In addition, we also measured FRC in the infants to testour further hypothesis that CXR thoracic areas andFRC results would differ according to the severity ofBPD developed. A further aim was to determine whichassessment better predicted BPD development.

MATERIALS AND METHODS

Infants born before 32 weeks of gestation were enteredinto a study determining factors predictive of BPDdevelopment if their parents gave informed writtenconsent. The study protocol included analysis of chestradiographs obtained for clinical purposes and serial FRCmeasurements. For this study, the results of infants whohad a chest radiograph taken in the first 48 hr after birthwere analyzed and the results of their FRC measurementson day three assessed.

Anteroposterior chest radiographs obtained for clinicalpurposes were analyzed. If an infant had requiredsurfactant administration, their first chest radiograph afterthe course of surfactant was complete was analyzed,otherwise their initial chest radiograph was analyzed.The radiograph plate was placed directly under the supineinfant and the X-ray beam tightly collimated. The chestradiographs were taken at a standard distance of 1 m abovethe incubator and at end inspiration as per the standardneonatal unit protocol. The chest radiographs wereimported as digital images from the Picture Archivingand Communicating System (PACS) (Centricity, GEMedical Systems, Barrington, IL) hospital computernetwork. CXR thoracic areas were calculated using GECentricity PACSTM software after free hand tracing of theperimeters of the thoracic area, outlined by the diaphragmand rib cage (Fig. 1). The results were related to birth

weight, as the BPD infants were expected to be of lowerbirthweight than the ‘‘no BPD’’ infants. To determine theintra and inter-coefficient of repeatability of the technique,chest radiographs from the first twenty infants were re-examined by the same observer and by two independentobservers. The intra-observer coefficient of variation was1.9% and the inter-observer coefficient of variation was3.9%.

FRC was assessed using a helium gas dilution techniqueand a specially designed infant circuit (total volume95 ml). The FRC system (Equilibrated Biosystems, Inc.,Series 7700, Melville, NY) included a 500 ml re-breathingbag containing a mixture of helium (10%) and oxygen.The bag was connected to a three-way valve to which afacemask or an endotracheal tube could be attached. Inventilated infants, the ventilator manifold was connectedto a port on the three-way valve, such that with the valve inthe standby position, the infant was ventilated directlyfrom the ventilator. The valve was actuated at endexpiration, which switched the infant from being directlysupported from the ventilator to the rebreathing bag.Positive pressure support was maintained by the simulta-neous redirection of airflow from the ventilator to theairtight cylinder containing the rebreathing bag. If theinfant was not ventilated, a facemask attached to the three-way valve was held snugly over their nose and mouth;silicone putty was used around the mask to achieve anairtight seal. The gas in the rebreathing bag was circulatedcontinuously through the helium analyzer during meas-urement to provide an instantaneous measurement of thehelium concentration, which was displayed in real timeon a LCD monitor. The on line display of the heliumdilution curve allowed precise determination of gasequilibration which was defined as no change in thehelium concentration over a 30-sec period. The initial andequilibration helium concentrations were used in thecalculation of FRC, which was corrected for oxygenconsumption (assumed to be 7 ml/kg min)13 and to bodytemperature under pressure saturated conditions. FRCwas estimated three times in each infant. The FRC wasexpressed as the mean of the three measurements andrelated to body weight. In this population of preterminfants, the coefficient of variation of FRC in ventilatedinfants was 5.6% and in nonventilated infants was 4.7%.

Analysis

Differences between infants with and without BPDwere assessed for statistical significance using KruskalWallis or the Mann–Whitney U test as appropriate. Toassess if there were statistically significant differencesaccording to the severity of BPD, Kruskal Wallis withDonn’s post hoc correction for multiple comparisons wasused. The strength of the relationship between CXRthoracic areas and FRC was determined by calculating a

Fig. 1. Illustration of free hand tracingof the perimeter of the CXR

thoracic area.

Pediatric Pulmonology

Chest Radiograph Thoracic Areas 81

Spearman’s correlation coefficient. Logistic regressionanalysis was performed to determine if CXR thoracicareas or FRCs were significantly associated with BPDdevelopment independent of know risk factors for BPDdevelopment, that is, gestational age, birthweight, growthretardation, and gender. To determine whether FRC orCXR thoracic areas were predictive of BPD developmentreceiver operator characteristic curves were constructed14

and the areas under the curves calculated.15

Patients

Fifty-three patients, median gestational age 28 (range24–32) weeks and birthweight 1.074 (0.474–1.950) kgwere included in the study. The infants were ventilated viashouldered, uncuffed endotracheal tubes, which have beendemonstrated to have minimal or no leaks16 and received3 cmH2O of positive end expiratory pressure. Infants werediagnosed as having BPD if they were oxygen dependentbeyond 28 days and further classified as having mild BPDif they were no longer oxygen dependent at 36 weekspost menstrual age (PMA) and moderate/severe BPDif they required supplementary oxygen with or withoutrespiratory support at 36 weeks PMA.10 The study wasapproved by the King’s College Hospital Research EthicsCommittee.

RESULTS

Thirty-two infants developed BPD, they were signifi-cantly more immature and lighter at birth than the restof the cohort. In addition, they had significantly higherCXR thoracic areas (Fig. 2) and lower FRCs (Table 1).The median FRC of the ventilated infants (12.8, range6.3–20.6 ml/kg) was significantly lower than that of thenonventilated infants (17.6, range 12.8–23.8 ml/kg)(P< 0.001). The infants who developed mild ormoderate/severe BPD were more immature (P< 0.001,P< 0.001) and lighter at birth than those who did not

develop BPD (P< 0.05, P< 0.001 respectively) (Table 2).Significantly more of the moderate/severe and the mildBPD group had required intubation (P< 0.001, P< 0.01respectively) than the no BPD group and more ofthe moderate/severe BPD group required surfactant(P< 0.001) than the no BPD group. The durations ofventilation were longer in the moderate/severe BPD thanthe no BPD (P< 0.001) and the mild BPD (P< 0.05)groups and their oxygen dependency duration was alsolonger than the no BPD group (P< 0.001), as was that ofthe mild BPD group (P< 0.05). The CXR thoracic areasof the moderate/severe group (P< 0.001) and the mildgroup (P< 0.05) were significantly greater than those ofthe no BPD group (Table 2). The infants who had mildor moderate/severe BPD had lower FRCs (P< 0.05,P< 0.001 respectively) than the no BPD group and theFRCs of the moderate/severe BPD group were lower thanthose of the mild BPD group (P< 0.05).

There was a significant negative correlation between theCXR thoracic areas and FRCs (r¼�0.459, P¼ 0.001).Regression analysis demonstrated that FRCs were(P¼ 0.015), but CXR thoracic areas were not, signifi-cantly associated with BPD development after accountingfor gestational age, birthweight, growth retardation, andgender. Calculation of areas under receiver operatorcurves (ROC) demonstrated that FRC results were morepredictive than CXR thoracic areas of BPD development(Table 3).

DISCUSSION

We have demonstrated that chest radiograph thoracicareas were significantly greater in infants who developedBPD, particularly in those who developed moderate/severe BPD, compared to those who did not. In contrast,the FRC results were significantly lower in the infants whodeveloped BPD compared to those who did not developBPD.

FRC was measured using a helium gas dilutiontechnique (FRCHe), which does not measure areas of thelung that are ventilating slowly or not at all. That problemis particularly marked in patients who have ventilationinhomogeneity or gas trapping. In affected patients, lungvolume using a gas dilution technique will be under-estimated and in adults may introduce an error as largeas 480 ml.17 In healthy supine adults, however, a goodcorrelation was found between helium gas dilution FRCand planimetric lung field areas.18 In addition, a moderateto strong positive correlation between the CXR lung areaand FRC assessed by helium dilution was demonstrated in14 preterm ventilated infants in both the prone and supinepositions.19 Using a gas dilution technique, however,lung volumes can be under-recorded, as suboptimal gasdiffusion occurs in lung units with a long time constant dueto an elevated resistance. Under-recording can be avoided

Fig. 2. Scatter plot of CXR thoracic lung areas related to gesta-

tional age. Individual data are shown (*) Infants who developed

BPD (*) Infants who did not develop BPD.

Pediatric Pulmonology

82 May et al.

if a protocol is used which stipulates that the measurementcontinues until there is no change in the heliumconcentration, as this will allow equilibration of thehelium even in areas of the lung with long time constants.In this study, we used the same protocol for all infants andonly deemed equilibration of the helium to have occurredif there was no change in the helium concentrationover a 15-sec period. The infants included in this studywere examined in the first days after birth, but even atthat time infants who develop BPD have a high respi-ratory resistance,20–22 although this is not a consistentfinding.23,24 The measurement of FRC by whole bodyplethysmography (FRCpleth) is not subject to errors relatedto ventilation inhomogeneity or gas trapping and neither isthe CXR thoracic area. Indeed, in one study lung volumeassessed radiographically correlated better with FRCpleth

than FRCHe.25 We did not use a plethysmographic

technique to measure lung volume as this was impracticalin the infants requiring respiratory support/intensive care.

The FRC (and CXR thoracic areas) were related toweight, only a very small proportion (<15%) of the infantswere small for gestational age. The expression of FRCresults as ml/kg has been stated to be appropriate in theneonatal period, as the regression of FRC on weight isrelatively linear and passes close to the origin.12 Inaddition, it is not practical to relate FRC to length, asuse of stadiometer is not feasible in prematurely borninfants receiving intensive care and use of a tape measureis inappropriate.12 It has been suggested25 that whenassessing lung areas from the radiograph it is unnecessaryto calculate a magnification factor if a large distancebetween the tube and radiograph is used. We assessed

TABLE 2— Demographics, FRC and CXR Areas According to BPD Status and BPD Severity

No BPD Mild BPD Mod/severe BPD

N 21 11 21

Gestational age (weeks) 31 (25–32) 26 (24–29) 27 (24–30)

Birthweight (g) 1,450 (860–1,950) 1,060 (660–1,230) 780 (470–1,580)

Male (%) 33% 55% 76%

Small for gestational age (%) 0% 0% 24%

Antenatal steroids (n) 71% 100% 86%

Maternal antenatal smoking 43% 36% 52%

Postnatal surfactant (n) 38% 73% 95%

Ventilated (n) 29% 82% 95%

PDA 0% 36% 62%

Duration of ventilation (days) 1 (0–6) 3 (0–40) 16 (1–161)

Duration of supplementary oxygen (days) 5 (0–16) 55 (29–72) 87 (16–365)

Septic episodes (n) 1 (0–2) 1 (1–6) 4 (2–9)

Postnatal steroids (%) 0% 9% 19%

CXR area (cm2/kg) 18.4 (14.8–32.4) 22.9 (17.8–30.4) 23.5 (19.8–35.8)

FRC (ml/kg) 20.2 (12.8–30.4) 14.5 (10.6–19.5) 12.1 (6.3–16.1)

Data are demonstrated as median (range) or percentage.

TABLE 1— Comparison of Infants With and Without BPD

No BPD BPD P

N 21 32

Gestational age (weeks) 31 (25–32) 26.5 (24–30) <0.0001

Birthweight 1,450 (860–1,950) 895 (470–1,580) <0.0001

Male 33% 69% 0.012

SGA 0% 16% 0.059

Antenatal steroids 71% 91% 0.071

Maternal smoking 43% 47% 0.776

Surfactant 38% 88% <0.0001

Ventilated 29% 94% <0.0001

PDA 0% 34% 0.003

Duration of ventilation 1 (0–6) 10.5 (0–161) <0.0001

Duration of supplementary oxygen (days) 5 (0–16) 65.5 (28–365) <0.0001

Septic episodes 1 (0–2) 3 (1–9) <0.0001

Postnatal steroids 0% 16% 0.059

CXR area (cm2/kg) 18.4 (14.8–32.4) 23.5 (17.8–35.8) <0.0001

FRC (ml/kg) 20.2 (12.8–30.4) 12.9 (6.3–19.5) <0.0001

Data are demonstrated as median (range), n or percentage.

Pediatric Pulmonology

Chest Radiograph Thoracic Areas 83

chest radiographs taken for clinical purposes, but a filmto focus distance of 1 m was used as per our standardprotocol.

The present results demonstrating infants who went onto develop BPD had significantly greater CXR thoraciclung areas, but lower FRCs, indicate the infants may havehad gas trapping and low functional lung volumes dueto ventilation inhomogeneity. This is further highlightedby the significant negative correlation between the CXRlung areas and FRCs. The chest radiographs which wereanalyzed were obtained for clinical purposes as it is not thepolicy of the unit to obtain a chest radiograph if the infant’srespiratory status is improving, hence a proportion ofinfants did not have chest radiographs taken after 48 hrfrom birth. Thus, the chest radiograph thoracic areas werecalculated on chest radiographs obtained in the first 48 hr.The lung volumes using the helium gas dilution FRCtechnique were measured on day 3. Assessment of FRCprior to day 3, however, may have yielded lower lungvolumes in those with more severe disease26 and hencemore likely to develop BPD and there would have been aneven greater discrepancy between the FRC and chestradiograph thoracic areas. More of the infants who wenton to develop BPD were ventilated. Over distension frommechanical ventilation, however, does not explain ourresults as the BPD compared to the non BPD infants hadsignificantly lower FRCs. In an earlier study27 we hadfound a significant positive correlation between CXRlung area and FRC (r¼ 0.60, P< 0.0001). In that study,however, the lung area was the area demarcated by the ribcage and diaphragm minus the cardiac, mediastinaland thymic densities and areas of perihilar and lobarconsolidation and thus the CXR lung area approximatedto the ‘‘functional’’ lung area. In contrast, in this studythe CXR thoracic area included the cardiac, mediastinaland thymic densities and areas of perihilar and lobarconsolidation, indeed all the areas contained inside thethoracic area. We did this, as in infants with severe RDS,the lungs may be so opaque it is difficult to distinguishbetween the lung fields and cardiac silhouette, theso-called ‘‘whiteout.’’ In infants poorly responsive tosurfactant, the whiteout may still be present despite that

treatment. Regression analysis demonstrated that the CXRlung areas were not significantly associated with BPDdevelopment after accounting for gestational age, birth-weight, growth retardation and gender. One explanationis that the very immature infants, who are the ones whodevelop BPD, have relatively broader and less deepthoraces than more mature infants. Nevertheless, theinfants who went on to develop moderate/severe BPD hadsignificantly lower FRCs than those who went on todevelop mild BPD.

There are various explanations for the low FRCs wereport in the infants who developed BPD. Significantlymore of the infants who developed BPD were ventilatedand received surfactant and thus likely to have respiratorydistress syndrome (RDS). RDS is characterized by non-compliant, low volume lungs. We, however, analyzedFRC measurements and chest radiographs obtained onceno further surfactant doses were given and surfactantadministration reduces RDS28 and improves lung vol-umes.29 Our results suggest ventilation inhomogeneityis present after surfactant administration. It would beinteresting to determine how long this abnormalitypersists, but clearly it would be inappropriate to undertakeserial CXRs only for such a purpose. Another explanationfor the lower FRCs in the BPD infants is poorer antenatallung growth. There was a tendency for more of the BPDgroup to have been exposed to antenatal steroids and, in apreclinical model30 antenatal betamethasone administra-tion was associated with almost a 30% reduction in thetotal alveolar number. The present results do not excludeabnormal lung growth contributing to the low FRC resultsin the infants developing BPD and at 36 weeks PMA, wepreviously highlighted that infants with BPD compared tothose without BPD had lower FRC results.31 It is possible,however, at that postmenstrual age the lower FRCs in theBPD infants reflect gas trapping.32,33 At 36 weeks PMA,FRC measured using a nitrogen washout technique wassignificantly lower in BPD infants compared to controls,but FRCpleth was significantly higher.32 The results of afurther study33 also demonstrated that infants with BPD at36 weeks, investigated using the raised volume rapidthoraco-abdominal compression as well as body plethys-mography, had gas trapping.

We have previously demonstrated a low FRC in theperinatal period is predictive of BPD development31 andconfirm those findings in this study. The CXR thoracicareas were less predictive of BPD development andregression analysis demonstrated that FRCs, but not CXRthoracic areas, were significantly associated with BPDdevelopment after accounting for gestational age, birth-weight, growth retardation and gender. Whether a specificassessment of ventilation inhomogeneity would betterpredict BPD development merits testing as in infantsstudied at 35 weeks PMA, BPD infants compared tocontrols, had impaired gas mixing.34

TABLE 3— Prediction of BPD Development and BPDSeverity

BPD severity

BPD Mild Moderate/severe

Gestational age 0.91 0.93 0.90

Birthweight 0.91 0.88 0.93

CXR area 0.81 0.79 0.83

FRC 0.90 0.81 0.95

Data are shown as areas under ROC curves.

Pediatric Pulmonology

84 May et al.

In conclusion, we have demonstrated that CXR thoracicareas and lung volume assessed by measurement ofFRC differed significantly in the first 72 hr after birthbetween infants who did and did not develop BPD. Thesignificantly greater CXR thoracic areas in associationwith lower FRC results in infants developing BPD likelyindicates gas trapping and low functional lung volumesdue to ventilation inhomogeneity.

ACKNOWLEDGMENTS

Dr. Caroline May was supported by the CharlesWolfson Charitable Trust.

REFERENCES

1. Johnson AH, Peacock JL, Greenough A, Marlow N, Limb ES,

Marston L, Calvert SA. High-frequency oscillatory ventilation for

the prevention of chronic lung disease of prematurity. N Engl J

Med 2002;347:633–642.

2. Greenough A, Hird MF, Gamsu HR. Home oxygen therapy

following neonatal intensive care. Early Hum Dev 1991;26:

29–35.

3. Greenough A, Cox S, Alexander J, Lenney W, Turnbull F,

Burgess S, Chetcuti PA, Shaw NJ, Woods A, Boorman J, Coles S,

Turner J. Health care utilisation of infants with chronic lung

disease, related to hospitalisation for RSV infection. Arch Dis

Child 2001;85:463–468.

4. Koumbourlis AC, Motoyama EK, Mutich RL, Mallory GB,

Walczak SA, Fertal K. Longitudinal follow-up of lung function

from childhood to adolescence in prematurely born patients with

neonatal chronic lung disease. Pediatr Pulmonol 1996;21:28–34.

5. Northway WH Jr, Rosan RC, Porter DY. Pulmonary disease

following respirator therapy of hyaline-membrane disease.

Bronchopulmonary dysplasia. N Engl J Med 1967;276:357–368.

6. Vrijlandt EJ, Gerritsen J, Boezen HM, Duiverman EJ. Gender

differences in respiratory symptoms in 19-year-old adults born

preterm. Respir Res 2005;6:117.

7. Husain AN, Siddiqui NH, Stocker JT. Pathology of arrested acinar

development in postsurfactant bronchopulmonary dysplasia. Hum

Pathol 1998;29:710–717.

8. Margraf LR, Tomashefski JF Jr, Bruce MC, Dahms BB.

Morphometric analysis of the lung in bronchopulmonary

dysplasia. Am Rev Respir Dis 1991;143:391–400.

9. Coalson JJ, Winter VT, Gerstmann DR, Idell S, King RJ, Delemos

RA. Pathophysiologic, morphometric, and biochemical studies of

the premature baboon with bronchopulmonary dysplasia. Am Rev

Respir Dis 1992;145:872–881.

10. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J Respir

Crit Care Med 2001;163:1723–1729.

11. Kavvadia V, Greenough A, Dimitriou G. Early prediction of

chronic oxygen dependency by lung function test results. Pediatr

Pulmonol 2000;29:19–26.

12. Hulskamp G, Pillow JJ, Dinger J, Stocks J. Lung function tests in

neonates and infants with chronic lung disease of infancy:

Functional residual capacity. Pediatr Pulmonol 2006;41:1–22.

13. Hey EN. The relation between environmental temperature and

oxygen consumption in the new-born baby. J Physiol 1969;200:

589–603.

14. Altman DG, Bland JM. Diagnostic tests 3: Receiver operating

characteristic plots. Br Med J 1994;309:188.

15. Hanley JA, McNeil BJ. The meaning and use of the area under

a receiver operating characteristic (ROC) curve. Radiology 1982;

143:29–36.

16. Hird M, Greenough A, Gamsu H. Gas trapping during high

frequency positive pressure ventilation using conventional

ventilators. Early Hum Dev 1990;22:51–56.17. Pierce RJ, Brown DJ, Holmes M, Cumming G, Denison DM.

Estimation of lung volumes from chest radiographs using shape

information. Thorax 1979;34:726–734.

18. Ries AL, Clausen JL, Friedman PJ. Measurement of lung volumes

from supine portable chest radiographs. J Appl Physiol 1979;47:

1332–1335.19. Kumar P, Leonidas JC, Ashtari M, Napolitano B, Steele AM.

Comparison of lung area by chest radiograph, with estimation of

lung volume by helium dilution during prone and supine position-

ing in mechanically ventilated preterm infants: A pilot study.

Pediatr Pulmonol 2005;40:219–222.20. Desai N, Cunningham MD, Boyer DJ, Gagel DM. Predisposing

pulmonary dynamics from birth for infants with bronchopulmo-

nary dysplasia. Clinic Res 1983;31:908A.

21. Goldman SL, Gerhardt T, Sonni R, Feller R, Hehre D, Tapia JL,

Bancalari E. Early prediction of chronic lung disease by

pulmonary function testing. J Pediatr 1983;102:613–617.

22. Lui K, Lloyd J, Ang E, Rynn M, Gupta JM. Early changes in

respiratory compliance and resistance during the development of

bronchopulmonary dysplasia in the era of surfactant therapy.

Pediatr Pulmonol 2000;30:282–290.23. Bhutani VK, Abbasi S. Relative likelihood of bronchopulmonary

dysplasia based on pulmonary mechanics measured in preterm

neonates during the first week of life. J Pediatr 1992;120:605–

613.

24. Freezer NJ, Sly PD. Predictive value of measurements of

respiratory mechanics in preterm infants with HMD. Pediatr

Pulmonol 1993;16:116–123.

25. Fumey MH, Nickerson BG, Birch M, McCrea R, Kao LC. A

radiographic method for estimating lung volumes in sick infants.

Pediatr Pulmonol 1992;13:42–47.

26. Dimitriou G, Greenough A, Kavadia V. Early measurement

of lung volume—A useful discriminator of neonatal respiratory

failure severity. Physiol Meas 1996;17:37–42.

27. Dimitriou G, Greenough A, Kavvadia V, Shute M, Karani J. A

radiographic method for assessing lung area in neonates. Br J

Radiol 1999;72:335–338.

28. Soll RF. Prophylactic natural surfactant extract for preventing

morbidity and mortality in preterm infants. Cochrane Database

Syst Rev 2000;2:CD000511.

29. Dimitriou G, Greenough A, Giffin FJ, Karani J. The appearance of

‘‘early’’ chest radiographs and the response to surfactant replace-

ment therapy. Br J Radiol 1995;68:1177–1180.

30. Willet KE, Jobe AH, Ikegami M, Newnham J, Brennan S,

Sly PD. Antenatal endotoxin and glucocorticoid effects on

lung morphometry in preterm lambs. Pediatr Res 2000;48:782–

788.

31. Greenough A, Dimitriou G, Bhat RY, Broughton S, Hannam S,

Rafferty GF, Leipala JA. Lung volumes in infants who had mild to

moderate bronchopulmonary dysplasia. Eur J Pediatr 2005;164:

583–586.

32. Wauer RR, Maurer T, Nowotny T, Schmalisch G. Assessment

of functional residual capacity using nitrogen washout and

plethysmographic techniques in infants with and without

bronchopulmonary dysplasia. Intensive Care Med 1998;24:

469–475.

33. Robin B, Kim YJ, Huth J, Klocksieben J, Torres M, Tepper RS,

Castile RG, Solway J, Hershenson MB, Goldstein-Filbrun A.

Pulmonary function in bronchopulmonary dysplasia. Pediatr

Pulmonol 2004;37:236–242.

34. Shao H, Sandberg K, Hjalmarson O. Impaired gas mixing and low

lung volume in preterm infants with mild chronic lung disease.

Pediatr Res 1998;43:536–541.

Pediatric Pulmonology

Chest Radiograph Thoracic Areas 85