Pediatric Pulmonology 48:1206–1213 (2013)

Bronchopulmonary Dysplasia:Clinical Grading in Relation to Ventilation/Perfusion MismatchMeasured by Single Photon Emission Computed Tomography

Malin Kjellberg, MD,1,2* Karin Bjorkman, MD,3 Malin Rohdin, MD, PhD,1,4

Alejandro Sanchez-Crespo, MSc, PhD,5 and Baldvin Jonsson, MD, PhD1,2

Summary. Bronchopulmonary dysplasia (BPD) is a significant cause of morbidity in the pre-

term population. Clinical severity grading based on the need for supplemental oxygen and/or

need for positive airway pressure at 36 weeks postmenstrual age does not yield reproducible

predictive values for later pulmonary morbidity. Single photon emission computed tomography

(SPECT) was used to measure the distribution of lung ventilation (V) and perfusion (Q) in

30 BPD preterm infants at a median age of 37 weeks postmenstrual age. The V and Q were

traced with 5 MBq Technegas and Technetium-labeled albumin macro aggregates, respectively,

and the V/Q match–mismatch was used to quantify the extent of lung function impairment. The

latter was then compared with the clinical severity grading at 36 weeks, and time

spent on mechanical ventilation, continuous positive airway pressure (CPAP) and supplemental

oxygen. Of those with mild and moderate BPD 3/9 and 3/11 patients, respectively, showed

significant V/Q mismatches. By contrast, 4/10 patients with severe BPD showed a satisfactory

V/Q matching distribution. An unsatisfactory V/Q match was not correlated with time spent

on supplemental oxygen or CPAP, but was significantly negatively correlated with time spent

on mechanical ventilation. SPECT provides unique additional information about regional lung

function. The results suggest that the current clinical severity grading can be improved and/or

complemented with SPECT. Pediatr Pulmonol. 2013; 48:1206–1213. � 2013 Wiley Periodicals, Inc.

Key words: bronchopulmonary dysplasia; ventilation; perfusion; V/Q; preterm; lung

function; SPECT; BPD; clinical grading.

Funding source: none reported.

INTRODUCTION

Bronchopulmonary dysplasia (BPD) is one of themost significant causes of morbidity within the pretermpopulation, with an incidence of 20–75%. It is the mostcommon chronic lung disease of infancy.1–3 BPD canresult in a significantly reduced lung function in child-hood and adult life.4

The increased survival of extremely preterm infantswith the advent of prenatal steroid and surfactant ad-ministration has led to a change in the presentation ofBPD. The older form of BPD showed pathology domi-nated by disrupted lung architecture, atelectasis, hyper-inflation, and fibrosis. The new form of BPD isbelieved to be more consistent with a homogeneouslyarrested lung development with less of the histopatho-logical changes seen in the older form.5

We lack a precise definition for the diagnosis ofBPD. The historical clinical diagnosis of BPD6 wasmodified by Sheenan et al. in 1988, when oxygendependency at 36 weeks postmenstrual age showeda good correlation to future pulmonary morbidity.

1Department of Woman and Child Health, Karolinska Institute, Stock-

holm, Sweden.

2Department of Neonatology, Karolinska University Hospital, Stockholm,

Sweden.

3Department of Physiology and Pharmacology, Karolinska Institute,

Stockholm, Sweden.

4Astrid Lindgren Children’s Hospital, Karolinska University Hospital,

Stockholm, Sweden.

5Department of Nuclear Medicine, Karolinska University Hospital, Stock-

holm, Sweden.

Conflict of interest: None.

Alejandro Sanchez-Crespo and Baldvin Jonsson contributed equally to

this work.

*Correspondence to: Malin Kjellberg, MD, Department of Neonatology,

Karolinska University Hospital, Stockholm SE 176 34, Sweden.

E-mail: malin.kjellberg@karolinska.se

Received 28 June 2012; Accepted 21 December 2012.

DOI 10.1002/ppul.22751

Published online 28 January 2013 in Wiley Online Library

(wileyonlinelibrary.com).

� 2013 Wiley Periodicals, Inc.

Subsequently, this time point for diagnosis became thefavored one.7 A revised clinical diagnosis of BPD wasproposed by the National Institutes of Health (NIH)workshop on BPD in 2001. Accordingly, diagnosis isnow based on the need for supplemental oxygen at �28days of age, and clinical grading on the need for oxy-gen/positive pressure support at 36 weeks postmenstrualage.8 The oxygen reduction test (ORT) was developedlater on to standardize grading of BPD in clinical tri-als.9,10 A concern is that the clinical oxygen and/or pos-itive pressure criteria at 36 weeks may not identifyimportant pulmonary abnormalities that can lead to areduced lung capacity in BPD survivors.11 A review ofseveral clinical studies revealed sensitivity and positivepredictive values of 11–75% for pulmonary morbidityin later childhood, based on the criteria of supplementaloxygen at 36 weeks postmenstrual age. Similar valueswere obtained by applying later classification criteria.12

The clinical grading and evaluation of BPD has notbeen improved by Computer Tomography (CT)imaging.13

The need for supplemental oxygen in BPD may bedue to an abnormal ventilation (V) and perfusion (Q)matching (V/Q distribution), a shunt, or an oxygen dif-fusion deficit. The need for oxygen may also be influ-enced by external factors such as feeding difficulties,fluid intake, infection etc. Increasing the FiO2 canimprove oxygenation in situations of reduced V/Q with-out a shunt, but is inefficient in a situation with an iso-lated shunt. Little is known about which componentpredominates in infants with BPD. Recent technicaladvances opened the possibility of applying single pho-ton emission computed tomography (SPECT) for simul-taneous V/Q scans in non-cooperative patients.14

SPECT imaging of the infant lungs offers the uniquecapability of quantifying the three dimensional distribu-tion of the extent of V/Q matching, and provides afunctional assessment of the lung at a regional level.The regional pulmonary V/Q distribution has not beenadequately described in BPD patients. We hypothesizedthat there is a correlation between clinical severity andextent of the V/Q mismatch. Thus, the specific aims ofthis study were to quantify the extent of lung functionimpairment with SPECT and correlate this with the cur-rent standard clinical severity grading. We also investi-gated whether the V/Q correlates with the need formechanical ventilation, continuous positive airway pres-sure, and/or supplemental oxygen.

METHODS

Patients

The study was conducted from 2006 to 2010 at theDepartment of Neonatology, Astrid Lindgren Children’s

Hospital, Karolinska University Hospital, Stockholm,Sweden. The patient criteria for study inclusion werebased on:

i) A clinical diagnosis of BPD according to the NIHcriteria. The diagnosis was assigned if an infantless than 32 weeks gestation required supplemen-tal oxygen for 28 days or more.8 The ORT wasthen used to refine the BPD diagnosis at 36 weekspostmenstrual age9 if the infant required fractionalinspired oxygen (FiO2) < 0.3. The effective oxy-gen delivery was first calculated. The ORT thenconsists of a stepwise reduction in either FiO2 orflow to room air, with a lower saturation limit of90% for each step. A failed step results in a returnto the original FiO2, the result is recorded as afailure, and the infant is assigned a moderate BPDgrading.

ii) Infants were judged to be sufficiently stable totolerate transport to the Nuclear Medicine facility(5 min away).

iii) A recent negative echocardiogram for right-to-leftheart shunt. The purpose of this was to minimizethe risk of leakage of macro aggregate humanalbumin (MAA) particles to other organs via anextra-pulmonary shunt.

A total of 32 patients were available for the studyafter verbal and written parental consent was obtainedand BPD grading at 36 weeks postmenstrual age wascompleted.

At 36 weeks all infants in the mild group (n ¼ 9)breathed room air with saturations >90%; the ORT wasnot performed in this group. All infants classified asmoderate (n ¼ 11) failed the ORT. All infants in thesevere group (n ¼ 10) were on CPAP; five were still onCPAP at the time of SPECT. The SPECT examinationwas scheduled as soon as possible after clinical gradingat a median age of 37 weeks postmenstrual age for theentire group (n ¼ 30). Specifically, the median age atexamination was 36.7, 36.6, and 38.1 week’s post-menstrual age for mild, moderate and severe patients,respectively.

The SPECT examination was successful in 30 patients.Two patients were excluded from SPECT analysis dueto body movement.

Clinical parameters including total number of dayson ventilator, with supplemental oxygen and on contin-uous positive airway pressure (CPAP) were recordedfor each patient. The regional ethical- and radiationprotection committees of the Karolinska UniversityHospital, Stockholm, Sweden, approved the study inpatients with a minimum age of 36 weeks postmenst-rual age and a total maximal administered radioactivityof 10 Mega Bequerel (MBq).

Bronchopulmonary Dysplasia: Clinical Grading in Relation to Ventilation 1207

Pediatric Pulmonology

SPECT Examination

The SPECT examinations were performed at theDepartment of Nuclear Medicine of the KarolinskaUniversity Hospital, Stockholm, Sweden, following theprocedure described by Sanchez-Crespo et al. in 2008.A three-headed gamma camera (TRIAD XLT, TrionixResearch Laboratory, Twinsburg, OH) equipped withlow-energy, high-resolution parallel-hole collimatorswas used in this study. SPECT acquisitions were per-formed with 90 projections, 25 sec per projection(12.5 min total acquisition time) and 256 � 256 matrixsize with a pixel size of 1.7 mm. The spatial resolutionof the reconstructed SPECT images was approximately7 mm (full width at half maximum).15

The patient arrives to the Department of NuclearMedicine in quiet sleep after being feed at the Depart-ment of Neonatology of the same hospital. However,two of the patients needed sedation with Chloral hy-drate (30 mg/Kg). Oxygen saturation and heart ratewere continuously monitored with a pulse oximeter dur-ing the entire SPECT examinations.

The SPECT examination procedure begins with theimmobilization of the patient in the gamma camerabench with a Styrofoam filled vacuum bag. Thereafter,between 3 and 4 MBq of technetium 99 m (99mTc) la-beled-MAA (Mallinckrodt Medical, Petten, TheNetherlands) are administered through a peripheral ve-nous catheter. The Q–SPECT scan was then started usingthe image acquisition protocol previously described. Im-mediately after the Q scan and with the patient in thesame position, approximately 5 MBq of Technegas aero-sol (radioactive carbon nano-particles suspended in argongas, Tetley Manufacturing Ltd., Sydney, Australia) areadministered to the immobilized and spontaneouslybreathing sleeping infant using facemasks with 1 or 2 mldead space. The infant breathed their appropriate FiO2 orroom air via the facemask for approximately 1 min be-fore the Technegas aerosol administration commenced.This period is chosen because applying the face masktypically causes a short-lasting (40–50 breaths) disruptionof breathing pattern with an increase in minute ventila-tion, tidal volume, and cycle duration.16 Air with oxygenand Technegas aerosol is then mixed in the modifiedTechnegas generator14 and flowed continuously towardsthe facemask at 6–10 L/min. An end paper-filter of negli-gible resistance collected the radioactive exhaust. Patientson CPAP also breathed via a facemask with a pause inCPAP during the inhalation. The FiO2 was adjusted, dur-ing Technegas aerosol administration, to keep patient sat-urations >90%. After completion of the Technegasadministration, the V–SPECT scan was performed withthe same scanning parameters used for the Q–SPECTscan. In order to remove the contribution fromthe 99mTc—MAA in the V–SPECT scan, the initial

Q–SPECT scan is pixel-wise subtracted at the projectionplane from the V–SPECT scan. Filtered back-projectionwas used for image reconstruction of the (pure) V and Qimages. The total procedure time at the Department ofNuclear Medicine was approximately 30 min and thepatients received an estimated total dose of 1.5 milli-Sie-vert (mSv).17

Image Quantification and Data Analysis

The SPECT image quantification was performedblinded to the clinical BPD severity grading. For eachpatient, the lung volume was first segmented using thereconstructed images from the second SPECT scan in-cluding both V and Q. In these images, lung functionaltissue was considered to be all pixels with intensity val-ues above the background maximum level. Pixel-wiseV/Q ratios were then performed. The frequency distri-bution of V/Q ratio values was scored for every patientand at BPD group level and statistical differences be-tween BPD groups were compared (paired t-test).

The total lung functional volume with V/Q ratio val-ues within [0.6–1.4] was scored for every patient withineach of the clinically graded groups. Patients were con-sidered to have a satisfactory SPECT scan if more than50% of the total lung volume was within V/Q [0.6–1.4]. This 50% threshold is conservative, derived fromprevious observations of healthy adults.15

Additionally, single linear regression analysis wasused to separately investigate the correlation betweenfunctional lung volume and the total number of days onmechanical ventilation, supplemental oxygen and CPAPfor each patient. A correction for gestational age andthe time span between termination of mechanical venti-lation and SPECT scan was applied.

Differences between clinical parameters for the threeBPD groups were evaluated by t-test. We used t-score(slope of linear regression/standard error of the slope)to determine whether the slope of the regression dif-fered from zero. Regression coefficients represent theindependent predictive contribution of each clinical var-iable to the SPECT parameter. A P-value < 0.05 wasconsidered statistically significant.

RESULTS

Clinical Characteristics

Clinical BPD grading of the 30 patients resulted in9 mild, 11 moderate, and 10 severe cases. The SPECTexamination was performed at a median age of 37 weekspostmenstrual age. One examination was abandoneddue to apnea, otherwise the procedure was well tolerat-ed. The average gestational age for the group overall

1208 Kjellberg et al.

Pediatric Pulmonology

was 26.8 (23–32) weeks and birth weight 976 (578–1,700) g, with no significant differences between thethree BPD groups. The clinical characteristics of theincluded patients are summarized in Table 1.

The infants with severe BPD received significantlymore postnatal steroids (P < 0.05), and required a lon-ger time on mechanical ventilation (P < 0.03), CPAP(P < 0.01), and supplemental oxygen (P < 0.03)

compared with infants with mild and moderate BPD.All patients with severe BPD were on CPAP at 36 weekspostmenstrual age.

Association Between SPECT and Clinical BPDGrading

Figure 1 shows a SPECT example for each clinicalseverity category. The images presented were chosen to

TABLE 1— Clinical Characteristics of the Patients Within Each of the Clinical BPD Severity Graded Groups

Mild (n ¼ 9) Moderate (n ¼ 11) Severe (n ¼ 10)

Gestational age (weeks) 27 (25–29) 27 (23–30) 26 (23–32)

Birth weight (g) 1,130 (697–1,542) 874 (578–1,475) 950 (623–1,700)

Sex (male/female) 8/1 11/0 6/4��

Prenatal steroids (full course) 89% 73% 90%

Apgar 5 (min) 7 8 5,5

Patients treated with surfactant 8/9 9/11 9/10

Patients treated with postnatal steroids 0/9 2/11 5/10þþ

Days on ventilator 10 (0–29) 16 (0–33) 78� (11–478)

Days on CPAP 27 (7–48) 31 (3–48) 50z (29–80)

Days with extra oxygen 46� (31–58) 97 (42–200) 228 (76–557)

Supplemental oxygen at SPECT examination (%) No 25 (21–45) 31 (21–65)

Body weight at SPECT examination (kg) 2.5 (1.9–3.0) 2.4 (1.7–4.6) 2.6 (2.3–3.2)

BPD severity scoring was performed using the NIH workshop definition and the oxygen reduction test. Values are shown as number of

patients, percentage or median and range.�P < 0.03 for pooled mild/moderate versus severe.��P < 0.05 for males versus females.þþP < 0.05 for pooled mild/moderate versus severe.zP < 0.01 for pooled mild/moderate versus severe.

Fig. 1. Representative coronal SPECT slices of lung ventilation (V), perfusion (Q), and corre-

sponding ratio (V/Q) for one individual from each BPD group. Left panel, mild BPD patient

showing a good V/Q matching in both lungs. Middle panel, moderate BPD with more uneven

V/Q matching. Right panel, severe BPD showing an uneven V/Q matching.

Bronchopulmonary Dysplasia: Clinical Grading in Relation to Ventilation 1209

Pediatric Pulmonology

demonstrate the largest expected V, Q degree of mis-match for each clinical severity grading group. The av-erage of the lung volume fractional V/Q distribution foreach of the clinically graded BPD groups is shown inFigure 2. No statistically significant difference wasfound between the mild and moderate patients(P ¼ 0.16), or between those with moderate and severeBPD (P ¼ 0.18). However if those with mild and mod-erate BPD were pooled and compared with the severeBPD group, significant differences were observed(P ¼ 0.04). At a group level the average percentagelung volume with V/Q ratio values within the interval[0.6–1.4] was 59.3 � 12.9, 53.7 � 12.2, and48.1 � 15.4 for mild, moderate, and severe patients, re-spectively (Table 2). No significant statistical differen-ces were found between the groups (P > 0.1). Aspreviously stated, a value of 50% of the total lung vol-ume with a V/Q ratio within the interval [0.6–1.4] de-fined a satisfactory/unsatisfactory SPECT outcome. Formild and moderate BPD patients, 3/9 and 3/11,

respectively, had an unsatisfactory SPECT. Among thesevere BPD patients, only 6/10 had an unsatisfactorySPECT. Consequently, based on this threshold, wefound a weak association between clinical BPD gradingand SPECT results (Table 2).

The Significance of Each Clinical Parameter on theSPECT Findings

Figure 3 shows the linear correlation between theSPECT outcome and the different clinical parametersfor the infants overall (n ¼ 30). Analysis of regressionslopes revealed no significant correlation between V/Qand days on supplemental oxygen (R-squared ¼ 0.02,P ¼ 0.54), nor between SPECT and days on CPAP(R-squared ¼ 0.01, P ¼ 0.95). However, a significantnegative correlation between SPECT and days on me-chanical ventilation was obtained (R-squared ¼ 0.16,P ¼ 0.04).

Fig. 2. SPECT measured lung ventilation/perfusion ratio (V/Q)

distribution at group level. The BPD clinical severity grading is

shown as mild, moderate and severe.

TABLE 2— SPECT Outcome Versus Clinical Severity Grading

Clinical severity grading

Mild Moderate Severe

Satisfactory SPECT1 6 8 4

Unsatisfactory SPECT1 3 3 6

Lung volume with V/Q within [0.6, 1.4] (%) 59.3 � 12.9 53.7 � 12.2 48.1 � 15.4

Values are shown as number of patients, or mean � standard deviation.1Satisfactory if more than 50% of the total lung volume is within V/Q [0.6–1.4] and unsatisfactory otherwise.

Fig. 3. Linear correlation between the SPECT obtained fraction

of the total lung volume with a ventilation/perfusion ratio (V/Q)

within the interval [0.6, 1.4] and days with supplemental oxy-

gen, days with CPAP and days with assisted ventilation.

1210 Kjellberg et al.

Pediatric Pulmonology

DISCUSSION

Association Between SPECT and Clinical BPDGrading

Overall 40% of the entire cohort showed an unsatis-factory V/Q distribution (mismatch), suggesting thatthey had significant ongoing pulmonary V/Q abnormali-ties. In the mild BPD group 33% of the babies had anunsatisfactory V/Q match. In many recent clinical trials,using the ORT, this group of infants are considered as‘‘non-BPD.’’18 The obvious risk with both the clinicaland physiological definition of mild or ‘‘non’’-BPD isthat it implies a degree of normality when these patientsmay in fact have substantial ongoing pulmonary V/Qabnormalities. These patients are at risk for adverse pul-monary outcome and need close follow-up. They maybenefit from early treatment for airway infections, hy-per-reactive airway disease, and emphysema as adults.Patients who are still on nasal CPAP at week 36 areregarded as having severe BPD according to currentlyaccepted grading criteria. However, their need forCPAP at that time may be for reasons not directly relat-ed to lung expansion and V/Q matching abnormalities.This could explain the apparently normal V/Q distribu-tion in 40% of the severe patients in our cohort. Fur-thermore 27% of the infants diagnosed with moderateBPD required FiO2 < 0.3 but not CPAP yet stillshowed an abnormal V/Q ratio distribution. Consideringboth findings, we question the reason for applying thenasal CPAP criteria when scoring BPD at 36 weekspostmenstrual age.

One way of interpreting our findings in relationship tothe observed differences in V/Q between the differentgroups is that the left shift in the V/Q with increasingseverity grading seen in Figure 2 indicates the presenceof a shunt due to distal airway disease reducing ventila-tion. The flatter curve in the severe group indicates amore heterogeneous distribution of V/Q ratios, probablyindicating impairment in ventilation as well as perfusion(Fig. 1). Another possibility is that patients with highoxygen requirement showing a good V/Q match mayhave a reduced oxygen diffusion capacity.

Correlation Between SPECT Findings and ClinicalParameters

No correlation was found between days on supple-mental oxygen and V/Q distribution in this study. Thisdoes not necessarily mean that supplemental oxygen isa poor marker of disease severity in this group ofpatients. It may simply confirm the large variations inthe prescription and dosing of supplemental oxygenbetween institutions and practitioners.10 The need foroxygen may also reflect other factors such as feedingdifficulties, fluid intake etc.

We found a significant negative correlation betweenV/Q matching and time spent on mechanical ventila-tion. The initiation of mechanical ventilation is stronglylinked with the development of BPD19, whilst CPAPuse has been linked with a decreased incidence ofBPD.20 In animal studies using the baboon modelCPAP appears to improve lung volume and preservelung architecture, whilst ventilated lungs show dis-rupted lung architecture.21 The relevance of our findingsfrom this relatively small cohort with regard to clinicalparameters and long-term pulmonary outcome isat present unclear. A future challenge will be to corre-late long-term clinical outcome with SPECT and pul-monary function.

SPECT Methodological Issues

In this study, the fraction of the total lung volumewith V/Q values in the interval [0.6–1.4] was used asthe SPECT interval of merit. V/Q values outside thisinterval represent lung regions with either severely al-tered ventilation or perfusion and consequently reducedgas exchange. Our choice was based on our previousSPECT studies of healthy adults, for whom about 85%of the entire lung volume lie within the V/Q interval[0.6–1.4].15 However, there is a lack of established, re-gional V/Q reference values for healthy preterm infantsdue to ethical considerations concerning unnecessary ex-posure to radiation. Together with published lung func-tion data that demonstrates impairments associated withdecreased gestational age and BPD in patients with andwithout BPD, this suggests that the functional lung vol-ume as defined here could be lower.22,23 Consequently,we chose a conservative total lung volume V/Q thresh-old value to define ‘‘normal’’ matching. This was sup-ported by our findings in the mild BPD group, in whichwe observed lung functional volumes of 42–76%.

A limitation of our technique is that we could notmeasure oxygen diffusion capacity, since SPECT relieson the deposition of aerosolized particles in the alveolithat cannot diffuse across the alveolar-capillary mem-brane at the time of the examination.24

The severe BPD patients were examined withSPECT, on average, 1 week later than mild or moderateBPD patients. This delay was partly due to clinical in-stability that precluded transport to the nuclearmedicine department. The effect of this delay on theSPECT V/Q comparison between BPD groups is un-clear, further longitudinal clinical studies could providean answer.

Five patients were still on CPAP at the time of theSPECT examination. Those infants were in the weaningphase of CPAP treatment and were able to tolerate peri-ods without CPAP during nursing procedures. TheCPAP need at that time was probably not so important

Bronchopulmonary Dysplasia: Clinical Grading in Relation to Ventilation 1211

Pediatric Pulmonology

for lung expansion and thus 1–3 min without CPAPshould not affect the result of gas distribution or perfu-sion. However, even though technically the Technegasaerosol could be administered through the CPAP de-vice, we wanted to standardize the inhalation procedurefor all patients. Most examinations were successful inpostprandial quiet sleep. Chloral Hydrate should beused rarely and with caution in this population. Howev-er, the low dose for sedation we used for two infantswould not be expected to significantly affect respirationand delivery of the aerosol.25

Alternatives to SPECT for BPD Grading

The limitations of the SPECT technique used in thiswork have been thoroughly presented previously.14,15

The major drawback is the use of radiation, which atthis early age and despite the extremely low radiationdoses involved, remains a concern. Quine et al.26 devel-oped an alternative non-invasive indirect measurementof V/Q by plotting the oxygen saturation (SaO2) againstFiO2. This method detects a right shift of the oxygendissociation curve and permits the degree of V/Q and/or right to left shunt to be estimated. Also, in otherstudies a higher gradient between capnography and cap-illary pCO2 has been described in infants with BPD anda suggestion made that these can partly be attributed toa V/Q mismatch.27 However, a more objective way ofdescribing the degree of lung physiologic impairment isto directly measure the reduced V/Q ‘‘mismatch’’ viaSPECT. Perhaps SPECT can be used to evaluatethe reliability and usefulness of the alternative, non in-vasive methods.

Several studies have examined the anatomical picturein infants with BPD, both in the pre- and post-surfac-tant era. In these studies, grading systems have beendeveloped that combine radiology and clinical severity.Soler et al.28 studied the pulmonary perfusion of 10infants with BPD as an adjunct to radiological and clin-ical severity. They found a good correlation betweenperfusion abnormalities and BPD severity. The advan-tage with the SPECT technique used in our study is thatit offers a functional assessment at the regional level ofboth ventilation and perfusion, combined with a newerclinical grading system. In addition this SPECT tech-nique is readily available and only uses a fraction of theradiation dose compared to conventional CT. In the fu-ture, we may be able to combine SPECT with a mea-surement of diffusion capacity in a single investigationto provide an even more complete picture of lung func-tion. Furthermore, no study has yet compared SPECTwith lung function measurements of infants or olderchildren with BPD. The combination of two measure-ments could provide valuable additional informationabout actual lung function.

CONCLUSION

SPECT provides unique additional informationregarding regional lung function. This is not currentlyavailable by any other clinical method. Our findingsmust be confirmed in a larger cohort, but they suggestthat the current clinical severity grading can be im-proved and/or complemented with SPECT.

ACKNOWLEDGMENTS

We would like to thank associate Professor MiramKatz-Salomon for critical advice and Gary Cohen, PhD,for reviewing the manuscript.

REFERENCES

1. Fellman V, Hellstrom-Westas L, Norman M, Westgren M,

Kallen K, Lagercrantz H, Marsal K, Serenius F, Wennergren M.

One-year survival of extremely preterm infants after active

perinatal care in Sweden. JAMA 2009;301:2225–2233.

2. Bhandari A, Bhandari V. Pitfalls, problems, and progress

in bronchopulmonary dysplasia. Pediatrics 2009;123:1562–

1573.

3. Lemons JA, Bauer CR, Oh W, Korones SB, Papile LA, Stoll BJ,

Verter J, Temprosa M, Wright LL, Ehrenkranz R, et al. Very

low birth weight outcomes of the National Institute of Child

health and human development neonatal research network,

January 1995 through December 1996. NICHD Neonatal

Research Network. Pediatrics 2001;107:E1.

4. Landry JS, Chan T, Lands L, Menzies D. Long-term impact of

bronchopulmonary dysplasia on pulmonary function. Can Respir

J 2011;18:265–270.

5. Bancalari E, Claure N, Sosenko IR. Bronchopulmonary dyspla-

sia: changes in pathogenesis, epidemiology and definition.

Semin Neonatol 2003;8:63–71.

6. Bancalari E, Abdenour GE, Feller R, Gannon J. Bronchopulmo-

nary dysplasia: clinical presentsation. J Pediatr 1979;95:819–

823.

7. Shennan AT, Dunn MS, Ohlsson A, Lennox K, Hoskins EM.

Abnormal pulmonary outcomes in premature infants: Prediction

from oxygen requirement in the neonatal period. Pediatrics

1988;82:527–532.

8. Jobe AH, Bancalari E. Bronchopulmonary dysplasia. Am J

Respir Crit Care Med 2001;163:1723–1729.

9. Walsh MC, Wilson-Costello D, Zadell A, Newman N, Fanaroff

A. Safety, reliability, and validity of a physiologic definition of

bronchopulmonary dysplasia. J Perinatol 2003;23:451–456.

10. Walsh MC, Yao Q, Gettner P, Hale E, Collins M, Hensman A,

Everette R, Peters N, Miller N, Muran G, et al. Impact of a

physiologic definition on bronchopulmonary dysplasia rates. Pe-

diatrics 2004;114:1305–1311.

11. Merritt TA, Deming DD, Boynton BR. The ‘‘new’’ bronchopul-

monary dysplasia: Challenges and commentary. Semin Fetal

Neonatal Med 2009;14:345–357.

12. Lefkowitz W, Rosenberg SH. Bronchopulmonary dysplasia:

Pathway from disease to long-term outcome. J Perinatol

2008;28:837–840.

13. Wilson AC. What does imaging the chest tell us about broncho-

pulmonary dysplasia? Paediatr Respir Rev 2010;11:158–161.

14. Sanchez-Crespo A, Rohdin M, Carlsson C, Bergstrom SE,

Larsson SA, Jacobsson H, Lindahl S, Jonsson B. A technique

for lung ventilation-perfusion SPECT in neonates and infants.

Nucl Med Commun 2008;29:173–177.

1212 Kjellberg et al.

Pediatric Pulmonology

15. Sanchez-Crespo A, Petersson J, Nyren S, Mure M, Glenny RW,

Thorell JO, Jacobsson H, Lindahl SH, Larsson SA. A novel

quantitative dual-isotope method for simultaneous ventilation

and perfusion lung SPET. Eur J Nucl Med Mol Imaging

2002;29:863–875.

16. Katz-Salamon M, Hertzberg T, Lagercrantz H. CO2-sensitivity

in newborn and young infants tested by the rebreathing method.

Methodological aspects. Biol Neonate 1991;59:126–132.

17. Valentin J, editor. Radiation dose to patients from radiopharma-

ceuticals (addendum 2 to ICRP publication 53). Oxford: Elsev-

ier Science; 1998. pp 1–126.

18. Mercier JC, Hummler H, Durrmeyer X, Sanchez-Luna M,

Carnielli V, Field D, Greenough A, Van Overmeire B, Jonsson

B, Hallman M, et al. Inhaled nitric oxide for prevention of bron-

chopulmonary dysplasia in premature babies (EUNO): A ran-

domised controlled trial. Lancet 2010;376:346–354.

19. Van Marter LJ, Allred EN, Pagano M, Sanocka U, Parad R,

Moore M, Susser M, Paneth N, Leviton A. Do clinical markers

of barotrauma and oxygen toxicity explain interhospital varia-

tion in rates of chronic lung disease? Pediatrics 2000;105:1194–

1201.

20. Young TE, Kruyer LS, Marshall DD, Bose CL. Population-

based study of chronic lung disease in very low birth weight

infants in North Carolina in 1994 with comparisons with 1984.

The North Carolina Neonatologists Association. Pediatrics

1999;104:e17.

21. Thomson MA, Yoder BA, Winter VT, Martin H, Catland D,

Siler-Khodr TM, Coalson J. Treatment of immature baboons for

28 days with early nasal continuous positive airway pressure.

Am J Respir Crit Care Med 2004;169:1054–1062.

22. Hjalmarson O, Sandberg KL. Lung function at term reflects se-

verity of bronchopulmonary dysplasia. J Pediatr 2005;146:

86–90.

23. Hjalmarson O, Sandberg K. Abnormal lung function in

healthy preterm infants. Am J Respir Crit Care Med 2002;

165:83–87.

24. Wiebert P, Sanchez-Crespo A, Seitz J, Falk R, Philipson K,

Kreyling WG, Moller W, Sommerer K, Larsson S, Svartengren

M. Negligible clearance of ultrafine particles retained in healthy

and affected human lungs. Eur Respir J 2006;28:286–290.

25. Baldwin DN, Pillow JJ, Stocks J, Frey U. Lung-function tests

in neonates and infants with chronic lung disease: Tidal breath-

ing and respiratory control. Pediatr Pulmonol 2006;41:391–

419.

26. Quine D, Wong CM, Boyle EM, Jones JG, Stenson BJ. Non-

invasive measurement of reduced ventilation:perfusion ratio and

shunt in infants with bronchopulmonary dysplasia: A physiolog-

ical definition of the disease. Arch Dis Child Fetal Neonatal Ed

2006;91:F409–F414.

27. Lopez E, Mathlouthi J, Lescure S, Krauss B, Jarreau PH, Mori-

ette G. Capnography in spontanously breathing preterm infants

with bronchopulmonary dysplasia. Pediatr Pulmonol 2011;46:

896–902.

28. Soler C, Figueras J, Roca I, Perez JM, Jimenez R. Pulmonary

perfusion scintigraphy in the evaluation of the severity of bron-

chopulmonary dysplasia. Pediatr Radiol 1997;27:32–35.

Bronchopulmonary Dysplasia: Clinical Grading in Relation to Ventilation 1213

Pediatric Pulmonology