radiology of diffuse interstitial pulmonary disease in children

11
Published online: 30 January 2004 © Springer-Verlag 2004 ECR 2004 – Categorical Course “Pediatric Radiology” Abstract Diffuse interstitial lung disease (DILD) represents a hetero- geneous group of disorders charac- terised by restrictive lung function and impaired gas exchange. As these diseases occur on a background of the developing lungs and immune system, the clinical presentation and disease progression is modified by comparison with their adult equiva- lents; thus, often differs markedly in presentation, clinical features and progress from ILD in adults, and it is not safe to extrapolate from adults to children. It is important to under- stand the normal growth and devel- opment of the lungs in children to understand the development of inter- stitial lung disease. As the chest ra- diograph is often non-specific, HRCT has been shown in adults and children to increase the accuracy at diagnosis of diffuse lung disease. The trade-off in sensitivity and spec- ificity of HRCT over chest X-ray is related to radiation dose which is significantly higher with convention- al spiral or volumetric CT; however, the use of low-dose (50 mA, 0.75 s) limited (1-mm slices every 15–20 mm) HRCT in inspiration with three expi- ratory supplementary scans allows accurate assessment of the presence and extent of diffuse lung disease at a dose equivalent to approximately ten chest radiographs. Images are re- constructed on a high spatial resolu- tion algorithm and displayed with a wide window setting, at a width of 1500 Hounsfield units (HU) and at a level of 500 HU. Keywords Diffuse paediatric interstitial lung disease · Idiopathic intersitial pneumonitis · Non-specific intersitial pneumonitis · Desquamative intersitial pneumonitis · Lymphocytic intersitial pneumonitis · Chronic pneumonitis of infancy Eur Radiol (2004) 14:L2–L12 DOI 10.1007/s00330-003-2037-y PEDIATRIC Catherine Owens Radiology of diffuse interstitial pulmonary disease in children Introduction Diffuse interstitial lung disease (DILD) represents a het- erogeneous group of disorders characterised by restric- tive lung function and impaired gas exchange. As these diseases occur on a background of the developing lungs and immune system, the clinical presentation and disease progression is modified by comparison with their adult equivalents [1]; thus, often differs markedly in presenta- tion, clinical features and progress from ILD in adults, and it is not safe to extrapolate from adults to children. It is important to understand the normal growth and devel- opment of the lungs in children to understand the devel- opment of ILD [2]. Normal growth and development of the lung and immune system Interstitial lung disease may present soon after birth, which suggests an antenatal onset in some cases; thus, an understanding of the early changes in the morphology and biology of the primitive lung is important. The ma- ture airway pattern is determined by week 16 of gesta- C. Owens ( ) Great Ormond Street Hospital for Children NHS Trust, London, WC1 N 3JH, UK e-mail: [email protected] Tel.: +44-207-4059200 Fax: +44-207-8298665

Upload: catherine-owens

Post on 14-Jul-2016

221 views

Category:

Documents


3 download

TRANSCRIPT

Published online: 30 January 2004© Springer-Verlag 2004

ECR 2004 – Categorical Course “PediatricRadiology”

Abstract Diffuse interstitial lungdisease (DILD) represents a hetero-geneous group of disorders charac-terised by restrictive lung functionand impaired gas exchange. As thesediseases occur on a background ofthe developing lungs and immunesystem, the clinical presentation anddisease progression is modified bycomparison with their adult equiva-lents; thus, often differs markedly inpresentation, clinical features andprogress from ILD in adults, and it isnot safe to extrapolate from adults tochildren. It is important to under-stand the normal growth and devel-opment of the lungs in children tounderstand the development of inter-stitial lung disease. As the chest ra-diograph is often non-specific,HRCT has been shown in adults andchildren to increase the accuracy atdiagnosis of diffuse lung disease.The trade-off in sensitivity and spec-ificity of HRCT over chest X-ray is

related to radiation dose which issignificantly higher with convention-al spiral or volumetric CT; however,the use of low-dose (50 mA, 0.75 s)limited (1-mm slices every 15–20 mm)HRCT in inspiration with three expi-ratory supplementary scans allowsaccurate assessment of the presenceand extent of diffuse lung disease ata dose equivalent to approximatelyten chest radiographs. Images are re-constructed on a high spatial resolu-tion algorithm and displayed with awide window setting, at a width of1500 Hounsfield units (HU) and at alevel of −500 HU.

Keywords Diffuse paediatric interstitial lung disease · Idiopathicintersitial pneumonitis · Non-specificintersitial pneumonitis · Desquamativeintersitial pneumonitis · Lymphocytic intersitial pneumonitis · Chronic pneumonitisof infancy

Eur Radiol (2004) 14:L2–L12DOI 10.1007/s00330-003-2037-y P E D I AT R I C

Catherine Owens Radiology of diffuse interstitial pulmonary disease in children

Introduction

Diffuse interstitial lung disease (DILD) represents a het-erogeneous group of disorders characterised by restric-tive lung function and impaired gas exchange. As thesediseases occur on a background of the developing lungsand immune system, the clinical presentation and diseaseprogression is modified by comparison with their adultequivalents [1]; thus, often differs markedly in presenta-tion, clinical features and progress from ILD in adults,and it is not safe to extrapolate from adults to children. Itis important to understand the normal growth and devel-

opment of the lungs in children to understand the devel-opment of ILD [2].

Normal growth and development of the lung and immune system

Interstitial lung disease may present soon after birth,which suggests an antenatal onset in some cases; thus, anunderstanding of the early changes in the morphologyand biology of the primitive lung is important. The ma-ture airway pattern is determined by week 16 of gesta-

C. Owens (✉)Great Ormond Street Hospital for Children NHS Trust,London, WC1 N 3JH, UKe-mail: [email protected].: +44-207-4059200Fax: +44-207-8298665

tion [3 ] and is driven by growth factors produced bymesenchyme and extracellular matrix (including fibro-nectin, tenascin, the integrins, syndecans and cadherins)and epipmorphin and epidermal platelet-derived and in-sulin-like growth factors [4, 5]. Mature alveoli begin todevelop late in fetal life, but most appear in the first2 years of life [6]. Numbers increase more slowly in laterchildhood, and adult levels are attained by approximately8 years of age [7, 8]. Treatment with steroids and oxygenin early life has been shown to interfere with alveolar de-velopment in the rat model [9]. Data from humans sug-gest that only limited catch-up growth may occur in laterchildhood following neonatal interference with lunggrowth [10]; thus, ILD and its treatment, particularlyearly in life, may profoundly affect long-term function.

The immune system is immature at birth and the im-munopathogenesis of ILD therefore is different in chil-dren compared with adults. Briefly, T-cell receptor acti-vation is slow in the neonate and suppressor K-cell activ-ity higher [11]. The pattern of cytokine secretion is vari-able: levels of production of interleukin 2 (IL2) are con-troversial [12], but secretion of IL4 and interferon-gam-ma is markedly reduced [13].

B-cell function is reduced, and whether this is intrin-sic or secondary to reduced T-cell function is unclear.Peripheral blood monocyte function, by contrast, seemsrelatively mature, although alveolar macrophages appearto be functionally deficient [14].

Morbidity and mortality associated with paediatricILD is high (range 14–39%) with a higher mortality inyounger infants.

The diagnosis may be delayed, but once suspected,the aim is to confirm the presence and severity of ILD,to uncover any predisposing factors and to identify thedominant pathology of the ILD.

Although it is sometimes possible to make a diagnosiswithout biopsy, the majority of patients require histologi-cal studies using either open-lung biopsy, video-assistedthoracoscopic (VATS) biopsy or high-resolution comput-ed tomography (HRCT)-guided percutaneous biopsy.

The general classification includes disorders ofknown aetiology (aspiration, chronic infection, hyper-sensitivity pneumonitis, lipid storage diseases), unknownprimary pulmonary disorders with lung involvement orsystemic disorders with pulmonary involvement, and fi-nally those diseases unique to childhood.

HRCT technique

As the chest radiograph is often non-specific, HRCT hasbeen shown in adults and children to increase the accura-cy at diagnosis of diffuse lung disease.

The tradeoff in sensitivity and specificity of HRCTover chest X-ray is related to radiation dose which is sig-nificantly higher with conventional spiral or volumetric

CT; however, the use of low-dose (50 mA, 0.75 s) limit-ed (1-mm slices every 15–20 mm) HRCT in inspirationwith three expiratory supplementary scans allows accu-rate assessment of the presence and extent of diffuselung disease at a dose equivalent to approximately tenchest radiographs. Images are reconstructed on a highspatial resolution algorithm and displayed with a widewindow setting, at a width of 1500 Hounsfield units(HU) and at a level of −500 HU.

If a child is unable to breathhold, the scans can beperformed during quiet breathing and decubitus scans re-place expiratory scans (the dependent lung behaving asthe “expiratory lung”).

The role of HRCT in paediatric ILD is evolving. Inadults, the diagnostic accuracy of HRCT has led to a de-crease in the number of lung biopsies. The histospecificaccuracy of HRCT compared with chest radiographs inmaking a correct first-choice diagnosis in an adult popu-lation with diffuse lung disease ranges from 46 to 75%for HRCT and 38 to 63% for chest radiography [15, 16,17]. The HRCT has a reported diagnostic accuracy of56% for a confident first-choice diagnosis in one seriesof 20 children with ILD [18]. This is comparable to amore recent series of 20 children with biopsy-provenILD from a single institution [19]. A correct first-choicediagnosis was made in 61% of the cases on HRCT com-pared with 34% on chest radiographs.

The diseases that were correctly diagnosed on HRCTwith a high degree of confidence were alveolar proteino-sis, pulmonary lymphangiectasia and idiopathic pulmo-nary haemosiderosis. Differentiation between nonspecif-ic interstitial pneumonitis (NSIP), desquamative intersti-tial pneumonitis (DIP) and lymphocytic interstitial pneu-monitis (LIP) was, however, less reliable.

There are several pitfalls in the interpretation ofHRCT in children. One of the most important pitfalls isin distinguishing diffuse ground-glass infiltration fromincreased lung attenuation resulting from a suboptimalinspiration. In the upper zones, the position of the poste-rior tracheal membrane is helpful in distinguishing be-tween the two. The posterior tracheal membrane is con-vex outwards in inspiration, and appears horizontal orslightly concave on expiration. The other difficulty thatmay be encountered is in determining which areas of thelung are abnormal when an investigation revealswidespread “mosaic” pattern of lung attenuation. Decid-ing on whether the areas of diminished attenuation re-present, for example, small airways disease or whetherthe areas of increased attenuation (ground-glass opacity)represent diffuse infiltration can be challenging. Al-though expiratory images may be helpful, obtaining im-ages at known phases of respiration is usually notachievable in children. Use of lateral decubitus imagingCT with the dependent lung simulating expiration is im-portant. In this way the non-dependent lung can also bereliably assessed in apparent inspiration.

L3

L4

HRCT features of diffuse ILD

Idiopathic interstitial pneumonitis

Non-specific interstitial pneumonitis (NSIP), desquama-tive interstitial pneumonitis (DIP; Fig. 1) and lympho-cytic interstitial pneumonitis (LIP) in childhood appearto share common CT appearances. Widespread ground-glass attenuation is the most frequent feature of interstiti-al pneumonitis/fibrosis [20]. There are no discriminatingfeatures for the radiological appearances of DIP whichincludes widespread ground-glass opacification and has

a wide differential: sub-acute extrinsic allergic alveolitis,opportunistic infections such as Pneumocystis cariniipneumonia and sarcoidosis which is rare in children. OnHRCT ground-glass opacification is the dominant pat-tern and often has a lower zone predominance, long-termfollow-up and usually shows complete resolution; how-ever, the few reported cases of childhood NSIP appear tohave a distinct pattern of involvement. In a review ofCTs of six cases of biopsy-proven NSIP [21], a distinctupper-zone predominance of honeycomb pattern and pa-renchymal distortion on a background of diffuse ground-glass opacification was seen in three children. The hon-

Fig. 1a–d Desquamative nterstitial pneumonitis (DIP).a Chest X-ray shows extensivebilateral consolidation. High-resolution CT (HRCT) con-firms the extent and distribu-tion of disease showing a verywidespread consolidative pro-cess. Histopathological speci-men of the lungs taken at avideo-assisted thoracoscopicprocedure (VATS) confirms thediagnosis of desquamative in-terstitial pneumonitis withevenly dispersed macrophageaccumulation of the alveolarspaces with relatively mild, ifany, interstitial fibrosis. Thereis temporal homogeneity [incontrast to the temporal hetero-geneity of usual interstitialpneumonitis (UIP)]. b TheHRCT shows dependent con-solidation in the right and leftlower lobes and a shallow rightanterior pneumothorax postlung biopsy. c Histology: speci-men from VATS lung biopsyshows evenly dispersed macro-phages accumulating in alveo-lar spaces (c1), reticulin stainwith only mild interstitial fibro-sis (c2). d Schematic diagramof DIP shows main histopatho-logical elements of DIP.1 Intensive macrophage accu-mulation within alveolar sacs.2 Minimal interstitial inflam-matory cell infiltrate

L5

eycomb pattern resembles emphysematous destructionwith some cystic change (Fig. 2). Seely et al. described asimilar HRCT appearance in their review of paediatricidiopathic interstitial fibrosis [20] but did not providepathological correlation. Ground-glass opacification,honeycombing and peripheral consolidation were fea-tures seen in three other cases of childhood NSIP. In thestudies by Copley et al. [17], one case of DIP resembledNSIP with predominant upper lobe involvement. The ex-istence of UIP (usual interstitial pneumonitis) in child-hood is controversial and said to be exceedingly rare.The adult form of UIP has a poor prognosis and HRCTfindings include a subpleural reticular pattern of honey-combing which may eventually result in thick-walledcystic spaces and irregularity of the pleural surface withunderlying traction bronchiectasis and bronchiolectasis.

Lymphocytic interstitial pneumonitis (LIP; Fig. 3) is acondition characterised histologically by a diffuse inter-stitial infiltrate of polyclonal lymphocytes, plasma cellsand histocytes. LIP is associated with viral infections in-cluding Ebstein-Barr virus (EBV) and acquired immuno-deficiency syndrome (AIDS). LIP is present in up to

30% of children with HIV disease, especially those fromthe sub-Saharan African continent where EBV is endem-ic. This is unlike the situation in adult HIV disease,where LIP is rare. Other associations include auto-im-mune diseases and lymphoproliferative disease second-ary to underlying congenital immunodeficiency.

Chest radiographs show reticulo-nodular change withor without areas of consolidation (Fig. 3a). High-resolu-tion CT shows interstitial reticulo-nodular change ofvarying degrees (Fig. 3b). There is usually marked reso-lution of the striking radiological appearances with ther-apy for HIV disease; however, if there has been chronicinterstitial disease, fibrosis and ensuing traction bronchi-ectasis may result.

Follicular bronchiolitis is a term coined to describe aform of LIP where the lymphoid aggregates congregatearound the small airways.

Chronic pneumonitis of infancy (CPI) is a recentlydescribed pathological entity and its HRCT appearanceshave been reported in a single biopsy-proven case.HRCT shows widespread ground-glass opacificationwithout focal areas of consolidation or cystic change.

Fig. 2a, b Nonspecific intersti-tial pneumonitis (NSIP).a1 Chest X-ray. a2, a3 TheHRCTs in a child with NSIPand connective tissue diseaseshows extensive parenchymaldistortion and fibrosis with ar-eas of cystic destruction of thelung parenchyma in the left up-per and right middle lobes.b Simplified line diagramshowing main features of NSIP.1 Patchy variable fibrosis com-bined with moderate (2), inter-stitial (3), alveolar inflammato-ry cell accumulation

L6

Connective tissue disorders

The prevalence of lung disease in paediatric connectivetissue disorders is high. In a study by Seely et al. [20] 10of 11 children with systemic sclerosis who underwentHRCT had evidence of interstitial lung disease. HRCTfeatures included ground-glass opacification, subpleuralmicronodules, linear opacities and honeycomb lung(consistent with UIP histologically).

SLE-induced vasculitis is manifest as patchy areas ofground-glass attenuation. Organising pneumonia (OP)with subsequent pulmonary fibrosis has been described(Fig. 4).

Aspiration pneumonitis secondary to abnormal oe-sophageal peristalsis may result in acute and chronic pa-

renchymal change with bronchiectasis in patients withsystemic sclerosis and mixed connective tissue diseasewith abnormal oesophageal peristalsis.

Pulmonary alveolar proteinosis

Congenital pulmonary alveolar proteinosis (PAP) is aninherited autosomal-recessive disorder, due to surfactantprotein B (SP/B) deficiency. This condition presents ear-ly in a full-term neonate with clinical and radiologicalfeatures indistinguishable from hyaline membrane dis-ease (HMD).

The differential diagnoses (with very similar features)include congenital viral infection and neonatal fibrosing

Fig. 3a–c Lymphocytic inter-stitial pneumonitis (LIP).a Chest X-ray shows extensivereticulo-nodular change moreconfluent in the lower lobe.b The LIP–HRCT showswidespread reticulo-nodular in-terstitial changes throughout allof the lung. c Histopathologicalspecimens of normal lung andlung biopsy from a child withLIP. Normal lung shows rela-tively sparse cellular popula-tion within the pulmonary in-terstitium, whereas the childwith LIP has diffuse infiltrationwith polyclonal lymphocytesand plasma cells

L7

alveolitis. The recently described chronic pneumonitis ofinfancy (CPI), characterised by alveolar septal thicken-ing, alveolar pneumocyte hypoplasia and an alveolar exudate containing macrophages and cellular debris, alsohave similar appearances.

The other later form of PAP usually occurs between a few months of age to several years of age [22]. Chest X-rays show bilateral air-space disease with ground-glass opacification. Air-space filling with low-attenua-tion proteinaceous material and a superimposed “crazy-paving” pattern, which represents thickened interlobularsepta, are characteristic HRCT features enabling a confi-dent diagnosis to be made in most cases (Fig. 5).

Fig. 4a, b Organising pneumo-nia (OP). Serial HRCTs takenin a teenager with OP admittedbecause of severe respiratorydistress related to systemic lupus erythematosis (SLE) with juvenile dermatomyositis(JDM) overlap. Oesophagealdysmotility resulted in aspira-tion pneumonitis. a1, a2 High-resolution CTs: the dependentlung on admission showedpatchy consolidation related to aspiration. There is also pe-ripheral ground-glass shadow-ing in the non-dependent areas(upper) pulmonary lobes. Thisis related to OP. Note the dou-ble-oblique fissure sign on theleft due to motion artefact relat-ed to tachypnoea. b1, b2 Sub-sequent scans at the same lev-els taken 6 months later follow-ing steroid and immunosup-pressant therapy show markedimprovement in the lower lobeconsolidation with some resid-ual peribronchial thickeningand dilatation. Persistent peripheral interstitial change isnoted in the non-dependent upper lobes related to scarring

Fig. 5a–c Pulmonary alveolar proteinosis (PAP). a Chest X-rayshows bilateral air space shadowing with associated upper lobeground-glass opacification. b The HRCT confirms the widespreadconsolidation with air spaces filled with low-attenuation materialand the superimposed “crazy-paving” pattern due to thickened in-ter-lobular septa. c Histopathological specimen shows amorphouspink proteinaceous material within the alveolar units (correspond-ing to consolidation) and also cellular infiltration into the pulmo-nary interstitium (corresponds to crazy-paving pattern and inter-lobular septal thickening)

L8

Congenital lymphangiectasia/diffuse pulmonary lymphangiomatosis

Congenital lymphangiectasia and diffuse pulmonarylymphangiomatosis represent a minority of chronic ILDin childhood and result from abnormal proliferation anddilatation of lymphatic channels (Fig. 6). The CT appear-ances of lymphangiomatosis can be easily predictedknowing the distribution of the lymphatic system withinthe lungs with smooth thickening of the interlobular sep-ta, peribronchial thickening, patchy ground-glass opaci-fication and increased attenuation of the mediastinal fatwith bilateral pleural effusions and pleural thickening(Fig. 6) [23]. This constellation is highly suggestive ofthe diagnosis. The condition is allied with congenitallymphangiectasia where there is abnormal dilatation ofthe lymphatics. The CT findings of these two conditionsare indistinguishable.

Idiopathic pulmonary haemosiderosis

Idiopathic pulmonary haemosiderosis (IPH), a rare dis-ease occurring in children and adults, is characterised byrecurrent episodes of pulmonary haemorrhage withhaemoptysis. The aetiology is unknown. The childhoodform of IPH usually presents before the age of 3 yearswith no specific gender prevalence [24].

The children usually present before 3 years of age andon HRCT acute haemorrhage is characterised by ground-glass opacification or consolidation. In the subacute/chron-ic phase discrete pulmonary nodules of uniform size areseen throughout the lung and are a characteristic feature.Interlobular septal thickening can also occur (Fig. 7).

Sarcoidosis

Childhood sarcoidosis is rare and there appears to betwo distinct types of sarcoidosis in children [25, 26]

Fig. 6a, b Pulmonary lymph-angiectasia. a1 Spiral CT anda2 HRCT show diffuse intersti-tial change with thickening ofthe interlobular septa in thepresence of bilateral pleuralfluid collections. b1 Coronaland b2 axial T2-weighted spin-echo MRI scans/coronal MRIscans show high signal materialwithin the pulmonary interstiti-um associated with pleural ef-fusions in a child with pulmo-nary lymphangiectasia andlarge left cervical cystic hygro-ma

L9

Fig. 8a, b Langerhan’s cell histiocytosis (LCH). a The HRCT in achild with pulmonary LCH showing extensive, multiple, thin-walled cysts distorting pulmonary architecture (<1 cm in size)within the upper lobes of the lung. b Further HRCT taken at thelevel just below the carina shows destructive thin-walled cystswithin the parenchyma with relative sparing of the lung adjacentto the anterior pleural reflections. Note area of subpleural scarringanteriorly in the right lung related to lung biopsy

Fig. 7a, b Idiopathic pulmona-ry haemoiderosis (IPH).a Chest X-ray in a child withIPH shows bilateral mid- andlower-zone ground-glasschange. b1, b2 The HRCTs atlung mid and basal zonesshowing poorly definedground-glass nodules in apatchy distribution related tobiopsy proven IPH

with the majority presenting at 13–15 years of age with a multisystem disease inseparable from the adulttype. Lymphadenopathy and pulmonary involvementare common with systemic symptoms of fever and mal-aise, and HRCT appearances identical to the adult dis-ease.

By contrast, children under 4 years of age [25, 26]present with the distinct form of the disease character-ised by rash, uveitis and arthritis/tenosynovitis. Althoughpulmonary involvement has been reported, it is usuallynot present [25]. The diagnosis can be confused clinical-ly with juvenile idiopathic arthritis (JIA).

Extrinsic allergic alveolitis

Extrinstic allergic alveolitis (hypersensitivity pneumoni-tis) is uncommon in children and is due to an immuno-logical response to a variety of allergens. The HRCT fea-tures are identical to those seen in adults [27]. In theacute/subacute phase diffuse ground-glass shadowing isseen in the lungs with small, poorly defined centrilobularnodules [28]. Patchy air trapping may also occur due tosmall-airway inflammation [29]. In chronic phases inter-stitial fibrosis supervenes [30].

L10

Langerhan’s cell histiocytosis

Pulmonary involvement in children with Langerhan’scell histiocytosis (LCH) is usually part of a disseminateddisease. Approximately 50% of children with multisys-tem LCH have pulmonary involvement, but this is not al-ways associated with an adverse outcome [29, 30]. Iso-lated pulmonary LCH is extremely rare in children.

In children under 10 years of age, pulmonary involve-ment tends to regress spontaneously (V. Nandhuri, pers.commun.), but in older children the acute features resem-ble those seen in adults, and HRCT shows multiple thin-walled cysts usually less than 1 cm in size (Fig. 8) pre-dominantly within the upper and mid zones of the lung,with relative sparing of the costophrenic angles. Multiple

nodules measuring 1–3 mm in size occur and may beperibronchial in distribution. Nodules may cavitate andare the precursors of thin-walled cystic lesions. The dis-tribution and profusion of the nodules correlates withdisease activity [31].

Aspiration pneumonitis

Children with chronic gastro-oesophageal reflux with si-lent or clinically overt aspiration of gastric contents candevelop very severe pulmonary dysfunction with associat-ed radiological changes, initially showing air-space dis-ease on the plain chest radiograph with subsequent inter-stitial fibrosis. In the infant recurrent episodes of unex-

Fig. 9a–c Aspiration pneumo-nitis. a Chest X-ray, b1, b2 spi-ral CT, and c1, c2 HRCTs in achild with unexplained respira-tory distress. There are diffuseairway and interstitial changeswith peripheral pulmonary fi-brosis, best appreciated onHRCT over spiral CTs. At lungbiopsy numerous fat-ladenmacrophages confirmed the di-agnosis of aspiration-inducedpulmonary fibrosis

L11

plained consolidation in the dependent lobes of the lungshould raise the suspicion of aspiration, and this should besought at flouroscopy and appropriately treated with anti-reflux medication±Nissen’s fundoplication to prevent thelong-term sequelae of pulmonary fibrosis (Fig. 9).

Depositional lung diseases

Lipid storage diseases

Niemann-Pick disease is an inherited defect in sphingo-myelinase production and has five clinical variants re-sulting in sphingomyelin deposition within the lung, liver,bone marrow and brain. The HRCT shows a diffusereticulo-nodular pattern with nodules varying from ap-proximately 1–2 mm in size due to the accumulation ofaggregates of large multi-vacuolated “foam” cells de-posited in the lungs (Fig. 10a).

Gaucher’s disease is due to deficiency of beta-gluco-sidase (the enzyme catablizer of glucosylceramide) re-sulting in the accumulation of the latter in reticulo-endo-thelial cells of the liver, spleen, lymph nodes, bones andbrain (especially in the infantile form). The Gauchercells then accumulate in the alveolar, interstitial and ad-jacent air spaces causing diffuse miliary or reticulo-nod-ular patterns (Fig. 10b) which may be associated withlytic rib lesions.

Conclusion

Diffuse interstitial lung disease in children is a rare butheterogeneous group of conditions, and many have noknown cause. There is a clear need for definitive histo-logical classification, which will aid with the increasingexperience in HRCT in these unusual, confounding con-ditions.

Fig. 10 Depositional lung diseasesincluding Niemann-Pick disease.a Chest X-ray and a1, a2 HRCTshow diffuse reticulo-nodular pat-tern due to aggregates of multi-vacuolated “foam” cells. b ChestX-ray and b1 HRCT show exten-sive infiltration of the lungs withaccumulation of Gaucher’s cellsin the alveolar interstitium and ad-jacent diffuse air spaces causingmiliary and reticulo-nodularchange throughout the lungs

References

1. Koh DM, Hansell DM (2000) Comput-ed tomography of diffuse interstitiallung disease in children. Clin Radiol55:659–667

2. Bush A, du Bois R (1996) Con-genital and paediatric interstitial disease. Curr Opin Pulm Med2:347–356

3. Reid L (1984) Lung growth in healthand disease. Br J Dis Chest 74:113–139

4. McGowan SE (1992) Extracellular ma-trix and the regulation of lung develop-ment and repair. FASEB J 11:2895–2904

5. Minoo O, King RJ (1994) Epithelial-mesenchymal interactions in lung development. Annu Rev Physiol56:13–45

6. Hislop A, Wigglesworth JS, Desai R(1986) Alveolar development in thehuman fetus and infant. Early HumDev 13:1–11

7. Thurlbeck WM (1982) Postnatal hu-man lung growth. Thorax 37:564–571

8. Dunnill MS (1982) The problem oflung growth. Thorax 37:561–563

L12

9. Thibeault DW, Heimes B, Rezaiekhaligh M, Mabry S (1993)Chronic modifications of lung andheart development in glucocorticoid-treated newborn rats exposed to hyper-oxia or room air. Pediatr Pulmonal6:81–88

10. Chan KN, Noble-Jamieson CM, Elliman A, Bryan EM, Silverman M(1989) Lung function in children oflow birthweight. Arch Dis Child64:1284–1293

11. Lawler SD, Ukaejiojo, Reeves BR(1975) Interaction of maternal and neo-natal cells in mixed-lymphocyte cul-tures. Lancet 2:1185–1187

12. Miyawaki TH, Seki H, Taga K, Sato H,Taniguchi N (1985) Dissociated pro-duction of interleukin-2 and immunegamma interferon by phytohaemag-glutinen stimulated lymphocytes inhealthy infants. Clin Exp Immunol59:505–511

13. Lewis DB, YU CC, Meyers J, EnglishBK, Kahn SJ, Wilson CB (1991) Cellular and molecular mechanisms forreduced interleukin 4 and interferongamma production by neonatal T cells.J Clin Invest 87:194–202

14. McCarthy KM, Gong JL, Telford W,Schneeberger EE (1992) Ontogeny of1a+ accessory cells in fetal and new-born rat lung. Am J Respir Cell MolBiol 6:349–356

15. Grenier P, Valeyre D, Cluzel P et al.(1991) Chronic diffuse interstitial lungdisease: diagnostic value of chest radiography and high resolution CT. Radiology 179:123–132

16. Padley SPG, Hansell DM, FlowerCDR, Jennings P (1991) Comparativeaccuracy of high resolution CT andchest radiography in the diagnosis of diffuse infiltrative lung disease. Clin Radiol 44:222–226

17. Copley SJ, Coren M, Nicholson AG,Rubens MB, Bush A, Hansell DM(2000) Diagnostic accuracy of thin-section computed tomography andchest radiograph in paediatric inter-stitial lung disease. Am J Roentgenol174:549–554

18. Fan L (1998) Pediatric interstitial lungdisease. In: Schawrz ML, King TE(eds) Interstitial lung disease. Decker,Hamilton, pp 103–118

19. Copley SJ, Padley SP (2001) HRCT ofthe paediatric lung disease. Eur Radiol11:2564–2575

20. Seely JM, Effmann EC, Muller NL(1997) High resolution CT of the paediatric chest: imaging findings. Am J Roentgenol 168:1269–1275

21. Fishback N, Koss M (1996) Update on lymphoid interstitial pneumonitis.Curr Opin Pulm Med 2:429–433

22. Albafouille V, Sayegh N, CoudenhoveSD et al. (1999) CT scan patterns ofpulmonary alveolar proteinosis in children. Pediatr Radiol 28:147–152

23. Swensen SJ, Tashijian JH, Myers JL et al. (1996) Pulmonary venooclusivedisease. CT findings in eight patients.Am J Roentgenol 167:937–940

24. Cheah FK, Sheppard MN, Hansell DM(1993) Computed tomography of diffuse pulmonary haemorrhage withpathological correlation. Clin Radiol48:89–93

25. Fink CW, Cimaz R (1997) Early onsetsarcoidosis: not a benign disease. J Rheum 24:174–177

26. Tsagris VA, Liapi-Adamidou G (1999)Sarcoidosis in infancy: a case with pulmonary involvement as a cardinalmanifestation. Eur J Paedr 158:258–260

27. Gurney JW (1992) Hypersensitivitypneumonitis. Radiol Clin North Am30:1219–1230

28. Hansell DM, Wells AU, Padley SP,Muller NL (1996) Hypersensitivitypneumonitis: correlation of individualCT patterns with functional abnormali-ties. Radiology 199:123–128

29. Adler BD, Padley SP, Muller NL,Remy-Jardin M, Remy-J (1992)Chronic hypersensitivity pneumonitis:high resolution CT and radiographicfeatures in 16 patients. Radiology185:91–95

30. Ha SY, Helms P, Fletcher M, Broadbent V, Pritchard J (1992) Lunginvolvement in Langerhans’ cell histio-cytosis: prevalence, clinical featuresand outcome. Paediatrics 89:466–469

31. Egeler RM, D’Angio GJ (1995)Langerhan cell histiocytosis. J Paediatr127:1–11