Introduction
Hypersensitivity pneumonitis (HP) - in Europe called extrinsic or exogenous allergic alveolitis (EAA) - is a complex syndrome incited by numerous inhaled agents including agricultural dusts, bio-aerosols, fungal-, bacterial- or protozoan microorganisms, and certain reactive chemicals. In children it is a relatively uncommon condition and the two major inciting allergens are bird (avian) allergens including down and inhaled particles derived from fungi, like thermophilic actinomycetes, or rarely fusarium [
1], aureobasidium [
2,
3] and epicoccum [
4].
A previous NHLBI/ORD Workshop has summarized state of the art and the needs and opportunities for research in HP [
5]. It was stated that because pediatric cases of HP are rarely recognized or reported, knowledge is limited and is based mostly on case reports and small series of patients. Between 1960 and 2005, 95 cases of HP in children have been reported in the literature [
6‐
8]. In contrast to the data from adults, 95% of the cases were males and 25% had a family history of HP [
5]. This hints to some reporting bias. The finding of clubbing in 31% (10/32) of the children, suggested that in the past the disease was recognized late in its clinical course. Importantly, as deaths from HP have been reported in children as well as adults [
5,
6], treatment may be more difficult than anticipated. This is highlighted by the fact that 3% of the children were not treated with removal from the exposure. Also, treatment with corticosteroids is very controversial. In addition to oral long term therapy most frequently done (about 66%) [
5], methylprednisolone pulse therapy [
9,
10] or inhaled budesonide [
11] were suggested. Current recommendations in adults, as well as our personal practice in children, clearly suggest no steroid treatment at all, if possible [
12].
Due to its rarity, many paediatricians and general practitioners are likely not to be very familiar with the clinical presentation and diagnosis of HP, and many pediatric pulmonologists may not use up to date treatment. All these issues, including an unknown proportion of cases of interstitial lung disease in children which may represent undiagnosed HP, suggested that current information on the diagnosis and treatment of HP is warranted. We used a survey that we performed to determine the incidence of pediatric diffuse parenchymal lung diseases in Germany, to learn more about the current status of HP in children in Germany [
13]. The goal of this study was to assess current means of diagnosis and therapy of HP, to compare this to the recommendations and to propose ways to improve future management with the help innovative strategies for rare lung diseases.
Discussion
From this work 3 main conclusions can be derived; (1) diagnosis of pediatric HP is frequently late, despite characteristic symptoms at presentation, (2) allergen avoidance as the principal treatment is not always followed very strictly, and (3) prolonged and high dose steroid treatment is often used.
Although HP is the most frequent chronic interstitial lung disease in children [
13], a quarter of the children with HP presented here and incident during a 3 y period in Germany, were diagnosed in chronic disease state. This together with the fact that lung biopsies were done at a relatively high frequency, suggests that some difficulties have to be overcome to diagnose this condition in childhood. All children, except one who was not measured, but was hypoxemic at rest (O2-saturation 80%), presented with a severely (≤65% pred.) restricted lung function and almost all had reduced oxygen saturation at rest or with exercise (Additional file
1: Table S3). These findings are very characteristic for presenting HP [
17], nevertheless in 70% of the children an atypical pneumonia was diagnosed and empiric antibiotic treatment was started. Lung biopsy helped in all 6 instances when performed to substantiate the diagnosis and showed in all cases variable mural and luminal lymphocytic alveolitis as well as bronchiolitis, whereas loosely formed non-caseating histiocytic granulomata with multinucleated giant cells where found only in three cases; although in retrospect, diagnosis might have been possible from the combination of characteristic clinical features including the presence of serum precipitins, BAL lymphocytosis (except for 1 case), and FVC below 50% pred.
CT scans have recently been demonstrated to significantly contribute to the initial diagnosis of HP in adults, with specificity of 81% [
5]. Their value in children has not yet been demonstrated. In this series in only 13 of 23 children CT’s were done, i.e. a diagnosis was reached in more than 40% without a CT scan. In addition, a review of the CT scans obtained demonstrated studies of variable quality. Only 9 of 13 cases with CT had a quality judged as good, and the quality was moderate in 3 and poor in 1. Major reasons for the failure were artifacts from respiratory motion. As already proposed for adults, these results highlight the need for and compliance with standardized protocols [
18]. The characteristic radiologic features known from adult cases of HP were seen in these pediatric patients to similar extents [
19]. Of interest was the high degree of concordance in the frequency of abnormal chest radiograph and CT findings. However it must be cautioned as only the chest CT findings “are characteristic” of HP.
The overwhelming majority of HP in the pediatric population is due to bird and fungus allergens; one of the challenges may be to identify alternative sources of these allergens, like fungus contaminated indoor hydroponics [
3], misting fountains [
20], basement showers [
4] or possibly even wild city pigeons [
21]. Up to 25% of the cases reported previously in the pediatric literature [
5] and few additional case reports [
2,
3,
22] were “familial” cases. Reasons may include (1) publication bias from preferring familial cases; (2) insensitivity of history to detect alone standing early disease, or (3) a higher likelihood for serologic testing in families with affected other members. Although common genetic predisposing factors may also be involved, they have not been demonstrated by now. Familiarity in the context of HP is most likely due to a common exposition to the antigens.
Identification of the responsible allergen is critical for avoidance and the causal and principal treatment [
12]. Of interest, almost all children identified here were treated with systemic steroids. The advantage of such an approach is a more rapid therapeutic response with steroid treatment; however this introduces the impossibility to monitor the completeness of allergen avoidance measures. Lack of antigen avoidance and non-effectiveness of corticosteroid therapy results in pulmonary fibrosis and end-stage lung disease with death [
23] or the need of lung transplant at young age [
24]. Thus for treatment primarily careful allergen avoidance with associated clinical improvement is warranted. This proposition needs to be tested in a randomized controlled trial using steroids or placebo.
A weakness of the study may be that we recovered most, but not all incident cases, as one, published as a single case, and came to our attention during manuscript preparation [
3]. Secondly, it is possible that some chronic forms might have been misdiagnosed as severe, steroid-resistant asthma [
21]. Although there was a very high preponderance of male children (95%) in the pediatric cases reported until 2005 [
5], evened out sex ratios were demonstrated recently in a population based study [
25]. Therefore the male to female ratio of 9/14 found in the present study, suggests an almost homogenous sampling. Lastly, long term follow up was not possible with the design of the study, based on the German Surveillance Unit which is not prepared for longitudinal follow up. This can be achieved with register studies.
Among the strengths of the study is the relatively large cohort of contemporary and newly detected pediatric HP cases collected over a brief period of time. The observed current approach to diagnosis and therapy clearly demonstrates that the index of suspicion for HP needs to be increased substantially and that an early diagnosis must be established by much more quality controlled assessment of the clinical, radiological and laboratory findings. In particular, allergen avoidance is key for management. The role of corticoids which are generally used needs to be defined in prospective clinical trials.
These issues are surpassingly suited to be assessed in an international study of this rare entity. Capture and long term follow up of the widely scattered cases of HP in children occurring in remote places can be easily overcome by web-based studies within rare disease registries (
http://www.kids-lung-register.eu). As such, the recently funded proposal “Orphans Unite: chILD better together – European Management Platform for Childhood Interstitial Lung Diseases” (
http://www.childeu.net) under the FP7 program will provide an excellent base for this important task.
Acknowledgements
Supported by Kids-Lung-Register (KLR) and BMBF, GOLDnet.
The authors are indebted to all the individuals who participated, and without whom the study would not have been possible.
National EAA Study Group
Tobias Ankermann, Children’s Hospital, University of Kiel, Germany; ankermann@pediatrics.uni-kiel.de
Katja Breuel, Children’s Hospital, University of Rostock, Germany; katja.breuel@med.uni-rostock.de
Achim Freihorst, Children’s Hospital, Aalen, Germany; Achim.Freihorst@ostalb-klinikum.de
Joern-Lorenz Gröbel, Klinikum Detmold, Germany; joern-lorenz.groebel@klinikum-lippe.de
Christoph Härtel, Children’s Hospital, University of Lübeck, Germany; haertel@paedia.ukl.mu-luebeck.de
Heinz Huprich, Children’s Hospital Datteln, Germany; H.Huprich@kinderklinik-datteln.de
Wolfgang Kamin, Children’s Hospital Hamm, Germany; sk-paediatrie@evkhamm.de
Joachim Lemke, Wilhelmstift Hamburg, Germany; j.Lemke@kkh-wilhelmstift.de
Frank Mandelkow, Kreiskrankenhaus Hagenow, Germany; f.mandelkow@krankenhaus-hagenow.de
Walther Mihatsch, Children’s Hospital, Schwäbisch Hall, Germany; walter.mihatsch@diaksha.de
Claus Pfannenstiel, Kinderarztpraxis Aachen, Germany; pfannenstiel@kinderarztpraxis-laurensberg.de
Michael Rau, Oberschwabenklinik, Ravensburg, Germany; michael.rau@Oberschwabenklinik.de
Johannes Schulze, Children’s Hospital, University of Frankfurt, Germany; Johannes.Schulze@kgu.de
Nicolaus Schwerk, Hannover Medical School, Germany; schwerk.nicolaus@mh-hannover.de
Jens-Oliver Steiß, Children’s Hospital, University of Gießen, Germany; Jens-Oliver.Steiss@paediat.med.uni-giessen.de
Heino Skopnik, Children’s Hospital Worms, Germany; paediatrie@klinikum-worms.de
Christian Teusch, München, Germany; c.teusch@me.com
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
MG designed the study protocol, analyzed the data, and wrote the draft of the manuscript, and is the study guarantor, with full responsibility for the finished article, access to any data, and control of the decision to publish. MH and DH collected and analyzed the data, VT and JGP performed analysis of the imaging data, FB did the pathological analysis. MH, DH, VT, JGP and FB helped to write the manuscript. All the members of the National EAA study group contributed cases and supplied clinical data. All authors read and approved the final manuscript.