Childhood Interstitial Lung Disease (ChILD) Workup

Updated: Feb 21, 2020
  • Author: Rebekah J Nevel, MD, MSCI; Chief Editor: Girish D Sharma, MD, FCCP, FAAP  more...
  • Print

Laboratory Studies

General diagnostic approach

The process and pace of evaluation depend on several factors, and no single algorithm applies to the diverse clinical settings in which interstitial lung disease (ILD) can occur. Considerations influencing the diagnostic approach include age at presentation, immunocompetence, chronicity, severity of disease, duration of illness, family history, and trend toward improvement. For example, the full-term newborn with respiratory failure is approached differently from the young child with tachypnea of insidious onset and hypoxemia with feeding or sleep.

As outlined below, some types of ILD may be diagnosed on the basis of genetic testing, and laboratory studies may provide clues for others, particularly in association with systemic disorders. Chest CT patterns may suggest certain diagnoses, although most forms of ILD require surgical lung biopsy for definitive diagnosis.

The following laboratory studies may be included in the workup:

  • Complete blood cell (CBC) count and differential: Anemia and reticulocytosis are seen in pulmonary hemorrhage. Polycythemia may be seen in chronic hypoxia. Peripheral eosinophilia suggests parasitic disease, hypersensitivity, eosinophilic syndromes, or other immune dysfunction.

  • Urinalysis may indicate coincident glomerulonephritis in patients with pulmonary-renal syndromes.

  • Stool Hemoccult results may be positive in patients with idiopathic pulmonary hemorrhage, celiac disease, or inflammatory bowel disease.

  • Sweat chloride test and cystic fibrosis (CF) genotyping may be required to exclude CF.

  • Serologic testing for Mycoplasma pneumoniae may be used.

  • Fungal serologic testing may be used.

  • Respiratory viral studies may be used.

  • Markers of inflammation, including erythrocyte sedimentation rate (ESR) and C-reactive protein (CRP) levels, may be elevated in inflammatory disorders.

Workup for immunodeficiency

The workup for immunodeficiency includes testing for immunoglobulins, immunoglobulin G (IgG), IgG subclasses, specific antibodies to vaccine antigens (tetanus, diphtheria, polyvalent pneumococcal vaccine, and Haemophilus influenzae type B), complement (C3, C4, CH50), and lymphocyte subsets.

The following tests may be included in the workup:

  • An anergy skin test panel should be considered.

  • Lymphocyte markers may be useful.

  • Immunoglobulin E (IgE) may be used to evaluate for parasitic disease, allergic bronchopulmonary aspergillosis (ABPA), and eosinophilic syndromes.

  • Human immunodeficiency virus (HIV) testing is indicated if lymphocytic interstitial pneumonia (LIP), Pneumocystis carinii pneumonia (PCP), or disseminated histoplasmosis is present.

  • Markers of rheumatic disorders, including rheumatoid factor (RF), antinuclear antibody (ANA)/anti–double-stranded DNA (anti-dsDNA) antibody, antineutrophil cytoplasmic antibodies (ANCAs), and anti–basement membrane antibodies, should be measured to determine connective tissue disease or autoimmune etiologies.

  • Levels of angiotensin-converting enzyme (ACE) and lysozyme may be elevated in patients with sarcoidosis, but these findings are neither sensitive nor specific.

  • Serum precipitin results may be positive in patients with hypersensitivity pneumonitis but do not prove disease causality.

Genetic testing

Genetic testing is evolving rapidly, and consultation with a geneticist or ILD specialist is recommended. Genetic testing is increasingly used for identification of mutations resulting in altered surfactant production (SFTPB, SFTPC, ABCA3, and NKX2.1) or surfactant catabolism (CSF2FA, CSF2RB, SLC7A7, MARS, GATA2, and OAS1).

Testing for SFTPB and ABCA3 mutations should be performed in infants with unexplained severe neonatal respiratory distress, particularly if a family history of respiratory disease is known. Testing for SFTPC and ABCA3 mutations should be considered in infants and children with childhood interstitial lung disease (chILD) syndrome, particularly if they exhibit digital clubbing, diffuse ground-glass opacities, or reticular changes on high-resolution computed tomography (HRCT) or if they have a family history of chronic lung disease.

Other genetic etiologies of chILD syndromes include the following [36, 37, 38] :

  • FOXF1, which is associated with alveolar capillary dysplasia with misalignment of the pulmonary veins
  • COPA
  • FLNA (filamin A), which is associated with X-linked periventricular nodular heterotopia, alveolar simplification, and pulmonary hypertension
  • TBX4, which is  associated with lung development alterations

Clinical genetic testing is available through Clinical Laboratories Improvement Act (CLIA)–certified diagnostic laboratories.

Other laboratory tests

Serum and urine amino acids may be measured if metabolic conditions, such as lysinuric protein intolerance, are suspected. Brain natriuretic peptide (BNP) may be used in the screening and trending of right heart strain in the setting of pulmonary hypertension.


Imaging Studies

Chest radiography

Chest radiographs are often the first imaging study performed for evaluation of possible ILD. Although they rarely allow for a specific diagnosis, they can be useful in identifying other causes of respiratory symptoms that may present similarly to ILD. [39, 40] Findings on plain chest images may be normal in the presence of active disease and abnormal in the absence of symptoms. Numerous radiographic patterns are associated with ILD, including ground-glass opacities; reticular, nodular, or reticulonodular infiltrates; and honeycombing. The ground-glass appearance is consistent with active alveolitis, and honeycombing is consistent with advanced fibrosis.

Radiographic findings are usually described as interstitial infiltrates, although predominantly nodular (alveolar) and mixed reticulonodular patterns have been described, as have nonspecific findings (hyperinflation).

CT scanning

Computed tomography (CT) scanning, specifically high-resolution CT (HRCT), provides a noninvasive means for determining the patterns, extent, and distribution of changes associated with ILD. The imaging appearance of diffuse abnormalities varies by specific chILD diagnosis; findings include ground-glass attenuation, a tree-in-bud appearance, lobular air trapping, reticular attenuations, and centrilobular nodules. [41] CT is especially useful in demarcating the most appropriate areas for tissue biopsy. [22]

Disadvantages of traditional HRCT include the need for sedation in uncooperative infants and young children and the relatively high radiation exposure. Newer, more rapid acquisition algorithms have decreased these problems.

Long et al developed a method by using a combination of sedation and controlled ventilation providing a controlled pause in respiration to allow scanning of the lung with 1-mm sections. [42] Without controlled ventilation, respiratory motion may mimic ground-glass opacities, and dependent atelectasis may occur mimicking parenchymal changes. When possible, both inspiratory and expiratory images should be obtained, as expiratory images can be important to assess for air trapping and to evaluate the extent of ground-glass opacity.

Because imaging technology is changing rapidly, consultation with a radiologist who is knowledgeable about imaging children's lungs is highly advised. Whether serial HRCT provides any benefit in monitoring disease progression or response to therapy is unclear.

In a study of HRCT in 20 children with ILD, specific patterns were correlated with certain types of pathology, with little overlap. [23]  A characteristic CT pattern appears to be associated with neuroendocrine cell hyperplasia of infancy (NEHI). [43, 44] Consolidative patterns were seen in aspiration syndromes, cryptogenic organizing pneumonia (COP), and vasculitides. Characteristic thin-walled, heterogeneous cysts, alternating with small nodules, were seen only in patients with Langerhans cell histiocytosis (LCH).

In another study, investigators evaluated the ability of expert readers to correctly diagnose pediatric diffuse lung disease with HRCT. [40] The correct first-choice diagnosis of ILD was made in 61% of CT observations, and the conditions correctly diagnosed with greatest frequency were alveolar proteinosis, idiopathic pulmonary hemosiderosis, and pulmonary lymphangiectasia.

Studies in patients with NEHI or bronchiolitis obliterans suggest that characteristic HRCT patterns are often noted in these entities and may be used for diagnosis in the appropriate clinical context. [43, 44, 45]

Barium swallow studies

Barium swallow studies, videofluoroscopic swallow studies, or radionuclide "milk" scans may demonstrate evidence of dysphagia or aspiration.


Echocardiography should be included in the initial diagnostic workup. Special attention should be paid to depiction of all four pulmonary veins, because partial anomalous pulmonary venous return may be present in patients with respiratory distress and chest radiograph findings of interstitial infiltrates.

Evidence of pulmonary hypertension (based on the tricuspid regurgitant jet velocity or septal geometry) and right ventricular hypertrophy may be evident. A systematic review found that the frequency of pulmonary hypertension in patients with chILD ranged from 1% to 64%. [34]

Magnetic resonance imaging

The use of thoracic magnetic resonance imaging (MRI) has been proposed, given the absence of ionizing radiation; however, because of the relatively low proton content of the lungs, conventional MRI has less utility for evaluating the pulmonary parenchyma than it has for assessing the mediastinal structures. [46]  MRI has been used specifically to evaluate sarcoidosis in children. [47]

MRI technology is rapidly evolving. Techniques such as the use of inhaled hyperpolarized gas for a contrast agent and ultrashort echo time MRI are increasingly applied in clinical settings. [46, 48, 49, 50]



Other Tests

Pulse oximetry

Decreased oxyhemoglobin saturation more often reflects ventilation-perfusion mismatching, rather than diffusion abnormalities, because of the remodeling of distal airspaces characteristic of most chILDs. In the early stages of ILD, oxyhemoglobin saturation may be relatively normal at rest but may worsen dramatically with exercise or sleep.

Pulmonary function testing

In children and adolescents who can perform spirometry and plethysmography, total lung capacity (TLC), forced vital capacity (FVC), and forced expiratory volume in 1 second (FEV1) are all often reduced. Although TLC maybe reduced, functional residual capacity (FRC) and residual volume (RV) are often normal or elevated, resulting in increased FRC/TLC and RV/TLC ratios. Airflow limitation, as indicated by a reduced FEV1/FVC ratio, is present in as many as one half of children with ILD. Compliance of the respiratory system (Crs) is reduced.

Diffusing capacity for carbon monoxide is usually low, although this value often returns to normal when corrected for lung volume and hemoglobin. In pulmonary hemorrhage syndromes, diffusing capacity may be elevated because of the affinity of carbon monoxide for sequestered hemoglobin.

Pulmonary function tests (PFTs) have been safely performed in sedated infants at some pediatric centers. Results of PFTs in infants, if available, usually show reduced Crs using both multiple occlusion and end-inspiratory occlusion techniques. PFTs have been used to monitor the response to treatment in some studies. [51]

Additional markers of pulmonary function, including lung clearance index (LCI) and oscillometry, have been used in other pulmonary disorders; however, they have not been systematically studied in chILD.

Results of arterial blood gas analysis may be normal, but typical changes include decreased arterial partial pressure of oxygen (PaO2) and respiratory alkalosis.

Exercise testing

In children old enough to cooperate, exercise testing may reveal exercise-related desaturation, even when oxyhemoglobin saturation is normal during rest. Exercise testing, or a 6-minute walk test, may provide an objective indicator of disease progression.

pH/impedance probe testing

pH or impedance probe testing may be required to demonstrate gastroesophageal reflux (GER), predisposing patients to aspiration. GER may occur as a secondary complication of ILD.


Electrocardiography (ECG) readings may show evidence of cor pulmonale, specifically right atrial and ventricular enlargement, and right axis deviation.



Bronchoalveolar lavage

Bronchoscopy with bronchoalveolar lavage (BAL) is useful in diagnosing certain conditions in the differential diagnosis of ILD, including alveolar proteinosis, aspiration syndromes, pulmonary hemosiderosis, eosinophilic syndromes, and various infections. Occasionally, results of cytologic analysis may be diagnostic—for example, when Langerhans cells are present, indicating histiocytosis. Most authorities believe BAL should precede biopsy.

Problems with the use of BAL include the lack of a standardized methodology in children, the paucity of reference values for differential cell counts, the variability of BAL findings at different times in a disease course, and the lack of correlation between BAL and histologic findings.

Fluid should be sent for differential cell counts, cultures and special staining for bacteria (including mycobacteria) and fungi, cytologic analysis (including oil red O staining for lipid-laden macrophages and staining for hemosiderin), pepsin levels, and viral diagnostic studies.

BAL findings can be diagnostic of pulmonary alveolar proteinosis (PAP), demonstrating a cloudy or milky appearance of the fluid with periodic acid-Schiff (PAS)–positive amorphous debris. Electron microscopy may reveal lamellar membranous structures in a fibrous and granular matrix with degenerative cells. [52] Increased eosinophils (>30% of total) are consistent with eosinophilic pneumonia syndromes, whereas predominant lymphocytosis can be associated with hypersensitivity pneumonitis. CD1a-positive cells are diagnostic for Langerhans cell histiocytosis. [53]

A study evaluated quantitation of levels of surfactant proteins B (SP-B) and C (SP-C) in BAL fluid for the diagnosis of chILD and reported that low SP-C levels may suggest diseases caused by mutations in NKX2.1, SFTPC, ABCA3, and other genes involved in surfactant metabolism. SP-B levels may be used to screen for SP-B deficiency; however, given the availability of genetic testing, this approach is not the standard of care for diagnosis. [54]

Lung biopsy

Analysis of tissue obtained during lung biopsy can be the best route to a diagnosis if it cannot be established by noninvasive means. For NEHI, an expert clinician can be confident of a presumptive diagnosis based on the clinical history and characteristic HRCT findings, but this is not true of most other chILD entities. An increasing number of chILDs can be diagnosed by genetic testing.

Much of the classification of ILD, especially in disorders of unknown cause, is based on histopathology. However, it is important to note that a histopathologic pattern is not synonymous with a diagnosis in many cases. To maximize the diagnostic yield, a pediatric lung biopsy protocol has been developed and supported by the ChILD Research Network. [55] However, a diagnosis is not reached in all patients, even after biopsy is performed.

The number of biopsy procedures performed and the method used (eg, open vs thoracoscopic) have little influence on diagnostic yield. Diagnostic yield may be enhanced if HRCT is used to direct the biopsy sites. The biopsy sample should be taken from a region of involvement. If diffuse involvement is found, any site (except the tip of the right middle lobe or lingua) is appropriate.

Communication between the clinician, surgeon, pathologist, and radiologist before the biopsy is useful and appropriate for determining biopsy sites and prioritizing use of the tissue.

Compared with open lung biopsy, thoracoscopic biopsy shortens the surgical time, duration of chest tube placement, and hospital stay without substantially altering the diagnostic yield. The choice between thoracoscopic and open approaches should be left to the consulting surgeon. Transbronchial biopsies are not recommended for chILDs.

Regardless of the method used, biopsy samples should be processed for bacterial, fungal, and mycobacterial cultures and staining, including special staining, light microscopy, immunofluorescence, and electron microscopy. Immunostains, such as bombesin for NEHI and vimentin for pulmonary interstitial glycogenesis (PIG), may aid in the diagnosis of specific forms of ILD. [56]

Cardiac catheterization

This procedure should be considered in any child with noninvasive evidence of pulmonary hypertension but especially in children with a history of hemoptysis or absence of crackles on examination. These findings have been correlated with pulmonary veno-occlusive disease.


Histologic Findings

Histologic findings on routine hematoxylin and eosin staining remain the criterion standard for the diagnosis and classification of many types of ILD and may indicate the underlying cause, if a cause is present, or may suggest associated systemic illnesses (eg, noncaseating granulomas characteristic of granulomatous infections or sarcoidosis). However, this area has been rife with confusion because of different classification schemes, inexact nomenclature, and important differences between children and adults. Consultation with pathologists experienced in chILD is critical and ideally should be sought before biopsy specimens are obtained.

ILD classification systems

Adult histologic classifications for ILD are of questionable relevance to chILD, because particular patterns may have differing diagnostic and prognostic significance in adults and children. A classification system for infants with ILD that is based on a multicenter retrospective review of histopathology divides chILD entities into disorders more prevalent in infancy, disorders not unique to infancy, and unclassified disorders. [1]


Histologic Patterns of Interstitial Lung Diseases More Prevalent in Infancy

Follicular bronchitis/Bronchiolitis

Lung biopsy findings in follicular bronchiolitis demonstrate follicular lymphocytic infiltration surrounding and infiltrating the bronchial walls. This pattern may be seen in immunodeficiency disorders and connective tissue diseases. [43, 57, 58, 59]  See the image below.

Follicular bronchiolitis (A) Chest high-resolution Follicular bronchiolitis (A) Chest high-resolution CT (HRCT) scan from a patient with common variable immunodeficiency with history of anemia, thrombocytopenia, recurrent pneumonia, chronic cough, and exercise intolerance. Mosaic attenuation is present diffusely throughout the lungs. Extensive hilar and mediastinal lymphadenopathy is also present. Air-trapping was seen on expiratory images (not shown). (B) Lung histopathology demonstrates severe airway-centric lymphocytic inflammation with reactive follicles, which infiltrates and obscures most bronchioles.

Cellular interstitial pneumonitis/Pulmonary interstitial glycogenosis (PIG)

Lung biopsy findings reveal interstitial proliferation of histiocytic type cells with minimal to no infiltration. [57]  Electron microscopy demonstrates primitive interstitial cells with abundant cytoplasmic glycogen. [60]  PIG presents in infancy and is associated with alveolar growth abnormality. [61, 62]

Genetic abnormalities of surfactant function/Chronic pneumonitis of infancy

Biopsy findings reveal alveolar septal thickening, prominent pneumocyte hyperplasia, and alveolar exudates with numerous macrophages along with rare eosinophils and cholesterol clefts. [57, 63]

The typical histologic pattern for SP-B deficiency is interstitial thickening, abundant type II cell hyperplasia, and eosinophilic PAS-positive granular material in the alveolar space. The eosinophilic granular material is typical of PAP. Immunohistochemical staining may demonstrate the absence of SP-B. [64, 65]

Adults with SP-C deficiency can have biopsy findings consistent with those of usual interstitial pneumonia (UIP), desquamative interstitial pneumonia (DIP), or nonspecific interstitial pneumonia (NSIP). [57, 66]

The histopathology of ABCA3 deficiency can widely vary and appears to be age-dependent, with infants exhibiting patterns consistent with PAP, chronic pneumonitis of infancy, or DIP, and older children exhibiting features of NSIP. [25]  The only well-documented case of UIP in a child was an adolescent with ABCA3 deficiency. [67]

Neuroendocrine cell hyperplasia of infancy (NEHI)

Lung biopsy in NEHI reveals mild, nonspecific changes, including airway smooth muscle hyperplasia, increased alveolar macrophages, and increased airway clear cells. Immunostaining of the cells demonstrates strong staining for bombesin and serotonin, which identifies these cells as pulmonary neuroendocrine cells (PNECs).

See the image below.

Neuroendocrine cell hyperplasia of infancy (NEHI) Neuroendocrine cell hyperplasia of infancy (NEHI) (A) Chest high-resolution CT (HRCT) scanning (at total lung capacity) in a 6-month-old infant with tachypnea, hypoxemia, and failure to thrive. Sharply defined areas of ground glass opacity are seen most prominently in the right middle lobe and lingula. Diffuse air-trapping was seen on expiratory images (not shown). No additional abnormalities were identified. (B) Hematoxylin and eosin staining of the lung biopsy specimen reveals near-normal lung architecture. (C) Bombesin immunostaining demonstrates increased numbers of neuroendocrine cells.

Histologic Patterns of Interstitial Lung Disease Not Limited to Children

Pulmonary alveolar proteinosis (PAP)

This histologic pattern may be seen in disorders of surfactant dysfunction, granulocyte-macrophage colony-stimulating factor (GM-CSF) signaling dysfunction or autoantibodies, and lysinuric protein intolerance. On histologic examination, PAP appears as dilated air spaces and amorphous material with scattered pulmonary macrophages. Electron microscopy may demonstrate characteristic findings. Silver staining should be performed to rule out Pneumocystis pneumonia.

Desquamative interstitial pneumonia (DIP)

In adults, DIP is usually seen in cigarette smokers. In children, the DIP histologic pattern is associated with some of the surfactant dysfunction mutations, such as ABCA3 deficiency, and with chronic pulmonary hemorrhage syndromes.

The histologic pattern is uniform and diffuse. Alveoli are filled with accumulations of macrophages, which were originally believed to be desquamated alveolar epithelial cells, hence the name.

Nonspecific interstitial pneumonia (NSIP)

NSIP is a confusing term (a specific pattern despite the inclusion of “nonspecific”) for a microscopically homogeneous pattern of inflammation and fibrosis, although gross involvement may be patchy. Honeycomb changes are rare. Patchy ground-glass attenuations are depicted on HRCT scans. This histologic pattern may be seen in connective tissue diseases and rarely in SFTPC mutations. See the image below.

Nonspecific interstitial pneumonitis. (A) Chest hi Nonspecific interstitial pneumonitis. (A) Chest high-resolution CT (HRCT) scanning from a 10-year-old with systemic sclerosis and progressive exercise intolerance. (B) Lung biopsy showed multiple abnormalities including a relatively diffuse interstitial process with mild chronic inflammation, abundant fibroblastic tissue and patchy dense interstitial fibrosis. Accumulation of alveolar macrophages is seen in the airspaces, with rare foci of organizing pneumonia. Pulmonary arteries demonstrated focal intimal hyperplasia and medial hypertrophy, and the pleura contains patchy chronic inflammation. This overall constellation of findings is generally classified as mixed cellular and fibrotic nonspecific interstitial pneumonia (NSIP) and is a pattern most commonly seen in the setting of underlying collagen vascular disease.

Diffuse alveolar damage (DAD)/Acute interstitial pneumonia (AIP)

AIP is seen following acute lung injury, with a histologic pattern of diffuse active fibrosis (many proliferating fibroblasts, little collagen) with uniform alveolar septal thickening. Features of acute lung injury and a proliferative state of DAD are present, such as hyaline membranes, acute inflammation, and bronchial epithelial atypia.

Organizing pneumonitis/Cryptogenic organizing pneumonitis (COP)

COP (previously bronchiolitis obliterans organizing pneumonia [BOOP]) appears as patchy areas of granulation tissue in conducting airways and alveolar ducts with inflammation (primarily macrophages) in the surrounding alveoli. In affected areas, the appearance is uniform without significant distortion of lung parenchyma. Buds of myofibroblasts in collagenous stroma (Masson bodies) may extend into adjacent airspaces, demonstrating a characteristic butterfly pattern. This pattern may be associated with a broad differential diagnosis (including HIV infection and bone marrow transplantation) and may occur with other histologic patterns. See the image below.

Bronchiolitis obliterans. (A) Chest CT scanning fr Bronchiolitis obliterans. (A) Chest CT scanning from an 8-year-old demonstrates irregular large mosaic regions of ground-glass opacity and air-trapping, as well as the presence of peribronchial thickening and bronchiectasis. (B) Pathology demonstrates focal areas of fibrosis with polypoid plugs of fibroblastic cells and fibrin filling distal bronchioles and airspaces (hematoxylin and eosin).

Lymphocytic interstitial pneumonia (LIP)

LIP is characterized by monotonous, diffuse, lymphoplasmacytic cell infiltrates in the interstitium and distal airspaces. Mononuclear cells and histiocytes are also seen. Occasionally, lymphoid aggregates are found in a lymphatic or angiocentric distribution. LIP has been described as a pulmonary manifestation of HIV infection in children. In addition, LIP is associated with Sjögren syndrome, chronic active hepatitis, and juvenile rheumatoid arthritis. Epstein-Barr virus (EBV)-genomic DNA occasionally can be identified in LIP. Lymphoproliferative diseases, lymphomas, and underlying immunodeficiency must be considered.

Usual interstitial pneumonitis (UIP)

UIP is the histopathologic pattern associated with idiopathic pulmonary fibrosis (IPF), a disease of adulthood. UIP is characterized by a heterogeneous appearance at low magnification, with alternating areas of normal lung, inflammation, fibrosis, and honeycomb changes, which are most prominent in the peripheral subpleural areas. Fibrotic areas contain dense collagenous deposits and characteristic foci of proliferating fibroblasts (fibroblastic foci), which have negative prognostic importance in IPF.

UIP is extremely rare in children, and its presence on a pediatric biopsy should be confirmed by a pathologist with significant expertise in chILD. There is only one well-characterized report of UIP in a child, an adolescent with ABCA3 deficiency. [67]  UIP at an early age has also been described in dyskeratosis congenita, with the onset of interstitial lung involvement in childhood in some cases. [68]



No widely used staging system is available for chILD, which is appropriate because the spectrum of possible diagnoses is large.

In adults, a scoring system is available for IPF, based on clinical, radiographic, and pathologic findings (ie, CRP scoring system).

Fan devised a simple scoring system for chILD. A score of 5 indicates the worst outcome, with a 38% survival rate at 60 months. A score of 2, 3, or 4 indicates a survival rate of 76%. Data from Cox proportional hazards modeling suggests a 140% increase in risk of death with each unit increase in score. The score can be applied to most ILD diagnoses with the likely exception of NEHI, which has a good eventual prognosis in the majority of cases.

The Fan severity scoring system is as follows [69] :

  1. Asymptomatic

  2. Symptomatic with normal oxyhemoglobin saturation

  3. Symptomatic with nocturnal or exercise-induced desaturation

  4. Desaturation at rest

  5. Pulmonary hypertension