Pathophysiology

Genetic mutations affecting SFTPB, SFTPC, ABCA3 and TTF1 have been identified as the main causes of SPDS. Experimental and clinical findings have suggested several possible pathophysiological mechanisms, resulting in parenchymal inflammation and interstitial lung disease (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Hypothesised pathophysiology of surfactant protein dysfunction syndromes. Abnormal surfactant protein function or production with resultant atelectasis, abnormal intracellular trafficking, cellular metabolism and activation of signalling cascades resulting in apoptotic, and abnormal morphogenesis have all been postulated to be involved. ABCA3, ATP binding cassette number A3; EM, electron microscopy; ER, endoplasmic reticulum; SP-A, surfactant protein B.

SFTPB mutations are inherited in an autosomal-recessive manner and result in production of abnormal protein and a failure to function normally in reducing surface tension. Interestingly, surfactant from SFTPB infants also lacks mature SP-C, which is thought to be the result of failure to process pro-SP-C to SP-C. Electron microscopy shows abnormal or absent lamellar body structure or absence altogether.8

SFTPC mutations can be inherited in an autosomal-dominant manner with variable penetrance or can arise from de novo mutations with no family history. Disease is not due to loss of function, but to deleterious effects of accumulation of misfolded proteins. Gene mutations result in production of misfolded protein, triggering a cellular response that leads directly to endoplasmic reticulum stress, proteasome dysfunction and AEC2 apoptosis.9

ABCA3 mutations are inherited in an autosomal-recessive manner and result in the production of protein that functions abnormally. The precise mechanism with which ABCA3 mutations result in the respiratory phenotype is not fully understood, though studies in patients have shown lower levels of SP-B and SP-C10 ,11 and the production of abnormal lamellar bodies.10 ,12

Mutations to NKX2.1 that result in inactivation of a single allele are sufficient to cause disease. In vitro experiments show that common NKX2.1 mutations result in production of protein with reduced DNA binding and transcriptional activation,13 ,14 with reduced SP-B, SP-C and ABCA3 levels in bronchoalveolar lavage (BAL) fluid.14 ,15 Other histopathological findings in patients with NKX2.1 mutations include emphysematous changes and low alveolar counts, together with reduced airway generations suggesting abnormal lung morphogenesis in disease pathogenesis.15 ,16 Surfactant protein function, genetic mutation and inheritance pattern are summarised in table 1.

Clinical features

The clinical presentation of SPDS varies greatly depending on both the protein involved, the precise genetic mutation and the phenotypic expression (tables 2 and 3). The timing of presentation and degree of severity can give useful insights into the underlying genetic mutation. For example, while SFTPB presents in the neonatal period with acute respiratory distress, SFTPC and ABCA3 can present at any age with varying severity of clinical features, ranging from young children with cough, tachypnoea and failure to thrive, to asymptomatic adults or those with features of an interstitial lung disease (eg, breathlessness on exertion, restrictive lung function testing, cough, tachypnoea and hypoxaemia). NKX2.1 classically presents with a triad that includes neurological features and hypothyroidism.

Mutations in SFTPB are extremely rare, with an estimated incidence of <1 in one million live births. The disease presents in early infancy with clinical and radiographic features similar to RDS of the newborn or persistent pulmonary hypertension of the newborn (PPHN).22 However, while premature infants with RDS or PPHN improve with standard therapies, children with SFTPB fail to do so, with the majority dying within the first 3–6 months of life despite supportive therapy.3 ,22 There have been several reports of children with alternative SFTPB mutations and milder phenotypes, though this is rarely so.23 ,24 Such milder disease phenotypes are thought to result from mutations that allow for production of enough functionally normal SP-B to prevent development of severe lung disease.

SFTPC has a highly variable age of onset and disease course with a greater tendency to present later in childhood.3 Neonatal lung disease due to SFTPC is less common than in SFTPB and ABCA3. One case series of 22 children (median age 3 months, range 0–24 months) demonstrated that initial presentation ranges from neonatal respiratory distress to persistent cough, tachypnoea and failure to thrive, though almost all patients were tachypnoeic and hypoxaemic at presentation.29 SFTPC may also present with recurrent non-resolving respiratory-syncytial virus (RSV)-positive bronchiolitis and pulmonary exacerbations,29 with in vivo SFTPC mouse models showing increased severity of RSV-induced pulmonary inflammation.34 SFTPC has been found to be a rare cause of idiopathic pulmonary fibrosis in adults, with many presenting only with mild respiratory symptoms.31 One case series of five patients demonstrated survival and near normal lung function in all patients after 27 years follow-up.30

Mutations in ABCA3 appear to be the most common cause of SPDS.3 ,12 Compared with SFTPB, disease caused by ABCA3 mutations varies in its presentation and severity. Clinical presentation in infancy may be identical to that in SFTPB with acute severe neonatal respiratory disease and high mortality.12 Alternatively, children may develop a subacute progressive lung disease that results in early mortality or pulmonary function may stabilise and even improve.3 ,27 A significant number of cases present later in childhood or early adulthood with cough, dyspnoea, exercise intolerance and chest wall deformities being the most commonly reported symptoms.11 ,27 Heterozygosity for ABCA3 mutations has been shown to modify the severity of lung disease associated with an SFTPC mutation.35

Autosomal-dominant mutations of NKX2.1 classically present with a triad (brain–lung–thyroid syndrome) of neurological features, hypothyroidism and lung disease.21 Chorea is the most specific neurological manifestation, though other findings include ataxia, developmental delay and hypotonia. Lung involvement may present in the neonatal and infant period with acute respiratory distress and progression to early respiratory failure,15 ,33 while other cases have described a chronic phenotype with survival into adulthood.

Radiology

All neonates and children presenting with respiratory distress, significant or unexplained respiratory symptoms should undergo chest radiography. Findings typically include diffuse alveolar or interstitial infiltrates affecting all lung fields (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 2

Chest radiograph and high-resolution CT (HRCT) of a 3-year-old child with SFTPC. Chest X-ray shows diffuse interstitial shadowing. HRCT showing generalised ground-glass opacification.

High-resolution computed tomography (HRCT) is used for further characterisation of radiographic abnormalities. While HRCT is unable to give a precise diagnosis of SPDS, findings may provide support or increase suspicion when taken in context with the clinical picture. Characteristic lesions found in SPDS include ground-glass opacification,27 ,29 and thickening of interlobular and intralobular septa27 ,36 (figure 2). These appearances may evolve with time, with resolution of ground-glass opacities and development of parenchymal cysts in later life.27 ,31

A volumetric HRCT scan during inspiration with thin-slice expiratory scan provides the best quality, high spatial resolution images for children with suspected SPDS.1 In younger children unable to reliably hold their breath in inspiration or expiration, control of ventilation will require a general anaesthetic or sedation with bag-mask ventilation. While the risks associated with anaesthesia must be considered, for infants and young children, doing HRCT with ventilatory control increases the likelihood of optimal images, providing important evidence towards a diagnosis of SPDS.4

Genetic testing

Genetic testing now enables non-invasive diagnosis of SPDS to be made, negating the need for histology on lung biopsy. Testing for surfactant protein mutations is currently available in Europe, the USA and Australia, and was introduced in the UK in 2006.3 A list of clinical laboratories that perform genetic testing is available through the GeneTests website (http://www.genetests.org). While there have been increasing numbers of referrals for genetic testing, a retrospective review of all such referrals found that an identifiable mutation was identified in 7.5% of cases, 25 out of 427 referrals over a 5-year period.3 Of the new cases, the majority (23 out of 25) were born after 37 weeks gestation. One infant born to consanguineous parents at 32 weeks gestation was found to have SFTPB.3

Genetic testing of the index case gives important prognostic information on the affected individual as well as obviating the need to undergo invasive lung biopsy.19 A positive result allows for avoidance of unnecessary and ineffective empirical therapies, may inform families of appropriate expectations of therapy and goals, and counsel families on risk of recurrence in future pregnancies and the availability of prenatal diagnosis if desired.

As is the case in genetic diagnosis for any rare condition, there are a number of important limitations that physicians should be aware of. From a practical perspective, genetic testing is relatively expensive, with current UK costs of £460 for combined SFTPB and SFTPC and £675 for ABCA3 sequencing. However, with advances in sequencing technology, these costs are anticipated to drop to around £650 for all three genes. Results of genetic testing may take several weeks to return, making its clinical relevance less useful in cases where children are critically unwell and require aggressive pulmonary support. The sensitivity and specificity of genetic testing has yet to be formally evaluated, and as such the false positive and false negative rates are unknown. While a large array of genetic mutations have been identified, there are reported cases of patients with no identified genetic mutation despite clinical, histological and family history suggestive of an SPDS.37 Chromosomal deletions affecting large exon regions or non-translated intron regions resulting in a significant phenotype may go undetected on current PCR-based sequencing assays. Specific assays can detect the gene dosage required in cases with a high clinical suspicion but unremarkable genetic testing.38 ,39 Future sequencing strategies should include deep intronic mutations, which are likely to account for SPDS with negative genetic studies.39 Furthermore, there are likely to be mutations affecting yet undiscovered genes that result in similar phenotypes to those that cause known SPDS. Sometimes, the phenotype may not be consistent with the results of genetic testing (incomplete penetrance), though in practice these results would have less clinical relevance.

Bronchoscopy with BAL

Bronchoscopy with BAL is frequently undertaken in infants and children with unexplained or suspected interstitial lung disease. It is readily availability and relatively safe. However, analysis of BAL fluid in SPDS does not currently yield important or clinically significant results. Its main role is predominantly for research. SP-A and SP-D levels are raised in chILD and levels appear to correlate with disease severity;40 however, their precise relationship with SPDS has not been formally investigated. One single study suggested that SP-C levels in BAL fluid are increased in patients with SFTPC29 while a few reports have tentatively suggested that SP-B, SP-C and surfactant phospholipid levels are reduced in NKX2.1.33 Despite the lack of a clear biomarker for SPDS, BAL fluid analysis may be useful in diagnosing alternative causes of respiratory distress, such as infection, eosinophilic lung disease, alveolar proteinosis and pulmonary haemorrhage syndromes.4 ,41 Bronchoscopy under general anaesthesia can be scheduled to coincide with an HRCT scan or lung biopsy to obviate the need for repeated anaesthetics. Protocols for flexible bronchoscopy have been recently described by the chILD-EU collaboration.4

Lung biopsy and histopathological diagnosis

Lung biopsy can provide a histopathological diagnosis in cases where the genetic diagnosis is equivocal or where rapidity of disease progression does not allow time for genetic testing. Desquamative interstitial pneumonitis is classically associated with the genetic mutations that disrupt normal surfactant production and function (table 3).11 Other histopathological findings may include chronic pneumonitis of infancy and pulmonary alveolar proteinosis, which is particularly associated with SFTPB and ABCA3.12 ,27 The risks of invasive tissue biopsy have led to increasingly early genetic testing for suspected SPDS. These include risks associated with the anaesthesia, which can be significant in an acutely unwell child with respiratory distress, and local complications such as bleeding, infection and pneumothorax.

Tissue may be obtained through several procedures: open thoracotomy, video-assisted thoracoscopy (VATS), transbronchial or percutaneous needle biopsy. VATS and open thoracotomy provide a clinical diagnosis in over half of cases but are more invasive procedures.42 Endobronchial and transbronchial approaches are not recommended for suspected SPDS due to the difficulties in obtaining sufficient quantities of tissue for histopathological examination. At least two superficial sites should be biopsied and samples should be transported directly to an experienced pathologist for further analysis.4 ,43 Depending on the clinical presentation, histopathological analysis might include H&E staining, microbiological studies (bacterial, viral and fungal), immunohistochemistry (CD1a-positive cells are suggestive of Langerhans cell histiocytosis and an elevated CD4/CD8 ratio is suggestive of pulmonary sarcoid), vascular staining (CD34 and elastic van Gieson) and examination for signs of pulmonary haemorrhage (Perls' stain, Oil Red O stain and Periodic acid–Schiff).4

Laboratory

The role of biomarkers in the diagnosis or determination of prognosis of SPDS in children has not been established. No biomarkers are available in routine clinical practice. One small study identified the potential use of serum glycoprotein KL-6 levels in differentiating SPDS from neuroendocrine cell hyperplasia (NEHI) in children presenting with interstitial lung disease. Levels are markedly elevated in children with SFTPC and ABCA3 but normal in those with NEHI.28

Recommendations for when to consider genetic testing

SFTPB and ABCA3 testing should be considered in term neonates (more than 36 weeks gestation) who present with unexplained respiratory distress, particularly when there is rapidly progressive disease or disease that fails to respond to conventional treatments (box 1). Routine testing is clearly inappropriate in premature infants who are most likely to have RDS of prematurity. The threshold for testing should be reduced in patients with a positive family history of an SPDS, history of unexplained respiratory symptoms in infancy and childhood, or parental consanguinity.3 ,5 Genetic testing for SFTPC and ABCA3 should also be considered in older children who present with chronic interstitial lung disease without a clear aetiology, particularly if there is family history of chronic lung disease or previous acute respiratory distress in infancy. Infants or children presenting with respiratory symptoms, together with neurological features and/or hypothyroidism, should be tested for NKX2.1.

Other cases where genetic testing should be considered include children with diffuse disease involving the entire lung on HRCT, histopathological findings that demonstrate alveolar proteinosis and findings of abnormal or absent lamellar bodies on electron microscopy on a lung biopsy.

Supportive care

In the acute setting, respiratory support may be required due to increased respiratory effort and hypoxaemia. This can range from oxygen supplementation, to non-invasive or invasive ventilation and extracorporeal membrane oxygenation. There is some evidence in infants with SFTPB that the use of high-frequency oscillatory ventilation with neuromuscular blockade improves oxygenation and chest radiographic appearances, and may help to stabilise infants for long enough to allow for lung transplantation.45

Somatic growth is also frequently impaired in children with SPDS. A dietician should review children and institute long-term nutritional supplementation as required.1

Pharmacotherapy

Most neonates with SPDS presenting acutely with respiratory distress would initially receive surfactant replacement therapy. While this may lead to a clinical response initially, it is not maintained in the longer term. Case series and small, uncontrolled studies show possible benefits associated with the use of hydroxychloroquine in the treatment of SFTPC31 and ABCA3.46 One literature review identified improvements in 9 of 21 patients with SFTPC and 2 of 3 with ABCA3.47 Azithromycin has also been used in patients with SFTPC29 and ABCA3.46

Using a Delphi process, the chILD EU has suggested that hydroxychloroquine can be started at a dose of 10 mg/kg (or 6.5 mg/kg in children <6 years of age).4 Clinical response to treatment should be assessed after 3–4 weeks in children who are ventilated or after 3 months in non-ventilated patients. Alternatively, azithromycin 10 mg/kg three times per week can be used as a second-line therapy with a response expected after 3 months of treatment.4 Due to the lack of either randomised or controlled studies, and the significant publication bias associated with small studies, it is difficult to predict exactly which cohort of patients will respond to treatment with hydroxychloroquine or azithromycin. Enrolment into a chILD EU-run international, randomised, placebo-controlled trial of hydroxychloroquine has begun and will provide much needed evidence.

Immunosuppression with corticosteroids is often started in children with SPDS.3 ChILD-EU suggests that intravenous methylprednisolone should be started at 10 mg/kg or 500 mg/m² (some centres use up to 30 mg/kg) daily on three consecutive days, assessing the clinical response to treatment after 7 days in children who are ventilated or at 28 days in non-ventilated children. Pulses may be repeated monthly for up to 6 months. Oral prednisolone at a dose of 1 mg/kg/day can be used between pulses of methylprednisolone in children who are ventilated or a dose of 2 mg/kg/day may be used instead of methylprednisolone in non-ventilated children.4 The ABCA3 gene promoter contains sterol-response elements, and in vitro studies have shown that the addition of glucocorticoids increase ABCA3 expression.48 ,49 While this rationale for use of steroids in ABCA3 is encouraging, response to corticosteroids is variable in these patients.27 Lack of robust trial data again makes it difficult to know the true clinical effectiveness of corticosteroid treatment in SPDS and mandates careful monitoring for side effects associated with long-term corticosteroid use.

Lung transplantation

Lung transplantation may be the only option for prolonging survival in children with severely affected and progressive disease that is unresponsive to other therapies and is the only successful treatment for children with severe SP-B disease.25 ,26 ,50 Data from US registries and small case series show clinical effectiveness and a 5-year survival of approximately 50%, which is comparable to the outcomes for age-matched children transplanted for other indications.25 ,26 ,50 Despite a lack of high-quality evidence, the poor outcomes associated with SPDS and, in particular, SFTPB and ABCA3 make lung transplantation an option to be considered on a case-by-case basis.