Pulmonary cysts and pneumothorax

Histology

BHD patients often develop pulmonary cysts and have an increased risk for pneumothorax Predina et al., 2011; Zbar et al. 2002). Lung anatomy and histology generally appears normal in individuals with BHD, and despite the presence of multiple pulmonary cysts, lung function is usually unaffected (Toro et al., 2007; Tobino et al., 2012). BHD lung cysts are distinct from those cysts that develop in LAM. BHD patients generally have fewer lung cysts, which are located in the lower lobes of the lungs and can affect male or female FLCN mutation carriers of any age; high resolution CT scans can be used to differentiate between these two and other cystic lung diseases (Ferreira Francisco et al., 2015, Gupta et al., 2015, Ha et al., 2015)

In 12 BHD patients analysed by Tobino et al. (2011), the number of lung cysts were found to vary from roughly 30-400, and the size varied from a few millimetres to more than 2 cm. The cysts were generally irregularly-shaped and most commonly found in the lower medial zone of the lungs (Tobino et al., 2012). Another study confirmed the presence of multiple lung cysts, mainly in the lower lungs, that varied in size and shape (Agarwal et al., 2011). Koga et al. (2009) have hypothesised that these cysts represent an aberrant cystic alveolar formation. Subsequently, Furuya et al. (2012) used histopathological analysis to show that the BHD-associated lung cysts were lined with differentiated pneumocytes and had alveolus-like structures within them, indicating that their histopathology is distinct from non-specific blebs and bullae. The clinical and pathological features of the lung symptoms seen in BHD syndrome were recently reviewed by Furuya and Nakatani (2012).

A study comparing the lung pathology seen in BHD patients with that seen in spontaneous pneumothorax cases caused by smoking, found distinct histological and radiological differences between the two groups (Fabre et al., 2013). BHD patients generally had numerous (more than 20) punch-out type cysts lined with CK7 and TTF1 expressing pneumocytes, showing no signs of inflammation and basal predominance while the control cohort had fewer cysts showing apical predominance. The control cohort also had a greater prevalence of smoking-related changes such as fibroelastotic scars, respiratory bronchiolitis and emphysema, which BHD patients generally did not.

However, one BHD patient, who was a current smoker, had pulmonary changes more similar to the smoking cohort. This suggests that although smoking does not cause the pulmonary changes seen in BHD, it is likely to exacerbate lung problems in BHD patients increasing the risk of pulmonary tumours and should be strongly discouraged. However, a diagnosis of BHD wasn’t ruled out and TTF1 expression was not clearly reported in the control cohort used in this study, suggesting that further studies are required before clinicians modify their diagnostic approach in these cases (Johannesma et al., 2014c).

Several cases of pulmonary tumourigenesis in BHD patients have been reported (Gunji et al., 2007Furuya & Nakatani 2013, Nishida et al., 2015) and appear to correlate with a history of smoking. No FLCN second hits have been reported in these tumours and further research is required to determine if there is a link between smoking and pulmonary tumours in BHD patients.

Kumasaka et al., (2014) analysed 229 lung cysts in resections from 50 BHD patients and compared them with 117 lung cysts from 34 PSP patients, comparing samples for number, size, location and presence of inflammation. They found that BHD lung cysts were found in both subpleural and intrapulmonary areas; often found at interlobular septa and 40% had venules; and only showed signs of inflammation if located in the subpleura. The authors suggest that loss of FLCN causes alveoli walls to become weak and vulnerable to disruption by mechanical stress during breathing, which causes cysts to form. Furthermore, based on the observation that inflamed cysts tended to be larger and located in the sub-pleura, they suggest that pneumothorax inflames existing cysts, causing them to grow and subsume neighbouring cysts.

Several recent reviews discuss the differential diagnosis of BHD, and other cystic lung diseases, based on imaging (Ha et al., 2015, Gupta et al., 2015, Richards et al., 2015).

Prevalence

The presence of pulmonary cysts in BHD syndrome was first described by Toro et al. (1999). Three of thirteen BHD patients examined had pulmonary cysts, and one of these three patients also developed pneumothorax (Toro et al., 1999).

Pulmonary cysts are the most common manifestation of BHD, seen in up to 90 % of patients (Predina et al., 2011), suggesting that FLCN has a significant role in normal lung physiology. Pneumothorax is strongly correlated with number of lung cysts, indicating that the presence of lung cysts may cause pneumothoraces (Johannesma et al., 2014f). It is thought that pulmonary cysts increase the risk of pneumothorax by rupturing and releasing air into the chest cavity (Furuya and Nakatani, 2013Johannesma et al., 2014e).

Using a cohort of 111 BHD haplotype carriers and 112 family members without the BHD haplotype as controls, Zbar et al. (2002) identified a 50-fold increase in the risk of pneumothorax for BHD-affected individuals (OR = 50.3, adjusted for age). Toro et al. (2007) found that following a single episode of spontaneous pneumothorax, recurrent events were more common.

A study by Houweling et al. (2011) found that 28 of 115 (24%) FLCN mutation carriers had a medical history of pneumothoraces, and by analysing 21 BHD families estimated the life-time risk of pneumothorax among BHD patients to be 29%, with a mean age for the first event of 36 years. It is believed that lung pathology may be the earliest symptom of BHD syndrome, as pneumothoraces have been reported in BHD patients as young as seven, fourteen and sixteen years of age (Bessis et al., 2006; Gunji et al., 2007; Johannesma et al., 2014d). However, there is one case of a BHD presenting with her first pneumothorax at the age of 73, suggesting that the age of onset can be quite variable (Kunogi Okura et al., 2013).

Cohort studies of primary spontaneous pneumothorax (PSP) patients have found that up to 10% of these patients carry FLCN mutations, often without any skin manifestations or renal tumours (Ren et al., 2008Johannesma et al., 2015). Suspicion of BHD should be higher in families with a history of PSP (Painter et al. 2005, Gunji et al. 2007, Hayashi et al. 2010, Ding et al., 2015) especially if there is also a history of renal cancer of dermatological growths. Identification of BHD patients based on a PSP presentation enables the patients and families to access regular monitoring which can result in earlier tumour treatment.

A recent study found that 12 in 190 BHD patients (6.3%) suffered 13 episodes of pneumothorax within one month of taking a commercial flight, suggesting that air travel may cause pneumothorax in a small proportion of patients (Postmus et al., 2014).

There is currently no evidence of a genotype-phenotype correlation regarding the lung manifestations of BHD (Kunogi et al., 2010).