Birt-Hogg-Dubé Syndrome: A Large Single Family Cohort

Kate Skolnik; Willis H. Tsai; Kimberly Dornan; Renée Perrier; Paul W. Burrowes; Warren J. Davidson


Respiratory Research. 2016;17(22) 

In This Article


This is the largest single family BHD cohort reported in the literature to date. There are approximately 430 families with BHD syndrome documented in the literature. We found that all individuals had pulmonary cysts, pneumothoraces were common, and cyst size and lower lobe predominant disease were associated with pneumothorax. Lung function was generally preserved and not affected by a high cyst burden.

Previous studies included multiple families and were characterized by several genotypes.[1,8,9,12,16–24] In contrast, our cohort consisted of a single Caucasian family with a FLCN c.59delT mutation; thus providing a more homogenous phenotype.

In our cohort, every individual had pulmonary cysts on CT chest imaging. The high penetrance of lung cysts was in keeping with existing BHD literature (83 to 100 %).[9,12,16–21,23,24] The prevalence of pneumothorax was slightly higher than that observed in the four largest BHD studies, which ranged from 24 to 32 %.[1,8,9,12] However, pneumothorax prevalence as low as 8 %[22,24] and as high as 93 %[20] has been described. Cyst size and extent of lower lung zone disease were found to be associated with pneumothorax. This is supported by previous literature, although cyst volume, number, and right middle location have also been reported to be associated with pneumothorax risk.[9] In contrast to Zbar and colleagues' finding,[8] age was not associated with pneumothorax.

Our recurrent pneumothorax rate was 42 %, which falls between the previously reported rates of 25 and 75 %.[8,9] Most of the documented pneumothoraces required management with an interventional procedure, similar to prior studies.[9] The recurrence rates were substantially reduced post treatment, especially post surgery. Furthermore, it has been suggested that chemical pleurodesis or pleural ablation (which can be used to prevent recurrence of primary and secondary pneumothoraces) would be of even greater benefit in BHD given the often diffuse nature of the lung disease.[25] Others have extrapolated from lymphangioleiomyomatosis (LAM) data that chemical pleurodesis and surgical intervention have comparable outcomes in preventing pneumothorax recurrence.[26]

The optimal method and timing of treating BHD-associated pneumothorax is unclear and controversies exist as randomized prospective trials are lacking. Some reviews suggest that BHD-associated pneumothorax should be managed the same as in the general population;[27] others argue that with the lower likelihood of spontaneous resolution and the high likelihood of recurrence, pneumothoraces should be managed more aggressively in this population.[26] Given the available data[9] in combination with our findings we would suggest the latter; clinicians should have a lower threshold to treat BHD-associated pneumothorax with tube thoracostomy (where indicated) as well as consider early referral for a surgical opinion.

Our study had the largest collection of BHD subjects to date who underwent lung function testing. Interestingly, we found the majority had preserved DLCO, with a subset having supranormal values. This was not explained by test technique, equipment problems, body surface area, or asthma exacerbation. Although the reason for these outliers is unclear, this finding could possibly be explained by factors that were not systematically assessed (such as polycythemia, early heart failure, pulmonary hemorrhage, pretest exercise, or the occurrence of a Mueller maneuver during testing). In contrast, prior studies documented a low normal or mildly reduced DLCO in the context of BHD.[17,18,28] Lung volumes were preserved in our cohort, consistent with existing literature.[17,18,28] Our study suggests that individuals with BHD may have even milder impairments in lung function that previously thought.[17,18] A novel finding was the observation that burden of cystic lung disease did not affect lung function.

With respect to extrapulmonary manifestations, dermatological findings were noted in only 53 % of the cohort, which was lower than the 79–100 % prevalence of skin involvement documented in most BHD cohorts.[1,9,12,24] This could be related to the relatively young average age at diagnosis in this cohort, as BHD related skin findings are more common starting in the third to fourth decade of life.[27] Renal cysts were found in 26 % of those with imaging, which was higher than the 15 % observed in the only other study that specifically quantified renal cysts.[24] In contrast, the 3 % prevalence of renal tumors documented in our cohort was significantly lower than the 12–27 % seen in the four largest BHD studies.[1,8,9,12] This may be explained by sample size as some smaller cohorts also quoted 5 to 6 % prevalence of renal cancer.[20,22]

There are a few limitations to this study. First, this is a retrospective review. However, the clinical assessments were very detailed with pulmonary function testing and imaging performed in most subjects. Second, the sample size is smaller than previously reported pooled cohorts. However, this is the largest single-family cohort described in the literature with the same genotype, allowing analysis of a uniquely homogenous population. Furthermore, our study highlights the concept that understanding the clinical phenotype of an individual with BHD might predict the phenotype of other affected family members and aid in personalizing risk estimates and tailoring management.

Although rare, BHD can have significant clinical implications. There are no well-validated studies of BHD prevalence,[29] but some sources estimate 9 per 1 million.[30] Of note, up to 5 % of apparent primary spontaneous pneumothorax have been attributed to underlying BHD.[31] This highlights the importance of inquiring about family history, skin abnormalities, and renal disease in individuals with pneumothorax. It is also important to recognize that individuals with BHD cystic lung disease may be underdiagnosed since lung function is generally well preserved (in contrast to conditions such as LAM).