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



The study was a retrospective chart review for a case series of a single family. Genetic testing, pulmonary function tests, and imaging were performed as part of the patients' routine clinical assessment. Our study did not proceed to formal ethics review given the retrospective and case series nature of the project.


The proband was diagnosed with BHD in the context of several unprovoked pneumothoraces and a strong family history of spontaneous pneumothoraces. A heterozygous mutation in the FLCN gene (c.59delT), resulting in premature protein truncation, was identified in this individual. Following the diagnosis of BHD syndrome, cascade genetic testing was offered to family members with possible clinical features of BHD and on a predictive basis for asymptomatic at-risk individuals, with 68 people agreeing to formal genetic testing. Inclusion criteria for the case series included: (1) age 18 and greater (2) direct relation to the proband (not through marriage) and (3) documentation of the FLCN gene mutation.

A family history was compiled using information provided by the proband and multiple relatives. The pedigree included over 150 individuals from six generations originating from a matriarch and patriarch born in the late nineteenth century (Fig. 1). The proband was one of 13 siblings.

Figure 1.

Pedigree of BHD affected family. The proband is indicated by the arrow on the far left. BHD affected individuals are highlighted in black

Genetic Assessment and Confirmation of BHD Syndrome

Genetic testing was performed using the same standardized assay and equipment in the Alberta Children's Hospital Molecular Diagnostics Laboratory. Initial sequencing on the proband was performed following Polymerase Chain Reaction (PCR) amplification using primers flanking each exon allowing for analysis of exonic plus flanking intronic sequence. Subtraction analysis using Mutation Surveyor (SoftGenetics) software was used to identify differences between the patient and a reference sequence. The Molecular Diagnostic Lab then performed targeted sequence analysis to detect this mutation in consenting family members. Genomic DNA isolated from whole blood was PCR amplified for exon 4 of the FLCN gene and sequenced using the Sanger method to identify the presence of the FLCN c.59delT mutation.

Clinical Assessment

All individuals with the FLCN mutation were offered a clinical assessment by a respirologist including history, physical exam, CT scan of the chest, CT or magnetic resonance imaging (MRI) scan of the abdomen, and complete pulmonary function tests. Lung function testing, including spirometry and diffusing capacity for carbon monoxide (DLCO), was performed in accordance with ATS standards and using the well-validated Knudson 1983 normset.[13–15] Lung function tests performed at the time of the individual's BHD diagnosis were used for data analysis. Clinical data was not collected from family members who did not have the gene mutation or undergo genetic testing.

Chest Imaging

Non-contrast, chest CT with 5 mm slices were performed. Images were reviewed by a thoracic radiologist who was blinded to the clinical history. CT scans were scored according to the following criteria: (1) cyst size (based on greatest diameter on axial sections), (2) cyst number (1, 2 to 5, or numerous), (3) extent of upper lung zone involvement (0–5 %, 5–25 %, 25–50 %, 50–75 %, or greater than 75 % of area affected by cysts), (4) extent of lower lung zone involvement (0–5 %, 5–25 %, 25–50 %, 50–75 %, or greater than 75 % of area affected by cysts), and (5) cyst distribution (upper lung zone predominant, lower lung zone predominant, right side predominant, left side predominant, or equal through-out). The lungs were divided into upper and lower zones at the level of the right main pulmonary artery. There is no existing rating system for cystic lung disease. Therefore, the divisions for extent of upper and lower lung involvement as well as cyst number were arbitrary and made a priori.

Statistical Analysis

Descriptive analysis was applied to the patient characteristics. Simple logistic regression was performed to identify variables associated with pneumothorax. A full model was constructed from variables identified as associated with pneumothorax on univariate analysis (p <0.05). Backwards selection was performed by successively dropping the least significant variables until all remaining variables were statistically significant (Stata STAT/IC 13 software).