What Are the Best Criteria for Birt–Hogg–Dubé Syndrome‐related Pneumothorax?

H. Rosie Xing,Deruo Liu,Chaoyang Liang
DOI: https://doi.org/10.1111/resp.13717
IF: 6.9
2019-01-01
Respirology
Abstract:We read with great interest the study by Ebana et al.1 in the April 2018 issue of Respirology. A scoring system was established to separate patients with Birt–Hogg–Dubé (BHD) syndrome from patients with primary spontaneous pneumothorax. We agree that this system could be quite useful in clinical practice, but we have a further suggestion. In this study, only patients who were suspected of BHD syndrome underwent sequencing of the FLCN gene. Potential BHD syndrome was judged by the familial history of pneumothorax and/or characteristic appearance of multiple lung cysts on chest computed tomography (CT). So, pneumothorax patients without family history or typical CT finding were grouped as non-BHD related directly. However, according to the study by Schmidt et al.,2 in a group of 53 previously diagnosed BHD families, 25 families had members who developed spontaneous pneumothorax. Among the 25 families, 13 had one member who developed pneumothorax while 12 had more than one, which means that over half of the BHD-related pneumothorax presented as sporadic pneumothorax (without a clear family history of pneumothorax). Furthermore, in this cohort, not all BHD patients had lung cysts on chest CT scan, even when pneumothorax had occurred (see family 246 in table 1 from Reference 2). The ‘typical’ chest CT finding of BHD patients is not a widely accepted or standard definition. It has been reported that the lung cysts and pneumothorax in BHD patients are age-dependent manifestations and lung cysts are not necessarily related to pneumothorax.3 Hence, we believe that the inclusion criteria used by the authors could not select BHD-related pneumothorax patients effectively from their study cohort. From the perspective of medical statistics, in order to set up a new diagnostic tool, there should be an already existing gold standard, which could be used to obtain a definitive diagnosis for a particular disease, such as pathology result for cancer or culture result for infection. In this case, in order to set up a prediction model, the gold standard should be used to definitely clarify a pneumothorax patient to be BHD-related or not, then the new prediction model could be compared and the sensitivity and specificity could be calculated. For genetic disorders, DNA sequencing is the gold standard for diagnosis. So, we believe that the authors should perform FLCN sequencing for every pneumothorax patient in their study cohort, regardless of the family history or the CT findings. In conclusion, we propose that DNA sequencing should be used in this study as gold standard in order to establish a reliable prediction model.
What problem does this paper attempt to address?