Diagnostic yield was favorable for TBLB in the diagnosis of LAM with varying degrees of lung involvement. The major finding of this work is a 57% positive rate of diagnosis and 5.6% TBLB-related pneumothorax. The positive rate was different between patients who had 1–2 biopsied samples and 3 or more.
Pathology is essential for LAM patients who cannot be definitely diagnosed based on the clinical presentations, VEGF-D, and radiological appearance. In 1991, Pedreira et al. first reported that TBLB was used in the diagnosis of LAM(8). In 1993, Bonetti et al recommended that HMB45 should be used to distinguish LAM from other smooth muscle proliferation; the positive rate of HMB-45 in LAM was 92%, and the histological diagnosis of LAM could be made when only a transbronchial biopsy is available(9). In 2012, Torre and Harari reported that TBLB was performed in seven patients and was diagnostic in six and did not result in complications(10). Meraj et al. found that TBLB was positive in 6 of 10 patients and had a yield of approximately 60% in LAM, with 52/217 (14%) complications, including pneumothorax (6%), bleeding (4%), chest pain (2%) and pneumonia (2%)(11). In recent years, some researchers reported that the positive rate of TBLB in patients with LAM was 70–78%, and there were no serious adverse events such as pneumothorax or uncontrollable bleeding due to TBLB(5, 6). Our results offered similar experience of the value of TBLB in LAM. The positive rate of TBLB in our results was similar to those previously reported(5, 6, 12). In addition, the complications of TBLB were also similar to previous reports. It was reported that the severity of LAM in CT scans will affect the positive rate of TBLB, but in our results, there was no difference in diagnostic yield in patients with different degrees of severity, although the trend existed(6). We first found that the number of biopsy specimens influenced the positive diagnosis rate. We recommended that at least 3 samples should be biopsied during TBLB.
Immunostaining with HMB-45 antibody is considered to be the most reliable method for the specific identification of LAM cells(13). In our study, the positive yield of HMB-45 in the diagnostic biopsies was 93%, which is similar to the previously reported rate of 92%(9). HMB-45 was not positive in all LAM cells, but it was a valuable marker in LAM(14). Gao et al reported that in LAM, the expression of progesterone receptor is frequently higher than that of estrogen receptor(14). However, our pathological reports showed that the positive yields of estrogen receptor (ER) and progesterone receptor (PR) were quite similar (84% and 78%, respectively). In patients with LAM, the severity of LAM in CT scans will affect the positive rate of TBLB(6). In our reports, mild LAM patients had a lower positive rate than moderate/severe LAM patients. However, there was no significant difference between mild, moderate, and severe LAM patients. This may result in a limited number of patients, especially few mild and moderate LAM patients. We found that with the increasing number of specimens, the positive rate will improve. Increasing the number of specimens did not increase the risk of complications in our reports. We recommended that the minimum number of samples biopsied should be three or more unless complications occur during the procedure.
Four patients (5.6%) reported pneumothorax, and no one reported major hemorrhage. Previous reports of TBLB in LAM found 0%-6% pneumothorax and 0%-4% hemorrhage(5, 6, 12).
De Fenoyl et al. reported that pneumothorax occurred in 3.4% of patients with interstitial lung disease who underwent TBLB(15). The complication of pneumothorax in LAM may be slightly higher than in patients with interstitial lung disease. Evaluation before, during and after the procedure is critical in the early finding of pneumothorax. Large bulli should be avoided during the procedure.
TBLB is effective and safe for patients with LAM; however, some patients cannot be diagnosed with TBLB. Recently, transbronchial cryobiopsy (cryo-TBB) has been introduced as a technique to obtain biopsy samples of lung parenchyma that exceed the size and quality of forceps biopsy samples(16–18). Cryo-TBB yields larger biopsy specimens (43 to 64 mm2) than transbronchial biopsy (5.8 mm2, range 0.58 to 20.88 mm2)(16, 19, 20). Cryo-TBB in patients with central, peribronchial or diffuse interstitial processes is now used clinically. The major complications of cryo-TBB included pneumothorax (12 to 28%), moderate bleeding (0 to 39%), and rarely death (0.3%)(17, 18, 20). However, there is no evidence of LAM in using cryo-TBB for diagnosis. Pneumothorax risk is a major concern of cryo-TBB in LAM patients.
There are some limitations to this study. First, it involved a small group of mild and moderate patients; therefore, we do not clearly know whether the severity of CT grades will influence the positive rate of TBLB. Second, how to evaluate the risk of pneumothorax or hemorrhage before the TBLB procedure has not been investigated.