Background CTGB is widely used to sample lung masses. One of the important complications is pneumothorax which could potentially lead to a prolonged hospital stay. The aim of the study was to examine the prevalence and potential predictors associated with pneumothorax following GTGB.

Methods We retrospectively reviewed CTGB data over a two year period (August 2014 to July 2016) for all patients who underwent CTGB. The data collected included age, sex, co-morbidities, smoking history, spirometry, performance status, presence of emphysema, thickness, depth of the needle, size of the lesion and lobe of the lesion.

Results 227 patients underwent CTGB with an overall diagnostic yield of 93.8% (213/227). The incidence of pneumothorax was 61/227 (26.9%). Of the patients with pneumothorax, 8/61 (13.1%) needed chest drain insertion with a median hospital stay of 4.5±2.1 days. There was no difference in diagnostic yield between both pneumothorax and the no pneumothorax group. Overall 89.2% (190/213) of the positive biopsies were malignant while 10.8% (23/213) were benign. There was no difference in the performance status, severity of airflow obstruction or lobe of the lesion between groups. Binary logistic regression analysis showed the size of the lesion as a determinant of developing pneumothorax (p=0.022). The risk of developing a pneumothorax was 27.1% for a lesion ≤10 mm and 18.7% for a lesion ≤20 mm.

Conclusion The incidence of pneumothorax following CTGB was 26.9% but only 3.5% of all patients undergoing CTGB had a chest drain inserted for their pneumothorax. While CTGB is a safe procedure with a good diagnostic yield one needs to be watchful of the risk of pneumothorax. The size of the lesion correlated with the development of pneumothorax with a smaller size associated with a higher risk.