OBJECTIVE Glossopharyngeal insufflation (GI) is a technique practised by competitive breath-hold divers to enhance their performance. Using the oropharyngeal musculature, air is pumped into the lungs to increase the lung volume above physiological TLC. Experienced breath-hold divers can increase their lung volumes by up to 3 L. Although the potential for lung injury is evident, there is limited information available. The aim of this study was to examine whether there is any evidence of lung injury following GI, independent of diving. METHODS Six male, competitive breath-hold divers were studied. CT of the thorax was performed during breath-holding at supramaximal lung volumes following GI (CT(GI)), and subsequently at baseline TLC (CT(TLC)). CT scans were performed a minimum of 3 days apart. Images were analysed for evidence of pneumomediastinum or pneumothorax by investigators who were blinded to the procedure. RESULTS None of the subjects showed symptoms or signs of pneumomediastinum. However, in five of six subjects a pneumomediastinum was detected during the CT(GI). In three subjects a pneumomediastinum was detected on the CT(GI), but had resolved by the time of the CT(TLC). In two subjects a pneumomediastinum was seen on both the CT(GI) and the CT(TLC), and these were larger on the day that a maximal GI manoeuvre had been performed. The single subject, in whom a pneumomediastinum was not detected, was demonstrated separately to not be proficient at GI. CONCLUSIONS Barotrauma was observed in breath-hold divers who increased their lung volumes by GI. The long-term effects of this barotrauma are uncertain and longitudinal studies are required to assess cumulative lung damage.