Tracheobronchomalacia in infants: the use of non-breath held 3D CT bronchoscopy

To evaluate the use of a non-breath held 3D-CT-bronchoscopy in detecting tracheobronchomalacia in infants.

The study was based on 17 infants who received both bronchoscopy and 3D-CT-bronchoscopy within 1 week at our institution. 3D-CT-bronchoscopy consisted of an axial-enhanced-chest-scan and a 3D-volume-external rendering (VR) image of the airways and was scanned with a consciously sedated non-breath held protocol, using a 64-channel-multidetecter-CT scanner. VR images were classified by two radiologists as normal, luminal narrowing or complete obliteration. All patients were confirmed with bronchoscopy and the sensitivity, specificity, negative predictive value (NPV), positive predictive value (PPV) was calculated.

Ten (M:F¼7:3, mean 1 month) out of 17 infants were confirmed of tracheobronchomalacia. The sensitivity was <75% in detecting laryngomalacia, tracheomalacia and bronchomalacia. However, the specificity and NPV were higher than 75% for laryngomalacia, tracheomalacia and bronchomalacia and the PPV was 100% in laryngomalacia and tracheomalacia. False-positive cases included secretion plugging, extrinsic compression and foreign body, which were distinguishable in axial scans. The effective radiation dose was 0.73_0.49mSv.

A volume rendering image of the airways can be used to evaluate tracheobronchomalacia and stenosis in infants. Although, non-breath held MDCT is not recommended for screening airway lesions, narrowing of the larynx, and trachea and patency of the bronchus are reliable findings.