Assessment of Bullae with High-Resolution CT in Patients with Spontaneous Pneumothorax: Comparison with Video-Assisted Thoracoscopy

Kim, Kyoung Rae; Oh, Yu Whan; Noh, Hyung Jun; Cho, Kyu Ran; Lee, Ki Yeol; Kang, Eun Young; Kim, Jung Hyuk

Abstract

PURPOSE: The purpose of this study was to compare the findings on high-resolution CT (HRCT) of the chest with those on video-assisted thoracoscopy for the detection of bullae in patients who had undergone an operation for spontaneous pneumothorax, and we also wished to evaluate the relationship between the characteristics of bullae on HRCT and development of spontaneous pneumothorax.
MATERIALS AND METHODS
Fifty patients with spontaneous pneumothorax who had undergone both HRCT of the chest and video-assisted thoracoscopic surgery were included in the study. Spontaneous pneumothoraces were classified as either primary or secondary pneumothorax, and as initial or recurrent pneumothorax. The HRCT scans were obtained with 1 mm slice thickness and a 5 mm scan interval. Two radiologists retrospectively compared the HRCT findings of the chest with those findings on video-assisted thoracoscopy for the detection of bullae, and they evaluated the value of HRCT for diagnosing bullae. In addition, we assessed the size and number of bullae in these patients, and we also evaluated the relationship between those findings of bullae and the development of spontaneous pneumothorax.
RESULTS
Bullae were detected in 40 patients by using video-assisted thoracoscopy, and HRCT showed bullae in 38 of these patients. Bullae were not identified with video-assisted thoracoscopy in the remaining ten patients, and among these ten patients, bullae were not demonstrated by HRCT in eight of them. Therefore, the sensitivity and specificity of HRCT for the detection of bullae were 95% (38/40) and 80% (8/10), respectively. The average size of the bullae of the affected hemithorax and the contralateral un-affected hemithorax was 1.97 cm+/-2.30 and 1.24 cm+/-1.46, respectively. Pneumothorax was more frequently observed in the hemithorax with larger bullae (p<0.05). The average numerical grade of the bullae (3.38+/-1.60) was higher in the affected hemithorax than in the contralateral un-affected hemithorax (2.96+/-1.86), but there was no statistically significant difference between both groups of hemithoraces (p>0.05). The average size of bullae in patients with secondary pneumothorax and those bullae of patients with primary pneumothorax was 4.44 cm+/-4.06 and 1.42 cm+/-1.26, respectively. The bullae were significantly larger in the patients with secondary pneumothorax than in those patients with primary pneumothorax (p<0.05). Although the average numerical grade of bullae was higher in the patients with secondary pneumothorax (4.00+/-1.58) than in those patients with primary pneumothorax (3.24+/-1.61), the difference between two groups was not statistically significant (p>0.05).
CONCLUSION
HRCT of the chest would be a useful modality for detecting the bullae of those patients with spontaneous pneumothorax. The development of spontaneous pneumothorax is associated with the size of bullae rather than the number of bullae.