Thoracic ultrasound versus artificial pneumothorax in complications of medical thoracoscopy—a propensity score matching analysis

Junjun Huang, Yan Hu, Xiangdong Mu, Jiping Liao, Xi Wang, Hong Zhang, Guangfa Wang


Background: Evaluation and location of the approaches is the key step of medical thoracoscopy. The previous standard for the step in many countries is artificial pneumothorax (AP). Recently, thoracic ultrasound (TU) has been considered as one of the choices for the development of technology. While there was a lack of data in investigating the complications of medical thoracoscopy locating approach with AP comparing TU.
Methods: A total of 108 patients who underwent medical thoracoscopy were retrospectively observed in Peking University First Hospital from January 2011 to April 2017, including 92 patients of the AP group and 16 patients of the TU group. Propensity score matching (PSM) was used to balance the covariance between the two groups. And the complications of the procedures between the groups were compared.
Results: Before PSM, there was one unbalanced covariates in the two groups. The overall complication rate was 7.6% (7/92) in the AP group, comparing 6.2% (1/16) in the TU group. There was no significant difference between the two groups (P=0.848). After balancing the covariate with 2:1 matched, the overall complication rate was 9.4% (3/32) in the AP group, comparing 6.2% (1/16) in the TU group. There was still no significant difference between the two groups (P=0.712).
Conclusions: TU is a locating approach method which does not increase the complication rate comparing with AP. And it could be a good choice in medical thoracoscopy.