Ultrasound for confirmation of thoracostomy tube placement by emergency medicine residents

Abstract

Introduction Thoracostomy tubes are placed in the Emergency Department for numerous indications, including hemothoraces, pneumothoraces, and empyemas. After insertion, a portable single view chest radiograph is typically performed minutes later to confirm thoracostomy tubes position. However, up to 2.6% of thoracostomy tubes are ultimately determined to be misplaced. Failure to adequately drain the chest in a timely manner may have disastrous consequences. Ultrasonography by expert sonographers has been previously described to evaluate thoracostomy tubes position. The purpose of this study was to assess the accuracy of ultrasound for confirmation of thoracostomy tubes placement by Emergency Medicine residents. Methods We conducted a prospective, randomized, blinded study using a cadaveric model for ultrasound confirmation of thoracostomy tube placement by resident physicians. Thirty-five Emergency Medicine residents performed a total of 140 confirmations. The primary outcome of the study was the sensitivity and specificity of EM resident-performed ultrasonography to correctly confirm thoracostomy tube placement. Secondary outcomes included time to identification, operator confidence, and subgroup analysis by resident training level. Results The study demonstrated an overall sensitivity of 100% (95% CI 94–100%) and specificity of 96% (95% CI 87–99%) for intrathoracic placement. Post-graduate year (PGY) 1 EM residents demonstrated 100% (95% CI 76–100%) sensitivity and 100% (95% CI 76–100%) specificity. PGY 2 EM residents demonstrated 100% (95% CI 87–100%) sensitivity and 94% (95% CI 79–99%) specificity. PGY 4 EM residents demonstrated 100% (95% CI 80–100%) sensitivity and 95% (95% CI 75–100%) specificity. The total time to identification was 16 seconds (95% CI 13–19). Overall operator confidence was 4.0/5.0 (95% CI 3.8–4.1). Conclusion Emergency medicine residents were able to quickly identify thoracostomy tube location using ultrasound with a high degree of accuracy in a cadaveric model after a brief educational session.