External Validation of a Tube Thoracostomy Complication Classification System.

BACKGROUND: Tube thoracostomy (TT) is a commonly performed procedure which is associated with significant complication rates. Currently, there is no validated taxonomy to classify and compare TT complications across different populations. This study aims to validate such TT complication taxonomy in a cohort of South African trauma patients. METHODS: Post hoc analysis of a prospectively collected trauma database from Pietermaritzburg Metropolitan Trauma Service (PMTS) in South Africa was performed for the period January 2010 to December 2013. Baseline demographics, mechanism of injury and complications were collected and categorized according to published classification protocols. All patients requiring bedside TT were included in the study. Patients who necessitated operatively placed or image-guided TT insertion were excluded. Summary and univariate analyses were performed. RESULTS: A total of 1010 patients underwent TT. The mean age was (±SD) of 26 ± 8 years. Unilateral TTs were inserted in n = 966 (96%) and bilateral in n = 44 (4%). Complications developed in 162 (16%) patients. Penetrating injury was associated with lower complication rate (11%) than blunt injury (26%), p = 0.0001. Higher complication rate was seen in TT placed by interns (17%) compared to TT placed by residents (7%), p = 0.0001. Complications were classified as: insertional (38%), positional (44%), removal (9%), infective/immunologic (9%), and instructional, educational or equipment related (0%). CONCLUSIONS: Despite being developed in the USA, this classification system is robust and was able to comprehensively assign and categorize all the complications of TT in this South African trauma cohort. A universal standardized definition and classification system permits equitable comparisons of complication rates. The use of this classification taxonomy may help develop strategies to improve TT placement techniques and reduce the complications associated with the procedure. LEVEL OF EVIDENCE: V. STUDY TYPE: Single Institution Retrospective review.