To drain or not to drain? Predictors of tube thoracostomy insertion and outcomes associated with drainage of traumatic hemothoraces

Bryan J Wells; Derek J Roberts; Sean Grondin; Pradeep H Navsaria; Andrew W Kirkpatrick; Michael B Dunham; Chad G Ball

doi:10.1016/j.injury.2015.04.032

To drain or not to drain? Predictors of tube thoracostomy insertion and outcomes associated with drainage of traumatic hemothoraces

Injury. 2015 Sep;46(9):1743-8. doi: 10.1016/j.injury.2015.04.032. Epub 2015 May 7.

Authors

Bryan J Wells¹, Derek J Roberts², Sean Grondin³, Pradeep H Navsaria⁴, Andrew W Kirkpatrick², Michael B Dunham¹, Chad G Ball⁵

Affiliations

¹ Department of Surgery, University of Calgary, Canada; Department of Critical Care Medicine, University of Calgary, Canada.
² Department of Surgery, University of Calgary, Canada.
³ Division of Thoracic Surgery, Department of Surgery, University of Calgary, Canada.
⁴ Department of Surgery, University of Cape Town, South Africa.
⁵ Department of Surgery, University of Calgary, Canada. Electronic address: ball.chad@gmail.com.

PMID: 25983221
DOI: 10.1016/j.injury.2015.04.032

Abstract

Introduction: Historical data suggests that many traumatic hemothoraces (HTX) can be managed expectantly without tube thoracostomy (TT) drainage. The purpose of this study was to identify predictors of TT, including whether the quantity of pleural blood predicted tube placement, and to evaluate outcomes associated with TT versus expected management (EM) of traumatic HTXs.

Patients and methods: A retrospective cohort study of all trauma patients with HTXs and an Injury Severity Score (ISS) ≥12 managed at a level I trauma centre between April 1, 2005 and December 31, 2012 was completed. Mixed-effects models with a subject-specific random intercept were used to identify independent risk factors for TT. Logistic and log-linear regression were used to compute odds ratios (ORs) for mortality and empyema and percent increases in length of hospital and intensive care unit stay between patients managed with TT versus EM, respectively.

Results: A total of 635 patients with 749 HTXs were included in the study. Overall, 491 (66%) HTXs were drained while 258 (34%) were managed expectantly. Independent predictors of TT placement included concomitant ipsilateral flail chest [OR 3.03; 95% confidence interval (CI) 1.04-8.80; p=0.04] or pneumothorax (OR 6.19; 95% CI 1.79-21.5; p<0.01) and the size of the HTX (OR per 10cc increase 1.12; 95% CI 1.04-1.21; p<0.01). Although the adjusted odds of mortality were not significantly different between groups (OR 3.99; 95% CI 0.87-18.30; p=0.08), TT was associated with a 47.14% (95% CI, 25.57-69.71%; p<0.01) adjusted increase in hospital length of stay. Empyemas (n=29) only occurred among TT patients.

Conclusions: Expectant management of traumatic HTX was associated with a shorter length of hospital stay, no empyemas, and no increase in mortality. Although EM of smaller HTXs may be safe, these findings must be confirmed by a large multi-centre cohort study and randomized controlled trials before they are used to guide practice.

Keywords: Empyema; Expectant management; Haemothorax; Trauma; Tube thoracostomy.

Publication types

Observational Study

MeSH terms

Adult
Chest Tubes
Critical Care / statistics & numerical data
Drainage* / methods
Empyema / etiology
Empyema / therapy*
Female
Hemopneumothorax / etiology
Hemopneumothorax / pathology
Hemopneumothorax / therapy*
Humans
Length of Stay / statistics & numerical data
Male
Middle Aged
Practice Guidelines as Topic
Retrospective Studies
Risk Factors
Thoracic Injuries
Thoracostomy / methods*
Treatment Outcome
Wounds, Nonpenetrating / complications*
Wounds, Nonpenetrating / pathology
Wounds, Nonpenetrating / therapy