Highlights from #AAST2015 - tracheal injuries

Do tracheal injuries require tracheostomy?  Not necessarily according to the group 

AIRWAY MANAGEMENT FOLLOWING REPAIR OF CERVICAL TRACHEAL INJURIES: A RETROSPECTIVE, MULTICENTER STUDY

John A. Harvin MD, Bryan A. Cotton* MD,MPH, Jason Brocker MD, Deborah M. Stein* MD,MPH, Evren Dilektasli MD, Kenji Inaba* MD, Michael A. Vella MD, Oscar Guillamondegui* MD, Lisa M. Kodadek MD, Elliot R. Haut* MD,Ph.D., Cory R. Evans MD, Jordan A. Weinberg* MD, Michael D. Goodman MD, Bryce R. Robinson* MD, John B. Holcomb* MD, University Of Texas Health Science Center At Houston 

Introduction: Optimal airway management following repair of cervical tracheal injuries is unknown. “Protective” tracheostomy had been commonly employed, but recent studies question the practice. This study aims to describe the current airway strategies being used and determine the optimal airway management following cervical tracheal injury repair. Methods: Patients with cervical tracheal injuries admitted 01/2000-01/2014 at seven U.S. Level I trauma centers were identified. Patients were placed into one of three groups depending on the post-operative airway management: immediate or early extubation (≤24 hours, EXT), prolonged intubation (>24 hours, INT), and immediate tracheostomy (TRACH). Following univariate analysis, a multivariate model was developed to evaluate for surgical site infection (SSI) and ICU-free and ventilator-free days, comparing INT
and TRACH to EXT as the reference. Continuous variables presented as median (IQR). Results: Over the study period, 382,529 patients were admitted to seven Level I trauma centers. 594 (0.16%) had a laryngotracheal injury with 120 (0.03%) cervical tracheal injuries. Ten patients were excluded for incomplete data and seven died within 24 hours of admission, leaving 103 patients included in the study. Patients were grouped based on airway management: 40 (39%) in EXT, 30 (29%) in INT, and 33 (32%) in TRACH. There were no differences in demographics or injury mechanism. The INT and TRACH groups were more severely injured than the EXT group (ISS INT 25 [16, 29] and TRACH 17 [12, 33] vs EXT 16 [10, 17], p<0.01). The INT and TRACH groups had a trend towards higher rates of destructive injuries (INT 20% vs TRACH 34% vs EXT 13%, p=0.08). Despite a higher SSI rate, the TRACH group had a lower mortality and more hospital-, ICU-, and ventilator-free days compared to the INT cohort. On multivariate analysis, tracheostomy was associated with an increased risk in the odds of SSI (OR 9.56, 95% CI 1.35-67.95) compared to both EXT and INT, while INT was associated with fewer ICU-free days (corr. coef. -9.64, 95% CI -12.66 to -6.62) and ventilator-free days (corr. coef. -9.24, 95% CI -12.30 to -6.18) compared to both EXT and TRACH.

	EXT	INT	TRACH	p
Surgical Site Inection	2 (5%)	4 (13%)	7 (21%)	0.11
Pneumonia	0 (0%)	7 (23%)	3 (9%)	<0.01
Hospital-free days	27 (24, 28)	12 (5, 22)	16 (10, 22)	<0.01
ICU-free days	29 (28, 30)	22 (8, 25)	26 (21, 29)	<0.01
Ventilator-free days	30 (29, 30)	25 (12, 27)	28 (28, 30)	<0.01
In-hospital mortality	0 (0%)	4 (13%)	0 (0%)	<0.01

Conclusion: In patients with a cervical tracheal injury, immediate or early extubation was common and safe. However, among those with more severe injuries, immediate tracheostomy versus prolonged intubation presents a risk-benefit decision. While immediate tracheostomy placement is associated with increased risk of SSI, prolonged intubation is associated with higher risk of pneumonia and mortality and fewer ICU-free and ventilator-free days.