Surgical Management of Tracheostomy in Patients with Severe Burns and Cervical Involvement: Impact on Cervical Wound-Healing Disturbance

TL;DR

This study examines how tracheostomy timing and wound coverage methods affect healing in patients with severe neck burns.

Contribution

The study identifies that burn depth, not tracheostomy timing or grafting method, is the key factor in wound complications.

Findings

Advanced dressings had lower cervical wound-healing disturbance rates compared to skin grafts.

Tracheostomy timing and grafting techniques did not significantly affect complication rates.

Cervical burn depth was the only independent predictor of wound-healing disturbances.

Abstract

Background: Major cervical burns often require tracheostomy (TT); however, it remains unclear whether TT timing relative to cervical wound coverage (WC) and the choice of coverage modality affect peristomal wound healing. Methods: In this retrospective single-center cohort study, we included 48 adults with thermally induced cervical burns who underwent TT between 2015 and 2024 in a specialized burn ICU. Eleven patients died before decannulation and were excluded from the primary endpoint analysis. Cervical wound-healing disturbance (CWHD) as a primary endpoint was assessed in the remaining 37 patients, including 13 treated with advanced cervical dressings and 24 treated with split-thickness skin grafts (STSG). Results: CWHD occurred in 2/13 (15.4%) with advanced dressings and 11/24 (45.8%) after STSG. Within the grafted subgroup, complication rates did not differ significantly according to TT sequencing relative to WC, TT–WC interval, grafting technique (mesh vs. Meek), or tracheostoma closure method (all p > 0.05). In multivariable logistic regression, only cervical burn depth independently predicted CWHD. Conclusion: In patients with cervical burns requiring TT, burn depth/severity, rather than TT timing, grafting strategy, or closure technique, appears to drive peristomal wound complications, supporting individualized planning instead of rigid algorithms.

Figures