Timing of Tracheostomy in Patients with Intracerebral Haemorrhage: A Propensity-Matched Analysis

Although early tracheostomy (ET) is recommended for patients with severe stroke, the optimal timing of tracheostomy for patients with intracerebral haemorrhage (ICH) remains controversial. This study aimed to explore the clinical characteristics, risk factors and timing of tracheostomy in patients after tracheal intubation using a propensity-matched analysis.

We conducted a retrospective database search and assessed 267 consecutive patients who underwent endotracheal intubation (175 of whom underwent tracheostomy) and ICH between July 2017 and June 2021. A logistic regression model was applied to identify the critical factors influencing the decision for tracheostomy by comparing factors in a tracheostomy group and a nontracheostomy group. Patients were divided into an early (≤5 days) or a late (>5 days) group according to the median time of tracheostomy. Propensity score matching was performed to adjust for possible confounders and investigate differences in outcomes between ET and late tracheostomy (LT).

Among the 267 enrolled patients with ICH and endotracheal intubation, 65.5% received tracheostomy during hospitalisation, and 52.6% received ET. The independent risk factors for tracheostomy included National Institute of Health Stroke Scale (NIHSS) (odds ratio [OR]: 1.179; 95% confidence interval [CI]: 1.028-1.351; P = 0.018), aspiration (OR: 2.171; 95% CI: 1.054-4.471; P = 0.035) and infiltrates (OR: 2.149; 95% CI: 1.088-4.242; P = 0.028). Using propensity matching, we found that ET was associated with fewer antibiotic-using days (15 vs. 18; P < 0.001) and sedativeusing days (6 vs. 8; P < 0.001), shorter intensive care unit (ICU) Length of Study (LOS) (9 vs. 12; P < 0.05) and reduced in-ICU costs (3.59 vs. 7.4; P < 0.001) and total hospital costs (8.26 vs. 11.28, respectively; P < 0.001). Muscle relaxants (31.8% vs. 60.6%) were used less frequently in patients with ET (P = 0.001). However, there were no differences between the ET and LT groups in terms of modified Rankin Scale (mRS) (4 vs. 4; P = 0.932), in-general-ward costs (4.74 vs. 4.37; P = 0.052), mechanical ventilation days (6 vs. 6; P = 0.961) and hospital LOS (23 vs. 23; P = 0.735) as well as the incidences of ventilator-associated pneumonia (28.8% vs. 37.9%; P = 0.268), tracheostomyrelated complications (16.7% vs. 19.7%; P = 0.652), respiratory failure (24.2% vs. 31.8%; P = 0.333), all-cause deaths (15.2% vs. 16.7%; P = 0.812) and pneumonia (77.3% vs. 87.9%; P = 0.108).

We recommend ET for high-risk patients with ICH. Although ET cannot reduce inhospital mortality or improve patient prognosis, it may help reduce hospital costs and ICU LOS as well as the use of antibiotics, sedatives and muscle relaxants.