Exploring the Influence of Dysphagia and Tracheostomy on Pneumonia in Patients with Stroke: A Retrospective Cohort Study

Background: Pneumonia is common in patients with tracheostomy and dysphagia. However, the influence of dysphagia and tracheostomy on pneumonia in patients with stroke remains unclear. The aim of this study was to explore the risk factors related to pneumonia, and the association between dysphagia, tracheostomy and pneumonia in patients with stroke was investigated. Methods: Patients with stroke who experienced tracheostomy and dysphagia were included and divided into two groups based on record of pneumonia at discharge. Clinical manifestations and physical examination were used to diagnose pneumonia, whereas clinical swallowing examination, and videofluoroscopy swallowing studies (VFSS) were used to evaluate swallowing function. Results: There were significant differences between the pneumonia group and the no pneumonia group in total tracheostomy time (6.3 ± 5.9 vs. 4.3 ± 1.7 months, p = 0.003), number of instances of ventilator support (0.41 ± 0.49 vs. 0.18 ± 0.38, p = 0.007), PAS score (5.2 ± 1.92 vs. 4.3 ± 1.79, p = 0.039), impaired or absent cough reflex (76.4 vs. 55.6%, p = 0.035), oropharyngeal phase dysfunction (60.6 vs. 40.8%, p = 0.047), length of hospital stay (36.0 ± 7.2 vs. 30.5 ± 11.7 days, p = 0.025) and direct medical costs (15,702.21 ± 14,244.61 vs. 10,923.99 ± 7250.14 United States dollar [USD], p = 0.042). Multivariate logistic regression showed that the total tracheostomy time (95% confidence interval [CI], 1.966–12.922, p = 0.001), impaired or absent cough reflex (95% CI, 0.084–0.695, p = 0.008), and oropharyngeal phase dysfunction (95% CI, 1.087–8.148, p = 0.034) were risk factors for pneumonia. Spearman’s correlation analysis demonstrated that PAS scores were significantly correlated with cough reflex dysfunction (r = 0.277, p = 0.03), oropharyngeal phase dysfunction (r = 0.318, p < 0.01) and total tracheostomy time (r = 0.178, p = 0.045). The oropharyngeal phase dysfunction was significantly correlated with cough reflex (r = 0.549, p < 0.001) and UES opening (r = 0.643, p < 0.01). Conclusions: Tracheostomy and dysphagia increased the risk of pneumonia in patients with stroke. Total tracheostomy time, duration of ventilator support, degree of penetration and aspiration, and oropharyngeal phase dysfunction are risk factors. Given this, we also found that there may be a correlation between tracheostomy and dysphagia.