Results
Forty-three patients with a history of severe-to-critical COVID-19 completed the evaluations. Patients consulted in our laryngology unit on average 51.6 ± 30.2 days after their hospitalization in intensive care units (UCI, range: 2-5 months). The mean age of cohort patients was 59.0 ± 11.3 yo. There were 10 females (23.3%) and 33 males (76.7%). Twenty-two patients were intubated less than 2 weeks (Group 1), while 21 were intubated more than 2 weeks (Group 2), respectively. The epidemiological and clinical features of patients are described in Table 1. Patient groups were comparable regarding gender proportion, mean age, tobacco history, comorbidities and symptoms. The most prevalent comorbidities included hypertension, diabetes and obesity (Table 1). The mean intubation duration was 9.9 ± 3.7 days in group 1, while patients of group 2 were intubated during 26.3 ± 6.8 days. In our medical center, the position of patients in the ICU bed were changed daily (stomach/back). There were 3 and 5 tracheostomies in groups 1 and 2, respectively. The mean duration of tracheostomy decannulation was 34.9 ± 22.0 days. One patient was not decannulated at the time of the consultation.
Dysphonia (100%), dyspnea (44.1%) dysphagia (20.9 %), and neck pain (9.3%) were the most prevalent symptoms (Table 1). The videolaryngostroboscopy examination reported posterior commissure hypertrophy, posterior glottis stenosis, laryngeal diffuse edema and granuloma as the most prevalent laryngeal abnormalities in patients (Table 2). Figure 1 reported some laryngeal findings of patients. Two laryngeal examinations were considered as normal by the judges. The proportions of posterior commissure hypertrophy and laryngeal edema were significantly higher in group 1 compared with group 2 (p<0.001), while posterior glottic stenosis was more prevalent in group 2 compared with group 1 (p<0.001). The posterior commissure hypertrophy occurred concurrently to another abnormality in 16 cases, e.g. diffuse laryngeal edema (N=8), granuloma (N=5), bilateral vocal fold insufficiency (N=2), and posterior glottic stenosis (N=2). According to the classification of Bogdasarian,12 the posterior glottic stenoses of group 1 were type 3 and 4, while the posterior glottic stenoses of group 2 were type 2, 3, and 4 in 1, 6, and 5 cases, respectively. Patients with posterior glottic stenosis, laryngeal necrosis and subglottic stenosis benefited from neck and chest CT-scan, which did not find another respiratory abnormality. Patients with a history of tracheostomy developed posterior glottic stenosis (N=4), laryngeal edema (N=4) associated with granuloma in two cases.
The surgical approaches performed in patients are reported in Table 2. Laryngeal necroses and posterior glottic stenoses were treated with corticosteroids and antibiotics, which led to effective results in 7 cases. Patients with posterior commissure hypertrophy, laryngeal diffuse edema and granuloma received antireflux diet, PPIs and alginate. Two patients with lack of improvement of posterior commissure hypertrophy and laryngeal erythema benefited from HEMII-pH, which confirmed the LPR diagnosis. Surgical treatment was proposed to patients who did not improve symptoms and findings posttreatment. The following surgical approaches were performed in 16 patients: granuloma laser excision (N=4), Montgomery-type calibration tube placement (N=4), dilatation (N=3), laser posterior transverse cordotomy (N=2), laser flange resection (N=1), and vocal fold fat medialization (N=1). One patient benefited from concurrent posterior transverse cordotomy and laser excision of granuloma. All patients benefited from speech therapy prior and after treatment. The number of intubation days was significantly higher in patients with posterior glottic stenosis (26.1 ± 9.4) compared with those presenting posterior commissure hypertrophy (11.5 ± 2.9) or granuloma (15.1 ± 5.8; p<0.001). Hypertension and diabetes were not associated with the development of any laryngeal injury.