OBJECTIVE: The aim of this study was to compare narrow band imaging (NBI) endoscopy and Storz Professional Image Enhancement System (SPIES) in observing epithelial and/or subepithelial microvascular irregularities and pathologies. METHODS: A total of 73 patients with laryngeal or hypopharyngeal lesions were investigated using high-definition NBI endoscopy preoperatively in local anesthesia and using SPIES system intraoperatively in general anesthesia from August 2016 to October 2017. Superficial vascular structures were classified preoperatively (NBI) and intraoperatively (SPIES) according to descriptive guidelines of vascular changes by Arens. All lesions were endoscopically evaluated and divided according to the histological examination into four groups (A-benign lesions, B-recurrent respiratory papillomatosis, C-low-grade dysplasia, D-high-grade dysplasia, carcinoma in situ or invasive squamous cell carcinoma), and results were compared with NBI and SPIES optical biopsy. RESULTS: Benign lesions (polyps, cysts, chronic inflammation, hyperkeratosis) were histologically confirmed in 26/73 (35.6%) cases and identified by NBI in 20/26 lesions (76.9%) and in 20/26 cases (76.9%) by SPIES, respectively. Recurrent respiratory papillomatosis was confirmed in 16/73 (21.9%) and detected in 15/16 cases (93.8%) by NBI and in 16/16 cases (100.0%) by SPIES. Low-grade dysplasia (mild and moderate dysplasia) was histologically detected in 7/73 patients (9.6%) and accurately identified by NBI in 6/7 (85.7%) and by SPIES in 6/7 (85.7%) cases, respectively. Histopathological features of severe dysplasia, carcinoma in situ or invasive squamous cell carcinoma were detected in 24/73 (32.9%) patients. According to the NBI endoscopy the suspected vascular neoangiogenesis was recognized in 19/24 cases (79.2%) and in 18/24 cases (75.0%) using SPIES endoscopy. Sensitivity and specificity of NBI endoscopy and SPIES system in correct prediction of histological diagnosis of already detected lesions were 83.0 and 98.0% and 86.0 and 96.0%, respectively. Results of NBI/SPIES endoscopy and histopathological features of laryngeal and hypopharyngeal lesions were compared and the level of agreement was 81.43%, kappa index κ = 0.7428 (95% CI 0.682-0.832) (p < 0.001) by NBI endoscopy and 81.16%, kappa index κ = 0.7379 (95% CI 0.638-0.880) (p < 0.001) by SPIES endoscopy, respectively. The agreement was confirmed as substantial and strong. Level of agreement of both endoscopic methods was 92.54%, kappa index κ = 0.8965 (95% CI 0.877-0.954) (p < 0.001), agreement was confirmed as almost perfect. Between NBI and SPIES endoscopic imaging methods is no significant differentiation. CONCLUSION: Both methods, NBI endoscopy and SPIES system, are comparable in detection and analysis of superficial neoangiogenesis, typical for benign lesion and for precancerous or cancerous changes in larynx and hypopharynx.

Department of Otorhinolaryngology and Head and Neck Surgery University Hospital Ostrava 17 listopadu 1790 708 52 Ostrava Czech Republic

Department of Otorhinolaryngology and Head and Neck Surgery University Hospital Ostrava 17 listopadu 1790 708 52 Ostrava Czech Republic Department of Craniofacial Surgery Faculty of Medicine University of Ostrava Ostrava Czech Republic

References provided by Crossref.org