Advances in Optical Adjunctive Aids for Visualisation and Detection of Oral Malignant and Potentially Malignant Lesions

Table 3

Published papers on the use of NBI in the oral cavity.

Paper	Purpose	Type of study	Sample population	Sensitivity; (%)	Specificity; (%)	PPV; (%)	NPV; (%)	Accuracy; (%)	Notes
Katada et al. [63]	Used to detect superficial SCC	Case report	2 patients with oesophageal cancer.	—	—	—	—	—	Coincidental finding of OSCC at floor of mouth.
Piazza et al. [68]	Used to detect dysplasia and SCC	Prospective cohort study	96 patients with biopsy-confirmed or previous-treated OSCC or oropharyngeal SCC.	96	100	100	93	97	Combined the use of NBI with high-definition television.; 27% (26 of 96) patients had a diagnostic benefit with the use of NBI and high-definition television.
Takano et al. [61]	Investigated the types of IPCL patterns	Prospective cohort study	41 patients with normal mucosa, or nonneoplastic or neoplastic lesions.	—	—	—	—	—	Devised the IPCL classification of oral mucosa.
Chu et al. [69]	Used to detect dysplasia and OSCC	Prospective cohort study	101 patients with treated OSCC.	95%	97%	91%	99%	97%	Had difficulty diagnosing hyperkeratotic lesions, tumours at the tongue base, and recurring tumours.; Chronic inflammation, postoperative radiation, colouration, or mucosal staining interfered with diagnosis.
Lin et al. [70]	Investigated the visibility of brownish spots in different types of epithelium	Prospective cohort study	125 patients with CIS or SCC in the head and neck.	—	—	—	—	—	Areas with nonkeratinized thin stratified squamous epithelium had a significantly higher prevalence of brownish spots than areas with keratinized epithelium or epithelium thicker than 500 𝜇m.
Tan et al. [64]	Used NBI to influence management of oral erythroplakia	Case report	1 patient with erythroplakia.	—	—	—	—	—	NBI was used to determine resection margins, which were beyond the clinically visible margins.
Yang et al. [65]	Correlated NBI clinical findings with histopathology	Retrospective cohort study	154 patients with oral leukoplakia.	—	—	—	—	—	The IPCL classification had a significant statistical association with the severity of pathology.
Yang et al. [67]	Evaluated the use of NBI for assessing and managing oral leukoplakia	Retrospective cohort study	160 patients with clinical homogenous oral leukoplakia.	—	—	—	—	—	All cases of thin leukoplakia had IPCL Type I and were confirmed as squamous hyperplasia.; Thick leukoplakia had IPCL Type I, II, or III, and a significant correlation between pathology and NBI images was present (𝑃<0.0001).
Yang et al. [71]	Used IPCL patterns made visible by NBI to diagnose high-grade dysplasia, CIS, and OSCC	Retrospective case-control study	414 patients with oral leukoplak.ia	15%	60%	7%	79%	53%	Criteria used was “brownish spots and demarcation line with irregular microvascular pattern”.
				77%	55%	24%	93%	58%	Criteria used was “well-demarcated brownish area with thick dark spots and/or winding vessels”.
				85%	95%	75%	97%	93%	Criteria used was “the intraepithelial papillary capillary loop (IPCL) Type III … and Type IV”.
Yang et al. [66]	Investigated the IPCL morphology of OSCC and correlated the pattern with infiltration depth and disease severity	Retrospective cohort study	80 patients with-biopsy confirmed OSCC	—	—	—	—	—	The IPCL pattern moved from tortuous and dilated to twisted and elongated to angiogenesis and destruction of IPCL as the severity of OSCC increased.; Depth of infiltration increased with the degree of severity.
Nguyen et al. [72]	Compared white light, autofluorescence, and NBI to detect moderate dysplasia or worse	Prospective cohort study	73 patients with known or treated head and neck SCC	96	79	85	94	—	Autofluorescence and NBI were significantly more sensitive than white light.; NBI and white light were more specific than autofluorescence.

NBI: narrow band imaging; OSCC: oral squamous cell carcinoma; SCC: squamous cell carcinoma; IPCL: intrapapillary capillary loop; and CIS: carcinoma in situ.