0501 Comparing Ultrasound to MRI for Volumetric Analysis of Tongue. (27th April 2018)

• Record Type: Journal Article
• Authors: Wiemken, A; Keenan, B; Schwab, R
• Is Part Of: Sleep. Volume 41(2018)Supplement 1
• Page Start: A188
• Page End: A188
• Publication Date: 2018-04-27
• Subjects: Sleep -- Physiological aspects -- Periodicals; Sleep disorders -- Periodicals; Sommeil -- Aspect physiologique -- Périodiques; Sommeil, Troubles du -- Périodiques; Sleep disorders; Sleep -- Physiological aspects; Sleep -- physiological aspects; Sleep Wake Disorders; Psychophysiology; Electronic journals; Periodicals; 616.8498
• DOI: 10.1093/sleep/zsy061.500
• Languages: English
• ISSNs: 0161-8105

Abstract

Abstract: Introduction: Tongue size is an important risk factor for obstructive sleep apnea (OSA). Magnetic resonance imaging (MRI) is the gold-standard method for determining tongue volume; however, it is expensive and time-consuming. Ultrasound is non-invasive and can be performed in a 5–10 minute timeframe. Validating measures from ultrasound against MRI-based measurement could lead to significant clinical applications in predicting sleep apnea risk and oropharyngeal characteristics. We hypothesized that tongue volume derived from MRI and ultrasound would be similar. Methods: Thirteen participants (7 female, 5 apneic, 5 African-American, 8 Caucasian, median age 37 [IQR=30–53], median BMI 26.8 [IQR=23.0–35.3]) underwent both axial T1 Spin-Echo MRI of the upper airway and submental ultrasound to measure tongue volume. Amira software was used to obtain MRI measurements. Tongue length and diameter were derived from ultrasound images using ImageJ software, and used in a formula modeling the tongue as two joined conical frustums to obtain volume. A Bland-Altman agreement analysis was performed to compare ultrasound to gold-standard MRI volumes; differences were calculated as tongue volume on ultrasound minus volume on MRI. Results: In our sample of 13 patients, there was no significant difference between tongue volume obtained on ultrasound as compared to MRI (mean [95% CI] difference = 3352 [-2546, 9249] mm 3 ; p=0.2393). There was a positive, but non-significant, correlation between the difference and mean value on the two techniques (rho=0.349, p=0.2424), suggesting no systematic bias in differences. Bland-Altman limits of agreement (equal to ±2 standard deviations) ranged from -16167 to 22870 mm 3 . Conclusion: Based on these preliminary data, ultrasound appears to be a promising high-throughput method of determining tongue volume; results should be examined in larger samples. Ultrasound could be used as a screening method to determine tongue size and apnea risk in population cohorts or for determining candidates for alternative treatments such as hypoglossal nerve stimulation. Support (If Any): Research reported in this publication was supported under NIH award number P01HL094307-06.