Does the Severity of Vocal Fold Paresis on Laryngeal Electromyography Correlate With Radiographic Findings on Cross Sectional Imaging?

Document Type


Publication Date



OBJECTIVES: Unilateral vocal fold paresis or paralysis (UVFP) is a condition for which radiographic evaluation and laryngeal electromyography (LEMG) are valuable to evaluate severity of injury and direct treatment plans. Despite extensive research involving findings suggestive of UVFP with radiographic imaging, no study has attempted to determine which findings suggestive of UVFP on imaging are clinically significant and suggest a need for intervention. The purpose of this study was to evaluate whether the severity of vocal fold paresis/paralysis affects the likelihood of encountering radiographic findings suggestive of UVFP. We also aimed to determine which findings suggestive of UVFP on imaging were clinically significant and were associated with surgical intervention.

MATERIALS AND METHODS: A retrospective chart review was conducted of patients who had been diagnosed with unilateral vocal fold paresis or paralysis and had been evaluated by CT scan and/or magnetic resonance imaging and laryngeal electromyography (EMG) between the dates of January 1, 2017 and January 9, 2018. Fisher's exact testing with Monte Carlo Simulation was utilized to determine statistical significance of identified relationships. Univariate analysis was conducted to assess for individual relationships between imaging results and the potential predictor variables. Chi square analysis was conducted with the various categorical variables to assess for any potential relationships to imaging results. Statistical significance was determined utilizing chi square analysis.

RESULTS: After data collection, 130 patients were included in the study population. Of the 112 patients with documented MRI results, 17% had a reported imaging abnormality suggestive of true vocal fold paresis or paralysis (VFP). Of the 71 patients with documented CT Neck results, 15.4% had an abnormality potentially concerning for true VFP. The average decrease in recruitment of the right and left SLN was 23.8% and 26.1%, respectively. The average decrease in recruitment of the right and left RLN was 37.3% and 57.78%, respectively. Seventy four percent of patients who exhibited abnormal MRI were found to have isolated SLN weakness, and 21% of patients were found to have a combined SLN and ipsilateral RLN weakness. In patients with abnormal CT scans 45% were found to have isolated SLN weakness, and 35% were found to have a combined SLN and RLN weakness. MRI imaging again failed to display any significant degree of paresis. However, abnormal CT results displayed severe CN X paresis in 36.84% vs 1.96% in normal scans. The chance of an abnormal MRI and CT result was 2.78 and 5.55 times greater, respectively, for each increase in the degree of severity of CN X paresis. When looking at the ability of imaging to predict the chance of a patient undergoing surgery, 34.8% of patients with an abnormal MRI underwent surgery compared to just 14.61% of those with normal scans. For CT scans, 35% of patients with an abnormal scan underwent surgery, compared with only 15.69% with normal imaging. When pooled, over 33% of patients with any abnormal imaging underwent a laryngeal procedure compared to 13% of patients with normal imaging.

CONCLUSIONS: There is a relationship between severity of vocal fold paresis found on laryngeal EMG and likelihood of detection on imaging. While CT was more likely to find characteristics of UVFP than MRI, patients who had an abnormal finding on either modality were more likely to undergo surgical intervention. These findings highlight the importance of early referral of patients with abnormal laryngeal imaging to an otolaryngologist for evaluation and possible intervention.


This article was published in Journal of Voice.

The published version is available at

Copyright © 2022 The Voice Foundation. Published by Elsevier Inc.

Publication Title

Journal of Voice

PubMed ID


This document is currently not available here.