Diagnostic accuracy of the Gugging Swallowing Screen and exploratory submental ultrasonographic findings in dysphagia screening for Parkinson disease: a retrospective cross-sectional study using videofluoroscopy as reference standard

Background: Oropharyngeal dysphagia is common in Parkinson's disease, but it may remain clinically silent until complications occur. Bedside screening is widely used in routine care, although its diagnostic accuracy remains uncertain. Submental ultrasonography has emerged as a non-invasive adjunct for bedside assessment of swallowing-related structures and movement. Methods: In this retrospective cross-sectional study, institutional records of adults with Parkinson's disease assessed between October 2021 and June 2023 were analyzed. Bedside swallowing screening was performed with the Gugging Swallowing Screen. Submental ultrasonography was used to measure tongue thickness and hyoid-thyroid approximation during swallowing. Videofluoroscopic swallowing study served as the reference standard. Swallowing safety was classified as safe or unsafe according to the Penetration-Aspiration Scale, with unsafe swallowing defined as a score of 3-8. The primary outcome was the diagnostic performance of the Gugging Swallowing Screen for identifying unsafe swallowing on videofluoroscopic swallowing study. Results: Thirty-nine participants were enrolled, and 36 completed videofluoroscopic swallowing study. A positive Gugging Swallowing Screen result was observed in 7 of 36 participants, corresponding to 19.4%, whereas unsafe swallowing on videofluoroscopic swallowing study was observed in 12 of 36 participants, corresponding to 33.3%. The Gugging Swallowing Screen showed moderate discriminative ability, with an area under the curve of 0.792. Sensitivity was 58.3%, specificity was 100%, overall accuracy was 86.1%, positive predictive value was 100%, and negative predictive value was 82.8%. Participants with unsafe swallowing had lower tongue thickness and shorter resting and minimum hyoid-thyroid distances. Conclusions: In Parkinson's disease, the Gugging Swallowing Screen may help prioritize referral for instrumental assessment, particularly when abnormal, but it cannot reliably exclude unsafe swallowing when used in isolation. Submental ultrasonographic measures differed between videofluoroscopy-defined swallowing-safety groups; however, their diagnostic utility, optimal cut-offs, and incremental value over the Gugging Swallowing Screen require prospective validation.