STARD 2015: An Updated List of Essential Items for Reporting Diagnostic Accuracy Studies

Is EAT-10 Useful to Assess Swallowing during the Chemo-Radiotherapy Phase in Patients with Head and Neck Cancer? A Pilot Study

OBJECTIVE/HYPOTHESIS

The 10-item Eating-Assessment Tool (EAT-10) is a dysphagia screening test. In HNC patients, screening and diagnosis of dysphagia are not well-established. To determine the metrological properties of the EAT-10 compared with videofluoroscopy in non-surgical HNC-patients and to assess the relationship between EAT-10 scores and patients' self-reported symptoms.

STUDY DESIGN

Prospective cohort study.

METHODS

Forty-six HNC-patients recently diagnosed and referred to chemoradiotherapy (CRT). Main outcome was evidence of dysphagia according to EAT-10 score, self-perception on a Visual Analog Scale (VAS) of impaired swallowing, severity on the Penetration-Aspiration Scale (PAS), and the Functional Oral Intake Scale (FOIS). Patients were assessed at baseline, before-CRT, after-CRT, and at 3-month follow-up.

RESULTS

A strong baseline correlation between EAT-10, VAS, and FOIS was observed. All 3 values decreased in weeks 6 to 9 after CRT initiation; a poor correlation of EAT-10 with VAS was observed at 3-month follow-up. A receiver operating characteristic curve determined new cut-off points (sensitivity/specificity) for safe swallowing: baseline 3 (86%, 77%); post-CRT, 15 (62.5%, 80%); and 3-month follow-up, 4 (83%, 75%).

CONCLUSIONS

New safe-swallow EAT-10-points are suggested for this population during screening and the oncological follow-up. A poor correlation between EAT10-score and patient self-reported symptoms was observed at the end-RT and at 3-month follow-up, highlighting the need for an objective evaluation instrument.