Validation of the Brazilian Portuguese Version of the Reflux Finding Score

Abstract

Laryngopharyngeal reflux (LPR) poses a diagnostic challenge. Clinical diagnosis, based on suggestive symptoms and laryngoscopic signs of inflammation, should be acceptable, as long as diligent differential diagnosis is sought. In order to minimize subjectivity, a number of diagnostic instruments have been proposed, being the most common the Reflux Symptom Index and the Reflux Finding Score (RFS). The latter has been translated into several languages including Portuguese, but it still has not been properly validated in this language. To validate the Brazilian Portuguese version of the RFS. For validity and internal consistency, 172 adults were studied (106 with LPR and 66 healthy controls). Flexible transnasal laryngoscopy images were randomly examined twice by each of the two experienced otolaryngologists with a 72-hour interval. Strict exclusion criteria were applied to avoid other possible known causes of chronic laryngitis. For assessment of reproducibility and temporal stability, a random sample of 108 subjects (53 patients and 55 controls) were tested and retested. A statistically significant difference was observed in the mean RFS between patients with LPR (10.26±3.58) and controls (5.52±1.34) (P < 0.001). The interclass correlation coefficient comparing test and retest for both raters was high (R1=0.956; R2= 0.948). The Brazilian Portuguese version of the RFS proved to be a reliable and reproducible instrument for the diagnosis of LPR with a sensitivity of 82.08%, a specificity of 93.94%, a positive predictive value of 95.60%, and a negative predictive value was 76.54%.