RESUMO
OBJECTIVE: to investigate the reliability of laryngeal reflux finding score (RFS) and symptom index (RSI) in assessing gastroesophageal reflux (GER) in children and infants. METHODS: patients with laryngeal or respiratory symptoms, who underwent laryngoscopy and esophageal pH-impedance (MII-pH) were recruited. RSI and RFS were correlated to MII-pH results. A RSI>13, RFS>7, acid exposure index>7%, total reflux episodes>100/24 h in infants or>70/24 h in children, or a positive symptom index or association probability, were considered pathological. Analysis considering age (12 months) was performed. Sensitivity, specificity, positive (PPV) and negative predictive values (NPV) of the laryngeal scores were calculated. RESULTS: 197 children (median 53 months, 61 infants) were enrolled. MII-pH was pathological in 5/10 patients with RFS>7, and in 17/31 with RSI>13. RFS>7 had 3.9% sensitivity, 93% specificity, a PPV of 50 and a NPV of 34 in predicting GER disease. RFS was inversely associated to weakly acidic and proximal GER. RSI>13 had 13% sensitivity, 83% specificity, and a PPV and NPV similar to RFS. RSI was significantly associated with the number of acid reflux episodes, and, in infants, with bolus exposure index. CONCLUSIONS: RSI and RFS aren't accurate in predicting GER in infants and children. Acid reflux relates to laryngeal symptoms, but neither acid, nor proximal and weakly acidic GER relate to laryngeal alterations.