Letter: symptom indexes in reflux monitoring - two are better than one

SIRS, We read with interest the article by Jonasson et al.1 on the use of a symptom-based questionnaire (GerdQ) as diagnostic tool for gastro-oesophageal reflux disease (GERD). The authors concluded that GerdQ is a valid complementary tool for the diagnosis of GERD, objectively identified by erosive oesophagitis (EO) at endoscopy and/or pathological oesophageal acid exposure and/or positive symptom association probability with acid reflux (SAP+ if 95%), and can be used to reduce the need for upper endoscopy and improve resource utilisation. Moreover, the study provided additional relevant data on the limited value of the proton pump inhibitor trial as a diagnostic test for GERD, as we have recently demonstrated with the current state-of-the-art technique to diagnose GERD.2, 3 However, we believe that the strength of their experimental findings would have improved if the results of symptom index (SI) were also reported. Indeed, it is still debated which symptom association index should be better used in clinical practice. Ward et al. firstly introduced the SI that has the advantage to be very easy to calculate, but has a major shortcoming as it does not take into account the total number of reflux episodes with the likelihood risk that a symptom is found to be associated with reflux by chance.4 On the other hand, the more recent SAP, as a statistical parameter, is not brought by chance and better explores the relationship between symptoms and reflux,5 but recent studies have questioned this index.6 Given the above data, we believe that the use of both symptom indexes, which can be calculated using both pH-metry and impedance-pH monitoring, are necessary to corroborate the final diagnosis of GERD. Moreover, at variance with the caution suggested by Slaughter et al., we have recently shown in patients with EO and nonerosive reflux disease (NERD) that they are both helpful in increasing the possibility of identifying patients with reflux disease.7, 8 Therefore, in conclusion, we believe that both symptom indexes should be included in pathophysiological studies using impedance-pH monitoring to improve the diagnostic yield.