When is gastroesophageal reflux the cause of symptoms?

Abstract

To the Editor: We read with great interest the article by Rudolph (J Pediatr Gastroenterol Nutr 2000;30:3–4) regarding the best approach to determining whether a specific symptom is caused by gastroesophageal reflux (GER). Gastroesophageal reflux disease is characterized by pathologic GER and associated symptoms. Twenty-four-hour esophageal pH monitoring is the best available diagnostic test for discriminating between physiological reflux and pathologic acid exposure. In adults, patients with normal esophageal exposure to acid and a significant association between reflux episodes and symptoms suggestive of GER disease have been described (1–3). Thus “acid-hypersensitive esophagus” or “sensitive esophagus” has been identified. In children, Orenstein (4) stressed the importance of investigating in every pH tracing the temporal relation between reflux episodes and symptoms. The Working Group of the European Society of Pediatric Gastroenterology and Nutrition stated that “for some individuals and for some clinical situations, it is more important to relate events (e.g., coughing, wheezing, apnea) to pH changes than to know if global results are within a normal range or not”(5). The North American Society for Pediatric Gastroenterology and Nutrition stated that “the symptom index (number of occurrences of a symptom with pH <4 divided by total number of occurrences of that symptom during an esophageal pH monitoring study) can be calculated to assess whether symptoms are due to GER”(6). Recently, we have reported three patients with “acid-hypersensitive esophagus” as the cause of recurrent abdominal pain (RAP) (7), chronic cough (8), and wheezing (9). In the three patients, the results of 24-hour pH monitoring were in the normal range; thus, GER disease was excluded. However, in the three patients, esophageal pH tracings demonstrated that all the symptoms were preceded by GER episodes. Fisher's exact test showed a significant relation between GER episodes and symptoms. To exclude the probability that symptoms coincided with GER episodes only by chance, in the patient affected by RAP, a calculated complex statistical approach based on the binomial law (10) produced positive results. In the patients with chronic cough and wheezing, both specificity and sensitivity indexes (11) were high. The three patients, after undergoing medical treatment for GER, became symptom free. As stressed by Rudolph, numerous patients with normal pH probe measurements improve after medical or surgical treatment of GER. Further studies are needed to indicate the role played by “acid-hypersensitive esophagus” in reflux-related symptoms in patients with normal esophageal exposure to acid. An emerging question is which statistical approach may be used to evaluate the relation between GER episodes and symptoms, in that, as indicated by Rudolph, the analysis methods used to calculate the probability that symptoms are associated with GER are not perfect and, in clinical practice, no method for demonstrating a symptom–reflux association has been prospectively evaluated against an independent criteria, such as the response to pharmacologic therapy. Moreover, in establishing a cause–effect relationship, there is an inherent difficulty in indicating the cause-and-effect link between GER episodes and symptoms, because the two events (GER episodes and symptoms) both could be the result of a third event—the true cause (12). Giovanni Corrado Marisa Cavaliere Claudia Pacchiarotti Mariassunta Porcelli Ettore Cardi