Frequency scale symptoms for gastroesophageal reflux disease (Frequency Scale for Symptoms of GERD) predicts need for addition of prokinetics to proton pump inhibitor therapy

Abstract

Proton pump inhibitors (PPI) are highly efficacious in providing symptomatic relief, healing erosions and improving quality of life in patients with gastroesophageal reflux disease (GERD).1 Furthermore, PPI achieve the above objectives faster than any other comparable medical therapy.2 Prokinetics have been used in conjunction with PPI to treat GERD, although their efficacy was thought to be modest, at best.3 Conflicting evidence exists regarding the role of cisapride in improving esophageal clearance or increasing lower esophageal sphincter pressure, although its effect on gastric motility appears more convincing. In contrast to cisapride, which acts only via 5HT4 receptors, mosapride is also a 5HT3 antagonist.4 Thus, it is believed that mosapride has the potential to have a greater mechanistic effect on protective motor functions involved in GERD. In the May issue of the Journal of Gastroenterology and Hepatology, Miyamoto and colleagues sought to assess whether a GERD questionnaire can predict a lack of response to PPI therapy. One hundred and sixty-three subjects with symptomatic GERD were given a Frequency Scale for Symptoms of GERD (FSSG) questionnaire which had been developed in-house. They were then treated with rabeprazole (10 mg daily).5 Reassessment of symptoms was carried out at 12 and 24 weeks. Patients ‘satisfied’ with their treatment were given three patient-directed options: (i) stop PPI; (ii) downgrade to H2-receptor antagonists (H2RA); or (iii) continue with the current dose of rabeprazole. Approximately 20% of patients who were dissatisfied with their symptom control were treated with additional prokinetic therapy (mosapride 5 mg t.d.s.). A second assessment occurred after a further 12-week period (i.e. 24 weeks from the start of rabeprazole therapy). At 24 weeks, almost all patients (98%) were satisfied with their symptom control. The authors claim that this study shows that the addition of prokinetic therapy is beneficial in a subset of patients who are inadequately controlled with PPI monotherapy, and that the FSSG questionnaire they have developed is useful in predicting potential therapy failures at the outset. This interesting article evokes several considerations before conclusions can be summarized. As Asia–Pacific guidelines recently published in the Journal endorse, therapeutic goals for GERD management outline that subjects should be started on the most effective therapy at the beginning; step-up therapy is an antiquated approach as it impacts adversely on quality of life.6 Although, PPI are extremely effective in healing erosive esophagitis and obtaining symptomatic remission, a suboptimal response has been observed in a limited number of patients. A small proportion of patients with severe erosive GERD (Los Angeles Grades C and D) may not achieve adequate healing with standard doses of PPI.7 Additionally, a subset of patients with non-erosive GERD also appears to have heartburn that may prove refractory to PPI. Increasing the PPI dose (or twice daily dosing) appears to improve healing rates in patients with severe erosive reflux disease. Moreover, subjects with severe erosions are easy to recognize at endoscopy. Thus, they can be flagged for dose escalation right at the outset. However, identifying potential non-responders within the non-erosive GERD population has proven to be difficult.8 Also, doubling the PPI dose may not exhibit positive impact on symptomatic control in patients with non-erosive GERD.9 Questions incorporated in the FSSG were distilled from a long symptom questionnaire given to a GERD focus group. FSSG includes questions about symptoms that are typical for GERD, and also those pertinent to dyspepsia/dysmotility. Then, what specifically does the FSSG instrument identify as predictors? Why does FSSG succeed? Recent studies clearly show an overlap between GERD and irritable bowel syndrome (IBS).10 Conceptually, bloating, a question incorporated in FSSG, and a symptom potentially amenable to mosapride treatment, could have been a predictor. A small subset of patients with GERD has been recognized to have poor gastric emptying. So, is it patients with bloating or early satiety that are recognized by this questionnaire as being putative non-responsive to PPI therapy? Factor analysis is a statistical technique that reduces a large number of interrelated variables (such as the questions in the FSSG) to a small number of underlying common domains that are mainly responsible for covariation of the data. Exploratory factor analysis could have provided important insights into this issue. Perhaps it may have explained the degree of variance accounted by the key items in their instrument. Unfortunately, those data are not provided in the Miyamoto article. Last, authors have only shown results of univariate analysis. An important question is whether these results (especially usefulness of the FSSG questionnaire) would retain their significance in multivariate analyses? Many Western countries use 20 mg rabeprazole as the standard dose for inducing remission of GERD symptoms. Step down to 10 mg can be undertaken if and when symptom resolution has been achieved. The authors argue that their population are low-acid secretors and, hence, they used a smaller dose of PPI (10 mg). Furthermore, they state that the economic climate in Japan favors addition of a prokinetic agent rather than doubling of PPI dose. However, formal cost–benefit analyses should be carried out, together with sensitivity analyses that consider whether assumptions made and conclusions drawn are likely to be applicable to other countries. Treatment effect in symptomatic GERD can be assessed as complete or incomplete. Dekkers et al. showed that, at 8 weeks, 38% of patients have complete resolution of their symptoms, whereas 73% have significant improvement from baseline.11 In the current study, outcome assessments by patients themselves were categorized as satisfied versus dissatisfied. Agreement between physician and patient regarding their assessment of symptomatic GERD can be poor.12 Thus, patients' own assessments, as carried out in this study, may be more acceptable. A much higher response rate was shown by Foch et al., who treated in 163 GERD patients in Singapore with 10 mg rabeprazole; they reported 98% satisfaction at 4 weeks.13 How do we reconcile the results of these two studies? It is unclear from the article whether the unsatisfactory response reported in the current study was specific for heartburn and acid regurgitation or whether the symptom response was influenced by ancillary symptoms assessed collectively with heartburn and acid regurgitation. This distinction is important as it is conceivable that it was the lack of improvement in dysmotility rather than GERD symptoms that could have led to the dissatisfaction. Support for such a hypothesis is given by the fact that female gender, low body mass index and functional constipation were also found to be predictors of refractoriness to PPI therapy. An alternative way of interpreting the data is that typical GERD symptoms could have improved in almost all patients, but that such improvement led to the unmasking of dysmotility symptoms in a subset, thereby leading to dissatisfaction in that group. The value of promotility agents was evaluated at the Asia–Pacific GERD consensus meeting.6 It was determined that prokinetics may have a role in the treatment of GERD, although there is limited evidence to support their use.6 In a randomized controlled study in northern India, the combination of mosapride and pantoprazole was compared to pantoprazole alone in the treatment of GERD.14 That study found that the addition of mosapride significantly improved symptom control in patients with erosive reflux disease. However, the combination therapy was not superior to pantoprazole alone in providing symptomatic relief in the non-erosive reflux disease group. Moreover, addition of mosapride did not improve the healing rates of erosions. Although PPI are very effective in the therapy of GERD, there are still unmet clinical needs.15 Adequate symptomatic control and patient satisfaction can be lacking in a small minority. The dissatisfaction in this group may occur due to low or non-acid reflux, or may stem from ancillary symptoms which may be functional or dysmotility-related (IBS). Whether, mosapride can claim a rightful place in the treatment algorithm clearly needs more study. The FSSG questionnaire could be an interesting novel instrument for assessing additional approaches for symptom relief in patients deemed to have GERD, but its effectiveness needs to be reproduced by other independent studies before it can be applied more generally.