The main finding of this study is that patients with typical reflux symptoms and abnormal acid exposure have a high response rate to standard esomeprazole regardless of whether they have ERD or NERD. Two thirds (22/31) of patients with typical GERD-related symptoms had an abnormal acid exposure in esophageal BRAVO pH metry. Including 2 patients with erosive changes but normal pH metry, 24 patients were eligible for PPI treatment in our study.
88% of this well selected patient group achieved complete symptom relief on esomeprazole standard dose for 4 weeks. Symptomatic response was similarly obtained in patients with ERD and NERD (Table 2; Figure 2). The claim that patients with NERD would have a worse response to PPI is therefore most likely due to the inclusion of patients without abnormal gastroesophageal reflux in previous studies. Misdiagnosis of GERD – NERD in particular – might also explain reasonably the high PPI failure rate in previously published data. Weijenborg and colleagues systematically reviewed previous outcome studies and found only 2 studies defining NERD by both negative endoscopy and a positive pH-test. In contrast to poor response rates in empirical treated or endoscopy-negative patients, the pooled estimate rate of complete relief of heartburn after 4 weeks of for those accurately diagnosed NERD was 0.73 (95% CI 0.69-0.77) and comparable to patients with ERD . This clinical data indicates to careful asses the diagnosis of NERD and differentiate especially from functional heartburn to predict a therapeutic success of current PPI therapy.
We excluded patients with normal acid exposure as there is no rationale for PPI treatment. This category of patients is likely to account for the frequent reports with up to 30-40% PPI failure to standard dose [15, 21]. In routine practice in Germany, the response rate to PPI is 60% . For patients not responding to PPI in presence of typical symptoms, functional testing is performed to test the initial diagnosis and to further investigate for conditions that might explain PPI refractoriness. Among them, persistent acid or non-acid reflux episodes have been reported to be responsible for incomplete symptom relief [6, 23–26].
In a further subset of patients, reflux symptoms may be unrelated to reflux episodes at all and related to a functional syndrome (functional heartburn) . Although unable to determine the proportion of “non acidic” reflux episodes by BRAVO pH metry, our study reemphasizes the importance of the patients’ interview and interpretation of symptoms to distinguish between acid-related symptoms and functional disorders that often overlap and requires different medical treatment [27, 28].
For patients not responding to PPI, pH metry should be considered to confirm the diagnosis of abnormal gastroesophageal reflux. Mechanisms involved in symptom generation or perpetration are either hypersensitivity to visceral stimuli or weakly acidic reflux episodes, a fast hepatic metabolism of PPIs  or duodenogastroesophageal reflux (DGER) [15, 30]. Intestinal proteases in the refluxate and interaction with epithelial protease-activated receptors are also involved in the pathogenesis of mucosal inflammation in GERD pathogenesis [31, 32].
The shortcomings of the study are the missing control group and the small sample size. This was mainly due to the inclusion criteria of PPI naive patients in a referral centre. As calculated before, the recruitment was finalized after having screened 40 patients.
In spite of the small sample size, the results indicate daily clinical practice. Nevertheless, our study has the true advantage of having included truly PPI-naive patients, a fact that is very hard in routine clinical practice, as most physicians administer PPI very quickly based on current guidelines. However, this “aggressive” approach might need to be rethought, as we believe that many patients receiving PPI do not suffer from NERD or ERD, and thus being over treated. Thus, a careful initial assessment of symptoms combined with functional testing may identify the patients who respond well to PPI therapy. This fact needs to be reconsidered in the interpretation of many clinical trials concerning response to PPI therapy, especially in NERD .