Review article: intragastric and oesophageal pH monitoring in patients with gastro-oesophageal reflux disease


Dr P. O. Katz, Division of Gastroenterology, Albert Einstein Medical Center, AEMC, Klein Building, 5501 Old York Road, Suite 331, Philadelphia, PA 19141, USA.


Treatment of gastro-oesophageal reflux disease with acid suppressive therapy is based on the principle that effective control of intragastric pH (a marker of acid control) leads to healing of erosive oesophagitis and relief of gastro-oesophageal reflux disease-associated symptoms.

Most patients with gastro-oesophageal reflux disease can be managed successfully with current antisecretory therapy. In difficult-to-treat patients, oesophageal pH monitoring is a useful technique to assess pH control and to investigate the association of reflux with refractory symptoms.

Intragastric pH monitoring allows direct assessment of acid suppression achieved with an agent, and is the most useful for head-to-head comparisons of antisecretory therapies, evaluating variability in individual gastric pH response, assessing dose timing and food effect, and determining alternate dosing strategies; as such, it is useful in the research laboratory, and may be useful clinically to guide clinicians in dose titration and in evaluating the effect of switching agents.

This article reviews these and other uses of these tests in an effort to explore the question of how to optimally use oesophageal pH monitoring and gastric pH monitoring in patient management.

Oesophageal pH monitoring in reflux disease diagnosis

The availability and widespread use of proton pump inhibitor (PPI) therapy has dramatically changed the use of oesophageal pH monitoring, such that in modern practice, there are few reasons to perform oesophageal pH monitoring as a primary diagnostic intervention in gastro-oesophageal reflux disease (GERD). Off-therapy pH monitoring is used in preoperative documentation of reflux in patients who are being considered for antireflux surgery or endoscopic therapy. In patients with either typical or atypical symptoms of GERD, without evidence of oesophageal erosions, oesophageal pH monitoring off-therapy is used to confirm the presence or absence of abnormal reflux in patients refractory to acid suppression therapy, in whom on-therapy pH monitoring has failed to clarify the clinical picture.1

Despite the decrease in use as a primary diagnostic test for GERD, oesophageal pH monitoring has been invaluable in documenting the severity and pattern of acid reflux, and correlating these with GERD pathology. Though multiple studies have been performed, only a few will be highlighted. Frazzoni et al. used ambulatory oesophageal pH monitoring to explore the relationship between the pattern of reflux in patients with GERD, specifically levels of supine reflux, and the severity of disease.2 The study enrolled 220 patients with GERD of varying severity: 25% (n = 56) had what the investigators termed complicated reflux disease [Barrett's oesophagus (BE) or ulcerative oesophagitis], 35% (n = 76) of patients had erosive oesophagitis (EE), and the remaining 40% (n = 88) had non-erosive reflux disease (NERD). Median time between 24-h pH monitoring and last endoscopy equals 2 months. The results showed a graded increase in oesophageal acid exposure (EAE) with increasing disease severity (Figure 1).2 Mean total percentage of the 24-h period with acid reflux was significantly greater in patients with complicated GERD (19%) compared with patients with EE (15%; P = 0.024) and NERD (12%; P = 0.000). Also, patients with EE had a significantly higher total per cent acid reflux time vs. patients with NERD (P = 0.017). The difference was because of a larger percentage of supine nocturnal acid reflux in more severe disease; there was no difference between patient groups in upright diurnal reflux. As in other investigations with oesophageal pH monitoring, some of the patients with endoscopically documented disease in this trial had normal pH-metry results: 13% (n = 10) of patients with EE and 4% of patients with complicated GERD, for a total false-negative rate of 9%. However, this rate decreased to 2% when the investigators considered the results of the supine period alone.2

Figure 1.

Mean total percentage acid reflux time over a 24-h period in patients with varying disease severity.2

Campos et al. looked at the association between acid reflux and disease severity to determine if the pattern of reflux could be used as a risk factor to identify patients with more severe GERD.3 To enter the study, patients were required to have had pathological acid reflux documented with a prior 24-h pH monitoring test. Patients (n = 401) were divided into four groups: (i) postprandial refluxers, defined as patients with abnormal EAE in the 2 h after a meal, but with normal EAE in the upright and supine positions; (ii) upright refluxers, defined as patients with abnormal EAE in the upright position, regardless of EAE in the postprandial period; (iii) supine refluxers, defined as patients with abnormal EAE only in the supine position; and (iv) bipositional refluxers, defined as patients with abnormal upright and supine EAE. This group found that the prevalence of mucosal injury was significantly higher in patients with bipositional reflux compared with the other groups (P ≤ 0.01), and that disease severity progressed linearly across the postprandial, upright, supine and biposition reflux groups (Figure 2).3

Figure 2.

Relationship between the prevalence of oesophageal mucosal injury and pattern of oesophageal acid reflux.3

Although the association between extent and pattern of abnormal EAE and disease severity has been fairly well documented, the relationship between symptoms and pH monitoring test results is less clear. This is highlighted by a study by Baldi et al. of 17 EE patients using oesophageal pH probes, placed distally and proximally in the oesophagus (5 and 10 cm), which showed that, although a minority of reflux episodes provoked symptoms, the majority of symptom episodes could be associated with acid reflux. Overall, 85% of acid reflux events, regardless of disease severity, failed to result in symptoms,4 raising questions as to what characteristics of reflux are needed to produce symptoms.

Several indices have been devised to attempt to relate reflux events to symptoms, the oldest and simplest of which is the symptom index (SI). This calculation was introduced by Ward et al.5 in 1986, and then validated in a study by the same group as a means to quantify the association between GERD symptoms and oesophageal pH <4.6 The SI equals the number of reflux-related episodes divided by the number of total symptom episodes and then multiplied by 100%. Most of the patients in this study had scores either above 75% or below 25%. A good association was found between high and low SI percentages and the presence and absence, respectively, of gastro-oesophageal reflux: over 90% of patients with a high SI had abnormal acid reflux, and >70% of patients with a low SI had normal pH monitoring results. The index was not particularly sensitive for predicting muscosal injury, however, as almost 30% of patients with abnormal pH monitoring and high SI percentages had normal endoscopies. This aspect of the SI is of special importance in patients with NERD, aiding clinicians in making a positive GERD diagnosis in the absence of endoscopic evidence of disease.6

Although the SI is the easiest to use in clinical practice, it has inherent problems in patients who have a very small number of symptom episodes or who have intermittent symptoms. Singh et al., in a retrospective study of 153 patients, found that an SI score of ≤50% had high sensitivity (93%) and good specificity (71%) for GERD diagnosis, especially in patients who reported multiple symptom episodes.7 Patients presenting with either oesophageal (heartburn, regurgitation) or extraoesophageal (chest pain, hoarseness) symptoms were divided into three groups: those with abnormal pH monitoring and normal endoscopies, those with normal results for both tests, and those with abnormal results for both tests. Although the reported occurrence of heartburn did not differ between the groups, the SI for this symptom was significantly higher in patients with abnormal pH tests (P < 0.001). The SI is not as useful in patients with non-cardiac chest pain (NCCP). Singh et al. also analysed the association between reflux and chest pain.7 As with heartburn, episodes of chest pain were similar between the three patient groups; however, unlike with heartburn, the mean SI was also similar, and an optimal threshold for using the SI to delineate acid reflux-related chest pain could not be determined. An SI of ≥25% was fairly sensitive at 68%, but specificity was only 50%, whereas an SI of ≥50% was fairly specific (68%) but with poor sensitivity (56%) for GERD.7 Another study found that a positive SI (≥50%) was fairly uncommon in a cohort of patients (n = 94) with NCCP. These patients were referred by cardiologists after a cardiac cause for their symptoms had been excluded and were divided about equally into GERD-positive (positive endoscopy or pH monitoring test) and GERD-negative groups (both test results negative). Although a slightly higher percentage of patients in the GERD-positive group had a positive SI (19%), the number of patients was small (n = 9) and the difference was not statistically significant from the percentage of patients in the GERD-negative group with a positive SI (11%; n = 5).8

Other symptom indices used to relate reflux episodes to symptoms are the symptom sensitivity index (SSI) and the symptom association probability (SAP) score. Introduced by Breumelhof and Smout, the SSI factors in the total number of reflux episodes to assign a value to a patient's sensitivity for reflux.9 Whereas the SI gives the ratio of reflux-related symptom episodes to total symptom episodes, the SSI equals the number of symptom-associated reflux episodes divided by the total number of reflux episodes, then multiplied by 100%.9 As noted, the SI can be difficult to interpret and is subject to chance in a patient with a score of 100% based on a total of only one recorded symptom episode. The SSI provides a measure of a patient's ability to sense a reflux episode (i.e. to experience symptoms during acid reflux). Therefore, a low score on both the SI and SSI indicates that a patient's symptoms are not caused by reflux, whereas a high score on both scales indicates a strong causal relationship. Patients with a low SI and a high SSI have some GERD-related symptoms, but other factors are also involved and may need to be investigated. A high SI and a low SSI may mean that the SI score is because of chance if based on a single symptom episode. In a patient with multiple symptom episodes, a high SI and a low SSI indicate GERD-related symptoms, but that the patient is unable to sense the majority of reflux episodes. The SSI score is a lower number than the SI score: in the study by Breumelhof and Smout, no patient had an SSI above 50%, and the authors deemed an SSI of ≥10% as high.9 The SAP is a statistical calculation that tries to account for times with no reflux vs. times with reflux. A pH monitoring tracing is divided into 2-min intervals, and each period is evaluated for the occurrence of reflux. Two minutes periods preceding the onset of symptom episodes are also analysed and categorized as either reflux positive or negative. A 2 × 2 contingency table is then constructed from the intervals (four fields: symptom and reflux positive, symptom and reflux negative, symptom positive but reflux negative, and symptom negative but reflux positive) and Fisher's exact test is used to calculate the probability (P) that a symptom association is not because of chance. The SAP equals 1 minus this P value, which is then multiplied by 100%.10 The SAP is obviously a much more complicated index to calculate than the SI or SSI and, as such, is more difficult for clinicians to use in patient evaluation.

Colas-Atger et al. found that the feasibility of using the SAP in a population of patients presenting with digestive gastro-oesophageal symptoms with either NERD or Savary–Miller grade 1 oesophagitis was fairly poor because of the variability of patient symptoms over time (96 of 244 patients in their study reported no symptoms during oesophageal pH monitoring) and issues with patients who had symptoms during pH testing but failed to record them accurately (38 of 148), limiting the population for which the SAP could be calculated.11 In this study, however, there was a statistically significant correlation between the SAP and the SSI with reflux (P < 0.05). Also, the correlation between these two indices allowed for a more accurate determination of oesophageal sensitivity [e.g. patients with a normal EAE but significant SAP (mean SSI = 15%) and patients with abnormal EAE and a significant SAP (mean SSI = 10%)].11

The purpose of any SI is to aid in the diagnosis of GERD and, therefore, guide treatment. One study of note by Taghavi et al. evaluated all three symptom indices against symptomatic response to omeprazole therapy as an independent measure of diagnostic accuracy.12 After assessment with oesophageal pH monitoring and calculation of symptom indices, patients (predominant symptom of heartburn; n = 38) received omeprazole 40 mg in the morning and 20 mg at night for 7 days. Symptom scores were also recorded at baseline and after treatment. By manipulating the cutoff value of the SSI (1.3 on a receiver operating characteristic curve vs. the 5% or 10% calculation often used), this index and the SAP had statistically significant associations with omeprazole test results, but both had high rates of discordance (26% and 32%, respectively). At no cutoff value were the results using the SI significantly related to the results of omeprazole treatment.12

This trial underscores many of the difficulties and controversies in this area, in particular the lack of a ‘gold standard’ of GERD diagnosis. Although it has limitations, oesophageal pH monitoring,1 combined with an SI that is easy to calculate and interpret in the clinical setting (such as the SI), provides a reasonably reliable indication of the association of symptoms with acid reflux in most patients, especially in those experiencing multiple episodes of heartburn on a regular basis but who have no endoscopically verifiable signs of mucosal injury.

Oesophageal pH monitoring in patients on antisecretory therapy

Oesophageal pH monitoring is the most useful in evaluating patients taking antisecretory therapy but who are still symptomatic. In these patients, pH monitoring is used to document the presence or absence of continued acid reflux despite treatment. Symptom association, or lack thereof with continued reflux, can be documented. A negative 24-h pH study, performed on therapy, makes it unlikely that acid reflux is a major cause of continued symptoms.

However, a negative oesophageal pH study does not completely rule out reflux as a cause of symptoms, as this technology cannot assess the potential for non-acid reflux to cause GERD symptoms. Vela et al. illustrated this using combined oesophageal pH and impedance monitoring in 12 patients with heartburn before and after omeprazole treatment.13 In the first session, patients were tested in the right lateral decubitus position for 2 h after being given a refluxogenic meal. Patients then were treated with omeprazole 20 mg b.d. for 7 days, and the testing procedure was repeated. Of the 217 reflux episodes detected in the first testing session, 45% were acidic and 55% were non-acidic. With omeprazole therapy, the percentage of acid reflux and non-acid reflux events changed significantly to 3% (P = 0.02) and 97%, respectively (P = 0.03).13 In this study, symptoms were recorded in a subset of patients. Heartburn and regurgitation were reported during acid and non-acid reflux events. Overall, a similar number of total symptom episodes were reported before and during omeprazole treatment. These findings indicate symptoms can be associated with periods of non-acid reflux, although they are more common with acid reflux.13

The likelihood of an abnormal EAE test in symptomatic GERD patients taking twice-daily PPI therapy is low. Charbel et al. conducted a retrospective review of oesophageal pH tracings in patients who experienced typical or extraoesophageal GERD symptoms while taking once- or twice-daily PPI therapy to evaluate the clinical application of the test in this patient subset.14 Patients with BE, prior fundoplication or poor compliance with treatment were excluded. From an initial pool of 2291 pH tracings, they identified 250 tests that had been conducted in patients taking PPI therapy. Of these, 119 patients had been taking once-daily PPI therapy, and 131 had received twice-daily PPI treatment. Forty-six per cent of patients complained of extraoesophageal symptoms (predominantly chest pain, cough, hoarseness, sore throat and asthma), and 54% had typical GERD symptoms (heartburn and regurgitation). Those with extraoesophageal symptoms were significantly more likely to undergo oesophageal pH testing while taking twice-daily PPI therapy [odds ratio (OR) = 3; P < 0.01]. Thirty-seven per cent of patients on once-daily PPI had abnormal studies. The overwhelming majority of patients taking PPI therapy twice daily had normal oesophageal pH tests: 93% of patients with typical symptoms and 99% of patients with extraoesophageal symptoms. The OR for a normal test in patients taking twice-daily PPIs was 11 times higher than for patients taking PPIs once daily (OR =11; P < 0.01) (Table 1).14 It is apparent that ambulatory pH monitoring in patients on twice-daily PPI therapy may not add considerably to the management of this patient group, except to reinforce the need to seek an alternate explanation for these patients’ symptoms.14 Symptom association was not assessed. The role of non-acid reflux in these patients should not be overlooked, as this study did not include impedance monitoring.

Table 1.  Results of oesophageal pH monitoring in patients with symptoms refractory to once- or twice-daily proton pump inhibitor (PPI) therapy14
PPI dose (n patients)Abnormal pH n (%)Normal pH n (%)OR (95% CI)P-value
  1. CI, confidence interval; OR, odds ratio.

Once daily (n = 119)37 (31)82 (69)
Twice daily (n = 131)5 (4)126 (96)12 (4, 25)<0.01

Although oesophageal pH testing in patients on therapy is prompted by refractory symptoms, even patients who report symptom control with antisecretory therapy may still experience pathologic levels of acid reflux. This was documented in a study by Milkes et al. in 50 patients with GERD whose symptoms were controlled with PPIs either once or twice daily.15 Although patients did not experience symptoms, 50% of them had abnormal levels of EAE. The clinical implications of these findings are unclear, but give insight to the prevention of disease progression in BE.15

Normalization of EAE

The clinical utility, or even the necessity, of normalizing EAE in patients with reflux disease is a subject for debate. This type of aggressive acid control may only be necessary in certain patient groups, such as those with BE, refractory oesophagitis and extraoesophageal symptoms.16 Results with endoscopic therapies have shown these techniques to be mediocre at best in normalizing EAE (30% of patients, for example, with the endoscopic plication system),17 with surgery perhaps the most successful technique (80% with laparoscopic procedures).18

There is a paucity of data concerning EAE normalization with therapeutic doses of antisecretory treatment. In the study by Charbel et al., normalization of oesophageal pH was seen in >90% of patients.14 Two studies with rabeprazole in patients with BE illustrate that this is a difficult goal even with high-dose therapy (Figure 3).19, 20 One study by Maqbool et al., conducted in 30 patients, investigated the effects of twice-daily doses of rabeprazole 20 and 40 mg on EAE and oesophageal bile exposure. With the 20-mg dose, 77% of patients tested normal for both measures. The 40-mg dose improved this percentage to nearly 100%.19 A second study by Sharma et al., evaluating rabeprazole 20 mg b.d. in 44 patients, produced similar results, with about 75% of patients achieving normal EAE. These results are quite good, and are as high as those seen with surgery, but required twice and four times the indicated dose for EE healing, respectively.20

Figure 3.

Normalization of oesophageal acid exposure (EAE) with high-dose rabeprazole therapy in patients with Barrett's oesophagus.19, 20

Normalization of EAE is of particular interest in patients with BE. Although PPIs have proven very useful in treating the symptoms of these patients, complete symptom control does not necessarily translate into complete oesophageal acid control. Katzka and Castell showed that, in a small population of BE patients (n = 5) reporting complete symptom relief with 20–60 mg of omeprazole daily, four patients had persistent abnormal levels of reflux.21 In a larger study, Ouatu-Lascar and Triadafilopoulos examined the effect of relatively low doses (15–30 mg) of lansoprazole in 30 patients with BE.22 In this study, normal EAE was defined as pH <4 for 8%, 3% and 6% of the upright, supine and total time periods, respectively. Symptomatic remission was achieved with lansoprazole 15 mg daily in 12 patients and with 30 mg daily in the other 18. However, overall, 40% of these patients continued to exhibit pathologic, bipositional acid reflux of moderate severity.22 Finally, Fass et al. tested a short course (mean 8 days) of very high doses of omeprazole (40 mg b.d.) in 25 BE patients. The results from this study were better: 16% of patients post-therapy had an overall abnormal pH test, with an additional 2% having abnormal supine reflux.23 These results indicate that persistent abnormal EAE in patients with BE is not uncommon, even in those treated with very high doses of acid suppressive therapy or who report complete symptom relief. The question remains whether aggressive acid control is good for patients, especially BE patients, or whether a PPI dose that provides symptom relief is adequate treatment.

Intragastric pH monitoring

An often-cited study linking the efficacy of control of acid secretion and healing of EE is that of Bell et al., in which a correlation was shown between the duration of gastric acid suppression and EE healing (P < 0.05; r = 0.87).24 Further study is needed to put these findings into better perspective, but, as with oesophageal pH testing, intragastric pH monitoring can be used to provide clinicians with a marker against which to judge an agent's potential (or relative potential) efficacy in GERD treatment. As such, control of intragastric pH is a valid endpoint in head-to-head comparisons of acid suppressive therapy, and numerous studies have been conducted investigating this measure among the available PPIs.

Of note are two crossover studies of omeprazole, lansoprazole, pantoprazole and rabeprazole, and a five-way crossover trial that included esomeprazole. In the first study, Tutuian et al. conducted intragastric pH monitoring in a population of healthy volunteers (n = 26) on day 7 of therapy with omeprazole 20 mg (n = 12), lansoprazole 30 mg (n = 24), pantoprazole 40 mg (n = 26) and rabeprazole 20 mg (n = 6). No statistically significant difference was found between any treatment for per cent time with intragastric pH >4 for the daytime, night-time or total 24-h period.25 Miner et al. conducted a five-way crossover trial with the same doses of the same four PPIs with the addition of esomeprazole 40 mg.26 Thirty-four patients with GERD symptoms received each PPI 30 min before a standardized breakfast for 5 days (10-day washout between treatments), and intragastric pH monitoring was conducted on day 5. In this trial, esomeprazole was statistically superior to the other agents in mean hours with intragastric pH >4 (P ≤ 0.001).26 Finally, Pantoflickova et al. investigated the effect on intragastric pH of single doses of omeprazole capsule 20 mg, omeprazole multiple unit pellet system tablet 20 mg, lansoprazole 30 mg, pantoprazole 40 mg and rabeprazole 20 mg in 18 healthy subjects in a crossover fashion (14- to 28-day washout period). Overall intragastric pH and time with pH >4 were significantly greater with rabeprazole (P ≤ 0.04).27 In general, the data from the Tutuian and Miner studies appear to support clinical studies of the relative efficacy of PPIs in EE treatment, with equivalent rates among omeprazole, lansoprazole and rabeprazole,25 and slightly higher healing rates with esomeprazole.28, 29 As well, the results from the Robinson study appear to support the rapid symptom relief seen with rabeprazole.30

However, differences in intragastric pH do not guarantee differences in treatment efficacy in individual patients. Intersubject variability in control of intragastric pH has been observed with PPIs and illustrates another important application of this measurement. A crossover study of 20 healthy volunteers given twice daily omeprazole 20 mg and lansoprazole 30 mg for 7 days found a wide intersubject and intrasubject variability in per cent of time with pH <4 (Figure 4).31 Intrasubject variability appears to be uncommon, but may explain the occasional patient who responds when PPI therapy is switched.16

Figure 4.

Intersubject variability in control of gastric pH with omeprazole 20 mg and lansoprazole 30 mg b.d.31

Intragastric pH monitoring is also useful to assess dose timing and food effect and evaluate alternative dosing strategies. Hatlebakk et al. showed that the median percentage of time with gastric pH <4 was lower (17%) when a PPI (in this case, omeprazole or lansoprazole) was taken with breakfast compared with that seen when the PPI was taken without a meal (42%; P = 0.01).32 This same group of investigators evaluated the efficacy of three dosing regimens for controlling nocturnal gastric acid breakthrough, defined as night-time periods of gastric pH <4 for ≥1 h.33 Healthy subjects (n = 18) received 7 days of omeprazole 40 mg taken before breakfast or before dinner or as a split dose (20 mg before breakfast and 20 mg before dinner) in a crossover fashion. Nocturnal acid breakthrough was significantly more common with the single morning dose compared with the other two regimens (P < 0.05), and the percentage of time with gastric pH <4 in the night-time hours was significantly lower (P < 0.02).33 In another crossover study, 20 patients with GERD were given rabeprazole 20 mg once daily in the morning or evening for 7 days, with a 7-day washout between treatments.34 Combined gastric and oesophageal pH monitoring was conducted at baseline and on day 7. Not only did the evening regimen provide significantly better control of nocturnal recumbent reflux, but it was also more effective in normalizing EAE overall.34

Finally, intragastric pH monitoring can identify patients who may require higher doses of PPIs to control acid secretion. In a study by Leite et al., 88 GERD patients experienced persistent symptoms, despite treatment with omeprazole 20 mg b.d., and 19% (n =17) were found to have inadequate pH control (defined as a pH <4 for >50% of a 24-h period).35 This group of ‘omeprazole failures’ were compared with 19 other patients from the original cohort and 19 healthy controls who received placebo and omeprazole 20 mg b.d. The gastric pH profiles of the patients with GERD were similar to those of the healthy volunteers when taking omeprazole. However, the profiles of the omeprazole failure group most closely resembled those of the control group when taking placebo. Median time with gastric pH <4 significantly decreased in six of seven patients in the omeprazole failure group when the dosage level was increased to 80 mg daily (P < 0.02).35 Although uncommon, in some patients with PPI refractory symptoms, standard doses of PPI therapy are not adequate to control gastric pH, and these patients may respond to an increased dose.


Oesophageal pH monitoring can assess the presence, severity and pattern of acid reflux in the oesophagus. Combined with an SI, oesophageal pH monitoring provides clinicians with an indication of the association of symptoms with acid reflux, especially in patients with multiple, regular heartburn episodes. It is of particular utility as a diagnostic tool in patients with no endoscopic evidence of disease, in documenting continued acid reflux and/or association of symptoms with reflux in patients with refractory symptoms on therapy, or in verifying acid suppression in these patients. A negative oesophageal pH study does not rule out reflux as a cause of symptoms, however. Day-to-day variability in EAE is great, necessitating a large sample size and crossover study design to assess antisecretory drug effects on EAE. Normalization of EAE in patients with reflux disease is a subject for debate, and questions remain as to whether or not oesophageal pH monitoring should be used instead of, or in combination with, symptomatic remission to assess therapeutic endpoints in BE patients as a way of managing the disease.

Because GERD is a disease of the oesophagus, one may argue that intragastric pH monitoring is of limited use in GERD diagnosis and management. Why should clinicians investigate stomach conditions when their ultimate area of concern is the injurious effect of EAE and correlating oesophageal pH with disease manifestations and symptoms? Part of the answer to this question concerns the current paradigm for GERD treatment, the mainstay of which is antisecretory therapy. These agents’ mechanism of action is gastric acid suppression. Intragastric pH monitoring, therefore, is the appropriate index for measuring their efficacy in suppressing acid, just as investigating lower oesophageal sphincter (LES) pressure would be an appropriate measure for an agent or technique that treats acid reflux by acting on the LES. Intragastric pH monitoring also provides a marker against which to measure the potential of an antisecretory agent to effectively treat GERD.

One may also question the importance of detecting the presence of abnormal acid levels in the stomach in patients with symptoms refractory to PPI therapy who have normal EAE. As the study by Charbel et al. illustrated, the likelihood of an abnormal oesophageal pH test for symptomatic GERD patients taking twice-daily PPI therapy is small.14 The aetiology of GERD symptoms is complex, and researchers are beginning to recognize that EAE is only one component in symptom development. Oesophageal sensitivity in individual patients and the connection between symptoms and non-acid reflux13 underscore the shortcomings of not only gastric pH investigations, but also oesophageal pH monitoring in the clinical management of GERD.

Gastric pH monitoring is useful in assessing dose timing and food effect and to evaluate alternative dosing strategies, particularly for the control of nocturnal acid breakthrough on therapy. Although differences in control of intragastric pH in head-to-head comparisons of PPIs reveal the relative therapeutic potentials of these agents, these differences do not guarantee differences in GERD treatment efficacy in individual patients, and wide intersubject variability in control of intragastric pH with PPIs has been documented. Finally, some patients with refractory symptoms may respond to an increased PPI dose. The lesson here is that one cannot presume control of intragastric pH with a fixed dose of a PPI—any PPI.