Impedance‐pH monitoring in proton pump inhibitor resistant patients: ready for clinical application?

Short abstract

Impedance‐pH monitoring represents a useful tool for studies and clinical research in oesophageal disorders, and may hold promise for the group of patients dissatisfied with their antireflux treatment

Gastro‐oesophageal reflux disease (GORD) is a very common disorder in which the reflux of gastric (or gastroduodenal) contents causes troublesome symptoms and/or lesions of the oesophageal mucosa. The harmful effects of acid injury on the oesophageal mucosa have been established for decades by experimental and clinical studies.¹ From a clinical and therapeutic viewpoint, the central role of acid in the pathogenesis of reflux symptoms and lesions of the oesophagitis was most convincingly confirmed by the introduction of acid suppressive therapy; the development of proton pump inhibitors (PPIs) achieved a level of efficacy not previously reached by antacids or H₂ blockers.²

Despite the progress made with PPIs, it has now however become clear that up to 40% of patients with GORD symptoms are either not completely satisfied with their acid suppressive therapy or may be resistant to PPIs.³ Several reasons may underlie an individual's resistance to his or her PPI regimen. Firstly, it may be that symptoms are not related to acid reflux; this may be the case when atypical and non‐specific symptoms are reported and endoscopy does not show mucosal breaks, for example. In addition, acid injury may not be completely eliminated by PPI treatment and acid exposure remains abnormal. The level of acidity necessary to cause symptoms is not precisely defined and is likely to vary from subject to subject or even in the same individual. Several studies have shown that “weakly acidic” or short lasting reflux episodes may cause symptoms similar to those reported in classical GORD despite an overall normal 24 hour oesophageal acid exposure time.⁴,⁵ In these patients with an apparently hypersensitive oesophagus, an increased dose of PPI (for example, omeprazole twice daily) proved to be effective in the subgroup of individuals with a significant association between symptoms and acid reflux episodes.⁶ In practice, demonstration of such a temporal relationship requires analysis of symptoms and acid events recorded during 24 hour oesophageal pH monitoring. Different symptom indices and probability calculations are commonly used to express the strength of this association between acid reflux and symptomatic events.⁷,⁸,⁹

However, despite the significant role of acid in GORD pathogenesis, other components of the reflux material can also exert a deleterious effect on the oesophageal mucosa; they can act either alone or synergistically with acid. These molecules include biliary acids and enzymes such as trypsin, normally present in the duodenum, which can reflux into the stomach and then the oesophagus (duodenogastro‐oesophageal reflux (DGOR)).¹ Non‐chemical stimuli such as mechanical distension of the oesophagus by the reflux bolus can also trigger symptoms, which are likely to include heartburn and regurgitation. Finally, demonstration of the role of reflux in the pathogenesis of symptoms is frequently a difficult issue. Although 24 hour pH monitoring in patients off therapy may provide some clues, the sensitivity of this technique is relatively poor in non‐erosive GORD, and probably does not exceed 60–70% at best.¹⁰ With respect to pH monitoring, the problem is further complicated in patients who continue to report symptoms while receiving acid suppression; the classical threshold of acid exposure (that is, per cent time pH less than 4) may be too stringent for the detection of abnormalities when monitoring patients who continue to receive PPI therapy.¹¹ In fact, an oesophageal pH less than 4 cannot be considered a reliable marker for the detection of reflux episodes in a patient on PPI therapy or in those with spontaneously low gastric acidity, such as frequently fed infants.

Addition of the detection of bilirubin (used as a marker of DGOR) to pH monitoring has been shown to be useful in patients poorly responsive to a once daily dose of a standard PPI.¹² This method is however hampered by the special monitoring conditions required (liquid diet) and limited access to the technique in clinical practice. In summary, neither standard pH monitoring nor bilirubin monitoring can satisfactorily address the situation of the so‐called “non‐acid reflux”, especially in patients with persistent symptoms on PPI therapy.

What really is “non‐acid reflux”? No definition is universally accepted and this terminology actually regroups a number of different conditions with no well defined corresponding clinical entities.¹³ New concepts have recently emerged with the development of a novel technology combining impedance and pH measurement. Impedance is a parameter which describes resistance to the flow of an electric current between adjacent pairs of electrodes mounted on an oesophageal catheter; when adjacent electrodes are linked by a liquid bolus, conductivity is high and therefore impedance is low. In contrast, when electrodes are separated by gas, such as during a belch, conductivity is low and impedance increases. In recent years several groups have applied this technique to the detection of reflux episodes in both normal volunteers and reflux patients.¹⁴,¹⁵,¹⁶ These studies have provided convincing evidence that impedance‐pH monitoring can accurately detect and characterise nearly all reflux episodes. Following this technical progress, a tentative classification of non‐acid reflux has been proposed by a panel of experts.¹⁷ According to this technology based description system, the term non‐acid reflux includes both “weakly acidic reflux” (reflux events that result in an oesophageal pH between 4 and 7, or 6.5 in reference to physiological neutrality) and “weakly alkaline reflux” (reflux detected by a change in impedance level but with pH remaining above the oesophageal value of 7 or 6.5).

In this issue of Gut, Mainie and colleagues¹⁸ report on the clinical application of impedance‐pH monitoring to patients with symptoms that persist despite acid suppressive therapy (see page 1398). This large series of 168 patients was recruited on the basis of therapeutic failure after at least one month of a twice daily PPI regimen. Interestingly, persistent symptoms were categorised as typical (heartburn, regurgitation, or chest pain) or atypical (for example, cough, hoarseness, abdominal discomfort). The primary endpoint of the study was the symptom index (SI), defined as the number of symptoms associated with reflux episodes divided by the total number of symptomatic events. Symptom association probability (SAP) was also determined according to previously described methodology.⁸ Surprisingly, the time windows used to determine whether or not a symptom was reflux related differed for SI (reflux within five minutes prior to symptom occurrence) and SAP (two minute segments with and without reflux). The most original aspect of the work, for which the authors should be complimented, relies on the fact that impedance‐pH was monitored in patients on therapy. This experimental design allowed more appropriate analysis of the mechanisms of symptoms occurring during the average 20 hour recording period. As expected,¹¹ very few patients (11%) continued to have abnormal acid exposure while receiving a PPI twice daily and only 17% of the reflux episodes detected by impedance were acid reflux (with pH below 4). In contrast, more than 80% of reflux episodes were non‐acid, as defined previously, the majority of which (90%) being categorised as weakly acidic and rarely, weakly alkaline (10%). It is of note that of the 144 patients who reported symptoms during the monitoring period, nearly half had a negative SI, suggesting a mechanism other than reflux in the pathogenesis of their symptoms. For those with a positive SI, non‐acid reflux was the predominant mechanism underlying residual symptoms. Finally, typical GORD symptoms in patients on PPI therapy were more likely to be associated with a positive SI (primarily non‐acid) compared with atypical symptoms which were more frequently associated with the absence of reflux.

The results of the study by Mainie and colleagues¹⁸ are consistent with other studies which have already assessed the additional yield of oesophageal impedance monitoring in patients on and off therapy.¹⁴,¹⁹,²⁰ It has been confirmed that standard 24 hour pH monitoring alone is of limited value in understanding the mechanism of PPI resistance as the large majority of these patients have normal or even low acid exposure values while receiving PPI therapy.¹¹ It is of note that Mainie and colleagues¹⁸ used especially low thresholds to define abnormal acid exposure times in PPI receiving patients. While the addition of impedance to pH monitoring seems of relatively limited clinical value in patients not receiving therapy,¹⁹,²⁰ the diagnostic yield of the combined measurement seems to be substantially improved in treated patients.¹⁸,¹⁹ The role of weakly acidic reflux episodes in symptom pathogenesis has now been established beyond doubt, both in typical (for example, regurgitation on therapy) and atypical GORD; for example, an association with chronic cough has now been reported in patients either off²¹ or on therapy.¹⁹ However, as for pH monitoring alone, controversies remain regarding the best method of expression of the strength of association between symptoms and reflux episodes; this issue seems particularly relevant as the aim is to best predict therapeutic response to acid suppression.²² It is indeed again surprising to see that the correlation between the SAP and SI results in the Mainie and colleagues study¹⁸ was so poor, a result that is however consistent with our own experiences.¹⁹

The choice of SI rather than SAP, as well as the time window for symptom analysis (two or five minutes) remains largely based on personal and arbitrary choices; this may be an area of research that deserves further attention. Standardisation of both the technique and parameters and validation of software required for reducing the analysis time are first necessary to further increase clinical application. Finally, the exact role of pH‐impedance monitoring will depend on its diagnostic relevance and ability to influence the management of GORD patients, particularly those with predominant extraoesophageal symptoms. Data from Mainie and colleagues¹⁸ and other groups concerning the outcome of surgery in PPI resistant patients that are carefully selected on the basis of objective impedance‐pH analysis is eagerly awaited. Application of this recent technology will be particularly relevant to the pharmacological assessment of new drugs aimed at reduction of transient lower oesophageal relaxations, such as GABA‐b agonists and glutamate ligands.²³,²⁴,²⁵

In conclusion, at its present stage of development, impedance‐pH monitoring represents a useful tool for studies and clinical research in oesophageal disorders. It is probably too early for it to be considered as ready for routine clinical application but recent studies hold promise, particularly for the group of patients dissatisfied with their antireflux treatment. Further studies should more directly measure the impact of results on patient management.