Kandulski A., Weigt J., Caro C., Jechorek D., Wex T., Malfertheiner P. Esophageal intraluminal baseline impedance differentiates gastroesophageal reflux disease from functional heartburn. Clin Gastroenterol Hepatol 2015; 13: 1075–1081.
Esophageal intraluminal baseline impedance differentiates gastroesophageal reflux disease from functional heartburn
Arne Kandulski1, Jochen Weigt1, Carlos Caro1, Doerthe Jechorek2, Thomas Wex1,3 and Peter Malfertheiner1
BACKGROUND & AIMS: Mucosal integrity can be assessed in patients with gastroesophageal reflux disease (GERD) by measuring intraluminal baseline impedance (BI). However, it is not clear whether BI is abnormal in patients with functional heartburn (FH), or can be used to distinguish them from patients with GERD. We compared differences in BI between patients with FH vs GERD.
METHODS: We performed a prospective study of 52 patients (16 men; mean age, 55 y; range, 23–78 y) seen at a tertiary university hospital from February 2009 through December 2012. Thirty-five patients had GERD (19 had nonerosive reflux disease [NERD], 16 had erosive reflux disease [ERD]) and 17 had FH. All patients discontinued proton pump inhibitor therapy and then underwent esophagogastroduodenoscopy and multichannel intraluminal impedance and pH monitoring. BI was assessed at 3, 5, 7, 9, 15, and 17 cm proximal to the lower esophageal sphincter in recumbent patients. Biopsy specimens were taken from 3 cm above the gastroesophageal junction; histology analysis was performed to identify and semiquantitatively score (scale, 0–3) dilated intercellular spaces.
RESULTS: Baseline impedance in the distal esophagus was significantly lower in patients with NERD or erosive reflux disease (ERD) than FH (P = .0006). At a cut-off value of less than 2100 Ω, BI measurements identified patients with GERD with 78% sensitivity and 71% specificity, with positive and negative predictive values of 75%. Also in the proximal esophagus, reduced levels of BI levels were found only in patients with ERD. There were negative correlations between level of BI and acid exposure time (r =-0.45; P = .0008), number of acidic reflux episodes (r=-0.45; P =.001), and proximal extent (r = -0.40; P = .004). Biopsy specimens from patients with NERD or ERD had significant increases in dilation of intercellular spaces, compared with those from patients with FH; there was an inverse association between dilated intercellular spaces and BI in the distal esophagus (r = -0.28; P = .06).
CONCLUSIONS: Measurement of BI in the lower esophagus can differentiate patients with ERD or NERD from patients with FH (78% sensitivity and 71% specificity), and therefore should be considered as a diagnostic tool for patients with proton pump inhibitor–refractory reflux. Low levels of BI are associated with increased exposure to acid and dilation of intercellular spaces, indicating that BI is a marker of mucosal integrity.
Keywords: MII-pH; Esophageal Mucosa; Acid-Suppressive Therapy; Diagnosis.
Abbreviations used in this paper: AET, acid exposure time; BI, baseline impedance; CI, confidence interval; DIS, dilated intercellular spaces; ERD, erosive reflux disease; FH, functional heartburn; GERD, gastroesophageal reflux disease; LES, lower esophageal sphincter; MII-pH, impedance pH monitoring; NERD, nonerosive reflux disease; PPI, proton pump inhibitor; SAP, symptom association probability; SI, symptom index.
© 2015 by the AGA Institute 1542-3565/ ht tp://dx.doi.org/10.1016/j.cgh.2014.11.033
The esophageal squamous epithelium is a tight protective barrier against luminal components. Disruption of this epithelial defense is a common phenomenon in gastroesophageal reflux disease (GERD), even in the absence of lesions visible at endoscopy (nonerosive reflux disease [NERD]). Microscopic alterations and dilation of cell–cell contacts usually are found in GERD and are associated with impaired mucosal integrity.1–3 In addition, in NERD, altered microscopic architecture with dilated intercellular spaces has been linked to impaired transepithelial permeability in several functional studies.4 Besides the measurements of transepithelial electrical resistance and permeability in Ussing chambers, impaired mucosal integrity has been associated with in vivo assessment of esophageal baseline impedance (BI).5,6 In a rabbit model, perfusion with acidified solution was found to reduce BI levels, which persisted even beyond the end of perfusion. Ex vivo, BI levels correlated with transepithelial electrical resistance in Ussing chambers and with dilated intercellular spaces.5 Patients with GERD have lower BI impedance levels compared with asymptomatic controls as well as with symptomatic patients with normal acid exposure of the distal esophagus. Furthermore, these changes have normalized with increasing BI levels after acidsuppressive therapy.6
In clinical practice, NERD is the most frequently diagnosed entity of GERD but poses a diagnostic challenge to conditions that are not GERD related (ie, functional heartburn [FH]).7 In case of a normal pHimpedance analysis without symptom association, the diagnosis most likely is FH with no or only a weak response to acid-suppressive therapy.8 Several studies have addressed the assessment of morphologic changes in esophageal mucosa and were able to distinguish NERD from FH by using transmission electron microscopy as well as standard histopathologic evaluation.9–11
To date, functional investigations to assess intramucosal BI levels showing impaired mucosal integrity with respect to histomorphologic alterations have not been performed to differentiate FH from GERD. The aim of our study was to assess BI levels in patients with FH and to differentiate them from GERD. We further aimed to evaluate histomorphologic alterations such as dilated intercellular spaces (DIS) to correlate with BI levels as a parameter of mucosal electrical conductivity and integrity.
MethodsStudy Subjects and Study Protocol
Fifty-two consecutive patients (16 men, 36 women; Q10 age, 55 y; [23–78]) were referred to our outpatient department and functional gastrointestinal laboratory and investigated for typical reflux symptoms (heartburn and acid regurgitation). In this prospective study we enrolled 17 patients with FH (age, 53.8 y; [23–78 y]). By definition, these patients suffered from proton pump inhibitor (PPI)-refractory heartburn with less than 50% symptom improvement and a past medical history of a PPI double standard dose for at least 6 weeks. Diagnostic criteria for FH were a normal endoscopic appearance of the gastroesophageal junction in combination with normal acid exposure time without any symptom association (negative symptom index [SI] and symptom association probability [SAP]) (see later). In addition, 16 patients with erosive reflux disease (ERD) (age, 53.8 y; [23–78 y]) and 19 patients with NERD (age, 64.9 y; [56–72 y]), including patients with esophageal hypersensitivity, were investigated.
All patients were interviewed and clinically characterized before planning further diagnostic steps. The patients were asked to taper and stop potential acid-suppressive medication for at least 3 weeks before endoscopy and impedance pH monitoring (MII-pH) to minimize effects of potential acid hypersecretion on BI levels and histology. Symptoms were recorded using the validated reflux disease questionnaire translated into German,19 and all patients were scheduled to be investigated endoscopically and by MII-pH monitoring on the same day.
The study protocol was performed according to the Declaration of Helsinki and approved by the local ethical committee. Eligible patients (>18 y) were included after providing informed consent. None of the patients had an esophagogastroduodenoscopy or functional diagnostics previously, but all patients had heartburn as a typical GERD symptom based on the Montreal classification.12 Previous upper gastrointestinal surgery, alarm symptoms, gastric or duodenal ulcer disease, Barrett’s esophagus, or esophageal motility disorders were considered exclusion criteria.
Upper Gastrointestinal Endoscopy and Esophageal Biopsy Specimens
After an overnight fast, all patients underwent an esophagogastroduodenoscopy under intravenous conscious sedation using midazolam (Dormicum V 5 mg/ mL; Roche Deutschland Holding GmbH, Penzberg, Germany) and/or 1% propofol (Propofol-Lipuro 10 mg/mL; Braun Melsungen AG, Melsungen, Germany) with a standard videogastroscope (GIFQ180; Olympus Optical Europe, Hamburg, Germany).
Endoscopic esophageal landmarks were defined as the gastroesophageal junction, with the beginning of the gastric folds and the Z-line as the squamocolumnar junction and diaphragmatic pinch. In the distal esophagus, 2 esophageal biopsy specimens were taken from 3 to 5 cm above the gastroesophageal junction, not including visible changes (no erosions), and immediately transferred to 4% neutral-buffered formalin for later embedding in paraffin.
Combined 24-Hour Impedance pH Monitoring and Assessment of Intraluminal Baseline Impedance Levels
After endoscopy, the MII-pH catheter (Sandhill Scientific, Highland Ranch, CO) was inserted and located with esophageal pH electrodes 5 cm above the gastroesophageal junction (lower esophageal sphincter [LES]). Manometry was not performed in all patients to localize the LES. In a subset of patients, localization of the LES and placement of the MII-pH catheter was performed after endoscopy by using a transversion factor of 4 cm,13 noting that there is a difference of 0.9 cm between the endoscopically determined squamocolumnar junction and manometrically detected LES.14
Esophageal impedance electrodes were located at 3, 5, 7, 9, 15, and 17 cm above the LES. Patients were asked to eat 3 meals and beverages at fixed times and not to lie down in the supine position during the day. Event markers were set for meal times, body posture, and the occurrence of specific symptoms. Manual analysis of the tracings was performed by 2 experienced operators (A.K. and C.C.) independently. In case of disagreement, the case was discussed with a third experienced investigator (J.W.).
Reflux episodes were defined as a decrease of more than 50% from baseline impedance moving from the distal to the proximal extend. They were characterized as liquid, mixed, or gas, and by the pH of the refluxate. Acid exposure time (AET) was defined as abnormal when a pH less than 4 was measured for more than 4.2% of the time over 24 hours. SAP and SI were assessed as previously described for acid, weakly acidic, and weakly alkaline reflux events.15,16 NERD was diagnosed if no erosions were visible endoscopically but AET was increased (>4.2%). Patients with a normal AET and a positive SI or SAP have a hypersensitive esophagus and were considered to have a diagnosis of NERD according to the Rome III criteria.6 A normal endoscopic appearance of the gastroesophageal junction in combination with a normal AET without any symptom association (negative SI and SAP) was considered FH.
BI levels were assessed at 3, 5, 7, 9, 15, and 17 cm above the lower esophageal sphincter. Swallows and reflux-induced changes were excluded. In the recumbent position at night, we identified a period with a stable and constant BI signal without any severe changes or interference over time. During these periods the BI levels were analyzed for at least 30 minutes.
Esophageal tissue specimens were embedded in paraffin and submitted for histopathologic examination with H&E and periodic acid–Schiff staining. The degree of basal cell hyperplasia, presence of papillary elongation, and dilated intercellular spaces was assessed10,23,24 and semiquantitatively scored as either 0 (absent), 1 (mild), 2 (moderate), or 3 (severe), as described previously.10,17 The expert pathologist (D.J.) was blinded to endoscopic data and results from MII-pH.
Data Collection and Statistical Analysis
All collected data were entered in an Excel sheet (Microsoft Corporation, Redmond, WA) and statistically analyzed using GraphPad Prism software (GraphPad Software, La Jolla, CA). Data are expressed as mean ± SE or 95% confidence intervals (CIs), if not stated otherwise. One-way analysis of variance was applied for comparisons among the groups (FH, NERD, and ERD). If significant differences were identified, Bonferroni analysis for multiple testing for post hoc analysis was performed to calculate differences between the histopathologic items. Correlation analysis was performed using the Pearson correlation test. All tests were 2-sided with a P level less than .05 indicating significance.
We performed a receiver operating characteristic analysis for distal baseline impedance at 3 cm above the LES. Sensitivity, specificity, and predictive values for the discrimination between NERD and FH were calculated with a cut-off value of less than 2100 Ω using the Fisher exact t test.
ResultsTwenty-Four–Hour Impedance pH Monitoring and Patient Characteristics
According to the reflux disease questionnaire, there were no differences in symptom severity between the groups for either heartburn, regurgitation, or dyspepsia.
Results from MII-pH monitoring are shown in Table 1, which summarizes the most important parameters in the distinction between ERD, NERD, and FH. Within NERD patients, 10 patients had no pathologic AET but a positive symptom association (esophageal hypersensitivity).
Table 1. Comparison of Symptoms and 24-Hour MII-pH in Patients With FH and Patients With GERD
Intraluminal Baseline Impedance Analysis in the Distal and Proximal Esophagus
BI levels in the distal esophagus at 3 and 5 cm above the LES differed significantly between FH and GERD (Figure 1). At 3 cm, BI levels of ERD (994.0 ± 182.2 Ω) and NERD (1558 ± 362.3 Ω) were significantly lower than in patients with FH (2884 ± 364.8 Ω) (P = .0006). Similar measurements were obtained at 5 cm above the LES (Table 2), but not for the more proximal impedance electrodes of the catheter.
In the proximal esophagus at 15 and 17 cm above the LES, BI levels of only patients with ERD were significantly lower (P = .02–.0003). No differences in BI levels were obtained between NERD and FH (Table 2).
Table 2. Comparison of BI Levels in Patients With FH, NERD, and ERD at 3, 5, 15, and 17 cm Above the LES
Correlation analyses of BI levels in the distal esophagus with functional parameters from MII-pH measurements showed significant associations of low BI levels with parameters of acidic reflux. Exemplary displayed for BI at 3 cm, we found a negative correlation with AET (r = -0.45; P = .008) and esophageal acid percentage time defined by MII (r = -0.45; P = .001), with the numbers of acidic reflux episodes (r = -0.45; P = .001) and numbers of proximal reflux episodes (r = -0.4; P = .003) (Figure 2).
Figure 2. Negative association of distal baseline impedance levels at 3 cm (U) and (A) acid exposure time (r = -0.45; P = .008), (B) acidic reflux episodes (r = -0.45; P = .001), and (C) numbers of proximal reflux episodes (r = -0.4; P = .003).
For discrimination between NERD and FH, we performed a receiver-operated model with an area under the receiver operating characteristic curve of 0.73 ± 0.09 (95% CI, 0.55–0.91; P = .01). With a cut-off value of less than 2100 Ω for discriminating NERD from FH, sensitivity was 0.79 (95% CI, 0.54–0.94) and specificity was 0.71 (95% CI, 0.44–0.90); with a positive negative predictive value of 0.75 (95% CI, 0.51–0.91) as well as a negative predictive value 0.75 (95% CI, 0.48–0.93) (Table 3).
Table 3. Discrimination of Patients With NERD From FH by Assessment of Distal Baseline Impedance Levels at 3 cm Above LES
Histopathologic Evaluation of Dilated Intercellular Spaces Correlating With Baseline Impedance
Histologically, patients with FH had less dilated intercellular spaces than NERD and ERD (Figure 3A). Furthermore, low BI levels correlated with DIS evaluated with standard light microscopy that marginally missed statistical significance (r = -0.28; P = - .06) (Figure 3B).
DiscussionAssessment of BI levels allows us to distinguish patients with FH from patients with GERD. Our findings confirm previous observations of low BI levels in patients with GERD including NERD,6,18,19 but there are very few publications regarding basal impedance in FH. Vaezi et al developed an impedance probe that determines mucosal baseline impedance by direct contact of the probe with the mucosa during endoscopy. They found higher impedance levels in patients with a normal endoscopic appearance and normal acid exposure time during pH-metry.20 Placed under endoscopic guidance, direct mucosal impedance is likely to be comparable with measurement of BI levels as conducted in our study. However, the distance between the electrodes and the size of the electrical field certainly has an impact on the measurement. The size of the electrical field is dependent on the distance between the electrodes. With a larger distance between impedance electrodes on the conventional MII-pH catheter, the electrical field of BI measurements might be enlarged and therefore often is speculated to measure impedance signals that are influenced by deeper layers of the esophageal wall. However, based on structural components of the mucosa (ie, tight junctions) forming a tight barrier for electrodes, the mucosal contribution to BI levels probably is most important for the impedance signal. Both methods have their specific limitations (ie, localization, pressure, and time of contact), but direct comparisons of both methods have not been performed.
Recently, Martinucci et al21 described higher BI levels in patients with characteristics of FH as well. Impaired BI levels in the distal esophagus have been associated with parameters of acidic reflux in MII-pH in our study. Again, this was similarly reported by other groups for patients with GERD.6,22–24 These publications clearly showed a more significant decrease of BI levels associated with more moderate to severe reflux episodes and also with more severe forms of esophagitis shown by the direct impedance technique.20
Similar to the BI levels in our study, Woodland et al22 described comparable low BI levels in the distal esophagus of patients with NERD (1669 ± 182 Ω), but not with FH (2384 ± 211 Ω). In addition, Woodland et al22 described slower impedance recovery rates in patients with NERD when compared with FH.
Although our study cohort included a rather small sample size, we calculated a cut-off level of less than 2100 Ω for differentiation between NERD and FH with a sensitivity and specificity of 78% and 71%, respectively. Including all patients with GERD (NERD as well as ERD), the sensitivity and specificity were 83% and 71%, respectively (data not shown). Therefore, analysis of the BI level should be considered a useful additional parameter during MII-pH analysis in the differential diagnosis of NERD and FH.
Morphologically, DIS was associated with ERD as well as NERD, but not with FH. Direct assessment of impaired mucosal integrity includes measurement of increased paracellular permeability, which also is associated with the presence of DIS,3,25 as well as impaired mucosal integrity, which was linked functionally to reduced BI levels and was induced by acidic perfusion in a rabbit model and in healthy volunteers.5 PPI therapy increases baseline impedance levels as well as microstructural changes of dilated intercellular spaces, a further indication for acidic-induced abnormalities.6 Our study also shows the differences of BI levels in the proximal esophagus. This was significant for patients with ERD when compared with FH. Structural changes of the esophageal mucosa also can be induced by infusion with acidic solutions even in the proximal esophagus.5,26,27 Because measurements were assessed at 15 and 17 cm above the LES these abnormalities were unlikely to be associated with mucosal erosions.
Recently, in patients with GERD, low baseline impedance values were described to be associated with dilated intercellular spaces by light microscopy and with the expression of claudin-1 and occludin.28 Both proteins are involved in the structural integrity of tight junctions and were found to be increased in patients with GERD.17,29
What actually causes changes of BI levels needs to be investigated further. Besides structural changes with increased paracellular conductivity, inflammatory changes also need to be considered. We did not find an association with histopathologic assessment of inflammatory cell infiltration, although inflammation-related mucosal edema needs to be discussed as a possible explanation for altered baseline impedance as well. We further showed a significant association of low BI with parameters of acidic reflux and with dilated intercellular spaces in light microscopy.
Impaired mucosal integrity assessed by BI levels is a characteristic finding in patients with GERD, but not in patients with FH. These results provide us with a greater insight into the pathophysiological differences between the entities. Although the sole coincidence of refluxassociated mucosal abnormalities needs to be considered, induced paracellular conductivity in the presence of DIS to date is the most likely explanation for low BI levels in patients with GERD.
In conclusion, we found that patients with FH can be distinguished from patients with GERD, in particular NERD, based on baseline impedance levels. Based on the test parameters calculated in our study, the measurement of BI as a single parameter does not allow us to adequately differentiate patients with FH from patients with GERD. The increasing number of PPI-refractory patients has lead to growing medicoeconomic efforts and costs, and thus accurate differential diagnosis is crucial. Measurement of BI levels is therefore a complementary tool in addition to endoscopy, conventional reflux monitoring, and esophageal biopsies. In difficult cases all methods should be considered for an accurate diagnosis to guide further adequate therapeutic management.
Conflicts of interest.The authors disclose no conflicts.
Назад в раздел
Популярно о болезнях ЖКТ читайте в разделе "Пациентам"
Информация на сайте www.gastroscan.ru предназначена для образовательных и научных целей. Условия использования.