Poprawa efektywności diagnostycznej pomiaru pH w przełyku przy wykorzystaniu wielokanałowego impedancyjnego pH-metru przy pomocy oceny impedancji podstawowej oraz porefluksowych fal perystaltycznych indukowanych przełknięciem

Sergii G. Melashchenko, Oleh O. Ksenchyn, Viacheslav M. Chernobroviy

Department of family and internal medicine, National Pirogov Memorial Medical University, Vinnytsia, Ukraine

ABSTRACT

Introduction: The postreflux swallow-induced peristaltic wave (PSPW) index and methods of baseline impedance estimation are novel parameters proposed to improve diagnostic yield of intraluminal impedance-pH monitoring (MII-pH-M). Measuring intraluminal baseline impedance level (BIL) reflects mucosal integrity of distal esophagus. We offer original approach with automatic generating a histogram of impedance and visual identification of peak (mode) corresponding to BIL.

The aim: To check the possibilities of introduction of methods for determining the index of PSPW and the average night basal level of impedance (MNBI), BIL-moda on the domestic MII-pH-M technique.

Materials and methods:Patients were divided on 2 groups: 1st [nonerosive GERD] – 11 women and 11 men, mean age (±S.E.M.) of 48.0±3.1 y.o.; 2nd [Reference] – 11 women and 10 men, mean age 47.9±3.5 y.o. Reference group included patients with non-ulcer dyspepsia. All patients MII-pH-monitoring with measuring current 6 μA at a frequency of 1 kHz and sampling frequency of 50 Hz. BIL was assessed at 3 cm proximal to the lower esophageal sphincter.

Results: In our study we obtained the best cut-off point for PSPW index – 58.2% with sensitivity 86.4% and specificity 81.0% (AUC=0,85). Our BIL-moda parameter demonstrated the best cut-off point <2,65 kΩ with sensitivity 90.9% and specificity 85.7% (AUC=0,89).

The MNBI indicator showed accuracy in the best cut-off point – 3,967 kΩ with sensitivity 95.5% and specificity 76.2% (AUC=0,87).

Conclusions: Applying the PSPW index, MNBI and our BIL-moda showed an obvious gain in diagnostic sensitivity – in 10% cases of NERD.

Wiad Lek 2018, 71, 8, -1536

Introduction

According to the latest global recommendations of the World Gastroenterology Organization (WGO-2015) for the treatment and diagnostics of gastroesophageal reflux disease (GERD), multichannel intraluminal impedance-pH monitoring (MII-pH-M) is the best tool in case of refractory patient to PPI, atypical (extra-esophageal) manifestations and determination of indication for antireflux endoscopic or surgical intervention [1]. Particular emphasis is made on the relevance of this functional test for nonerosive GERD (NERD), which not only predominates in the structure of the disease variants, but also more often demonstrates resistance to potent inhibitors of gastric acidity than endoscopy-positive forms. This clinical situation puts the diagnosis of GERD in doubt, because it can be caused by functional heartburn (FH), which is related to somatoform autonomic dysfunctions and requires a different treatment. A promising direction in solving this problem may be applying of modern methods of visualization with minimal damage to the distal esophagus mucosa, which are present in patients with NERD in contrast to functional heartburn. The most advanced endoscopic diagnostic techniques (confocal laser endomicroscopy, endoscopy with auto-fluorescence, I-scan endoscopy) are still the subject of further research [2]. Sensitivity and specificity of these methods are acceptable and varies within 90%, but the obstacle is the high cost and low prevalence of necessary equipment.

MII-pH-M is a method that is available in most tertiary centers of gastroenterological care in developed countries. Its wide practical use in the last decade has brought significant advantages over the conduct of conventional pH monitoring. The traditional determination of acid exposure in the esophagus as a fraction of time in a day with pH <4 units, the number of acid refluxes, as well as the integrative index DeMeester score allows verifying reflux disease in only half of patients with NERD [3]. The most significant increase to this sensitivity (10-15%) provides analysis of association of symptoms with reflux and the determination of the coefficients of SAP (symptom related probability) and SI (symptom index) [4,5]. In this case, the simultaneous determination of non-acid reflux by means of impedance significantly improves the quality of the diagnosis. During monitoring patients don’t always have complaints of heartburn or other manifestations of the disease. Consequently, the analysis of the association of symptoms with reflux is impossible. The situation can be improved by conducting a 48-hour study instead of 24-hour one with the use of wireless systems [6,7], but there aren’t commercial equipment with simultaneous impedance recording.

Therefore, the search of gastroenterologists are aimed at identifying other criteria of pathological gastroesophageal reflux in MII-pH-M, which could increase the sensitivity and specificity of the method. In the last seven years the so-called baseline impedance level (BIL) has been actively studied. BIL is reduced by delay of saliva in the lower third of the esophagus, edema of the inflamed esophagus mucus, decreased muscle tone and what is more important due to impaired mucosal integrity. All of these factors are present in patients with NERD but absent in patients with FH. The conventional approach of determining the BIL is time-consuming due to the necessity to select “stable minute” (one minute without a reflux episode or swallow) every hour and calculate the median value of 24 samples [8]. Therefore, there are efforts to simplify this task by automatic counting. For example, we tested the method of visual estimation of the distribution curve (histogram) of impedance measurements per day with subsequent determination of the peak value (the most often value – mode) [9]. In several publications Frazzoni and co-workers prove the rationality of determining a stable impedance in three 10-minute time frames during night time recumbent period – around 1:00 AM , 2:00 AM , and 3:00 AM [10,11,12,13]. This significantly simplifies the definition of BIL because it is necessary to calculate the mean of only three values. The authors proposed the name of this indicator – «mean nocturnal baseline impedance» (MNBI).

The aforementioned Italian researcher M. Frazzoni drew attention to an interesting feature of the motility of the esophagus in GERD. Patients have impaired chemical and mechanical clearance of the esophagus when the refluxates reach the organ. If among healthy people and patients with heartburn of functional origin the retrograde bolus entering almost immediately provokes a peristaltic transit of saliva (swallowing), in patients with GERD this event is delayed – the reflex does not work. The authors named the phenomenon of «post-reflux swallow-induced peristaltic wave» (PSPW). PSPW is defined as anterograde reduction in impedance by 50% within 30 seconds after end of gastroesophageal reflux – bolus output (Fig.1). This drop starts from the most proximal impedance measurement segment (Z4), reaching distal one and then returning to the baseline by at least 50%. The PSPW index is obtained dividing the number of refluxes followed within 30 sec by a PSPW on the number of total refluxes. The authors examined a wide range of patients with FH and NERD. The large part of them had negative results of classical MII-pH-M with a normal number of reflux, acid exposure and without proven association of symptoms with refluxes. Regardless of the above, the PSPW index with threshold 61% detected the presence of GERD or hypersensitive esophagus and distinguished these patients from patients with FH and healthy individuals [14].

In Ukraine domestic MII-pH-M technology including original pH-metric probes with the ability to simultaneously measure impedance was developed. The necessity for the introduction of this equipment is caused not only widespread GERD in Eastern Europe [1], but also by the prestige of our country. We were interested to take advantage of the innovations of the Italian researchers using the Ukrainian MII-pH-M technique in the examination of the contingent of patients, which are most often referred to the functional diagnosis of the esophagus.

THE AIM

The aim of the study: To check the possibilities of introduction of methods for determining the index of post-reflux droplet waves (PSPW) and the average night basal level of impedance (MNBI) on the domestic MII-pH-M technique and to compare their accuracy with traditional indicators.

MATERIALS AND METHODS

MII-pH monitoring was carried out using a tungsten pH sensor, placed on a 1.8 mm diameter probe similar to ПЕ-2рН (СКБ «МЕД», м. Кам’янець-Подільський), equipped with additional 6 stainless steel electrodes for impedance measurements. The adjacent electrodes form 4 impedance measuring pairs (Z1-4) located at 3, 5, 7 and 12 cm above from the lower oesophageal sphincter (LOS). The probe is inserted through the inferior nasal conchae to the level of the distal esophagus with the pH sensor placed 5 cm above the LOS. Previously a location of the LOS was determined by the following formula:

L=45+((H – 175)/4) (cm),

where L- distance from the patient’s nostril (cm); H – height of the patient (cm).

After the insertion of the probe the position of LOS was further defined by means of distal impedance electrode pair slipping along the esophagogastric junction. Resistance changes were displayed on the monitor of a connected laptop using a specialized program. The “jump“ of impedance from 0.5-0.8 kΩ to 1.2-1.8 kΩ corresponds to the location of LOS.

In the study we used the computer system of the MII-pH-monitoring AG-3pH-4R (ТОВ «Старт», Vinnytsia), which consisted of a mobile digitrapper, a laptop, a software for storage and math processing information.

The impedance measuring block contained an AC generator with a frequency of 1000 Hz, a commutator that consistently switched the voltage between 1-2-3-4-th impedance pairs with a discreteness of measurements 50 Hz, a voltmeter with analog-to-digital converter. The current strength of the measurement was 6 μA.

Patients had lunch at 14:00 with an approximate caloric value of 700 kcal, dinner at 19:30 (800 kcal), breakfast at 08:00 (500 kcal) and the night’s sleep from 22:00 to 07:00. The intake of proton pump inhibitors and histamine H2-receptor antagonists was withdrawn one week before the study.

The data received by the digitrapper were transferred to a personal computer where they were recorded and visualized using the graphical interface of the program. The operator looked at all recording channels (1 pH and 4 Z) with the subsequent analysis of events recorded on the charts (Fig.1).

In addition to the calculation of the above-described post-reflux swallow-induced peristaltic wave (PSPW) and the mean nocturnal baseline impedance (MNBI), the calculation of acidic and total liquid-mixed refluxes, fraction time with pH <4 (AET – acid exposure time) were performed. Data analysis was performed on liquid and liquid-gas (mixed) reflux episodes for acidic (nadir pH <4), weakly acidic (nadir pH between 4 and 7), and weakly alkaline refluxes (nadir pH not below 7); meal times were excluded. The symptom association probability (SAP) and the symptom index (SI) were considered positive with threshold ≥95% and ≥50% respectively.

BIL was assessed at 3 cm proximal to the lower esophageal sphincter (LOS) by using the most distal pair of stainless steel electrodes with diameter 2 mm, length 4 mm and distance 20 mm. A histogram of impedance values was generated for each recording using bin-width =0.1 kΩ (Fig.2). Visual assessment allowed to establish peak (the most often value – mode) of histogram corresponded to BIL and therefore excluded influences of swallowing, refluxes and etc.

The study was carried out in Diagnostic gastroenterologic laboratory at the Department of internal and family medicine in Vinnytsia National Medical University within 2013-2017.

Two groups of patients were selected for which MII-pH-monitoring is most relevant and causes the most problems in differential diagnosis – NERD and functional disorders of the upper part of the digestive tract. In the second (reference) group the diagnosis of non-ulcerous dyspepsia (including functional) was dominated alone or in combination with other conditions. Detailed information on age, gender is presented in Table I. Diagnosis of GERD and FH were established by the criteria of the Montreal (2006) and Rome-III (2005) consensuses. The analysis of the association of symptoms with refluxes by SAP/SI indexes were the determining factor for the diagnosis of FH and esophageal hypersensitivity in NERD.

All patients were previously undergoing esophagogastroduodenoscopy, ultrasound examination of the abdominal cavity, routine laboratory tests. The exclusion criteria were: age up to 20 years or above 75 years, pregnancy and lactation, peptic ulcer (active or healed less than 12 months ago), resection of the stomach, Zollinger-Ellison syndrome, ulcerative colitis, Crohn’s disease, chronic heart failure above IIIrd FC by NYHA, chronic kidney disease with GFR less than 30 mL/min, pulmonary failure above II st., hepatic encephalopathy above Ist degree, portal hypertension II – IV Baveno stages.

Statistical data processing was carried out using program «MedCalc 11.3.3.0» (MedCalc software bvba, Holland) by calculation of the arithmetic mean, its standard deviation and building ROC(Receiver operating characteristic)-curves to summarize the accuracy of reflux predictions.

RESULTS AND DISCUSSION

The values of the indices studied in the NERD group significantly differed from those in the reference group (Tabl.II). Almost all of these indicators, with the exception of the total number of fluid refluxes, are considered as criteria for the presence of pathological gastroesophageal refluxes.

The most comprehensive view about accuracy of each indicator is the building ROC-curves with the calculation of its parameters (see Fig. 3). The values of the area under the ROC curves (AUC) presented in Table III indicate that all of them are those that have a good characteristic (> 0.90) or are very close to it.

The main thing is that the negative results from different indicators did not coincide in one observation at the same time. There were individuals in which some indicators operated, but others didn’t. Thus, the composite application of various indices allows in general to increase the sensitivity of research on the identification of pathological gastroesophageal reflux.

Applying the PSPW index, we practically confirmed the results of M.Frazzoni regarding the threshold value of the number of effective sips (58.2 versus 61%) [14]. The most valuable thing was that in 2 of 22 cases, the GERD was confirmed only by PSPW index and not by the conventional indicators of reflux, associations with symptoms and acid exposure. The lesser sensitivity and specificity PSPW index received by us (AUC 0.85 vs 0.977) presumably were due to the fact that in Italian trial the non-responders to PPI were dominated in both groups. We didn’t have many such patients.

It is worth pointing out that 3 patients with a confirmed NERD, but with high PSPW index, which differed from the rest of the patients in poor tolerance of transnasal monitoring. They particularly complained that the 24-hour study was extremely burdensome for them. It provoked a sleep disorder, very often sipping in comparison with regular patients. Perhaps this fact must be taken into account in the further analysis and should ignore PSPW index in above-described patients.

The MNBI indicator showed similar accuracy with trial of M.Frazzoni and co-workers [14]. But for us the best cut-off point (threshold between normal and abnormal condition) was almost twofold higher – 3,967 versus 2,292 kΩ . The explanation for this discrepancy may be in different designs of probes – the size of the measuring Z-electrodes, their shape, the composition of the metal alloy, etc. The equipment developed by us used the same parameters of current strength as in most foreign analogues, but probes have significant differences. This fact is important for the implementation of the indicators obtained on one equipment and usage on another.

The low specificity of MNBI and similar to our BIL can be explained by the presence of a significant number of patients with functional dyspepsia in the reference group. From scientific publications it is known that almost one third of these patients have pathological reflux without heartburn or extraesophageal manifestations [15]. Taking into account that this indicator is the most sensitive, it probably reflects the presence of latent unexpressed gastroesophageal reflux in patients with dyspepsia.

CONCLUSIONS

Thus, the detection of post-reflux swallow-induced peristaltic wave (PSPW) index and the mean nocturnal baseline impedance (MNBI) are reliable tools for functional diagnosis of pathological gastroesophageal reflux, which significantly increase the diagnostic capabilities of MII-pH-M. The use of domestic monitoring equipment makes some corrections regarding the application of standards, which is obviously due to the structural features of Ukrainian probes.

References

1. Hunt R., Armstrong D., Katelaris P. et al. World Gastroenterology Organisation Global Guidelines: GERD Global Perspective on Gastroesophageal Reflux Disease. J.Clin. Gastroenterol. 2017; 51(6):467-478.

2. Chu C., Du Q., Li C., et al. Ambulatory 24-hour multichannel intraluminal impedance-pH monitoring and high resolution endoscopy distinguish patients with non-erosive reflux disease from those with functional heartburn. PLoS One. 2017;12(4):e0175263.

3. Long J.D, Orlando R.C. Nonerosive reflux disease. Minerva Gastroenterol Dietol. 2007;53, N.2:127-141

4. Bredenoord A.J, Weusten B.L. et al. Addition of esophageal impedance monitoring to pH monitoring increases the yield of symptom association analysis in patients off PPI therapy. Am J Gastroenterol. 2006;101, N.3:453-459.

5. Kline M.M., Ewing M., Simpson N. et al. The utility of intraluminal impedance in patients with gastroesophageal reflux disease-like symptoms but normal endoscopy and 24-hour pH testing. Clin Gastroenterol Hepatol. 2008;6 N.8:880-885.

6. Penagini R., Sweis R., Mauro A. et al. Inconsistency in the Diagnosis of Functional Heartburn: Usefulness of Prolonged Wireless pH Monitoring in Patients With Proton Pump Inhibitor Refractory Gastroesophageal Reflux Disease. J. Neurogastroenterol. Motil. 2015;V.21(2):265–272.

7. Chander B., Hanley-Williams N., Deng Y. et al. 24 Versus 48-hour bravo pH monitoring. J Clin Gastroenterol. 2012;V.46, N.3:197-200.

8. de Bortoli N., Martinucci I., Savarino E. et al. Association Between Baseline Impedance Values and Response Proton Pump Inhibitors in Patients With Heartburn Clinical Gastroenterology and Hepatology. 2015;V.13:1082–1088.

9. Melashchenko S.G. Opredelenie bazalnogo impedansa v nizhnem segmente pischevoda kak diagnosticheskogo kriterija gastroezophagealnoy refluksnoj bolezni [Determination of basal impedance in the lower segment of the esophagus as a diagnostic criterion of gastroesophageal reflux disease] Russian Journal of Gastroenterology, Hepatology, Coloproctology. 2015;Vol.25, №5 (Annex 46 – Materials of the XXI United Gastroenterological Week). Moscow:10.

10. Frazzoni M, Manta R, Mirante VG, et al. Esophageal chemical clearance is impaired in gastroesophageal reflux disease – A 24 h impedance-pH monitoring assessment. Neurogastroenterol. Motil. 2013;25:399–406.

11. Frazzoni M, Savarino E, de Bortoli N, et al. Analyses of the post-reflux swallow-induced peristaltic wave index and nocturnal base-line impedance parameters increase the diagnostic yield of patients with reflux disease. Clin. Gastroenterol. Hepatol. 2016;14:40–46.

12. Martinucci I, de Bortoli N, Savarino E, et al. Esophageal base-line impedance levels in patients with pathophysiological char-acteristics of functional heartburn. Neurogastroenterol Motil. 2014;26:546–555.

13. de Bortoli N, Martinucci I, Savarino E, et al. Association between baseline impedance values and response to proton pump in-hibitors in patients with heartburn. Clin Gastroenterol Hepatol. 2015;13:1082–1088.

14. Frazzoni M., de Bortoli N., Frazzoni L. et al. Impedance-pH monitoring for diagnosis of reflux disease: new perspectives Dig. Dis. Sci. 2017;62:1881–1889.

15. Tack J., Caenepeel P., Arts J. et al. Prevalence of acid reflux in functional dyspepsia and its association with symptom profile. Gut. 2005;54:1370-1376.

This article was carried out within the framework of research work on the department of internal and family medicine, Vinnytsia National N.I. Pirogov Memorial Medical University- «Esophagogastroduodenal peptic (acid-related) and Hp-associated diseases with comorbidity: the ways of improvement in diagnostics, treatment and prevention by means of multichannel intraluminal impedance-pH-monitoring.»

Authors’ contributions:

According to the order of the Authorship.

Conflict of interest:

The Authors declare no conflict of interest.

CORRESPONDING AUTHOR

Oleh O. Ksenchyn

Department of family and internal medicine

National Pirogov Memorial Medical University

Pirogova str. 56, 21018, Vinnytsia, Ukraine

e-mail: vinshura@gmail.com

Received: 06.08.2018

Accepted: 29.10.2018

Fig.1. A fragment of impedance-pH-gram showing patterns of acid refluxes (RF) and swallows (SW). The distance between the exit of the reflux bolus from the distal portion of the esophagus and the subsequent swallow is marked by PSPW. (own observation).

Fig.2. A histogram of disriburtion impedance measurements gathered during whole day from Z1-channel (3 см above LOS) in patient with NERD and index BIL* (baseline impedance level) 1,8 kΩ.

Table I. Characteristics of clinical groups

NERD

Reference group

Number of people

22

21

males/ females

11/11

10/11

Mean age (M±m)

48.0±3.1

47.9±3.5

Diafr. hernia (n/%)

8/(36.4%)

2/(9.5%)

Clinical manifestations (n/%):

Heartburn (>1 weekly)
– 17(77.3%)
Cough – 3(13.6%)

Cardialgia -3(13.6%)

Dyspepsia – 9(40.3%)

Non-ulcer dyspepsia – 15(71.4%)

Functional heartburn – 4 (19.0%)

Cough (unassociated with refluxes) – 2(9.5%)

Cardialgia (unassociated with refluxes) – 2(9.5%)

Globus hystericus (unassociated with rf) – 2(9.5%)

Burning mouth syndrome – 2(9.5%)

Results SAP or SI:

+/ – / asymptomatic

12 / 0 / 10

0 / 9 / 12

Table II. Indicators of MII-pH-M, studied in clinical groups

Index

NERD

Reference group

(M±m); the proportion of patients with a positive GERD test

Acidic refluxes

64.1±5.4; 18/22

19.6±4.3 3/21

Total refluxes

92.9±7.5; 14/22

55.7±5.2; 3/21

Percent time with рН<4 (AET)

9.17±1.0;

3.05±0.9;

BIL-moda

2.02±0.19; 3/22

3.81±0.23; 3/21

PSPW index

46.9±3.0; 19/22

68.18±3.2; 4/21

MNBI

2.68±0.24; 1/22

4.31±0.20; 5/21

Table III. Value of indicators of MII-pH-M, studied in clinical groups

Index

The

best cut-off point

AUC (95%CI)

Sensitivity

Specificity

Acidic refluxes

>40

0,92 (0,79-0,98)

86,4

90,5

Total refluxes

>77

0,84 (0,70-0,94)

63,6

90,5

Percent time with рН<4 (AET)

>4%

0,90 (0,77-0,97)

90,9

81,0

BIL

<2,65 kΩ

0,89 (0,76-0,97)

90,9

85,7

PSPW index

<58,2%

0,85 (0,70-0,94)

86,4

81,0

MNBI

<3,96 kΩ

0,87 (0,72-0,95)

95,5

76,2

Fig.3. ROC-curves for Index post-reflux swallow-induced peristaltic wave (PSPW – а) and mean