Alimentary Pharmacology and Therapeutics
Dilated intercellular space diameter as marker of reflux-related mucosal injury in children with chronic cough and gastrooesophageal reflux disease O. Borrelli*, V. Mancini†, N. Thapar*, M. Ribolsi‡, S. Emerenziani‡, G. de’Angelis†, B. Bizzarri†, K. J. Lindley* & M. Cicala‡
*Department of Gastroenterology, Neurogastroenterology and Motility Division, Great Ormond Street Hospital and Institute of Child Health, London, UK. † Division of Pediatric Gastroenterology, Parma University, Parma, Italy. ‡ Department of Digestive Disease, Campus Bio Medico University of Rome, Rome, Italy.
Correspondence to: Dr O. Borrelli, Department of Paediatric Gastroenterology, Division of Neurogastroenterology & Motility, Great Ormond Street Hospital, Great Ormond Street, WC1N 3HZ London, UK. E-mail: [email protected]
Publication data Submitted 12 December 2013 First decision 2 January 2014 Resubmitted 18 January 2014 Accepted 18 January 2014 EV Pub Online 11 February 2014 This article was accepted for publication after full peer-review.
SUMMARY Background The diagnostic corroboration of the relationship between gastro-oesophageal reflux disease (GERD) and chronic cough remains challenging. Aims To compare oesophageal mucosal intercellular space diameter (ISD) in children with GERD, children with gastro-oesophageal reflux (GER)-related cough (GrC) and a control group, and to explore the relationship between baseline impedance levels and dilated ISD in children with GER-related cough. Methods Forty children with GERD, 15 children with GrC and 12 controls prospectively underwent oesophagogastroduodenoscopy (EGD) with oesophageal biopsies taken 2–3 cm above squamocolumnar junction. ISD were quantified using transmission electron microscopy. Impedance-pH monitoring with evaluation of baseline impedance in the most distal impedance channel was performed in both patient groups. Results A significant difference in mean ISD values was found between GrC patients (0.9 0.2 lm) and controls (0.5 0.2 lm, P < 0.001), whereas there was no difference between GrC and GERD group (1 0.3 lm, NS). No difference was found in the mean ISD between GrC children with or without pathological oesophageal acid exposure time (1 0.3 vs. 0.9 0.2 lm), and there was no correlation between ISD and any reflux parameter. Finally, there was no correlation between ISD and distal baseline impedance values (r: 0.35; NS). Conclusions In children with reflux-related cough, dilated intercellular space diameter appears to be an objective and useful marker of oesophageal mucosal injury regardless of acid exposure, and its evaluation should be considered for those patients where the diagnosis is uncertain. In children with reflux-related cough, baseline impedance levels have no role in identifying reflux-induced oesophageal mucosal ultrastructural changes. Aliment Pharmacol Ther 2014; 39: 733–742
ª 2014 John Wiley & Sons Ltd doi:10.1111/apt.12652
733
O. Borrelli et al. INTRODUCTION Gastro-oesophageal reflux disease (GERD) has been recently categorised into oesophageal and extra-oesophageal syndromes, as consequences of the effects of gastric refluxate in the oesophagus or beyond it.1 However, the relationship between GERD and respiratory symptoms is still a subject of considerable deliberation, and although the role of reflux in the pathophysiology of respiratory manifestations is supported by several epidemiological studies, there are still limited data to corroborate or rebut cause and effect.2, 3 In adults, GERD is commonly identified as a cause of chronic cough.4 In childhood, the proportion of patients with unexplained chronic cough attributed to GER ranges between 15% and 45%, and in a subgroup they appear to be inter-related as sole cause or manifestation.5, 6 The correct diagnosis of GERD in patients with chronic cough, however, remains challenging due to the poor specificity and sensitivity of the available diagnostic tools. Both paediatric and adult studies have suggested that dilated intercellular space diameter (ISD) of the oesophageal epithelium represents a useful and sensitive ultrastructural marker of mucosal impairment in patients with GERD regardless of presence of abnormal acid exposure or classical histological changes of oesophagitis.7, 8 Furthermore, it has been proposed that during resting, the assessment of baseline impedance using oesophageal impedance-pH monitoring might provide useful information of oesophageal mucosal integrity.9 As far as we know, the measurement of ISD to assess children with GERD-related cough (GrC) has not been previously evaluated, nor whether changes in baseline impedance in this patient group parallel the degree of dilatation of intracellular spaces. In the present study, we compared oesophageal mucosal ISD of children with GERD with that of patients with GrC and controls. A secondary objective was to explore whether in patients with GrC any relationship exists between baseline impedance levels, reflux pattern and the presence or degree of dilated ISD. MATERIALS AND METHODS Patients and study protocol Between January 2009 and January 2010, we conducted a prospective longitudinal study including three groups of subjects namely: (i) controls; (ii) GERD patients; and (iii) GERD-related cough (GrC) patients. The control group consisted of 12 children with positive coeliac 734
serology who underwent oesophagogastroduodenoscopy (OGD) to verify the diagnosis. Patients were recruited as controls for the measurement of oesophageal ISD if they had no history of foregut symptoms nor had received previous treatment with acid suppression therapy. GERD patients consisted of children with typical reflux symptoms (as their chief complaints) lasting at least 8 weeks. They were used as controls for the measurement of both oesophageal ISD and functional parameters measured using impedance-pH monitoring. GrC patients were identified following referral by respiratory paediatricians. All patients had been formally reviewed and the most common oro-pharyngeal and respiratory causes of cough excluded. This process comprised an internationally accepted diagnostic protocol comprising a detailed medical history and examination, pulmonary function testing, chest X-ray, ENT evaluation including laryngoscopy and sinus CT and/or MRI imaging.10 Exclusion criteria included anatomical abnormalities of the lungs, airways, heart and gastrointestinal tract, known systemic and infectious diseases, previous surgery on the gastrointestinal tract, neurological impairment (e.g. cerebral palsy), food and respiratory allergies, immune disorders, suspected inhalation of foreign body, cystic fibrosis and idiopathic pulmonary fibrosis. Subjects that had received H2 receptor antagonist, proton pump inhibitors or medications affecting the function of the lower oesophageal sphincter (i.e. bronchodilators) prior to the study were excluded as were patients who smoked. The study was conducted in accordance with the Helsinki Declaration, and ethical approval was obtained by the Ethics Committees of Universities of Rome and Parma. In all patients and controls prior to procedures, written parental consent was obtained, and a statement of assent was signed by all children older than 12 years of age. All eligible children with unexplained chronic cough underwent 24-h impedance-pH monitoring, and those with abnormal tests were subsequently invited to participate in the study and to undergo oesophagogastroduodenoscopy (OGD), which was performed within a week (median 2 days, range 1–5 days) of the 24-h impedance-pH monitoring. Conversely, as per protocol, those children with normal impedance-pH monitoring were not considered for endoscopic examination. All eligible children with symptoms of GERD who consented to take part in the study after baseline assessment underwent OGD followed within 1 week (median 3 days, range 1–6 days) by 24-h impedance-pH monitoring. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children Upper GI endoscopy and histomorphometric analysis All study subjects (controls and patients) underwent OGD, with a paediatric endoscope (Olympus XQ260, outer diameter: 9.8 mm, bioptic channel: 2.0 mm) under short-acting anaesthesia with propofol. Endoscopic appearances of the oesophageal mucosa were graded using the Los Angeles system.11 Mucosal biopsies were taken from the duodenum, stomach and oesophagus and fixed in formalin for later embedding in paraffin; the tissue sections were subsequently stained with haematoxylin and eosin for routine histology. Of four oesophageal biopsies obtained 2–3 cm above the squamocolumnar junction, two were used to corroborate the presence of reflux oesophagitis only when the endoscopic oesophageal appearance was within normal limits and/or rule out the presence of eosinophilic oesophagitis. The remaining two oesophageal biopsies were processed for transmission electron microscopy by fixing in 2% glutaraldehyde in saline phosphate buffer at 7.2 and embedding in epoxy resin before cutting of ultrathin sections and lead citrate staining.12 Prepared slides were then examined using a Morgagni 268D digital transmission electron microscope (FEI Company, Eindhoven, Netherlands). EM Images of at least of 10 representative fields were obtained and magnified to 50009. Photographs of 10 fields with an internal scale marker, each showing a whole cell with opposing cell membranes from the basal and prickle layers of oesophageal epithelium, were randomly selected. Morphometric analysis of intercellular spaces was performed using EndoxPro System (Casti imaging, Medra-Venice, Italy). On each of the fields, 10 perpendicular transect lines were randomly drawn across selected areas of intercellular spaces leaving at least 5 lm between two adjacent transects and 100 transects available for measurements from each patients were obtained.13, 14 The following parameters were noted: (i) the mean ISD value; (ii) the mean maximum ISD value; (iii) the minimum ISD value. ISD values were assessed by two investigators blinded to both clinical and impedance-pH data (VM and MR). 24-h impedance-pH monitoring All patients underwent ambulatory 24-h impedance-pH monitoring (Sleuth, Sandhill Scientific, Inc; Highland Ranch, CO, USA). The technique, methodology and recording performance of this system have been previously described.15 The definition and classification of the reflux episodes were in accordance with previous standardised criteria.16 Both pH and impedance data were utilised for determining reflux patterns, including the Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
type, number and characteristics of reflux episodes, and for quantifying the acid exposure time (AET, percentage), acid clearance time (ACT, sec) and baseline impedance levels.17 Acid exposure time (AET) was considered abnormal when pH5% or SAP was >95%.6 The diagnosis of GERD was performed in the presence of either an abnormal impedance-pH monitoring based on quantitative and/or quantitative analysis or endoscopic and/or histological oesophagitis.8 MII-pH monitoring was considered quantitatively abnormal if the oesophageal acid exposure time was >5% of the total recording time, whereas it was considered qualitatively positive if SAP was >95%.6, 18 Sample size estimation was performed for the evaluation of ISD by using a 2-sided a of 0.05 and 80% power based on previous studies reporting dilated ISD in GERD patients as compared with controls. A sample size of 48 patients (16 in each group) was scheduled to be included to show a 0.2 unit difference among groups. Data were analysed using Prism for windows version 5.00 (GraphPad, San Diego, CA, USA), and data were expressed as median (25th to 75th), if not otherwise stated. The characteristics of the patients were evaluated by simple descriptive analysis. One-way ANOVA followed by Dunn’s Multiple Comparison Test for post hoc analyses was used to analyse differences among groups. Chi-square test, Fisher exact test and Mann–Whitney test were used when appropriate. Correlations between different 735
O. Borrelli et al. parameters were performed using Spearman’s correlation test. All statistical tests were 2-tailed, and P < 0.05 was indicative of statistical significance.
RESULTS Figure 1 summarises the flow of GrC patients throughout the study. Over a period of 12 months, forty-eight children with chronic cough were considered eligible for the study, of whom 15 with abnormal impedance-pH monitoring underwent OGD and were analysed. Twenty-two children with chronic cough did not undergo OGD due to a normal impedance-pH monitoring (Figure 1). However, they were used as controls for the evaluation of baseline impedance values. During the same recruitment period, 56 children with typical GERD symptoms were considered eligible, and the complete assessment was available in 44 children. Four patients with normal impedance-pH monitoring and without mucosal injury at endoscopy and oesophagitis at histology were excluded from the final analysis. Twelve asymptomatic children (median age: 8.3 years, range 2–13) with positive coeliac screening were recruited during the study period and served as controls for the assessment of oesophageal ISD. Cough patients assessed for eligibility (n = 48) Refused to undergo MII-pH imp (n = 4) Patients undergone MII-pH imp (n = 44) Excluded for technical issues (n = 2) Excluded for negative MII-pH imp (n = 22) Patients considered for the study (n = 20) Refused to participate (n = 5) Patients enrolled into the study (n = 15) Abnormal AET (n = 6)
SAP + (n = 2)
AR (n = 2)
SAP – (n = 4)
Normal AET (n = 9)
SAP + (n = 9)
SAP – (n = 0)
AR AR-Wac Wac-Walk (n = 5) (n = 3) (n = 1)
Figure 1 | Cough patient flow throughout the study. The number of patients is indicated in brackets. AET, acid exposure time; SAP, symptom association probability; AR, acid reflux; WAc, weakly acidic reflux; Walk, weakly alkaline reflux. 736
No differences were found among the patient groups with regard to age and gender frequency, but significant differences were found between GERD group and GrC group in term of epigastric pain (P < 0.05), regurgitation/vomiting (P < 0.001), pyrosis (P < 0.001), and chough (P < 0.001) (Table 1). Among GrC patients, thirteen (80%) did not complain any typical symptom of GERD. At endoscopy, none of the children with GrC had any apparent mucosal injury or hiatal hernia.
Histomorphometric analysis Figure 2 shows examples of electron microscopy photographs demonstrating normal and dilated ISD. A significant increase in mean ISD values was found in both patient groups as compared with controls, whereas there were no differences between the GrC and GERD groups [GrC: 0.9 0.2 lm (95% CI: 0.8 1), GERD Table 1 | Demographics and baseline disease characteristics of the two groups of patients
No. of cases Age (years; median and ranges) M/F Disease duration (weeks; median and ranges) Symptoms and signs (number and percentage) Regurgitation/Vomiting Pyrosis Epigastric pain Dysphagia Wheezing/Asthma Haematemesis Cough Presence of Erosive Oesophagitis** (number and percentage) Grade A Grade B Grade C Grade D Hiatal Hernia Presence of Oesophagitis† (number and percentage) Abnormal AET (number and percentage)
GrC
GERD
15 7.7 (2–15)
40 8.9 (3–17)
9/6 9 (6–15)
27/17 13 (8–18)
2/15 1/15 2/15 – – – 15/15 –
30/40 26/40 19/40 7/40 5/40 3/40 9/40 18/40
(75%)* (65%)* (47.5%)& (17.5%) (12.5%) (7.5%) (22.5%)* (45%)
– – – – – 6/15 (40%)
10/40 7/40 1/40 – 2/40 28/40
(25%) (17.5%) (2.5%)
6/15 (46%)
27/40 (67.5%)
(13%) (7%) (13%)
(100%)
(5%) (70%)
GERD, gastro-oesophageal reflux disease; GrC, gastro-oesophageal reflux-related cough. *P < 0.001; **P < 0.05; according to Los Angeles classification. † According to histology. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children (a)
(b)
(c)
Figure 2 | Examples of transmission electron photomicrographs of oesophageal biopsy specimens from GERD-related cough children (a), GERD children (b), and controls (c). Each photograph is selected from an area of epithelium near the luminal surface. On each field, 10 transect lines were randomly drawn across area of intercellular spaces, thus obtaining 100 transects, available for measurement, from each individuals.
1 0.3 lm (95% CI: 0.9 1), Controls 0.5 0.2 lm (95% CI: 0.3 0.7), P < 0.001 by one-way ANOVA]. Likewise, there was a statistically significant difference in both maximum and minimum ISD values between patient groups and controls, but no difference between the GrC and GERD groups (NS) [Maximum Value: GrC: 1.5 0.4 lm (95% CI: 1.2 1.7), GERD 1.6 0.5 lm (95% CI: 1.5 1.8), Controls 0.9 0.3 lm (95% CI: 0.6 1), P < 0.001 by one-way ANOVA; Minimum value: GrC: 0.4 0.1 lm (95% CI: 0.4 0.5), GERD 0.5 0.1 lm (95% CI: 0.5 0.5), Controls 0.9 0.3 lm (95% CI: 0.2 0.3), P < 0.001 by one-way ANOVA]. Finally, segregating the GrC patients into those with abnormal acid exposure time and those with normal acid exposure time, there was no difference in the mean ISD values between the two patient subgroups (abnormal AET: 1 0.3, physiological AET: 0.9 0.2; NS).
24-h impedance-pH monitoring There was no difference between the GERD group and the GrC group in the proportion of children with abnormal AET (GERD: 27/40, 67.5%, GrC: 6/15, 40%; NS). The median number of total reflux episodes, the median number acid, weakly acidic and weakly alkaline reflux episodes did not differ between the GERD and GrC groups (Table 2). Conversely, a significant increase in AET, number of long-lasting reflux episodes and ACT was found in the GERD group as compared with the GrC group (Table 2). No difference was found between the two groups with regard to distal baseline impedance [GERD: 2968 (1708–4113) Ω, GrC: 3975 (2650–4220) Ω; NS] (Figure 3a). There was no statistical difference in impedance baseline values between GrC children and those with cough and negative impedance-pH Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
monitoring [4070 (3600–4569) Ω; NS]), whereas a statistically significant difference was found between the latter group and GERD group (P < 0.05). Finally, a statistically significant difference was found in the distal baseline values between GrC patients with abnormal AET and those with normal AET ([pathological AET: 2498 (1812–2927) Ω, physiological AET: 4217 (3981–4359) Ω; P < 0.001] (Figure 3b).
ISD and impedance-pH parameters In children with GrC, there was no correlation between distal mean ISD values and baseline impedance values (r: 0.3; NS) (Figure 4). On the other hand, a significant inverse correlation was found between AET and distal baseline impedance (r: 0.91; P < 0.001) (Figure 5a), and between the latter and number of reflux episodes >5 min (r: 0.87; P < 0.01) (Figure 5b), total number of reflux episodes (r: 0.7; P < 0.001), number of acid reflux (r: 0.73; P < 0.001) (Figure 5c) and ACT (r: 0.90; P < 0.001) (Figure 5d). There was no correlation between distal baseline impedance levels and weakly acidic or weakly alkaline reflux episodes (r: 0.1 and r: 0.1, respectively; NS). Finally, there was no correlation between ISD values and AET (r: 0.3, P = 0.2), as well as between the former and any other impedance-pH monitoring parameters. DISCUSSION Chronic chough is a common and debilitating complaint in children, representing one of the most frequent reasons for parents to seek medical advice.20 Even though in both adult and children it is generally accepted that gastric refluxate might be a cause of unexplained chronic cough, the diagnostic confirmation of this association remains a major challenge. More recently, it has been proposed that 737
O. Borrelli et al.
GERD
Total number of reflux 74 (64–89) 89.5 (72–123) episodes median [(25th to 75th)] Chemical composition of refluxate [median (25th–75th)] Acid episodes 40 (34–62) 52.5 (37–75) Weakly acidic episodes 11 (7–16) 17 (7.5–25.75) Weakly alkaline episodes 8 (3–10) 6 (1.25–15) pH-only reflux episodes 16 (8–28) 12 (7–19) Reflux Composition (percentage) Liquid 73% 70% Mixed 25% 29% Gas 2% 1% Oesophageal Acid Exposure 5 3.6* 7.8 4.3 Time (mean s.d.) Number of long-lasting 2 (0–5)* 4.5 (3–6) episodes (>5 min) [median (25th–75th)] Acid Clearance Time (s) 67 (38–154)* 145 (63.25–195) [median (25th–75th)] GERD, gastro-oesophageal reflux disease; GrC, gastro-oesophageal reflux-related cough. * P < 0.05 as compared to GERD group.
ultrastructural changes in the oesophageal mucosa may underlie the pathogenesis of typical symptoms in GERD. Amongst patients with GERD-related chronic cough, no such data are available for this or the relationship between oesophageal mucosal integrity and baseline impedance values. Therefore, we studied a group of children with unexplained chronic cough in whom GER was the most plausible co-factor for cough elicitation and the main findings were: (i) oesophageal mucosal ISD values were significantly higher in GERD-related cough (GrC) group compared with control group, (ii) the magnitude of ISD dilatation was similar between the groups of GrC and GERD; (iii) Dilated ISD was observed in GrC children despite physiological oesophageal acid exposure; (iv) in GrC patients, baseline impedance values as well as traditional parameters of pH-impedance were not helpful in identifying the oesophageal mucosal ultrastructural changes induced by gastric refluxate. It has repeatedly been shown that dilated oesophageal ISD is a useful and consistent marker of mucosal injury in GERD patients, both children and adult, irrespective of the presence of oesophagitis.7, 8, 12–14, 21 In the absence of visible oesophageal mucosal lesion, symptoms are believed to result from gastric refluxate, both acid and non-acid, being able to activate chemosensitive 738
Baseline impedance (Ω)
GrC
(a) 5000
4000 3000 2000 1000 0 GrC
GERD
Physiological AET
Pathological AET
(b) 5000
Baseline impedance (Ω)
Table 2 | Reflux characteristics in the two groups of patients
4000 3000 2000 1000 0
Figure 3 | Baseline impedance levels in GrC and GERD groups (a), and GrC children with abnormal and normal acid exposure time (b). Medians and interquartile range. GrC, gastro-oesophageal reflux disease-related cough; GERD, gastro-oesophageal reflux disease.
nociceptors (e.g. TRPV-1) on oesophageal submucosal sensory nerve terminals through dilated ISD. These terminals are, in turn, able to transmit signals to the brain for sensory processing and perception.22 Our study showed dilated ISD in children with GrC is of a similar magnitude to that in GERD patients. We show for the first time that ISD is also increased in GrC children having physiological acid exposure, but similar to those GrC children with abnormal acid exposure. Moreover, no statistical correlation was found between the degree of ISD dilatation and any reflux parameter. Thus, the occurrence of dilated ISD in children with GrC seems to be an objective morphological marker of reflux-impaired oesophageal mucosa integrity, as it discriminates this group of patients from controls regardless of the reflux pattern. The measurement of baseline impedance during impedance-pH monitoring has been proposed as a potential surrogate marker for oesophageal integrity and Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children therefore, mucosal damage. Oesophageal impedance-pH monitoring is a widely accepted method to diagnose GERD. It is capable of detecting both antegrade and 2.0
r: –0.3 NS
1.0
0.5
0.0 0
1000
2000
3000
4000
5000
Baseline impedance (Ω)
Figure 4 | Correlation between baseline impedance levels and intercellular space diameter (ISD) in GrC group. GrC, gastro-oesophageal reflux disease-related cough.
(a)
(b) 10
15
AET (percentage)
r: –0.87 P < 0.001 10
5
0 1000
2000
3000 4000 5000 Baseline impedance (Ω)
r: –0.80 P < 0.001
8 6 4 2 0 1000
6000
(c) 100
2000
3000 4000 5000 Baseline impedance (Ω)
6000
(d) 300 r: –0.85 P < 0.001
r: –0.73 P < 0.001
80
ACT (sec)
No of AR episodes
No of reflux episodes > 5 min
ISD (μm)
1.5
retrograde flow of the bolus within the oesophagus in a pH-independent fashion by measuring changes in impedance (i.e. resistance) to alternating current passed between pairs of metallic rings mounted along the length of a catheter. Given that the oesophagus at rest is an empty and narrow tube, baseline impedance, measured by contact of impedance-measuring segments with the oesophageal wall, is thought to provide information on the integrity of oesophageal mucosa.9 In our study, however, there was no correlation between dilated ISD and baseline impedance, suggesting that the latter cannot reliably identify ultrastructural changes in the oesophageal mucosa induced by gastric refluxate. Our results are in agreement with previous studies in GERD children where oesophageal mucosal ultrastructural changes were not predicted by baseline impedance values despite the presence of erosive oesophagitis.17 Moreover, our study showed that baseline impedance is affected by acid exposure as an inverse correlation was found between distal baseline impedance and both
60 40
200
100 20 0 1000
2000
3000
4000
Baseline impedance (Ω)
5000
6000
0 1000
2000
3000
4000
5000
6000
Baseline impedance (Ω)
Figure 5 | Correlation between baseline impedance values and acid exposure time (AET, a), prolonged episodes of reflux (>5 min) (b), acid reflux episodes (c), and acid clearance time (ACT, d). GrC, gastro-oesophageal reflux disease-related cough. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
739
O. Borrelli et al. number of acid reflux episodes and oesophageal acid exposure time, and there was no correlation between acid exposure time and ISD. Our findings corroborate data in adults showing a clear relationship between acid exposure time and baseline impedance, and suggest that the baseline impedance is often decreased in symptomatic GERD patients compared with healthy controls.23 In a recent double-blind, randomised, placebo-controlled trial in preterm and term infants with clinical symptoms and signs suggestive of GERD, Loots and co-workers showed that therapy with proton pump inhibitors (PPIs) increased oesophageal baseline levels, which were correlated with reduced acid exposure corroborating the close relationship between the two parameters.24 Finally, Woodland et al. in an elegant study showed that distal oesophageal acid infusion caused an abrupt fall in baseline impedance, the recovery from which was slower in non-erosive reflux disease (NERD) patients with low baseline impedance and high oesophageal acid exposure time compared with patients with functional heartburn, corroborating the close relationship between the amount and frequency of gastric acid within the oesophageal lumen and the baseline impedance values.25 Two major mechanisms by which GERD induces chronic cough have been proposed, including micro-aspiration of gastric refluxate into the respiratory tree, and activation of afferent sensory neurons by refluxate in the oesophageal lumen. A number of studies in adults and experimental models suggest that the activation of afferent neurons into the oesophagus, e.g. by acid, leads to peripheral or central sensitisation of the cough reflex. In adults with GERD and chronic cough, infusion of acid into the distal oesophagus increases the cough reflex (hyper) sensitivity, shown as a decreased threshold to inhaled capsaicin.26 Moreover, this sensitivity is heightened in patients with chronic cough and positive reflux–cough association as compared with those without, supporting the key role of vagally mediated mechanisms.27 Thus, we are tempted to suggest that in patients with GERD-induced cough, dilated ISD represents the ‘primum movens’ through which gastric contents sensitise the afferent pathway of the cough reflex within the oesophageal wall. Our study has some limitations. First, it is possible that the differences between patients and controls may represent a type II error. Although larger groups would strengthen the data, we do feel that the sample size was large enough to identify significant differences between study groups with regard to impedance parameters and ISD. Secondly, the site of the oesophageal biopsies taken 740
in our study could be contentious. However, it is a common practice in paediatric age to take oesophageal biopsies 2–3 cm above the squamocolumnar junction.28, 29 Furthermore, it has been shown in a large series of adult GERD patients that although the biopsies at squamocolumnar junction might increase the overall sensitivity in discriminating between GERD patients and controls, they result in a decreased specificity, as basal cell hyperplasia and papillae elongation are also present in high percentage of controls.30 Interestingly, the same authors reported similar prevalence of dilated ISD among different GERD patient groups at squamocolumnar junction and 2 cm above it, whereas the prevalence significantly decreases when the biopsies are taken 4 cm above the squamocolumnar junction.30 Moreover, Fiocca et al., in a recent study aiming to develop and standardise criteria for recognising microscopic oesophageal lesions, including dilated ISD, reported that the four biopsy specimens taken at each quadrant 2 cm above the squamocolumnar junction show similar degree of epithelial damage.31 However, further research in this area is warranted. Thirdly, for ethical reasons, we did not perform the morphometric evaluation in children with chronic cough and negative MII-pH monitoring analysis perhaps representing the most challenging group of patients for which the measurement of ISD could be highly beneficial, as also suggested by the observation that they display baseline impedance values similar to those of GrC children. The presence of dilated ISD in this group could provide a more robust tool to diagnose GERD-related cough and evaluate new alternative treatments directed specifically at mitigating the intercellular space dilatation. Further studies in this group of patients are warranted. In conclusion, our data provide evidence that in children with unexplained chronic cough and GERD, dilated ISD seems to be a useful and objective marker of oesophageal mucosal injury regardless of acid exposure, and its evaluation should be considered for those children when the diagnosis is uncertain. Conversely, baseline impedance levels are not useful in identifying the oesophageal mucosal ultrastructural changes induced by gastric refluxate. However, further larger studies aiming to determine the sensitivity and specificity of ISD in both adults and children with GERD-related chronic cough are warranted. Moreover, prospective, controlled study with long-term follow-up to evaluate whether the presence of dilated ISD can predict the efficacy of aggressive anti-reflux in those patients with chronic cough and negative pH-impedance test is required. Finally, the mechanisms underlying the pathophysiology of chronic cough Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children such as peripherally or centrally mediated oesophageal sensitivity deserve to be further investigated.
AUTHORSHIP Guarantor of the article: Osvaldo Borrelli. Author contributions: OB designed the research study, wrote the manuscript, performed data acquisition and analysed the data. VM and MR performed morphometric analysis and critically revised the manuscript. NT analysed the data and critically revised the manuscript. SE analysed the MII-pH monitoring data, analysed the
data and critically revised the manuscript. BB analysed the MII-pH monitoring data, critically revised the manuscript and approved the submission. GdA performed data acquisition, analysed the data and critically revised the manuscript and approved the submission. KJL & MC designed the research study, critically revised the manuscript and approved the submission. All authors approved the final version of the manuscript.
ACKNOWLEDGMENT Declaration of personal and funding interests: None.
REFERENCES 1. Sherman PM, Hassall E, Fagundes-Neto U, et al. A global, evidence-based consensus on the definition of gastroesophageal reflux disease in the paediatric population. Am J Gastroenterol 2009; 104: 1278–95. 2. El-Serag HB, Gilger M, Kuebeler M, et al. Extraoesophageal associations of gastroesophageal reflux disease in children without neurologic defects. Gastroenterology 2001; 121: 1294–9. 3. Tolia V, Vandenplas Y. Systematic review: the extra-oesophageal reflux symptoms of gastrooesophageal reflux disease in children. Aliment Pharmacol Ther 2009; 29: 258–72. 4. Irwin RS. Chronic cough due to gastroesophageal reflux disease: AACP evidence-based clinical practice guidelines. Chest 2006; 19: 80S–94S. 5. Khoshoo V, Edell D, Mohnot S, et al. Associated factors in children with chronic cough. Chest 2009; 136: 811–5. 6. Borrelli O, Marabotto C, Mancini V, et al. Role of gastro-esophageal reflux in children with unexplained chronic cough. J Pediatr Gastroenterol Nutr 2011; 53: 287–92. 7. van Malenstein H, Farre R, Sifrim D. Oesophageal dilated intercellular spaces (DIS) and nonerosive reflux disease. Am J Gastroenterol 2008; 103: 1021–8. 8. Mancini V, Ribolsi M, Gentile M, et al. Oesophageal mucosal intercellular space diameter and reflux pattern in childhood erosive and non-erosive reflux disease. Dig Liver Dis 2012; 44: 981–7. 9. Farre R, Blondeau K, Clement D, et al. Evaluation of oesophageal mucosa integrity by the intraluminal impedance technique. Gut 2011; 60: 885–92. 10. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in paediatrics:
Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
11.
12.
13.
14.
15.
16.
17.
18.
AACP evidence-based clinical practice guidelines. Chest 2006; 19: 260S–83S. Lundell L, Dent J, Bennett J, et al. Endoscopic assessment of esophagitis: clinical and functional correlates and further validation of Los Angeles classification. Gut 1999; 45: 172–80. Tobey NA, Carson JL, Alkiek RA, et al. Dilated intercellular spaces: a morphological feature of acid reflux– damaged human esophageal epithelium. Gastroenterology 1996; 111: 1200–5. Caviglia R, Ribolsi M, Maggiano N, et al. Dilated intercellular spaces of esophageal epithelium in nonerosive reflux disease patients with physiological esophageal acid exposure. Am J Gastroenterol 2005; 100: 543–8. Caviglia R, Ribolsi M, Gentile M, et al. Dilated intercellular spaces and acid reflux at the distal and proximal esophagus in patients with non-erosive gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2007; 25: 629– 36. Borrelli O, Battaglia M, Galos F, et al. Non-acid gastro-oesophageal reflux in children with suspected pulmonary aspiration. Dig Liv Dis 2010; 42: 115– 21. Sifrim D, Castell D, Dent J, et al. Gastro-oesophageal reflux monitoring: review and consensus report on detection and definitions of acid, nonacid, and gas reflux. Gut 2004; 53: 1024–31. Borrelli O, Salvatore S, Mancini V, et al. Relationship between baseline impedance levels and oesophageal mucosal integrity in children with erosive and non-erosive reflux disease. Neurogastroenterol Motil 2012; 24: 828– e394. Bredenoord AJ, Wuesten BL, Smouth AJ. Symptom association analysis in
19.
20.
21.
22.
23.
24.
25.
26.
ambulatory gastrooesophageal reflux monitoring. Gut 2005; 54: 1810–7. Omari TI, Schwarzer A, vanWijk MP, et al. Optimization of the refluxsymptom association statistics for use in infants being investigated by 24-hour pH impedance. J Pediatr Gastroenterol Nutr 2011; 52: 408–13. Marchant JM, Newcombe PA, Juniper EF, et al. What is the burden of chronic cough for families? Chest 2008; 134: 303–9. Guarino MP, Cheng L, Ma J, et al. Increased TRPV1 gene expression in oesophageal mucosa of patients with non-erosive and erosive reflux disease. Neurogastroenterol Motil 2010; 22: 746–51. Kandulski A, Jechorek D, Caro C, et al. Histomorphological differentiation of non-erosive reflux disease and functional heartburn in patients with PPI-refractory heartburn. Aliment Pharmacol Ther 2013; 38: 643–51. Kessing BF, Bredenoord AJ, Weijenborg PW, et al. Esophageal acid exposure decreases intraluminal baseline impedance levels. Am J Gastroenterol 2011; 106: 2093–7. Loots CM, Wijnakker R, van Wijk MP, et al. Oesophageal impedance baselines in infants before and after placebo and proton pump inhibitor therapy. Neurogastroenterol Motil 2012; 24: 758–62. Woodland P, Al-Zinaty M, Yazaki E, et al. In vivo evaluation of acidinduced changes in oesophageal mucosa integrity and sensitivity in non-erosive reflux disease. Gut 2013; 62: 1256–61. Javorkova N, Varechova S, Pecova R, et al. Acidification of the esophagus acutely increases the cough sensitivity in patients with gastro-oesophageal reflux and chronic cough.
741
O. Borrelli et al. Neurogastroenterol Motil 2008; 20: 119–24. 27. Smith JA, Decalmer S, Kelsall A, et al. Acoustic cough-reflux associations in chronic cough: potential triggers and mechanisms. Gastroenterology 2010; 139: 754–62. 28. Winter HS, Madara JL, Stafford RJ, et al. Intraepithelial eosinophils: a new diagnostic criterion for reflux esophagitis. Gastroenterology 1982; 83: 818–23.
742
29. Sch€appi M, Mougenot J, Belli DC. Upper gastrointestinal endoscopy. In Kleinman et al. editors Walker’s Pediatric Gastrointestinal Disease. 5th ed. Hamilton, ON, Canada: BC Decker Inc, 2008; 1265–83. 30. Mastracci L, Spaggiari P, Grillo F, et al. Microscopic esophagitis in gastroesophageal reflux disease: individual lesions, biopsy sampling, and clinical correlations. Virchows Arch 2009; 454: 31–9.
31. Fiocca R, Mastracci L, Riddell R, et al. Development of consensus guidelines for the histologic recognition of microscopic esophagitis in patients with gastroesophageal reflux disease: the Esohisto project. Hum Pathol 2010; 41: 223–31.
Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter as marker of reflux-related mucosal injury in children with chronic cough and gastrooesophageal reflux disease O. Borrelli*, V. Mancini†, N. Thapar*, M. Ribolsi‡, S. Emerenziani‡, G. de’Angelis†, B. Bizzarri†, K. J. Lindley* & M. Cicala‡
*Department of Gastroenterology, Neurogastroenterology and Motility Division, Great Ormond Street Hospital and Institute of Child Health, London, UK. † Division of Pediatric Gastroenterology, Parma University, Parma, Italy. ‡ Department of Digestive Disease, Campus Bio Medico University of Rome, Rome, Italy.
Correspondence to: Dr O. Borrelli, Department of Paediatric Gastroenterology, Division of Neurogastroenterology & Motility, Great Ormond Street Hospital, Great Ormond Street, WC1N 3HZ London, UK. E-mail: [email protected]
Publication data Submitted 12 December 2013 First decision 2 January 2014 Resubmitted 18 January 2014 Accepted 18 January 2014 EV Pub Online 11 February 2014 This article was accepted for publication after full peer-review.
SUMMARY Background The diagnostic corroboration of the relationship between gastro-oesophageal reflux disease (GERD) and chronic cough remains challenging. Aims To compare oesophageal mucosal intercellular space diameter (ISD) in children with GERD, children with gastro-oesophageal reflux (GER)-related cough (GrC) and a control group, and to explore the relationship between baseline impedance levels and dilated ISD in children with GER-related cough. Methods Forty children with GERD, 15 children with GrC and 12 controls prospectively underwent oesophagogastroduodenoscopy (EGD) with oesophageal biopsies taken 2–3 cm above squamocolumnar junction. ISD were quantified using transmission electron microscopy. Impedance-pH monitoring with evaluation of baseline impedance in the most distal impedance channel was performed in both patient groups. Results A significant difference in mean ISD values was found between GrC patients (0.9 0.2 lm) and controls (0.5 0.2 lm, P < 0.001), whereas there was no difference between GrC and GERD group (1 0.3 lm, NS). No difference was found in the mean ISD between GrC children with or without pathological oesophageal acid exposure time (1 0.3 vs. 0.9 0.2 lm), and there was no correlation between ISD and any reflux parameter. Finally, there was no correlation between ISD and distal baseline impedance values (r: 0.35; NS). Conclusions In children with reflux-related cough, dilated intercellular space diameter appears to be an objective and useful marker of oesophageal mucosal injury regardless of acid exposure, and its evaluation should be considered for those patients where the diagnosis is uncertain. In children with reflux-related cough, baseline impedance levels have no role in identifying reflux-induced oesophageal mucosal ultrastructural changes. Aliment Pharmacol Ther 2014; 39: 733–742
ª 2014 John Wiley & Sons Ltd doi:10.1111/apt.12652
733
O. Borrelli et al. INTRODUCTION Gastro-oesophageal reflux disease (GERD) has been recently categorised into oesophageal and extra-oesophageal syndromes, as consequences of the effects of gastric refluxate in the oesophagus or beyond it.1 However, the relationship between GERD and respiratory symptoms is still a subject of considerable deliberation, and although the role of reflux in the pathophysiology of respiratory manifestations is supported by several epidemiological studies, there are still limited data to corroborate or rebut cause and effect.2, 3 In adults, GERD is commonly identified as a cause of chronic cough.4 In childhood, the proportion of patients with unexplained chronic cough attributed to GER ranges between 15% and 45%, and in a subgroup they appear to be inter-related as sole cause or manifestation.5, 6 The correct diagnosis of GERD in patients with chronic cough, however, remains challenging due to the poor specificity and sensitivity of the available diagnostic tools. Both paediatric and adult studies have suggested that dilated intercellular space diameter (ISD) of the oesophageal epithelium represents a useful and sensitive ultrastructural marker of mucosal impairment in patients with GERD regardless of presence of abnormal acid exposure or classical histological changes of oesophagitis.7, 8 Furthermore, it has been proposed that during resting, the assessment of baseline impedance using oesophageal impedance-pH monitoring might provide useful information of oesophageal mucosal integrity.9 As far as we know, the measurement of ISD to assess children with GERD-related cough (GrC) has not been previously evaluated, nor whether changes in baseline impedance in this patient group parallel the degree of dilatation of intracellular spaces. In the present study, we compared oesophageal mucosal ISD of children with GERD with that of patients with GrC and controls. A secondary objective was to explore whether in patients with GrC any relationship exists between baseline impedance levels, reflux pattern and the presence or degree of dilated ISD. MATERIALS AND METHODS Patients and study protocol Between January 2009 and January 2010, we conducted a prospective longitudinal study including three groups of subjects namely: (i) controls; (ii) GERD patients; and (iii) GERD-related cough (GrC) patients. The control group consisted of 12 children with positive coeliac 734
serology who underwent oesophagogastroduodenoscopy (OGD) to verify the diagnosis. Patients were recruited as controls for the measurement of oesophageal ISD if they had no history of foregut symptoms nor had received previous treatment with acid suppression therapy. GERD patients consisted of children with typical reflux symptoms (as their chief complaints) lasting at least 8 weeks. They were used as controls for the measurement of both oesophageal ISD and functional parameters measured using impedance-pH monitoring. GrC patients were identified following referral by respiratory paediatricians. All patients had been formally reviewed and the most common oro-pharyngeal and respiratory causes of cough excluded. This process comprised an internationally accepted diagnostic protocol comprising a detailed medical history and examination, pulmonary function testing, chest X-ray, ENT evaluation including laryngoscopy and sinus CT and/or MRI imaging.10 Exclusion criteria included anatomical abnormalities of the lungs, airways, heart and gastrointestinal tract, known systemic and infectious diseases, previous surgery on the gastrointestinal tract, neurological impairment (e.g. cerebral palsy), food and respiratory allergies, immune disorders, suspected inhalation of foreign body, cystic fibrosis and idiopathic pulmonary fibrosis. Subjects that had received H2 receptor antagonist, proton pump inhibitors or medications affecting the function of the lower oesophageal sphincter (i.e. bronchodilators) prior to the study were excluded as were patients who smoked. The study was conducted in accordance with the Helsinki Declaration, and ethical approval was obtained by the Ethics Committees of Universities of Rome and Parma. In all patients and controls prior to procedures, written parental consent was obtained, and a statement of assent was signed by all children older than 12 years of age. All eligible children with unexplained chronic cough underwent 24-h impedance-pH monitoring, and those with abnormal tests were subsequently invited to participate in the study and to undergo oesophagogastroduodenoscopy (OGD), which was performed within a week (median 2 days, range 1–5 days) of the 24-h impedance-pH monitoring. Conversely, as per protocol, those children with normal impedance-pH monitoring were not considered for endoscopic examination. All eligible children with symptoms of GERD who consented to take part in the study after baseline assessment underwent OGD followed within 1 week (median 3 days, range 1–6 days) by 24-h impedance-pH monitoring. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children Upper GI endoscopy and histomorphometric analysis All study subjects (controls and patients) underwent OGD, with a paediatric endoscope (Olympus XQ260, outer diameter: 9.8 mm, bioptic channel: 2.0 mm) under short-acting anaesthesia with propofol. Endoscopic appearances of the oesophageal mucosa were graded using the Los Angeles system.11 Mucosal biopsies were taken from the duodenum, stomach and oesophagus and fixed in formalin for later embedding in paraffin; the tissue sections were subsequently stained with haematoxylin and eosin for routine histology. Of four oesophageal biopsies obtained 2–3 cm above the squamocolumnar junction, two were used to corroborate the presence of reflux oesophagitis only when the endoscopic oesophageal appearance was within normal limits and/or rule out the presence of eosinophilic oesophagitis. The remaining two oesophageal biopsies were processed for transmission electron microscopy by fixing in 2% glutaraldehyde in saline phosphate buffer at 7.2 and embedding in epoxy resin before cutting of ultrathin sections and lead citrate staining.12 Prepared slides were then examined using a Morgagni 268D digital transmission electron microscope (FEI Company, Eindhoven, Netherlands). EM Images of at least of 10 representative fields were obtained and magnified to 50009. Photographs of 10 fields with an internal scale marker, each showing a whole cell with opposing cell membranes from the basal and prickle layers of oesophageal epithelium, were randomly selected. Morphometric analysis of intercellular spaces was performed using EndoxPro System (Casti imaging, Medra-Venice, Italy). On each of the fields, 10 perpendicular transect lines were randomly drawn across selected areas of intercellular spaces leaving at least 5 lm between two adjacent transects and 100 transects available for measurements from each patients were obtained.13, 14 The following parameters were noted: (i) the mean ISD value; (ii) the mean maximum ISD value; (iii) the minimum ISD value. ISD values were assessed by two investigators blinded to both clinical and impedance-pH data (VM and MR). 24-h impedance-pH monitoring All patients underwent ambulatory 24-h impedance-pH monitoring (Sleuth, Sandhill Scientific, Inc; Highland Ranch, CO, USA). The technique, methodology and recording performance of this system have been previously described.15 The definition and classification of the reflux episodes were in accordance with previous standardised criteria.16 Both pH and impedance data were utilised for determining reflux patterns, including the Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
type, number and characteristics of reflux episodes, and for quantifying the acid exposure time (AET, percentage), acid clearance time (ACT, sec) and baseline impedance levels.17 Acid exposure time (AET) was considered abnormal when pH5% or SAP was >95%.6 The diagnosis of GERD was performed in the presence of either an abnormal impedance-pH monitoring based on quantitative and/or quantitative analysis or endoscopic and/or histological oesophagitis.8 MII-pH monitoring was considered quantitatively abnormal if the oesophageal acid exposure time was >5% of the total recording time, whereas it was considered qualitatively positive if SAP was >95%.6, 18 Sample size estimation was performed for the evaluation of ISD by using a 2-sided a of 0.05 and 80% power based on previous studies reporting dilated ISD in GERD patients as compared with controls. A sample size of 48 patients (16 in each group) was scheduled to be included to show a 0.2 unit difference among groups. Data were analysed using Prism for windows version 5.00 (GraphPad, San Diego, CA, USA), and data were expressed as median (25th to 75th), if not otherwise stated. The characteristics of the patients were evaluated by simple descriptive analysis. One-way ANOVA followed by Dunn’s Multiple Comparison Test for post hoc analyses was used to analyse differences among groups. Chi-square test, Fisher exact test and Mann–Whitney test were used when appropriate. Correlations between different 735
O. Borrelli et al. parameters were performed using Spearman’s correlation test. All statistical tests were 2-tailed, and P < 0.05 was indicative of statistical significance.
RESULTS Figure 1 summarises the flow of GrC patients throughout the study. Over a period of 12 months, forty-eight children with chronic cough were considered eligible for the study, of whom 15 with abnormal impedance-pH monitoring underwent OGD and were analysed. Twenty-two children with chronic cough did not undergo OGD due to a normal impedance-pH monitoring (Figure 1). However, they were used as controls for the evaluation of baseline impedance values. During the same recruitment period, 56 children with typical GERD symptoms were considered eligible, and the complete assessment was available in 44 children. Four patients with normal impedance-pH monitoring and without mucosal injury at endoscopy and oesophagitis at histology were excluded from the final analysis. Twelve asymptomatic children (median age: 8.3 years, range 2–13) with positive coeliac screening were recruited during the study period and served as controls for the assessment of oesophageal ISD. Cough patients assessed for eligibility (n = 48) Refused to undergo MII-pH imp (n = 4) Patients undergone MII-pH imp (n = 44) Excluded for technical issues (n = 2) Excluded for negative MII-pH imp (n = 22) Patients considered for the study (n = 20) Refused to participate (n = 5) Patients enrolled into the study (n = 15) Abnormal AET (n = 6)
SAP + (n = 2)
AR (n = 2)
SAP – (n = 4)
Normal AET (n = 9)
SAP + (n = 9)
SAP – (n = 0)
AR AR-Wac Wac-Walk (n = 5) (n = 3) (n = 1)
Figure 1 | Cough patient flow throughout the study. The number of patients is indicated in brackets. AET, acid exposure time; SAP, symptom association probability; AR, acid reflux; WAc, weakly acidic reflux; Walk, weakly alkaline reflux. 736
No differences were found among the patient groups with regard to age and gender frequency, but significant differences were found between GERD group and GrC group in term of epigastric pain (P < 0.05), regurgitation/vomiting (P < 0.001), pyrosis (P < 0.001), and chough (P < 0.001) (Table 1). Among GrC patients, thirteen (80%) did not complain any typical symptom of GERD. At endoscopy, none of the children with GrC had any apparent mucosal injury or hiatal hernia.
Histomorphometric analysis Figure 2 shows examples of electron microscopy photographs demonstrating normal and dilated ISD. A significant increase in mean ISD values was found in both patient groups as compared with controls, whereas there were no differences between the GrC and GERD groups [GrC: 0.9 0.2 lm (95% CI: 0.8 1), GERD Table 1 | Demographics and baseline disease characteristics of the two groups of patients
No. of cases Age (years; median and ranges) M/F Disease duration (weeks; median and ranges) Symptoms and signs (number and percentage) Regurgitation/Vomiting Pyrosis Epigastric pain Dysphagia Wheezing/Asthma Haematemesis Cough Presence of Erosive Oesophagitis** (number and percentage) Grade A Grade B Grade C Grade D Hiatal Hernia Presence of Oesophagitis† (number and percentage) Abnormal AET (number and percentage)
GrC
GERD
15 7.7 (2–15)
40 8.9 (3–17)
9/6 9 (6–15)
27/17 13 (8–18)
2/15 1/15 2/15 – – – 15/15 –
30/40 26/40 19/40 7/40 5/40 3/40 9/40 18/40
(75%)* (65%)* (47.5%)& (17.5%) (12.5%) (7.5%) (22.5%)* (45%)
– – – – – 6/15 (40%)
10/40 7/40 1/40 – 2/40 28/40
(25%) (17.5%) (2.5%)
6/15 (46%)
27/40 (67.5%)
(13%) (7%) (13%)
(100%)
(5%) (70%)
GERD, gastro-oesophageal reflux disease; GrC, gastro-oesophageal reflux-related cough. *P < 0.001; **P < 0.05; according to Los Angeles classification. † According to histology. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children (a)
(b)
(c)
Figure 2 | Examples of transmission electron photomicrographs of oesophageal biopsy specimens from GERD-related cough children (a), GERD children (b), and controls (c). Each photograph is selected from an area of epithelium near the luminal surface. On each field, 10 transect lines were randomly drawn across area of intercellular spaces, thus obtaining 100 transects, available for measurement, from each individuals.
1 0.3 lm (95% CI: 0.9 1), Controls 0.5 0.2 lm (95% CI: 0.3 0.7), P < 0.001 by one-way ANOVA]. Likewise, there was a statistically significant difference in both maximum and minimum ISD values between patient groups and controls, but no difference between the GrC and GERD groups (NS) [Maximum Value: GrC: 1.5 0.4 lm (95% CI: 1.2 1.7), GERD 1.6 0.5 lm (95% CI: 1.5 1.8), Controls 0.9 0.3 lm (95% CI: 0.6 1), P < 0.001 by one-way ANOVA; Minimum value: GrC: 0.4 0.1 lm (95% CI: 0.4 0.5), GERD 0.5 0.1 lm (95% CI: 0.5 0.5), Controls 0.9 0.3 lm (95% CI: 0.2 0.3), P < 0.001 by one-way ANOVA]. Finally, segregating the GrC patients into those with abnormal acid exposure time and those with normal acid exposure time, there was no difference in the mean ISD values between the two patient subgroups (abnormal AET: 1 0.3, physiological AET: 0.9 0.2; NS).
24-h impedance-pH monitoring There was no difference between the GERD group and the GrC group in the proportion of children with abnormal AET (GERD: 27/40, 67.5%, GrC: 6/15, 40%; NS). The median number of total reflux episodes, the median number acid, weakly acidic and weakly alkaline reflux episodes did not differ between the GERD and GrC groups (Table 2). Conversely, a significant increase in AET, number of long-lasting reflux episodes and ACT was found in the GERD group as compared with the GrC group (Table 2). No difference was found between the two groups with regard to distal baseline impedance [GERD: 2968 (1708–4113) Ω, GrC: 3975 (2650–4220) Ω; NS] (Figure 3a). There was no statistical difference in impedance baseline values between GrC children and those with cough and negative impedance-pH Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
monitoring [4070 (3600–4569) Ω; NS]), whereas a statistically significant difference was found between the latter group and GERD group (P < 0.05). Finally, a statistically significant difference was found in the distal baseline values between GrC patients with abnormal AET and those with normal AET ([pathological AET: 2498 (1812–2927) Ω, physiological AET: 4217 (3981–4359) Ω; P < 0.001] (Figure 3b).
ISD and impedance-pH parameters In children with GrC, there was no correlation between distal mean ISD values and baseline impedance values (r: 0.3; NS) (Figure 4). On the other hand, a significant inverse correlation was found between AET and distal baseline impedance (r: 0.91; P < 0.001) (Figure 5a), and between the latter and number of reflux episodes >5 min (r: 0.87; P < 0.01) (Figure 5b), total number of reflux episodes (r: 0.7; P < 0.001), number of acid reflux (r: 0.73; P < 0.001) (Figure 5c) and ACT (r: 0.90; P < 0.001) (Figure 5d). There was no correlation between distal baseline impedance levels and weakly acidic or weakly alkaline reflux episodes (r: 0.1 and r: 0.1, respectively; NS). Finally, there was no correlation between ISD values and AET (r: 0.3, P = 0.2), as well as between the former and any other impedance-pH monitoring parameters. DISCUSSION Chronic chough is a common and debilitating complaint in children, representing one of the most frequent reasons for parents to seek medical advice.20 Even though in both adult and children it is generally accepted that gastric refluxate might be a cause of unexplained chronic cough, the diagnostic confirmation of this association remains a major challenge. More recently, it has been proposed that 737
O. Borrelli et al.
GERD
Total number of reflux 74 (64–89) 89.5 (72–123) episodes median [(25th to 75th)] Chemical composition of refluxate [median (25th–75th)] Acid episodes 40 (34–62) 52.5 (37–75) Weakly acidic episodes 11 (7–16) 17 (7.5–25.75) Weakly alkaline episodes 8 (3–10) 6 (1.25–15) pH-only reflux episodes 16 (8–28) 12 (7–19) Reflux Composition (percentage) Liquid 73% 70% Mixed 25% 29% Gas 2% 1% Oesophageal Acid Exposure 5 3.6* 7.8 4.3 Time (mean s.d.) Number of long-lasting 2 (0–5)* 4.5 (3–6) episodes (>5 min) [median (25th–75th)] Acid Clearance Time (s) 67 (38–154)* 145 (63.25–195) [median (25th–75th)] GERD, gastro-oesophageal reflux disease; GrC, gastro-oesophageal reflux-related cough. * P < 0.05 as compared to GERD group.
ultrastructural changes in the oesophageal mucosa may underlie the pathogenesis of typical symptoms in GERD. Amongst patients with GERD-related chronic cough, no such data are available for this or the relationship between oesophageal mucosal integrity and baseline impedance values. Therefore, we studied a group of children with unexplained chronic cough in whom GER was the most plausible co-factor for cough elicitation and the main findings were: (i) oesophageal mucosal ISD values were significantly higher in GERD-related cough (GrC) group compared with control group, (ii) the magnitude of ISD dilatation was similar between the groups of GrC and GERD; (iii) Dilated ISD was observed in GrC children despite physiological oesophageal acid exposure; (iv) in GrC patients, baseline impedance values as well as traditional parameters of pH-impedance were not helpful in identifying the oesophageal mucosal ultrastructural changes induced by gastric refluxate. It has repeatedly been shown that dilated oesophageal ISD is a useful and consistent marker of mucosal injury in GERD patients, both children and adult, irrespective of the presence of oesophagitis.7, 8, 12–14, 21 In the absence of visible oesophageal mucosal lesion, symptoms are believed to result from gastric refluxate, both acid and non-acid, being able to activate chemosensitive 738
Baseline impedance (Ω)
GrC
(a) 5000
4000 3000 2000 1000 0 GrC
GERD
Physiological AET
Pathological AET
(b) 5000
Baseline impedance (Ω)
Table 2 | Reflux characteristics in the two groups of patients
4000 3000 2000 1000 0
Figure 3 | Baseline impedance levels in GrC and GERD groups (a), and GrC children with abnormal and normal acid exposure time (b). Medians and interquartile range. GrC, gastro-oesophageal reflux disease-related cough; GERD, gastro-oesophageal reflux disease.
nociceptors (e.g. TRPV-1) on oesophageal submucosal sensory nerve terminals through dilated ISD. These terminals are, in turn, able to transmit signals to the brain for sensory processing and perception.22 Our study showed dilated ISD in children with GrC is of a similar magnitude to that in GERD patients. We show for the first time that ISD is also increased in GrC children having physiological acid exposure, but similar to those GrC children with abnormal acid exposure. Moreover, no statistical correlation was found between the degree of ISD dilatation and any reflux parameter. Thus, the occurrence of dilated ISD in children with GrC seems to be an objective morphological marker of reflux-impaired oesophageal mucosa integrity, as it discriminates this group of patients from controls regardless of the reflux pattern. The measurement of baseline impedance during impedance-pH monitoring has been proposed as a potential surrogate marker for oesophageal integrity and Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children therefore, mucosal damage. Oesophageal impedance-pH monitoring is a widely accepted method to diagnose GERD. It is capable of detecting both antegrade and 2.0
r: –0.3 NS
1.0
0.5
0.0 0
1000
2000
3000
4000
5000
Baseline impedance (Ω)
Figure 4 | Correlation between baseline impedance levels and intercellular space diameter (ISD) in GrC group. GrC, gastro-oesophageal reflux disease-related cough.
(a)
(b) 10
15
AET (percentage)
r: –0.87 P < 0.001 10
5
0 1000
2000
3000 4000 5000 Baseline impedance (Ω)
r: –0.80 P < 0.001
8 6 4 2 0 1000
6000
(c) 100
2000
3000 4000 5000 Baseline impedance (Ω)
6000
(d) 300 r: –0.85 P < 0.001
r: –0.73 P < 0.001
80
ACT (sec)
No of AR episodes
No of reflux episodes > 5 min
ISD (μm)
1.5
retrograde flow of the bolus within the oesophagus in a pH-independent fashion by measuring changes in impedance (i.e. resistance) to alternating current passed between pairs of metallic rings mounted along the length of a catheter. Given that the oesophagus at rest is an empty and narrow tube, baseline impedance, measured by contact of impedance-measuring segments with the oesophageal wall, is thought to provide information on the integrity of oesophageal mucosa.9 In our study, however, there was no correlation between dilated ISD and baseline impedance, suggesting that the latter cannot reliably identify ultrastructural changes in the oesophageal mucosa induced by gastric refluxate. Our results are in agreement with previous studies in GERD children where oesophageal mucosal ultrastructural changes were not predicted by baseline impedance values despite the presence of erosive oesophagitis.17 Moreover, our study showed that baseline impedance is affected by acid exposure as an inverse correlation was found between distal baseline impedance and both
60 40
200
100 20 0 1000
2000
3000
4000
Baseline impedance (Ω)
5000
6000
0 1000
2000
3000
4000
5000
6000
Baseline impedance (Ω)
Figure 5 | Correlation between baseline impedance values and acid exposure time (AET, a), prolonged episodes of reflux (>5 min) (b), acid reflux episodes (c), and acid clearance time (ACT, d). GrC, gastro-oesophageal reflux disease-related cough. Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
739
O. Borrelli et al. number of acid reflux episodes and oesophageal acid exposure time, and there was no correlation between acid exposure time and ISD. Our findings corroborate data in adults showing a clear relationship between acid exposure time and baseline impedance, and suggest that the baseline impedance is often decreased in symptomatic GERD patients compared with healthy controls.23 In a recent double-blind, randomised, placebo-controlled trial in preterm and term infants with clinical symptoms and signs suggestive of GERD, Loots and co-workers showed that therapy with proton pump inhibitors (PPIs) increased oesophageal baseline levels, which were correlated with reduced acid exposure corroborating the close relationship between the two parameters.24 Finally, Woodland et al. in an elegant study showed that distal oesophageal acid infusion caused an abrupt fall in baseline impedance, the recovery from which was slower in non-erosive reflux disease (NERD) patients with low baseline impedance and high oesophageal acid exposure time compared with patients with functional heartburn, corroborating the close relationship between the amount and frequency of gastric acid within the oesophageal lumen and the baseline impedance values.25 Two major mechanisms by which GERD induces chronic cough have been proposed, including micro-aspiration of gastric refluxate into the respiratory tree, and activation of afferent sensory neurons by refluxate in the oesophageal lumen. A number of studies in adults and experimental models suggest that the activation of afferent neurons into the oesophagus, e.g. by acid, leads to peripheral or central sensitisation of the cough reflex. In adults with GERD and chronic cough, infusion of acid into the distal oesophagus increases the cough reflex (hyper) sensitivity, shown as a decreased threshold to inhaled capsaicin.26 Moreover, this sensitivity is heightened in patients with chronic cough and positive reflux–cough association as compared with those without, supporting the key role of vagally mediated mechanisms.27 Thus, we are tempted to suggest that in patients with GERD-induced cough, dilated ISD represents the ‘primum movens’ through which gastric contents sensitise the afferent pathway of the cough reflex within the oesophageal wall. Our study has some limitations. First, it is possible that the differences between patients and controls may represent a type II error. Although larger groups would strengthen the data, we do feel that the sample size was large enough to identify significant differences between study groups with regard to impedance parameters and ISD. Secondly, the site of the oesophageal biopsies taken 740
in our study could be contentious. However, it is a common practice in paediatric age to take oesophageal biopsies 2–3 cm above the squamocolumnar junction.28, 29 Furthermore, it has been shown in a large series of adult GERD patients that although the biopsies at squamocolumnar junction might increase the overall sensitivity in discriminating between GERD patients and controls, they result in a decreased specificity, as basal cell hyperplasia and papillae elongation are also present in high percentage of controls.30 Interestingly, the same authors reported similar prevalence of dilated ISD among different GERD patient groups at squamocolumnar junction and 2 cm above it, whereas the prevalence significantly decreases when the biopsies are taken 4 cm above the squamocolumnar junction.30 Moreover, Fiocca et al., in a recent study aiming to develop and standardise criteria for recognising microscopic oesophageal lesions, including dilated ISD, reported that the four biopsy specimens taken at each quadrant 2 cm above the squamocolumnar junction show similar degree of epithelial damage.31 However, further research in this area is warranted. Thirdly, for ethical reasons, we did not perform the morphometric evaluation in children with chronic cough and negative MII-pH monitoring analysis perhaps representing the most challenging group of patients for which the measurement of ISD could be highly beneficial, as also suggested by the observation that they display baseline impedance values similar to those of GrC children. The presence of dilated ISD in this group could provide a more robust tool to diagnose GERD-related cough and evaluate new alternative treatments directed specifically at mitigating the intercellular space dilatation. Further studies in this group of patients are warranted. In conclusion, our data provide evidence that in children with unexplained chronic cough and GERD, dilated ISD seems to be a useful and objective marker of oesophageal mucosal injury regardless of acid exposure, and its evaluation should be considered for those children when the diagnosis is uncertain. Conversely, baseline impedance levels are not useful in identifying the oesophageal mucosal ultrastructural changes induced by gastric refluxate. However, further larger studies aiming to determine the sensitivity and specificity of ISD in both adults and children with GERD-related chronic cough are warranted. Moreover, prospective, controlled study with long-term follow-up to evaluate whether the presence of dilated ISD can predict the efficacy of aggressive anti-reflux in those patients with chronic cough and negative pH-impedance test is required. Finally, the mechanisms underlying the pathophysiology of chronic cough Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
Dilated intercellular space diameter and reflux-related cough in children such as peripherally or centrally mediated oesophageal sensitivity deserve to be further investigated.
AUTHORSHIP Guarantor of the article: Osvaldo Borrelli. Author contributions: OB designed the research study, wrote the manuscript, performed data acquisition and analysed the data. VM and MR performed morphometric analysis and critically revised the manuscript. NT analysed the data and critically revised the manuscript. SE analysed the MII-pH monitoring data, analysed the
data and critically revised the manuscript. BB analysed the MII-pH monitoring data, critically revised the manuscript and approved the submission. GdA performed data acquisition, analysed the data and critically revised the manuscript and approved the submission. KJL & MC designed the research study, critically revised the manuscript and approved the submission. All authors approved the final version of the manuscript.
ACKNOWLEDGMENT Declaration of personal and funding interests: None.
REFERENCES 1. Sherman PM, Hassall E, Fagundes-Neto U, et al. A global, evidence-based consensus on the definition of gastroesophageal reflux disease in the paediatric population. Am J Gastroenterol 2009; 104: 1278–95. 2. El-Serag HB, Gilger M, Kuebeler M, et al. Extraoesophageal associations of gastroesophageal reflux disease in children without neurologic defects. Gastroenterology 2001; 121: 1294–9. 3. Tolia V, Vandenplas Y. Systematic review: the extra-oesophageal reflux symptoms of gastrooesophageal reflux disease in children. Aliment Pharmacol Ther 2009; 29: 258–72. 4. Irwin RS. Chronic cough due to gastroesophageal reflux disease: AACP evidence-based clinical practice guidelines. Chest 2006; 19: 80S–94S. 5. Khoshoo V, Edell D, Mohnot S, et al. Associated factors in children with chronic cough. Chest 2009; 136: 811–5. 6. Borrelli O, Marabotto C, Mancini V, et al. Role of gastro-esophageal reflux in children with unexplained chronic cough. J Pediatr Gastroenterol Nutr 2011; 53: 287–92. 7. van Malenstein H, Farre R, Sifrim D. Oesophageal dilated intercellular spaces (DIS) and nonerosive reflux disease. Am J Gastroenterol 2008; 103: 1021–8. 8. Mancini V, Ribolsi M, Gentile M, et al. Oesophageal mucosal intercellular space diameter and reflux pattern in childhood erosive and non-erosive reflux disease. Dig Liver Dis 2012; 44: 981–7. 9. Farre R, Blondeau K, Clement D, et al. Evaluation of oesophageal mucosa integrity by the intraluminal impedance technique. Gut 2011; 60: 885–92. 10. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in paediatrics:
Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
11.
12.
13.
14.
15.
16.
17.
18.
AACP evidence-based clinical practice guidelines. Chest 2006; 19: 260S–83S. Lundell L, Dent J, Bennett J, et al. Endoscopic assessment of esophagitis: clinical and functional correlates and further validation of Los Angeles classification. Gut 1999; 45: 172–80. Tobey NA, Carson JL, Alkiek RA, et al. Dilated intercellular spaces: a morphological feature of acid reflux– damaged human esophageal epithelium. Gastroenterology 1996; 111: 1200–5. Caviglia R, Ribolsi M, Maggiano N, et al. Dilated intercellular spaces of esophageal epithelium in nonerosive reflux disease patients with physiological esophageal acid exposure. Am J Gastroenterol 2005; 100: 543–8. Caviglia R, Ribolsi M, Gentile M, et al. Dilated intercellular spaces and acid reflux at the distal and proximal esophagus in patients with non-erosive gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2007; 25: 629– 36. Borrelli O, Battaglia M, Galos F, et al. Non-acid gastro-oesophageal reflux in children with suspected pulmonary aspiration. Dig Liv Dis 2010; 42: 115– 21. Sifrim D, Castell D, Dent J, et al. Gastro-oesophageal reflux monitoring: review and consensus report on detection and definitions of acid, nonacid, and gas reflux. Gut 2004; 53: 1024–31. Borrelli O, Salvatore S, Mancini V, et al. Relationship between baseline impedance levels and oesophageal mucosal integrity in children with erosive and non-erosive reflux disease. Neurogastroenterol Motil 2012; 24: 828– e394. Bredenoord AJ, Wuesten BL, Smouth AJ. Symptom association analysis in
19.
20.
21.
22.
23.
24.
25.
26.
ambulatory gastrooesophageal reflux monitoring. Gut 2005; 54: 1810–7. Omari TI, Schwarzer A, vanWijk MP, et al. Optimization of the refluxsymptom association statistics for use in infants being investigated by 24-hour pH impedance. J Pediatr Gastroenterol Nutr 2011; 52: 408–13. Marchant JM, Newcombe PA, Juniper EF, et al. What is the burden of chronic cough for families? Chest 2008; 134: 303–9. Guarino MP, Cheng L, Ma J, et al. Increased TRPV1 gene expression in oesophageal mucosa of patients with non-erosive and erosive reflux disease. Neurogastroenterol Motil 2010; 22: 746–51. Kandulski A, Jechorek D, Caro C, et al. Histomorphological differentiation of non-erosive reflux disease and functional heartburn in patients with PPI-refractory heartburn. Aliment Pharmacol Ther 2013; 38: 643–51. Kessing BF, Bredenoord AJ, Weijenborg PW, et al. Esophageal acid exposure decreases intraluminal baseline impedance levels. Am J Gastroenterol 2011; 106: 2093–7. Loots CM, Wijnakker R, van Wijk MP, et al. Oesophageal impedance baselines in infants before and after placebo and proton pump inhibitor therapy. Neurogastroenterol Motil 2012; 24: 758–62. Woodland P, Al-Zinaty M, Yazaki E, et al. In vivo evaluation of acidinduced changes in oesophageal mucosa integrity and sensitivity in non-erosive reflux disease. Gut 2013; 62: 1256–61. Javorkova N, Varechova S, Pecova R, et al. Acidification of the esophagus acutely increases the cough sensitivity in patients with gastro-oesophageal reflux and chronic cough.
741
O. Borrelli et al. Neurogastroenterol Motil 2008; 20: 119–24. 27. Smith JA, Decalmer S, Kelsall A, et al. Acoustic cough-reflux associations in chronic cough: potential triggers and mechanisms. Gastroenterology 2010; 139: 754–62. 28. Winter HS, Madara JL, Stafford RJ, et al. Intraepithelial eosinophils: a new diagnostic criterion for reflux esophagitis. Gastroenterology 1982; 83: 818–23.
742
29. Sch€appi M, Mougenot J, Belli DC. Upper gastrointestinal endoscopy. In Kleinman et al. editors Walker’s Pediatric Gastrointestinal Disease. 5th ed. Hamilton, ON, Canada: BC Decker Inc, 2008; 1265–83. 30. Mastracci L, Spaggiari P, Grillo F, et al. Microscopic esophagitis in gastroesophageal reflux disease: individual lesions, biopsy sampling, and clinical correlations. Virchows Arch 2009; 454: 31–9.
31. Fiocca R, Mastracci L, Riddell R, et al. Development of consensus guidelines for the histologic recognition of microscopic esophagitis in patients with gastroesophageal reflux disease: the Esohisto project. Hum Pathol 2010; 41: 223–31.
Aliment Pharmacol Ther 2014; 39: 733-742 ª 2014 John Wiley & Sons Ltd
