Chronic cough is a common and debilitating complaint in childhood, and represents one of the most common reasons for which parents seek medical care (1). Although the role of gastroesophageal reflux disease (GERD) in the pathogenesis of chronic cough in adults is widely accepted (2), in children there is less convincing evidence to support this relation (3,4).
GER may trigger cough through several mechanisms including the aspiration of acid gastric contents into the upper airways, vagally mediated cough upon the presence of acid in the esophageal lumen, and the sensitization of the central cough reflex (5). Recent studies have emphasized a role for nonacid reflux (AR) in the pathogenesis of atypical reflux symptoms (6,7). Although 24-hour pH monitoring is one of the current standard methods for GERD assessment in children (3), it fails to detect some types of reflux, especially when little or no acid is present in the refluxate. Multichannel intraluminal impedance and pH (MII-pH) monitoring combined can characterize the reflux episodes as acid or nonacid, as well as the height reached by the refluxate (8). Although data on the relation between non-AR and chronic cough have been published in adults (9,10), to our knowledge there are few published data regarding this issue in children (11).
In this prospective study we aimed to describe the type and physical characteristics of reflux episodes in a selected population of children with unexplained chronic cough and to compare the reflux pattern of the latter group with that found in children with erosive reflux disease (ERD).
PATIENTS AND METHODS
All of the children with unexplained chronic cough were considered eligible for the study. Unexplained chronic cough was defined as a daily cough lasting more than 4 weeks, in the absence of identifiable respiratory disease or known etiology (12). All of the patients had been consecutively referred by pneumologists after the exclusion of the most common respiratory and oropharyngeal diseases potentially responsible for the cough, according to a comprehensive international diagnostic protocol. The protocol included a detailed medical history, a careful physical examination, formal ear, nose, and throat (ENT) assessment, including laryngoscopic investigation, pulmonary function testing, chest x-ray, and chest and sinus computed tomography and/or magnetic resonance (12). Exclusion criteria were known infectious and systemic disorders, gastrointestinal surgery, structural abnormalities of the gastrointestinal tract, neurological impairment, previous esophageal or gastrointestinal surgery, allergic diseases, immunodeficiency, cardiac disease, suspected inhaled foreign body, suspected airway abnormalities, idiopathic pulmonary fibrosis, and cystic fibrosis. Before the study, none had received acid suppression therapy, such as H2 receptor antagonist and/or proton pump inhibitors, as well as drugs affecting lower esophageal sphincter function (ie, β-adrenergic agonist and theophylline). Patients who smoked were also excluded.
Twenty children (median age 8.2; range 2.5–16.8) with erosive esophagitis, defined according to the Los Angeles classification, were recruited during the study protocol and served as controls.
The study protocol was defined in accordance with the Declaration of Helsinki and was approved by the ethical committee of “La Sapienza” University of Rome. Written consent was obtained from both parents of all of the enrolled patients before the procedures; children older than 12 years signed a statement of assent.
Multichannel Intraluminal Impedance and pH Monitoring
All of the children underwent 24-hour MII-pH monitoring using an ambulatory system (Sleuth, Sandhill Scientific, Highlands Ranch, CO). The characteristics of this system and the performance of recording have been described in detail elsewhere (13).
Reflux was defined by impedance as liquid, gas, and mixed (combined liquid and gas). Liquid reflux was defined as a retrograde 50% fall in impedance on at least the 2 more distal impedance channels. Gas reflux was defined as a rapid increase in impedance >5000 Ω, occurring simultaneously in any 2 consecutive impedance channels. Mixed liquid-gas reflux was defined as gas reflux occurring during or immediately before a liquid reflux. According to the corresponding pH changes, impedance-detected reflux episodes were classified as acid (AR) (pH < 4.0), weakly acidic (WAc) (pH 4–7), and weakly alkaline (WAlk) (P > 7.0) (14). Reflux episodes detected only by pH electrodes were defined as pH-only reflux. The height of a reflux event was defined as proximal if it reached 1 or both of the 2 most proximal impedance channels, intermediate if it reached 1 or both of the 2 middle impedance channels, and distal if it reached the 2 distal impedance channels.
The following variables were analyzed: total number of reflux episodes; number of AR episodes; number of WAc episodes; number of WAlk episodes; height of reflux episodes; percentage of the recording time during which pH is less than 4.0 (ie, esophageal acid exposure time); pH monitoring was deemed abnormal if the esophageal acid exposure during 24 hours exceeded 5%; number of reflux episodes lasting more than 5 minutes per 24 hours; acid clearance time (ACT), defined as the time (in seconds) taken for the esophageal pH to reach pH ≥4 after an AR episode.
The association between MII-detected liquid-containing reflux and cough was evaluated using the symptom association probability (SAP) (15,16). A 2-minute time interval preceding or following the onset of a MII-detected reflux episode was used to delimit the time window for reflux-cough association. The SAP was calculated in a per-individual analysis dividing the 24-hour MII-pH trace into consecutive 2-minute segments. Each period was then assessed for the presence or absence of reflux episodes. Subsequently, each period was labeled as positive or negative for cough depending on whether the symptoms were documented on the diary. If cough occurred within a 2-minute window, either preceding or following, then the period was considered positive for both cough and reflux. Next, a contingency table for each patient was constructed, which included 4 fields in which the number of 2-minute segments with and without cough and with and without reflux were tabulated. Fisher exact test was used for calculating the probability (P) of a falsely significant relation between cough and reflux episodes. At this time, SAP was calculated as (1.0 − P) × 100%, and a value greater than 95% was considered statistically significant (15,16).
The diagnosis of chronic cough–related reflux was achieved on the basis of quantitative or quantitative analysis of MII-pH monitoring. The latter was considered quantitatively abnormal if the esophageal acid exposure was >5%, whereas it was considered qualitatively positive if SAP was >95%. Kruskal-Wallis 1-way analysis of variance by ranks, followed by post hoc analyses using the Dunn multiple comparison test, Wilcoxon signed rank test, χ2 test, or Fisher exact test was used when appropriate. All of the statistical tests were 2-tailed using 0.05 level of significance. Analysis was done by running the Prism software version 4.00 (GraphPad, San Diego, CA). Data were expressed as median (25th–75th), unless otherwise stated.
Sixty-three patients were considered for the trial during a recruitment period of 12 months. Fifteen children were excluded because of either applicability of exclusion criteria or they declined to participate. Three patients were excluded from the study because of technical problems in MII-pH monitoring. Analysis was therefore completed in 45 children (median age 7.8 years, range 1–16). Of those, 24 children had an abnormal test based on quantitative and/or qualitative analysis and were classified as having cough-related reflux (CRR), whereas the remaining 21 were classified as having cough-unrelated reflux (CUR). Twenty children with ERD were recruited during the study period and served as controls. As reported in Table 1, there were no differences among the patient groups in age and sex frequency, but patients with ERD significantly differed from either of the patient groups with cough regarding disease duration (P < 0.001) and type of symptom complaints (P < 0.001). Among children with CRR, 19 (79%) did not show any typical reflux symptoms. Although the upper gastrointestinal endoscopy was not part of the initial protocol, the remaining 5 patients (21%) complaining of regurgitation and/or epigastric pain underwent this procedure, and none revealed mucosal injury or hiatal hernia.
Comparison Among Groups
Table 2 summarizes the results of the 24-hour MII-pH monitoring in the 3 groups of patients. The median number of total reflux episodes was significantly higher in both CRR and ERD groups as compared with the CUR group (P < 0.001), but there was no difference between the former groups. Similarly, in both CRR and ERD groups the median number of AR, WAc, WAlk, and pH-only reflux episodes was significantly higher than in the CUR group, whereas no difference was found between the CRR and ERD groups (Fig. 1). Conversely, no differences were found among the groups in terms of percentage of AR, WAc, WAlk, and pH-only reflux episodes (Fig. 2).
Patients with ERD showed a significantly greater percentage of esophageal acid exposure time, a significantly higher number of long-lasting reflux episodes, and a significantly increased ACT as compared with either patients with CRR (P < 0.05, P < 0.05, and P < 0.05, respectively) or patients with CUR (P < 0.001, P < 0.001, and P < 0.01, respectively) (Table 2). The median values of both esophageal acid exposure time and number of long-lasting reflux episodes were also significantly higher in the CRR group as compared with the CUR group (P < 0.05 and P < 0.01, respectively), whereas although the former group showed an higher ACT, the difference did not reach statistical significance (Table 2).
With regard to reflux migration, the median number of reflux episodes reaching the proximal esophagus was significantly higher in both CRR and ERD groups as compared with the CUR group (P < 0.001), whereas no difference was found between the CRR and ERD groups (Table 3). The subgroup analysis revealed that the median number of AR, WAc, and WAlk episodes reaching the proximal esophagus was significantly higher in both CRR and ERD groups than in the CUR group, and no differences were found among the 3 groups regarding the median number of reflux episodes reaching the mid and the distal esophagus (Table 3).
pH-MII Analysis in Patients With Chronic CRR
Eleven patients (46%) had abnormal esophageal acid exposure time (median 8.4 [7–9.3]) (Table 4). In all but 1 patient cough episodes were detected during 24-hour pH-MII recording; however, this patient showed an abnormal esophageal acid exposure time. A total of 319 cough episodes were analyzed, 56 of which occurred during the night. The majority of patients (15, 62.5%) had cough only during daytime, whereas the remaining 9 had cough during both daytime and nighttime. Patients recorded a median of 11 cough episodes (25th–75th, 5–21). Of 319 cough episodes analyzed, 158 (50%) were considered temporally correlated with reflux episodes. Of these, 105 (66%) involved AR and were detected by both the pH and the impedance sensor, 28 (18%) involved WAc and 25 (16%) involved WAlk. Thus, 53 (34%) CRR episodes were detected by the impedance sensor only and would have been missed by esophageal pH monitoring alone.
Seventeen children had positive SAP and impedance-detected reflux, and 4 of these had abnormal esophageal acid exposure times. Seven children showed positive SAP for AR, and 3 of them had normal esophageal acid exposure time. Five children had positive SAP for both AR and WAc, 4 children for both WAc and WAlk, and only 1 patient for WAlk. All of the children with positive SAP involving WAc and/or WAlk showed normal esophageal acid exposure time.
The relation between respiratory symptoms and GER has long been debated. Although in adults it is widely accepted that the reflux of gastric contents is a cause of chronic cough, to date there are only limited data to support this in children. The main findings of this study were as follows: in a subgroup of children with unexplained chronic cough, otherwise asymptomatic GER is a possible etiologic factor; in children with cough-related GER, there is an increase in both AR and non-AR (ie, WAc and WAlk reflux); and only patients with erosive esophagitis showed an increased esophageal acid exposure time and delayed esophageal acid clearance, clearly demonstrating that the parameter “reflux index” as used in the analysis of esophageal pH studies is not suitable for evaluating reflux-related respiratory symptoms.
In adults, GER has been identified as a cause of chronic cough in patients, being otherwise clinically silent in up to 75% of the cases (2,17). In childhood, several studies have shown a positive association between GER and respiratory symptoms, and the percentage of children with unexplained chronic cough ascribed to AR ranges from 15% to 45% (18–21). Cough as sole manifestation of GER has been described in children rarely. In our study, about 80% of the patients did not complain of typical reflux symptoms such as regurgitation, heartburn, and epigastric pain, corroborating also in childhood the limited value of the typical GER symptom assessment in patients with chronic cough. Recently, Khoshoo et al (21) showed that GER is the single most common factor associated with chronic cough, accounting for 27.5% of the likely etiologic factors. Thus, it is noticeable at least in a subgroup of children that GER can be the sole cause of chronic cough and that chronic cough can be the sole presenting manifestation of GERD.
The majority of the studies addressing the relation between unexplained chronic cough and GER have been performed by using 24-hour pH monitoring, the diagnostic value of which is limited to AR. In the present study, we used MII-pH monitoring, which allows detection of acid, WAc, and WAlk reflux. We found that the patients with CRR showed increased total GER as compared with those classified as having CUR. Approximately half of the CRR group had abnormal distal esophageal acid exposure and 7 patients had positive temporal association between acid GER and cough. Moreover, we identified 5 patients in whom the SAP was positive for both acid and WAc reflux, 4 for both WAc and WAlk reflux and 1 for only WAlk reflux. Our findings are in agreement with previous studies that suggest a role for non-AR in patients with chronic respiratory diseases (13,22,23). Studies in both adults and children show a clear association between cough and non-AR in a significant subgroup of patients with unexplained chronic cough (9–11). Recently, by using a 24-hour ambulatory pressure-pH-impedance monitoring Blondeau et al (11) found that of 198 cough episodes preceded by reflux episodes, 41% involved AR, 55% WAc reflux, and only 4% WAlk reflux. Moreover, 10 of 26 patients of this series had positive SAP, 1 for AR, 6 for WAc reflux, and 3 for both acid and WAc reflux (11). Although WAlk reflux-cough sequences were found, they suggested that WAc reflux was unlikely to precede the cough. The reason for this difference in the role of WAc reflux is unclear; however, in our series a higher number of children had positive SAP (17/24). Moreover, the reflux-cough association may subject to a day-to-day variability. Larger studies are required to clarify this issue.
A high proportion of children with CRR exhibited a normal esophageal acid exposure time despite a positive association between cough and acid, WAc, and WAlk reflux. Hence, esophageal acid exposure time is a poor correlate of CRR. In addition, more than 40% of the patient group showing a positive SAP involving non-AR would have been missed by 24-hour pH monitoring alone. It has been speculated that the use of symptom indices has several limitations. First, because symptom indices are a statistical reflection, a positive symptom association does not necessarily imply causality between 2 parameters. Furthermore, they are based on the defined time window length for defining a symptom as reflux related, and although we used the widespread accepted 2-minute time window, it may not be optimal for assessment of symptoms such as cough. We used in our study the most appropriate method to characterize the association between reflux and cough, and a positive SAP (>95%) indicates that the observed relation between cough and GER is not brought about by chance (15,16).
When we sought to find out whether children with cough-related GER had different reflux patterns as compared with children with erosive esophagitis, we found that the latter group showed a significantly higher esophageal acid exposure time as well as an increased number of long-lasting AR episodes (ie, >5 minutes). In addition, children with ERD showed a significantly increased acid clearance time. Among patients with cough-related reflux, only 5 (21%) with typical GER symptoms underwent upper gastrointestinal endoscopy, and none of them showed mucosal injury. Several studies have shown that the majority of patients with GER-associated respiratory manifestation do not have esophagitis at endoscopy (3,24,25). Furthermore, studies in adults reported a progressive increase in both esophageal acid exposure time and number of long-lasting reflux with increased mucosal injury, and this abnormality paralleled the esophageal motility dysfunctions, which in turn affect the clearance of refluxed gastric contents from the esophagus (26–28).
Patients with CRR showed a higher median number of reflux episodes reaching the proximal esophagus than those classified as having CUR, suggesting that the proximal extent of refluxate could be a key event in symptom production. We found no differences between cough-related GER patients and ERD patients. Hence, although prolonged acid exposure time caused by reduced acid clearance provides an explanation why some patients develop esophagitis and others do not, our data suggest that the proximal propagation of reflux episodes does not distinguish patients who develop respiratory symptoms such as cough from those with more typical esophageal symptoms. It is likely that mechanisms other than the type or the proximal extension of the reflux episodes, such as the volume of refluxate, the degree of esophageal distension, and the activation of different types of mechanoreceptors, also may be involved in the generation of different types of symptoms (6,7).
Our study has some limitations. First, we did not perform simultaneous gastroesophageal pressure monitoring, which allows the precise analysis of temporal relation between cough and reflux. It is likely that sequences of cough-preceding reflux have been missed. Blondeau et al (11) showed that, overall, 84% of the event markers corresponded to cough episode detected using manometry, and there was a significant correlation between the number of event marker and the number of cough episodes recorded by manometry (r = 0.73, P < 0.001). Moreover, from all of the marker events that did not correspond to a manometric cough burst, 83% of them occurred in the same child. Second, the results of MII-pH monitoring are still difficult to interpret given the absence of normal pediatric reference values.
In conclusion, our data provide evidence that at least in a subgroup of children with unexplained chronic cough asymptomatic GER is an etiologic factor. Although AR is the main determinant in the genesis of cough-related GER, in a subgroup of children chronic cough is clearly associated with WAc and WAlk reflux. MII-pH monitoring with symptom association analysis increases the likelihood of demonstrating a temporal association between the cough and all types of reflux. Finally, the mechanisms underlying the pathogenesis of cough-related GER deserve to be investigated further.
1. Marchant JM, Newcombe PA, Juniper EF, et al. What is the burden of chronic cough for families? Chest 2008; 134:303–309.
2. Irwin RS. Chronic cough due to gastroesophageal reflux disease: AACP evidence-based clinical practice guidelines. Chest 2006; 19:80S–94S.
3. Vandenplas Y, Rudolph CD, Di Lorenzo C, et al. Pediatric gastroesophageal reflux clinical practice guidelines: joint recommendations of the North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition (NASPGHAN) and the European Society for Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN). J Pediatr Gastroenterol Nutr 2009; 49:498–547.
4. Sherman PM, Hassall E, Fagundes-Neto U, et al. A global, evidence-based consensus on the definition of gastroesophageal reflux disease in the pediatric population. Am J Gastroenterol 2009; 104:1278–1295.
5. Tolia V, Vandenplas Y. Systematic review: the extra-oesophageal reflux symptoms of gastroesophageal reflux disease in children. Aliment Pharmacol Ther 2009; 29:258–272.
6. Galmiche JP, Zerbib F, Bruley des Varannes S. Review article: respiratory manifestations of gastro-oesophageal reflux disease. Aliment Pharmacol Ther 2008;27:449–64.
7. Sifrim D, Mittal R, Fass R, et al. Review article: acidity and volume of the refluxate in the genesis of gastro-oesophageal reflux disease symptoms. Aliment Pharmacol Ther 2007; 25:1003–1017.
8. van Wijk MP, Benninga MA, Omari TI. Role of the multichannel intraluminal impedance technique in infants and children. J Pediatr Gastroenterol Nutr 2009; 48:2–12.
9. Sifrim D, Dupont LJ, Blondeau K, et al. Weakly acidic reflux in patients with chronic unexplained cough during 24 hour pressure, pH, and impedance monitoring. Gut 2005; 54:449–454.
10. Blondeau K, Dupont LJ, Mertens V, et al. Improved diagnosis of gastroesophageal reflux in patients with unexplained chronic cough. Aliment Pharmacol Ther 2007; 25:723–732.
11. Blondeau K, Mertens V, Dupont L, et al. The relationship between gastroesophageal reflux and cough in children with chronic unexplained cough using combined impedance-pH-manometry recordings. Pediatr Pulmunol 2010 Oct 22 [Epub ahead of print].
12. Chang AB, Glomb WB. Guidelines for evaluating chronic cough in pediatrics. AACP evidence-based clinical practice guidelines. Chest 2006; 19:260S–283S.
13. Borrelli O, Battaglia M, Galos F, et al. Non-acid gastro-oesophageal reflux in children with suspected pulmonary aspiration. Dig Liv Dis 2010; 42:115–121.
14. Sifrim D, Castell D, Dent J, et al. Gastro-oesophageal reflux monitoring: review and consensus report on detection and definitions of acid, non-acid, and gas reflux. Gut 2004; 53:1024–1031.
15. Wuesten BL, Roelofs JM, Akkermans LM, et al. The symptom-association propability: an improved method for symptom analysis of 24-hour esophageal pH data. Gastroenterology 1995; 107:1741–1745.
16. Bredenoord AJ, Wuesten BL, Smouth AJ. Symptom association analysis in ambulatory gastroesophageal reflux monitoring. Gut 2005; 54:1810–1817.
17. Irwin RS, French CL, Curley FJ, et al. Chronic cough due to gastroesophageal reflux. Clinical, diagnostic, and pathogenic aspect. Chest 1993; 104:1511–1517.
18. Holinger LD, Sanders AD. Chronic cough in infants and children: an update. Laryngoscope 1991; 101:596–605.
19. Jain A, Patwari AK, Bajaj P, et al. Association of gastroesophageal reflux in young children with persistent respiratory symptoms. Trop J Pediatr 2002; 48:39–42.
20. Thilmany C, Beck-Ripp J, Griese M. Acid and non-acid gastro-esophageal refluxes in children with chronic pulmonary diseases. Respir Med 2007; 101:969–976.
21. Khoshoo V, Edell D, Mohnot S, et al. Associated factors in children with chronic cough. Chest 2009; 136:811–815.
22. Rosen R, Nurko S. The importance of multichannel intraluminal impedance in the evaluation of children with persistent respiratory symptoms. Am J Gastroenterol 2004; 99:2452–2458.
23. Condino AA, Sondheimer J, Pan Z, et al. Evaluation of gastroesophageal reflux in pediatric patients with asthma using impedance-pH monitoring. J Pediatr 2006; 149:216–219.
24. Locke GR III, Talley NJ, Fett SL, et al. Prevalence and clinical spectrum of gastroesophageal reflux disease: a population-based study in Olmsted County, Minnesota. Gastroenterology 1997; 112:1448–1456.
25. Vandenplas Y. Asthma and gastroesophageal reflux disease. J Pediatr Gastroenterol Nutr 1997; 24:89–99.
26. Cucchiara S, Staiano A, Di Lorenzo C, et al. Esophageal motor abnormalities in children with gastroesophageal reflux disease and peptic esophagitis. J Pediatr 1986; 108:907–910.
27. Wu JCY, Cheung CMY, Wong VWS, et al. Distinct clinical characteristic between patients with nonerosive reflux disease and those with reflux esophagitis. Clin Gastroenterol Hepatol 2007; 5:690–695.
28. Chen CL, Yi CH, Cook IJ. Differences in oesophageal bolus transit between patients with and without erosive reflux disease. Dig Liv Dis 2008; 40:348–354.