© 2002 American Thoracic Society
Effects of Bronchial Obstruction on Lower Esophageal Sphincter Motility and Gastroesophageal Reflux in Patients with AsthmaService d'Hépato-gastroentérologie, Hôpital Saint-André, Bordeaux; Service des Maladies Respiratoires, Hôpital du Haut-Lévêque, Pessac; INSERM U 539, CHU Hôtel Dieu, Nantes; and INSERM E9937, Université Victor Segalen Bordeaux 2, Bordeaux, France Correspondence and requests for reprints should be addressed to Frank Zerbib, M.D., Ph.D., Service d'Hépato-gastroentérologie, Hôpital Saint-André, F-33075 Bordeaux Cédex, France. E-mail: frank.zerbib{at}chu-bordeaux.fr
The relationship between gastroesophageal reflux and asthma remains unclear. The aim of this study was to analyze the effect of bronchial obstruction on lower esophageal sphincter (LES) motility and reflux in patients with asthma. LES motility and esophageal pH were assessed in eight subjects with intermittent asthma and eight healthy volunteers during three consecutive 30-minute periods: baseline, methacholine-induced bronchospasm, and after inhalation of the ß2-agonist salbutamol. Healthy subjects inhaled 2 mg of methacholine, whereas subjects with asthma inhaled the dose of methacholine causing a 15% fall in FEV1, as determined by a previous methacholine challenge. LES motility, esophageal pH, and FEV1 were not significantly different between the three periods in healthy subjects. In patients with asthma, methacholine induced a 21.9 ± 2.6% decrease in FEV1 and a concomitant increase in the rate of transient LES relaxation (TLESR) and reflux episodes. Inhalation of salbutamol decreased the rate of TLESRs but not the number of reflux episodes. We conclude that in patients with asthma, methacholine-induced bronchospasm increases the rate of TLESR and the number of reflux episodes. These results support the belief that, in asthma, bronchial obstruction may be responsible for reflux or may aggravate reflux through a mechanism that remains to be further clarified.
Key Words: asthma • gastroesophageal reflux • lower esophageal sphincter • bronchoconstriction • manometry
Gastroesophageal reflux and asthma are very frequently associated. Esophagitis has been found in up to 39% of patients with asthma (1), and 82% seem to have abnormal esophageal-pH testing (2). The relationship between asthma and reflux is potentially complex, but the prevailing view is that gastroesophageal reflux could trigger or aggravate asthma. Indeed, the role of gastroesophageal reflux has been suggested in many experimental studies showing that gastroesophageal reflux can induce or exacerbate airflow obstruction in patients with asthma through a vagally mediated reflex and/or microaspiration of gastric contents (3). This concept is consistent with the benefit of potent acid suppression with proton pump inhibitors (4) or even antireflux surgery (5), which have been reported to decrease drug requirements in nearly 75% of patients. However, the relationship between gastroesophageal reflux and asthma remains poorly understood (6, 7). Conflicting results have been reported regarding the effects of esophageal acidification on pulmonary function (8), and association between gastroesophageal reflux and bronchial symptoms or peak expiratory flow reduction has rarely been demonstrated (9). In addition, most clinical data have been obtained from nonrandomized studies, and the results of a recent placebo-controlled trial with omeprazole show that only 35% of patients with asthma having gastroesophageal reflux disease experienced respiratory improvement after 8 weeks of treatment (10). The high prevalence of gastroesophageal reflux in patients with asthma may be the consequence of asthma itself, but data supporting this hypothesis are scarce and are not convincing. During an asthma attack, negative pleural pressure increases the transdiaphragmatic pressure gradient and may thus facilitate gastroesophageal reflux. Likewise, thoracic distention and air trapping may also result in an impairment of the diaphragmatic function, which is believed to be an important component of the antireflux barrier (11). However, these putative mechanisms have never been established. In patients with asthma, Ekstrom and Tibbling (12) have reported the lack of effect of histamine-induced bronchospasm on gastroesophageal reflux, whereas Moote and coworkers (13) found an increased number of reflux episodes during methacholine-induced bronchospasm. In the latter study, the authors failed to detect any effect of bronchospasm on resting lower esophageal sphincter (LES) pressure, so the mechanisms of asthma-induced reflux remained unclear (13). In fact, it is now clearly established that most reflux episodes in healthy subjects, as well as in individuals with gastroesophageal reflux disease, result from transient LES relaxations (TLESRs), which are spontaneous long-lasting relaxations of the LES not related to swallowing (14–17). Transient increase in intra-abdominal pressure that overcomes the resistance of the gastroesophageal junction (i.e., "stress reflux") and spontaneous reflux episodes because of a permanently low LES pressure (i.e., "free reflux") represent less frequent motor patterns responsible for reflux, mainly observed in patients with esophagitis or severe disease (18). The aim of this study was thus to determine the effect of bronchial obstruction induced by methacholine on LES motility and the occurrence of gastroesophageal reflux in both healthy subjects and patients with asthma.
Subjects Eight patients suffering from intermittent asthma (five males; mean age, 23 years) and eight healthy volunteers (six males; mean age, 22 years) gave their informed written consent to participate in the study. The protocol had been approved previously by the local Human Ethics Committee (Comité Consultatif pour la Protection des Personnes dans la Recherche Biomédicale Bordeaux A). No subjects were taking medications or had a history of gastrointestinal symptoms (especially reflux symptoms) or prior abdominal surgery, except appendectomy. Asthma was defined according to an American Thoracic Society statement (19), and its severity was assessed according to international guidelines (20). Patients were treated with short-acting ß2-agonists on an as-required basis and had not received any inhaled or systemic corticosteroid for the last 3 months. Healthy subjects had no history of respiratory symptoms or disease and were taking no medication.
Study Design On a separate day, the effects of inhaled methacholine on LES motility and gastroesophageal reflux were assessed. Each subject fasted for at least 8 hours before the study, and subjects with asthma did not use inhaled bronchodilators during the previous 6 hours. An esophageal motility catheter and a pH electrode were inserted through the nose. After 15 minutes for stabilization, LES motility (i.e., resting LES pressure and TLESRs) and esophageal pH were monitored for a 30-minute baseline period. In patients with asthma, baseline FEV1 was then determined, and the calculated dose of methacholine supposed to induce a 15% fall in FEV1 (PC15) was delivered using a Mefar dosimeter (Mefar, Bovezzo, Italy). When a 15% decrease in FEV1 was not reached, the patients were made to inhale the same dose of methacholine again to obtain a significant obstruction. By contrast, all control subjects inhaled 2 mg of methacholine without any alteration in the FEV1 value. Then, esophageal pH and LES motility were monitored for a second 30-minute period. During this period, the tolerance of bronchospasm was assessed clinically (wheezing) and by FEV1 measurements repeated every 10 minutes. In subjects with asthma, when FEV1 increased above 15% of baseline, the previous dose of methacholine (PC15) was inhaled again to maintain the obstruction. In control subjects, FEV1 was determined to verify the absence of a significant bronchospasm during this period. At the end of this period, subjects from both groups inhaled 200 µg of salbutamol through a spacer (Volumatic; GlaxoSmithKline, Marly-le Roi, France), and once FEV1 measurements had demonstrated the return to baseline values, esophageal recordings were started for a third 30-minute period.
Lung Function and MC Bronchial responsiveness was analyzed using an MC as described in a European Respiratory Society statement (21). Briefly, subjects were advised to avoid using any ß2-agonist for 8 hours before the test. Subjects were permitted nine attempts to provide at least two technically acceptable maneuvers. All those whose FEV1 was at least 80% predicted, and more than 1.5 L, were invited to undergo MC unless they reported that they had heart disease, epilepsy, were pregnant or breastfeeding, or were taking a ß-blocker. Bronchial challenge commenced with inhalation of saline diluent, and the maximum postdiluent FEV1 recorded 2 minutes later was used as the control value. Those whose control FEV1 was less than 90% of the baseline value were not challenged further. The Mefar dosimeter (Mefar) was set to deliver methacholine over a 1-second period. Those who denied respiratory symptoms suggestive of asthma received methacholine starting at a cumulative dose of 0.0078 mg and then in quadrupling doses (short schedule) until a 10% fall in FEV1 from the control value was recorded, after which doubling doses were used. All other subjects received an initial dose of 0.00195 mg and then received doubling doses (long schedule). Two minutes after each inhalation, subjects had up to five attempts to achieve two technically acceptable maneuvers. PC20 (i.e., the provocative dose of agonist corresponding to a 20% fall in FEV1) was calculated using multiple logistic regression on the variables. The mean PC20 in subjects with asthma was 0.6 ± 0.2 mg (0.08–1.2 mg). The baseline pulmonary function of control subjects and subjects with asthma is presented in Table 1 . Although FEV1/FVC was significantly lower in subjects with asthma than in control subjects, the difference in FEV1 fell short of statistical significance, thus reflecting the relatively mild severity of the disease.
LES Motility Recordings Our technique of esophageal motility recording has been extensively described elsewhere (22). A standard motility catheter fitted with a 6-cm sleeve (Dentsleeve PTY Ltd., Parkside, Australia) was used to monitor LES and esophageal pressure. The assembly was introduced through a nostril, swallowed, and positioned so that pressure could be recorded from the LES (sleeve), fundus (2 cm below the sleeve), esophageal body (side holes 5, 10, and 15 cm proximal to the sleeve), and pharynx (side hole 28 cm proximal to the sleeve to detect swallowing). The catheter was infused at 0.5 ml/minute using a low-compliance hydraulic capillary infusion system (Arndorfer Medical Specialties Inc., Milwaukee, WI) driven by a pressure head of nitrogen. The infusion system was connected to pressure transducers. Signals were recorded on a polygraph (Synectics Medical, Stockholm, Sweden), digitized, computer-processed, and stored using commercially available software (Polygram; Synectics Medical). All esophageal recordings were performed with subjects in sitting position.
Esophageal pH Monitoring
Data Analysis
LES motility.
Reflux episodes.
Statistics
Pulmonary Function In control subjects, methacholine never induced wheezing or significant FEV1 decrease as indicated in Figure 1 . The average FEV1 decrease during this period was 3.3 ± 1.0% of baseline. In this group, salbutamol did not significantly affect pulmonary function (1.9 ± 1.4% FEV1 increase from baseline).
In subjects with asthma, the average dose of methacholine inhaled to provoke a significant obstruction (> 15%) was only 0.5 ± 0.1 mg. After the first inhaled PC15 (0.4 ± 0.1 mg), two subjects needed a second inhalation of the same dose to obtain a FEV1 fall of at least 15% during the whole 30-minute period. All the other subjects performed the same numbers of forced expiratory maneuvers. Overall, methacholine induced a 21.9 ± 2.6% FEV1 decrease from baseline (Figure 1). Despite a mild wheezing in all patients, this mild bronchial obstruction was well tolerated by all the subjects. Salbutamol abolished wheezing and normalized pulmonary function with respect to FEV1 (3.8 ± 0.2 versus 3.7 ± 0.2 L during baseline, nonsignificant).
LES Motility
In control subjects, neither the mean resting LES pressure (Figure 2) nor the rate (Figure 3) and duration (Table 2) of TLESRs was affected by methacholine and salbutamol.
In patients with asthma, an increase in the rate of TLESRs (3.5 ± 0.6 versus 1.0 ± 0.5 per 30 minutes, p = 0.005) was observed after methacholine inhalation, whereas the mean LES pressure remained unchanged. Inhalation of salbutamol significantly decreased the rate of TLESRs (1.6 ± 0.3 per 30 minutes, p = 0.006 versus methacholine period, nonsignificant versus baseline) but did not affect the mean resting LES pressure. In subjects with asthma, the duration of TLESRs was not affected by methacholine and salbutamol (Table 2).
Gastroesophageal Reflux
In control subjects, neither methacholine nor salbutamol inhalation affected esophageal pH. In subjects with asthma, inhalation of methacholine significantly increased the number of reflux episodes as compared with baseline (0.9 ± 0.4 versus 0.3 ± 0.2 reflux per 30 minutes, p < 0.05) (Figure 4), but the duration of reflux episodes and time at pH below 4 did not change significantly (Table 2). After inhalation of salbutamol, the number of reflux episodes remained elevated (1.0 ± 0.4 episodes per 30 minutes, p = 0.02 versus baseline, nonsignificant versus methacholine period) (Figure 4), whereas the duration of reflux episodes and time at pH below 4 (Table 2) were not significantly changed.
Mechanisms of Gastroesophageal Reflux In subjects with asthma, during baseline, one episode of stress reflux and one of free reflux were observed. All other reflux episodes occurred during TLESRs. After methacholine inhalation, the increase in acid reflux number was clearly related to an increase in TLESR rate, but the proportion of TLESRs associated with reflux remained non–statistically significant. After salbutamol inhalation, the persistent increase in acid reflux number was related to a significantly higher proportion of TLESRs associated with reflux (54%) as compared with baseline (0%) and methacholine (25%, p < 0.03) periods.
The high prevalence of gastroesophageal reflux in patients with asthma has been widely reported (1, 2), but the relationships between these two conditions remain unclear, especially with respect to the impact of asthma on gastroesophageal reflux. In the present study, we demonstrate for the first time that, in asthma, bronchial obstruction elicits an increase in the rate of TLESRs, an effect that is reversed by the ß2-agonist salbutamol. In addition, the number of reflux episodes also increased after methacholine inhalation, but this effect was not reversed by salbutamol. For ethical and practical reasons, studies that use MC are difficult to conduct in healthy volunteers as well as in patients with asthma. In this respect, our study protocol was limited to three consecutive relatively short periods. Likewise, it was not possible to investigate the effects of salbutamol alone because patients involved in the present study had asthma of mild severity and required an MC to experience a significant obstruction. Therefore, methacholine and salbutamol periods could not be studied in different sets of experiments and had to be studied successively. These practical difficulties probably explain why data concerning the effects of bronchial obstruction on LES motility in humans are scarce. Our results on LES motility are consistent with the results of Moote and coworkers (13) regarding the lack of effect of methacholine on LES resting pressure. Our study was performed with a motility catheter fitted with a 6-cm sleeve (24) that allows prolonged recordings of LES motility and therefore provides additional data concerning TLESRs. The change in the rate of TLESRs is likely to be related to methacholine-induced bronchospasm and not to methacholine per se. Because atropine has been shown to inhibit TLESRs (25), the observed increase in TLESR rate may in theory be related to a pharmacologic effect of methacholine, which is a cholinergic agonist. However, this hypothesis can be ruled out because no significant effect was observed in control subjects, who received higher doses of methacholine than subjects with asthma. Moreover, methacholine did not induce any significant change in control subjects' pulmonary function but provoked a significant bronchial obstruction (21.9 ± 2.6% FEV1 decrease) in subjects with asthma. Finally, during the third period, after reversal of the bronchospasm by salbutamol, TLESR rate returned to baseline levels. In our study performed during fasting, the methacholine-induced increase in reflux episodes was statistically significant as compared with baseline, despite the relatively small number of subjects with asthma and the short duration of periods. As for TLESRs, the increase in reflux episodes seems rather related to bronchial obstruction than to the pharmacologic effect of methacholine because no change in esophageal pH was observed in control subjects, who did not experience significant bronchospasm. Again, our results are in agreement with those reported by Moote and coworkers (13), who applied a very similar study protocol. By contrast, Ekstrom and Tibbling (12) found that gastroesophageal reflux was not more pronounced during the provoked bronchospasm period. The different experimental conditions (e.g., histamine-induced bronchospasm and postprandial period) may account for these discrepancies. Surprisingly, in patients with asthma, the number of reflux episodes after salbutamol inhalation remained significantly higher than during baseline (Figure 4) despite the TLESR rate returning to baseline levels. This persistent increase in reflux episodes was related to a higher proportion of TLESRs associated with reflux. These findings are not easy to interpret, and different factors could interfere. A pharmacologic effect of salbutamol on LES motility cannot be excluded, but such bronchodilator therapy was not shown to affect either the number of reflux episodes or esophageal motility in healthy subjects (26, 27) and patients with asthma (27). However, we cannot exclude an effect of salbutamol on esophageal tonic motor activity. Indeed, Sifrim and coworkers (28) have suggested that inhibition of esophageal body tonic contractility during TLESRs may contribute to a higher prevalence of reflux. Another potential factor that may facilitate reflux may be related to the volume and pH of the gastric contents available for reflux during the third period of the tests. Indeed, salbutamol was administered after cholinergic stimulus (i.e., MC) and infusion of approximately 200 ml of water through the motility catheters into the stomach. However, the systemic effect of inhaled methacholine has, to our knowledge, never been reported so far, and it is likely that a large proportion of the water perfused has already emptied during the first hour of the test. Although our results strongly support the concept that bronchial obstruction can trigger or aggravate gastroesophageal reflux through an increase in TLESR rate, the mechanisms remain to be elucidated. It is well established that TLESRs are triggered by gastric distention (29) and by pharyngeal sensory stimulation (30, 31). In addition, meal ingestion represents another potent stimulus of TLESRs and reflux episodes (16, 32, 33). TLESRs are believed to be neurally mediated through the involvement of a vago-vagal reflex stimulated by receptors located mainly in the fundus (34) and the pharyngolarynx (35). Afferent fibers stimulated by gastric distention and pharyngeal stimulation are known to project via the vagus to the nucleus tractus solitarius and to the dorsal motor nucleus of the vagus (36). The inhibitory efferent pathway for TLESRs is also probably the vagus because TLESRs are completely abolished by cooling of the cervical vagus nerve (37). It has been suggested that the intermittent occurrence of TLESRs could be ascribed to a pattern generator in the brain stem that responds with intermittent output according to the level of input (35). A similar mechanism may be involved in the bronchospasm-induced TLESRs because there is both neurophysiologic (38) and neuroanatomic (36) evidence that pulmonary vagal afferents terminate within regions of nucleus tractus solitarius. During bronchospasm, bronchial and/or pulmonary mechanoreceptors may be stimulated and elicit TLESRs through these afferent pathways. A preliminary report supports this hypothesis, demonstrating that airway obstruction in cats induces LES relaxations that partly resemble TLESRs (39). Therefore, bronchospasm may represent a potent trigger for TLESRs, at least during the fasted state. The results of our study have been obtained under experimental conditions, and their clinical relevance remains to be confirmed. Bronchoconstriction may contribute, at least in part, to the high prevalence of gastroesophageal reflux in patients with asthma. Moreover, if reflux triggers asthma and asthma causes reflux, a vicious circle could arise with an increase in the severity of asthma. Therapy aimed at the restoration of bronchial diameter (e.g., with ß-agonists) may finally be less detrimental than previously believed or may be even beneficial with regard to gastroesophageal reflux. However, TLESRs represent only one mechanism potentially involved in the complex relationships between asthma and gastroesophageal reflux disease, and other abnormalities may have pathophysiologic or pharmacologic relevance.
Supported by grants from the Centre Hospitalier Universitaire de Bordeaux (PHRC Appel d'offre interne, 1998) and Institut Pneumologique d'Aquitaine. Received in original form October 17, 2002; accepted in final form January 21, 2002
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
INTRODUCTION
