Oral Corticosteroids Increase Esophageal Acid Contact Times in Patients With Stable Asthma*

Transcription

1 preliminary report Oral Corticosteroids Increase Esophageal Acid Contact Times in Patients With Stable Asthma* John P. Lazenby, MD; Melany R. Guzzo, BSN; Susan M. Harding, MD, FCCP; Patricia E. Patterson, MD; Lawrence F. Johnson, MD; and Laurence A. Bradley, PhD Study objectives: The prevalence of gastroesophageal reflux disease (GERD) is higher in people with asthma than in control populations. Predisposing factors for GERD development may include asthma medications such as prednisone. The objective of this study was to determine whether prednisone alters GERD parameters in people with asthma. Design: Prospective, single-blinded, placebo-controlled, crossover study. Setting: University medical center clinic. Participants: Twenty adults with stable, moderate persistent asthma with minimal esophageal reflux symptoms (less than three times a week) who were not receiving antireflux therapy. Intervention: Prednisone, 60 mg/d, for 7 days. Measurements and results: Asthma, esophageal reflux symptoms, and spirometry were measured during baseline, placebo, and prednisone phases, each 7 days in duration. Dual-probe esophageal ph monitoring, esophageal and respiratory manometrics (20 subjects), and basal and stimulated gastric acid secretion (4 subjects) were measured after placebo and prednisone phases. There were significant increases in esophageal acid contact times at the distal and proximal ph probes during the prednisone phase. Total percentage of time that ph was < 4.0 at the distal probe was % for placebo compared with % for prednisone (p < 0.002). Total percentage of time that ph was < 4.0 at the proximal probe was % for placebo and % for prednisone (p < ). There were no significant changes in subject weight, spirometry, asthma or esophageal reflux symptoms, manometrics, or basal or stimulated gastric acid secretion. Conclusion: Prednisone, 60 mg/d for 7 days, increased esophageal acid contact times in this small population of people with stable asthma; however, the mechanism for this finding is unclear. (CHEST 2002; 121: ) Key words: asthma; corticosteroids; esophageal ph monitoring; gastroesophageal reflux Abbreviations: ACG American College of Gastroenterology; FEF 25 75% forced expiratory flow during the middle half of FVC; FEV 1% FEV 1 /FVC; GERD gastroesophageal reflux disease; LES lower esophageal sphincter; PEF peak expiratory flow; UES upper esophageal sphincter Asthma is characterized by inflammation and heightened airway responsiveness ignited by allergic and nonallergic triggers. Gastroesophageal reflux disease (GERD) is a potential asthma trigger, *From the Division of Pulmonary, Allergy, and Critical Care Medicine (Drs. Lazenby, Harding, and Patterson, and Ms. Guzzo), the Division of Gastroenterology and Hepatology (Dr. Johnson), and the Division of Clinical Immunology and Rheumatology (Dr. Bradley), Department of Medicine, University of Alabama at Birmingham, Birmingham, AL. Presented in part at the American Thoracic Society International Meeting, May 10, 2000, Toronto, ON, Canada. Supported by Astra-Zeneca LP and National Institutes of Health- National Heart, Lung, and Blood Institute Sleep Academic Award Grant No. HL and acid suppressive therapy may improve asthma outcome in selected patients. 1 Gastroesophageal reflux symptom prevalence is higher in people with asthma compared with control populations, with 77% of people with asthma experiencing heartburn and 55% experiencing regurgitation (p 0.05). 2 Sontag et al 3 examined esophageal acid contact times Manuscript received November 8, 2000; revision accepted June 13, Correspondence to: Susan M. Harding, MD, FCCP, Associate Professor of Medicine, Division of Pulmonary, Allergy, and Critical Care Medicine, University of Alabama at Birmingham, 1900 University Blvd, THT Room 215, Birmingham, AL 35294; sharding@uab.edu CHEST / 121 / 2/ FEBRUARY,

2 using 24-h esophageal ph testing in 104 consecutive subjects with asthma and in 44 control subjects, and noted that people with asthma had higher esophageal acid contact times, more frequent reflux episodes, and longer esophageal acid clearance times compared with control subjects. They also noted that 82% of people with asthma had abnormal esophageal acid contact times consistent with the diagnosis of GERD. 3 Furthermore, in 186 consecutive, unselected people with asthma, 39% had biopsy-proven esophagitis. 4 One potential factor promoting GERD in patients with asthma is asthma medication. 5 Theophylline decreases lower esophageal sphincter (LES) pressure, increases gastric acid secretion and esophageal acid contact times, and may increase esophageal reflux symptoms. 6 9 Treatment with oral agonists, but not inhaled -agonists, increases esophageal reflux symptoms There are no published data examining the effect of oral corticosteroids on esophageal reflux parameters. In our clinical practice, we noticed that patients with asthma complain of new or worsening esophageal reflux symptoms when treatment with oral corticosteroids is initiated. These symptoms decrease as the corticosteroid dose is tapered or withdrawn. Also, Irwin et al 13 noted that GERD was an important factor in patients with difficult-to-control asthma, defined as people who require at least 10 mg of prednisone every other day. Is it possible that prednisone may have played a role in GERD development in these patients and, once GERD was present, GERD had an impact on airway reactivity, such that the asthma was difficult to control? A previous study 14 noted that esophageal acid augments airway reactivity. Furthermore, Strickland et al 15 noted in healthy subjects that 1 month of treatment with prednisolone, 20 mg/d, resulted in an increase in stimulated gastric acid secretion that could, theoretically, promote GERD. With this rationale, we hypothesized that oral corticosteroids could alter esophageal reflux parameters including esophageal acid contact times. This study examines the effect of 60 mg/d of oral prednisone for 7 days on esophageal acid contact times and esophageal reflux symptoms in 20 patients with stable asthma in a single-blinded, placebo-controlled trial. Respiratory and esophageal manometrics and basal and stimulated gastric acid secretion were evaluated to examine potential mechanisms of altered esophageal acid contact times. Subjects Materials and Methods Twenty adults with asthma who gave informed consent participated in this prospective study approved by the Institutional Review Board at the University of Alabama at Birmingham. All subjects were nonsmokers and had moderate, persistent asthma as defined by the Second Expert Panel for the Diagnosis and Management of Asthma and by the American Thoracic Society criteria, including: (1) a 12% improvement in the FEV 1 after bronchodilation, or a 20% decrease in FEV 1 after bronchoprovocation with methacholine; and (2) the absence of symptoms of chronic bronchitis or other forms of chronic lung disease Asthma symptoms and peak expiratory flow (PEF) rates were stable for at least 1 week before study enrollment, and all subjects had not received systemic corticosteroids for at least 3 months. Inhaled corticosteroids and other medications were continued throughout the study. Study enrollment was limited to subjects with minimal esophageal reflux symptoms (heartburn, indigestion, belching, and regurgitation occurring not more than three times a week), and no subject used histamine type 2-receptor antagonists, proton pump inhibitors, or prokinetic agents before study entry. Exclusion criteria included age 18 years or 70 years; history of Zollinger-Ellison syndrome; esophageal motility disorder; advanced heart disease; history of esophageal, gastric, pulmonary, or vagus nerve surgery; or history of GI anastomosis. Other exclusion criteria included the following: diabetes mellitus; BP 169/100 mm Hg; serum potassium 3.2 meq/l, or serum glucose 140 meq/l; pregnant or lactating women, or women of child-bearing potential not using reliable means of contraception; warfarin, phenytoin, or investigational drug use; previous significant side-effects from prednisone use; and inability to give informed consent. Study Design A single-blinded, placebo-controlled, crossover study was performed to examine the effect of 7 days of treatment with prednisone, 60 mg/d, on esophageal acid contact times, respiratory and esophageal manometrics, and esophageal reflux symptoms. The prednisone treatment phase followed the placebo phase so that potential treatment phase effects of the prednisone phase would not linger into the placebo phase. Helicobacter pylori serology was tested in the final 12 consecutive subjects who entered the study. In the final four patients, basal and stimulated gastric acid secretions were analyzed using Pentagastrin (Cambridge Laboratories; New Castle-on-Tyne, UK). With the exception of the esophageal testing period, subjects were administered antacid tablets if esophageal reflux symptoms occurred during the study period. Figure 1 outlines the study protocol. Esophageal and respiratory manometrics, esophageal ph monitoring, and basal and stimulated gastric acid secretion parameters were analyzed in a blinded manner with respect to subject number and treatment phase. During the baseline phase (days 1 through 7), subjects monitored their asthma, esophageal reflux symptoms, and PEF rates. On day 7, a baseline spirogram, physical examination, and laboratory studies (serum glucose, serum potassium, H pylori serology, and, in female subjects, a urine pregnancy test) were performed. During the placebo phase (days 8 through 14), subjects received one placebo capsule each morning for 7 days and continued to monitor their symptoms and PEF rates. On day 14, subjects underwent a physical examination, esophageal and respiratory manometrics, and 24-h dual probe esophageal ph monitoring. In four subjects, basal and stimulated gastric acid secretions were also measured the morning after esophageal ph monitoring was completed on day 15. During the prednisone phase (days 15 through 22), subjects were supplied with identical-appearing capsules containing prednisone, 60 mg, and instructed to ingest one capsule each morning 626 Preliminary Report

3 Figure 1. Flow diagram of the study design along a time axis showing baseline, placebo, and prednisone phases. PEFR PEF rate; a.m. morning; pts patients. for 7 days while continuing to monitor their symptoms and PEF rates. On day 21, spirometry, physical examination, esophageal and respiratory manometrics, and esophageal ph monitoring were repeated, and symptom diaries were collected. Once again, basal and stimulated gastric acid secretion analysis was performed in four subjects on day 22. Asthma and GERD Questionnaire Daytime asthma and esophageal reflux symptoms were scored according to subjects daily perceptions. Asthma scores were based on symptoms of cough, wheezing, shortness of breath, chest tightness, increased sputum production, and decreased exercise tolerance. Each asthma symptom was scored on a scale of 0 to 10 (10 being the most severe) for a total possible daily score of 60 points, and a total weekly score of 420 points. Individual esophageal reflux symptoms, including heartburn, regurgitation, indigestion, and belching, were also quantified on a scale of 0 to 10 with a weekly range of 0 to 70 for each individual symptom. Weekly total esophageal reflux symptom scores were totaled for a total possible daily score of 40 points and a total weekly score of 280 points. The number of antacids taken to relieve esophageal reflux symptoms was also tabulated in a weekly score. Esophageal and Respiratory Manometrics After an overnight fast, esophageal manometry was performed with patients in the supine position using a round polyvinyl catheter (diameter, 4.5 mm; Arndorfer Specialities; Greendale, WI) continuously perfused with distilled water at a rate of 0.5 ml/min by a low-compliance, pneumohydraulic capillary infusion system (Arndorfer Specialities). The location and mean resting pressure at end-expiration of the LES and upper esophageal sphincter (UES), the mean esophageal contraction amplitude in the esophagus (at 3 cm, 8 cm, 13 cm, and 18 cm above the LES), and the percentage of peristaltic contractions in response to ten 5-mL swallows of water were obtained and measured by previously described techniques. 19 Respiratory manometric measurements included transthoracic pressure gradient and diaphragmatic pinch pressure. Transthoracic pressure was defined as the pressure difference between the gastric pressure (abdomen) and the pleural pressure (mid-esophagus) at the end of expiration. Diaphragmatic pinch pressure, reflecting the diaphragmatic crura s contribution to LES pressure, was measured by determining the amplitude of the three largest consecutive respiratory excursions (end-expiration to end-inspiration) within the first 1 to 2 cm of the caudad margin of the esophagogastric high pressure zone Twenty-four Hour Esophageal ph Testing Standardized methods of ambulatory 24-h esophageal ph testing were performed on all subjects. Immediately after esophageal manometry, a 2.5-mm in diameter monocrystalline catheter with two antimony ph electrodes (Medtronic Upper Airway; Minneapolis, MN) was passed nasally and positioned with the distal electrode 5 cm above the proximal border of the LES and the proximal electrode just below the UES. The proximal probe was placed within 3 cm of the UES using both commercially available and custom-made probes (Medtronic Upper Airway) with interprobe distances of 10, 12, 15, and 18 cm. The electrodes were calibrated at ph 7 and ph 1 using a buffer solution (Fisher Scientific; Fairlawn, NJ) before and at the completion of each study. A reference electrode was placed on the anterior chest. Both electrodes were connected to a digital recorder that stored ph data every 4 s. Subjects were sent home with instructions to record meal times, time of assuming the supine position for sleep, and time of arising in the morning. Subjects were instructed not to ingest antacid tablets during esophageal ph testing. Subjects were not given a specific diet and were instructed to eat normally, except for restrictions noted as follows: fruits (except bananas), tomato sauce or tomato-based foods, candy, gum, and beverages with a ph 4.0. Subjects were encouraged to perform their normal daily activities. After at least 18 h of recording, data were downloaded into an personal computer and analyzed separately for the proximal and distal esophageal ph electrodes. Based on 110 healthy control subjects using 95th percentile data in our laboratory, abnormal amounts of acid reflux were present in the distal esophagus if the total percentage of time that ph was 4 was 5.8% during the 24-h study period, upright-position acid exposure was 8.2%, or supine-position acid exposure was 3.5%. 25 Based on studies in 20 healthy volunteers, the amount of proximal reflux was deemed abnormal if the total percent time that ph was 4 exceeded 1.1%, if upright-position acid exposure was 1.7%, or if supineposition acid exposure was 0.6%. 26 CHEST / 121 / 2/ FEBRUARY,

4 Basal and Stimulated Gastric Acid Secretion Gastric acid analysis was performed in the basal and stimulated state using Pentagastrin Injection BP (Cambridge Laboratories). After an overnight fast, a 12F Sump tube (Bard; Covington, GA) was inserted nasally and placed fluoroscopically in the gastric antrum. Gastric contents were aspirated, and the volume, ph, and acid concentration were measured. During the basal hour phase, gastric secretions were collected, and ph acid concentration and acid output were measured. Next, 6 g/kg of Pentagastrin Injection BP was administered subcutaneously, and gastric secretions were again collected and analyzed. During the basal and stimulated or maximal acid output hour, gastric secretions were collected in aliquots at six 10-min intervals. Normal gastric output values for our laboratory in the basal state were 0.1 to 4 meq/h and in the stimulated state were 10 to 33 meq/h. 27 H Pylori Testing Because H pylori infection status could affect gastric acid secretion, H pylori testing was performed by measuring serum IgG antibody (Pharmacia Upjohn; Kalamazoo, MI). 28 Pulmonary Function Testing Spirometry was performed using a SensorMedics VMax 22 Series V6200 Autobox (SensorMedics; Yorba Linda, CA). FVC, FEV 1, mean forced expiratory flow during the middle half of FVC (FEF 25 75% ), PEF, and FEV 1 /FVC (FEV 1% ) were determined in accordance with American Thoracic Society Guidelines. 18 To ensure reproducibility, flow volume curves were monitored with each effort. Multiple forced expiratory curves were obtained. Reference values were obtained using Morris/ Polgar normal predicted equation sets based on height, age, weight, sex, and race. 29 Statistics Descriptive statistics were performed on demographic data, including subject weight, asthma status, esophageal reflux symptom scores, spirometry, esophageal and respiratory manometrics, esophageal ph data, and gastric acid analysis. Data from the placebo and prednisone phases were compared using paired t tests. Data from 24-h esophageal ph monitoring were nonparametric; therefore, data were logarithmically transformed (log 10 ) before statistical comparison with paired t tests. For symptom scores and spirometry, one-way analysis of variance was also used to compare baseline placebo with prednisone phases. This statistical analysis was used to assess asthma stability throughout all study phases. There were no differences in the significance of the data when one-way analysis of variance was used; therefore, reported p values were obtained from the paired t tests analysis. Because the treatment phase order was not randomized, the possibility of a treatment phase effect that is related to tolerability of the esophageal ph probe during the prednisone phase exists. To determine whether the experience of undergoing esophageal ph testing before this study had any effect on esophageal acid contact times, using unpaired t test analyses we compared the difference between the total distal esophageal acid exposure (during the prednisone and the placebo phases) in subjects who had esophageal ph testing against the difference found in the subjects who had not undergone esophageal ph testing before study entry. This same analysis was performed on the total proximal esophageal acid contact times. Because antacids were given to relieve esophageal reflux symptoms and could interfere with prednisone drug absorption, antacid use was closely monitored. To determine whether antacid use altered esophageal acid times, unpaired t tests to determine the difference between total distal esophageal acid exposures between the prednisone phase and the placebo phase were compared in subjects who received less than seven antacid tablets during the prednisone week (less than one tablet per day) with subjects who received seven or more antacid tablets a week. This same analysis was performed for the total proximal esophageal acid contact times. Data were considered statistically significant if p Values are expressed as mean SEM. Demographics Results Twenty-five subjects were screened for participation, of whom 20 subjects met entrance criteria and 19 subjects completed the protocol. One patient did not complete the protocol because of pregnancy before the second esophageal ph test. Eighteen participants were female, and the mean ( SEM) age of the study population was years (range, 20 to 60 years). Table 1 summarizes group demographics and asthma medication use. H pylori antibody results were positive in 5 of 12 consecutive subjects tested. Seven patients underwent a previous esophageal ph test before entry into the current study. Spirometry and Asthma Symptom Scores Table 2 reviews spirometry data at baseline, at the end of the placebo phase, and at the end of the prednisone phase. Spirometry and asthma symptom Table 1 Demographics of 20 Subjects With Moderate Persistent Asthma* Variables Data Demographics Age, yr Subjects White 10 (50) Black 10 (50) Body mass index, kg/m Asthma duration, yr Childhood-onset asthma 11 (55) Nocturnal asthma symptoms 17 (85) Asthma medication use Inhaled 2 -agonists 19 (95) Inhaled corticosteroids 10 (50) Inhaled anticholinergics 3 (15) Leukotriene modifiers 4 (20) Oral corticosteroids 0 (0) Oral 2 -agonists 0 (0) Oral theophylline 2 (10) No. of asthma medications used per day *As defined by American Thoracic Society guidelines and the Second Expert Panel for the Diagnosis and Management of Asthma. 16,17 Data are presented as mean SEM or No. (%). 628 Preliminary Report

5 Table 2 Spirometry and Asthma Symptom Scores at Baseline, End of Placebo Phase, and End of Prednisone Phase* Spirometry Baseline Placebo Prednisone p Value FVC, L FVC, % predicted FEV 1,L FEV 1, % predicted FEV 1% FEF 25 75%, L/s FEF 25 75%, % predicted PEF, L/s PEF, % predicted Weekly asthma symptom scores, of a possible 420 points *Data are presented as mean SEM. scores were used to assess asthma stability and eliminate the possibility that uncontrolled asthma might influence the results of the study. There were no significant changes noted in FVC, FEV 1,FEV 1%, mean FEF 25 75% and PEF. Data are shown both as actual values and percent predicted. There was also a trend toward improvement in weekly asthma symptom scores, but statistical significance was not reached. Esophageal Acid Contact Times and Esophageal Reflux Symptoms Esophageal acid contact times increased significantly after 7 days of treatments with prednisone, 60 mg/d, with statistical significance found at both the distal (5 cm above the LES) and the proximal (within 2 cm of UES) probes (total time esophageal ph 4, upright and supine positions). Table 3 shows mean esophageal ph data, and Figures 2 and 3 display individual responses at the distal and proximal esophageal ph probes. Because prednisone could result in weight gain and could potentially influence esophageal acid contact times, individual weights were monitored throughout the study period. There was no significant change in weight during the prednisone phase (p 0.08; Table 4). Three subjects had a weight gain of 3 lb during the prednisone phase. To see whether subjects who underwent esophageal ph testing before study entry exhibited alterations in esophageal acid contact times, we compared the difference in total distal esophageal acid contact times between the prednisone phase and the placebo phase for the 7 subjects who underwent a previous test with the 13 subjects who did not. There was no significant difference between groups (p 0.22). The same analysis was performed for total proximal esophageal acid contact times, which also showed no significant difference (p 0.10). Also, because antacid use has the potential to decrease prednisone absorption, a similar analysis was performed, and subjects who received seven or less antacid tablets a week (n 11) were compared with subjects who received seven or more antacid Table 3 Esophageal Acid Contact Times at the Distal and Proximal ph Probe During the Placebo and Prednisone Phases* Variables Placebo Prednisone p Values Distal probe, % Total (normal 5.8%) Upright (normal 8.2%) Supine (normal 3.5%) No. of episodes 5 min (normal 4) Longest episode, min (normal 18) Proximal probe, % Total (normal 1.1%) Upright (normal 1.7%) Supine (normal 0.6%) No. of episodes 5 min (normal 0) Longest episode, min (normal 3) *Data are presented as mean SE. CHEST / 121 / 2/ FEBRUARY,

6 Figure 3. Individual esophageal acid contact times during the total monitoring period at the proximal esophageal probe located within 3 cm below the UES during the placebo phase and the prednisone phase. Mean SE and p value are also displayed. Figure 2. Individual esophageal acid contact times during the total monitoring period at the distal esophageal probe located 5 cm above the LES during the placebo phase and the prednisone phase. Mean SE and p value are also displayed. tablets a week (n 8) during the prednisone phase. There was no significant difference noted in the total distal esophageal acid contact times during the prednisone phase between the antacid usage groups (p 0.95). There was also no significant difference noted between the prednisone and the placebo phases between groups (p 0.65). The identical analysis examining the total proximal esophageal acid contact times between the antacid usage groups showed no significant difference in the total proximal esophageal acid contact times during the prednisone phase (p 0.88), and there was no significant difference in the amount of change in esophageal acid contact times between the prednisone and placebo phases (p 0.29). Despite changes in esophageal acid contact times, esophageal reflux symptom scores were similar between the placebo and prednisone phases. Also, individual esophageal reflux symptoms, including heartburn, regurgitation, indigestion, belching, and antacid use, were similar between phases. Review of subject diaries during esophageal ph testing showed that no subject reported difficulty eating their normal diet, and no one received antacids. Table 4 displays esophageal reflux symptoms throughout the study period. Esophageal and Respiratory Manometrics and Basal and Stimulated Gastric Acid Secretion Manometric and gastric acid secretion analyses were performed to examine potential mechanisms of increased esophageal acid contact times during the prednisone phase. Table 5 displays manometric data as well as basal and stimulated gastric acid secretion during the placebo and prednisone phases. During the prednisone phase, a trend was noted toward higher LES pressures and higher amplitude of peristaltic contractions at 13 cm above the LES; however, no significant differences were noted between treatment phases in any of the respiratory or esophageal manometric measurements. Gastric analysis showed a significant decrease in gastric volume collected during the basal hour of the prednisone phase. There were no significant differences noted in the lowest gastric ph measurement, gastric acid concentration, or gastric acid output between treatment phases. 630 Preliminary Report

7 Table 4 Weekly Esophageal Reflux Symptom Scores, Subject Weight, and Weekly Number of Antacids Tablets Used* Symptoms Baseline Placebo Prednisone p Values Heartburn Regurgitation Indigestion Belching Total esophageal reflux symptom scores Antacid use Subject weight *Data are expressed as mean SE. Range for weekly esophageal reflux symptom scores 0to70. Discussion The purpose of this study was to determine whether prednisone alters esophageal acid contact times. Treatment with oral prednisone, 60 mg/d for 1 week, resulted in significant increases in esophageal acid contact times in subjects with stable, moderate persistent asthma. This finding is most likely not secondary to altered asthma control because asthma symptoms and spirometry variables did not change significantly throughout the study period. Other factors that could potentially contribute to the esophageal acid contact times findings include weight gain during the prednisone phase, which could promote GERD; familiarity with the esophageal ph probe test (because the subjects were not randomized to treatment phase); and antacid use. When analyzing these potential factors, none of the factors resulted in significant differences. Four of 19 subjects (21%) had abnormal esophageal ph parameters during the placebo phase, which increased to 13 of 19 subjects (68%) during the prednisone phase. The low prevalence of abnormal esophageal acid contact times at baseline is most likely caused by Table 5 Respiratory and Esophageal Manometry and Basal and Stimulated Gastric Acid Secretion During Placebo and Prednisone Phases* Variables Placebo Prednisone p Value Respiratory manometry, mm Hg Transdiaphragmatic pressure gradient Diaphragmatic pinch pressure Esophageal manometry, mm Hg UES pressure LES pressure Peristaltic contractions, % Peristaltic contraction amplitude, mm Hg 3 cm above LES cm above LES cm above LES cm above LES Basal and stimulated gastric acid secretion Fasting analysis Volume, ml Lowest, ph Acid concentration, meq/l Basal acid secretion Volume, ml Lowest ph Acid concentration, meq/l Acid output, meq/h Stimulated acid secretion Volume, ml Lowest ph Acid concentration, meq/l Acid output, meq/h *Data are presented as mean SE. Utilizing Pentagastrin, 6 g/kg, subcutaneous injection. CHEST / 121 / 2/ FEBRUARY,

8 study subjects not having a diagnosis of GERD or frequent esophageal reflux symptoms at study entry. Interestingly, esophageal reflux symptoms did not change significantly between treatment phases. Previous work 13,30 has shown that significant abnormal esophageal acid contact times may be present in the absence of esophageal reflux symptoms in people with asthma. The mechanism for the increase in esophageal acid contact times remains unclear. There were no significant differences in respiratory or esophageal manometrics, including LES pressure, UES pressure, peristaltic contractions, transdiaphragmatic pressure gradient, or diaphragmatic pinch pressure between the placebo and prednisone phases. The small sample size may have limited the power to examine potential mechanisms. Similarly, gastric acid secretion analysis was not able to demonstrate any differences in gastric acid output, acid concentration, or lowest ph measurement in the basal or stimulated state during the two treatment phases. The significance of the decrease in basal gastric volume during the prednisone phase is not known. Theoretically, this finding could be protective to GERD development. Previous studies 15,31 34 evaluating the effect of systemic corticosteroids on gastric acid secretion in animals and in human subjects have demonstrated variable results, which might be attributable to the length of time corticosteroids were administered. For instance, Strickland et al 15 showed in a placebo-controlled trial that treatment with prednisolone, 20 mg/d for 1 month, in 14 healthy men increased betazole hydrochloride-stimulated gastric acid secretion and not basal acid secretion. However, Steinert et al, 34 in a double-blind, randomized, placebo-controlled trial examining the effect treatment with oral prednisone, 60 mg/d for 6 days, in 14 healthy control subjects found no significant differences in basal or Pentagastrin-stimulated gastric acid secretion. Our findings agree with the results of Steinert et al. 34 Other potential mechanisms for the increase in esophageal acid contact times with prednisone include an increase in transient LES relaxations or delayed gastric emptying, which were not measured in our study population. Mittal et al 20,35,36 demonstrated that transient LES relaxations are an important mechanism for GERD development. Although there were significant increases in esophageal acid contact times during the prednisone phase, mean values for the entire group during the prednisone phase were in the abnormal range only for the total distal and distal supine-position variables. This may be partially explained because our subject group did not have the diagnosis of GERD or significant esophageal reflux symptoms at study entry. A potential weakness of this study is the lack of randomization in the order of the treatment phases. The prednisone phase was evaluated last because of potential treatment phase effects and the lack of information regarding the proper length of a prednisone washout phase. We designed a 3-week study to improve the probability that the subjects asthma would remain stable throughout the study period. Potentially, a change in asthma control could alter esophageal acid contact times. H pylori antibody testing showed that 5 of 12 subjects had a positive test result. The prevalence of H pylori infection has not been rigorously tested in an asthma population. Epidemiologic studies show that H pylori infection prevalence in an asymptomatic population in the United States is 52% and that it increases with age (approximately 1%/yr) and is higher in black people (70%) compared with white people (34%). 37 Fifty percent of our study population tested for H pylori were black, which may explain our positivity rate of 42%. The American College of Gastroenterology (ACG) guidelines for the management of H pylori infection conclude that H pylori infection is common in the general population, and that testing for H pylori is indicated in patients with a history of active peptic ulcer disease or gastric cancer (mucosa-associated lymphoid tissue lymphoma). 38 None of our patients met ACG guideline criteria for testing. H pylori status was examined in our population because H pylori can alter gastric acid secretion. We do not recommend H pylori testing or treatment for people with asthma who do not meet ACG guideline criteria. The clinical significance of increased esophageal acid contact times with prednisone use is not known. Esophageal acid triggers bronchoconstriction in selected patients with asthma. Previous studies have shown that esophageal acid may lower PEF and may increase specific airway resistance in selected people with asthma. Esophageal acid events also correlate with airway reactivity. 14,42 For instance, Cuttitta et al 42 noted a correlation between reflux episode duration and increased respiratory resistance when monitoring nocturnal esophageal ph monitoring with polysomnography. Furthermore, Vincent et al 14 showed in 105 consecutive patients with asthma that the provocative dose of methacholine causing a 20% fall in FEV 1 correlated with the number of reflux episodes during 24-h esophageal ph testing. Esophageal acid events correlated with respiratory symptoms in 98% of coughing episodes and 65% of wheezing and shortness of breath episodes associated with esophageal acid during 24-h 632 Preliminary Report

9 esophageal ph testing. 43 Also, acid suppressive therapy using omeprazole improved asthma outcomes, including symptoms, PEF, and spirometry, in selected patients. 44 Future studies examining what effect acid suppressive therapy might have on asthma outcomes in patients requiring prednisone during an asthma exacerbation would also add valuable clinical information. ACKNOWLEDGMENT: The authors thank Dr. Basil Hirschowitz, Jean Price, MT (ASCP), and Susan Irwin, RN, BSN, for assistance with gastric analysis, and Ms. Arren Graf for editorial assistance. References 1 Harding SM. Gastroesophageal reflux and asthma: insight into the association. J Allergy Clin Immunol 1999; 104: Field SK, Underwood M, Brant R, et al. Prevalence of gastroesophageal reflux symptoms in asthma. Chest 1996; 109: Sontag SJ, O Connell S, Khandelwal S, et al. Most asthmatics have gastroesophageal reflux with or without bronchodilator therapy. Gastroenterology 1990; 99: Sontag SJ, Schnell TG, Miller TQ, et al. Prevalence of oesophagitis in asthmatics. Gut 1992; 33: Harding SM. GERD, airway disease, and the mechanisms of interaction. In: Stein MR, ed. Gastroesophageal reflux disease and airway disease: lung biology in health and disease (vol 129). New York, NY: Marcel Dekker, 1999; Johannesson N, Andersson KE, Joelsson B, et al. Relaxation of lower esophageal sphincter and stimulation of gastric secretion and diuresis by antiasthmatic xanthines. Am Rev Respir Dis 1985; 131: Stein MR, Towner TG, Weber RW, et al. The effect of theophylline on the lower esophageal sphincter pressure. Ann Allergy 1980; 45: Berquist WE, Rachelefsky GS, Kadden M, et al. Effect of theophylline on gastroesophageal reflux in normal adults. J Allergy Clin Immunol 1981; 67: Ekström T, Tibbling L. Influence of theophylline on gastroesophageal reflux and asthma. Eur J Clin Pharmacol 1988; 35: Christensen J. Effects of drugs on esophageal motility. Arch Intern Med 1976; 136: DiMarino AJ, Cohen S. Effect of an oral 2 -adrenergic agonist on lower esophageal sphincter pressure in normals and in patients with achalasia. Dig Dis Sci 1982; 27: Schindlbeck NE, Heinrich C, Huber RM, et al. Effects of albuterol (salbutamol) on esophageal motility and gastroesophageal reflux in healthy volunteers. JAMA 1988; 260: Irwin RS, Curley FJ, French CL. Difficult-to-control asthma: contributing factors and outcome of a systematic management protocol. Chest 1993; 103: Vincent D, Cohen-Jonathan AM, Leport J, et al. Gastroesophageal reflux prevalence and relationship with bronchial reactivity in asthma. Eur Respir J 1997; 10: Strickland RG, Fisher JM, Taylor KB. Effect of prednisolone on gastric function and structure in man. Gastroenterology 1969; 56: National Asthma Education and Prevention Program. Expert panel report 2. Guidelines for the diagnosis and management of asthma. Bethesda, MD; National Institutes of Health, 1997; publication No A 17 American Thoracic Society. Standards for the diagnosis and care of patients with chronic obstructive pulmonary disease (COPD) and asthma. Am Rev Respir Dis 1987; 136: American Thoracic Society. Standardization of spirometry, 1994 update. Am J Respir Crit Care Med 1995; 152: Dalton CB, Castell DO, Richter JE, eds. The manometric study. In: Esophageal motility testing. New York, NY: Elsevier, 1987; Mittal RK, Balaban DH. The esophagogastric junction. N Engl J Med 1997; 336: Hein K, Dent J, Mittal RK. Anatomical relationship between the crural diaphragm and the lower esophageal sphincter: an electrophysiologic study. J Gastrointest Motil 1993; 5: Welch RW, Gray JE. Influence of respiration on recording of LES pressure in humans. Gastroenterology 1982; 83: Boyle JT, Altschuler SM, Nixon TE, et al. Role of the diaphragm in the genesis of lower esophageal sphincter pressure in the cat. Gastroenterology 1985; 88: Delattre JF, Palot JP, Ducasse A, et al. The crura of the diaphragm and diaphragmatic passage. Anat Clin 1985; 7: Richter JE, Bradley LA, DeMeester TR, et al. Normal 24-h ambulatory esophageal ph values: influence of study center, ph electrode, age and gender. Dig Dis Sci 1992; 37: Sinclair JW, Dalton CB. The performance of ambulatory esophageal ph monitoring in adults. In: Richter JE, ed. Ambulatory esophageal ph monitoring: practical approach and clinical applications. New York, NY: Igaku-Shoin, 1991; Hirschowitz BI. Apparent and intrinsic sensitivity to Pentagastrin of acid and pepsin secretion in peptic ulcer. Gastroenterology 1984; 86: el-omar EM, Perman P, Ardill JE, et al. Helicobacter pylori infection and abnormalities of acid secretion in patients with duodenal ulcer disease. Gastroenterology 1995; 109: Morris JF, Koski A, Johnson LC. Spirometric standards for healthy nonsmoking adults. Am Rev Respir Dis 1971; 103: Harding SM, Guzzo MR, Richter JE. The prevalence of gastroesophageal reflux in asthma patients without reflux symptoms. Am J Respir Crit Care Med 2000; 162: Berndt V, Konrad RM, Störmer B, et al. Die potezierung der betazol-stimulierten magensaftsekretion durch prednisolon beim hund. Res Exp Med 1978; 172: Mertz DP, Papavassiliou K. Akute wirkung von prednisolon auf die säuresekretion der stimulierten menschlichen magenschleimhaut. Münch Med Wochenschr 1975; 117: Jacobson ED, Price WE. Effect of hydrocortisone on gastric mucosal blood flow and secretion. Gastroenterology 1969; 57: Steinert S, Singer MV, Eysselein VE, et al. Einflu einer einwächagen gabe von prednison auf die magensaureskretion und die freissetzung von gastrin beim gesundern menschem. Z Gastroenterol 1990; 28: Mittal RK, Holloway RH, Penagini R, et al. Transient lower esophageal sphincter relaxation. Gastroenterology 1995; 109: Mittal RK, Rochester DF, McCallum RW. Electrical and mechanical activity in the human lower esophageal sphincter during diaphragmatic contraction. J Clin Invest 1988; 81: Graham DY, Malaty HM, Evans DG, et al. Epidemiology of Helicobacter pylori in an asymptomatic population in the CHEST / 121 / 2/ FEBRUARY,

10 United States: effect of age, race, and socioeconomic status. Gastroenterology 1991; 100: Howden CW, Hunt RH. Guidelines for the management of Helicobacter pylori infection. Am J Gastroenterol 1998; 93: Schan CA, Harding SM, Haile JM, et al. Gastroesophageal reflux-induced bronchoconstriction: an intraesophageal acid infusion study using state-of-the-art technology. Chest 1994; 106: Harding SM, Schan CA, Guzzo MR, et al. Gastroesophageal reflux-induced bronchoconstriction: is microaspiration a factor? Chest 1995; 108: Field SK. A critical review of the studies of the effects of simulated or real gastroesophageal reflux on pulmonary function in asthmatic adults. Chest 1999; 115: Cuttitta G, Cibella F, Visconti A, et al. Spontaneous gastroesophageal reflux and airway patency during the night in adult asthmatics. Am J Respir Crit Care Med 2000; 161: Harding SM, Guzzo MR, Richter JE. 24-h esophageal ph testing in asthmatics: respiratory symptom correlation with esophageal acid events. Chest 1999; 115: Harding SM, Richter JE, Guzzo MR. Asthma and gastroesophageal reflux: acid suppressive therapy improves asthma outcome. Am J Med 1996; 100: Preliminary Report