| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Dr. Richter is Chairman and Professor of Medicine, Department of Gastroenterology, Cleveland Clinic Foundation, Cleveland, OH.
Correspondence to: Joel E. Richter, MD, Ther Cleveland Clinic Foundation, Department of Gastroenterology, 9500 Euclid Ave, Cleveland, OH 44195-0002
In 1892, Sir William Osler1 first observed the association between worsening asthma and a distended stomach. However, it was only since the publication by Kennedy2 in 1962 that attention has focused on the potential causal association between gastroesophageal reflux disease (GERD) and asthma. In the current issue of CHEST (see page 1257), the study by Kiljander and colleagues allows us the opportunity to look at this possible association and to critique the problems inherent in defining its true relationship.
To strengthen the cause-and-effect relationship between GERD and asthma, three criteria should be met. First, patients with GERD should have a higher prevalence of asthma than patients without GERD. This is the case since these reflux symptoms are reported in up to 77% of asthmatics,3 while 32 to 82% of asthmatics have abnormal pH studies.4 5 Additionally, "silent reflux" may be as common as symptomatic reflux, with reports suggesting that 25 to 50% of asthmatics have no reflux complaints but abnormal pH studies.5 6 Second, the pathophysiologic mechanisms between GERD and asthma should help explain how the disease processes interact (ie, esophageal acid should exacerbate asthma). Animal and human studies have shown that GERD can aggravate asthma through several mechanisms, including the following: (1) vagally mediated reflex triggered by acid in the esophagus, (2) heightened bronchial reactivity, and (3) microaspiration of gastric acid resulting in bronchoconstriction.7 Lastly, if GERD causes asthma, then antireflux therapy, either medical or surgical, should improve or even resolve the asthma in many patients. Unfortunately, this is where the causal association begins to fall apart. For example, Field and Sutherland8 recently reviewed 12 studies involving 326 asthma patients whose reflux disease was aggressively treated with medical therapy and could only conclude that "medical antireflux therapy improves asthma symptoms, may reduce asthma medication used, but has minimal or no effect on lung function."
I believe that a number of methodologic limitations exist in the currently published literature relating GERD and asthma, which contributes to the confusion in this area.9 In many of the published studies, there is no attempt to optimize conventional, "standard" therapy for the underlying asthma. Today, this means using daily inhaled corticosteroids. The inadequate use of inhaled corticosteroids is a major treatment shortcoming for many patients with poorly controlled asthma10 and an important source of variability in the published series.7 8 The current study from Finland had patients receiving optimal asthma therapy, including 89% taking inhaled steroids and 91% taking ß-sympathomimetic drugs. On the other hand, too-tight control of asthma may not allow a sufficient margin for any new therapy to show improvement, increasing the chances of a false-negative study unless large number of patients are investigated. The authors of the current study believe that this may have been a factor making it difficult to reach a 20% increase in pulmonary function (the definition of a responder) or a statistically significant improvement in asthma symptoms after omeprazole therapy.
A second major limitation in most studies is the lack of an objective assessment of acid suppression while asthmatics are being treated for GERD. The only published study addressing this issue was performed by Harding et al,11 using serial pH studies in patients treated with the proton pump inhibitor, omeprazole. They found that 73% of patients only required omeprazole at 20 mg/d, but 20% required omeprazole at 40 mg/d, 7% needed 60 mg/d, and one patient was eliminated from the study because her GERD was still not controlled. In the absence of documented control of acid reflux, one cannot conclude that therapy for GERD did not improve the associated asthma. This may explain the overall better efficacy of antireflux surgery in relieving symptoms of asthma. A recent review of 10 surgical studies involving 318 asthmatics with GERD found that 253 patients (80%) had their asthma improved. Of this group, over half were "cured" of their disease (ie, they did not require further asthma medication); many were previously taking oral steroids.7
A third limitation is the absence of a control group in some of these studies. Based on data from a variety of experimental anti-inflammatory therapies for chronic steroid-dependent asthma, we know the placebo arm can improve from 20 to 40% simply by participating in a clinical study (ie, the "Hawthorne effect"12 ). A separate control group doubles the number of patients needed and may make recruitment difficult, because patients may not want to risk receiving a placebo for an extended period of time. As done in the current study, a crossover design eliminates these problems. However, this type of design has its own inherent problems, including carryover effect and order effect. Carryover effect implies that active therapy, if given first, may continue to influence the disease even during the placebo phase. As done in this study, this can usually be eliminated by an adequate washout period. On the other hand, an order effect maybe more difficult to control and explain. A review of Figures 1–2 in the study by Kiljander and colleagues suggests that this may have been a factor. In contrast to patients receiving omeprazole first, those receiving placebo for 8 weeks improved their pulmonary symptoms when switched to omeprazole, 40 mg, each morning after a 2-week washout period.
A fourth limitation among many of these studies is that the duration of acid suppression may be too short to predictably show a response in symptoms and pulmonary function tests. As illustrated in Figure 3 of the current study, patients who were regarded as responders showed improvement in their pulmonary symptoms only after several weeks of omeprazole treatment. This is in accordance with several previous medical treatment studies,11 13 and is further confirmed by the surgical series in which those with longer follow-up usually found improvement in pulmonary function tests.7 The optimal duration of therapy should be at least 3 months and could even be as long as 6 months, when taking into consideration seasonal variations in asthma.
Finally, predictors need to be identified to help characterize which patients to treat aggressively with antireflux therapy. Harding and colleagues11 found that responders to omeprazole therapy were characterized by the presence of proximal reflux on pH testing and acid regurgitation into the mouth at least one time per week. These are markers of more severe disease with possible microaspiration. Likewise, the responders in the study by Kiljander and colleagues had greater amounts of total and upright acid reflux on pH testing in comparison to the nonresponders. Other studies have suggested a wide array of predictors, including the presence of nonallergic asthma; difficult-to-control asthma; nocturnal asthma; other respiratory or ear, nose, and throat complaints; and esophagitis healing on medical therapy.7 However, none of these predictors have systematically been reexamined in an independent study population. Future studies need to be "enriched" with patients who, in all likelihood, have GERD exacerbating their asthma. If this cannot be accomplished, then the prevalence of asthma caused by GERD in the individual studies, rather than the efficacy of the drug treatment, will be the deciding factor in the outcome of these studies.
Many questions remain regarding the relationship and proper treatment for GERD-associated asthma. A large multicenter study is needed to adequately address this issue. I would suggest using a proton pump inhibitor at a very high dosage (ie, omeprazole, 40 mg bid, or lansoprazole, 60 mg bid), possibly adding a bedtime H2 antagonist to better control nocturnal acid secretion.14 This approach would avoid individual titration by serial pH testing that would be impossible to perform in a large study. The study duration should be at least 6 months. Flexibility in asthma control should be allowed, and patients with difficult-to-control asthma should be included in the study.6 A detailed demographic analysis as well as a comprehensive pulmonary and GERD evaluation (endoscopy, manometry, pH testing) should be obtained in all patients. A study with an adequate sample size (probably 150 to 200 patients per group) will help to validate or refine suggested predictors for a therapeutic response to acid suppression. Finally, cost analysis and quality of life studies will be necessary to assess the cost tradeoffs (ie, expensive antireflux medication vs less asthma medicines), improvement in quality of life, and health-care utilization in these patients. Hopefully, this "perfect" study will bring truth to this difficult issue.
References
This article has been cited by other articles:
|
|
 |
|
  J. J. Leggett, B. T. Johnston, M. Mills, J. Gamble, and L. G. Heaney Prevalence of Gastroesophageal Reflux in Difficult Asthma: Relationship to Asthma Outcome Chest, April 1, 2005; 127(4): 1227 - 1231. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. E. Richter Not the Perfect Study, but Helpful Wisdom for Treating Asthma Patients With Gastroesophageal Reflux Disease Chest, April 1, 2003; 123(4): 973 - 975. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
