Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Meta-analysis in chronic cough
- Alyn H Morice (24 January 2006)
-
Is treatment for chronic cough associated with gastro-oesophageal reflux evidence based?
- Ms. Natasha Choudhury, Mr. Saumabha Mandal (1 February 2006)
-
Not all reflux is gastro-oesophageal
- Jonathan C Hobson, Nicholas J Kay (7 February 2006)
-
Reflux and cough: a phenomenon that needs clarification
- Silvia Salvatore, Bruno Hauser, Yvan Vandenplas (15 February 2006)
-
clarification of data
- Anne B Chang, Toby J. Lasserson, Toni O. Kiljander, Frances L. Connor, Luke A Garske ) and Chris Cates (17 February 2006)
|
|
|||
|
Alyn H Morice, Professor of Respiratory Medicine University of Hull, Castle Hill Hospital, HU16 5JQ
Send response to journal:
|
Re: Systematic review and meta-analysis of randomised controlled trials in gastro-oesophageal reflux interventions for chronic cough associated with gastro-oesophageal reflux Anne Chang and colleagues (1) point out the inadequacy of trials of proton pump inhibitors in cough associated with gastroesophageal reflux. They then compound the confusion by performing a meta-analysis on the manifestly deficient studies in this area. When performed on adequate numbers of properly powered studies, preferably with prospectively agreed common end points, then meta-analysis may be a useful tool. However, when poor quality studies are cobbled together then spurious conclusions may be generated. Indeed, even well performed retrospective meta-analyses published in high quality journals are little better than tossing a coin in predicting the outcome of subsequent large randomised studies (2). The inadequacy of the studies used are not primarily the fault of the investigators. The treatment of reflux associated cough is an evolving area. Thus, it is now clear that a greater degree of acid suppression is required in treating cough as compared to heartburn. As the authors point out, response to anti reflux medication may be delayed and in retrospect studies with a cross-over design are inappropriate. Their inclusion in a meta analysis will inevitably lead to erroneous conclusions. Chang and colleagues criticise the international guidelines that recommend the use of PPI therapy in the treatment of reflux cough. In the construction of the guidelines cognisance was taken of all of the available evidence. Non randomised studies have shown PPIs to be effective in approximately 50% of patients who are thought to have reflux cough. In our own prospective experience cough was significantly improved in 23 out of 39 patients taking twice daily Rabeprazole (3) . Chang and her colleagues dismiss this sort of experience as possibly being due to a placebo effect. Their reference however is disingenuous since it deals with the well recognised placebo effect in the treatment of acute cough. Those of us dealing with patients with chronic cough with its high degree of co-existent morbidity and depression would be only too pleased if there were such a dramatic placebo response! In reality it is the belief that systematic analysis of the existing literature according to arbitrary criteria provides a better evidence base for the construction of guidelines than the summation of the evidence provided by those expert in the field. We have previously commented at length about this widespread but dangerous fallacy (4). Chang and colleagues would have done better to learn from the errors in the preceding studies and constructed a properly designed trial, particularly in paediatrics where no adequate studies exist. Our own randomised controlled parallel group study is nearing completion and we plan to submit the manuscript to the BMJ, whose invention of a new syndrome of chronic non specific cough (5) clearly alludes to their interest in this area. Reference List (1) Chang AB, Lasserson TJ, Kiljander TO, Connor FL, Gaffney JT, Garske LA. Systematic review and meta-analysis of randomised controlled trials of gastro-oesophageal reflux interventions for chronic cough associated with gastro-oesophageal reflux. Br Med J 2006; 332(7532):11-17. (2) LeLorier J, Gregoire G, Benhaddad A, Lapierre J, Derderian F. Discrepancies between meta-analyses and subsequent large randomized, controlled trials. N Engl J Med 1997; 337(8):536-542. (3) Smith GM, Ramsay JR, Everett CF, Kastelik JA, Morice AH. The effect of rabeprazole on cough related to gastro-oesophageal reflux. European Respiratory Journal. Supplement Vol 26 [Supplement 49, Pages 1S - 832S], 302s. 2005. (Abstract) (4) Morice AH, Parry-Billings M. Evidence based guidelines-a step too far? Pulm Pharmacol Ther 2005. (5) Anonymous. Role of gastro-oesophageal reflux treatment for chronic cough may be in doubt. Br Med J 2006; 332(No 7532). Competing interests: None declared |
|||
|
|
|||
|
Ms. Natasha Choudhury, SpR ENT Southend General Hospital, Mr. Saumabha Mandal
Send response to journal:
|
Dear Editor, We read the article by Chang et al with great interest(1). Chronic cough is a common symptom encountered in ENT. It is frequently attributed to some degree of underlying gastro-oesophageal reflux, particularly in those patients who do not have any underlying laryngeal pathology. Empirical medical treatment for this with acid suppressing medication is commonly employed. However, patients’ response to such empirical treatment is variable(2) and the evidence supporting the recommended length of treatment and its efficacy is limited(3). Chang et al’s publication claims to objectively demonstrate the efficacy of treatment for gastro-oesophageal reflux disease on chronic cough in adults and children with no primary lung disease(1). However, we feel that there are a number of limitations of this review. The authors have performed a meta-analysis of randomised controlled trials of gastro- oesophageal reflux interventions for chronic cough. However, only 8 studies were included in adults and 3 in children. In our opinion, many of these individual studies were poorly designed. For example, not all patients that were included had cough as a symptom of their gastro- oesophageal reflux. Some adult studies included patients with ‘laryngitis symptoms,’ which included any symptoms of cough, hoarseness, throat- clearing or globus sensation. In addition, the presence of gastro- oesophageal reflux disease was not confirmed objectively in all the studies that were analysed, despite the defined objective of the publication being to demonstrate the efficacy of treatment for this on chronic cough. Also, patients with primary lung disease should clearly have been excluded as the review aims to establish the efficacy of treatment interventions for reflux disease on chronic cough in patients without an underlying respiratory disorder. However, at least one paediatric study included children with cough, with or without co-existing asthma. We believe that these are fundamental flaws in some of the study designs, and such studies should not have been included for the purpose of this meta-analysis. With the limited number and quality of paediatric studies reviewed, we agree that this part of the meta-analysis is not possible. However, we would add that even for the adult studies, there is considerable variation between them in terms of the spectrum of treatment modalities being evaluated, as well as the duration of treatment. We feel that pooling data together from such a limited number of studies that analyse different treatment interventions for gastro-oesophageal reflux undermines the value of this meta-analysis. We agree that this paper addresses an important subject matter, as patients with chronic cough associated with gastro-oesophageal reflux are commonly reviewed in both the primary care and hospital setting, and indeed by a number of specialties including paediatrics, ENT, respiratory medicine and gastroenterology. A more critical meta-analysis of well designed, high powered studies may go some way in adding to recommendations for treatment guidelines for this important clinical problem. References (1).Chang AB, Lasserson TJ, Kiljander TO, Connor FL, Gaffney JT, Garske LA. Systematic review and meta-analysis of randomised controlled trials of gastro-oesophageal reflux interventions for chronic cough associated with gastro-oesophageal reflux. BMJ 2006; 332:11-14 (2).Irwin RS. Chronic cough due to gastroesophageal reflux disease: ACCP Evidence –based clinical practice guidelines. Chest 2006; 129(1): 80S -94S (3).Boulet LP. Future directions in the clinical management of cough: ACCP Evidence –based clinical practice guidelines. Chest 2006; 129(1): 287S-292S Competing interests: None declared |
|||
|
|
|||
|
Jonathan C Hobson, ENT SpR Stepping Hill Hospital, Stockport, SK2 7JE, Nicholas J Kay
Send response to journal:
|
We read with interest Chang et al's article on chronic cough associated with gastro-oesophageal reflux (GORD). They state that objective confirmation of GORD is preferable to empirical therapy in patients with chronic non-specific cough and no gastrointestinal symptoms. The problem with this approach is that cough (along with hoarseness and globus symptoms) is likely to be a manifestation of extra-oesophageal reflux; namely laryngo-pharyngeal reflux (LPR). Otolaryngologists diagnose LPR on a daily basis with the use of the flexible nasendoscope, but without always resorting to contrast studies, pH monitoring or gastroscopy. Our experience is that treatment of such patients is often beneficial and this has been documented elsewhere(1). LPR patients have head and neck symptoms, but heartburn is uncommon(2). We agree with Chang et al. that the effect of drugs to reduce the effect of gastric acid in GORD is less universal than suggested in cohort studies. We would like to re-emphasise that LPR can exist in the absence of GORD and in these patients cough may be a presenting symptom of reflux in the absence of a ‘positive’ oesophagogastroscopy. (1) Ford CN “Evaluation and management of laryngopharyngeal reflux.” JAMA. 2005 Sep 28;294(12):1534-40 (2) Koufman JA “Laryngopharyngeal reflux is different from classic gastroesophageal reflux disease” Ear Nose Throat J. 2002 Sep;81(9 Suppl 2):7-9 Competing interests: None declared |
|||
|
|
|||
|
Silvia Salvatore, Lecturer in Pediatrics Clinica Pediatrica, Università dell'Insubria, Ospedale F. Del Ponte, Piazza Biroldi, 21100 Varese,, Bruno Hauser, Yvan Vandenplas
Send response to journal:
|
The relation between gastro-oesophageal reflux disease(GORD) and cough has long been considered, but despite major advances in acid treatment and higher clinical awareness is still incompletely clarified. Chang et al [1] concluded that treatment of GORD (with proton pump inhibitors) in chronic cough is expected to produce “some effect in some adult”. The uncertainty of the message derives from multiple factors which were just briefly considered but should be emphasized. The restricted number of studies eligible, the selection of patients (all presenting with GORD symptoms), the different diagnostic criteria (for GORD), the duration of treatment, the various cough scales, and the limited follow-up may all justify the not significant result of the meta-analysis for the main outcomes. Data in children were even more limited, and the benefit of proton pump inhibitors have never been evaluated, thus no clear conclusion could be drawn. Furthermore the studies included in the meta-analysis used event markers or diary annotation to identify cough, selected various “windows” of temporal association and symptom indexes, and did not consider non-acid reflux. However the role of non-acid reflux, the availability of improved investigations and the spectrum of GORD patients have recently added crucial insights.[2] Combined oesophageal impedance with pH-monitoring allows now detection and classification of all reflux episodes (especially postprandial, weakly and non-acid) and simultaneous use of gastro- oesophageal manometry provides an accurate and objective recognition of cough bursts.[3] The importance of identifying subgroup of patients with events associated with less acidic content, the discrimination of cough-inducing or provoking reflux, and a more homogeneous selection of symptom “index” (as the emerging proposed symptom probability association) and time relation could be crucial to establish the appropriate treatment and clarify the relation between the two phenomena. 1. Chang AB, Lasserson TJ, Kiljander TO, Connor FL, Gaffney JT, Garske LA. Systematic review and meta-analysis of randomised controlled trials of gastro-oesophageal reflux interventions for chronic cough associated with gastro-oesophageal reflux. BMJ 2006;332:11-17 2. Sifrin D. Acid, weakly acidic and non-acid gastro-oesophageal reflux: differences, prevalence and clinical relevance. Eur J Gastroenterol Hepatol 2004;16(9):823-30. 3. Sifrim D, Dupont L, Blondeau K, Zhang X, Tack J, Janssens J. Weakly acidic reflux in patients with chronic unexplained cough during 24 hour pressure, pH, and impedance monitoring. Gut 2005;54:449–454. Competing interests: None declared |
|||
|
|
|||
|
Anne B Chang, Paediatric Respiratory Consultant Royal CHildren;s Hospital, Brisbane, Australia, Toby J. Lasserson, Toni O. Kiljander, Frances L. Connor, Luke A Garske ) and Chris Cates
Send response to journal:
|
We thank Prof Morice and Ms Choudhury for their interest in our paper.[1] One of the difficulties in the issue of GOR and respiratory diseases (cough, asthma, etc) is that it is controversial; there are some believers and some sceptics.[2,3] An objective look at randomised evidence overcomes bias associated with personal anecdotal experiences.[4] Our findings are indeed similar to that of evidence based guidelines of GOR[5] clearly stating the lack of good evidence for cause and effect (as opposed to an association). Better and larger randomised controlled trials are clearly needed for a definitive conclusion on the efficacy of GORD treatments for cough. Prof Morice criticises the inclusion of cross over studies in the review. Clarification of this point requires pointing out that the studies were prospective and not retrospective, GIV analysis is a well accepted statistical method and that only the first arm of the study was used in the meta-analysis in accordance to current statistical standards. The limitations of any systematic review and meta-analysis have been well described and that specific for our review was clearly elaborated in the Cochrane review.[6] In the absence of adequate randomised controlled data, it is impossible to defend the possibility of a placebo effect explaining positive results of non-randomised studies, and the natural history of waxing and waning of chronic cough provides further potential explanation for the observed improvement in non-randomised studies. Cochrane reviews do not aim to conduct studies as suggested by Prof Morice. We applaud and look forward to including Prof Morice’s study in updates of our review. Ms Choudhury fails to realise that a systematic review is not equivalent to a meta-analysis and none of the pediatric data was included in the meta- analysis, precisely for the reasons she described, among other reasons. Also key issues of publication bias, use of sensitivity analysis, quality scores etc were not understood. Our Cochrane review highlighted what is known in other circles about the conundrum[7,8] and current poor evidence on the relationship and efficacy of GORD treatments on the symptom of isolated cough. We add to Gregor’s call for the rational use of GOR treatments such as proton pump inhibitors.[9] References (1) Chang AB, Lasserson TJ, Kiljander TO, Connor FL, Gaffney JT, Garske LA. Systematic review and meta-analysis of randomised controlled trials of gastro-oesophageal reflux interventions for chronic cough associated with gastro-oesophageal reflux. BMJ 2006; 332(7532):11-17. (2) Ours TM, Richter JE. Re: Irwin editorial--cough and GERD. Am J Gastroenterol 2000; 95(7):1833-1836. (3) Weinberger M. Gastroesophageal reflux disease is not a significant cause of lung disease in children. Pediatr Pulmonol Suppl 2004; 26:197-200. (4) Altman DG, Bland JM. Statistics notes. Treatment allocation in controlled trials: why randomise? BMJ 1999; 318(7192):1209. (5) Rudolph CD, Mazur LJ, Liptak GS, Baker RD, Boyle JT, Colletti RB et al. Guidelines for evaluation and treatment of gastroesophageal reflux in infants and children: recommendations of the North American Society for Pediatric Gastroenterology and Nutrition. J Pediatr Gastroenterol Nutr 2001; 32 Suppl 2:S1-31. (6) Chang AB, Lasserson T, Gaffney J, Connor FC, Garske LA. Gastro- oesophageal reflux treatment for prolonged non-specific cough in children and adults. The Cochrane Database of Systematic Reviews 2005;Issue 2. (7) Richter JE. Ear, nose and throat and respiratory manifestations of gastro-esophageal reflux disease: An increasing conundrum. Eur J Gastro Hepatology 2004; 16:837-845. (8) Rudolph CD. Supraesophageal complications of gastroesophageal reflux in children: challenges in diagnosis and treatment. Am J Med 2003; 115 Suppl 3A:150S-156S. (9) Gregor JC. Acid suppression and pneumonia: a clinical indication for rational prescribing. JAMA 2004; 292:2012-2013. Competing interests: None declared |
|||