There is overwhelming evidence linking NSAIDs to a variety of gastrointestinal toxicities (Fries 1990, Stalnikowicz 1993, Smalley 1996, Fries 1991, Griffin 1988, Bollini 1992, McMahon 1997, Gabriel 1991, Langman 1994, MacDonald 1997, Armstrong 1987, Silverstein 1995). Minor side effects such as nausea, and dyspepsia are relatively common, but these correlate poorly with serious adverse GI events (Silverstein 1995, Larkai 1987). Endoscopic ulcers can be documented in as many as 40% of chronic NSAID users (Stalnikowicz 1993), however, it is estimated that as many as 85% of these ulcers never become clinically symptomatic (Silverstein 1995, Maetzel 1998). Serious NSAID induced GI complications such as hemorrhage, perforation or death are much less common, occurring collectively with an incidence of about 2% per year (Silverstein 1995). However, due to the vast numbers of individuals using these drugs, they have been linked directly to over 70,000 hospitalizations and over 7,000 deaths annually in the United States alone (Fries 1991). The relative risk of upper GI hemorrhage or perforation with NSAID use varies in the literature from 4.7 in hospital based case control studies to 2.0 in cohort studies (Griffin 1988, Bollini 1992, McMahon 1997), and is also dependent on well recognized patient based risk factors such as advancing age, and previous peptic ulcer history therapy (Fries 1991, Bollini 1992, Gabriel 1991, Silverstein 1995, Hallas 1995, Hansen 1996, Laporte 1991, Rodriguez 1997, Hochain 1995).
1. They assess these drugs in the prevention of NSAID induced upper GI toxicity.
2. The active intervention is compared to placebo or to another of the agents thought to be effective in this setting.
3. The duration of NSAID exposure is greater than 3 weeks (Chronic NSAID exposure).
4. The proportion of patients with endoscopic ulcers or actual clinical GI events can be determined.
5. Endoscopic ulcers are defined as at least 3mm in diameter and/or can be distinguished from erosions based on the authors descriptions. Note will be made of whether endoscopy was performed based on symptoms or as part of a set protocol.
Studies will be considered ""secondary"" prophylaxis trials if they enrolled patients who had NSAID ulcers that were healed in a preceding healing phase study. ""Primary"" prophylaxis studies include patients that did not have an ulcer at an initial screening endoscopy.
Standard dose H2RA will be defined as the equivalent of ranitidine 150mg Bid, while double dose H2RA will be defined as the equivalent of ranitidine 300mg twice daily. Misoprostol daily dosages will be classified as low (400ug), medium (600ug) or high (800ug).
Short term studies ( <3 weeks) on healthy volunteers will not be considered, since the clinical importance of acute NSAID induced erosions and submucosal hemorrhages is felt to be minimal.
See: Collaborative Review Group search strategy
Published clinical trials of prostaglandin analogues for the prevention of NSAID induced upper GI toxicity will be identified through a search of MEDLINE, Current Contents, EMBASE, and the Cochrane Controlled Trials Register. The MEDLINE search will be performed in accordance with published recommendations (Haynes 1994, Hunt 1997). Content experts will be contacted for additional studies, and for information regarding ongoing clinical trials. Additionally the pharmaceutical manufacturers of the drugs of interest will be contacted for any unpublished or ongoing trial information. No language restrictions will be used.The reference lists of all potentially relevant articles including reviews, meta-analyses, consensus statements, and conference proceedings will be reviewed for the identification of potential studies.
All stages of reference collection and data extraction will be performed entirely in duplicate by two independent reviewers (AR, VW).Selection of relevant studies:
The results of the search strategies will be searched for a list of potentially relevant articles. These references will be retrieved and assessed for eligibility.
Data extraction:
For each included trial, data will be collected regarding the trial design, patient characteristics, intervention drugs, dosages, NSAIDs used, and outcomes using a predetermined extraction form. Differences will be resolved by referring back to the studies in question and establishing consensus. Any remaining differences will be resolved with the help of a gastroenterology (CD) or statistical (GW) content expert. When necessary, information will be sought from the authors of the primary studies. Included studies will be further classified as primary or secondary prophylaxis trials and by the time periods of outcome measures.
Quality assessment:
All studies will be scored for quality by two independent reviewers (AR, VW) using a modification of the scale developed by Jadad et al. (Jadad 1996). For each study the adequacy of randomization, blinding, and completeness of follow-up will be determined. Differences will be resolved by consensus.
Statistical analysis:
We will use REVMAN v3.1 to analyse the dichotomous (binary) data obtained from this review. The results will be expressed as the relative risk of the outcome with the intervention as compared with placebo or the active comparator using a fixed effects model. If necessary, a random effects model will be used if heterogeneity is detected. We will express our results as relative risks rather than the standard odds ratios since clinicians find the concept of risk easier to understand than odds. The absolute risk difference will also be presented.
Heterogeneity:
Heterogeneity will be tested using a Chi square test with (N-1) degrees of freedom, where N equals the number of trials contributing data. L Abbé plots and standard forest plots will also be used to explore the observed heterogeneity graphically (L Abbe 1987).
Publication Bias:
The existence of publication bias will be explored through the use of the inverted funnel plot technique based on the data for the primary outcome of endoscopic gastric ulcer.
Subgroup Analyses:
Subgroup analyses will be performed based on the: dose of the intervention; duration of therapy; and primary vs secondary prophylaxis trials. Analyses by dosages are particularly important for the PAs where higher doses are associated with more side-effects, and for the H2RAs where standard doses may be ineffective at reducing NSAID ulcer risk. The risk of an adverse GI event while on an NSAID is now felt to be constant over time (MacDonald 1997, Silverstein 1995, Garcia 1994). Therefore we feel it is appropriate to pool the results of studies with follow-up times of 3 months or greater together. However, patients developing NSAID complications at less than 3 months may represent a particularly susceptible population and as such will be analysed separately. The study population age is a recognized reflection of the study population s inherent risk of GI complications while on NSAIDs (Fries 1991, Bollini 1992, Gabriel 1991, Silverstein 1995, Hallas 1995, Hansen 1996, Laporte 1991, Rodriguez 1997). Therefore we aim to assess the relationship between mean age with the observed effect of the therapy on endoscopic ulcers using the techniques of meta-regression. The relationship of Helicobacter pylori (HP) to NSAID induced GI toxicity is controversial. If possible analyses based on HP status will be performed.
Sensitivity Analyses:
Sensitivity analyses will be performed to test the robustness of the results to analyse by the: study quality using each of the subsets of quality (blinding, randomization, and completeness of follow) as well as by using the mean quality score as a cut-off to define high and low quality studies; definition of endoscopic ulcer size ( > 3 mm vs > 5 mm); type of NSAIDs used; and by varying the obtained point estimates from efficacy to intention to treat and through the use of ""best"", ""worst"" and ""realist"" case estimates since we predict that protocol compliance with an end of study endoscopy may be a problem. The worst case will set all patients lost to follow up as having suffered the outcome. The best case assumes that none of the lost patients had the outcome, while the ""realistic"" case assumes that those lost to follow up suffered an event rate similar to that in the followed study population.
Additional references
Armstrong 1987 Armstrong CP, Blower AL. Nonsteroidal antiinflammatory drugs and life threatening complications of peptic ulceration. Gut 1987; 28: 527-532. Bollini 1992 Bollini P, Rodriguez G, Gutthann S. The impact of research quality and study design on epidemiologic estimates of the effect of nonsteroidal anti-inflammatory drugs on upper gastrointestinal tract disease. Archives of Internal Medicine 1992; 152: 12891295-. Fries 1990 Fries JF, Miller SR, Spitz PW, Williams CA, Hubert HB, Bloch DA. Identification of patients at risk for gastropathy associated with NSAID use. Journal of Rheumatology - Supplement 1990; 20: 12-19. Fries 1991 Fries JF. NSAID gastropathy: the second most deadly rheumatic disease? Epidemiology and risk appraisal. Journal of Rheumatology - Supplement 1991; 28: 6-10. Gabriel 1991 Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious gastrointestinal complications related to use of non-steroidal anti-inflammatory drugs: A meta-analysis. Annals of Internal Medicine 1991; 115: 787-796. Garcia 1994 Garcia RL, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual non-steroidal antiinflammatory drugs (see comments) (published erratum appears in Lancet 1994 Apr 23;343(8904):1048). Lancet 1994; 343: 769-772. Griffin 1988 Griffin MR, Ray WA, Schaffner W. Nonsteroidal anti-inflammatory drug use and death from peptic ulcer in elderly persons. Annals of Internal Medicine 1988; 109: 359-363. Hallas 1995 Hallas J, Lauritsen J, Villadsen HD, Gram LF. Nonsteroidal antiinflammatory drugs and upper gastrointestinal bleeding, identifying high-risk groups by excess risk estimates. Scandinavian Journal of Gastroenterology 1995; 30: 438-444. Hansen 1996 Hansen JM, Hallas J, Lauritsen JM, Bytzer P. Non-steroidal antiinflammatory drugs and ulcer complications: a risk factor analysis for clinical decision-making. Scandinavian Journal of Gastroenterology 1996; 31: 126-130. Haynes 1994 Haynes RB, Wilczynski N, McKibbon KA, Walker CJ, Sinclair JC. Developing optimal search strategies for detecting clinically sound studies in MEDLINE. Journal of the American Medical Informatics Association 1994; 1: 447-458. Hochain 1995 Hochain P, Berkelmans I, Czernichow P, Duhamel C, Tranvouez JL, Lerebours E, et al. Which patients taking non-aspirin nonsteroidal anti-inflammatory drugs bleed? A case-control study. European Journal of Gastroenterology & Hepatology 1995; 7: 419426-. Hunt 1997 Hunt D, McKibbon K. Locating and appraising systematic reviews. Annals of Internal Medicine 1997; 126: 532-538. Jadad 1996 Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Controlled Clinical Trials 1996; 17: 1-12. LAbbe 1987 L Abbe KA, Detsky AS, O Rourke K. Meta-analysis in clinical research. (Review) (73 refs). Annals of Internal Medicine 1987; 107: 224-233. Langman 1994 Langman MJ, Weil J, Wainwright P, Lawson DH, Rawlins MD, Logan RF, et al. Risks of bleeding peptic ulcer associated with individual non-steroidal anti-inflammatory drugs (see comments) (published erratum appears in Lancet 1994 May 21;343(8908):1302). Lancet 1994; 343: 1075-1078. Laporte 1991 Laporte JR, Carne X, Vidal X, Moreno V. Upper gastrointestinal bleeding in relation to previous use of analgesics and nonsteroidal anti-inflammatory drugs. Lancet 1991; 337: 85-89. Larkai 1987 Larkai EN, Smith Jl, Lidsky MD. Gastroduodenal mucosa and dyspeptic symptoms in arthritic patients during chronic nonsteroidal anti-inflammatory drug use. American Journal of Gastroenterology 1987; 82: 1153-1158. MacDonald 1997 MacDonald TM, Morant SV, Robinson GC. Association of upper Gastrointestinal toxicity of non-steroidal anti-inflammatory drugs with continued exposure: Cohort Study. BMJ 1997; 315: 1333-1337. Maetzel 1998 Maetzel A, Ferraz MB, Bombardier C. The cost-effectiveness of misoprostol in preventing serious gastrointestinal events associated with the use of nonsteroidal antiinflammatory drugs. Arthritis & Rheumatism 1998; 41: 16-25. McMahon 1997 McMahon AD, Evans JM, White G, Murray FE, McGilchrist MM, McDevitt DG, et al. A cohort study (with re-sampled comparator groups) to measure the association between new NSAID prescribing and upper gastrointestinal hemorrhage and perforation. Journal of Clinical Epidemiology 1997; 50: 351-356. PHS 1989 Anonymous. Steering Committee of the physicians health study research group: final report on the aspirin component of the ongoing physician s health study. N Eng J Med 1989; 321: 129-135. Patr 1991 Patr O. Aspirin as an anti-platelet drug. N Eng J Med 1994; 330:1287-1294.Stroke prevention in atrial fibrillation investigators. Stroke prevention in atrial fibrillation study: final results. Circulation 1991; 84: 527-539. Rodriguez 1997 Rodriguez LA. Nonsteroidal anti-inflammatory drugs, ulcers and risks: A collaborative meta-analysis. Seminars in Arthritis & Rheumatism 1997; 26: 16-20. SALT 1991 The SALT collaborative study group. Swedish aspirin low-dose trial of 75mg aspirin as secondary prophylaxis after cerebrovascular ischemic events. Lancet 1991; 338: 1345-49. Silverstein 1995 Silverstein FE, Graham DY, Senior JR, Davies HW, Struthers BJ, Bittman RM, et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A randomized, double-blind, placebo-controlled trial (see comments). Annals of Internal Medicine 1995; 123: 241-249. Smalley 1996 Smalley WE, Griffin MR, Fought RL, Ray WA. Excess costs from gastrointestinal disease associated with nonsteroidal antiinflammatory drugs. Journal of General Internal Medicine 1996; 11: 461-469. Stalnikowicz 1993 Stalnikowicz R, Rachmilewitz D. NSAID-induced gastroduodenal damage: is prevention needed? A review and metaanalysis. Journal of Clinical Gastroenterology 1993; 17: 238-243. Stroke 1991 Stroke prevention in atrial fibrillation investigators. Stroke prevention in atrial fibrillation study: final results. Circulation 1991; 84: 527-539. Wallace 1997 Wallace JL. Nonsteroidal anti-inflammatory drugs and gastroenteropathy: the second hundred years. Gastroenterology 1997; 112: 1000-1016. Extramural sources of support to the review Intramural sources of support to the reviewReviewer(s) Rostom A, Welch V, Wells G, Tugwell P, Dubé C, Lanas A, McGowan J Date of most recent amendment 24 February 1999 Date of most recent substantive amendment 18 November 1998 Date review expected 01 October 1999 Contact address Dr Alaa Rostom MD, FRCPC
Division of Gastroenterology
University of Ottawa
A1 - Endoscopy Unit, Ottawa Hospital - Civic Campus
1053 Carling Ave.
Ottawa
Ontario
CANADA
K1Y-4E9
Telephone: 613-798-5555 extension: 6167
Facsimile:
E-mail: alrostom@cyberus.caCochrane Library number CD001448 Editorial group Cochrane Upper Gastrointestinal and Pancreatic Diseases Group Editorial group code HM-UPPERGI This protocol should be cited as :
Rostom A, Welch V, Wells G, Tugwell P, Dubé C, Lanas A, McGowan J. Prostaglandin analogues, H2-receptor antagonists and proton pump inhibitors for the prevention of chronic NSAID induced upper gastrointestinal toxicity in adults (Protocol for a Cochrane Review). In: The Cochrane Library, Issue 2, 1999. Oxford: Update Software.Sources of support
Comment, Reply and Editorial notes
Review produced in collaboration with the Musculoskeletal Group.