U.S. Preventive Services Task Force banner
U.S. Preventive Services Task Force

Nonsteroidal Anti-inflammatory Drugs and Cyclooxygenase-2 Inhibitors for Primary Prevention of Colorectal Cancer

Systematic Review

Alaa Rostom, M.D., M.Sc.a; Catherine Dubé, M.D., M.Sc.a; Gabriela Lewin, M.D.b; Alexander Tsertsvadze, M.D., M.Sc.b; Nicholas Barrowman, Ph.D.b; Catherine Code, M.D.c; Margaret Sampson, M.L.I.S.b; and David Moher, Ph.D.b

The authors of this article are responsible for its contents, including any clinical or treatment recommendations. No statement in this article should be construed as an official position of the U.S. Agency for Healthcare Research and Quality (AHRQ) or the U.S. Department of Health and Human Services.

Address correspondence to: Alaa Rostom M.D., M.Sc (Epi), FRPC, Division of Gastroenterology, University of Calgary Medical Clinic, 3330 Hospital Drive, NW G176, Calgary, Alberta, Canada T2N 4N1, E-mail: arostom@calgary.ca

This systematic review was first published in the Annals of Internal Medicine. Select for copyright and source.




Purpose: To examine the benefits and harms of nonaspirin (non-ASA) nonsteroidal anti-inflammatory drugs (NSAIDs) and cyclooxygenase (COX-2) inhibitors for the prevention of colorectal cancer (CRC) and adenoma.

Data Sources: MEDLINE® (1966 to 2006), EMBASE (1980 to 2006), Cochrane Central Register of Controlled Trials, Cochrane Collaboration's registry of clinical trials, Cochrane Database of Systematic Reviews.

Study Selection: Randomized, controlled trials and case-control and cohort studies of the effectiveness of NSAIDs for the prevention of CRC and colorectal adenoma were identified by multilevel screening by 2 independent reviewers. Systematic reviews of harms were sought.

Data Extraction: Data abstraction, checking, and quality assessment were completed in duplicate.

Data Synthesis: A single cohort study showed no effect of non-ASA NSAIDs on death due to CRC. Colorectal cancer incidence was reduced with non-ASA NSAIDs in cohort studies (relative risk, 0.61 [95% CI, 0.48 to 0.77]) and case-control studies (relative risk, 0.70 [CI, 0.63 to 0.78]). Colorectal adenoma incidence was also reduced with non-ASA NSAID use in cohort studies (relative risk, 0.64 [CI, 0.48 to 0.85]) and case-control studies (relative risk, 0.54 [CI, 0.4 to 0.74]) and by COX-2 inhibitors in randomized, controlled trials (relative risk, 0.72 [CI, 0.68 to 0.77]). The ulcer complication rate associated with non-ASA NSAIDs is 1.5% per year. Compared with non-ASA NSAIDs, COX-2 inhibitors reduce this risk but, in multiyear use, have a higher ulcer complication rate than placebo. Cyclooxygenase-2 inhibitors and nonnaproxen NSAIDs increase the risk for serious cardiovascular events (relative risk, 1.86 [CI, 1.33 to 2.59] for COX-2 inhibitors vs. placebo).

Limitations: Heterogeneity in the dose, duration and frequency of use necessitated careful grouping for analysis.

Conclusions: Cyclooxygenase-2 inhibitors and NSAIDs reduce the incidence of colonic adenomas. Nonsteroidal anti-inflammatory drugs also reduce the incidence of CRC. However, these agents are associated with important cardiovascular events and gastrointestinal harms. The balance of benefits to risk does not favor chemoprevention in average-risk individuals.

Return to Contents


In the United States, cancer is the second leading cause of death after heart disease and is the leading cause of death in persons younger than 65 years of age. Colorectal cancer (CRC) is the second and third leading cause of cancer-related deaths in men and women, respectively, and overall, is the third most common type of cancer in men and women. In 2006, it was estimated that 148,610 new cases of CRC occurred and that 51,170 patients died of the disease.1,2

The U.S. Preventive Services Task Force (USPSTF) strongly recommends screening for men and women 50 years of age or older for CRC ("A" recommendation).3 Biannual fecal occult blood testing can reduce CRC-related death by 21%, and it has been reported that flexible sigmoidoscopy reduces death by 60% for lesions within reach of the instrument. Further, data suggest that sigmoidoscopy followed by colonoscopy when polyps are found could decrease CRC incidence by up to 80%.4 Despite evidence of the effectiveness of several screening methods, adoption of routine CRC screening by eligible individuals, using any method, continues to be low in the United States.5-8

A CRC chemoprophylactic strategy may be used as a complement to or instead of a screening strategy. Several basic science, population-based, and experimental studies have suggested a protective effect of aspirin (ASA) and non-ASA nonsteroidal anti-inflammatory drugs (NSAIDs), including cyclooxygenase (COX)-2 inhibitors, on colorectal adenomas and CRC. However, 2 long-term, randomized, controlled trials, the Physicians' Health Study9 and the Women's Health Study,10 did not show a beneficial effect of low-dose ASA on CRC incidence. Furthermore, these agents are not without harms. Clinically significant gastrointestinal hemorrhage can occur with all of these agents, although it is substantially lower with COX-2 inhibitors.

More recently, interest has focused on a potentially prothrombotic effect of selective COX-2 inhibitors and nonnaproxen NSAIDs. In fact, during the conduct of our systematic review, 2 COX-2 inhibitors (rofecoxib and valdecoxib) were withdrawn from the U.S. market because of concerns about their cardiovascular toxicity, leaving only celecoxib remaining and uncertainty about the future of others, such as lumiracoxib and etoricoxib. These developments have resulted in uncertainty about the safety of COX-2 inhibitors and non-ASA NSAIDs when used longterm, such as in the setting of CRC prevention.11

At the request of the Agency for Healthcare Research and Quality (AHRQ), the Centers for Disease Control and Prevention (CDC), and the USPSTF, we conducted a systematic review to ascertain the effectiveness of non-ASA NSAIDs and COX-2 inhibitors in the chemoprevention of colorectal adenomas, CRC, and CRC-related death in average- to higher-risk individuals. We also examined the harms associated with these agents.

Return to Contents


Data Sources

We developed the search strategy in MEDLINE® and modified it for other databases. The search was limited to English-language reports of human studies. We searched the following databases: MEDLINE® (1966 to December [week 4] 2006), EMBASE (1980 to the 14th week of 2005 publication years 2003 to 2005), Cochrane Central Register of Controlled Trials (CENTRAL), and Cochrane Library Issue 4, 2004. Beyond these dates, we surveyed several sources to ascertain additional potentially eligible studies. PubMed Cancer subset was searched for non-MEDLINE® material.

Search terms were derived from the National Cancer Institute (NCI) Cancer topic searches for "colorectal cancer" and "adenomatous polyps." We derived a comprehensive retrieval strategy from the indexing in MEDLINE® and EMBASE, investigator-nominated terms, and previous reviews.12-14

We developed a search strategy in MEDLINE® (2003 to the third week of December 2006) to detect recent systematic reviews that appeared to address the harms of non-ASA NSAIDs and COX-2 inhibitors. We implemented a weekly monitoring strategy to detect emerging information on cardiovascular harms associated with COX-2 inhibitors. We also monitored the U.S. Food and Drug Administration News Digest and Health Canada's Health Product Information mailing list for announcements related to COX-2 inhibitors and cardiovascular harms (monitoring dates, 14 January 2005 to 26 May 2005). Beyond these dates, we surveyed several sources to ascertain additional potentially eligible studies.

Study Selection

Citation records were screened to identify potentially relevant articles and retained records were assessed for relevance to identify articles meeting inclusion criteria. A third screening phase was included to discriminate between the different study designs. At each screening stage, 2 members of the review team selected articles for inclusion after an initial calibration exercise. Conflicts were resolved by consensus.

We considered randomized, controlled trials (RCTs); controlled, clinical trials; and observational studies (cohort and case-control studies) of the efficacy of non-ASA NSAIDs and COX-2 inhibitors for inclusion if they fulfilled the population and outcome criteria.

We considered studies for inclusion if participants were at average risk for CRC (that is, no known risk factors for colorectal adenoma or CRC, other than age). We also considered studies of higher-risk individuals with a personal or family history of colorectal adenoma or a family history of sporadic CRC. Included studies addressed the incidence of colorectal adenomas, CRC, or both and CRC-related death or overall death. We excluded studies of high-risk patients with familial adenomatous polyposis or hereditary nonpolyposis colon cancer syndromes (Lynch I or II) and secondary prevention studies of patients with a personal history of CRC.

We sought existing systematic reviews to address the gastrointestinal, cardiovascular, and renal harms associated with the use of non-ASA NSAIDs and COX-2 inhibitors considering the number of reviews already done on these topics.

Data Extraction and Quality Assessment

Several members of the team extracted data independently by using a Web-based system (SRS 4.0, TrialStat Corp., Ottawa, Ontario, Canada). We extracted data by using the PICOS (participant, intervention and exposure, comparator, outcome, and study design) approach.

We used predefined criteria from the USPSTF to assess the quality of included systematic reviews, clinical trials, and observational studies, which we rated as good, fair, or poor.11 This scale relies on 4, 6, 7, and 7 criteria for systematic reviews, case-control studies, cohort studies, and RCTs, respectively. A good rating was given when all criteria were met; a fair rating when at least 80% were met and the study had no fatal flaws; and a poor rating when less than 80% of the criteria were met, when there was a fatal flaw, or both.

Data Synthesis and Analysis

We used an analytical framework to facilitate study grouping and subsequent data analysis in an effort to minimize clinical heterogeneity. We initially grouped studies by disorder (that is, colorectal adenoma or CRC), study design, study population, and medication exposure and subsequently subcategorized studies based on measures of dose effect, duration of exposure, and secondary outcomes (when reported). Definition of categories, such as "regular use," can be found elsewhere.11

We summarized and presented harms data from the included systematic reviews as a qualitative synthesis. We combined results numerically only if clinically and statistically appropriate. We chose relative risk as the effect measure. In case-control studies, a direct estimate of the relative risk is not possible. However, when event rates are low, as was the case in our review, the odds ratio provides a close approximation of the relative risk. In what follows, we simply refer to the relative risk. We assessed heterogeneity by using the I2 statistic. We combined studies when I2 was 50% or less.15 We directly abstracted point estimates of the adjusted relative risks and their 95% CIs from the reports of primary studies. One source of heterogeneity may be study-to-study variation in the method of selecting confounders for which to adjust and the final set of confounders chosen.

In Appendix Tables 1 and 2, we summarize these characteristics for each study. Further, a detailed discussion of the methodological considerations is presented in the USPSTF report.11 We computed standard errors by dividing the CI width by 2 X 1.96. We conducted quantitative synthesis by using inverse variance weighting and a random-effects model.17

Role of the Funding Sources

The evidence synthesis on which this article was based was funded by the CDC, AHRQ, and the USPSTF. Its design, conduct, and reporting was based on specific directives from these agencies.

Return to Contents


Study Selection

Our literature search yielded 1790 potentially relevant bibliographic records that addressed the use of ASA, COX-2 inhibitors, and other non-ASA NSAIDs.11 For non-ASA NSAIDS, we retrieved 364 articles for relevance assessment, and 29 studies met final inclusion criteria. One study of rofecoxib18 and 2 studies of celecoxib19,20 were published after completion of the task force report.11 and we include them herein.

A CRC-related death in 1 cohort study21 was reported. The chemoprophylaxis of CRC was addressed in 10 case-control studies22-31 and 3 cohort studies.32-34 The chemoprophylaxis of colorectal adenoma was addressed in 10 case-control studies,31,35-43 1 cohort study,44 and 4 RCTs.18-20,45

The Figure46,47 describes the flow of reports through our review, and Appendix Tables 3, 4, and 5 describe the included studies. A table of duplicate and companion articles is available in the AHRQ report.11

Study Quality and Methodological Considerations

The understanding of the important sources of heterogeneity among the included observational studies is key to interpreting the results of this review and the ASA review,11 also in this issue of Annals of Internal Medicine. This was discussed in detail elsewhere,11 and we present it here in brief. We produced an a priori, hierarchical framework that identified key characteristics that were expected to be common to all the included studies. We used this framework to facilitate study grouping and subsequent data analysis.

We anticipated certain key characteristics, such as the dose across studies, to show important heterogeneity. Measuring the dose effect depended on the intervention dose, the frequency and duration of use, and whether the use was current and ongoing or had occurred at some time in the past. For example, some studies defined specific dose levels, whereas in other studies, researchers reported dose effect in terms of frequency of use, such as number of pills per week or prescription refills in a given time period, thereby combining the effects of dose and duration. One way to handle this inconsistency across studies was to define regular use and specific duration intervals in the developed framework11 to group studies with similar dose effects. Other sources of inconsistency also existed, such as the methods and timing of ascertainment of exposure (for example, questionnaires, patient records, and databases) and outcome (for example, colonoscopy, patient records, and databases). Lastly, the type of NSAIDs used varied among studies between non-ASA NSAIDs alone, ASA included among NSAIDs (herein referred to as "any NSAIDs"), and COX-2 inhibitors alone. We analyzed the data separately for each of these 3 types of exposures. In some situations, individual study differences precluded statistical pooling.

The quality of the included studies was good for 3 of the 4 RCTs, good to fair for the 5 cohort studies, and fair for most of the case-control studies (5 good, 11 fair, and 4 poor).

Colorectal Cancer Mortality

A single cohort study of fair quality assessed the effect of ibuprofen on CRC mortality.21 The study used an administrative database to identify 113,538 participants who filled at least 1 ibuprofen prescription over a 6-year period. A statistically significant increase in all-cause mortality was observed with ibuprofen, but no effect on death due to bowel or rectal cancer was observed (Table 1).

Colorectal Cancer Incidence

Table 1 summarizes the effects of regular use of non-ASA NSAIDs on CRC incidence. The available data are limited to observational studies.

Cohort Studies

Three cohort studies assessed the effect of non-ASA NSAIDs on CRC incidence.32-34 The Nurses' Health Study34 was a large, good-quality, 20-year prospective followup of average-risk U.S. women.34 It showed a statistically significant dose-dependent protective effect of non-ASA NSAIDs on CRC. The magnitude of the relative risk reduction was up to 30% in colon cancer, whereas no benefit was observed for rectal cancer alone. When specific dose subgroups were analyzed, patients receiving less than 6 tablets per week or those receiving non-ASA NSAIDs irregularly did not seem to show a reduction in CRC incidence. Two other large administrative database studies of fair quality showed a statistically significant protective effect of regular non-ASA NSAIDs on the incidence of CRC.32,33

Case-Control Studies

The regular use of non-ASA NSAIDs and of any NSAIDs was associated with statistically significant reductions in CRC frequency in the pooled analyses (relative risk, 0.70 for non-ASA NSAIDs22,23,25,49 vs. 0.57 for any NSAIDs26-29,31). Two other case-control studies (1 large prescription database study of good quality30 and 1 study of fair quality24) demonstrated statistically significant reductions in CRC frequency, but their method of quantifying regular NSAID use prevented statistical pooling with the other studies.

Dose and Duration of Use

In cohort studies33,34 and case-control studies22,30,31 higher dose levels of any NSAIDs were generally associated with statistically significant relative risk reductions in CRC frequency, whereas lower dose levels were not (Tables 2 and 3 ). Two studies of fair quality25,28 demonstrated inconsistent dose effects, which may be due to underpowered subgroup analyses.

Similarly, longer durations of non-ASA NSAID use (that is, beyond 2 to 5 years) generally resulted in statistically significant reductions in risk for CRC, whereas lower durations of use did not.22,25,26,30 The largest and best-quality study in the group demonstrated a statistically significant reduction in risk for CRC with non-ASA NSAID use of at least 11 years but not for shorter durations.30 Small studies of poor quality did not demonstrate a consistent duration effect.27,31

Colorectal Adenoma

Randomized, Controlled Trials

In patients with a history of colorectal adenomas, 3 recent, good-quality RCTs on COX-2 inhibitor (celecoxib19,20 and rofecoxib18) demonstrated statistically significant reductions in the incidence of all adenomas and advanced adenomas over a 3-year followup (pooled relative risk, 0.72 [95% CI, 0.68 to 0.77] vs. 0.56 [CI, 0.42 to 0.75], respectively) (Table 1). A nonsignificant trend was observed toward a greater relative risk reduction in advanced versus all adenomas for celecoxib.19 However, patients with advanced adenoma seemed to derive less benefit from rofecoxib than those without advanced adenomas.18 Patients also seemed to have a reduced benefit with rofecoxib over time. Further, in a small subgroup of randomly assigned patients who agreed to undergo colonoscopy in year 4 post-study completion, patients in the rofecoxib group had a higher risk for adenomas than those in the placebo group, suggesting a possible rebound effect.18

Another small RCT (45) of fair quality found that 4 months of sulindac, 30 mg/d (non-ASA NSAID), did not cause a statistically significant regression of colorectal adenomas (.1.0 cm), which were initially identified by using flexible sigmoidoscopy.

Cohort Studies

In a single cohort study of good quality,44 regular use of any NSAID significantly reduced the incidence of colorectal adenomas in patients with a history of colorectal adenoma (relative risk, 0.64 [CI, 0.48 to 0.85).

Case-Control Studies

The regular use of non-ASA NSAIDs36,38,50,51 and any NSAID31,36,41-43 in average-risk individuals was associated with statistically significant reductions in frequency of colorectal adenoma (relative risk, 0.54 [CI, 0.4 to 0.74] vs. 0.57 [CI, 0.46 to 0.71], respectively).

Dose and Duration of Use

A nonstatistically significant trend for greater reduction in adenoma incidence was observed with celecoxib, 800 mg/d, compared with celecoxib, 400 mg/d, in 1 RCT.19 In 3 case-control studies,31,37,42 higher NSAID doses were associated with statistically significant reductions in frequency of colorectal adenoma, whereas lower doses were not (Table 2).

The use of any NSAID had less consistent duration effects on adenoma prevention than on CRC prevention. Two studies36,39 demonstrated statistically significant reductions in adenoma frequency with the use of any NSAID for at least 5 years, whereas another study42 demonstrated a nonsignificant trend toward greater adenoma reduction with more than 19 years of use of any NSAID compared with fewer than 10 years of use of any NSAID. The remaining studies31,38,41 demonstrated inconsistent results mostly because of underpowered subgroup analyses (Table 3).

Harms Due to Non-ASA NSAIDs and COX-2 Inhibitors

All-Cause Mortality

Three reviews52-54 reported no statistically significant differences in all-cause mortality between different NSAIDs or between NSAIDs and placebo. Compared with placebo, neither less selective COX-2 inhibitors (etodolac, meloxicam, nabumetone, or nimesulide) used in 51 RCTs (relative risk, 0.68 [CI, 0.3 to 1.6]) nor selective COX-2 inhibitors (celecoxib and rofecoxib) used in 17 RCTs (relative risk, 1.02 [0.6 to 1.9]) were associated with a difference in mortality.52 No deaths were reported in 3 RCTs comparing celecoxib with placebo or other NSAIDs,53 and mortality rates were similar between rofecoxib (0.5%) and naproxen (0.4%) in the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial54 and between rofecoxib (0.93%) and placebo (0.92%) in the Adenomatous Polyp Prevention on Vioxx (APPROVe) trial18 However, 1 administrative database study of fair quality21 and a systematic review using a biologic progression model55 found a small, statistically significant increase in all-cause mortality with non-ASA NSAIDs.

Cardiovascular Harms

Eight systematic reviews53,54,56-61 addressed the magnitude of cardiovascular harms associated with the use of COX-2 inhibitors. They reported on RCT data, thereby providing high-level evidence, and 1 review59 also included observational studies. Two of the reviews56,59 extracted cardiovascular harms of non-ASA NSAIDs. Cardiovascular events reported across the systematic reviews included death due to such events, serious cardiovascular events (overall), acute myocardial infarction (MI), acute stroke, arterial hypertension, congestive heart failure, edema, and thrombotic events (Table 4).

Four reviews found no significant differences in death due to cardiovascular events with the use of a COX-2 inhibitor compared with placebo, nonnaproxen NSAIDs, or naproxen.53,54,56,59

Three reviews56,58,59 reporting overall serious cardiovascular events consistently demonstrated an excess risk for these events with the use of COX-2 inhibitors compared with use of placebo or naproxen. The risk for cardiovascular events was greatest in patients at high risk for such events (patients for whom aspirin is indicated).58 The risk associated with the use of nonnaproxen non-ASA NSAIDs (mostly high-dose diclofenac and ibuprofen) seemed similar to that shown with the use of COX-2 inhibitors.

Six reviews54,56,58-61 reported on the risk for acute MI in patients taking COX-2 inhibitors or non-ASA NSAIDs. The results consistently demonstrated statistically significant increases in the relative risk for MI with the use of COX-2 inhibitors compared with placebo or naproxen. High-dose, nonnaproxen, non-ASA NSAIDs (mostly diclofenac and ibuprofen) seemed to have a similar risk for MI as that of COX-2 inhibitors.56 One of the identified reviews59 showed a statistically significant protective effect of naproxen on MI; however, that analysis demonstrated significant heterogeneity.

Five reviews reported on acute stroke.54,56,58,59,61 The results consistently showed no statistically significant increased risk for stroke with COX-2 inhibitors compared with placebo, nonnaproxen NSAIDs, or naproxen. One high-quality review56 demonstrated a statistically significant lower risk for acute stroke with COX-2 inhibitors than nonnaproxen NSAIDs in an analysis primarily driven by the effect of high-dose diclofenac.

The risks for hypertension and renal toxicity may also be elevated with COX-2 inhibitors and are reported elsewhere.62

Gastrointestinal Harms

The included systematic reviews of the gastrointestinal harms of NSAIDs summarized data from RCTs,12,50,55,63,64 cohort studies,55,65,66 and case-control studies.55,63,65 Two of the systematic reviews of RCTs12,52 focused primarily on prevention of NSAID-induced upper gastrointestinal toxicity through the use of prophylactic agents or the use of COX-2 inhibitors. One of these12 reported the rate of gastrointestinal complications in patients taking NSAIDs. Twelve systematic reviews assessed COX-2 inhibitors with data on celecoxib,12,52,53,66-69 rofecoxib,12,50,54,57,61 valdecoxib,60,70,71 and meloxicam.12,50,72 Rostom and colleagues12 updated their COX-2 inhibitor review to include data for lumiracoxib, valdecoxib, and etoricoxib. The updated review is currently in press, and the pooled estimates remain similar to those presented here.

All of the included studies reported an increased risk for peptic ulceration and gastrointestinal hemorrhage with non-ASA NSAID use. The risk for complicated peptic ulcers (perforation, obstruction, or bleeding) in those receiving NSAIDs compared with those who were not was elevated in pooled analyses for RCTs (odds ratio, 5.36 [CI, 1.79 to 16.1]), cohort studies (relative risk, 2.7 [CI, 2.1 to 3.5]), and case-control studies (odds ratio, 3.0 [CI, 2.5 to 3.7]).63 The best RCT evidence of the risk for perforation, obstruction, or bleeding with NSAIDs was derived from the original Misoprostol Ulcer Complications Outcome Safety Assessment (MUCOSA) study12,13,73 and corroborated with recent data from the NSAID groups of the COX-2 inhibitor trials.12,74-76 A risk for perforation, obstruction, and bleeding of approximately 1.5% to 2% per year was observed in average-risk individuals taking standard non-ASA NSAIDs. The risk for perforation, obstruction, or bleeding can reach 10% or more in higher-risk individuals, including those who have had previous peptic ulcers; who are older; and who have comorbid conditions, such as cardiovascular disease.12,13,70,74

We estimated the absolute risk difference of perforation, obstruction, or bleeding for patients taking NSAIDs compared with those not taking NSAIDs to be 0.48% for the included RCTs and 0.22% for the included cohort studies.

The risk for upper gastrointestinal toxicity due to non-ASA NSAID use can be reduced through the use of a concomitant gastroprotective agent. Misoprostol was associated with a statistically significant 40% relative risk reduction in clinical ulcer complications due to combined NSAID use.12,13,73 Histamine-2-receptor antagonists (H2RAs) and proton-pump inhibitors have only been evaluated in endoscopic ulcer studies.12,13 Double-dose H2RAs (equivalent to ranitidine, 300 mg twice daily) and standard dose proton-pump inhibitors were associated with statistically significant reductions in the risk for NSAID-induced duodenal and gastric ulcers. Standard-dose H2RAs were not effective at reducing the risk for NSAID-induced gastric ulcers.12,13

The use of a COX-2 inhibitor compared with a non-ASA NSAID (ibuprofen, diclofenac, or naproxen) results in statistically significant relative risk reductions for the following: the incidence of endoscopically detected gastroduodenal ulcers by approximately 75%;12,52-54,61,66,67 clinically significant ulcer complication (perforation, obstruction, or bleeding and symptomatic ulcers) by 40% to 60%;12,52-54,61,66,67,70,72 and gastrointestinal symptoms, such as dyspepsia.12,52,61,67,71 The effects were similar when non-ASA NSAIDs were pooled and when each was compared separately with COX-2 inhibitors.12

In several systematic reviews, no statistically significant difference in gastrointestinal bleeding or ulceration was reported when COX-2 inhibitors were compared with placebo.12,53,54,60,67,70 However, 1 review66 showed that patients receiving celecoxib, 200 mg/d, were not at an increased risk for endoscopic ulcers compared with those receiving placebo but patients receiving celecoxib, 400 mg/d, were at increased risk (relative risk, 2.35; CI, 1.02 to 5.38).66 Compared with placebo, rofecoxib was associated with a statistically significant increased risk for total adverse events (relative risk, 1.32 [CI, 1.11 to 1.56]) and total gastrointestinal events accrued at 6 weeks (relative risk, 3.39 [CI, 1.47 to 7.84]).61 The APPROVe study found that the risk for symptomatic ulcer, bleeding, perforation or obstruction was higher with rofecoxib than with placebo over a 3-year followup period (relative risk, 4.9 [CI, 1.98 to 14.5]).

The Celecoxib Long-term Arthritis Safety Study (CLASS),74 Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET) (lumiracoxib),76 and the valdecoxib trial70 assessed the use of a COX-2 inhibitor in a subgroup of patients receiving ASA. In patients taking ASA the frequency of clinically important ulcer complications was not different in those who received COX-2 inhibitors or non-ASA NSAIDs. The combination of ASA and celecoxib resulted in a 4-fold increase in ulcer complications over celecoxib alone,12,74 and the combination of valdecoxib and ASA resulted in a 9-fold increase in ulcer complications over valdecoxib alone.70 Although the data for these estimates were derived from post hoc subgroup analyses and may be subject to important bias, one needs to keep them in mind when considering a strategy for combining a COX-2 inhibitor with ASA for cardioprotection.

Return to Contents
Proceed to Next Section


USPSTF Program Office   540 Gaither Road, Rockville, MD 20850