SEARCH

SEARCH BY CITATION

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Aim : In a predefined analysis, data were pooled from eight blinded, randomized, controlled trials, and separately from three long-term, open-label trials to determine the rate of upper gastrointestinal ulcer complications with the cyclo-oxygenase-2 selective inhibitor, valdecoxib, vs. non-selective non-steroidal anti-inflammatory drugs.

Methods : In randomized, controlled trials, 7434 osteoarthritis and rheumatoid arthritis patients received placebo (n = 973), valdecoxib 5–80 mg daily (n =4362), or a non-selective non-steroidal anti-inflammatory drug (naproxen, ibuprofen or diclofenac; n =2099) for 12–26 weeks. In long-term, open-label trials, 2871 patients received valdecoxib 10–80 mg daily for up to 1 year. All potential events were reviewed by a blinded, independent review committee based on a priori definitions of ulcer complications (perforations, obstructions, bleeds).

Results : In randomized, controlled trials, 19 of 955 potential events were adjudicated to be ulcer complications. Valdecoxib was associated with a significantly lower ulcer complication rate than non-selective non-steroidal anti-inflammatory drugs (0.68% vs. 1.96%, all patients; 0.29% vs. 2.08%, non-aspirin users; P < 0.05). In long-term, open-label trials, seven of 310 potential events were adjudicated to be ulcer complications; the annualized incidence for valdecoxib was 0.39% (seven of 1791 patient-years) for all patients and 0.2% (three of 1472 patient-years) for non-aspirin users.

Conclusions : Valdecoxib, including above recommended doses, is associated with a significantly lower rate of upper gastrointestinal ulcer complications than therapeutic doses of non-selective non-steroidal anti-inflammatory drugs.


Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Non-selective non-steroidal anti-inflammatory drugs (NSAIDs) are widely used to treat acute pain, and the inflammation and pain associated with osteoarthritis (OA) and adult onset rheumatoid arthritis (RA).1 However, non-selective NSAIDs are associated with the development of endoscopic gastroduodenal ulcers and upper gastrointestinal (UGI) ulcer complications, including bleeding, perforations and gastric outlet obstruction.2–18 Non-selective NSAIDs inhibit both cyclo-oxygenase (COX)-1 and -2 iso-enzymes.19, 20 While the roles of COX-1 and -2 in initiating and mediating pain and inflammation are evolving and controversial,21 COX-2 inhibition has been forwarded as only one of many mechanisms explaining the analgesic and anti-inflammatory efficacy of non-selective NSAIDs.19–22 The UGI injury associated with non-selective NSAID use is thought to be, in part, the consequence of reduced protective mucosal prostaglandin production resulting from COX-1 inhibition.23–26 Non-selective COX-1 inhibition also inhibits platelet thromboxane production, leading to impaired platelet aggregation.8, 27, 28 This may possibly add to an increased risk of bleeding, including gastrointestinal bleeding.29 Based on observational and prospective, randomized trials, the long-term use of non-selective NSAIDs is associated with an expected 1–2% annualized rate of ulcer complications.2, 30–36

Cyclo-oxygenase-2 selective inhibitors, such as celecoxib, rofecoxib, etoricoxib, lumiracoxib and valdecoxib, are as clinically effective as non-selective NSAIDs in the treatment of the signs and symptoms of OA and RA.37–45 In contrast to non-selective NSAIDs, and consistent with their COX-1 sparing properties, COX-2 selective inhibitors, like placebo, do not inhibit platelet aggregation.28,46–49 Furthermore, and of clinical relevance, in prospective, blinded, head-to-head comparisons with non-selective NSAIDs, celecoxib and rofecoxib have been associated with a reduced incidence of UGI ulcer complications (perforations, obstructions and bleeding) and symptomatic ulcers.32–35, 50, 51

The COX-2 selective inhibitor, valdecoxib, is indicated for the relief of OA and RA signs and symptoms.11, 43, 44 The approved therapeutic doses of valdecoxib for OA and RA are 10 mg once daily (qd) in the United States and up to 20 mg qd in other regions of the world.

The UGI safety profiles of celecoxib and rofecoxib have been demonstrated in clinical trials.32, 33 Valdecoxib similarly demonstrates COX-2 selectivity,11, 43, 44, 52, 53 and is associated with lower rates of endoscopic ulcers compared with non-selective NSAIDs studied.11, 13 Therefore, it was hypothesized that the use of valdecoxib would also result in a significantly lower incidence of UGI ulcer complications, compared with non-selective NSAIDs. Therefore, in the present prospectively defined, adjudicated pooled analysis of various safety and efficacy trials, the annualized rates of ulcer complications were compared in eight randomized, controlled trials (RCTs) of OA and RA patients taking a range of therapeutic and above recommended doses of valdecoxib, therapeutic doses of non-selective NSAIDs [naproxen 500 mg b.d., ibuprofen 800 mg three times daily (tid) or diclofenac 75 mg b.d.] or placebo for 12–26 weeks. In addition, in a separate predefined analysis, the annualized rates of UGI ulcer complications were pooled from three long-term, open-label safety trials in which patients were treated with valdecoxib 10–80 mg daily for up to 1 year.

Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Investigator-reported suspected UGI ulcer complications and selected UGI-related adverse events were collected from 11 trials involving patients with either OA or RA (Table 1). Eight of the trials were blinded RCTs involving treatment with valdecoxib, placebo and a non-selective NSAID.11, 43, 44, 54 Three of the studies were long-term, open-label trials (LTOLs), in which all arthritis patients received valdecoxib at various therapeutic and above recommended doses. All studies were conducted at multiple sites under the direction of a site-specific primary investigator. The data regarding the endpoint of ulcer complications for individual studies have not been published previously.

Table 1.  Design features of the eight randomized, controlled and three long-term, open-label trials
StudyNumber of patients randomizedDosage groupsDuration of treatment (weeks)
  1. * Study 8 is an ongoing 52-week trial, and 26-week data are presented.

  2. † Patients enrolled in study 10 had previously participated in either study 4 or study 5.

  3. ‡ Study 11 was a continuation of study 6.

  4. UGI, upper gastrointestinal; OA, osteoarthritis; RA, rheumatoid arthritis.

Phase III, randomized, controlled trials
 (1) UGI safety in OA of the hip541052Placebo (n = 210) Valdecoxib 10 mg qd (n = 204) Valdecoxib 20 mg qd (n = 219) Ibuprofen 800 mg tid (n = 207) Diclofenac 75 mg b.d. (n = 212)12
 (2) Pivotal efficacy and UGI safety in OA of the hip43467Placebo (n = 118) Valdecoxib 5 mg qd (n = 120) Valdecoxib 10 mg qd (n = 111) Naproxen 500 mg b.d. (n = 118)12
 (3) Pivotal efficacy and UGI safety in OA of the knee111019Placebo (n = 205) Valdecoxib 5 mg qd (n = 201) Valdecoxib 10 mg qd (n = 206) Valdecoxib 20 mg qd (n = 202) Naproxen 500 mg b.d. (n = 205)12
 (4) Pivotal efficacy and UGI safety in RA441090Placebo (n = 222) Valdecoxib 10 mg qd (n = 209) Valdecoxib 20 mg qd (n = 212) Valdecoxib 40 mg qd (n = 221) Naproxen 500 mg b.d. (n = 226)12
 (5) Pivotal efficacy and UGI safety in RA1093Placebo (n = 220) Valdecoxib 10 mg qd (n = 226) Valdecoxib 20 mg qd (n = 219) Valdecoxib 40 mg qd (n = 209) Naproxen 500 mg b.d. (n = 219)12
 (6) UGI and renal safety in OA and RA1218Valdecoxib 20 mg b.d. (n = 399) Valdecoxib 40 mg b.d. (n = 404) Naproxen 500 mg b.d. (n = 415)26
 (7) Pivotal efficacy and UGI safety in RA722Valdecoxib 20 mg qd (n = 246) Valdecoxib 40 mg qd (n = 237) Diclofenac 75 mg b.d. (n = 239)26
 (8) Pivotal efficacy and UGI safety in OA of the knee or hip*784Valdecoxib 10 mg qd (n = 259) Valdecoxib 20 mg qd (n = 261) Diclofenac 75 mg b.d. (n = 264)26
Long-term, open-label safety trials
 (9) OA1167Valdecoxib 10 to 20 mg qd (n = 1167)48 weeks
 (10) RA†1306Valdecoxib 10 to 40 mg qd (n = 1306)Up to 15 months
 (11) OA and RA‡398Valdecoxib 40 mg b.d. (n = 398)Up to 12 months

Randomized, controlled trials

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Studies 1–8 were multicentre, randomized, double-blind, parallel-group trials that were designed primarily to assess efficacy and/or safety in treating OA and RA patients (Table 1). Each study employed a balanced randomization process whereby treatment groups were planned to be approximately the same size. Valdecoxib arms included one or more doses of 5, 10, 20 or 40 mg qd, 20 or 40 mg b.d. There was at least one non-selective NSAID comparator in each study (naproxen 500 mg b.d., ibuprofen 800 mg tid or diclofenac 75 mg b.d.). Five of the studies included both a placebo and non-selective NSAID comparator (Table 1). Study duration ranged from 12 to 26 weeks.

In all RCTs, the diagnosis of OA or RA was made using American College of Rheumatology criteria and/or was confirmed by the investigator. Patients with significant co-morbidities or other factors that might interfere with their ability to complete the trials were excluded from study participation. Risk factors for UGI ulcer complications were tabulated based on the patients’ medical histories obtained from each enrolled subject at the baseline visit. In all trials, patients taking low-dose aspirin (≤325 mg daily) for presumed cardiovascular (CV) prophylaxis for at least 30 days before the first dose of study drug were permitted to continue the same dose regimen during the study. RA patients were allowed to take oral glucocorticoids (prednisone or equivalent, up to a maximum daily dose of 10 mg daily) as long as they did not begin taking them or change the dosing regimen within 4 weeks before receiving the first dose of study medication.

Long-term, open-label trials

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Studies 9–11 were designed to assess the long-term safety of valdecoxib 10–80 mg daily in OA or RA patients (Table 1). Patients enrolled in study 10 had previously participated in either study 4 or study 5. Study 11 was a continuation of study 6. In study 9, the diagnosis of OA was confirmed by the investigator, and patients with significant co-morbidities or other factors that might interfere with their ability to complete the trial were excluded from study participation. Patients could increase the dosage of valdecoxib from 10 to 20 mg qd as needed in study 9, and from 10 to 40 mg qd as needed in study 10. In study 11, patients were treated with valdecoxib 40 mg b.d.

Monitoring for UGI ulcer complications

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Investigators at the local sites of all trials were instructed to monitor subjects for signs or symptoms that might indicate the presence of a UGI ulcer or ulcer complication (broadly defined as a perforation, bleeding or gastric outlet obstruction). Investigators were instructed to evaluate patients with potential events based on their clinical judgement using usual standard of care. There was no predefined protocol for local principal investigators with respect to the work up or management of subjects with potential ulcer complications. All potential events were reported by investigators and all reported events were reviewed according to a priori definitions by at least one member of an independent Gastrointestinal Events Committee (GEC), who were blinded to treatment and study. In the situation where the report was inconsistent with a UGI ulcer complication, the case was classified as a negative event. If any of the data and/or the results were suggestive of a UGI ulcer complication, all clinical data were reviewed by the full GEC, which consisted of four expert consultant gastroenterologists who are co-authors on this paper (Jay L. Goldstein; Glenn M. Eisen; Naurang Agrawal; William F. Stenson) working under a prospectively signed charter.

Upper gastrointestinal bleeding.  The UGI bleeding was categorized as one of the following seven clinical presentations: (i) haematemesis with a lesion (ulcer or large erosion) at endoscopy or X-ray; (ii) lesion (ulcer or large erosion) at endoscopy with evidence of active bleeding or stigmata of a recent haemorrhage (visible vessel or clot attached to the base of an ulcer); (iii) melena with a lesion (ulcer or large erosion) at endoscopy or X-ray; (iv) haemoccult-positive stools with a lesion (ulcer or large erosion) at endoscopy or X-ray with evidence of serious bleeding, which included at least one of the following: (a) reduction in haematocrit ≥ 5% or haemoglobin ≥ 1.5 g/dL from baseline; (b) signs of postural vital sign changes (increase in pulse rate of ≥ 20 beats/min and/or decreases in systolic blood pressure of ≥ 20 mmHg and/or decrease in diastolic blood pressure of ≥ 10 mmHg; (c) transfusion of ≥ 2 units of blood; (d) blood in the stomach.

Perforation.  This was defined as an opening in the wall of the stomach or duodenum that required surgery. It could involve a laparoscopic repair, but only if evidence of the perforation was unequivocal, such as free air in the abdomen visible by X-ray or peritoneal signs upon physical examination.

Gastric outlet obstruction.  Diagnosis of an obstruction was required to be made by the investigator, and the diagnosis was required to be supported by endoscopy (e.g. ulcer with a tight oedematous pyloric channel) or X-ray results (e.g. a dilated stomach or delayed barium emptying with clinical evidence).

Statistical methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

Although the process of identifying and adjudicating UGI ulcer complications in these analyses was predetermined, the individual studies included were not primarily designed to assess rates of these events per se. The sample size of each study was calculated based on its primary objectives. A power calculation (for statistical evaluation, not for sample size determination) estimated that the approximate number of patient-years of experience was 200 for placebo-, 500 for non-selective NSAID-, and 2000 for valdecoxib-treated patients. Assuming an annualized UGI ulcer complication rate of 1.6% with non-selective NSAIDs and 0.2% with valdecoxib, with an approximate 20% reduction in patient-year exposure because of early withdrawal, there would be at least 80% power to detect a treatment difference using Chi-square test (alpha = 0.05, two-sided). The treatment groups were compared using the Chi-square test for categorical variables and two-way analysis of variance with treatment and study effects.

All analyses were based on the intent-to-treat cohort, which comprised all subjects who were randomized and received at least one dose of the study drug. Kaplan–Meier rates of UGI ulcer complications by time interval were calculated and comparisons of annualized incidence (events per total patient-years of exposure) were made between treatment groups by log-rank test. In addition to crude incidence rates, analyses of event rates by length of exposure were calculated based on patient-years of exposure. Analyses were performed on all patients and also on those patients not using low-dose aspirin. All statistical tests were performed at a 0.05 alpha level (two-sided test).

Patient data.  A total of 7434 patients with OA or RA were enrolled in studies 1–8 and received at least one dose of placebo (n = 973; 146 patient-years of exposure), valdecoxib 5–80 mg daily (n = 4362; 1183 patient-years), or a non-selective NSAID (n = 2099; 563 patient-years). Patient numbers for each non-selective NSAID treatment group and the valdecoxib treatment group by dose are listed in Table 1. The number of patient-years of exposure in each group is listed in Table 2. The majority of valdecoxib-treated patients received approved doses of 10–20 mg qd (58.9%), while 33.6% were treated with above recommended doses (Table 2).

Table 2.  Demographic and disease characteristics of patients in the randomized, controlled trials
 PlaceboValdecoxibNon-selective NSAIDs
  1. * Cardiovascular (CV) disease included history of angina, atherosclerotic CV disease, congestive heart failure, hypertension, and myocardial infarction.

  2. NSAIDs, non-steroidal anti-inflammatory drugs; OA, osteoarthritis; RA, rheumatoid arthritis; UGI, upper gastrointestinal.

Number of patients randomized
 Total, n97543652105
 By treatment group, n (%) 5 mg qd, 321 (7.4) 10 mg qd, 1214 (27.8) 20 mg qd, 1358 (31.1) 20 mg b.d., 399 (9.1) 40 mg qd, 667 (15.3) 40 mg b.d., 403 (9.2)Naproxen 500 mg b.d., 1181 (56.3) Diclofenac 75 mg b.d., 711 (33.9) Ibuprofen 800 mg tid, 207 (9.9)
Exposure, patient-years
 Total1461183563
 By treatment group 5 mg qd, 61 10 mg qd, 289 20 mg qd, 380 20 mg b.d., 140 40 mg qd, 175 40 mg b.d., 142Naproxen 500 mg b.d., 275 Diclofenac 75 mg b.d., 248 Ibuprofen 800 mg tid, 40
Age, years
 Mean (range)58.8 (19–88)57.8 (18–90)58.5 (18–90)
 n (%)
  <25 years4 (0.4)20 (0.5)   9 (0.4)
  25–34 years20 (2.1)146 (3.3)  52 (2.5)
  35–44 years90 (9.2)450 (10.3) 210 (9.9)
  45–54 years239 (24.5)1073 (24.6) 505 (24.0)
  55–64 years280 (28.7)1296 (29.7) 615 (29.2)
  65–74 years257 (26.4)1045 (24.0) 533 (25.3)
  ≥ 75 years85 (8.7)335 (7.7) 181 (8.6)
Sex, n (%)
 Women687 (70.6)3129 (71.7)1513 (72.1)
 Men286 (29.4)1233 (28.3) 586 (27.9)
Race, n (%)
 Caucasian784 (80.6)3552 (81.4)1736 (82.7)
 Black81 (8.3)337 (7.7) 158 (7.5)
 Hispanic99 (10.2)319 (7.3) 126 (6.0)
 Asian3 (0.3)52 (1.2)  32 (1.5)
 Other6 (0.6)102 (2.3)  47 (2.2)
Primary disease, n (%)
 OA531 (54.6)2182 (50.0)1205 (57.4)
 RA442 (45.4)2180 (50.0) 894 (42.6)
Disease history, n (%)
 UGI bleeding13 (1.3)53 (1.2)  35 (1.7)
 Gastroduodenal ulcer100 (10.3)417 (9.6) 224 (10.7)
 CV disease*457 (47.0)2028 (46.5)1031 (49.1)
Concurrent medications, n (%)
 Aspirin (≤325 mg qd)126 (12.9)584 (13.4) 294 (14.0)
 Corticosteroids190 (19.5)1015 (23.3) 442 (21.1)

There were no clinically significant differences in baseline demographic and disease characteristics between treatment groups (Table 2). Most patients (78.3%) were aged between 45 and 74 years. Notably, the prevalence of the most well-characterized risk factors for NSAID-related ulcer complications were not different across the three arms of the analysis. Table 2 also shows the percentage of patients who took low-dose aspirin or glucocorticoids during study participation.

UGI ulcer complications.  Among the 7434 patients treated in the RCTs, 955 potential UGI ulcer complications were identified by the primary investigators, representing 908 patients. A total of 657 cases were determined to be either an isolated UGI symptom or anaemia without further evidence of a potential ulcer complication. The remaining 298 cases were forwarded for adjudication by the full GEC resulting in a final decision of 19 confirmed UGI ulcer complications, 88 uncomplicated, symptomatic gastroduodenal ulcers, and 191 cases judged to represent neither an ulcer complication nor a symptomatic ulcer. Full diagnoses of all cases adjudicated by the full GEC are summarized in Table 3. All confirmed UGI ulcer complications occurred within the bleeding end-point category; there were no perforations or gastric outlet obstructions.

Table 3.  Classification of adjudicated events in the intent-to-treat populations
CategoryEight randomized, controlled trialsThree long-term, open-label trials
Placebo (n = 973) (%)Valdecoxib 5–80 mg/day (n = 4362) (%)Non-selective NSAIDs (n = 2099) (%)Valdecoxib 5–80 mg/day (n = 2871) (%)
  1. NSAIDs, non-steroidal anti-inflammatory drugs; UGI, upper gastrointestinal.

Total cases of adjudicated events2113514295
Cases not meeting the definition of a clinically significant UGI event2112713188
 Symptomatic ulcers2285810
 Non-UGI events, non-ulcers19997378
  Oesophageal disease418169
  Gastroduodenitis317273
  Colonic or small bowel disease25512
  Non-ulcer bleeding218617
  Non-specific GI symptoms4331322
  Anaemia1428
  Miscellaneous3447
Cases meeting the definition of a clinically significant UGI event08117
 UGI bleeding08117
  <25 years0/4 (0)0/20 (0)0/9 (0)0/7 (0)
  25–34 years0/20 (0)0/146 (0)0/52 (0)0/55 (0)
  35–44 years0/90 (0)1/450 (0.22)1/210 (0.48)0/240 (0)
  45–54 years0/239 (0)1/1073 (0.09)4/505 (0.79)0/646 (0)
  55–64 years0/280 (0)1/1296 (0.08)1/615 (0.16)3/867 (0.35)
  65–74 years0/257 (0)3/1045 (0.29)3/533 (0.56)3/752 (0.40)
  ≥ 75 years0/85 (0)2/335 (0.60)2/181 (1.10)1/308 (0.32)
 Perforation0000
 Gastric outlet obstruction0000

Descriptions of the 19 confirmed UGI ulcer complications are described in Table 4. They occurred in 11 patients taking non-selective NSAIDs (naproxen 500 mg b.d., n = 7; diclofenac 75 mg b.d., n = 4) and eight patients taking valdecoxib (10 mg daily, n =3; 40 mg daily, n = 3; 80 mg daily, n = 2). None of the patients receiving placebo or ibuprofen 800 mg tid experienced a confirmed UGI ulcer complication. A dose–response for UGI ulcer complications was not observed within the valdecoxib treatment group. Of the 19 patients experiencing a confirmed ulcer complication, three were found to have haematemesis with a lesion, five had a lesion with evidence of active bleeding, seven had melena with a lesion, and four had haemoccult-positive stools with a lesion and reduction from baseline in haematocrit or haemoglobin.

Table 4.  Definite upper gastrointestinal (UGI) ulcer complications in the intent-to-treat population of the randomized controlled trials*
 Valdecoxib daily dosageNon-selective NSAID
5 mg (n = 321) 10 mg (n = 1214) 20 mg (n = 1358) 40 mg (n = 1066) 80 mg (n = 403)Naproxen 500 mg b.d. (n = 1181)Diclofenac 75 mg b.d. (n = 711)Ibuprofen 800 mg tid (n = 207)
  1. * All confirmed UGI ulcer complications occurred within the bleeding end-point category; there were no UGI perforations or gastric outlet obstructions.

  2. NSAIDs, non-steroidal anti-inflammatory drugs.

Haematemesis with lesion21
Lesion with bleeding14
Melena with lesion2212
Haemoccult-positive stools with lesion and haematocrit or haemoglobin reduction1111
Total03032740

Figure 1 shows time-to-event analyses in the eight RCTs. Rates of ulcer complications, expressed as annualized incidences, were significantly lower (approximately threefold) in the valdecoxib group (0.68%) than with non-selective NSAID treatment (1.96%). This difference was statistically significant by log-rank test (P = 0.018). No event occurred in the placebo group, and there was no significant difference in ulcer complication rates between valdecoxib- and placebo-treated patients (P = 0.271).

image

Figure 1. Time to development of confirmed upper gastrointestinal ulcer complications in the eight randomized, controlled trials. Placebo is not shown because no event occurred in this group. NSAIDs, non-steroidal anti-inflammatory drugs.

Download figure to PowerPoint

The use of low-dose aspirin, and patients’ history of UGI bleeding, gastroduodenal ulcers, or UGI-related NSAID intolerance had a significant effect on the development of UGI ulcer complications (Table 5). Within the valdecoxib treatment group, a UGI ulcer complication was ninefold more common in aspirin users (P < 0.001). Interestingly, low-dose aspirin use was not a risk factor for developing a UGI ulcer complication in patients receiving non-selective NSAIDs. The annualized incidence of UGI ulcer complications in valdecoxib-treated patients who were non-aspirin users was 0.29% (three events/1029 patient-years of exposure), and significantly lower than that seen in patients receiving a non-selective NSAID (2.08%; 10 events/481 patient-years of exposure; P < 0.001).

Table 5.  Risk factor analysis of clinically significant upper gastrointestinal (UGI) events in the intent-to-treat populations of the randomized, controlled trials, n (%)
  Placebo (n = 973) (%) Valdecoxib (n = 4362) (%)Non-selective NSAIDs (n = 2099) (%)
  1. Significant vs. no aspirin use or disease history: *P < 0.05, **P < 0.01, ***P < 0.001.

  2. NSAIDs, non-steroidal anti-inflammatory drugs.

Aspirin use
 Yes05/584 (0.9)***1/294 (0.3)
 No03/3778 (<0.1)10/1805 (0.6)
History of UGI bleeding
 Yes00/53 (0)2/35 (5.7)***
 No08/4309 (0.2)9/2064 (0.4)
History of gastroduodenal ulcer
 Yes03/417 (0.7)*4/224 (1.8)**
 No05/3945 (0.1)7/1875 (0.4)
History of UGI-related NSAID intolerance
 Yes05/372 (1.3)***3/159 (1.9)*
 No03/3990 (<0.1)8/1940 (0.4)

A history of UGI bleeding significantly increased the risk of UGI ulcer complications in patients receiving non-selective NSAIDs (P < 0.001), but not in patients taking valdecoxib (P > 0.05). Both non-selective NSAID- and valdecoxib-treated patients had an increased risk of UGI ulcer complications if they had a history of gastroduodenal ulcers or GI-related NSAID intolerance (P < 0.05).

The development of UGI ulcer complications in each treatment group was not significantly affected by age, gender, disease type (OA or RA), history of CV disease, or Helicobacter pylori status, which was based on serological testing by the CLOtest.

Patient data.  In total, 2871 patients with OA or RA were enrolled in studies 9–11, and received at least one dose of valdecoxib, corresponding to 1791 patient-years of exposure. Of the total number of patients, 2361 participated for ≥ 12 weeks, 2105 for ≥ 24 weeks, 1168 for ≥ 46 weeks, and 85 for ≥ 52 weeks. Daily valdecoxib dosages ranged from 10 to 20 mg (n = 1167; 810 patient-years exposure), and 10–40 mg (n = 1306; 813 patient-years) to 80 mg (n = 398; 167 patient-years).

The demographics and baseline characteristics of patients recruited into the LTOLs are shown in Table 6. The data describing the patients who elected to enter this trial were similar to the population studied in the RCTs.

Table 6.  Demographic and disease characteristics of patients in the long-term, open-label trials
 Valdecoxib (n = 2871)
  1. OA, osteoarthritis; RA, rheumatoid arthritis; UGI, upper gastrointestinal; CV, cardiovascular.

Age, years
 Mean (range)59.6 (19–92)
 n (%)
  <25 years7 (0.2)
  25–34 years55 (1.9)
  35–44 years240 (8.3)
  45–54 years646 (22.5)
  55–64 years867 (30.2)
  65–74 years752 (26.2)
  ≥ 75 years308 (10.7)
Sex, n (%)
 Women2079 (72.4)
 Men792 (27.6)
Race, n (%)
 Caucasian2534 (88.3)
 Black186 (6.5)
 Hispanic112 (3.9)
 Asian16 (0.6)
 Other23 (0.8)
Primary disease, n (%)
 OA1378 (48.0)
 RA1493 (52.0)
Disease history, n (%)
 UGI bleeding52 (1.8)
 Gastroduodenal ulcer295 (10.3)
 CV disease1333 (46.4)
Concurrent medications, n (%)
 Aspirin (≤325 mg qd)488 (17.0)
 Corticosteroids978 (34.1)

UGI ulcer complications.  Among the 2871 patients enrolled in the long-term, open-label studies, 310 potential UGI ulcer complications were identified, representing 299 patients. A total of 215 cases were either isolated UGI symptoms or anaemia without further evidence of a potential event. Ninety-five cases were adjudicated by the full GEC, and comprised seven confirmed UGI ulcer complications, 10 uncomplicated, symptomatic gastroduodenal ulcers, and 78 cases judged to represent neither a UGI ulcer complication nor a symptomatic ulcer. Full diagnoses of all cases adjudicated by the full GEC are summarized in Table 3. All seven confirmed UGI ulcer complications occurred within the bleeding end-point category; there were no perforations or gastric outlet obstructions.

Of the seven confirmed UGI ulcer complications, four were categorized as lesions with evidence of active bleeding, and three were classified as melena with a lesion. Two of the events occurred in a patient receiving valdecoxib 80 mg daily, four events in patients taking valdecoxib 10–20 mg daily, and one event occurred in a patient taking valdecoxib 10–40 mg daily. A dose–response for UGI ulcer complications was not observed with long-term valdecoxib treatment.

The overall annualized incidence of UGI ulcer complications following treatment with valdecoxib 10–80 mg daily in LTOLs was 0.39% (seven events/1791 patient-years). These data include the 17% (n = 488) of patients receiving low-dose aspirin (≤325 mg daily). Among non-aspirin users, the annualized incidence of UGI ulcer complications following long-term treatment with valdecoxib 10–80 mg daily was 0.2% or three events/1472 patient-years. Among the subpopulation of low-dose aspirin users, the annualized incidence of UGI ulcer complications following long-term valdecoxib treatment was 1.25% or four events/319 patient-years.

The development of UGI ulcer complications in each treatment group was not significantly affected by age, gender, disease type (OA/RA) or history of CV disease, UGI bleeding, gastroduodenal ulcers or GI-related NSAID intolerance.

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

These data confirm and extend the UGI safety profile of COX-2 selective inhibitors, and specifically of valdecoxib, compared with non-selective NSAIDs. In eight RCTs of 12–26 weeks’ duration, the annualized incidence of UGI ulcer complications (perforation, bleeding and gastric outlet obstruction) was threefold lower with valdecoxib 5–80 mg daily (0.68%) compared with therapeutic doses of non-selective NSAIDs (1.96%; P = 0.018). Furthermore, the rate was sevenfold lower in non-aspirin users taking valdecoxib vs. that seen with non-selective NSAIDs (0.29% vs. 2.08%; P < 0.001). In parallel, the data from three LTOLs showed that the low annualized incidence rate of UGI ulcer complications with valdecoxib 10–80 mg daily was maintained for up to 1 year (0.39%). These data are qualitatively consistent with previous studies of similar pooled analyses of prospective, randomized trials.30, 32, 33, 55

It is well established that non-selective NSAID use is associated with significant UGI injury, which is thought to be the result of COX-1 inhibition in the gastroduodenal mucosa. The annualized incidence of NSAID-related UGI ulcer complications in the present pooled analysis (1.96%) is consistent with the rate found with NSAIDs in the previous prospective Misoprostol Ulcer Complications Outcomes Safety Assessment (MUCOSA) trial (1.9%),30 observational Arthritis Rheumatism and Aging Medical Information System (ARAMIS) data (1.3%),2 pooled data from 14 RCTs with celecoxib (1.68%),34 the Celecoxib Long-term Arthritis Safety Study (CLASS; 1.45%),32 and the Vioxx Gastrointestinal Outcomes Research (VIGOR) study (1.4%).33

In the present analysis, the COX-2 selective inhibitor, valdecoxib, was associated with a low rate of UGI ulcer complications: 0.68% in RCTs and 0.39% in LTOLs. The incidence rates with valdecoxib agree with those found with celecoxib in the CLASS trial (0.76%), a pooled analysis of 14 RCTs (0.2%), and a LTOL (0.18%).32, 34 In the VIGOR study, the rate of complicated, confirmed UGI ulcer complications (perforation, obstruction and severe UGI bleeding) after 9 months in approximately 8000 RA patients not taking aspirin was significantly lower in rofecoxib vs. non-selective NSAID-treated patients (0.6% vs. 1.4% per 100 patient-years; P = 0.005).33 Similarly, in a pooled analysis of 5435 OA patients in eight RCTs, rofecoxib 12.5–50 mg daily was associated with a significantly lower incidence of confirmed UGI ulcer complications (perforation, obstruction and severe UGI bleeding) compared with non-selective NSAIDs (1.33% vs. 2.60% per 100 patient-years; P = 0.046).35 In addition, etoricoxib 60–120 mg daily was associated with a reduced incidence of UGI ulcer complications compared with non-selective NSAIDs in 4922 patients with OA, RA or low back pain (1.20% vs. 2.85% per 100 patient-years; P = 0.03).36

Patients taking placebo in the present pooled RCTs did not develop UGI ulcer complications. However, as all of the studies were conducted in patients with symptomatic arthritis, there was a significantly higher rate of early drop out because of lack of efficacy and, as such, this group had a relatively low exposure to treatment. Nevertheless, the rate of UGI ulcer complications associated with valdecoxib treatment was not significantly different vs. placebo treatment in this analysis. It is interesting to note that the rates of UGI ulcer complications associated with valdecoxib in this analysis and COX-2 selective inhibitors in other similar analyses are consistent with the rates found in patients not receiving NSAIDs: 0.27% in one study,2 and 0.1–0.2% in another.56

The present analysis has some potential limitations. The pooled trials had different designs with varying end-points and durations of treatment, and were not specifically designed to compare UGI ulcer complications. Moreover, data for various different non-selective NSAIDs were pooled. It is recognized that the various non-selective NSAID comparators have their own specific pharmacokinetic and pharmacodynamic properties, and that the use of these agents may be associated with different rates of ulcer complications. However, we feel justified in pooling the three non-selective NSAIDs (naproxen, ibuprofen, and diclofenac) for the purposes of this analysis because they are all associated with increased rates of ulcer complications compared with non-NSAID users.8, 31, 35, 57 In addition, some patients (<5%) in the RCTs received a sub-therapeutic dose of valdecoxib. However, we believe that the present analysis and conclusions are valid for several reasons. A large number of patients were involved, representing a high number of total patient-years of valdecoxib exposure. Furthermore, a retrospective power calculation showed that the analysis had sufficient power to detect differences in incidence rates. Regarding the different durations of the trials, all of them involved treatment for at least 12 weeks, and a patient-years analysis was included to allow for the differences. Importantly, the definitions of UGI ulcer complications and the monitoring system were prospectively defined and remained constant throughout the studies. In addition, there is consistency in the adjudication process as all potential UGI ulcer complications were adjudicated by the same committee members. Four of the RCTs (studies 1, 3, 6 and 7) and one of the LTOLs (study 11, which was a continuation of study 6) incorporated scheduled endoscopies. However, in the event of an incidental finding of an asymptomatic ulcer at a scheduled endoscopy, investigators were instructed not to submit the case as a potentially significant event for the purposes of this analysis.

There are several risk factors that predispose patients to UGI ulcer complications. These include age, history of ulcers or GI bleeding and previous CV disease.30 In the present analysis, the use of low-dose aspirin, and a history of UGI bleeding, gastroduodenal ulcers or UGI-related NSAID intolerance increased the risk of ulcer complications in the pooled RCTs. However, age and history of CV disease were not found to be risk factors.

In conclusion, the COX-2 selective inhibitor, valdecoxib, is associated with a significantly lower annualized incidence of UGI ulcer complications vs. non-selective NSAIDs in eight RCTs of 12–26 weeks’ duration. Pooled data from three LTOLs show that the low rate of ulcer complications with valdecoxib is maintained for up to 1 year. Therefore, valdecoxib may represent an improved therapeutic option compared with non-selective NSAIDs for the long-term management of arthritis pain.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References

This study was sponsored by Pfizer Global Pharmaceuticals and Pharmacia Corporation. Dr Jay L. Goldstein has served as a consultant to Pfizer, has received travel expenses and speaker honoraria, and has been the recipient of research grants. Dr Glenn M Eisen has served as a consultant to, and has received research support from, Pfizer. Dr Naurang Agrawal has served on a Pfizer advisory board, and has received research support and honoraria from Pfizer. Dr William F. Stenson has served as a consultant to Pfizer. Dr Jeffrey D. Kent was an employee of Pfizer Global Research and Development during the time that the study was conducted, and is currently an employee of Abbott Immunology (Abbott Park, IL, USA). Dr Kenneth M. Verburg is an employee of Pfizer Global Research and Development.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Methods
  5. Randomized, controlled trials
  6. Long-term, open-label trials
  7. Monitoring for UGI ulcer complications
  8. Definitions of UGI ulcer complications
  9. Statistical methods
  10. Results
  11. Randomized, controlled trials
  12. Long-term, open-label trial
  13. Discussion
  14. Acknowledgements
  15. References
  • 1
    Blackburn WD. Management of osteoarthritis and rheumatoid arthritis: prospects and possibilities. Am J Med 1996; 100: 24S30S.
  • 2
    Singh G, Rosen Ramey D. NSAID induced gastrointestinal complications: the ARAMIS perspective 1997. Arthritis, Rheumatism, and Aging Medical Information System. J Rheumatol 1998; 51(Suppl.): 816.
  • 3
    Thomas J, Straus WL, Bloom BS. Over-the-counter nonsteroidal anti-inflammatory drugs and risk of gastrointestinal symptoms. Am J Gastroenterol 2002; 97: 22159.
    Direct Link:
  • 4
    Gabriel SE, Jaakkimainen L, Bombardier C. Risk for serious gastrointestinal complications related to use of nonsteroidal anti-inflammatory drugs. A meta-analysis. Ann Intern Med 1991; 115: 78796.
  • 5
    Lichtenstein DR, Syngal S, Wolfe MM. Nonsteroidal antiinflammatory drugs and the gastrointestinal tract. The double-edged sword. Arthritis Rheum 1995; 38: 518.
  • 6
    Borda IT, Koff R. NSAIDs: A Profile of Adverse Effects. Philadelphia: Hanley and Belfus, Inc., 1995.
  • 7
    Griffin MR. Epidemiology of nonsteroidal anti-inflammatory drug-associated gastrointestinal injury. Am J Med 1998; 104: 23S29S.
  • 8
    Garcia Rodriguez LA, Jick H. Risk of upper gastrointestinal bleeding and perforation associated with individual nonsteroidal anti-inflammatory drugs. Lancet 1994; 343: 76972.
  • 9
    Goldstein JL, Correa P, Zhao WW, et al. Reduced incidence of gastroduodenal ulcers with celecoxib, a novel cyclooxygenase-2 inhibitor, compared to naproxen in patients with arthritis. Am J Gastroenterol 2001; 96: 101927.
    Direct Link:
  • 10
    Simon LS, Lanza FL, Lipsky PE, et al. Preliminary study of the safety and efficacy of SC-58635, a novel cyclooxygenase 2 inhibitor: efficacy and safety in two placebo-controlled trials in osteoarthritis and rheumatoid arthritis, and studies of gastrointestinal and platelet effects. Arthritis Rheum 1998; 41: 1591602.
  • 11
    Kivitz A, Eisen G, Zhao WW, et al. Randomized placebo-controlled trial comparing efficacy and safety of valdecoxib with naproxen in patients with osteoarthritis. J Fam Pract 2002; 51: 5307.
  • 12
    Goldstein JL, Kent JD, Zhao W, Recker D, Verburg KM. Reduced incidence of gastroduodenal ulcers with valdecoxib compared to ibuprofen and diclofenac in patients with rheumtoid arthrtits: a multicenter trial [Abstract]. Am J Gastroenterol 2001; 96(Suppl.): 9.
  • 13
    Goldstein JL, Talwalker S, Verburg KM, Recker D, Hubbard RC. Comparative gastroduodenal mucosal effects of valdecoxib, a potent COX-2 specific inhibitor, compared with naproxen and placebo [Abstract]. Gut 2001; 49(suppl. III): Abstract 2116.
  • 14
    Rordorf C, Kellett N, Mair S, et al. Gastroduodenal tolerability of lumiracoxib vs placebo and naproxen: a pilot endoscopic study in healthy male subjects. Aliment Pharmacol Ther 2003; 18: 53341.
  • 15
    Hunt RH, Harper S, Callegari P, et al. Complementary studies of the gastrointestinal safety of the cyclo-oxygenase-2-selective inhibitor etoricoxib. Aliment Pharmacol Ther 2003; 17: 20110.
  • 16
    Hawkey CJ, Laine L, Harper SE, et al. Influence of risk factors on endoscopic and clinical ulcers in patients taking rofecoxib or ibuprofen in two randomized controlled trials. Aliment Pharmacol Ther 2001; 15: 1593601.
  • 17
    Atherton C, Jones J, McKaig B, et al. Pharmacology and gastrointestinal safety of lumiracoxib, a novel cyclooxygenase-2 selective inhibitor: an integrated study. Clin Gastroenterol Hepatol 2004; 2: 11320.
  • 18
    Lanza FL, Rack MF, Simon TJ, et al. Specific inhibition of cyclooxygenase-2 with MK-0966 is associated with less gastroduodenal damage than either aspirin or ibuprofen. Aliment Pharmacol Ther 1999; 13: 7617.
  • 19
    Baek SJ, Wilson LC, Lee CH, Eling TE. Dual function of nonsteroidal anti-inflammatory drugs (NSAIDs): inhibition of cyclooxygenase and induction of NSAID-activated gene. J Pharmacol Exp Ther 2002; 301: 112631.
  • 20
    Turini ME, DuBois RN. Cyclooxygenase-2: a therapeutic target. Annu Rev Med 2002; 53: 3557.
  • 21
    Lipsky PE, Brooks P, Crofford LJ, et al. Unresolved issues in the role of cyclooxygenase-2 in normal physiologic processes and disease. Arch Intern Med 2000; 160: 91320.
  • 22
    Seibert K, Zhang Y, Leahy K, et al. Pharmacological and biochemical demonstration of the role of cyclooxygenase 2 in inflammation and pain. Proc Natl Acad Sci USA 1994; 91: 1201317.
  • 23
    Smith WL, Garavito RM, DeWitt DL. Prostaglandin endoperoxide H synthases (cyclooxygenases)-1 and -2. J Biol Chem 1996; 271: 3315760.
  • 24
    Kargman S, Charleson S, Cartwright M, et al. Characterization of Prostaglandin G/H Synthase 1 and 2 in rat, dog, monkey, and human gastrointestinal tracts. Gastroenterology 1996; 111: 44554.
  • 25
    Kuehl FA, Egan RW. Prostaglandins, arachidonic acid, and inflammation. Science 1980; 210: 97884.
  • 26
    Crofford LJ, Lipsky PE, Brooks P, et al. Basic biology and clinical application of specific cyclooxygenase-2 inhibitors. Arthritis Rheum 2000; 43: 413.
  • 27
    Born GVR, Cross MJ. The aggregation of blood platelets. J Physiol (Lond) 1963; 168: 17895.
  • 28
    Leese PT, Talwalker S, Kent JD, Recker DP. Valdecoxib does not impair platelet function. Am J Emerg Med 2002; 20: 27581.
  • 29
    Wolfe MM, Lichtenstein DR, Singh G. Gastrointestinal toxicity of nonsteroidal antiinflammatory drugs. N Engl J Med 1999; 340: 188899.
  • 30
    Silverstein FE, Graham DY, Senior JR, et al. Misoprostol reduces serious gastrointestinal complications in patients with rheumatoid arthritis receiving nonsteroidal anti-inflammatory drugs. A randomized, double-blind, placebo-controlled trial. Ann Intern Med 1995; 123: 2419.
  • 31
    MacDonald TM, Morant SV, Robinson GC, et al. Association of upper gastrointestinal toxicity of non-steroidal anti-inflammatory drugs with continued exposure: cohort study. BMJ 1997; 315: 13337.
  • 32
    Silverstein FE, Faich G, Goldstein JL, et al. Gastrointestinal toxicity with celecoxib vs. nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. Celecoxib Long-term Arthritis Safety Study. JAMA 2000; 284: 124755.
  • 33
    Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. VIGOR Study Group. N Engl J Med 2000; 343: 15208.
  • 34
    Goldstein JL, Silverstein FE, Agrawal NM, et al. Reduced risk of upper gastrointestinal ulcer complications with celecoxib, a novel COX-2 inhibitor. Am J Gastroenterol 2000; 95: 168190.
    Direct Link:
  • 35
    Langman MJ, Jensen DM, Watson DJ, et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282: 192933.
  • 36
    Harper S, Lee M, Curtis S, et al. A lower incidence of upper-GI perforations, ulcers and bleeds (PUBs) in patients treated with etoricoxib vs. nonselective cyclooxygenase inhibitors (NSAIDs). Arthritis & Rheumatism 2001; 44: S318. Abstract 1590.
  • 37
    Gierse JK, McDonald JJ, Hauser SD, et al. A single amino acid difference between cyclooxygenase-1 (COX-1) and -2 (COX-2) reverses the selectivity of COX-2 specific inhibitors. J Biol Chem 1996; 271: 158104.
  • 38
    Simon LS, Weaver AL, Graham DY, et al. Anti-inflammatory and upper gastrointestinal effects of celecoxib in rheumatoid arthritis: a randomized controlled trial. JAMA 1999; 282: 19218.
  • 39
    Masferrer JL, Zweifel BS, Manning PT, et al. Selective inhibition of inducible cyclooxygenase 2 in vivo is antiinflammatory and nonulcerogenic. Proc Natl Acad Sci USA 1994; 91: 322832.
  • 40
    Collantes E, Curtis SP, Lee KW, et al. A multinational randomized, controlled, clinical trial of etoricoxib in the treatment of rheumatoid arthritis. BMC Fam Pract 2002; 3: 10.
  • 41
    Leung AT, Malmstrom K, Gallacher AE, et al. Efficacy and tolerability profile of etoricoxib in patients with osteoarthritis: a randomized, double-blind, placebo and active-comparator controlled 12-week efficacy trial. Curr Med Res Opin 2002; 18: 4958.
  • 42
    Matsumoto AK, Melian A, Mandel DR, et al. A randomized, controlled, clinical trial of etoricoxib in the treatment of rheumatoid arthritis. J Rheumatol. 2002; 29: 162330.
  • 43
    Makarowski W, Zhao W, Bevirt T, Recker D. Efficacy and safety of the COX-2 specific inhibitor valdecoxib in the management of osteoarthritis of the hip: a randomized, double-blind, placebo-controlled comparison with naproxen. Osteoarthritis Cartilage. 2002; 10: 2906.
  • 44
    Bensen W, Weaver A, Espinoza L, et al. Efficacy and safety of valdecoxib in treating the signs and symptoms of rheumatoid arthritis: a randomized, controlled comparison with placebo and naproxen. Rheumatology 2002; 41: 100816.
  • 45
    Bensen WG, Fiechtner JJ, McMillen JI, et al. Treatment of osteoarthritis with celecoxib, a cyclooxygenase-2 inhibitor: a randomized controlled trial. Mayo Clin Proc 1999; 74: 1095105.
  • 46
    Leese PT, Recker DP, Kent JD. The COX-2 specific inhibitor valdecoxib does not impair platelet function in the elderly: results of a randomized controlled trial. J Clin Pharmacol 2002; 43: 50413.
  • 47
    Leese PT, Hubbard RC, Karim A, et al. Effects of celecoxib, a novel cyclooxygenase-2 inhibitor, on platelet function in healthy adults: a randomized, controlled trial. J Clin Pharmacol 2000; 40: 12432.
  • 48
    Greenberg HE, Gottesdiener K, Huntington M, et al. A new cyclooxygenase-2 inhibitor, rofecoxib (VIOXX), did not alter the antiplatelet effects of low-dose aspirin in healthy volunteers. J Clin Pharmacol 2000; 40: 150915.
  • 49
    Riendeau D, Percival MD, Brideau C, et al. Etoricoxib (MK-0663): preclinical profile and comparison with other agents that selectively inhibit cyclooxygenase-2. J Pharmacol Exp Ther 2001; 296: 55866.
  • 50
    Bensen WG, Zhao SZ, Burke TA, et al. Upper gastrointestinal tolerability of celecoxib, a COX-2 specific inhibitor, compared to naproxen and placebo. J Rheumatol 2000; 27: 187683.
  • 51
    Weaver J, Bonnel RA, Karwoski CB, Brinker AD, Beitz J. GI events leading to death in association with celecoxib and rofecoxib. Am J Gastroenterol 2001; 96: 344950.
  • 52
    Talley JJ, Brown DL, Carter JS, et al. 4-[5-Methyl-3-phenylisoxazol-4-yl]- benzenesulfonamide, valdecoxib: a potent and selective inhibitor of COX-2. J Med Chem 2000; 43: 7757.
  • 53
    Daniels S, Talwalker S, Torri S, et al. Valdecoxib, a COX-2 specific inhibitor, is effective and well tolerated in treating women with primary dysmenorrhea. Obstet Gynecol 2002; 100: 3508.
  • 54
    Sikes DH, Agrawal NM, Zhao WW, et al. Incidence of gastroduodenal ulcers associated with valdecoxib compared with that of ibuprofen and diclofenac in patients with osteoarthritis. Eur J Gastroenterol Hepatol 2002; 14: 110111.
  • 55
    Goldstein JL, Eisen G, Stenson WF, et al. Significant reduction in serious upper gastrointestinal (UGI) events with celecoxib, a COX-2 specific inhibitor, compared with conventional NSAIDs: The SUCCESS-1 trial. Ann Rheum Dis 2001; 60(Suppl. 1): Abstract THU0182.
  • 56
    Gutthann SP, Garcia Rodriguez LA, Raiford DS. Individual nonsteroidal antiinflammatory drugs and other risk factors for upper gastrointestinal bleeding and perforation. Epidemiology 1997; 8: 1824.
  • 57
    Langman MJ, Weil J, Wainwright P, et al. Risks of bleeding peptic ulcer associated with individual non-steroidal anti-inflammatory drugs. Lancet 1994; 343: 10758.