[ Editorials | Letters | Selected Articles | Signature Series | Patient Handouts | Index ]

2002 SelectedArticles

Effective Use of Nonsteroidal Anti-inflammatory Drugs

Nicole T. Ansani, PharmD; Terence W. Starz, MD

Nonsteroidal anti-inflammatory drugs (NSAIDs) have been used for the treatment of musculoskeletal (M-S) diseases since the time of Hippocrates in the fourth century BCE. Sodium salicylate, derived from willow bark and other plants, was the first NSAID, but its use was limited by gastrointestinal (GI) upset. In 1897, Felix Hoffmann, a chemist for Friedrich Bayer & Co in Germany, synthesized aspirin by adding acetic acid to sodium salicylate, thus initiating the modern era of NSAID use.¹ Currently, there are more than 25 NSAIDs available, and their role in M-S and other therapies has evolved significantly. In 2001, more than 50 million Americans used NSAIDs, accounting for some 70 million prescriptions—or approximately 3% of all prescriptions.^2,3 These drugs are effective for treating many acute and chronic M-S conditions, including osteoarthritis (OA) and rheumatoid arthritis (RA); regional disorders such as tendonitis, bursitis, and acute low back pain (ALBP); and other acute pain states, such as headache, injuries, and dysmenorrhea.⁴ As a class, NSAIDs have similarities in efficacy, safety, and pharmacokinetic characteristics, but there are differences in the adverse drug reaction (ADR) profiles between nonselective NSAIDs and the newer cyclo-oxygenase-2 (COX-2) selective agents, especially with regard to GI side effects. In choosing NSAID therapy, it is important to consider these inherent properties, as well as differences in ADR profiles and cost.

back to top

MECHANISM OF ACTION

All NSAIDs have both analgesic and anti-inflammatory properties. Until the 1970s, their mechanism of action was not well understood.¹ Although NSAIDs have a number of physiologic effects, their principal action is the inhibition of the COX enzyme. This enzyme is responsible for the production of prostaglandins, which are derived from arachidonic acid, a phospholipid present in all human cell membranes. Prostaglandins are important mediators of normal homeostatic functions and of the inflammatory response, primarily via their influence on vascular permeability, immune reactions, and platelet function.^5,6 In 1991, two isoforms of the COX enzyme were identified (COX-1 and COX-2). These enzymes share approximately 60% structural homogeneity and are coded on two different genes.^5,6 While their patterns of tissue expression and regulation are different, their physiologic effects on various body functions are overlapping.^2,5-9

Prostaglandins produced by COX-1 are primarily involved in homeostatic (“housekeeping”) activities such as maintaining the GI mucosal barrier, renal hemodynamics, platelet function, and vascular homeostasis. They may also play some role in inflammation.^5,6,8,9 The COX-2 enzyme is induced by inflammation, resulting in prostaglandin production by fibroblasts, macrophages, endothelial cells, and synoviocytes. In addition, COX-2 may also be expressed normally in small amounts in the kidney, small intestine, ovary, uterus, bone, and brain.^2,5,9

Nonselective NSAIDs (eg, ibuprofen, naproxen) inhibit both COX-1 and COX-2 enzymes, thereby decreasing both inflammatory and homeostatic prostaglandin production. Certain nonselective NSAIDs (eg, etodolac, nabumetone, meloxicam) have been associated with some degree of COX-2 selectivity, albeit much less than the COX-2 selective agents.⁹ The selective COX-2 agents (eg, celecoxib, rofecoxib, valdecoxib) have shown a COX-2 selectivity of more than 50-fold, and exert their actions primarily in inflammatory processes.^8,10

back to top

PRESCRIBING CONSIDERATIONS

Pharmacologic Characteristics

All NSAIDs are weakly acidic drugs, and can be divided into categories based on chemical structure (Table 1). Gastrointestinal absorption of NSAIDs is rapid, but variables such as product formulations (eg, enteric coating, delayed-release preparations) and metabolic characteristics lead to differences in absorption, time-to-peak effect, and drug half-life. The NSAIDs as a class are highly plasma-protein bound, hepatically metabolized primarily through the cytochrome P450 and/or glucuronidase enzymes, and eliminated by renal and biliary excretion.¹¹ These properties influence the potential for drug side effects. For example, impaired renal or hepatic function may prolong drug half-life and extend the NSAID’s effect on the GI tract or kidneys, thereby increasing the risk of toxicity.¹² Furthermore, interactions with other drugs via these pharmacokinetic properties may result in adverse reactions.

 

View this table

Table 1. Characteristics of Commonly Prescribed NSAIDs4,46

Patient Response

Musculoskeletal disorders can be generally classified as mechanical (noninflammatory) problems such as OA, regional disorders, and inflammatory diseases (eg, RA, gout). At times, an inflammatory component may be associated with mechanical M-S problems.

The most common indications for NSAIDs have been rheumatic conditions and acute pain states, including ALBP, headache, and dysmenorrhea. The analgesic effects of NSAIDs are often achieved with doses lower than those required for an anti-inflammatory response. For example, the pain of OA may respond to 6 to 8 aspirin tablets per day, while 10 or more tablets are usually required to influence the inflammatory synovitis of RA. More recently, the use of NSAIDs for postoperative pain has increased in popularity, but their possible impact on healing and the antiplatelet effect of nonselective NSAIDs must be considered. For acute pain relief, the onset of NSAID action is typically within 30 to 60 minutes. However, 2 weeks of therapy may be required to achieve a full anti-inflammatory and analgesic response.¹³

The effectiveness of an NSAID and the need for continued therapy must be assessed periodically. As a rule, the lowest effective dose of the NSAID should be used for the shortest duration required to control the problem. For chronic M-S conditions such as OA and RA, though, long-term NSAID use is often necessary.¹³

In clinical trials, all NSAIDs have demonstrated clinical efficacy in relieving pain and inflammation.^14,15 Selective COX-2 inhibitors have been compared with naproxen, ibuprofen, and diclofenac in patients with OA, RA, and various acute pain states.^14,15 There is interpatient variability in response to any given agent, with approximately 70% to 80% of individuals responding to a particular NSAID. Lack of response to one NSAID does not preclude a response to another.^13,16 It is not clear why there are differences in individual responsiveness. All agents inhibit the COX-enzyme-mediated conversion of arachidonic acid to prostaglandins, albeit to varying degrees. Because of the variations in response to NSAIDs, a 10- to 14-day therapeutic trial is necessary to assess efficacy.^13,16

If the response to an NSAID is inadequate, factors such as patient compliance, dosing strategy, and disease mechanism should be considered. Once- or twice-daily dosing significantly enhances patient compliance. For example, the use of aspirin for arthritis has decreased substantially because of TID or QID dosing and the large number of tablets required to achieve a clinical effect. Prescribing more than one NSAID concurrently is not recommended because of additive side effects with little or no demonstrated additional benefit. Characteristics common to all classes of NSAIDs are summarized in Table 2.

 

View this table

Table 2. Key Considerations with All NSAIDs

back to top

Adverse Drug Reaction Profile

In addition to efficacy, the ADR profiles of NSAIDs are important in considering drug selection. In general, the ADR profiles of all nonselective NSAIDs are similar, with GI problems being the most common. Nonselective NSAIDs can cause GI irritation by direct, local injury to the mucosal surface and disturbance of the mucosal barrier via inhibition of COX-1 “housekeeping” effects. With the introduction of COX-2 agents, there is a decreased risk of GI ADRs due to the COX-1-sparing effect at therapeutic dosing.^2,8

Gastrointestinal Effects.—Between 15% and 30% of patients taking chronic nonselective NSAIDs experience GI symptoms such as nausea, dyspepsia, or abdominal pain.¹⁷ In the majority of these individuals, the mucosal injury is superficial, and is not associated with a serious adverse GI event such as bleeding. Nonselective NSAID therapy for 1 to 3 months or longer may result in upper GI mucosal irritation or ulceration in more than 20% of patients on endoscopy, many of whom are asymptomatic and never have a major complication.² Lastly, not all patients with dyspepsia have abnormalities on endoscopy, and the origin of their symptoms is not clear.

When significant nonselective-NSAID-related upper GI bleeding requiring hospitalization occurs (estimated to be 103,000 cases per year), the mortality is 5% to 10%.¹⁷ Age over 50 years is a continuously increasing variable in the development of GI ADRs with nonselective NSAIDs. A relative risk (RR) analysis for serious nonselective NSAID-associated GI complications indicated an RR of 1.6 at ages 50 to 59 years, 3.1 at ages 60 to 69 years, and 5.6 at ages 70 to 80 years.¹⁸ In comparison, the RR is similar to that associated with corticosteroids (4.4) and high-dose NSAID use (5.8), but less than the RR for anticoagulant use (12.7) or history of GI toxicity (13.5).¹⁸ Tablet formulation and administration of nonselective NSAIDs with food may decrease local GI irritation, but may also reduce drug absorption.

Studies have compared the GI effects of the COX-2 agents celecoxib, rofecoxib, and valdecoxib with those of naproxen and other nonselective NSAIDs.^19-24 With celecoxib, the CLASS and other studies showed an equal or lower rate of GI complications than with nonselective NSAIDs.^19,20 The primary endpoints of the CLASS study¹⁹ (GI bleeding, outlet obstruction, and perforation) revealed a numerical but not statistical difference in serious GI complications between celecoxib and nonselective NSAIDs. However, when symptomatic ulcers were added to the analysis with the primary endpoints, a significant benefit was found with celecoxib. Further analysis of the CLASS data evaluated the incidence of GI complications in patients who were not using concomitant aspirin therapy. Results of this subanalysis of an aspirin-free population revealed a significantly lower rate of GI complications and GI complications plus symptomatic ulcers when celecoxib was compared with diclofenac or ibuprofen. Despite these data, the US Food and Drug Administration (FDA) still requires celecoxib labeling similar to that for nonselective NSAIDs.

GI tolerability of rofecoxib was compared with nonselective NSAIDs in the VIGOR study²² and a subsequent meta-analysis.²³ These studies showed an improved GI complication profile of rofecoxib compared with naproxen, but low-dose aspirin use was not allowed.

Differences in study methodology (duration, subject selection, primary and secondary outcome endpoints), patient populations (OA and RA in CLASS, RA in VIGOR) and concomitant therapy (no aspirin in VIGOR) preclude valid comparisons between the CLASS and VIGOR data (Table 3). With valdecoxib, the most recently introduced COX-2 NSAID, GI endoscopy with therapeutic and supratherapeutic doses demonstrated similar lower rates of gastroduodenal ulcers compared with nonselective NSAIDs.²⁴ Currently, there are no published data evaluating the clinical GI profile of valdecoxib.

 

View this table

Table 3. Data From the CLASS and VIGOR Studies19,22

back to top

Cardiovascular Effects.—The risk of cardiovascular ADRs with COX-2 agents compared with nonselective NSAIDs remains controversial.^25-29 The VIGOR trial²² reported a significant increase in the incidence of myocardial infarction (MI) with rofecoxib compared with naproxen, whereas the CLASS study¹⁹ showed no difference in cardiovascular endpoints when celecoxib was compared with ibuprofen and diclofenac. Again, there were differences in study methodology (no aspirin allowed in VIGOR) and patient populations, making it difficult to compare these data (Table 3). Unlike nonselective NSAIDs, COX-2 agents have no antiplatelet effects. Therefore, the occurrence of more cardiovascular events in the VIGOR trial rofecoxib group may be due to the antiplatelet effect of naproxen plus the ban on aspirin use, and not to an adverse vascular effect of rofecoxib per se.²⁸ In the CLASS study, celecoxib was compared with ibuprofen and diclofenac, which have less antiplatelet activity than naproxen.³⁰ Further investigations are needed to fully assess the cardiovascular effects of COX-2 agents.

Renal Effects.—The estimated incidence of sodium retention and associated edema ranges from 2% to 5% with both nonselective and COX-2 NSAIDs.^24,31-33 These findings appear to be caused primarily by renal inhibition of COX-2.³¹ Rofecoxib has a higher reported incidence of lower-extremity edema, which appears to be dose-related.³³ A study comparing celecoxib, 200 mg/d, with rofecoxib, 25 mg/d, showed a higher incidence of edema (4.9% versus 9.5%) with rofecoxib and an absolute increase in systolic blood pressure of 3.1 mm Hg with rofecoxib compared to celecoxib.³⁴

Differences between the influence of COX-1 and COX-2 agents on hypertension are not clear. Data from COX-2 studies resulted in hypertension labeling similar to that for nonselective NSAIDs, and an additional precaution with higher doses of rofecoxib.^31,33 Patients with cardiovascular disease, congestive heart failure, hypertension, renal or hepatic insufficiency, or advanced age should be monitored for fluid retention and effect on blood pressure when given any NSAID (including COX-2 selective) therapy.³¹ In addition, drug/drug interactions with nonselective NSAIDs or COX-2 agents and antihypertensive drugs (eg, angiotensin-converting enzyme inhibitors, b-blockers, diuretics) may accentuate the inhibition of renal prostaglandin production.¹¹

In addition to fluid retention and hypertension, renal ADRs such as renal insufficiency and (less commonly) electrolyte abnormalities have been associated with NSAID use. Both COX-1 and COX-2 are constitutively expressed in the kidney, predisposing to renal problems with both NSAID classes. The ADRs with nonselective NSAIDs and COX-2 agents regarding renal function appear to be similar.³⁵ These effects on renal homeostasis are most pronounced in patients with renal insufficiency and volume depletion. In addition, patients with concurrent hypertension, congestive heart failure, edema, or diuretic use may be particularly sensitive.^36,37 Thus, similar prescribing precautions should be taken with nonselective NSAIDs and COX-2 agents.^35,37 As with nonselective NSAIDs, the renal ADRs of COX-2 agents tend to occur early in therapy, and are usually reversible on discontinuation of the drug.^35,37

Allergic Reactions.—Hypersensitivity reactions, including bronchospasm, occur rarely with NSAID use, and are more common in individuals with nasal polyps and asthma. Allergic reactions such as bronchoconstriction, nasal polyps, rhinitis, and urticaria have been associated with all nonselective NSAIDs and COX-2 agents. Therefore, use of nonselective NSAIDs and COX-2 agents should be avoided in patients with a history of bronchoconstriction/allergic reactions to other NSAIDs due to the potential for cross-reactivity. Despite reports of cross-reactivity with the NSAID class, choline magnesium trisalicylate has been shown to be well tolerated in aspirin-sensitive asthmatic patients.³⁸ Recent data suggest a role of altered COX-2 regulation associated with the aspirin-intolerant asthma/rhinitis syndrome, but these data require further investigation.³⁹ Because of their chemical structural similarity to sulfonamides, celecoxib and valdecoxib are contraindicated in patients with a history of allergic cutaneous and other reactions to these drugs.^32,40 Overall ADR considerations for nonselective and COX-2 NSAIDs are summarized in Table 4.

 

View this table

Table 4. Key ADR Considerations2,8,25-29,31,32,35-37,40

back to top

Patient Profile/Risk Assessment

Although the risk of GI toxicity increases with nonselective NSAIDs during the first 3 months of use, the incidence of GI toxicity remains constant with prolonged use.¹⁸ Patient-specific risk factors must be evaluated carefully when determining NSAID selection. Risk factors for GI toxicity that have consistently met evidence-based medicine criteria in multiple clinical trials are summarized in Table 5.^18,41-45

 

View this table

Table 5. Risk Factors for NSAID GI Toxicity18,41-45

Data have implicated cigarette smoking, alcohol use, and Helicobacter pylori as possible risk factors for nonselective NSAID GI toxicity.⁴⁰ In addition, low-dose aspirin therapy increases the risk of serious GI complications when given concomitantly with either nonselective or COX-2 NSAIDs.

Individuals with an increased risk of developing nonselective-NSAID-related GI toxicity may benefit from using COX-2 agents that have demonstrated a more favorable GI risk profile. Use of selective COX-2 agents should be considered in patients with these risk factors.^18,41-45 Nonetheless, the preferential use of COX-2 agents over nonselective NSAIDs in patients with a low risk of GI or other toxicities is not warranted. The age threshold of 50 years and above for an increased risk of serious NSAID GI toxicity is based on RR, and does not alone preclude considering a nonselective NSAID for older patients.

Other, less common ADRs associated with NSAID use include hepatic abnormalities (elevated liver function values), headache, confusion (especially in older patients), sleep disturbances, and tinnitus. In rare cases, NSAIDs have been implicated in causing Reye’s syndrome (in children) and aseptic meningitis.^11,13

Cost

After considering efficacy and safety in NSAID selection, cost must be evaluated.⁴⁷ Factors include brand-name versus generic, cost associated with ADRs, the patient’s health insurance formulary, and drug monitoring requirements. The relative risk of GI toxicity has a major role in determining whether a nonselective or selective NSAID is chosen. Use of selective COX-2 inhibitors may result in lower overall costs in particular populations due to the decreased incidence of GI toxicity.^47,48 In patients at average risk of GI toxicity, the costs of COX-2 agents are similar to those of nonselective agents plus a proton-pump inhibitor or misoprostol.^47,48 The use of COX-2 agents in high-risk patients has the potential to improve tolerance and compliance and lower overall costs.⁴⁷ Studies have demonstrated that the savings from decreased GI toxicity of COX-2 agents in high-risk patients can offset the higher acquisition costs of these drugs.⁴⁸

Chronic Use

Finally, NSAIDs are not curative for rheumatic disorders and other indicated chronic conditions, and they are often used in conjunction with nonpharmacologic measures such as physical and occupational therapy, weight control, disease education, and joint protection. After selecting the initial agent, the patient should be counseled about possible adverse effects and the importance of reporting them. Toxicities of NSAIDs have the potential for morbidity and (rarely) mortality. NSAIDs vary in their incidence of adverse effects, but most toxicities are common to all agents. With chronic NSAID use, efficacy and safety profiles should be determined at baseline, and periodically thereafter by evaluating clinical and laboratory parameters. The precise frequency of laboratory testing for NSAID toxicity is not fully defined in package inserts, but a complete blood count with renal and liver function testing should be considered initially, and periodically thereafter in women requiring chronic therapy. The primary considerations for prescribing NSAIDs in terms of efficacy, safety, and cost are summarized in Table 6.

 

View this table

Table 6. Key Efficacy, Safety, and Cost Considerations

back to top

CONCLUSION

NSAIDs are a common class of medications indicated for a variety of rheumatic and other conditions. In general, the lowest effective dose of the NSAID should be used for the shortest possible duration. Before initiating NSAID therapy, the following factors must be considered: diagnosis, patient characteristics, efficacy, side effects, and cost. The patient should be counseled on potential ADRs and the importance of communicating symptoms to the physician. Appropriate clinical and laboratory follow-up is necessary, especially for patients on chronic NSAID therapy. When selecting an NSAID, the efficacy profile should be balanced with the toxicity profile and patient characteristics to provide the most appropriate, safe, and cost-effective therapy. In addition, the dosage range for each NSAID must be individualized at baseline according to patient characteristics and disease mechanism, and adjusted according to the results of follow-up monitoring.

Nicole T. Ansani, PharmD, is an assistant professor and associate director of drug information, Department of Pharmacy and Therapeutics University of Pittsburgh School of Pharmacy, Pa. Terence W. Starz, MD is a clinical professor of medicine, Division of Rheumatology, University of Pittsburgh School of Medicine, Pa.

back to top

REFERENCES

1. Vane JR, Botting RM. Mechanism of action of non-steroidal anti-inflammatory drugs. Am J Med. 1998;104(3A):2S-8S.
2. Verburg KM, Maziasz TJ, Weiner E, et al. COX-2 specific inhibitors: definition of a new therapeutic concept. Am J Ther. 2001;8:49-64.
3. Boyce EG, Takiya L. Nonsteroidal anti-inflammatory drugs: review of factors guiding formulary selection. Formulary. 2000;35:142-168.
4. Murray L, Kelly GL, eds. Physician’s Desk Reference. Montvale, NJ: Medical Economics; 2002.
5. Buttar NS, Wang KK. The “aspirin” of the new millennium: cyclooxygenase-2 inhibitors. Mayo Clin Proc. 2000;75: 1027-1038.
6. FitzGerald GA, Patrono C. Drug therapy: the coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med. 2001; 345:433-442.
7. Penglis PS, James MJ, Cleland LG. Cyclooxygenase inhibitors: any reservations? Intern Med J. 2001;31:37-41.
8. Fiorucci S, Meli R, Bucci M, Cirino G. Dual inhibitors of cyclooxygenase and 5-lipoxygenase. A new avenue in anti-inflammatory therapy. Biochem Pharmacol. 2001;62: 1433-1438.
9. Kaplan-Machlis B, Klostermeyer BS. The cyclooxygenase-2 inhibitors: safety and effectiveness. Ann Pharmacother. 1999; 33:979-988.
10. Vane JR, Botting RM. Anti-inflammatory drugs and their mechanism of action. Inflamm Res. 1998;47:S78-S87.
11. Garnett WR. Clinical implications of drug interactions with coxib. Pharmacotherapy. 2001;21:1223-1232.
12. Brune K, Neubert A. Pharmacokinetic and pharmacodynamic aspects of the ideal COX-2 inhibitor: a pharmacologist’s perspective. Clin Exp Rheumatol. 2001;19:S51-S57.
13. Pollison R. Nonsteroidal anti-inflammatory drugs: practical and theoretical considerations in their selection. Am J Med. 1996;100:2A31S-2A36S.
14. Cannon GW, Breedveld FC. Efficacy of cyclooxygenase-2-specific inhibitors. Am J Med. 2001;110:6S-12S.
15. Emery P. Cyclooxygenase-2: a major therapeutic advance? Am J Med. 2001;110:42S-45S.
16. Hardman JG, Limbird LE, eds. Goodman and Gillman’s The Pharmacological Basis of Therapeutics. New York: McGraw-Hill; 1996;617-642.
17. Wolfe MM, Lichtenstein DV, Singh G. Medical progress: gastrointestinal toxicity of nonsteroidal anti-inflammatory drugs. N Engl J Med. 1999;340:1888-1899.
18. Schoenfeld P. An evidence-based approach to the gastrointestinal safety profile of COX-2-selective anti-inflammatories. Gastroenterol Clin North Am. 2001;30:1027-1045.
19. 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. JAMA. 2000;284:1247-1255.
20. Simon LS, Weaver AE, Graham DY, et al. Anti-inflammatory and upper gastrointestinal effects of celecoxib in rheumatoid arthritis: a randomized controlled trial. JAMA. 1999; 282: 1921-1928.
21. Burke TA, Zabinski RA, Pettitt D, et al. A framework for evaluating the clinical consequences of initial therapy with NSAIDs, NSAIDs plus gastroprotective agents, or celecoxib in the treatment of arthritis. Pharmacoeconomics. 2001;19:33-47.
22. Bombardier C, Laine L, Reicin A, et al. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med. 2000;343:1520-1528.
23. Langman MJ, Jensen DM, Watson DJ, et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA. 1999;282:1929-1933.
24. Valdecoxib (Bextra™) Manufacturer Product Information. Chicago: GD Searle; November 2001.
25. Solomon DH, Glynn RJ, Levin R, Avorn J. Nonsteroidal anti-inflammatory drug use and acute myocardial infarction. Arch Intern Med. 2002;162:1099-1104.
26. Watson DJ, Rhodes T, Cai B, Guess HA. Lower risk of thromboembolic cardiovascular events with naproxen among patients with rheumatoid arthritis. Arch Intern Med. 2001;162:1105-1110.
27. Rahme E, Pilote L, LeLerier J. Association between naproxen use and protection against acute myocardial infarction. Arch Intern Med. 2002;162:1111-1115.
28. Dalen JE. Selective COX-2 inhibitors, NSAIDs, aspirin, and myocardial infarction. Arch Intern Med. 2002; 162:1091-1092.
29. Simon LS, Smolen JS, Abramson SB, et al. Controversies in COX-2 selective inhibition. Consensus Conference Report. J Rheumatol. 2002;29:1501-1510.
30. Mukherjee D, Nissen SE, Topol EJ. Risk of cardiovascular events associated with selective COX-2 inhibitors. JAMA. 2001;286:954-959.
31. Frishman WH. Effects of nonsteroidal anti-inflammatory drug therapy on blood pressure and peripheral edema. Am J Cardiol. 2002;89:18D-25D.
32. Celecoxib (Celebrexç™) Manufacturer Product Information. Chicago: GD Searle; June 2002.
33. Rofecoxib (Vioxx®) Manufacturer Product Information. Whitehouse Station, NJ: Merck; April 2002
34. Whelton A, Fort JG, Puma JA, et al. Cyclooxygenase-2-specific inhibitors and cardiorenal function: a randomized, controlled trial of celecoxib and rofecoxib in older hypertensive osteoarthritis patients. Am J Ther. 2001;8:85-95.
35. Komers R, Anderson S, Epstein M. Renal and cardiovascular effects of selective cyclooxygenase-2 inhibitors. Am J Kidney Dis. 2001;38:1145-1157.
36. Breyer MD, Hao C, Qi A. Cyclooxygenase-2 selective inhibitors and the kidney. Curr Opin Crit Care. 2001;7:393-400.
37. Harris CJ, Brater DC. Renal effects of cyclooxygenase-2 selective inhibitors. Curr Opin Nephrol Hypertens. 2001;10:603-610.
38. Choline magnesium trisalicylate (Trilisate®) Manufacturer Product Information. Stamford, CT: Purdue Frederick; January 2001.
39. Picado C, Fernandez-Morata JC, Juan M, et al. Cyclooxygenase-s mRNA is downexpressed in nasal polyps from aspirin-sensitive asthmatics. Am J Respir Crit Care Med. 1999;160:291-296.
40. Dear Health Care Professional Letter. Safety Information Pertaining to Bextra™ (valdecoxib). Pharmacia, Peapack, NJ, September 2002.
41. Wolfe F, Anderson J, Burke TA, et al. Gastroprotective therapy and risk of gastrointestinal ulcers: risk reduction by COX-2 therapy. J Rheumatol. 2002;29:467-473.
42. Hawkey CJ. Gastrointestinal safety of COX-2 specific inhibitors. Gastroenterol Clin North Am. 2001;30:921-937.
43. Simon LS. COX-2 inhibitors: are they nonsteroidal anti-inflammatory drugs with a better safety profile? Gastroenterol Clin North Am. 2001;30:1011-1025.
44. Lane JM. Anti-inflammatory medications: selective COX-2 inhibitors. J Am Acad Orthop Surg. 2002;10:75-78.
45. Lipsky PE. Recommendations for the clinical use of cyclooxygenase-2-specific inhibitors. Am J Med. 2001;110:3S-5S.
46. Drug Topics Red Book. Montvale, NJ: Medical Economics; 2002.
47. Peloso PM, Scheiman JM. The economic implications of cyclooxygenase-2-specific inhibitors. Am J Med. 2001; 110:50S-54S.
48. Pellissier JM, Straus WL, Watson DJ, et al. Economic evaluation of rofecoxib versus nonselective nonsteroidal anti-inflammatory drugs for the treatment of osteoarthritis. Clin Ther. 2001; 23:1061-1079.

back to top