Drugs Aging 2005; 22 (3): 185-200 1170-229X/05/0003-0185/$34.95/0

CURRENT OPINION

© 2005 Adis Data Information BV. All rights reserved.

Cyclo-Oxygenase-2 Inhibitors When Should They Be Used in the Elderly? Ruth Savage New Zealand Pharmacovigilance Centre, Department of Preventive and Social Medicine, University of Otago, Dunedin, New Zealand

Abstract

Chronic pain in the elderly is frequently a result of arthritic disorders, particularly osteoarthritis. The cyclo-oxygenase (COX)-2 inhibitors are as effective as standard NSAIDs for the relief of pain and for improving function in elderly patients with osteoarthritis and rheumatoid arthritis. COX-2 inhibitors increase the risk of serious gastroduodenal adverse reactions but there is evidence that they carry a lower risk for these adverse effects than standard NSAIDs, except when there is concurrent aspirin use. Since gastroduodenal disorders are the most frequently reported serious adverse effects of NSAIDs and these disorders occur more frequently in the elderly, COX-2 inhibitors offer an alternative to standard NSAIDs in this age group. However, they are not appropriate for many patients with cardiovascular and renal disease. The adverse reaction profile of the COX-2 inhibitors has confirmed the role of the COX-2 enzyme in renal function, salt and water homeostasis and the vascular endothelium. Thus, like standard NSAIDs, COX-2 inhibitors can cause renal failure, hypertension and exacerbation of cardiac failure. Of note is that these disorders are dose related. Thus, there are good reasons to avoid high doses of COX-2 inhibitors in the elderly. Clinical trials indicate that daily doses of rofecoxib 12.5mg, celecoxib 100–200mg, valdecoxib 10mg and etoricoxib 60mg are the minimum effective doses of these agents. Data from the New Zealand Intensive Medicines Monitoring Programme indicate that celecoxib 200 mg/day and rofecoxib 25 mg/day are/were the most commonly prescribed doses and that 6% of patients had taken rofecoxib 50 mg/day for longer than recommended. Recent research indicates that COX-2 inhibitors have a thrombotic potential, especially in high doses and when use is prolonged, and this further limits the extent to which they can be used in the elderly. Important interactions with COX-2 inhibitors in the elderly include those with warfarin, which can result in loss of control of anticoagulation, and those with ACE inhibitors, angiotensin II type 1 receptor antagonists and diuretics, which can result in loss of control of blood pressure and cardiac failure and, in hypovolaemic conditions, renal failure. The clinical significance of an interaction between celecoxib and aspirin to reduce the antiplatelet effect of the latter drug is unknown. Preliminary information from spontaneous reporting systems indicates that there may be differences in the risk of cardiac failure and hypertension between

186

Savage

standard NSAIDs and COX-2 inhibitors and between rofecoxib and celecoxib. More formal studies using equivalent doses are needed to test this observation. Use of COX-2 inhibitors may be considered in the elderly to reduce the risk of gastroduodenal complications associated with standard NSAIDs but only when consideration has first been given to use of less toxic medicines as alternatives or supplements, the appropriate dose of the COX-2 inhibitor or standard NSAID, the presence and possible impact of co-morbidities, and the implications of taking COX-2 inhibitors with any concomitant medications. Equally important is regular monitoring of the patient taking a COX-2 inhibitor for efficacy and adverse effects, and ensuring that the patient has a continuing need to keep taking the drug. Close attention also needs to be paid to intercurrent illnesses and new prescriptions that may reduce the safety of the COX-2 inhibitor. A standard NSAID plus a proton pump inhibitor may be equally effective as a COX-2 inhibitor in reducing the risk of gastroduodenal toxicity and if used the same prescribing advice applies. Current knowledge concerning the thrombotic potential of COX-2 inhibitors suggests that this combination, if tolerated, may be preferable to a COX-2 inhibitor, particularly where prolonged use is required. This knowledge also indicates that for patients with or at high risk of ischaemic heart disease or stroke, COX-2 inhibitors are contraindicated.

Chronic pain adversely impacts the quality of life of many elderly patients. Much of this pain is a result of rheumatic disorders, predominantly osteoarthritis.[1] NSAIDs often afford a welcome degree of pain relief for these patients.[2] However, adverse effects have limited their usefulness and caused a considerable burden of gastroduodenal, renal and cardiovascular disease.[3,4] The prevalence of NSAID use in patients >65 years of age is high. A case-control study conducted in New Zealand[5] indicated that approximately 15% of people in this age group were using NSAIDs. Similar findings have been reported more recently from other countries.[6,7] In addition, a Canadian study has provided evidence that cyclo-oxygenase (COX)-2 inhibitors are replacing standard NSAIDs in patients who had previously discontinued them and are also being taken by elderly new users.[7] The questions that now need addressing are whether this rapid uptake of COX-2 inhibitors is of greater benefit to the elderly than the use of standard NSAIDs and what guidelines can be given for their use. Age itself and disorders occurring most often in the elderly increase the risk of NSAID-related adverse effects.[4,5] Concomitant use of potentially in© 2005 Adis Data Information BV. All rights reserved.

teracting medicines adds to the difficulty of deciding whether a prescription is safe and appropriate. Because celecoxib and rofecoxib have been the most widely used and studied of the NSAIDs that are most selective for COX-2, this article focuses mainly on these two COX-2 inhibitors. Notwithstanding the fact that rofecoxib has now been withdrawn, the findings from research into this medicine remain relevant to our understanding of other COX-2 inhibitors. In this article, the term ‘standard NSAIDs’ refers to NSAIDs with less (i.e. than celecoxib and rofecoxib) or no COX-2 selectivity. The broader term ‘NSAIDs’ includes both standard NSAIDs and COX-2 inhibitors. 1. Is an NSAID Indicated? The first consideration when deciding whether to use a COX-2 inhibitor is whether an NSAID is indicated. The indications for NSAID treatment that are relevant to the elderly are pain and inflammation as a result of osteoarthritis or inflammatory arthritis and some musculoskeletal injuries. Also, in some circumstances, postoperative pain relief and palliative care may be indications. Low doses of ibuprofen can be used for simple analgesia. Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

Clinical trials have provided evidence of comparable efficacy between COX-2 inhibitors and standard NSAIDs. The United Kingdom National Institute for Clinical Excellence (NICE)[8] indicated in 2001 that 32 randomised, controlled trials of rofecoxib, celecoxib or both had provided evidence that they are more effective than placebo and of equivalent efficacy to standard NSAIDs in their ability to reduce resting and active pain and improve physical and global function in patients with osteoarthritis or rheumatoid arthritis. Despite this equivalence, the effectiveness of both standard NSAIDs and COX-2 inhibitors relative to other forms of treatment or even to no treatment needs to be considered before prescribing these drugs for individual patients. Studies comparing the effectiveness of NSAIDs and maximum daily doses of paracetamol in patients with osteoarthritis of the hip and knee found that paracetamol was as effective as NSAIDs for the relief of pain in mild to moderate arthritis in one-third of patients studied.[9] Another important observation from these studies was that the effect of either medicine in these patients was modest, with improvements of no more than 30mm on the 100mm arthritis pain visual analogue scale being recorded. In severe osteoarthritis, NSAIDs were significantly more effective than paracetamol.[10] However, in a randomised, controlled study of 382 patients with osteoarthritis and a mean age of 62 years, no difference was seen between rofecoxib 12.5 mg/day, celecoxib 200 mg/ day and paracetamol 4 g/day in terms of improving resting pain, morning stiffness or function.[11] Rofecoxib 25 mg/day was significantly more effective for all parameters. The American Rheumatism Association has published guidelines for the treatment of osteoarthritis.[12] Underpinning the management of osteoarthritis of the hip and knee is nonpharmacological treatment, such as muscle strengthening exercises and orthotics. In patients with mild to moderate osteoarthritis, a trial of paracetamol is worthwhile, based on the evidence quoted above, before embarking on NSAID treatment (including COX-2 inhibitors). Additionally, or as an alternative to NSAIDs, intra© 2005 Adis Data Information BV. All rights reserved.

187

articular corticosteroids can be considered and mild opioids may be added to paracetamol. Where there is severe pain and joint destruction, surgery is often indicated. In rheumatoid arthritis, careful initial assessment is needed to identify predictors of disease outcome. Where joint destruction is likely, disease modifying antirheumatic drugs (DMARDs) should be started early in the course of the disease. DMARDs with the lowest toxicities, for example, methotrexate and salazopyrin, are now recommended.[13] In patients at high risk of NSAID toxicity, use of a DMARD with low toxicity may reduce or remove the need for NSAIDs.[13] 2. Efficacy of Cyclo-Oxygenase (COX)-2 Inhibitors Efficacy studies can provide valuable information for prescribers considering the indications and contraindications for COX-2 inhibitors in the elderly. The age of the trial subjects and the lowest doses employed that were effective should be considered. Efficacy studies also provide information about the dose equivalency of the COX-2 inhibitors, which is important when considering studies that compare adverse reaction rates. A summary of early efficacy trials of celecoxib in osteoarthritis and rheumatoid arthritis was published in 2000.[14] The trials used standard measures to assess initial status and effectiveness of interventions, including patient and physician global assessment of arthritis, osteoarthritis severity, pain and quality of life. Efficacy in patients ≥75 years of age was addressed by pooling data from studies of 12 weeks’ duration that were used to assess the response to celecoxib 100mg or 200mg twice daily in patients with osteoarthritis of the hip or knee. Significant improvements in functional status and quality of life were reported with both dose levels of celecoxib compared with placebo and there was comparable efficacy with naproxen 500mg twice daily. In patients with rheumatoid arthritis, celecoxib in doses of 100mg and 200mg twice daily was effective in improving pain, morning stiffness and number of swollen joints compared with placebo. Drugs Aging 2005; 22 (3)

188

Savage

There were no further improvements with 400mg twice daily. The outcomes were comparable to those with maximal daily doses of two standard NSAIDs – naproxen and diclofenac.[15] In 1999, Scott and Lamb[16] reviewed randomised, double-blind efficacy trials of rofecoxib. All trials used patient and physician global assessments of arthritis as well as pain while walking on a flat surface to measure response to treatment in patients with osteoarthritis. A dose-ranging study showed that rofecoxib 12.5 mg/day, 25 mg/day and 50 mg/ day was more effective than placebo for improving physical functioning in patients with osteoarthritis. Rofecoxib at dosages of 12.5 mg/day and 25 mg/day had similar efficacy to full anti-inflammatory doses of three standard NSAIDs in four studies over a 6-week period. In studies of postoperative dental pain, single doses of rofecoxib 25mg and 50mg were significantly more effective than rofecoxib 12.5mg.[17] Five double-blind, randomised, controlled trials of valdecoxib in 3777 patients with osteoarthritis demonstrated that valdecoxib was superior to placebo in improving a composite index of pain, stiffness and function and that dosages of 10 mg/day and 20 mg/day were equivalent to maximum recommended daily dosages of naproxen, diclofenac and ibuprofen. Similar findings were reported for valdecoxib compared with naproxen and diclofenac in four trials of 3475 patients with rheumatoid arthritis.[18] The results of these studies indicate that the lowest starting dosages likely to be effective in osteoarthritis and rheumatoid arthritis are celecoxib 100mg twice daily, rofecoxib 12.5 mg/day and valdecoxib 10 mg/day. The lowest recommended starting dose for etoricoxib is 60 mg/day.[19] 3. Pharmacology of COX-2 Inhibitors Some knowledge of the pharmacology of the COX-2 inhibitors is helpful for understanding the adverse reaction profile of these medicines. © 2005 Adis Data Information BV. All rights reserved.

3.1 Implications of COX-2 Inhibition

The clinical effects of NSAIDs are based on inhibition of the enzyme COX. COX-1 is the isoform of the COX enzyme that is constitutively expressed and catalyses the formation of the prostanoids important in homeostasis. The COX-2 isoform is inducible and increases at sites of inflammation. Therefore, NSAIDs with a high degree of selectivity for the COX-2 isoform were expected to reduce inflammation while avoiding the adverse effects of COX-1 inhibition on homeostasis, e.g. disruption of gastric mucosal integrity. However, it is now known that under basal conditions COX-2 is expressed in the brain, renal cortex and vascular endothelium. Furthermore, production of COX-2 can be upregulated at these sites, an important factor in the development of adverse reactions. Renal COX-2 is upregulated in high renin states, such as salt restriction, congestive cardiac failure and ACE inhibition, and is important in the maintenance of glomerular perfusion in these conditions.[20] COX-2 in the vascular endothelium is upregulated in the presence of vessel wall damage and other inflammatory stimuli[21] and catalyses the production of prostacyclin (PGI2), which has vasodilating and antiaggregatory properties. This provides a theoretical basis for a prothrombotic effect of COX-2 inhibitors, but not for standard NSAIDs, which do not have the opposing effect of inhibiting the production of proaggregatory thromboxane A2, a product of COX-1 activity.[22,23] 3.2 COX-2 Specificity

The following studies used whole blood assays of thromboxane A2 and B2 as a measure of COX-1 activity and lipopolysaccharide stimulation of PGI2 synthesis as a measure of COX-2 activity. Studies of volunteers indicated that rofecoxib exhibited dose-dependent inhibition of COX-2 activity with no significant inhibition of COX-1 activity at 12.5 mg/day and 25 mg/day[24] Celecoxib showed some COX-1 inhibitory activity at 800 mg/ day in that it significantly inhibited thromboxane B2 production. However it did not affect stimulated platelet aggregation.[25] Riendeau et al.[26] identified Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

and tabulated the order of COX-2 selectivity of a range of COX-2 inhibitors and NSAIDs and demonstrated that etoricoxib, rofecoxib and valdecoxib were more selective COX-2 inhibitors than celecoxib. 3.3 Pharmacokinetics

In the elderly, the area under the (plasma) concentration-time curve (AUC) is increased by 34% for celecoxib[27] and 30% for valdecoxib[18] compared with younger patients. Celecoxib undergoes extensive hepatic metabolism via cytochrome P450 (CYP) 2C9 and is also an inhibitor of CYP206. This is important because of the potential for interactions and because of genetic variability of CYP2C9 expression.Valdecoxib is metabolised by multiple pathways that include CYP3A4 and CYP2C9.[18] The metabolism of etoricoxib is also facilitated predominantly by CYP enzymes.[19] 4. Precautions and Contraindications to Use of COX-2 Inhibitors The gastrointestinal, renal and cardiovascular adverse effects of COX-2 inhibitors have particular relevance in the elderly. Most of these adverse effects also occur with standard NSAIDs and it is important to assess, if possible, if there are clinically important variations in the severity and incidence of these adverse reactions between standard NSAIDs and COX-2 inhibitors and between medicines within these groups. Prior to prescribing a COX-2 inhibitor, the physician should carefully assess the patient for co-morbidities that might increase the risk of adverse reactions. It is also important to scrutinise current treatment regimens for potentially interacting medicines. 4.1 Gastroduodenal Toxicity

Serious gastroduodenal toxicity is the most important adverse effect of NSAIDs[28] and, thus, has the greatest influence on the choice of NSAID. The risk of gastroduodenal toxicity with NSAIDs is increased when there is a past history of ulcer, bleeding or perforation, when warfarin or corticosteroids © 2005 Adis Data Information BV. All rights reserved.

189

are co-prescribed, when the patient has serious comorbidities, where prolonged use of maximum doses of standard NSAIDs is required, or when the patient is >65 years of age.[8] Thus, all elderly patients are at increased risk of gastroduodenal toxicity from NSAIDs.[8] Many studies have shown a decreased incidence of gastric erosions on endoscopy with celecoxib and rofecoxib compared with standard NSAIDs[8] but the importance of these as a surrogate endpoint for major adverse gastroduodenal effects is uncertain. Mortality as a result of gastroduodenal ulcer haemorrhage and perforation is approximately 10% and most deaths occur in the elderly.[3] Results from a meta-analysis of randomised trials comparing rofecoxib with standard NSAIDs in patients with osteoarthritis (n = 5435),[29] together with findings from the VIGOR (Vioxx Gastro-Intestinal Outcomes Research) study, a randomised, prospective trial of 8076 patients with rheumatoid arthritis,[30] showed a significant 50% reduction in risk of peptic ulcer complications and/or upper gastrointestinal bleeding with rofecoxib compared with standard NSAIDs. In another randomised trial, the CLASS (Celecoxib Long-Term Arthritis Safety Study) investigators compared maximum recommended daily doses of ibuprofen and diclofenac with supratherapeutic doses of celecoxib in 8059 patients with osteoarthritis.[31] After 6 months this study showed a significant reduction in risk of gastroduodenal bleeding, perforation or obstruction with celecoxib compared with ibuprofen and with combined NSAIDs but not diclofenac and only when aspirin users were excluded. At 12 months, again in nonaspirin users, there was a significant difference in risk of gastroduodenal bleeding, perforation or obstruction between celecoxib and ibuprofen but not diclofenac or combined NSAIDs.[32] Thus, the use of low-dose aspirin appears to negate the gastrointestinal advantage of the COX-2 inhibitors (see section 5.2.1). The investigators argued that the absence of a statistically significant difference in risk of gastroduodenal bleeding, perforation or obstruction between celecoxib and diclofenac was due to a larger Drugs Aging 2005; 22 (3)

190

proportion of patients withdrawing from the diclofenac group because of gastrointestinal adverse events.[32] In a large observational controlled cohort study of patients ≥66 years of age,[7] a 4-fold increase in the risk of upper gastrointestinal haemorrhage was observed in patients taking standard NSAIDs compared with controls not exposed to NSAIDs, a finding that was consistent with the results of other studies.[5] The same study also showed a 2-fold increase in risk with rofecoxib compared with controls, and no increase in risk with celecoxib. The difference in risk between the COX-2 inhibitors and standard NSAIDs was significant. This is the only study that has particularly addressed patients with risk factors, in this case the elderly. The recently published TARGET (Therapeutic Arthritis Research and Gastrointestinal Event Trial) trial included 18 325 patients randomised to receive maximum daily doses of lumiracoxib or ibuprofen in one arm of the study and lumiracoxib or naproxen in the other.[33] There was a significant 4-fold reduction in risk of ulcer complications with lumiracoxib compared with standard NSAIDs, but again no risk reduction was seen in patients also taking low-dose aspirin. No prospective long-term studies to comparing the incidence of serious clinically significant upper gastrointestinal events in patients taking valdecoxib[18] versus standard NSAIDs have been performed and similar studies with etoricoxib have not been identified in the literature. A recent multicentre, case-control study has ranked standard and newer NSAIDs according to their associated risk of upper gastrointestinal bleeding.[34] The numbers of cases and controls taking COX-2 inhibitors were small but the odds ratio rankings indicated that if diclofenac and ibuprofen rather than naproxen had been used as comparators in the VIGOR study the comparative risk reduction with rofecoxib would have been less. This was not evident in the TARGET study of lumiracoxib,[33] probably because ibuprofen was used in full antiinflammatory doses, which is not usual in clinical practice. © 2005 Adis Data Information BV. All rights reserved.

Savage

Adverse reaction reports to several national pharmacovigilance centres including those in New Zealand,[35] Australia[36] and the US[37] indicate that, as with standard NSAIDs, serious upper gastrointestinal reactions to COX-2 inhibitors occur predominantly, but not exclusively, in elderly patients. This is especially so in those of great age and those with other risk factors. Of concern is that there are no data on the degree of risk reduction associated with use of COX-2 inhibitors in patients with a past history of gastroduodenal ulcer, bleeding or perforation. However, one study has compared the incidence of recurrence of ulcer bleeding in patients treated with either celecoxib 400 mg/day or diclofenac 150 mg/day plus omeprazole.[38] Recurrences occurred in 4.9% and 6.4%, respectively. Thus, the regimens were almost equally effective but neither completely protected against recurrence. Most of the patients in this study were elderly and had co-existing medical conditions. Therefore, present knowledge suggests that a COX-2 inhibitor or a standard NSAID with a proton pump inhibitor would be of equal merit where there is a risk of gastroduodenal toxicity. The choice will be influenced by hypersensitivity and co-morbidity. Attention should be paid to the results of metaanalyses of standard NSAID studies[39] and the NICE guidelines,[8] which indicate that gastrointestinal toxicity with NSAIDs and COX-2 inhibitors is dose related. Also, if a standard NSAID is used, rankings indicate that diclofenac and ibuprofen at daily doses of ≤1200mg carry the lowest risk.[3] It is possible that haemorrhage caused by COX-2 inhibitors may be less severe than that caused by standard NSAIDs because COX-2 inhibitors do not inhibit platelet thromboxane A2 production, but there are no published investigations addressing this hypothesis. 4.2 Lower Gastrointestinal Tract Toxicity

Standard NSAIDs have been known, on rare occasions, to cause ulceration, haemorrhage and diaphragm-like strictures of the distal small intestine and the large intestine.[40] There is also a published Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

report of exacerbation of inflammatory bowel disease with celecoxib[41] and one report of a woman who developed diaphragm-like strictures of the colon while taking celecoxib.[42] For these reasons, patients who develop symptoms suggestive of gastrointestinal ulceration or obstruction should discontinue COX-2 inhibitors while awaiting investigation. 4.3 Renal Failure

Biochemical and clinical studies indicate that rofecoxib and celecoxib produce qualitative changes in urinary prostaglandin excretion, glomerular filtration rate, sodium retention and their consequences in a similar way to standard NSAIDs.[20] Renal prostaglandins control sodium and potassium homeostasis even in healthy individuals but their haemodynamic effects are only operative in stressed states.[20] A search of the US FDA Adverse Event Reporting System[43] revealed that both rofecoxib and celecoxib are associated with adverse renal events similar to those produced by standard NSAIDs. Patients at greatest risk were elderly and/or had pre-existing renal impairment, heart failure or hepatic dysfunction, or were taking ACE inhibitors and diuretics. These reports support observations that COX-2 may be important in modulating renal function in conditions of reduced effective circulating blood volume. In the study of the incidence of recurrence of gastroduodenal ulcer bleeding mentioned in section 4.1, 51.4% of patients taking celecoxib who had renal failure at the start of the study and 40.7% of similar patients taking diclofenac and omeprazole experienced renal adverse events during the study.[38] Thus, renal impairment, dehydration and other conditions leading to a reduced effective blood volume all increase the risk of acute renal failure in patients taking NSAIDs. Less predictable is interstitial nephritis, which is also a known adverse effect of standard NSAIDs. There are published reports of this disorder occurring with both rofecoxib and celecoxib.[44,45] 1

191

A widely quoted study based on a Bayesian analysis of the WHO International Drug Monitoring database indicated that the drug/adverse reaction combination rofecoxib/acute renal failure stood out statistically from the background data to a significantly greater extent than did the celecoxib/acute renal failure combination.[46] This study was derived from the database but written by independent investigators. The difference in reporting rates suggested that both medicines were likely to cause acute renal failure, but rofecoxib might do so to a greater degree. This study had some limitations in that doses could not be directly compared, and the COX-2 inhibitors were in the early post-marketing phase so that the measures carried out were still unstable. In June 2003 there was no statistical difference in the degree to which celecoxib (Information Component [IC], 1.52, IC-2 SD 1.33)1 and rofecoxib (IC 1.59, IC-2 SD 1.39) were reported to cause acute renal failure.[47] 4.4 Hypertension or Cardiac Failure

In support of a role for COX-2 in sodium and potassium homeostasis in physiological conditions, a study of salt replete elderly subjects found that rofecoxib 25mg, celecoxib 400mg and naproxen 1000mg reduced sodium excretion to a similar extent.[48] Another study in elderly subjects confirmed this finding.[49] Thus, COX-2 inhibitors are likely to elevate blood pressure and aggravate cardiac failure. Recently, two studies that attempted to quantify this adverse effect in patients taking standard NSAIDs reported a 10-fold increase in risk of relapse of cardiac failure, but not in first episodes, with standard NSAIDs compared with no NSAID use.[4,50] The magnitude of this effect means that it is important to be aware of any differences in risk of cardiac failure between NSAIDs and COX-2 inhibitors. Analysis of proportional reporting ratios[51] of spontaneous reports of cardiovascular adverse reactions attributed to rofecoxib and celecoxib and three standard NSAIDs in the WHO International Drug Monitoring database (figure 1) generated the hy-

Information Component (IC) is a measure of disproportionality that is considered significant when IC – 2 SD is >0.

© 2005 Adis Data Information BV. All rights reserved.

Drugs Aging 2005; 22 (3)

192

Savage

Celecoxib (29 998) Diclofenac (26 105) Ibuprofen (23 879) Naproxen (20 601) Rofecoxib (31 804)

Angina

Arrhythmia

Cardiac failure

Cerebral haemorrhage

Cerebral infarction

Embolism

Heart block

Myocardial infarction

0

0.2

0.4

0.6

0.8

1.0

1.2

Percentage Fig. 1. Proportions of cardiovascular reactions reported to the WHO compared with total reactions reported for each medicine (diclofenac, ibuprofen, naproxen 1990–2001; celecoxib, rofecoxib 1999–2001).[51]

pothesis that there may be differences in risk of cardiac failure between the COX-2 inhibitors and standard NSAIDs. In addition to a greater proportion of thrombotic events reported with COX-2 inhibitors, there was a 6-fold greater proportion of reports of cardiac failure with the COX-2 inhibitors overall. Furthermore, Bayesian analysis of the WHO International Drug Monitoring Database indicated that the IC values[47] for cardiac failure and hypertension were greater for rofecoxib than for celecoxib, indicating a higher reporting rate for the former. Mamdani et al.,[52] in a case-control study, also found a greater risk of exacerbation of heart failure with rofecoxib than with celecoxib or standard NSAIDs. The results of one clinical study suggested a greater incidence of hypertension and oedema with rofecoxib than with celecoxib in elderly patients,[53] © 2005 Adis Data Information BV. All rights reserved.

but again there was ensuing debate about dose equivalency.[54] A subsidiary analysis of the VIGOR study demonstrated a greater incidence of hypertension with rofecoxib than with naproxen. However, rofecoxib was given at supratherapeutic doses.[55] An analysis of reports in the FDA spontaneous adverse events database revealed that hospital admission because of acute hypertension was reported 3.8 times more often with rofecoxib than with celecoxib and there were no such reports for two standard NSAID comparators.[56] In a study of 7111 patients with osteoarthritis taking etoricoxib 90 g/ day compared with diclofenac 150 g/day, hypertensive adverse experiences occurred at a higher rate with etoricoxib.[19] In contrast, in the TARGET study,[57] the increase in systolic blood pressure was significantly less with lumiracoxib than with comparator NSAIDs. It is difficult to envisage a qualitaDrugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

tive difference between COX-2 inhibitors with regard to hypertension and cardiac failure and the differences found in voluntary reporting systems may reflect the doses used. When prescribing, the assumption has to be made that COX-2 inhibitors are as likely as standard NSAIDs to aggravate cardiac failure and hypertension. As they may also interact with the ACE inhibitor/diuretic combination frequently used to treat cardiac failure (see section 5.1.1), this condition is a contraindication for all NSAID use. 4.5 Myocardial Infarction

The VIGOR study demonstrated a 4-fold increase in risk of myocardial infarction with rofecoxib 50 mg/day compared with naproxen.[30] A retrospective cohort study of patients >50 years of age from the Tennessee Medicaid programme also found an increase in risk but only in new users taking >25 mg/day.[58] The APPROVe (Adenomatous Polyp Prevention with VIOXX) study was a randomised, placebo-controlled trial of 2600 patients conducted to evaluate rofecoxib 25 mg/day for the prophylaxis of adenocarcinoma in patients with familial colon polyposis. This study was discontinued early and led to the market withdrawal of rofecoxib because of a 2-fold increase in risk of myocardial infarction and stroke compared with placebo. In contrast to the Tennessee study, the increase in risk became significant only after patients had taken rofecoxib regularly for >18 months.[59] A nested case control study of myocardial infarction, sudden cardiac death and NSAIDs in a cohort of 1 394 764 patients derived from the Californian Kaiser Permanente managed care organisation database identified 8143 index patients.[60] Compared with both remote NSAID use and celecoxib use there was an increased risk of acute myocardial infarction with rofecoxib which became significant at daily doses of >25 mg/day. Compared with remote NSAID use the increase in risk with rofecoxib was 3-fold at >25 mg/day. Small significant increases in risk (odds ratio, 1.13) were found with standard NSAIDs, including naproxen, indicating that they are not protective against myocardial in© 2005 Adis Data Information BV. All rights reserved.

193

farction. These findings might be explained by the effect of NSAIDs on blood pressure.[60] Thus, both prolonged use of rofecoxib and doses >25 mg/day have been implicated in increased risk of myocardial infarction. As indicated in section 3.1, there is biological plausibility for COX-2 inhibitors having prothrombotic activity.[22] Thus, it is important to know if COX-2 inhibitors other than rofecoxib are also prothrombotic. Five published studies have not shown an increased risk of myocardial infarction with celecoxib compared with standard NSAIDs. The CLASS study did not show an increased incidence of cardiovascular disorders with celecoxib at high doses compared with diclofenac and ibuprofen after 12 months.[32] Two studies did not find a significant increase in risk with rofecoxib either[61,62] One of these studies was confined to patients >65 years of age. The Kaiser Permanente study did not show an increased risk of myocardial infarction with celecoxib compared with remote NSAID use and the risk with high dose rofecoxib was significantly greater than with celecoxib. Odds ratios with different doses of celecoxib were not stated.[60] An unpublished trial of celecoxib 400 mg/day versus placebo for the prevention of spontaneous colorectal polyps did not demonstrate an increased risk of cardiovascular adverse events with celecoxib.[62] However, a similarly designed trial, the Adenoma Prevention with Celecoxib (APC) trial in 2035 patients undertaken to assess the efficacy of celecoxib in preventing adenomatous colon polyp recurrence, demonstrated a 2.5-fold increase in risk of cardiovascular death, nonfatal myocardial infarction and stroke with celecoxib 400 mg/day and a 3.4-fold increase with 800 mg/day, compared with placebo. The patients were exposed to celecoxib for more than 2.5 years.[63] Prior to the findings of the APC study it was thought that one possible explanation for an increased risk of myocardial infarction with rofecoxib but not celecoxib was greater COX-2 specificity of rofecoxib (particularly as celecoxib was given at 800 mg/day in the CLASS study and, as such, may have had more COX-1 inhibitory activity). Drugs Aging 2005; 22 (3)

194

Lumiracoxib has greater COX-2 specificity than rofecoxib. Although the TARGET study[57] found an increased rate of myocardial infarction with lumiracoxib compared with naproxen this did not reach statistical significance and there was no increase compared with ibuprofen. These findings may not be conclusive for lumiracoxib despite the large cohort studied because the number of events was small and patients with myocardial infarction, stroke, previous coronary artery revascularisation and congestive heart failure were largely excluded from the study. Furthermore the study did not extend beyond 18 months. There are no relevant published long-term studies for valdecoxib or etoricoxib. However, there are two trials[64] in which a combination of a single dose of parecoxib, an intravenous prodrug of valdecoxib, followed by an oral dose of valdecoxib 40 mg/day was compared with placebo or placebo followed by valdecoxib for the relief of post-operative pain following coronary artery bypass graft surgery. In these studies, valdecoxib, with or without parecoxib, was associated with an approximately 3-fold increase in coronary and cerebrovascular adverse events and this became statistically significant when a meta-analysis was performed on the two studies. Thus, there is evidence for an increased risk of thrombotic cardiovascular disorders compared with placebo in patients taking ≥400 mg/day of celecoxib for prolonged periods and for high risk patients taking large doses of valdecoxib for 10–14 days. Data from cohort studies, such as the Intensive Medicines Monitoring Programme (IMMP), should provide useful direct comparisons between COX-2 inhibitors for patients in normal clinical practice. Although the evidence is less strong for celecoxib and lumiracoxib than for rofecoxib, and the other COX-2 inhibitors have not been studied as extensively, there is enough cause for concern to recommend that these medicines be contraindicated in patients at high risk of thrombotic cardio- and cerebrovascular events. In all patients, prolonged use and high doses should be avoided where possible. One would normally avoid in the elderly those doses of COX-2 inhibitors that have been associated with © 2005 Adis Data Information BV. All rights reserved.

Savage

increased risk, and particular attention should be paid to the cardiovascular status of those requiring treatment for prolonged periods. This is not only because of a possible prothrombotic effect but also because hypertension and cardiac failure may increase the risk of myocardial infarction. 4.6 Neuropsychiatric and Ophthalmic Disorders

Analysis of spontaneous reporting data in two countries revealed that neuropsychiatric reactions were the third most common type of standard NSAID-related adverse effects reported.[65] Although rarely severe, psychiatric events reported have included depression, hallucinations, nightmares, anxiety and amnesia. There are also published reports of exacerbations of psychotic and depressive illness associated with standard NSAIDs.[66] The New Zealand IMMP has received similar reports of deterioration in psychiatric disorders with COX-2 inhibitor use as well as the more minor neuropsychiatric adverse effects also seen with standard NSAIDs.[67] The New Zealand IMMP has also received reports of acute temporary visual disturbance, including visual field defects, in patients taking COX-2 inhibitors; again, these are adverse effects that have also been seen with standard NSAIDs.[68] As well as providing evidence for COX-2 activity in the brain and retina, these reactions are important because they can mimic naturally occurring diseases that are more common in older persons, such as dementia and thrombotic phenomena. A trial withdrawal of COX-2 inhibitors and standard NSAIDs is recommended should these conditions occur. 5. Drug Interactions A recent study from Italy of drug use in elderly outpatients listed ACE inhibitors, diuretics, NSAIDs and regular low-dose aspirin as the most frequently used medicines.[6] There are complex pharmacodynamic interactions between NSAIDs, or alternatively COX-2 inhibitors, and these three other commonly used medicines. Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

5.1 COX-2 Inhibitors and Diuretics, ACE Inhibitors and Angiotensin II Type 1 Receptor Blockers 5.1.1 Loss of Control of Hypertension and Cardiac Failure

By decreasing the synthesis of prostaglandins that affect vascular tone, fluid homeostasis and the renin-angiotensin system, COX-2 inhibitors, like standard NSAIDs, are likely to antagonise the effects of ACE inhibitors, angiotensin II type 1 receptor antagonists, diuretics and β-adrenoceptor antagonist agents,[69] leading to loss of control of blood pressure in some patients. It is also likely that loss of control of cardiac failure will occur.[4,50] 5.1.2 Acute Renal Failure

The Australian Adverse Reactions Advisory Committee (ADRAC) has noted that ACE inhibitors, diuretics and NSAIDs, individually or in combination, are involved in over 50% of the reports it has received of acute renal failure.[70] More recently, similar reports have been received involving COX-2 inhibitors instead of standard NSAIDs.[71] The mean age of the patients affected was 76 years. The theoretical basis for this adverse effect is that hypovolaemic states may predispose patients to renal insufficiency by lowering the afferent glomerular arteriolar pressure. Compensatory mechanisms include dilation of afferent arterioles caused by renal prostaglandin activity, and activation of the renin-angiotensin system with resultant constriction of the efferent arterioles and a further increase in prostaglandin activity. Renal failure may be precipitated by NSAIDs or COX-2 inhibitors that impair renal prostaglandin synthesis and by ACE inhibitors or angiotensin II type 1 receptor antagonists that reduce angiotensin II activity.[72] The elderly are at risk of lowered effective circulating blood volume because of renal impairment, congestive cardiac failure, diuretic-induced dehydration and hypotension, and inadequate fluid intake. Therefore, NSAIDs, including COX-2 inhibitors, should not be prescribed in patients with these conditions and, ideally, should not be prescribed at all in patients taking an ACE inhibitor or angioten© 2005 Adis Data Information BV. All rights reserved.

195

sin II type 1 receptor antagonist in combination with a diuretic. If the requirement for pain relief is such that this type of combination is considered desirable, the NSAID or COX-2 inhibitor should be withdrawn immediately if a hypovolaemic state develops. 5.2 COX-2 Inhibitors and Aspirin 5.2.1 Gastrointestinal Adverse Effects

Evidence that aspirin may mitigate against the advantage of COX-2 selectivity is derived from results over the first six months of the CLASS study for celecoxib[31] and from the TARGET study for lumiracoxib.[33] As noted in section 4.1, the CLASS study[31] showed a significant reduction in risk of gastroduodenal bleeding or perforation with celecoxib but only when aspirin users were excluded. Similarly, the TARGET study[33] reported a significant reduction in the risk of ulcer complications with lumiracoxib compared with standard NSAIDs, but only in patients not also taking low-dose aspirin. More data are needed to determine if a proton pump inhibitor should be given to patients taking a COX-2 inhibitor and aspirin and whether this would confer any greater protection against gastroduodenal toxicity than a standard NSAID with a proton pump inhibitor. 5.2.2 Prevention of Myocardial Infarction

As COX-2 inhibitors do not inhibit the synthesis of thromboxane, aspirin is indicated in patients taking COX-2 inhibitors who require prophylaxis for myocardial infarction. However, both aspirin and standard NSAIDs inhibit the formation of platelet thromboxane. Aspirin has an irreversible effect for the lifetime of the platelet while the effect of other NSAIDs is reversible.[73] Recently, it has been shown that aspirin and some NSAIDs may interact to cause a loss of effect of aspirin. In a randomised, cross-over study,[74] inhibition of serum thromboxane B2 production and platelet aggregation in patients given daily low-dose aspirin for 6 days was blocked when each aspirin dose was preceded by a therapeutic dose of ibuprofen. This was not the case when diclofenac, rofecoxib or paracetamol were given instead of ibuprofen. A biochemical study[73] Drugs Aging 2005; 22 (3)

196

Savage

also indicated that celecoxib, but not rofecoxib or etoricoxib, could inhibit aspirin activity at therapeutic doses. A recent re-analysis of the Physician’s Health Study showed an increased risk of myocardial infarction in patients taking NSAIDs regularly with aspirin, compared with aspirin alone, but intermittent NSAID use did not increase the risk.[75] However, a retrospective cohort study[76] of 234 769 patients discharged from hospital following a myocardial infarction did not show any difference in the rates of death in the year following discharge between patients taking aspirin alone, patients taking aspirin and ibuprofen, and patients taking aspirin and other NSAIDs. The investigators concluded that the current level of evidence was not sufficient to make definitive recommendations for or against the concomitant use of ibuprofen and aspirin. Published data to predict the clinical outcome of a celecoxib/ low-dose aspirin combination compared with a rofecoxib/low-dose aspirin prescription have not been identified. 5.3 COX-2 Inhibitors and Warfarin

The ADRAC and the New Zealand Pharmacovigilance Centre have received and published case reports of elevated International Normalised Ratios (INR) and haemorrhage when celecoxib and rofecoxib have been given to patients stabilised on warfarin. Small increases in INR have also been noted with valdecoxib.[77-79] It is possible that celecoxib competes with warfarin for the CYP2C9 substrate. It is not known if this interaction is confined to poor CYP2C9 metabolisers. There is no identifiable mechanism for a rofecoxib/warfarin interaction. Warfarin is usually considered a contraindication for standard NSAID use largely because of the increased risk of haemorrhage, particularly gastroduodenal. Although there are no data to confirm that co-prescription of COX-2 inhibitors with warfarin is safer than co-prescription of standard NSAIDs with warfarin, the lower risk of upper gastrointestinal haemorrhage and lack of effect on platelet thromboxane of COX-2 inhibitors should reduce the © 2005 Adis Data Information BV. All rights reserved.

risk of severe consequences. If there is co-prescription, careful monitoring of the INR is required. 5.4 Other Pharmacokinetic Interactions

Reduced renal clearance of methotrexate in patients also taking COX-2 inhibitors is important because this combination may be used in patients with rheumatoid or psoriatic arthritis. However, the clinical significance of the interaction is unknown.[69] A published case report indicates that a similar interaction with lithium is of clinical concern.[80] Celecoxib may inhibit the metabolism of CYP2D6 substrates and its metabolism may be altered by medicines affecting CYP2C9 isoenzymes. Plasma levels of valdecoxib have been increased by concomitant use of fluconazole and ketoconazole, probably due to inhibition of CYP3A4.[18] 6. Patterns of COX-2 Inhibitor Use Efficacy studies indicate that celecoxib 200 mg/ day is equivalent to rofecoxib 12.5 mg/day.[11] However, a Spanish study[81] and data from the New Zealand IMMP[82] suggest that the most commonly used daily doses are celecoxib 200mg and rofecoxib 25mg. Furthermore, 6% of patients in the IMMP study had taken rofecoxib 50 mg/day for more than 5 days, the maximum recommended duration for this dose. In the cohort of 6710 patients taking rofecoxib or celecoxib, 11% of patients were ≥80 years of age, 66% had a history of upper gastrointestinal disorders and 33% had a history of cardiovascular disorders. Notably, 5% had a history of cardiac failure. Solomon et al.[62] in the US found that, compared with standard NSAID users, COX-2 inhibitor users were less healthy at baseline and had more cardiovascular disorders. Thus, high doses of COX-2 inhibitors are being prescribed despite the fact that a large proportion of patients receiving them are predisposed to dose-related adverse reactions to these drugs. A lesson from experience with rofecoxib is that the starting doses of COX-2 inhibitors in the elderly must be the lowest effective doses found in clinical trials rather than the doses most heavily promoted. If Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

low doses are incompletely effective in the elderly then a trial of adding another agent, for example, paracetamol and/or codeine, should be considered. 7. Conclusion Inadequately treated pain in the elderly reduces quality of life and may impair mobility. Those who are still active may be unable to exercise and thus lose the ensuing health benefits. The widespread use of standard NSAIDs and, more recently, COX-2 inhibitors indicates a need for pain relief that is met more by these medicines than available alternatives. When considering prescribing a COX-2 inhibitor there should be a clear indication for NSAID use and consideration should have been given to alternatives, possible drug interactions and co-morbidities. The only advantage COX-2 inhibitors have over standard NSAIDs is a lower risk of serious gastroduodenal toxicity. For rofecoxib, which is now withdrawn, and for lumiracoxib, there is good evidence for a reduction in risk. The evidence with celecoxib is less secure and long-term studies of this endpoint have not been conducted for valdecoxib and etoricoxib. However, the study showing a similar rate of recurrence of ulcer bleeding with celecoxib compared with diclofenac and omeprazole suggests that, for the elderly, COX-2 inhibitors or standard NSAIDs, especially those shown to carry a lower risk,[39] together with a proton pump inhibitor, are of equal value where an NSAID is indicated. New prescribing advice is being considered by several regulatory authorities. The present evidence indicates that patients at high risk of ischaemic cardio- or cerebrovascular disease should avoid COX-2 inhibitors and those requiring prolonged NSAID treatment or high doses should use COX-2 inhibitors only if they cannot tolerate the NSAID/ proton pump inhibitor combination. Every effort should be made to avoid more than the lowest recommended doses of standard NSAIDs or COX-2 inhibitors in the elderly, for example, by supplementing with other analgesics. There is no advantage in prescribing a COX-2 inhibitor if the patient is taking aspirin. © 2005 Adis Data Information BV. All rights reserved.

197

COX-2 inhibitors, like standard NSAIDs, are contraindicated in active gastroduodenal disease and cardiac and renal failure. They are ideally contraindicated in renal impairment, remembering that in the elderly, renal impairment may be present when the plasma creatinine is normal, and in patients taking the combination of an ACE inhibitor or angiotensin II type 1 receptor antagonist with a diuretic. However, patients who obtain significant relief from chronic pain with NSAIDs may prefer to accept a degree of risk in exchange for quality of life. The advice to use the lowest effective dose of a COX-2 inhibitor, intermittently if possible, is particularly important for these patients. Close monitoring is required for patients taking warfarin and those with hypertension. There should be a trial withdrawal of COX-2 inhibitors if there is evidence of gastroduodenal or intestinal ulceration or obstruction and if psychiatric or visual disturbances occur. When prolonged use of COX-2 inhibitors is indicated there should be regular clinical and biochemical monitoring and re-evaluation of need.[8] An analysis of case histories in the WHO International Drug Monitoring database indicated that many patients taking COX-2 inhibitors or standard NSAIDs have a myocardial infarction or develop cardiac failure as the end result of a cascade of events which may include gastrointestinal haemorrhage, sepsis and renal failure. Thus, the prescriber needs to be alert for intercurrent conditions that may make the COX-2 inhibitor prescription unsafe. Acknowledgements The author acknowledges the assistance of Mrs Janelle Ashton of the New Zealand Pharmacovigilance Centre in collecting and collating references. The New Zealand Pharmacovigilance Centre operates through a grant from Medsafe a unit of the New Zealand Ministry of Health. The opinions expressed in this article are the personal opinions of the author. The author has no conflicts of interest that are directly relevant to the content of this review.

Drugs Aging 2005; 22 (3)

198

Savage

References 1. Brooks PM. Impact of osteoarthritis on individuals and society: how much disability? Social consequences and health economic implications. Curr Opin Rheumatol 2002; 14 (5): 573-7 2. Rang HP, Dale MM, Ritter JM. Anti-inflammatory and immunosuppressive drugs. In: Rang HP, Dale MM, Ritter JM, editors. Pharmacology. 4th ed. Sydney: Churchill Livingstone, 1999: 229-34 3. Langman MJS. Ulcer complications associated with anti-inflammatory drug use: what is the extent of the disease burden? Pharmacoepidemiol Drug Saf 2001; 10: 13-9 4. Page J, Henry D. Consumption of NSAIDs and the development of heart failure in elderly patients: an underrecognised public health problem. Arch Intern Med 2002; 160 (6): 777-84 5. Savage RL, Moller PW, Ballantyne CL, et al. Variation in the risk of peptic ulcer complications with nonsteroidal antiinflammatory drug therapy. Arthritis Rheum 1993; 36 (1): 84-90 6. Pilotto A, Franceschi M, Leandro G, et al. NSAID and aspirin use by the elderly in general practice: effect on gastrointestinal symptoms and therapies. Drugs Aging 2003; 20 (9): 701-10 7. Mamdani M, Rochon PA, Juurlink DN, et al. Observational study of upper gastrointestinal haemorrhage in elderly patients given selective cyclo-oxygenase-2 inhibitors or conventional non-steroidal anti-inflammatory drugs. BMJ 2002; 325 (7365): 624 8. National Institute for Clinical Excellence. Guidance on the use of cyclo-oxygenase (COX) II selective inhibitors, celecoxib, rofecoxib, meloxicam and etodolac for osteoarthritis and rheumatoid arthritis. Technology No. 27. London: Government Publication, 2001 9. Felson DT. The verdict favors nonsteroidal antiinflammatory drugs for treatment of osteoarthritis and a plea for more evidence on other treatments. Arthritis Rheum 2001; 44 (7): 1477-80 10. Pincus T, Koch GG, Sokka T, et al. A randomized, doubleblind, crossover clinical trial of diclofenac plus misoprostol versus acetaminophen in patients with osteoarthritis of the hip or knee. Arthritis Rheum 2001; 44 (7): 1587-98 11. Geba GP, Weaver AL, Polis AB, et al. Efficacy of rofecoxib, celecoxib, and acetaminophen in osteoarthritis of the knee: a randomized trial [published erratum appears in JAMA 2002 Feb 27; 287 (8): 989]. JAMA 2002; 287 (1): 64-71 12. American College of Rheumatology Osteoarthritis Guidelines Subcommittee. Recommendation for the medical management of osteoarthritis of the hip and knee. Arthritis Rheum 2000; 43 (9): 1905-15 13. Harrison A. Disease-modifying anti-rheumatic drugs (DMARDs) for rheumatoid arthritis: benefits and risks. Prescriber Update 1999; 18: 4-12 14. Clemett D, Goa KL. Celecoxib: a review of its use in osteoarthritis, rheumatoid arthritis and acute pain. Drugs 2000; 59 (4): 957-80 15. Emery P, Zeidler H, Kvien TK, et al. Celecoxib versus diclofenac in long-term management of rheumatoid arthritis: randomised double-blind comparison. Lancet 1999; 354 (9196): 2106-11 16. Scott LJ, Lamb HM. Rofecoxib. Drugs 1999; 58 (3): 499-505 17. Ehrich EW, Dallob A, De Lepeleire I, et al. Characterization of rofecoxib as a cyclooxygenase-2 isoform inhibitor and demonstration of analgesia in the dental pain model. Clin Pharmacol Ther 1999; 65: 336-47

© 2005 Adis Data Information BV. All rights reserved.

18. Bextra®. New Zealand data sheet: Pharmacia. 7th February 2003 [online]. Available from URL: http://www.medsafe.govt.nz [Accessed 2005 Feb 15] 19. Arcoxia®. New Zealand data sheet: Merck, Sharp & Dohme (NZ) Ltd. November 2004 [online]. Available from URL: http://www.medsafe.gov.nz [Accessed 2005 Feb 15] 20. Brater DC, Harris C, Redfern JS, et al. Renal effects of COX-2-selective inhibitors. Am J Nephrol 2001; 21: 1-15 21. Belton O, Byrne D, Kearney D, et al. Cyclooxygenase-1 and -2-dependent prostacyclin formation in patients with atherosclerosis. Circulation 2000; 102 (8): 840-5 22. Fitzgerald GA, Patrono C. The coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med 2001; 345 (6): 433-42 23. Clark DWJ, Layton D, Shakir SAW. Do some inhibitors of COX-2 increase the risk of thromboembolic events? Drug Saf 2004; 27 (7): 427-56 24. Van Hecken A, Schwartz JI, Depr´e M, et al. Comparative inhibitory activity of rofecoxib, meloxicam, diclofenac, ibuprofen, and naproxen on COX-2 versus COX-1 in healthy volunteers. J Clin Pharmacol 2000; 40 (10): 1109-20 25. McAdam BF, Catella-Lawson F, Mardini IA, et al. Systemic biosynthesis of prostacyclin by cyclooxygenase (COX)-2: the human pharmacology of a selective inhibitor of COX-2. Proc Natl Acad Sci U S A 1999; 96 (1): 272-7 26. 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: 558-68 27. Celebrex® New Zealand data sheet. Pfizer New Zealand Ltd [online]. Available from URL: http://www.medsafe.govt.nz [Accessed 2004 Apr 29] 28. Committee on Safety of Medicines: CSM update. Nonsteroidal anti-inflammatory drugs and serious gastrointestinal adverse reactions-2. BMJ (Clin Res Ed) 1986; 292: 1190-1 29. Langman MJ, Jensen DM, Watson DJ, et al. Adverse upper gastrointestinal effects of rofecoxib compared with NSAIDs. JAMA 1999; 282 (20): 1929-33 30. Bombardier C, Laine L, Reicin A, et al., on behalf of VIGOR Study Group. Comparison of upper gastrointestinal toxicity of rofecoxib and naproxen in patients with rheumatoid arthritis. N Engl J Med 2000; 343 (21): 1520-30 31. Silverstein FE, Faich G, Goldstein JL, et al., on behalf of The Celecoxib Long-term Arthritis Safety Study. Gastrointestinal toxicity with celecoxib vs nonsteroidal anti-inflammatory drugs for osteoarthritis and rheumatoid arthritis: the CLASS study: a randomized controlled trial. JAMA 2000; 284 (10): 1247-55 32. FDA Advisory Committee 2000. Celecoxib Medical Officer review (CLASS Study) [online]. Available from URL: http:// www.fda.gov/ohrms/dockets/ac/01/briefing/3677b1_03_ med.pdf [Accessed 2005 Feb 13] 33. Schnitzer TJ, Burmester GR Mysler E, et al. Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), reduction in ulcer complications: randomised controlled trial. Lancet 2004 Aug 21; 364: 665-74 34. Laporte J-R, Ibanez L, Vidal X, et al. Upper gastrointestinal bleeding associated with the use of NSAIDs. Drug Saf 2003; 27 (6): 411-20 35. Savage RL. Can patients stomach COX-2 inhibitors? Prescriber Update 2002; 23 (2): 26-7

Drugs Aging 2005; 22 (3)

COX-2 Inhibitors: When Should They Be Used in the Elderly?

36. Adverse Drug Reactions Advisory Committee (ADRAC). Serious gastrointestinal effects with celecoxib and rofecoxib. Aust Adv Drug Reactions Bull 2003; 22 (3): 15 37. Weaver J, Bonnel RA, Karwoski CB, et al. GI events leading to death in association with celecoxib and rofecoxib. Am J Gastroenterol 2001; 96 (12): 3449-50 38. Chan FKL, Hung LCT, Sven BY, et al. Celecoxib versus diclofenac and omeprazole in reducing the risk of recurrent ulcer bleeding in patients with arthritis. N Engl J Med 2002; 347 (26): 2104-10 39. Lewis SC, Langman MJS, Laporte J-R, et al. Dose response relationships between individual nonaspirin nonsteroidal antiinflammatory drugs (NANSAIDs) and serious upper gastrointestinal bleeding: a meta-analysis based on individual patient data. Br J Clin Pharmacol 2002; 54 (3): 320-6 40. Savage RL. NSAIAs can cause lower GIT damage. Prescriber Update 2001; 21: 25-7 41. Bonner GF. Exacerbation of inflammatory bowel disease associated with use of celecoxib. Am J Gastroenterol 2001; 96 (4): 1306-8 42. Eid RY, Quirk DM. Diaphragm-like strictures of the colon induced by the selective COX-2 inhibitor celecoxib: a case report and review of the literature. Am J Gastroenterol 2001; 96 Suppl.: S194. 43. Ahmad SR, Kortepeter C, Brinker A, et al. Renal failure associated with the use of celecoxib and rofecoxib. Drug Saf 2002; 25 (7): 537-44 44. Henao J, Hisamuddin I, Nzerue CM, et al. Celecoxib-induced acute interstitial nephritis. Am J Kidney Dis 2002; 39 (6): 1313-7 45. Rocha JL, Fernandez-Alonso J. Acute tubulointerstitial nephritis associated with the selective COX-2 enzyme inhibitor, rofecoxib. Lancet 2001; 357: 1946-7 46. Zhao SZ, Reynolds MW, Lejkowith J, et al. A comparison of renal related-adverse drug reactions between rofecoxib and celecoxib based on the World Health Organization/Uppsala Monitoring Centre safety database. Clin Ther 2001; 23 (9): 1478-91 47. Bate A, Lindquist M, Edwards IR, et al. Data derived from WHO International Drug Monitoring Database: a data mining approach for signal detection and analysis. Drug Saf 2002; 25 (6): 393-7 48. Schwartz JI, Vandormael K, Malice MP, et al. Comparison of rofecoxib, celecoxib, and naproxen on renal function in elderly subjects receiving a normal-salt diet. Clin Pharmacol Ther 2002; 72 (1): 50-61 49. Whelton A, Schulman G, Wallemark C, et al. Effects of celecoxib and naproxen on renal function in the elderly. Arch Intern Med 2000; 160 (10): 1465-70 50. Feenstra J, Heerdink ER, Diederick E, et al. Association of nonsteroidal anti-inflammatory drugs with first occurrence of heart failure and with relapsing heart failure. Arch Intern Med 2002; 162: 265-70 51. The WHO International Drug Monitoring Program. Data derived from the WHO International Drug Monitoring Database. In: Aronson JK, editor. Side effects of drugs, annual 26. Amsterdam: Elsevier Science, 2003: 548-57 52. Mamdani M, Juurlink D, Lee DS, et al. Cyclo-oxygenase-2 inhibitors versus non-selective non-steroidal anti-inflammatory drugs and congestive heart failure outcomes in elderly patients: a population-based cohort study. Lancet 2004 May 29; 363: 1751-6

© 2005 Adis Data Information BV. All rights reserved.

199

53. Whelton A, White WB, Bello AE, et al. Effects of celecoxib and rofecoxib on blood pressure and edema in patients > or = 65 years of age with systemic hypertension and osteoarthritis. Am J Cardiol 2002; 90 (9): 959-63 54. Weaver A, Alderman M, Sperling R. Blood pressure control and rates of edema following the administration of the cyclooxygenase-2 specific inhibitors celecoxib versus rofecoxib in patients with systemic hypertension and osteoarthritis. Am J Cardiol 2003; 91 (10): 1291-2 55. FDA Advisory Committee (2000). Rofecoxib statistical review (VIGOR study) [online]. Available from URL: http://www.fda.gov/ohrms/dockets/ac/01/briefing/3677b2_04_stats.pdf [Accessed 2005 Feb 13] 56. Brinker A, Goldkind L, Bonnel R, et al. Spontaneous reports of hypertension leading to hospitalisation in association with rofecoxib, celecoxib, nabumetone and oxaprozin. Drugs Aging 2004; 21 (7): 479-84 57. Farkouh M, Kirshner H, Harrington RA, et al. Comparison of lumiracoxib with naproxen and ibuprofen in the Therapeutic Arthritis Research and Gastrointestinal Event Trial (TARGET), cardiovascular outcomes: randomised controlled trial. Lancet 2004 Aug 21; 364: 675-84 58. Ray WA, Stein CM, Daugherty JR, et al. COX-2 selective nonsteroidal anti-inflammatory drugs and risk of serious coronary heart disease. Lancet 2002; 360: 1071-3 59. Bresalier RS, Sandler RS, Quan H, et al. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial. N Engl J Med 2005 Feb 15 [online]. Available from URL: http://www.nejm.org [Accessed 2005 Feb 15] 60. Graham DJ, Campen D, Hui R, et al. Risk of acute myocardial infarction and sudden cardiac death in patients treated with cyclo-oxygenase 2 selective and non-selective non-steroidal anti-inflammatory drugs: nested case-control study. Lancet 2005 Feb 5; 365 (9458): 475-81 61. Mamdani M, Rochon P, Juurlink DN, et al. Effect of selective cyclooxygenase-2 inhibitors and naproxen on short term risk of acute myocardial infarction in the elderly. Arch Intern Med 2003; 163: 481-6 62. Solomon DH, Schneeweiss S, Glynn RJ, et al. Relationship between selective cyclooxygenase-2 inhibitors and acute myocardial infarction in older adults. Circulation 2004; 109: 2068-73 63. Solomon SD, McMurray JJ, Pfeffer MA, et al. Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med. 2005 Feb 15 [online]. Available from URL: http://content.nejm.org/cgi/content/abstract/NEJMoa050405v1 [Accessed 2005 Feb 15] 64. Furberg CD, Psaty BM, FitzGerald GA. Parecoxib, valdecoxib, and cardiovascular risk. Circulation 2005; 111 (3): 249 65. Clark DWJ, Ghose K. Neuropsychiatric reactions to nonsteroidal anti-inflammatory drugs (NSAIDs). Drug Saf 1992; 7 (6): 460-8 66. Jiang H-K, Chang D-M. Non-steroidal anti-inflammatory drugs with adverse psychiatric reactions: five case reports. Clin Rheumatol 1999; 18: 339-45 67. Coulter DM. Acute psychiatric reactions with COX-2 inhibitors. Prescriber Update 2002; 23 (2): 21 68. Coulter DM, Clark DWJ, Savage RL. Celecoxib, rofecoxib and acute temporary visual impairment. BMJ 2003; 327: 1214-5 69. Garnett WR. Clinical Implications of drug interactions with the coxibs. Pharmacotherapy 2001; 21 (10): 1223-32

Drugs Aging 2005; 22 (3)

200

70. Thomas M, on behalf of ADRAC. Diuretics, ACE inhibitors and NSAIDs: the triple whammy. Med J Aust 2000; 172: 184-5 71. Boyd IW, Mathew TH, Thomas MC. COX-2 inhibitors and renal failure: the triple whammy revisited [published erratum appears in Med J Aust 2000; 173: 504]. Med J Aust 2000; 173: 274 72. Savage RL. A dangerous trio. Prescriber Update 2002; 23 (2): 20 73. Ouellet M, Riendeau D, Percival MD. A high level of cyclooxygenaase-2 inhibitor selectivity is associated with a reduced interference of platelet cyclooxygenase-1 inactivation by aspirin. Proc Natl Acad Sci U S A 2001; 98 (25): 14583-5 74. Catella-Lawson F, Reilly MP, Kapoor SC, et al. Cyclooxygenase inhibitors and the antiplatelet effects of aspirin. N Engl J Med 2001; 345: 1809-17 75. Kurth T, Glynn RJ, Walker AM, et al. Inhibition of clinical benefits of aspirin on first myocardial infarction by nonsteroidal antiinflammatory drugs. Circulation 2003; 108 (10): 1191-5 76. Curtis JP, Wang Y, Portnay EL, et al. Aspirin, ibuprofen and mortality after myocardial infarction: retrospective cohort study. BMJ 2003; 327 (7427): 1322-3 77. Adverse Drug Reactions Advisory Committee (ADRAC). Interaction of celecoxib and warfarin. Adv Drug React Bull 2001; 20 (1): 2 78. Medsafe Editorial Team. Interaction between COX-2 inhibitors and warfarin. Prescriber Update 2001; 22: 16-8

© 2005 Adis Data Information BV. All rights reserved.

Savage

79. Adverse Drug Reactions Advisory Committee (ADRAC). Interaction of rofecoxib and warfarin. Adv Drug React Bull 2002; 21 (1): 3 80. Slørdal L, Samstad S, Bathen J, et al. A life-threatening interaction between lithium and celecoxib. Br J Clin Pharmacol 2003 Apr; 55: 413-4 81. Collantes-Estevez E, Fernandez-Perez C. Improved control of osteoarthritis pain and self-reported health status in non-responders to celecoxib switched to rofecoxib: results of PAVIA, an open-label post-marketing survey in Spain. Curr Med Res Opin 2003; 19 (5): 402-10 82. Coulter DM. COX-2 inhibitor use in New Zealand: a problem of age and death. In: Davis EA, Sobey CG, editors. Proceedings of the Australasian Society of Clinical and Experimental Pharmacologists and Toxicologists; 2003 Nov 30-Dec 3; Sydney. Melbourne (VIC): ASCEPT, 2003: 10, 110

Correspondence and offprints: Dr Ruth Savage, New Zealand Pharmacovigilence Centre, Department of Preventive and Social Medicine, University of Otago, PO Box 913, Dunedin, New Zealand. E-mail: [email protected]

Drugs Aging 2005; 22 (3)