Sock's Toxicity And Drugs In RA:
NSAIDs
Home | Strategies | Life | Summary | Case Histories | Coping With Pain | Side Effects | Herbs | Supplements | Symptoms | Musculoskeletal | Disease Process | Chronic Pain | Management | Effects | Arthritis | Updates | Research | Alerts | Enbrel | Remicade | Arava | Surgery | Lab Tests | Treatment | Clinical Trials | Physical Care. | NSAIDs | Joint Injection | DMARDs | Factors | Steriods

The major effect of NSAIDs is to reduce acute inflammation thereby decreasing pain,and improving function. They help relieve the symptoms of RA but do not slow or arrest the progression of the disease. All of these types of  drugs have mild to moderate analgestic properties independent of their anti-inflammatory effect.
 
The older NSAIDs inhibit prostaglandin synthesis by blocking cyclooxygenase enzyme, Cox-1 (which normally protects the stomach).
 
 The newer NSAIDs are designed to selectively block the Cox-2 enzyme which is unregulated in inflammatory states but preserves Cox-1 activity, promise  to provide safer alternatives (less gastrointestinal side effects). Prostaglandins are important mediators of both pain and inflammation.

Aspirin is the oldest drug of the non-steroidal class. However,because of  its higher rate of GI toxicity,its narrow relationship between toxic and anti-inflammatory serum levels,and the inconvenience of mutiple daily doses, aspirin's use as the initial choice of drug therapy has largely been replaced by the newer NSAIDs.
 
There are now a large number of NSAIDs from which to choose,and at full doseage are potentially equally effective. Likewise, the toxicities of the currently available NSAIDs are similar. However, there is a great deal of variation in tolerance and response to a particular NSAID. Long acting NSAIDs that allow daily or twice daily dosing will improve compliance and help with morning stiffness.
 
If there is active inflammation,a full doseage of a NSAID should be prescribed. A lower doseage can be used if inflammation is mild,if mechanical pain is the major problem,if the patient is elderly or if the patient suffers from conditions that increase the risk for toxicity.
 
If a particular drug is ineffective after a t least 4 week trial or is not tolerated then another NSAID can be initiated. Some  papers on RA treatment claim that no one NSAID has been proven to be better than another. As a patient I disagree.
 
Nonsteriodal anti-inflammatory drugs are all different;they have different formulas. Therefore,their actions in different patients depend on the disease and the dose. They are also different in terms of effectiveness,side effects and ease of use.
 
As a patient of rheumatoid arthritis these factors are important considerations in choosing a NSAID. Most patients will react differently to individual NSAIDs. Sometimes,over time,they will lose their effectiveness and another NSAID should be implemented.
 
The most common toxicity of NSAIDs is gastrointestinal problems. Because prostagladins play a role in the regulation of renal blood flow and maintenance of glomerular filteration,NSAIDs can impair renal function in certain patients.
 
The patients at highest risk are those with fluid imbalances or with compromised renal function (e.g., heart failure,dieuric use, cirrhosis, dehydration, and renal insufficiency).
 
Patients taking the older NSAIDs with stomach problems should consider a stomach protectant such as misoprostol (Cytotec). Taking more than one NSAID at a time increases the possibility of side effects,particularily stomach problems such as ulcers and (GI) bleeding. At over the-the-counter,NSAIDs are among the best safer antidotes for arthritis pain.
 
At prescription doses,they are effective fighters of both pain snd inflammation. But this relief may come at a cost--stomach problems. NSAIDs reduce pain when taken at low dose,and relieve inflammation when taken at a higher dose. Patients should not take doses higher than prescribed in the mistaken belief that pain will be relieved. Always follow prescribed directions to avoid unnecessary side effects.
 
Cox-2 inhibitors--Celecoxib (Celebrex) and Rofecoxib (Vioxx) are new NSAID drugs known as Cox-2 inhibitors. Regular NSAIDs block two prostaglandin-producing enzymes called cyclooxygenase 1 and 2. (Cox-1 and Cox-2) The new drugs block Cox-2 which is resposible for most inflammatory effects,but not Cox-1 which normally protect the stomach.
 
Some research has shown that they may not provide protection against heart disease as a traditional NSAID might,ask the physician about the advisability of using it along with low-dose aspirin or other heart-protective medication, particularily if one have heart disease or risk factors for it. Generally more than one NSAID is not recommended at the same time as the manufacturer's labels state.
 
Upper Gwynedd,PA--8/22/01--In response to an article published in this week's Journal of the American Medical Association containing data from selected previously released studies of Vioxx  Tm (Rofecoxib) and Celebrex, Merck & Co.,Inc. said the Company stands behind the overall and cardiovascular safety profile and the favourable gastrointestinal (GI) profile of Vioxx.
 
The authors say more data are needed on the cardiovascular profile of Cox-2 inhibitors. However,Merck believes that extensive cardiovascular data already exists exist on Vioxx and that these data--which were not incorporated into the authors'analysis--suggest there is no increase in the risk of cardiovascular events as a result of treatment with Vioxx.
 
These cardiovascular data are from a meta-analysis of data from 19 controlled clinical trials with Vioxx involving more than 28,000 patients and showed the relative risks of serious cardiovascular events were similar with Vioxx and placebo,and with Vioxx and the widely orescribed NSAID's ibuprofen, diclofenac and nabumetone. This analysis were presented in February at the FDA Advisory Committee meeting and in Europe in June at the annual meeting of the EAR.
 
The authors of the JAMA article conced that their analysis "has several significant limitations". These limitations included:- The analysis looked at 3 of the many available studies of Vioxx and 1 study of Celebrex-The authors borrowed a "placebo" arm for their analysis from 4 older,separate aspirin studies that did not involve Vioxx or Celebrex,and compared data from that "placebo" group to data from separate studies of Vioxx and Celebrex. It is widely understood that an essential component in determing the validity of such a analysis is the similarity of the studies being compared.
 
-Specifically,none of the aspirin studies used by the authors to estimate the "placebo" rate of cardivascular events included patients with RA. The authors acknowledge this limitation in their article because RA is a disease that may increase one's risk of cardiovascular events.
 
Therefore,to compare a group of patients taking Vioxx for RA with a "placebo"  group of patients who do not have RA is inappropiate and misleading because one would expect a higher rate of cardiovascular events in the RA group, regardless of medications.
 
In fact, two of the aspirin studies cited by the authors involved healthy male physicians (physicians' Health Study and Doctors' Study) who likely had lower risk for cardiovascular events than those studied in the Cox-2 inhibitor studies cited in the article.
 
Based on the authors' analysis and the data from Merck's large-scale placebo-contolled clinical trials,Merk does not believe the author's conclusions in the article are scientifically supported by the totality of the data available.
 
In rare cases,serious problems,such as bleeding,can occur without warning symptoms. People who have had an allergic reaction such as asthma,to Vioxx,aspirin,or other arthritis medicines should not take Vioxx.
 
People who have had liver or kidney problems,or are pregnant, should tell their doctors. Also,Vioxx should not be used by women in late pregnency . Vioxx is not a substitute for aspirin for cardiovascular prophylaxix. Vioxx does not interfeare with the effects of low-dose aspirin on platelets.
 
Common side effects reported in clinical trials with Vioxx were upper-respiratory  infection,diarrhea,nausea,and high blood pressure. More than 40 million prescriptions have been written for Vioxx in the U.S. since its introduction.
 
"Our patients panicked--needlessly,"says Dr.Claire Bombardier a research-rheumatoligist at Mount Sinai Hospital in Toronto. "The JAMA analysis was flawed because it compared apples and oranges". "A more rigorous analysis -study showed that only group of coxib users who had a higher rate of heart attacks were those at high risk of heart problems to begin with".
 
Even with people at high risk of heart disease,there were no direct evidence that coxibs actually harmed the heart.according to Dr. Bombardier.
 
San Francisco,CA--11/15/01--For patients who require NSAID therapy,the cyclo-oxygenase 2 (Cox-2) inhibitors,celecoxib and rofecoxib,provide significantly enhanced upper gastrointestinal (GI) safety compared to traditioal NSAIDs.
 
In a presentation of clinical results on the GI tolerability of Cox-1 sparing agents,Jay Goldstein,MD,Department of Medicine,Vice-head of Clinical Affairs,University of Illionois at Chicago,highlighted recent data on celecoxib and rofecoxib,and on valdecoxib.
 
The CLASS (Celecoxib,Long-term Arthritis Safety Study) and VIGOR (Vioxx Gastrointestinal Outcomes Research) trials compared the incedence of clinically significant ulcers and ulcer complications associated with celecoxib 800mg/day (which is 2 to 4 times the recommended doses for OA and RA and that of ibuprofen 2,400mg/day or diclofenac 150mg/day over a minimum six-month treatment period.
 
Six month data showed that celecoxib was associated with a significantly lower annualized incidence of UGII complications--perforations,obstructions,and bleeding--and symptomatic ulcers,when compared to the NSAIDs used in the study. 2.08 vs 3.54 events per 100 patient-years of exposure,respectively,(p=0.02). The long-term data showed similar data.
 
Results from the VIGOR trial (Vioxx Gastrointestinal Outcomes Research) also showed a significant reduction in the rate of complicated confirmed events,defined as perforations,obstruction and severe UGI bleding,among patients receiving rofecoxib as compared with those receiving naproxen (0.6 and 1.4 events per 100 patients-years of exposure,respectively,p==0.005).
 
Dr. Goldstein also introduced data from the recently completed SUCCESS-1(Successive Celecoxib Efficacy and Safety Study). The 12-week, international, double-blind trial enrolled 13,274 patients with OA who were randomized to 200 or 400 mg/day of celecoxib,1000mg/day of naproxen,or 100mg/day of diclofenac.
 
The data show that the annualized rate of ulcer complictions was significant (p<0.05)lower in the celecoxib arm compared to the NSAID arm (0.01 vs ).8 ulcer complications per 100 patient-years of exposure, respectively).
 
Futuremore, the annualized rate of combined ulcer complications and symptomatic ulcers were significantly (p<0.05) lower in the celecoxib arm compared with the NSAID arm (1.0 vs 2.1 events per 100 patient-yers of exposure,respectively).
 
Dr.Goldstein cited data on the incidence of GI ulcers seen in patients with the newer Cox-2 inhibitor,valdecoxib. Results come from a 3-month trial in patients with OA,in which valdecoxib was compared to ibuprofen and diclofenac. Data show that 10mg/day and 20mg/day of valdecoxib achieved a statiscally significant lower incidence of GI ulcers (-<0.05) than ibuprofen 800mg 3 times daily,and diclofenac 75mg/ twice daily.
 
Futhermore,data from the 14 week trial of valdecoxib 20mg twice daily and 40mg twice daily in patients with OA or RA show a significantly statistically lower incidence of GI ulcers (p<0.05) compared to naproxen 500mg twice daily.
 
The FDA recently approved a supplemental application for the use of Vioxx (rofecoxib) for RA adding the indication to the previously approved indications for osteoarthritis and pain. The FDA has also approved new label text and precautions that are based on the results of the Vioxx Gastrointestinal Outcomes Research (VIGOR).---4/14/2002.
 
The new labeling information approved by FDA will advise doctors to use caution in prescribing Vioxx for patients with ischemic heart disease and notes that Vioxx 50 mg is not recommended for chronic use. The geniatric section of the label will reinforce information in the existing standard warning section of all NSAIDs indicating that the elderly are at higher risk of serious GI and renal events such as GI bleeding and acute renal failure.
 
Taken overall,these data strongly indicate that Cox-2 inhibitors are clinically superior to traditional NSAIDs with regard to UGI tolerability.
 
There has recently been concerns about the question of whether coxibs impair kidney function. The concern arises because the cyclooxygenase-2 enzyme, which coxibs inhibit,plays a role in maintaining blood flow in the kidney. But the concern is out of proportion to the reality says Dr. Hani El-Gabalawy, director of the Arthritis Centre at the University of Manitoba Health Sciences in Winnipeg. 
 
"Its important to understand that cyclooxygenase-2 kicks in primarily under stress conditions;e.g,when kidney function is already impaired," he says. "It stands to reason that coxibs should be avoided in such people,who need all the cyclooxygenase they can produce to keep the blood flowing through their kidneys.
 
But there's no reason for people with normal kidney's to avoid these drugs. By the same token,the finding that coxibs (as well as conventional NSAIDs) raise blood pressure by four to five units need not concern people whose blood pressure is in the normal range. It's only people with hypertension who risk losing control over their blood pressure by taking coxibs or other NSAIDs,according to Dr. El-Gabalawy.
 
He says that individuals with high blood pressure should use conventional NSAIDs with caution and only under a doctor's supervision,if at all. And if your overall risk of heart disease is especially high,he says "you'd be prudent to avoid all NSAIDs,especially the coxibs".
 
On the other hand,Dr. El-Gabalawy says he would recommend coxibs to people whose GI risk outweighs their cardiac fisk;e.g., an elderly person who has a four percent chance of developing a life-threatening GI bleed. It's also safe to take coxibs concurrently with A.S.A., meaning you can protect your heart and GI tract at the same time.
 
What about that old standby acetaminophen? "Preliminary results of a major new study have shown that both coxibs offer significantly better pain relief than high-dose acetaminophen for osteoarthritis patients. (ACR recommends the drug for osteoarthritis pain relief.)

VIOXX

CELEBREX

Anagesics:
 
Many patients with early RA take over-the-counter medications,such  as ibuprofen or acetaminophen,for a period of time before seeking medical attention. Since the clinical picture at this stage may still be vague,the primary care physician may simply prescribe a NSAID and recommend continued follow-up. Pain caused by inflammation is best treated with NSAIDs,the occasional addition of acetaminophen can be helpful.Chronic narcotic therapy should be avoided due to side effects such as diminished mental state, constipation etc.Dependency and addiction occur infrequently.but the clinician must be alert to these behaviour patterns.

Chaddsford,PA--11/26/01--Endo Pharmaceuticals Holdings Inc. announced the FDA's approval of the abbreviated new drug application to market Percocet (TM) 7.5/325 and 10/325 oxycodone/acetaminophen tablets that was filed by Endo Pharmaceuticals Inc., a wholly owned subsidiary of Endo Pharmaceuticals Holdings.
 
Percocet,the leading brand of oxycodone/acetaminophen on the market,is indicated for the treatment of moderate to moderately severe pain. The new reformulated Percocet 7.5/325 and 10/325 strengths provide pain relief with reduced acetaminophen content as compared with previous formulations of the pain-relieving drug.
 
"With this formulation,physicians can still take advantage of the synergistic action of the two analgesics combined,but without worrying about exceeding the daily limit of acetaminophen," said Nathaniel Katz,MD,Assistant Professor of Anesthia at Harvard Medical School. "This will really be helpful to patients who may need higher doses of oxycodone to relieve their pain".
 
The new formulated strengths,which are expected to be available in pharmacies  within the next week to ten days,will potentially enable patients to take their Percocet on a simpler dosing schedule,allowing them to take fewer tablets per day,and less frequently then the original 5 mg tablet. And due to the lower Acetaminophen levels,these new strengths may allow physicians to prescribe  Percocet 7.5/325 and 10/325 for more long-term use than previously acceptable.
 
"The new strengths continue to deliver the dual benefits of oxycodone and acetaminophen while lowering the risk of excessive acetaminophen exposure" .said Lori  Reisner,Pharm. D. and Associate Clinical Professor of Pharmacy at U. of Cal. in San Francisco. "The ever-increasing number of acetaminophen-containing products and other over-the-counter pain relievers requires pharmacists to be deligent in reminding physicians and patients about the appropriate level of usage."
 
The most frequently reported side effects of Percocet include lightheadedness, dizziness,sedation,nausea and vomiting. Oxycodone can produce drug dependence and has the potential for being abused. Physicians are reminded that the total daily dose of acetaminophen must be carefully considered in the use of any combination product.

For the individual patient one NSAID often appears to be best. To find one,sometimes,a sequence of NSAIDs is prescribed for relatively short trial periods of about 7 to 10 days each. By the time a couple or sometimes,more have been tried,the doctor has a pretty good idea of how much the patient is going to be helped by an NSAID,and which one is best. Trying more would waste time and money.
 
Ideally,the NSAID chosen will be cheap,well tolerated,easy to remember to take and effective. Few drugs meet all these criteria. The decision is also influenced by whether or not a patient has a drug plan to help with the cost,because some are expensive.
 
You can bleed on any of the traditional NSAIDs,(even without symptoms). particularily if;you are older (the risk increases with age),you have had a ulcer before,you are taking high doses of one NSAID,or combing,which should not be done,sometimes,if  you are taking a corticosteriod such as predisone (even some DMARDs-in different people ),you are on blood-thinning drugs and if you havea serious medical condition and,althought this is less certain,if you smoke cigarettes or drink alcohol. Your doctor is the expert,and he has to assess each individual situation,not you.
 
Choosing the right NSAID depends on a number of factors. Some are clearly more risky then others. People with a previous ulcer should probably be prescribed either ulcer protection (with misoprostal combined with an NSAID in a single pill or omeprazole added to an NSAID0 or one of the newer COX-2 specific agents. Older people and those on anticoagulant (blood thinners) should be individually considered And people with borderline kidney function should probably not take any type of NSAID.
 
In the old days,one consideration that may have been exaggerated is conveinence. Some NSAIDs need to be taken 4 times daily,others twice daily and a small number once daily. For most patients this is not important,but for others,especially if they're likely to forget a midday dose at work,it very definitely is.
 
The choice of the NSAID tried,and the order in which they are tried,is influenced by their safety profiles. If one don't seem to be working or a drug that once worked has lost its efficacy,let your doctor know.
 
In RA common practise is to find the best NSAID and then decide on a second-line drug to be added. It is seldom that the disease is so mild (non-erosive) as to be controlled by NSAIDs alone.
 
In OA there may be an inflammatory component that is helped by an NSAID,but the chief benefit seems to be pain control, Acetaminophen is often just as good as an NSAID,and safer and cheaper.
 
In some diseases NSAIDs ae very effective,in crystal arthritis,such as gout,NSAID response is terrific-unless treatment is started too late,or the dose is inadequate. In ankylosing spondylitis,an NSAID usually results in a patient's first good night of sleep in years.
 
Side effects of the older NSAIDs;rare side effects-skin rash,including photosensitivity (a rash provoked by going out in the sun) and hives,jaundice or other evidence of liver problems,bowel ulceration and bleeding,impaired production of blood cells (aplastic anemia ),asthma,significant kidney impairement,confusion,headaches and neck stiffness--VERY RARE SIDE EFFECTS.
 
COMMON SIDE EFFECTS are indigestion,stomach ulceration and bleeding
 
Some of the rarer side effects are linked to specific NSAIDs (such as alastic anemis with phenylbutazone),while others may be seen with any NSAID. For example,if an asthma attack is provoked by one NSAID,all (including ASA) should be avoided. In special circumstances,if the NSAID is badly needed,the patient can be desesitized.
 
All NSAIDs also appear to affect kidney blood flow. This is not a problem for most people,but its a big problem for (some) older people,people with kidney function that is alredy below normal people on some other drugs (such as dieuretics,given to rid of excess body fluid or to treat high blood pressure)
 
The common side effects,those affecting the stomach,are the real problem NSAID indigestion is common,about 10 to 20 % of users will experience it at one time or another. But it's an unreliable marker for an ulcer,which can bleed without any warning signs. And many with indigestion don't have ulcers.
 
Taking an NSAID increases the risk of developing an ulcer,having gastrointestinal bleeding or experiencing an ulcer perforation by about five times. The risk of a bleeding ulcer in an other wise healthy person taking one of the older NSAIDs has been estimated at 1 to 2 for every 100 patient years. In plain English,if 100 patients take the NSAID for a year,1 or 2 of then will develop an ulcer in that year
 
Only about half of ulcers will cause symptoms. Very often the first sign of an ulcer is evidence of bleeding,such as vomiting of blood,black bowel movements or rarely,light-headedness and fainting.
 
There are a number of ulcer-protective drugs available,but only misoprostol and omeprazole have been convincingly shown to reduce the risk
 
For indigestion alone,anti-ulcer drugs can be tried. They may or may not clear up the symptoms. Since the cause of non-ulcer indigestion isn't knwn (the true cause of back-pain is unknown in many cases ).it can be difficult to deal with,at times. Sometimes it's so bad the best solution is simply to stop the NSAID,but when that time approaches,your physician will know.
 
If one worries about everything,one won't take any medication for any medical purposes. The physician will weigh the risk and benefit ratio.
According  to my rheumatologist;taking an NSAID on an "as needed" basis (a pill or two today,then a few days skipped) is foolish. NSAIDs need several days to develop the ir full effect,and little can be expected from occasional doses except possibility of a heartburn or an ulcer. The ulcer doctors see much more ulcer bleeding in occasional NSAID users than they do in patients with arthritis, It's surprising how many people take ASA from stomach aches or insominia.
 
Most pharmacists and doctors recommend that an NSAID be taken with food. It's doubtful this really helps prevent an ulcer,but meals do help as reminders.
 
NSAIDs come in suppository forms as well as tablets and capsules,but there's no real evidence that supporitories cause fewer ulcers.
 
NSAIDs may interact with other drugs (this applies to all forms of drugs,herbs included),and can cause problems with blood-thinning agents,serious bleeding may result,with lithium,toxic blood levels of lithium can occur with a variety of blood pressure medications,high blood pressure may worsen Sulindac may possibly be the safest NSAID in this setting.
 
There is nothing magic about an NSAID If you don't think it's it's helping,and your disease control has improved to the point of little pain,stop taking it. If symptoms of pain or pain or even increased joint swelling reappear,you need it,and can restart it. Your doctor will determine this phase. You may have to wait a few days to feel its full effect,but if it was helping before,and nothing else has changed,it will work again.
 
From my personal experience; It was a question whether NSAID was helping me. My RA was so poor at that point that No NSAID seemed to help,this also applied to DMARD and predisone. It was not until later in my personal disese course that I relized the benefit of these medications. Simply,put the disease was so rampart that medication did not SEEM to help.
 
Another point;The decision to take Vioxx and Celebrex (in my case ) I tried Vioxx first,it seemed to work. Later Celebrex was suggested, (insurance company-at that time Celebrex was cheaper-pennies). I tried Celebrex. The problem with RA is that it waxes and wanes and sometimes,you think one NSAID is better,but in reality( in my case),the long term result is that there was no "real" difference. This has happened with other RA medications. Sometimes,we will be fooled and more is likely,then less. You may not have a choice with some insurance company health plans.
 

Heart researchers from the Cleveland Clinic Foundation say that the popular arthritis drugs Vioxx and Celebrex may increase the risk for heart attacks and are cautioning doctors to use the drugs very carefully, especially in people with heart disease.

But the makers of the drugs say that the researchers are misinterpreting data and are ignoring many favorable studies of the drugs.

Steven Nissen, MD, says an analysis of existing medical research of the drugs suggests that the people taking them are twice as likely to have heart attacks as people who take older anti-inflammatory drugs to treat arthritis. But Nissen, who is vice chairman of cardiology at the Cleveland Clinic, adds that even with this increased risk, the number of heart attacks is still very low.

The study is reported in the Aug. 22 issue of The Journal of the American Medical Association.

Celebrex and Vioxx are drugs that fight pain and inflammation in the body and are used for arthritis as well as other medical problems such as severe menstrual cramps. These drugs are called Cox-2 inhibitors and are newer versions of older painkillers, such as ibuprofen and naproxen. The newer medications are believed to cause less stomach problems like ulcers and bleeding, so many doctors have been giving their patients Cox-2 inhibitors over the older drugs.

Celebrex and Vioxx have been available since 1999, and with sales expected to top $6 billion this year, both drugs are big moneymakers for their manufacturers. So it is not surprising that these drug companies staunchly defended the drugs.

Steve Geis, MD, PhD, group vice president for clinical research at Pharmacia, the maker of Celebrex, says the study is flawed because Nissen and his colleagues were comparing apples and oranges. He says a better and more accurate approach would be to compare people who take Celebrex to people who are not taking aspirin. Using that method, says Geis, heart attacks would actually be less common in people taking Celebrex.

Laura Demopoulos, MD, senior director of cardiovascular clinical research at Merck, the maker of Vioxx, said that the researchers overlooked several studies that showed no increased chance of having a heart attack with Vioxx. She says 19 previous studies have shown that Vioxx is no more likely to cause a heart attack than traditional anti-inflammatory drugs or placebo.

In the current study, Nissen and his colleagues compared Vioxx to the traditional arthritis drug naproxen, which works in a way similar to aspirin. People on Vioxx were twice as likely to have a heart attack than patients on naproxen, says Nissen. But among the 8,000 people in the study, only 161 patients had heart attacks and 70% of them were taking Vioxx.

The second study compared Celebrex with ibuprofen and diclofenac, an older arthritis medicine sold under the names Voltaren and Cataflam. The scientists did not find that people on Celebrex were more likely to have a heart attack. However, the people in the study were allowed to take aspirin, which is known to protect against heart attacks. This made the results harder to interpret.

Nissen then compared Vioxx and Celebrex users to placebo groups in four large studies. Again, the researchers found that the people on the Cox-2 inhibitors were more likely to have a heart attack than people taking placebo, says Nissen.

Valentin Fuster, MD, former president of the American Heart Association, said  that this study was done to test the side effects of these drugs on the stomach and intestines and not to look at the drugs' effects on the heart. So experts really can't say for sure that these drugs really have any bad effects on the heart based on this study, he says.

But even though the study "is not perfect, the findings cannot be thrown into the wastebasket," says Fuster. Because so many people use the drugs -- more than 2 million prescriptions will be written this year -- he warned, "we had better pay attention to this observation."

Fuster adds that the drugs have made a huge impact on people's lives, often allowing them to become much more active after years of living with painful arthritis. And he says that among his patients, he is already using the drugs with caution because they can cause swelling and increase blood pressure in some people.