Healthy Skepticism Library item: 12555

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

Publication type: Journal Article

Miselli M.
Nimesulide : waiting for Godot
Informazioni sui Farmaci 2007; 4:

Full text:

In Italy, nimesulide, on the market since 1985, is now available in the form of 48 products; it is the most used anti-inflammatory drug. Last year, more than 3.5 million patients took the drug with prescriptions charged to the SSN (national social health organization), but the number of private users is even higher1.
Nimesulide recently returned to the attention of the media for a relapse of a presumed “crime”, hepatotoxicity. The story began on March 8th, 2002, when the Finnish health authority suspended the marketing of the drug because of the increase in reports of liver reactions related to its use – 66 cases of liver damage with two liver transplants and one death.
In May of the same year, Spain also withdrew nimesulide from the market. As praxi, when one member-state of the European Union revokes, suspends or modifies the marketing authorization of a drug for safety reasons, the procedure of referral is activated in which the scientific committee of the European agency of medicines (EMEA), the CHMP, is asked to express an opinion on the necessity of implementing, or not, regulatory actions in all of Europe. Italy, the country with the highest consumption, was given the task, along with Finland, to prepare a complete evaluation report on the safety profile of the drug2. At the conclusion of the evaluation in December, 2003, the CHMP favourably judged the benefit/risk ratio of nimesulide, limiting the therapeutic use to the treatment of acute pain, osteoarthritis and dysmenorrhea, limiting the maximum daily dose to 200 mg and adding the contraindication of use by patients with hepatic problems.
In May of this year, the case was reopened. The Irish minister of health, following 6 cases of serious hepatic insufficiency requiring liver transplants, and following a comprehensive review of the 53 national reports of adverse hepatic reactions, decided to suspend marketing3. For the second time in a period of little more than 3 years, the EMEA was called to review the hepatic safety of nimesulide and express an opinion on the decision to keep the drug on the market or not. In a press release, on September 21st, 2007, the EMEA reported the conclusion that “the benefits still outweigh the risks, but it is necessary to limit the duration of treatment and restrict the use to assure that the risk of developing hepatic problems is minimal”4. The maximum duration of treatment was set to 15 days; the companies were exhorted to remove packages containing more than 30 doses from the market and doctors to carefully evaluate the total risks for each patient. Therefore, the higher hepatotoxicity of nimesulide compared to other non-steroidal anti-inflammatory drugs was confirmed, but it is believed that with further limitations, the drug can be used without worry provided that “doctors and patients are aware of the possibility of occurrence of hepatic problems”5.
In the press release, the data is used in a questionable way to justify the decision, and regulatory solutions are stubbornly pursued (the modification of the drug leaflet, use limitations, added contraindications) that, in analogous situations in the past, proved to be insufficient in averting the risks of prescriptive malpractice.
It can be taken for granted that the benefits of a drug must outweigh the risks; the addition of the term “still” only indicates that the gap between the benefits and the risks has narrowed and does not help the comprehension on how safe nimesulide is. The benefit/risk ratio of a drug must be the more tilted in favour of the benefits, as less significant its therapeutic role is. If one is prepared to accept a lower safety profile for efficacious drugs in important, life-threatening diseases, similar acceptance cannot be granted to drugs destined to less important use, moreover, with numerous other alternatives available, as is the case for nimesulide.
In the document that accompanied the EMEA press release, the following is written: “the limitation of use to a maximum of 15 days” was dictated by the fact that “most of the hepatic effects arise after two weeks of treatment”, and “it is acknowledged that the mechanism that determines the hepatic effects of nimesulide is unknown, and this makes it difficult to predict if a particular patient who takes the drug could be at risk of developing hepatic reactions”5.
The hepatotoxicity of nimesulide is placed in an area of uncertainty of the physiopathological mechanism, which leaves one perplexed.
The Italian Drug Agency, in a press release issued at the end of August, referring to the reports of serious hepatotoxicity related to another NSAID (not yet marketed in Italy), lumiracoxib, states that “hepatotoxicity is a class effect of non-steroidal anti-inflammatory drugs and of the coxibs, even if there is some evidence of a difference of risk among the various active ingredients (as also indicated by the recent attention to nimesulide)”6.
But this case cannot be treated as a “class effect” because this terminology refers to an undesired side-effect (e.g., gastric ulcer, renal insufficiency, bronchoconstriction) bound to the mechanism of action (in this case, inhibition of cyclooxygenases), thus characteristic, even if in greater or lesser measure, of all of the NSAIDs. The “class effects” of the NSAIDs are the most frequent, their principle risk factors are known (high dosage, length of treatment, advanced age), and they can be prevented in part (complicated ulcers, haemorrhages, perforations) or completely (renal failure). The other side effects of the NSAIDs (e.g., haematic dyscrasia, serious skin reactions, liver damage) are, instead, of idiosyncratic type and, as such, unpredictable and not preventable.
In the past, 5 NSAIDs were taken off the market for problems concerning hepatotoxicity.
In recognising the unpredictability of the hepatotoxic effect in the individual patient, one acknowledges implicitly its idiosyncratic nature and the uselessness of a time limitation in the use of the drug. What data did the EMEA consult to exclude a possible idiosyncratic effect? Not the Italian ones which demonstrate that in most cases of hepatotoxicity, taking nimesulide was of short duration (from 1 to 8 days)1, such to “suggest the hypothesis of an idiosyncratic reacton”1. Between January 1st, 2001, and May 17th, 2007, 738 suspected adverse reactions to nimesulide were reported to the national system of drug supervision. Most reports were of young people (between 19 and 60 years old), mostly women, of whom 360 seriously affected (48.8%) and 19 fatally (2.6%)1. Of the 102 reports regarding hepatobiliary reactions (acute hepatitis, hepatitis, cholestatic hepatitis, fulminant hepatitis), 9 were fatal, 70 serious1.
The Italian data, instead, were useful to the EMEA to simulate the impact of suspending nimesulide on the number of side effects and to affirm that “the suspension could lead to a decrease in the number of hospitalisations related to hepatic problems, but it could also cause an increase in the number of hospitalisations due to gastrointestinal effects associated with the use of other NSAIDs5”. The data do not seem, however, to justify this prediction since also for nimesulide in 2006, a non-negligible number of gastrointestinal effects were reported: 20.1% compared to 28.2% for diclofenac, 17.9% for ketoprofen and 17.1% for ibuprofen1. Since the rate of reports of serious adverse effects per million DDD (defined daily dose?) was 0.16 for nimesulide, 0.30 for diclofenac, 0.23 for ketoprofen and 0.16 for ibuprofen (the 3 NSAIDs most prescribed after nimesulide), should we think that the other gastrointestinal effects of the other NSAIDs were less serious? Or has the EMEA considered only the data regarding ketorolac (a purely Italian case) for which the number of reports of serious adverse effects per million DDD is 7 times higher than the average of the other NSAIDs (1.02), more than one third of which can be correlated to its gastrointestinal effects. Ketorolac is not included, however, among the 10 NSAIDs most prescribed1 and would absorb only a minimal part of the prescription of nimesulide.
Finally, it is questionable to put a gastrointestinal reaction and hepatotoxicity on the same level, even if the impact and seriousness were equal, considering the different associated clinical outcome.
In summary, a drug remains on the market that we know to be dangerous, one that can be substituted with others that are equally efficacious and safer, provided that the patient is informed and the duration of treatment is limited. It is very clear how reasons different from more rational ones, namely caution and protection of the patient, prevailed in this decision.

Bibliography

1. Nimesulide ed epatotossicità. AIFA. BIF 2007; XIV:112-6.
2. Traversa G et al. Cohort study of hepatotoxicity associated with nimesulide and other non-steroidal anti-inflammatory drugs. BMJ 2003; 327:18-22.
3. Press Release from the Irish Medicines Board, 15 May 2007. www.imb.ie
4. EMEA. Press Release. Doc. Ref. EMEA/432604/2007. London, 21 September 2007.
5. EMEA. Questions and answers on the CHMP recommendation on nimesulide-containing medicines. Doc. Ref. EMEA/430988/2007. London, 21 September 2007.
6. Lumiracoxub: informazioni relative alla sicurezza d’uso. Comunicato AIFA del 23/08/2007.