February 28th, 2012

FDA Revises the Safety Labeling of Statins

The FDA today announced important new changes to the safety language on the labels of statins:

Routine periodic monitoring of liver enzymes is no longer recommended. Serious liver injury associated with statins is “rare and unpredictable in individual patients” and “routine periodic monitoring of liver enzymes does not appear to be effective in detecting or preventing this rare side effect,” according to the FDA.  The agency now says that liver enzyme tests only need to be performed before starting statin therapy and “as clinically indicated thereafter.”

Memory loss and confusion will now be included in the label. But, the FDA notes, reports about the cognitive effects of statins have usually “not been serious and the patients’ symptoms were reversed by stopping the statin.” In a story published on Forbes, Matt Herper writes that the “new labeling will lead patients and doctors to take memory issues on the drugs more seriously, and it will also lead patients to link normal lapses of memory to their drugs.”

The increased risk of hyperglycemia and type 2 diabetes will be mentioned in the label. The label will now reflect data showing the small increased risk for type 2 diabetes in people taking statins.

The FDA is also making specific recommendations about lovastatin, noting that when used with some other drugs, lovastatin can increase the risk for muscle injury.

9 Responses to “FDA Revises the Safety Labeling of Statins”

  1. Robert Castro, DO says:

    With these new revised safety labeling for statins, might this affect the physician’s decision to consider statin therapy for those patients without significant risk factors but with an hs-CRP level greater than 2 based on the JUPITER trial. Since screening for some of these patients, we have found a younger population of patients that would benefit with statin therapy with elevated hs_CRP levels but are still within LDL goals based on ATP III. Any thoughts or discussion is greatly appreciated.

  2. David Powell , MD, FACC says:

    I would appreciate it if anyone could lead me to the data regarding the effect of statins on cognitive
    function.. I am curious regarding the strength of this data.

  3. There are only few answers for David´s question. In 2004 Golomb outlined UCSD study at http://www.ncbi.nlm.nih.gov/pubmed/15020036 . No big results about cognitive effects of statins from the big number of 1016 subjects. The only paper I could find is the abstract in Circulation (2006): Do Low Dose Statins Affect Cognition? Results of the UCSD Statin Study – Their conclusion: Findings provide reassurance that low dose statins confer little average cognitive impact with short term use in a generally healthy population, regardless of lipophilicity. However possible effect modification was provisionally supported. Future studies should evaluate subgroups with potential for differential statin effects; and should assess the higher doses of statins increasingly used in lipid therapy. See: http://circ.ahajournals.org/cgi/content/meeting_abstract/114/18_MeetingAbstracts/II_289-d It is a pity that UCSD study did not examine long-term effects of statins at high doses. In 2009 Evans and Golomb published Statin-associated adverse cognitive effects: survey results from 171 patients. Their results „suggest that cognitive problems associated with statin therapy have variable onset and recovery courses, a clear relation to statin potency, and significant negative impact on quality-of-life.“ http://www.ncbi.nlm.nih.gov/pubmed/19558254 By the way, in some patients who reported stopping statin therapy, a diagnosis of dementia or Alzheimer’s disease reportedly was reversed. (Conflict of Interest: I like to read Uffe Ravnskov.)

  4. One important issue to add about the UCSD Statin Study. Its follow-up of six months is very short in view of long-term, many a time life-long, need for the treatment.

  5. David Kaufman, MD says:

    I will review the links above. Having placed many hundreds of patients on statins over the last 25 years I really question this association. I continue to see these patients and can count on one hand the patients that have had a memory issue, cognitive complaint etc. These are incredibly important medications with a wide range beneficial effects. Considering the numbers of prescriptions written over the years—in the millions–where is evidence for this. My fear, like the HIV/Rhabdo issue, is that will discourage patients and docs. Has already come up in my practice.
    Question for you docs out there: how many of you take a statin and why?????

  6. Thank you, Larry, for clearly reporting that the FDA are talking about infrequent issues of memory loss or confusion. In my country (Australia) several media outlets misinterpreted things and ran with stories informing readers that statins can cause dementia (e.g. http://www.canberratimes.com.au/national/diabetes-dementia-linked-to-cholesterol-medicine-20120229-1u3ti.html ). I believe this allegation is false.

    I have tried to look into this issue and failed to find much solid evidence to show that statins cause cognitive problems, despite the recurring media stories over the years about this issue. Most evidence of harm seem to come from case reports (see http://www.ncbi.nlm.nih.gov/pubmed/12885101 or even the survey by Evans and Golumb cited above, which seems to be a survey of self-selected individuals who believe they’ve had statin-associated cognitive problems). Of course, case studies do not control for any other causes of an association such as placebo effect, and are thus low-level evidence. Some observational studies actually suggested statins might prevent dementia, but these were probably biased in the other direction. The best evidence I’ve seen is from a pair of Cochrane reviews on statins for the prevention or treatment of dementia. One (http://www.ncbi.nlm.nih.gov/pubmed/19370582 – looking at prevention of dementia) identified two large randomised controlled trials (PROSPER and HPS) which reported on dementia and neuropsychological measures of cognition. Neither trial found any evidence of a positive or negative association between statins and dementia. The other Cochrane review (http://www.ncbi.nlm.nih.gov/pubmed/20687089 – looking at statins for treating dementia) found only three smaller trials which when meta-analysed showed no significant difference between statin and placebo on most of a range of cognitive measures. Further evidence is awaited from a trial which is apparently complete but unpublished (CLASP – http://clinicaltrials.gov/ct2/show/NCT00053599).

    In my view, it seems unlikely that statins have a major role in the causation, prevention or treatment of dementia. If there are real cases of cognitive impairment from statins, they are probably either so rare that they don’t affect the results of that Cochrane review, or offset by cognitive improvements in other people so as to leave no significant overall net effect on cognition.

    However, I would still be aware of possible idiosyncratic reactions to statins and listen to my patients’ concerns with an open mind. I would be quite prepared to trial taking patients off their statins if they were concerned – more so the lower their cardiovascular risk. This would need to be a shared decision with the patient, weighing all benefits, risks and uncertainties.

  7. It is a pity that Golomb did not sudy 1016 subjects for several years like most statin trials have done to achieve some endpoint. Side effects or adverse effects should be put under the same scrutiny as statistical analysis of primary or secondary endpoints.
    There is no problem for any metaanalysis to conclude that side-effects of the drug are negligible as long as the metaanalysis involves trials where clincal investigators had been instructed before starting the trial that statins´ side-effects are very rare. (Cochrane metaanalysis is then one bias gained by putting several pieces of bias together.)
    The proof of the pudding is in eating it or something like that. The patient who says their memory is back again is the proof. Is there anybody who would like to tell the patient who declares that he or she feels better that the patient is wrong? … Except for some mania. Let there be no mania about statins either.

  8. As with most decisions in life, treatment decisions are always “trade-offs.” The first is: what is the prognosis if the intervention is undertaken, vs the prognosis if nothing is done? Since no intervention is uniformly effective, this risk that remains after the therapy should be discussed. Secondly, the question of comparative effectiveness arises. Third, the risk to reward ratio of each intervention is considered.

    Within each subset, what is reported by the patient, by and large, may not be measureable (or, impractical to do so). Not only placebo effects occur, but nocebo effects are more common than usually recognized. When symptoms are protean, and may be encountered from day to day in healthy individuals (eg, memory, aches, “indigestion,” etc.) establishing causality is difficult. This is not to deny the obvious, but to observe that nocebo effects are the other sharp edge of the same knife as placebo, yet the former are generally unwelcome to the physician, while the latter is accepted with a smile.

    The blade cuts both ways.

    Richard Kones MD