Robert M. Califf, MD

Statins, Diabetes and Cognitive Function (a problem with the drug or the system?)

The FDA added information to the label for statins today.  This led to some interesting press activity.  The details can be found at www.fda.gov or in numerous press reports.  It was particularly personally interesting to see that Ron Winslow from the WSJ had sought views from 3 people: me, Eric Topol and Steve Nissen.  He asked 3 people and got 3 different opinions.  A subjective view on the 3 issues addressed by FDA is given below:

1.  Hepatic function tests are no longer recommended routinely.  All I can say is that it’s about time.  There is no question that statins cause elevated LFTs in some people, but it is also true that permanent hepatic damage just doesn’t occur.  The routine testing has costed huge amounts of money, long after the utility was known to nonexistent.  Once something like this gets in a drug label, it takes forever to get rid of it.

2.  The evidence that statins increase the risk of diabetes was added to the label.  While no individual trial has been designed to specifically test the effect of statins on glucose levels, there is convincing RCT evidence from many trials of an effect.  The reason is unknown.  I believe this is worth knowing.  My comment to the WSJ was that among the many people with known atherosclerosis, especially those with a previous hard event, this small increase in risk of diabetes is not a reason to give up the benefits of reduction of risk of death, stroke, heart attack, renal failure, and heart failure.  Eric chose to focus on the many people without a history of a hard event or known significant atherosclerosis who take a statin for primary prevention–here, the diabetes risk should be a consideration.

The whole issue of statins in primary prevention is fascinating, but its a topic for another day.

3.  The one that bothered me was a note in the label that statins are associated with some cases of cognitive dysfunction. All of the evidence is anecdotal, case series, or epidemiological studies.  No RCTs have shown this, including some that looked at the issue specifically in the hope that statins would slow the normal course of cognitive decline.  Basically, cognitive dysfunction happens–seemingly with equal frequency in those on or not on statins.  There may well be some people with a specific idiosyncratic  risk, but the evidence is flimsy.  The good news is that if someone develops cognitive dysfunction while on a statin a simple n of 1 trial is easy to do!

At our Clinical Trials Transformation Initiative (CTTI) meeting the past 2 days we focused on adverse event reporting.  We have a lot to learn about how to transmit information that improves the use of drugs without frightening people with false signals and shabby evidence.  I would hate to think that people with serious atherosclerosis might stop a life saving or stroke preventing treatment because of this kind of shabby evidence!

more...