Another strike against NSAIDs?

- Jennifer L. Middleton MD, MPH

One of the issues family docs deal with on a daily basis is pain control.   I usually think about pain medication as falling into one of three categories: acetaminophen, NSAIDs, and opioids.  I frequently recommend acetaminophen, but patients often tell me "it's not strong enough for me" (maybe an unintentional consequence of those commercials touting Tylenol's gentleness?).  And, of course, I defer opioid regimens if possible given the risks of addiction and diversion.

Perhaps you've already heard about The Lancet's NSAID meta-analysis article from about a month ago.   The authors performed a robust literature search and included hundreds of trials with several outcome measures, one of which was the rate of "major coronary events" (a composite of non-fatal myocardial infarction and coronary death).  The authors found that long-term use of all non-steroidal anti-inflammatory drugs (NSAIDs), selective COX-2 or non-selective, doubled the risk of heart failure.  I'd like to focus on two non-selective NSAIDs, ibuprofen and naproxen, for the rest of this post.

I found this meta-analysis unsettling, as I like having an option in between acetaminophen and opioids to offer my patients.  True, the authors only examined patients taking high dose NSAIDs (2400 mg ibuprofen/day and 1000 mg naproxen/day) for at least 4 weeks.  Is it safe, then, to extrapolate that lower doses and/or shorter periods of time are safer?

AFP had a nice article about osteoarthritis treatment last year that discussed the pros and cons of all of these medication classes. Rereading that article this past week reminded me that every 12th patient taking an NSAID, even short-term, will experience a gastrointestinal (GI) bleed, kidney problem, or elevated blood pressure (number needed to harm [NNH] = 12 for that composite outcome).  As NNHs go, that's a pretty impressive number.

For now, at least, NSAIDs probably should be off the table for patients at an increased risk of heart disease.  Myself, I will probably continue recommending NSAIDs in patients without a history of GI bleed, with normal kidney function, and without a history of heart disease, but I will recommend more modest doses and shorter periods of use.  I will probably spend more time counseling patients, too, about the risks of ibuprofen and naproxen.

I encourage you to take a look at these related AFP By Topic collections:
Heart Failure
Pain: Chronic 
Arthritis and Joint Pain (includes this AHRQ-EHC review's discussion of NSAID risks)

How frequently have you been recommending NSAIDs?  Will this meta-analysis change your NSAID prescribing?