Yet one more thing to remember…

One thing I’d like to start doing with the blog is a weekly “in the news” post.  I was looking online today and ran across the following article from  Antidepressants, OTC Painkillers Not a Good Combination

Imagine you are working in fast track and you see a 50yo female with joint pain.  Your assessment after you have talked with and examined the patient is that it is likely arthritis.  You notice the patient has a past medical history of depression and is on Celexa.  Your plan is to discharge the patient on naproxen.  Is this a reasonable plan?

Prior to reading the article below I would have done this without thinking twice.  I ran a drug interaction check between SSRIs and NSAIDs and only found an increased risk of bleeding, SIADH, and hyponatremia.  Has anyone else heard of NSAIDs reducing the effectiveness of SSRIs?  I know this isn’t journal club, but will knowing this change your practice?

Since this is the first “in the news” post (and because I wanted to practice using the “polls” feature, let me know if you would like to see more “in the news” posts.


6 thoughts on “Yet one more thing to remember…

    1. ASA has one of the best “NNT” of any medicine ever. And MI victims have a higher rate of depression. Until we have better drugs, I say: so what?
      Give me my aspirin. There’s a reason even Harrison’s has a large subsection about ASA in its coronary disease chapter.
      These articles tend to make people make poor choices (see the book “Blink” about how irrational most people’s decision-making is, and our collective inability to more coldly calculate what is needed, what doesn’t make any sense, and when not to do anything at all <– that third one is what really freaks out the nursing staff the most)…

  1. (I just drank a bunch of coffee, bear with me here…)

    1) Fox news, again, Don? Ehh…
    2) How about, to be rather unprofessional, NSAIDs just totally suck. I’m pretty sure GI fellows history stops after “Do you take ibuprofen or Aleve?” If the answer is yes, guess what, scope. You just hit criteria. While I’m on that, why don’t you guys look and see how many people come back with GI bleeds a few months after being diagnosed with an ulcer by EGD who never got an H pylori Ab or never had it followed up, and, abracadabra, they’re back with another ulcer as the problem was never treated, just masked. (I’ve had 2 cases of this since January).
    3) It’s unfortunate fast track ends up being primary care for folks who just don’t make an acute visit to see their established PCP. It’s unreal how many folks stroll through fast track when so many of them could be acute visits to PCPs… barring other stuff, ortho stuff, ophtho stuff, etc.
    4) SSRIs in general are on my list to run through when I see hyponatremia in the hospital. It’s a super short list: HCTZ, SSRI, thyroid status, adrenal status, volume status. Usually one of those comes up (hint: it’s always volume, unless it’s not, or unless they obviously have a mass or something somewhere). But in reality, I can probably count 2 people having SIADH with an influence of an SSRI that I’ve seen. Have I missed any? I don’t know, but I think volume is way more frequently the problem in a hospital, and it’s tough to label SIADH anyway (it’s done prematurely and inappropriately more often than not by an overzealous primary team sometimes — there is a reason it’s a diagnosis of exclusion).
    5) To your original point, a few subcomments:
    a) This public article is inappropriate reporting. The primary thing being studied is in mice. They extrapolated results to a study bank and are making findings as only as good as the statistics they are processing… It was based on self-report, memory of what folks are taking, and really, there seem to be abundant variables left out… I do not think this is a good study. And, I think conclusions in “news” articles tend to jump the gun like this all the time, and draw quite far-reaching conclusions from a quite limited amount of possibly-related data. My take home point would be: ask if any of the conclusions being drawn SHOULDN’T be drawn, and why, and what would have to be done to validate any of this, and what your gut feeling on quite hazy areas really are (hence, EBM stuff) particularly in articles like this, which are rampant weekly.
    b) No I would do nothing to change practice. Hyponatremia in outpatients… eh. Bleeding = common anyway, so long as people know the risks. Making drugs less/more effective = see last point. Also, who takes just 2 drugs anyway? Whatever happened to polypharmacy? Pointing to one drug-drug interaction is a joke anymore unless it involves the CYP subclasses or coumadin, considering the exponential possible interactions and differences in folks’ metabolisms, etc. How often are we REALLY able to pinpoint “Ah ha, it was this. This caused everything to not work.” The world is more like Mr. Burns’ body in the Simpsons — he has so many diseases, if one was modified too much, then his balance would break, and he would instantly die.

    6) Finally. A better headline: “Antidepressants and narcotic painkillers. REALLY not a good combination.” I’d say, all painkillers have “evil” to them. But for inflammatory-based pain, NSAIDs are supposed to be used for a reason. Otherwise, as I tell patients, “If you’re sitting in the middle of a burning house and feeling all that heat and pain, when we give you Percocet, all we’re doing is masking the heat, while the house just keeps on burning, and maybe we’re even fanning the flames (read: PO narcotics –> gastroparesis –> IV Dilaudid –> worse gastroparesis/n/v –> admission. Repeat x 4 / month)”

    — Vince.

  2. There are lots of questions here: Is there really an effect? Is it dose and/or time dependent (is a week of ibuprofen OK, but not a month)? Is it all SSRIs and all NSAIDs or just certain combinations? Is it really better to give depressed patients “non-NSAIDs” (often opioids)? There are some great points from clinicians in the WSJ article that the Fox News article missed.

    Until there’s an actual human-based clinical study designed to look at this, no, I will not change my practice. And I certainly won’t stop giving aspirin to patients with chest pain (NNT to prevent death in STEMI is 42 based on at least one study) simply because they’re on an SSRI.

Comments are closed.