Medicine is still hard

I'm passionate about medicine. I love practicing it, teaching it, learning it...everything about it. It's also damned scary sometimes. Knowledge changes; patients don't walk in with textbook diseases; and you have to be able to doubt yourself without drowning in indecision. Every time a patient brings me an article or monograph that looks relevant to their care, and it's new to me, I have to go through it and see if there really is something to it.

For example, there has been some data recently about popular medications for acid reflux (called PPIs, including such medications as Prilosec) and possible interactions with a potent anti-platelet agent called clopidogrel (Plavix). What does this mean for me and my patients?

This is not a trivial conundrum. Here's a fairly typical scenario: a patient comes in with a heart attack, and the cardiologist performs an intervention to save her. Knowing the data, the cardiologist decides to deploy a drug-eluding stent in the patient's blocked artery. Without the procedure, the patient is likely to be disabled or dead. In order to prevent the stent from re-occluding, leading to another heart attack, she must stay on clopidogrel for a long time (probably a year).

Then, a few months into the treatment, the patient develops rectal bleeding. Endoscopy reveals a bleeding ulcer, which the doctor cauterizes. In order to prevent further bleeding, the patient needs to be on proton pump inhibitor such as omeprazole (Prilosec). We can't just stop the clopidogrel, or the patient may have a heart attack. We also can't let the patient bleed to death.

The whole clopidogrel/omeprazole thing is pretty interesting. It turns out that clopidogrel is really a pro-drug, and must be metabolized in the liver by the cytochrome P450 (CYP) system into its active metabolite. Some people are naturally deficient in the necessary functional CYP, and continue to clot even on the drug. In the future, we may even be able to do genetic testing to see what patients may need a higher dose of the drug or a different approach altogether. Prilosec is available over the counter, and a patient may take it without my knowledge (and without needing it) complicating things further.

It turns out that omeprazole inhibits the same CYP enzyme that is needed by clopidogrel. Lots of patients are on ompeprazole (and other similar drugs called PPIs), so in clinical settings, we don't know if patients who get more blood clots on clopidogrel do so because of their genetics, because of a PPI, for other reasons, or just by chance. If the patient asks me what to do next, how do I give a reasonable answer?

This kind of thinking is going on continuously. Do I give this patient with a sinus infection antibiotics? If so, will there me a medication interaction? Is the patient at high risk for side effects, such as C. difficile colitis? And how can I relieve their symptoms without raising their blood pressure?

My decisions have real consequences. This is the reason that I get really, really angry when people who don't know what they are talking about spend a few minutes at Google U. and all of a sudden think that Potion X, Herb Y, or whatever is some really good shit. How do they know? Because "some dude" said so? People listen to trusted friends a lot more than to doctors, so the damage can be real and immediate. If I have a patient on warfarin to prevent a deadly blood clot, and their nephew tells them that Potion Zed is good for blood clots, and Potion Zed just happens to have some vitamin K in it, my patient may die. I'm not making this up---ask any doctor in practice for their latest horror stories.

Intervening in human health is serious business, and people will listen to advice from anyone, knowledgeable or not. Sometimes my patient's greatest enemy is a well-meaning friend.


More like this

Things have been rather busy at Pal's place. For whatever reason, the complexity of patients has been pretty high lately, so I haven't had a chance to get in my usual once or twice a day post. I'm told that "back in the day" internal medicine patients were a bit less complicated. These days, to…
Warfarin, a commonly used anti-clotting drug, sold under the brand name of Coumadin, has a been a poster child for the promise of pharmacogenomics and personalized medicine. The excitement has come from the idea that knowing a patient's genotype, in this case for the VKORC1 and CYP2C9 genes, would…
Yesterday I issued a challenge to naturopathic physicians to justify why they should be considered competent primary care physicians. The best and most comprehensive answer received so far is the one from "Mona". Here is my analysis. Her response, while not entirely "wrong", shows a frightening…
Or if he does, he's even a worse person than I'd realized. After suffering chest pain in Hawaii he was evaluated in a hospital. When discharged today, he held a briefing in which he praised the U.S. health care system as being the best in the world and remarked that he sees nothing wrong with it…

I've heard that prasugrel may get around some of these problems with clopidogrel, at the cost of adding various problems of it's own. Including (correct me if I'm wrong, this is off the top of my head) some anecdotal questions of an increased cancer risk.

Still a wonderful illustration of the complexity of primary care.

Another good example is the patient who is both a clot risk and a bleed risk: The patient with a history of stomach ulcers, with cardiac stents and a mechanical heart valve. I've seen these patients on warfarin, plavix, and aspirin before. For those who don't want to wiki, that's 3 drugs that make you more likely to bleed which are all appropriate for a patient with stents and a mechanical valve, but a big risk for a patient with stomach ulcers.

Of course, one of the times I saw a patient on that regimen, he had a stroke, because he was also taking "mega vitamin supplements" that a naturopath said would help his stomach ulcers... which contained vitamin K, and leading us right back to your example.

well-done post, Pal. Ars longa, vita brevis. Unless we're talking about dilettantes. Then it's ars brevis, and the hell with complexity. It's not just the nephew with potion zed. There's also the anti-vaccine nut, the chiropractor and the colon cleanser. You have accurately stated the burden of acting in a responsible way on behalf of your patients. You are a mensch.

Squeezing a balloon.

Best wishes,

By D. C. Sessions (not verified) on 16 Aug 2009 #permalink

Great article. I really like the way you give a human face to medicine- the responsibilities, the dilemmas. These are issues which most people would never otherwise think about. Thank you for doing what you do, both in the clinic and on this blog.

I've been wondering why it's not possible to develop a cheap and easy functional poly-CYP screen with a bunch of colored substrates (e.g. how long does your urine stay purple after eating this candy?) instead of trying to worry about all the specific mutations.

I also wonder if there could be some formulation of the ompeprazole such that it would *not* get metabolized in the liver, but get broken down specifically in the intestine where it is needed to prevent the ulcers (apparently CYP2C19 is expressed in the intestine).


My understanding is that Medicine is not hard.

Medicine, is a right bastard.

Or as Goldacre, being English, puts it "I think you'll find it's a bit more complicated than that"

By antipodean (not verified) on 16 Aug 2009 #permalink

Interesting. As I take Prilosec daily to control my GERD symptoms, and as I have several relatives with heart conditions (including one who is on warfarin all the time after several close calls with DVT), this information could one day save my life. Fortunately, my doctor knows about my Prilosec usage (I am taking it under her direction), but will an ER doctor know about it? More and more, I see the need for a common electronic records system.

Vitamin K and warfarin -- I never thought of that (dammit, I'm a software engineer, not a doctor), but it makes perfect sense. I've seen plenty of episodes of Emergency Vets where a dog got into rat poison, and the treatment was 1) vomiting, 2) force-feeding charcoal, and 3) intravenous vitamin K to counteract the warfarin. So I should have thought of that. I'm sure Joe Schmoe won't think of that at all when popping a multivitamin along with his daily Coumadin dose.

By Calli Arcale (not verified) on 17 Aug 2009 #permalink

Calli, if Joe doesn't know about vitamin K risks he either wasn't listening or had a crappy doctor. Probably the first.

when I turned up at the ER with bazillions of teeny tiny pulmonary emboli, and was out of danger, when they put me on warfarin, they told me all about monitoring my vitamin K intake. I got a handout that listed foods that contained it and was told not to necessarily AVOID them, but to ensure that once stabilized at that magic "therapeutically decoagulated" point, not to CHANGE my intake, either up or down.

Luckily here (in BC), as soon as my health card is scanned in the system, the doctor has a record of all my prescription medications, thanks to our PharmaNet program. So, they can see that I'm on pantoprazole (among other drugs).

Pal, is there the same risk with H2 antagonists like ranitidine or cimetidine as there is with PPIs? I know that my ulcers were kept at bay nicely by cimetidine for a very long time, and it's a LOT less expensive than the PPIs.

(Eventually the combination of ketoprofen and diet coke was more than the cimetidine could a switch to a COX-2 inhibitor and eventually a PPI was needed to prevent a third ulcer.)

By CanadianChick (not verified) on 17 Aug 2009 #permalink

AFIK, pantoprazole may not be as bad as omprazole (assuming there is clinical relevance to the omeprazole thing). I'm not aware of H2 blockers such as ranitidine having similar effects.

this is an extremely useful resource:

I keep a copy of the table chopped up in my palm pilot.

Ranitidine isn't much of a cytochrome inhibitor. Cimetidine inhibits several human cytochromes and has numerous potentially troublesome drug interactions.

As for pantoprazole, check this: Effects of pantoprazole and esomeprazole on platelet inhibition by clopidogrel.
Siller-Matula etal Am Heart J. 2009 v157(1):148.e1-5


Yeah, we really don't use any cimetidine. We use famotidine and rantidine instead.

Interesting that omeprazole appears to be the biggest baddie, given that it's the cheapest by far.

Thankfully, pantoprazole should come down in price soon.

I'm loving my EMRs more and more. The one I use has an interactor-detector dealie, with references.

The X-factor is a brute. Here, we have a high density of woo. One patient couldn't figure out why his sugars were out of control, and swore up and down he hadn't changed anything. We did a brown-bag and he was taking a load of supplements chock-a-block with lactose, fish-oil, and complex carbohydrate.

I can empathize with your position, but Google U -- or at least PubMed U -- saved my life (and I'm not kidding), through revealing a well-researched side effect of a prescribed medication that was seriously minimized in the info for practitioners or patients. It probably doesn't help that I like many healthy young women was getting a large portion of my primary care from a rotating cast of GYN NPs (NPs are awesome, and I prefer them for routine healthy-adult care, but it has not been my experience that they have the same basis for troubleshooting as a good physician).

Care to share some specifics for us? For conversation's sake?