Sunday, July 18, 2010

Heart and Soul

UCSF medical school is running an 8 year prospective cohort study following patients with heart disease. The goal of the study is to gain some perspective on how psychological states affect your heart. A number of papers have been published (free full text! Suweet!), and today I'm focusing on one called "Scared to Death? Generalized Anxiety Disorder and Cardiovascular Events in Patients with Stable Coronary Heart Disease."

Anxiety has never had quite as much robust medical research that its big brother depression has. Perhaps because depression is easier to quantify and easier to treat. Anxiety tends to start in your youth, as you learn how to cope. By the time you end up getting treatment for it, you've probably had it for many years, even decades, and it becomes a part of who you are.

So it's no surprise that it is already known that depression and heart disease go hand in hand. If you have depression, you are more likely to develop heart disease (1), and if you have depression and heart disease, your heart disease will likely be worse (2).

But what studies there are of anxiety and heart disease show that anxiety is common among those with heart disease, and anxiety symptoms predict the amount of disability you will have. In the Heart and Soul study, 1015 people (mostly veterans from the VA medical centers) were followed for an average of 5.8 years. Generalized Anxiety Disorder (there are always specific criteria for these things, but in general, someone with GAD will worry a lot and have physical symptoms associated with worry, to the point that daily functioning is impaired. Panic attacks can also occur) was tested for via a Diagnostic Interview Schedule (that's a good test - a lot of studies will just use regular old scales, but the diagnostic interview is really the gold standard). In addition, everyone in the study had cholesterol measured, exercise capacity tested, a 24 hour heart monitor, 24 hour urine to measure norepinephrine and cortisol (chemicals associated with stress), C-reactive protein, and red blood cell percentage composition of fatty acids such as omega 3s, saturated fats, and omega 6s (which is the best way to figure out the fatty acid composition of the diet). Other patient data was also taken into account - age, sex, race, education, smoking, exercise, height, weight, and medications. Whew. All and all, pretty comprehensive, and all the latest technology. So far so good.

Each year, the investigators called up the study participants and asked about heart trouble. If anyone had an EKG, or a heart attack, any other "heart event" (specifically stroke, heart failure, MI, TIA, and death), the investigators got a hold of a copy of the medical records. Then they subjected the data to a tortuous round with the statisticians to try to sort out any confounding variables, and at the end, we get a bunch of nifty tables of information.

So what did they find?

10.4% of the participants met criteria for Generalized Anxiety Disorder (that's about in line with the literature - about 1/8 of people who visit their primary care doctor have GAD). Those who were anxious were also more likely to be younger, female, depressed, have better heart function on echocardiogram, take antidepressants and anxiety medicines, have lower omega 3 fatty acid levels in the red blood cell membranes, be less likely to exercise, and more likely to smoke. Also, they were less "adherent to medications" (what doctors call "noncompliance").

And the "heart events"? The annual rate of cardiovascular events was 6.6% for the people without generalized anxiety, and 9.6% of the people with GAD (p=0.03). That's annual! Meaning in 5.8 average years followed, there were a lot of medical records for the poor investigators to pore over!

And the confounding variables? (things which might cloud the statistical correlation between anxiety and heart disease) - male sex, heart function itself, exercise capacity, certain medication use, level of physical activity, and heart rate variability. So if those variables are "adjusted" for, you end up with a 62% greater rate of cardiovascular events for someone with generalized anxiety disorder and "stable" coronary heart disease. The raw data leaves you with a 74% greater rate of cardiovascular events. Either way, if you have heart disease, you are better off if you are more or less serene.

What do we take away from this paper? They have a discussion at the end worth reading. There's always a question in the medical literature about patients with psychiatric conditions - maybe they are sicker because they are too depressed or anxious to take care of themselves properly. They eat garbage, smoke more, exercise less. But, time after time, the studies show there is more to the connection than just crappy self-care. Interestingly, this study didn't find any link between physiologic markers of stress (the 24 hour urine measures, CRP, and heart rate variability) and the increased risk of anxiety disorders and heart disease. Smokin' and being lazy didn't explain the correlation either.

The authors postulated that a 24 hour urine wouldn't capture the risky "catecholamine spikes" of stress hormone that would be more likely to precipitate a heart attack. That makes sense. Then they speculate that anxious people are less likely to seek medical care (I sincerely doubt that one!), or are more likely to seek medical care (more realistic), thus the increase in recorded events was due to anxious people being more likely to consult their doctor with symptoms. Except, unfortunately, people with anxiety were more likely to be dead at the end of the 5.8 years of follow up, and that is one condition that isn't likely to be missed or uncounted.

Here we go: "It is also possible that there exists a common background origin to GAD symptoms and risk of cardiovascular disease." Also, GAD was associated with lower omega 3 fatty acid levels and depression, and "there is a clear association between lower omega 3 fatty acid levels and cardiovascular risk."

The conclusion? Take care of your anxiety! And eat some wild-caught salmon tonight.


  1. What I'd love to see someone study is the connection between anxiety (with or without depression) and gluten intolerance. Lotsa celiac folk seem to report anxiety-flavored depression that abates with a gluten-free diet.

  2. Nice post. If I were a betting man, I'd be putting money on gluten being at least one of the factors in a 'common background origin'.
    Interesting re: the lower omega-3 fatty acid levels. It would have been good to see the full analysis of RBC's - n-3, n-6, & SFA. There is certainly a 'bias' toward n-3 being protective, thereby suggesting that increasing n-3 intake would be beneficial. It is possible that it is the lowering of n-6 intake that holds as much if not more benefit. There is a lot of focus on the ratio between the two, but the absolute amounts of each could be equally or more important.

  3. It's hard to tease apart the relationship between mental health and heart disease. I always look for studies that test whether TREATMENT of anxiety or depression actually reduces future cardiac event rates.

    Those studies are few and far between.

    First of all, you have to decide which single method you're going to use to treat all that anxiety and depression. You don't want too many variables. Talk therapy (e.g., cognitive therapy) always takes a back seat to drug therapy.

    Very difficult endeavor.

  4. Jacflash - a recent German study looked at that. Here's my snippit - "Germans (primarily women) with Celiac Disease on a gluten-free diet were compared with a group of Germans with inflammatory bowel disease (Crohn's or Ulcerative Colitis) and a group of normal controls. Women with IBD and celiac disease were both more anxious than the controls. Men seemed to be serene no matter what. Adherence to the gluten-free diet didn't seem to matter much with regards to level of anxiety. In a related study, anxiety improved in patients with celiac disease in a year on a gluten-free diet, but not depression."

    Jamie - yes, I certainly agree that the ratio is more important than the absolute amount of omega-3.

    Steve - I think SADHEART was the biggest and baddest trial (there are a couple of cutesy-named depression/cardiology trials and I can never remember off the top of my head the exact details of each one) - but the summary of the big trials came down to this - SSRIs like zoloft improve mortality in depressed individuals after a heart attack. Tricyclic antidepressants are a wash (and maybe do worse), even if the depression is treated successfully with the medications. So the improvement in depression has little to do with the mortality benefit, it was all about the type of drug. SSRIs have known anti-platelet effects and are likely cardioprotective. TCAs are known to be somewhat cardiotoxic (they increase heart rate, etc.). There's always that question of self-care - people who aren't depressed or anxious ostensibly take care of themselves better, but that doesn't seem to matter overall in the trials I remember.

    I think the most intriguing interpretation is that an underlying derangement causes both heart disease and depression (and anxiety, though that is trickier still to sort out, perhaps, because you get the rushes of stress hormones and subsequent arrhythmia and sudden cardiac death/toxicity, takutsobu cardiomyopathy is an extreme case of this - saw a case recently! Tough to know exactly what to do. I can't order people to relax, after all :). SSRIs happen to destickify platelets so incidentally improve mortality in heart disease (rather like the incidental anti inflammatory effect of statins in the appropriate candidate for them).

    I agree talk therapy takes an undeserved second place in the big trials. But then that's not where the money is.

  5. This came through this morning on Medscape if you have access... ties in with some of your post Emily. We have known that erectile dysfunction is perhaps a lead indicator for cardiovascular dysfunction for a while now.

    "Our results show that when evaluating patients for sexual dysfunction, doctors should think about general health as well. Erectile dysfunction may be the first disease or depression may be first disease, but we should look beyond these initial conditions to look at secondary consequences such as increased cardiovascular risk. If we treat depression and sexual dysfunction, we may be able to improve cardiovascular outcomes, too.

    What is important . . . is the broader concept of the sexual-medicine problem no longer being just about a man's performance in the bedroom, but about his psychological mood and his cardiovascular health."