nos-trum. pronunciation: \nos'-trum\. noun. Etymology: Latin, neuter of noster our, ours.
1. a medicine of secret composition recommended by its preparer but usually without scientific proof of its effectiveness.
2. a usually questionable remedy or scheme.
See here for more discussion.

Tuesday, April 6, 2010

Cardiac Treadmill Not Reliable? Misses The Point

An article appearing in the LA Times yesterday talks about the "unreliability" of the treadmill.

This is manufactured news, and a waste of newsprint.

The article talks about treadmill exercise tolerance testing (TM-ETT), or cardiac stress tests.  The allegation is that they are accurate only "60% of the time, meaning they miss 40% of the problems."  They quote an expert who notes that "they're even less accurate for women."

Sounds bad, doesn't it?  Some years ago, my boss looked at me in a meeting and angrily noted that a friend of his had undergone a normal exercise test on the treadmill, and dropped dead later that week.  He didn't blame me personally; I was a convenient outlet for his anger.

All of this misses the point of TM-ETT.  There is a large body of tests that doctors perform when they are unsure whether something bad is going on.  Say you suspect an illness, but there's no solid evidence for high suspicion, or any clinical signs that point only to that disease.  You don't want to subject your patient to a risky test, or expose them needlessly to radiation, so you choose a screening test.

Screening tests can solidify your suspicion if positive, but don't allay all of your concerns if they're negative.  That means if you get a negative test, you have to decide whether to take the next step...or not.

Screening tests, whether a treadmill, or a urine glucose (for diabetes), or a prostate specific antigen (PSA, for prostate cancer), are all non-specific things that get you some information without putting your patient through a cardiac catheterization (complication rate 1%), six-hour glucose tolerance test (takes most of day,  without eating), or a prostate ultrasound/biopsy, etc).  In addition to low risk and simplicity, they are cheaper.

For Onion Peelers, (skip if you don't like statistics)
The way to look at screening tests is in terms of sensitivity and specificity.  A sensitive test will pick up a disease most of the time, but there may be a lot of false positives.  A highly specific test may not pick up all the diseases, but if it is positive, it's almost always "spot-on."  Almost no tests are highly sensitive AND highly specific.  With a screening test, you don't get spot-on accuracy, but you hope to gather in as many of those people who might possibly have a disease, and just accept that sometimes the test will say a person has a disease but they don't.

Which is better?  A test that tags a patient with a disease they don't have, temporarily, until further testing is done?  Or a test that misses identifying patients who have the disease?

I vote for the former.  Yes, it creates a hassle for patients, but if you explain what's going on, they get it.

Back to cardiac stress tests.  They are a good starting point, not definitive, but helpful.  To expect more than they are designed to deliver is foolish.  And media reports that try to impugn their utility are just trying to sell newspapers.

And, yes, all doctors know that cardiac stress tests are less helpful in women.  Women sometimes exhibit coronary artery disease differently.  That means retaining a stronger suspicion of coronary disease until proven otherwise.

1 comment:

Anonymous said...

I passed my treadmill, but now I don't know what to think! I guess if I ever get short of breath, I should tell someone!

Post a Comment


What I'm Reading - Updated 3 May

Blog Archive