Coronary angiography is the "gold-standard" test for coronary disease. It's considered the final word on whether or not you have it. It involves threading a catheter to the arteries that supply the heart. A liquid that x-rays don't penetrate is injected into the blood, and so with an x-ray (really fluoroscopy, a continuous x-ray image over a few seconds) you can see the exact shape of the coronary arteries.
Source: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5143685/ (image creators do not endorse opinions in this post)
Multiple specialty societies caution against its overuse. For example, as noted on the Choosing Wisely website, the Society for Cardiovascular Angiography and Interventions says: "Avoid coronary angiography to assess risk in asymptomatic patients with no evidence of ischemia or other abnormalities on adequate non-invasive testing." That is, if a patient has no symptoms and no evidence of coronary disease on testing, try not to do an angiogram.
And yet there is evidence that, at least in the past, angiograms have been overused. A 2014 study of a database of angiograms in New York State in 2010 and 2011 found that 25% of them did not appear to be justified, based on established criteria.
In medical culture, there is a place for going outside guidelines. Providers are taught from the beginning to treat the individual patient according to the circumstances. "Guidelines are great, but there is no substitute for thinking," as I used to tell my students and residents.
I must say that after talking with a cardiologist recently, it sounds like this old information about NY state angiograms has been taken to heart by the community, and things have changed. And yet, given the structure of our health care system, with its financial incentives to do procedures, all the malpractice risk, and patients' and providers' desire for certainty, excess testing still occurs.
At Independent Health, we have access to claims data from the entire region. Claims data only reflects information sent to us with a request for payment. And yet it is somewhat useful in determining disease and treatment patterns on the ground. We put claims data through the Milliman Waste Calculator, which is the industry standard in assessing whether certain care might not have been necessary. The calculator has been used, for example, in Washington state, and has generated media reports.
When we looked at coronary angiography, the calculator identified 860 coronary angiograms that may have been inappropriate during the first half of 2018. The logic the calculator used can be summed up as: "patients who had an angiogram but did not have a significant heart condition or a positive stress test." If anyone would like a more granular look at the logic, please contact me.
What does this mean? It depends on your tolerance of uncertainty and risk.
Let's take the perspective that the guidelines are clear dividing lines between what should and should not be done. Even if the calculator is highly inaccurate, and only a fraction of the 860 angiograms did not meet guidelines and were unnecessary, that still means that patients were put at risk of stroke or cardiac complication for no reason. I think that putting someone at unnecessary risk is harming them.
But we can take the other perspective, that the cardiologists, assessing risk on the ground, based on experience and a multitude of details, made generally good decisions and only subjected patients to that procedure if the benefits of the information obtained clearly seemed to outweigh the risks. In that case, who is it that carefully assesses whether the cardiologist, in this subjective matter, has judged reasonably? And that he or she has carefully set aside the financial incentive, the malpractice risk, any pressure from the patient to "find an answer," and their own personal biases toward knowing what's going on? It's not me, I'll tell you that.
And that's probably why gravitate toward the guidelines.
The original source of the Milliman waste calculator logic on this topic is here: https://bit.ly/2SQRunP