About the ASCVD and ACC/AHA 2013 Calculators

With the launch of the ASCVD Calculator and the ASCVD algorithm we recently added to MDCalc (The difference? I’ll explain further down) we thought it might be nice to review the 2013 guideline. Let’s start at the beginning.

Before the ASCVD

A long time ago, in a galaxy far, far away, (2002) there were the ATP-III Guidelines — short for the “Adult Treatment Panel,” a group of cholesterol and lipid experts that attempted to figure out what the heck to do with patients with lipid issues. It really focused on LDL cholesterol and addressed trying to aggressively reduce it. Find high risk people with high LDL, and get that LDL down! They based this data on the Framingham studies that followed a bunch of people in Framingham, Massachusetts and tried to figure out what things predicted you having, say, a heart attack.

66384192

Here Comes the ASCVD

Fast forward 11 years later, and in 2013 this comes out: 2013 ACC/AHA Guideline on the Treatment of Blood Cholesterol to Reduce Atherosclerotic Cardiovascular Risk in Adults: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. This is a big, fancy, long document (with a big, fancy, long name) that a bunch of experts developed to attempt to re-examine cholesterol and outcomes — and importantly, also include stroke in the cardiovascular outcomes. (Previously, stroke wasn’t really included as a “cardiac outcome.”) So the authors looked at a number of studies of cholesterol and cardiovascular outcomes, and then this time, did something different: they came up with their own formula to determine people’s risk of having a bad cardiovascular outcome — and then told everyone to use that formula to decide who should get treated for their cholesterol issues.

But it’s never that simple.

This report and guideline isn’t just a calculator; you can’t just input your patient’s info and  spit out a recommendation. This is probably what busy clinicians would want: “Okay, I have this patient with a million medical problems, let me just at least get some simple help from the cholesterol guidelines that will tell me their recommendations.” This is also, unfortunately, what a lot of people think the guidelines do.

Let me say that again: these guidelines are not equivalent to the ASCVD calculator — not the one recommended by the ACC/AHA from Cardiosource and not even ours from MDCalc (that does the same thing as the ACC/AHA). Here’s the actual algorithm that was put together in the paper to try to summarize the guidelines:

Figure 2 from the 2013 ACC/AHA Guidelines.

Figure 2 from the 2013 ACC/AHA Guidelines.

But you really don’t start using the ASCVD algorithm until you get to the orange box:

F2.large.jpg

Figure 2, Modified to show where you apply the ASCVD Calculator (orange box).

Not only is this algorithm incredibly confusing, it’s not particularly intuitive for a practicing clinician. First, you have to know what they’re defining as “Clinical ASCVD,” and then go through the fairly complicated algorithm to figure it all out. After that, you’re also directed to two other algorithms (Figures 3 and 4) for your specific patient.

It’s an incredibly cumbersome and poor implementation strategy to try to get clinicians on-board with a new recommendation, complicated by the fact that there’s not really a clear reason they give to use cutoffs of 5%, 7.5%, and 10% CHD risk to determine who should or should not receive treatments.

The MDCalc ASCVD ACC/AHA Algorithm 

Patients don’t come in with a label on their foreheads, “No history of Clinical ASCVD, LDL <190 mg/dL,” so it seemed really confusing to try to build a calculator that only applies to a subset of patients based on history and lab values. Obviously everything has some inclusion or exclusion criteria, but they’re a little easier to follow. Because of this, we converted the above Figure 2 — as best we could — into our ASCVD ACC/AHA Algorithm page. The main difference? You start from the beginning with your patient, enter some basic information about them, like their history and age and LDL cholesterol level, and out comes a recommendation from the guideline on the other end. Users don’t have to remember some fairly complicated criteria about when to use the calculator and when not to use it when applying it to the entire adult population of the United States.

 

Next time, we’ll look at some of the evidence and controversies behind the calculation.