Mythbusting

HEART score predicts adverse events for no risk chest pain patients – for a whole 66 patients.

There has been some buzz about one particular HEART score paper on SoMe recently, and while touting the HEART score is all the rage, the excitement over a whopping 66 patients seems downright silly.

This study screened about 5000 patients and excluded roughly 4700 patients – without obvious good reason – as “inclusion criteria were satisfied if the provider ordered an ECG and troponin for the evaluation of ACS. Consistent with prior studies, patients were excluded for the following reasons: new ST -segment elevation > 1mm, hypotension, life expectancy <1 year, a non -cardiac medical, surgical, or psychiatric illness determined by the provider to require admission, prior enrollment, non -English speaking, and incapacity or unwillingness to consent.” Seems like most would be included then, right? Except that was not the case at all. I just do not see how >90% of patients could be excluded based on these criteria, in a study done in the United States.

So we are led to believe that this tertiary care center, which has >100,000 annual ED visits, somehow could only scrounge 282 patients with chest pain to recruit into this study. They randomized 1:1 into the HEART “pathway” (HEART score plus serial troponins – more on this later), with usual care per ACC/AHA guidelines – admit/obs, serial troponins, and provocative testing in consultation with cardiology. They assessed adverse events (MI, revascularization, cardiac death), objective testing (stress or angiography), and future hospitalizations / ED visits within the year.

While yes, all sixty-six patients with a HEART score of 3 or less and 2 negative troponins (yep, a whole 66 no/low risk patients at a 100k + visit ED), did have zero adverse events at 1 year, the question really is, who cares? This is not enough patients to be excited about, and besides, these are essentially no-risk chest pain patients that shouldnt have a high risk of adverse events anyway. They represent a group whom we should not be doing additional testing on anyway!

As for all the others, there was no change in adverse events between HEART vs usual care- about 10% of patients had adverse events in each arm. As such, I dont read this as the HEART pathway being awesome at reducing long term issues, but rather, more of the futility of advanced testing aimed at reducing adverse events; The HEART score is merely equivocal to our current tools at future risk stratification. While this is good, I doubt it will result in fewer downstream tests in “higher risk” patients, if only because old habits die hard, Americans like to do more, and ACC/AHA recommend doing so. Choose wisely, perhaps?

Laughably, the authors open their paper with this line: “Care patterns for the evaluation of Emergency Department (ED) patients with possible acute coronary syndrome (ACS) in the United States are heterogeneous, inefficient, and costly.” And yet, where did we get this “HEART pathway” from? The HEART score did not have two troponins. To boot, neither did the largest study published looking at the HEART score. So what does a second troponin get you? Well, if you believe this study of over 45,000 patients, in which *all* comers had 2 troponins and 2 ekgs, [patients only excluded if abnormal vital signs were present & defined as hypotension (systolic blood pressure <100 mm Hg), tachycardia (pulse >100 beats/min), tachypnea (respiratory rate >20 breaths/min), or hypoxemia (oxygen saturation level <95%), or if they had electrocardiographic ischemia, a left bundle branch block, or a paced rhythm], they only identified a primary end point event in 4 of 7266 patients, 2 of which were noncardiac and 2 of which were possibly iatrogenic.

So, I again ask, where did the two troponin issue come from? Well, here it is- in one singular study, not doing a second troponin would have missed 5 adverse events out of 899 patients. Because the total event rate was 12, the authors state that their sensitivity went from 58% to 100%. While this sounds nice, it can also be reframed as a 1.33% miss rate vs 0.78% miss rate – an overall improvement of…. 0.55% with a second troponin. And, this has not really been replicated to demonstrate the value of a second troponin. So, either use the HEART score as it was meant, or use two troponins and gestalt- both probably get you into trouble and add waste to the system and you reach the point of minimal gains. You’re testing 200 plus patients to find one more bumped troponin that may or may not get a meaningful intervention.

So I ask hospital administrators and EM & cardiology colleagues: is it worth the extra 4+ hours (3 hours for the second troponin, plus time for the draw to occur and test to get processed & resulted), plus all of the associated resources & space not allotted to other patients due to these 4+ hours that are given to thousands of patients annually… is a decrease of 0.55% worth a second troponin? I imagine if time and bed space was allocated to others, the 0.55% increase of adverse events would be more than made up for across the rest of the patients within the department by earlier recognition and treatment of just about every other disease entity within the department.  Sounds like, at a minimum, a time to introduce shared decision making (“hey, we catch one in two hundred with a second troponin, want to stay another 4 hours? and no, you can not go outside and smoke.  How about a nicotine patch and a tuna sandwich?”).

And lastly, no, you can not say the HEART score results in low risk of 1 year adverse events based on a single center study that failed to enroll more than 66 patients.

Standard

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s