Thread by @SatyaPatelMD

1/18 Chest pain is a frequently seen reason for admission. Here's my take on when to consider ACS in patients with chest pain!
#MedTwitter #CardioTwitter #MedEd #FOAMed #MedTweetorial @MedTweetorials

2/18 Our evaluation of ACS starts with 3 things:

1⃣ History
2⃣ EKG
3⃣ Troponin

The primary focus of this thread is going to be on the history (a heads up - the flowchart at the end will go a little bit out of order)!

3/18 When taking a chest pain history, we ask lots of questions about associated symptoms and alleviating/aggravating factors, mostly because we were taught to obtain and report this history. But is there a more focused way to approach this?

4/18 It's worth noting that an anginal equivalent is very vaguely defined and can vary significantly between male and female patients.

5/18 We are hoping that the answers to our questions can help us quantify our concern for ACS. Let’s split our line of questioning into three big baskets that we are trying to report:
1⃣ Diamond-Forrester classification
2⃣ HEART score
3⃣“Everything else”

6/18 The Diamond-Forrester classification was first introduced by Dr. Diamond in a JACC editorial - reader.elsevier.com/reader/sd/pii/…
He compared the angiographic prevalence of CAD with the probability of CAD based on "typical" features of chest pain.

7/18 Looking at Table 1 from that article, Dr. Diamond posited more classic symptoms ➡️ higher chance of CAD.

Thus, the Diamond-Forrester criteria were born.

8/18 Now let’s turn to the HEART score. This score was developed by Drs. Six, Backus, and Kelder as a way to help determine if a patient with chest pain in the ED has ACS (this was done in the Netherlands). ncbi.nlm.nih.gov/pmc/articles/P…

9/18 They found that the higher the HEART score, the more likely the chance that the patient had a combined endpoint of acute MI, PCI, CABG and death.

10/18 The last part is “everything else” and this includes things like radiation of pain to the neck or jaw, pleuritic pain, reproducible tenderness to palpation, etc. Interestingly, we may not be asking the most helpful questions to change our probability.

11/18 How do we use Likelihood Ratios to help us? Use a Fagan Nomogram! Here’s a great article about using it.
LR close to 1 ➡️ no change in probability
LR > 1 ➡️ increase in probability
LR < 1 ➡️ decreases probability

12/18 Enter the JAMA Rational Clinical Examination Series. Here is an aggregate of some of the findings, listed from highest LR to lowest LR. It’s interesting to see that “typical” chest pain based on Diamond-Forrester criteria only has an LR of 1.9.

13/18 Notably, an article in JACC suggests that a HEART score of 5 or less may not require further evaluation in the ED. jacc.org/doi/full/10.10….

14/18 Remember that the HEART score is only as good as your clinical suspicion (there is a reason that part of the score requires you to assign a point value to history. Don’t worry, your Spidey-sense will continue to get better as you see more and more patients).

15/18 The last major part of the history is to assess other pathologies that might explain the chest pain, such as pericarditis, myocarditis, sepsis, or heart failure. Patient’s can absolutely have ACS + another pathology, so be on the lookout! History is 🔑

16/18 As far as EKGs are concerned, here is a review of the classic progression of EKG changes in ACS (and some STEMI-equivalents).

17/18 When looking at cardiac biomarkers, it is interesting to see what was used before troponins.
Now we have hs-troponin, which might become a game-changer in ruling out ACS (outside the scope of this thread).

18/18 So how do we put all this info together? You can use this flowchart (note - the flowchart uses the exam and HEART score before the history)! Thanks to @jimenezd19 and @jiwenli for reviewing everything. Would love feedback!