Paucis Verbis card: TIMI risk score

Paucis Verbis card: TIMI risk score


Chest PainHow do you risk-stratify undifferentiated chest pain patients in the Emergency Department? There are a multitude of causes for chest pain. We are always taught to think of the 5 big life-threats: ACS, PE, aortic dissection, tension pneumothorax, and pericardial tamponade.

So how do YOU risk-stratify your patients for unstable angina (UA) and non-ST elevation myocardial infarction (NSTEMI)? STEMI’s are usually obvious. UA and NSTEMIs — not so much.

Fortunately a 2000 JAMA article and a followup Academic Emergency Medicine 2006 study have solidified the TIMI risk scoring system as a reasonable risk-stratification tool for all-comer ED patients with chest pain requiring an ECG.

Generally there is an upslope in risk at a TIMI score of 3 and greater.

PV Card: TIMI Risk Score

Adapted from 1,2
Go to the ALiEM Cards site for more resources.

Pollack C, Sites F, Shofer F, Sease K, Hollander J. Application of the TIMI risk score for unstable angina and non-ST elevation acute coronary syndrome to an unselected emergency department chest pain population. Acad Emerg Med. 2006;13(1):13-18. [PubMed]
Antman E, Cohen M, Bernink P, et al. The TIMI risk score for unstable angina/non-ST elevation MI: A method for prognostication and therapeutic decision making. JAMA. 2000;284(7):835-842. [PubMed]

Michelle Lin, MD
ALiEM Editor-in-Chief
Academy Endowed Chair of EM Education
Professor of Clinical Emergency Medicine
University of California, San Francisco
Michelle Lin, MD
Michelle Lin, MD

Latest posts by Michelle Lin, MD (see all)

  • Michelle,

    Another good source for the use of TIMI in the ED is the recently published meta-analysis:
    Canadian Medical Association Journal. 2010. Vol. 182, Iss. 10; pg. 1039.

    if we are going to use the score, wouldn’t it be better to split up the TIMI 0 and the TIMI 1 patients, as it would seem only the former are low enough risk to be sent home prior to provocative testing.


  • Thanks for the citation, Scott. It’s a free download at: As for your question, looks like the derivation group didn’t have enough “low risk” or “high risk” patients in their study (TIMI 0 or 1, TIMI 6 or 7) and so they combined 0 with 1 and 6 with 7. Tis a bummer.

    In the meta-analysis you mentioned, looks like even with a TIMI score of 0, the risk of adverse outcome in 30 days is still 1.8%. The paper concludes that TIMI scores should NOT be the only means of determining patient disposition.

  • Anonymous

    I am not sure how you use the TIMI score in the ED. The adverse outcomes are too high for me. Sending home 1.8% misses is high.

    Is there a study that uses physician gestalt + TIMI score for adverse outcomes? Similar to the PERC rules?

  • Good point. In all honesty, it mostly helps me (1) to teach students about the higher risk elements to consider in the history/labs and (2) in convincing the cardiology service to admit a patient with atypical symptoms but, perhaps, is on aspirin daily.

    I have not seen anything that incorporates physician gestalt with TIMI score unfortunately.

  • Less than 2% is probably all we can ask for. Klein actually did the hard math for PE, which is only tangentially applicable to MI/ACS, but trying to go below 2-2.5% in that disease resulted in greater harm from testing and false overdiagnosis. The number for MI/ACS is probably in the same ballpark, though with the problems of false + stress, it might even be worse. <1% is probably unobtainable unless we stress every single patient in the ED with vague complaints.

    The cardiologists will try to use AHA’s just released guidelines on low risk chest pain to try to convince us we should be sending home patients with even less use of validated risk strat than TIMI.

    Essentially, guidelines state that if no high risk features and normal ekg/enzymes, pt can go home for 72 hour stress. But when you look through the guidelines for evidence, there really isn;t any.

  • Thanks, Scott. I knew someone smarter than me would be able to address this question. Good to know that the new AHA guidelines are really based on consensus rather than rigorous studies. Keep up the great work at EMCrit blog.

  • Anonymous

    It may be unattainable overall to expect a <1% miss rate. However it shouldn't be unattainable in subgroups. I thought the miss rate for PE in low suspicion patients was <2%.

    In low risk ACS patients plus a physician gestalt might get you there.

    I hate to use TIMI alone because there are so many patients that have a 0 or 1 TIMI score that are still slam dunk admits to me. (And I am just about the avg on admitting chest pains in my ED)

  • Yes, I believe the miss rate for PE in Kline’s initial study in the Journal of Thrombosis and Haemostasis was <2% for low risk patients who were PERC negative.

    You make a good point about using the TIMI scores with caution. Worrisome historical elements should, of course, trump any TIMI score.

  • thank you

  • Hollander et al published several papers on ‘clear cut alternative dx’ and also in combination with timi that showed that this was still not enough to get people home without testing.
    Just found your blog and am definitely enjoying.

  • Hi Anna Marie: Looks like you were one of the authors on the paper! Thanks for the tip. Looks like: “The incidence of 30-day death, myocardial infarction, or revascularization for patients with a clinical impression of an alternative diagnosis and a TIMI score of 0 was 2.9% (95% confidence interval, 1.6%-5.0%).” That 2.9% is still way to high of a risk indeed.

    Campbell CF, Chang AM, Sease KL, Follansbee C, McCusker CM, Shofer FS, Hollander JE. Combining Thrombolysis in Myocardial Infarction risk score and clear-cut alternative diagnosis for chest pain risk stratification. Am J Emerg Med. 2009 Jan;27(1):37-42.

  • Why aspirin considered a risk factor in TIMI score ?

    • Great question. I don’t think anyone knows the answer to the “aspirin paradox”. One theory is the notion of aspirin resistance. Another theory is that those on aspirin already represent a higher risk cohort (presumably have known CAD). Hard to say since the TIMI database is a set of 16 heterogenous studies put together. See discussion:

  • Rob McNab

    The Aspirin question shows that the patient is having ua symptoms DESPITE being on an anti-platelet agent. (IE read- appropriate treatment.