Syncope in the ED: What the Canadian Syncope Risk Score Adds

Chester "Chet" Shermer, MD, FACEP · November 19, 2026

Syncope in the ED: What the Canadian Syncope Risk Score Adds

After a negative ED syncope workup, the Canadian Syncope Risk Score sorts 30-day risk and monitoring time. What the derivation, validation, and 2026 US data show.

The retired farmer passed out standing through a long hymn in a hot, packed church, and his wife says he looked gray and sweaty for a few seconds before he went down. He has a coronary stent from six years ago. In bay 9 he is back to baseline, his pressures since triage run between 128 and 146 systolic, the ECG shows sinus rhythm with a normal axis and a QRS of 96 ms, and his first troponin is below the 99th percentile. The admitting team wants him on telemetry overnight because of the stent.

Bottom line: once the ED evaluation finds no serious cause, calculate the Canadian Syncope Risk Score (CSRS) and let it guide disposition and monitoring time. For this patient the score is -2, the very-low-risk category, and he can go home.

Who the Score Applies To

The CSRS predicts serious outcomes within 30 days that were not evident during the index ED evaluation. That framing matters. Patients with a serious diagnosis found in the ED were excluded from the derivation and validation cohorts, so the score starts only after your workup is done [1][2].

The derivation cohort enrolled 4,030 adults aged 16 or older who presented within 24 hours of syncope to 6 Canadian EDs. Exclusions were loss of consciousness longer than 5 minutes, a change in mental status from baseline, an obvious witnessed seizure, head trauma causing loss of consciousness, major trauma requiring admission, intoxication, and a language barrier. Serious adverse events occurred in 147 patients (3.6%) [1].

The nine predictors and their points [2]:

  • Predisposition to vasovagal symptoms (warm crowded place, prolonged standing, fear, emotion, or pain): -1
  • History of heart disease (coronary or valvular disease, cardiomyopathy, heart failure, nonsinus rhythm, or device): +1
  • Any systolic pressure below 90 or above 180 mm Hg from triage to disposition: +2
  • Troponin above the 99th percentile: +2
  • QRS axis below -30° or above 100°: +1
  • QRS duration longer than 130 ms: +1
  • QTc longer than 480 ms: +2
  • ED diagnosis of vasovagal syncope: -2
  • ED diagnosis of cardiac syncope: +2

Scores run from -3 to 11. In the derivation paper, a score of -2 or lower carried an estimated 30-day risk below 1% [1].

How It Held Up in Validation

The Canadian validation study enrolled 3,819 patients at 9 EDs with the same entry criteria. Serious outcomes occurred in 139 (3.6%), including 13 deaths. Discrimination was high (AUC 0.91), and calibration matched predicted risk [2]. Observed 30-day serious outcomes by category were:

  • Very low: 3 of 1,631 (0.2%)
  • Low: 9 of 1,254 (0.7%)
  • Medium: 55 of 687 (8.0%)
  • High: 32 of 167 (19.2%)
  • Very high: 40 of 78 (51.3%)

No very-low-risk or low-risk patient died or had a ventricular arrhythmia. At a threshold score of -1, sensitivity was 97.8% and specificity 44.3% [2]. The authors recommended discharge for very-low-risk and low-risk patients, shared decision-making for medium risk, and a short hospitalization for high risk [2].

Two weaknesses deserve attention. Troponin was missing for 41.0% of the validation cohort and was imputed as normal, and all sites were academic Canadian EDs [2].

Outside Canada, With Caveats

An international cohort of 2,283 patients aged 40 or older, enrolled within 12 hours of syncope at EDs in 8 countries, found an AUC of 0.85 for the primary composite of serious clinical and procedural events, which occurred in 7.2%. The CSRS beat the OESIL score (AUC 0.74), and the composite occurred in 19 of 1,388 (0.6%) CSRS low-risk patients versus 17 of 1,104 (1.5%) OESIL low-risk patients. One finding cuts against the score: clinician classification of the syncope type alone (cardiac, vasovagal, or other) reached an AUC of 0.83. The authors concluded that the clinical utility of the CSRS remains uncertain [3].

A 2026 US cohort from 6 urban EDs enrolled 1,263 patients aged 40 or older with syncope or presyncope and no serious ED diagnosis. Serious adverse outcomes occurred in 74 (5.9%) [4]. A CSRS below 0 classified 380 patients (30.1%) as low risk, with a sensitivity of 91.9% and a negative predictive value of 97.5%. The FAINT score (heart failure, arrhythmia history, abnormal initial ECG, elevated NT-proBNP, elevated high-sensitivity troponin T) equal to 0 had a sensitivity of 96.7% for serious cardiac outcomes. Attending physicians' unstructured estimates below 2% had a sensitivity of 87.5% for any serious outcome and 79.2% for serious cardiac outcomes. FAINT, but not the CSRS, was significantly more sensitive than physicians for cardiac outcomes [4].

That cohort is older and sicker than the Canadian one, and only 55.3% of its patients were discharged [4]. The score's low-risk group shrinks in that population, and its sensitivity drops below what the Canadian data promised.

How Long to Monitor

A 5,581-patient cohort answered the telemetry question. CSRS risk sorted 30-day arrhythmic outcomes at 0.4% for low risk, 8.7% for medium risk, and 25.3% for high risk. Half of the arrhythmic outcomes in low-risk patients appeared within 2 hours of ED arrival, and half in medium-risk and high-risk patients appeared within 6 hours. After those points the residual risk was 0.2% for low risk, 5.0% for medium risk, and 18.1% for high risk. No low-risk patient had a ventricular arrhythmia or unexplained death [5].

Among medium-risk and high-risk patients, 91.7% of arrhythmic outcomes, including every ventricular arrhythmia, surfaced within 15 days. The authors proposed 15 days of outpatient rhythm monitoring for selected medium-risk and all high-risk patients after discharge [5].

Serial troponin adds little after a single value. In 4,996 patients from the CSRS cohorts, 10 of the 499 with repeat testing crossed the 99th percentile, 9 were reclassified by the CSRS and none of those 9 had a 30-day serious event, and ED stays ran 5.6 versus 3.8 hours [6]. The authors reserved repeat troponin for ongoing symptoms or ECG findings suggesting ischemia [6].

The 2017 ACC/AHA/HRS guideline frames all of this in plain terms. Risk scores are "not primary determinants for admission," the presence of a serious medical condition is "the key determinant," hospital-based evaluation is "unlikely to provide benefit" for presumed reflex syncope without another dangerous condition, and a structured ED protocol "can be effective as an alternative to inpatient admission" [7].

Key Takeaways

  • The CSRS applies only after the ED workup finds no serious cause; it excludes LOC over 5 minutes, altered mental status, witnessed seizure, and head trauma with LOC.
  • In Canadian validation, 30-day serious outcomes were 0.2% at very low risk and 0.7% at low risk, with no deaths or ventricular arrhythmias in either group.
  • Outside Canada, sensitivity fell to 91.9% at a threshold below 0 in older US patients, so treat a low score as one input alongside your own read of the syncope type.
  • Match monitoring to the score: about 2 hours on the monitor for low risk and about 6 hours for medium and high risk, the windows that caught half of arrhythmic outcomes, then 15 days of outpatient rhythm monitoring for selected medium-risk and all high-risk patients after discharge.

FAQ

What is the Canadian Syncope Risk Score?

The CSRS is a nine-item score that predicts serious outcomes within 30 days after ED evaluation for syncope when no serious cause was found in the ED. It combines vasovagal predisposition, heart disease history, extreme systolic pressures, troponin, three ECG intervals or axis findings, and the ED diagnosis of vasovagal or cardiac syncope, for a total from -3 to 11 [1][2]. In Canadian validation, very-low-risk and low-risk patients had 30-day serious outcome rates of 0.2% and 0.7% [2].

Which syncope patients can be discharged from the ED?

Patients whose ED evaluation finds no serious condition and who score in the CSRS very-low-risk or low-risk categories were recommended for discharge by the validation authors [2]. The 2017 ACC/AHA/HRS guideline states that hospital-based evaluation is unlikely to help presumed reflex syncope without another dangerous condition [7]. Patients with a serious condition found in the ED, such as symptomatic Mobitz II or third-degree block, sustained VT, severe aortic stenosis, or severe anemia from GI bleeding, are outside the score and need in-hospital care [7].

How long should syncope patients be on a cardiac monitor in the ED?

In a 5,581-patient cohort, half of arrhythmic outcomes appeared within 2 hours of arrival in CSRS low-risk patients and within 6 hours in medium-risk and high-risk patients [5]. Residual 30-day arrhythmic risk after those windows was 0.2% for low risk. The same authors proposed 15 days of outpatient monitoring for selected medium-risk and all high-risk patients [5].

Dr. Chet's Take

I have worked syncope as an ED attending at a Level I academic center and as a TelEmergency medical director covering small rural departments, and the CSRS is the first syncope tool I would hand a new attending without apology. The article gets the main point right. The score answers a narrow question, what happens after a clean workup, and answers it well. The trap is using it to skip the workup. A score of -2 means nothing if nobody listened for the murmur, checked the stool, or looked hard at the ECG for a short PR or a Brugada pattern, because those patients never belonged in the cohort in the first place. Do the evaluation, then do the math. The core argument holds, and it lets you defend a discharge on paper in a way gestalt never could.

That being said, the heaviest weight in the score sits on the least reproducible item. Calling an event vasovagal or cardiac swings the total by four points, and that call is a judgment that two good physicians will make differently on the same patient. The honest answer is that I do not know how much of the score's performance is the score and how much is the physician who filled it in. That makes the history the procedure. In the rural sites I support, the stakes run the other way too. Overnight telemetry there means a transfer, an ambulance out of the county, and a family driving two hours. A clean low score is the strongest tool I have for keeping a stable patient home without guessing.

If you are the attending, write the score and its components in the chart, and name the syncope type you chose and why. If you are discharging a medium-risk or high-risk patient, make the outpatient rhythm monitor a real order with a named person who reads it and calls the patient, before the patient leaves the department. If you are covering a rural ED by telemedicine, put the score and the monitoring plan in the virtual consult note so the local nurse, the patient, and the primary care physician see the same number and the same plan. Tell the patient in plain words what brings them back: another faint, chest pain, palpitations, or symptoms with exertion. A monitor nobody reads is a discharge with extra steps. Syncope disposition is defensible when the reasoning is written down and the follow-up has an owner.

— Chester Shermer, MD, FACEP | Emergency Medicine, 25+ Years Clinical Experience | State Surgeon

Continue Your Training

If you're an emergency physician (or any clinician treating patients daily) trying to understand how AI will actually impact your clinical practice — not just the hype — I put together a free practical guide. You can download the AI in EM Survival Guide here.

Browse All Courses at Global MedOps Command

Relevant Reading on Global MedOps Command:

How to Avoid Becoming an AI Casualty — Dr. Shermer's guide to navigating AI tools in clinical and operational settings without compromising judgment or patient outcomes.

Emergency Department Efficiency Playbook — Practical systems for throughput, triage optimization, and operational efficiency.

Read more from Dr. Shermer on Medium

Connect with Dr. Shermer: LinkedIn — Chester "Chet" Shermer, MD, FACEP.

References

  1. Thiruganasambandamoorthy V, Kwong K, Wells GA, et al. Development of the Canadian Syncope Risk Score to predict serious adverse events after emergency department assessment of syncope. CMAJ. 2016;188(12):E289-E298. PubMed 27378464
  2. Thiruganasambandamoorthy V, Sivilotti MLA, Le Sage N, et al. Multicenter emergency department validation of the Canadian Syncope Risk Score. JAMA Intern Med. 2020;180(5):737-744. PubMed 32202605
  3. Zimmermann T, du Fay de Lavallaz J, Nestelberger T, et al. International validation of the Canadian Syncope Risk Score: a cohort study. Ann Intern Med. 2022;175(6):783-794. PubMed 35467933
  4. Suh EH, Winskill C, Sacco DL, et al. Validation of 2 syncope risk scores and comparison with physician risk estimation. JAMA Netw Open. 2026;9(5):e2617362. PubMed 42154462
  5. Thiruganasambandamoorthy V, Rowe BH, Sivilotti MLA, et al. Duration of electrocardiographic monitoring of emergency department patients with syncope. Circulation. 2019;139(11):1396-1406. PubMed 30661373
  6. Leafloor CW, McRae AD, Mercier E, et al. Utility of serial troponin testing for emergency department patients with syncope. CJEM. 2024;26(8):535-542. PubMed 39095575
  7. Shen WK, Sheldon RS, Benditt DG, et al. 2017 ACC/AHA/HRS guideline for the evaluation and management of patients with syncope. Circulation. 2017;136(5):e60-e122. PubMed 28280231

Related simulation training: EMS-MedSim · Books by Dr. Shermer

Related reading