Clinical background
Syncope is a common presenting complaint in the emergency department in which the initial assessment often reveals no serious cause. A proportion of patients nonetheless suffer a serious event within 30 days, including arrhythmia, myocardial infarction, structural heart disease, pulmonary embolism and death. The decision to discharge a patient from the emergency department therefore becomes a trade-off between not missing a serious cause and not admitting patients unnecessarily. The Canadian Syncope Risk Score was developed to provide structured support for precisely this decision, with a particular focus on identifying patients whose risk is low enough for discharge to be safe.
Calculating the Canadian Syncope Risk Score
The score rests on nine variables collected during the emergency department visit: the history, clinical examination, the ECG, troponin, and the physician's assessment of the likely mechanism of the syncope. Two variables are negative, that is, they lower the score and hence the risk: a predisposition to vasovagal symptoms, and a diagnosis of vasovagal syncope in the emergency department. The other seven variables add points and reflect cardiovascular risk.
The score ranges from −3 to +11.
The derivation cohort consisted of 4,030 adult patients (≥16 years) presenting with syncope within 24 hours at six large Canadian emergency departments between September 2010 and February 2014 [1]. The mean age was 53.6 years, 55.5% were women, and 9.5% were admitted. The primary outcome was an adjudicated serious event within 30 days, which occurred in 147 patients (3.6%). Of 43 candidate predictors, nine were selected for the final model, which showed good discrimination with a c-statistic of 0.88 (95% CI 0.85 to 0.90) and good calibration (goodness-of-fit p = 0.11) [1].
Interpretation in practice
The score translates into risk categories with different recommendations for action. The category boundaries come from a pooled analysis of the derivation and validation cohorts comprising 8,233 patients [3].
| Risk category | Score | Observed 30-day risk | Recommended action |
|---|---|---|---|
| Very low risk | −3 to −2 | 0.3% | Discharge |
| Low risk | −1 to 0 | ≤1.0% | Discharge |
| Medium risk | +1 to +3 | 7.8% | Individual discussion with the patient; consider brief observation or investigation |
| High risk | +4 to +5 | >20% | Hospital admission |
| Very high risk | ≥+6 | 42.7% | Hospital admission |
In the pooled cohort, no patient in the very low and low risk groups died of a ventricular arrhythmia, and only one death (from sepsis) occurred in these groups [3]. This strengthens the role of the score as a tool for identifying patients who can be discharged safely.
The threshold of −1 has been studied particularly as a cut-off for low risk. In the validation cohort this threshold gave a sensitivity of 97.8% (95% CI 93.8 to 99.6%) and a specificity of 44.3% (95% CI 42.7 to 45.9%) [2]. In the derivation cohort the sensitivity was 97.7% at the same threshold [1].
Validation and performance
The large external validation was carried out prospectively at nine Canadian emergency departments between March 2014 and April 2018 and comprised 3,819 patients with syncope [2]. The mean age was 53.9 years and 54.7% were women. Serious events within 30 days occurred in 139 patients (3.6%), 13 of whom died. The AUC was 0.91 (95% CI 0.88 to 0.93), on a par with the derivation cohort. Calibration was excellent, with a calibration slope of 1.0 and no significant difference between predicted and observed risk (p = 0.26). The proportion of patients with serious events rose from 0.3% in the very low risk group to 51.3% in the very high risk group [2].
An American prospective validation at six emergency departments (2020 to 2024) comprising 1,263 patients aged ≥40 years with syncope or presyncope showed, by contrast, poorer performance [4]. The AUC for serious events was 0.72 (95% CI 0.67 to 0.78). At a threshold of CSRS <0, sensitivity was 91.9% (95% CI 85.7 to 98.1%) and the negative predictive value 97.5%. In this cohort the median age was higher (66 years) and the proportion with serious events greater (5.9%). The FAINT score, an alternative instrument for patients aged ≥40 years, was more sensitive than unstructured physician risk assessment, whereas the CSRS was not [4].
A systematic review and meta-analysis from 2025 identified the CSRS as one of three clinical decision rules for syncope validated in more than two studies [5]. The CSRS had been validated in five studies with a positive likelihood ratio of 1.15 to 2.58 and a negative likelihood ratio of 0.05 to 0.50. The review notes that the quality of evidence for most decision rules is low, but that the CSRS is among the best validated.
Limitations
The score applies to patients in whom no serious cause of the syncope is identified during the initial emergency department assessment. If a serious diagnosis is evident at the index visit — for example aortic dissection, gastrointestinal bleeding or acute myocardial infarction — the patient should be managed according to that diagnosis and not risk-stratified with the CSRS.
Two of the variables require clinical judgement: a predisposition to vasovagal symptoms, and a diagnosis of vasovagal or cardiac syncope in the emergency department. These are subjective and can vary between assessors, which reduces reproducibility. In the American validation, in which the patients were older and the event rate higher, discrimination was lower than in the Canadian studies [4]. This suggests that the score may perform less well in populations that are older and have a higher prevalence of cardiovascular disease than the original cohort.
Nor has the score been evaluated in children under 16 years or, in the original studies, in patients with presyncope without complete loss of consciousness, although the American validation did include presyncope [4]. The troponin value is defined against the local assay's upper limit of normal, which means that different laboratories may classify the same patient differently.
References
- Thiruganasambandamoorthy V et al. Development of the Canadian Syncope Risk Score to predict serious adverse events after emergency department assessment of syncope. CMAJ 2016. PMID: 27378464
- Thiruganasambandamoorthy V et al. Multicenter Emergency Department Validation of the Canadian Syncope Risk Score. JAMA Intern Med 2020. PMID: 32202605
- Thiruganasambandamoorthy V et al. Personalised risk prediction following emergency department assessment for syncope. Emerg Med J 2022. PMID: 34740890
- Suh EH et al. Validation of 2 Syncope Risk Scores and Comparison With Physician Risk Estimation. JAMA Netw Open 2026. PMID: 42154462
- Wakai A et al. Risk-stratification tools for emergency department patients with syncope: A systematic review and meta-analysis of direct evidence for SAEM GRACE. Acad Emerg Med 2025. PMID: 39496561
