Is the Canadian Syncope Risk Score Valid?
By Michael H. Crawford, MD
Professor of Medicine, Lucy Stern Chair in Cardiology, University of California, San Francisco
SYNOPSIS: Researchers found the Canadian Syncope Risk Score accurately predicts which patients are low risk for discharge. However, since it is driven largely by the physician’s final risk classification at ED discharge, the score’s clinical utility is uncertain.
SOURCE: Zimmermann T, de Lavallaz JF, Nestelberger T, et al. International validation of the Canadian Syncope Risk Score: A cohort study. Ann Intern Med 2022;175:783-794.
Various risk scores to predict outcomes in patients with syncope have been developed, but none are used widely. To address this issue, researchers designed the Basel Syncope Evaluation study (BASEL IX), an international, multicenter investigation.
Zimmerman et al enrolled patients older than age 40 years presenting to the ED with syncope within the last 12 hours to 14 hospitals in eight countries. The authors excluded patients with nonsyncopal loss of consciousness or those without an entry ECG. Researchers followed clinical guidelines to evaluate patients, and the treating physician made the final diagnosis of the cause of syncope. The authors determined the Canadian Syncope Risk Score (CSR), which is comprised of nine characteristics from the clinical evaluation; ECG; troponin; and the physician’s classification of syncope as vasovagal, cardiac, or other at discharge. Also, they calculated the Osservatorio Epidemiologico sulla Sincope nel Lazio (OESIL) score, which is based on four predictors: older than age 65 years, known cardiovascular disease, no prodrome, and an abnormal ECG. The OESIL score has been shown and validated to predict one-year mortality rates.
Patients were followed periodically for 24 months. The primary composite outcome was composed of serious clinical and procedural outcomes, including death, life-threatening arrhythmia, myocardial infarction, aortic dissection, pulmonary embolism, and procedural interventions to abrogate syncope. The secondary outcome was the primary outcome minus procedural interventions. Three risk groups were created: low risk (CSR = 0; OESIL 0-1), medium risk (CSR = 1-3; OESIL = 2), and high risk (CSR = ≥ 4; OESIL = 3-4). Patients were collected between 2010 to 2019. A total of 2,283 were eligible for the analysis (mean age = 68 years; 42% women).
Overall, 54% of patients were hospitalized. The primary outcome occurred in 7.2% of patients at 30 days. The secondary outcome, excluding the procedural intervention, occurred in 3.1% of patients. Discrimination of the CSR for the primary outcome at 30 days exhibited an AUC of 0.85 (95% CI, 0.83-0.88) and was superior to the OESIL AUC of 0.74 (95% CI, 0.71-0.78; P < 0.001). The authors observed similar results for the secondary outcome CSR (AUC = 0.80; 95% CI, 0.75-0.84) vs. OESIL (AUC = 0.69; 95% CI, 0.64-0.75; P < 0.001). The CSR triaged 61% to low risk vs. 48% with the OESIL score. The 30-day outcomes happened less often in the low-risk CSR group (primary outcome = 1.1% vs. 2.7%; secondary outcome = 0.6% vs. 1.5%, respectively). A multivariable analysis of the individual items in the CSR revealed the largest odds ratios (OR) for the clinician classification of syncope at discharge from the ED (OR = 16.3; 95% CI, 10.8-24.7). The event rate after 30 days was lower compared to the 30-day rate, but remained higher in the high-risk group compared to the others. The authors concluded this study validates the CSR for identifying low-risk patients with syncope and is superior to the OESIL score. However, since most of the power of the CSR resides in the physician’s classification of syncope at ED discharge, the utility of the CSR is unclear.
This study established the value of the CSR for identifying low-risk syncope patients who may not need hospitalization. The CSR classified almost two-thirds of patients as low risk. Among these patients, the rate of a 30-day adverse outcome was 1%. The long-term follow-up data supported this finding, with even lower rates of the primary outcome at two years. The strength of the CSR was shown to be in the ED physician’s final classification of the patient as either vasovagal, cardiac, or other causes of syncope. Of note, the OR for the physician’s classification of syncope as cardiac for predicting the primary outcome was 16. The next closest OR was 2.1, for a QRS > 130. Naturally, physicians at the point of discharge from the ED knew the results of all studies performed during the ED visit. Thus, the clinical value of calculating this complex score is questionable.
The OESIL score was not as good at predicting who was at low risk. Among those graded as low risk by this score, 2.7% experienced adverse 30-day outcomes vs. 1.1% with the CSR. The OESIL includes three clinical parameters and one ECG finding. It does not include the physician’s final diagnostic classification. Therefore, it is simple and does not depend on the physician’s experience or acumen. The patient cohort was relatively large for a syncope study. The authors focused on older patients in whom the use of decision support mechanisms is most useful because of the higher probability of serious underlying pathology in such patients. That said, these results may not apply to younger patients.
Elsewhere, one-third of patients fell into the intermediate group using the CSR and experienced intermediate outcomes compared to the low- and high-risk groups. It is unclear what to do with these patients. There may have been an ascertainment bias by including subsequent procedures as an outcome. Importantly, pacemaker placement was such an outcome and was the most common by far (36% by 30 days). However, after excluding procedures, the CSR still was good at predicting adverse outcomes and was superior to the OESIL score. In the final analysis, this study emphasizes the value of the final diagnosis made by a physician in triaging syncope patients. So far, no check box score has been shown to be superior to this.
Researchers found the Canadian Syncope Risk Score accurately predicts which patients are low risk for discharge. However, since it is driven largely by the physician’s final risk classification at ED discharge, the score’s clinical utility is uncertain.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.