(P069) CURRENT DEFINITIONS TO IDENTIFY SUDDEN CARDIAC DEATH AS A CAUSE FOR CARDIAC ARREST ARE INACCURATE: LIMITATIONS OF EMS RECORDS

Background: While Sudden Cardiac Death (SCD) remains a leading cause of death worldwide, differing definitions may result in unreliable estimates of incidence and outcomes. The term SCD refers to an unexpected cardiac arrest leading to death, due to an underlying cardiac/arrhythmic etiology. Most consensus documents define this phenomenon with time-based criteria (e.g., “WHO definition”). Without autopsies, often only EMS data are available in young patients to adjudicate SCD, and little is known about their accuracy. Our aim was to determine the accuracy of expert opinion and time-based criteria using EMS data when compared with comprehensive adjudication using autopsy data to determine SCD in young adults suffering an Out-of-Hospital Cardiac Arrest (OHCA).

METHODS AND RESULTS: Patients aged 18-45 years who had EMS attended OHCA in a large urban area between 2011-2012, designated as “no obvious cause” as per the Utstein criteria, were retrospectively evaluated for SCD by expert opinion and time-based criteria. Using available EMS data, a determination of “cardiac” vs “non-cardiac” sudden death was made by expert opinion or classified as “cannot determine”; consensus of 2 or more experts was used to categorize ambiguous cases. Time-based criteria were met in patients who had an OHCA within < 1h of symptom onset or in patients found dead < 24h of being asymptomatic, and classified as “cannot determine” if no information was available on timing. The gold-standard for determining the final cause of death was the use of autopsy, toxicology, and coroner investigation data. Sensitivity and specificity were calculated with the assumption of “cannot determine” being a false negative for cardiac cause and false positive for non-cardiac cause, respectively. Of 159 study patients, 53% had autopsy/toxicology proven SCD. Clinical characteristics were similar between autopsy-confirmed “cardiac” and “non-cardiac” sudden death. Table 1 outlines diagnostic test characteristics of expert opinion and time-based criteria in determining SCD when compared with gold-standard adjudication. Expert opinion had a sensitivity of 45.2% (29.4-61.0; 95% CI), and specificity of 56.1% (40.1-72.1; 95% CI). When expert opinion determined etiology, the specificity increased to 97.4% (86.2-99.99; 95% CI). By comparison, time-based criteria had a sensitivity of 66.7% (54.4-79.0; 95% CI) and specificity of 50.0% (33.0-67.0; 95% CI). These values were similar when excluding “cannot determine” from the calculation.

Conclusion: In the absence of autopsy data, expert opinion and time-based criteria perform poorly in identifying SCD in young adults suffering OHCA. Definitions used to identify SCD in the literature may be inaccurate in approximately half of patients.