Performance of the 'International Criteria' for interpretation of athletes' ECGs in a Dutch cohort

The prevalence of false-positive ECGs in cardiovascular pre-participation screening has historically been high. These false positives screening outcomes lead to additional health care costs. The most recent criteria for interpretation of athletes’ resting ECGs are the ‘International Criteria’ published in 2017. These criteria aim to reduce the amount of false positive ECGs, while preserving sensitivity. To assess the performance of these criteria in daily practice, evaluation across diverse cohorts is imperative.

Our study aimed to assess the performance of the International Criteria within our Dutch cohort. The primary objective was to evaluate the impact of the International Criteria on the number of abnormal ECG’s, compared to the previous criteria (ESC 2010, Seattle, and Refined).

In this prospective observational cohort study all preparticipation screenings in a Dutch hospital between April 2016 and March 2023 were recorded in a database. Retrospectively, all ECGs were systematically re-evaluated. Statistical analysis was performed using McNemar's test for paired binary data.

A total of 1862 ECGs were included. This cohort included both elite and amateur athletes aged 12-35 years. Application of the International Criteria classified 36 ECGs (1.9%) as abnormal, whereas use of the ESC 2010, Seattle, and Refined criteria resulted in 888 (47.6%), 70 (3.7%), and 73 (3.9%) abnormal ECGs, respectively. Applying the International Criteria significantly reduced the number of abnormal ECGs compared to the previous criteria (all p<0.001). This reduction was mostly due to optimalisation of the definitions used in the borderline category. The sensitivity for known pathology within our cohort was 100% when using the International Criteria.

Notably, the newer definition for pathologic Q waves, as described in the Refined and International Criteria, did not decrease the prevalence of abnormal Q waves in our cohort. Abnormal Q waves were the most frequent observed abnormality, accounting for 70% of observed abnormalities in our female population. Another finding was the presence of an extreme axis in 18 athletes (0.97%). However, this is not considered an abnormal finding in the International Criteria.

Use of the International Criteria shows a significant and clinically relevant reduction of the number abnormal ECGs within this Dutch cohort, without an apparent reduction in sensitivity. One of the aspects with considerable potential for optimalization for future criteria appears to be the definition used for abnormal Q waves. Another potential improvement could be defining criteria for interpretation of ECGs with an extreme axis.

Amount of abnormal ECGs and sensitivity.