Athletes with premature ventricular beats: correlation with cardiac MRI findings and validation of risk stratification algorithms outside tertiary sports cardiology screening centers
A Damele, V Carinci, P Di Donna, G Mascia, E Leci, G Casella, I PortoAbstract
Background
Premature ventricular beats (PVBs) are common in athletes and usually benign. When PVBs are the only abnormal finding, cardiac magnetic resonance (CMR) may reveal concealed myocardial abnormalities, including non-ischemic left ventricular (LV) scar. Current risk-stratification relies on clinical assessment, ECG, 24-hour Holter monitoring and exercise testing, with focus on PVB burden, morphology, complexity and response to exercise. However, most evidence comes from tertiary sports cardiology centers and highly selected competitive athletes.
Purpose
To evaluate whether simple clinical and electrocardiographic markers of risk for LV scar predict late gadolinium enhancement (LGE) in a real-world population of both competitive and non-competitive athletes.
Methods
We retrospectively enrolled 34 consecutive athletes referred to 2 sports cardiology centers because of incidentally detected or symptomatic PVBs, with no other clinical, electrocardiographic or echocardiographic evidence of heart disease, who underwent CMR with LGE imaging. Demographic and training data, 12-lead ECG, 24-hour Holter (PVB burden, morphology, complexity) and maximal exercise testing (exercise-induced PVBs and non-sustained ventricular tachycardia [NSVT]) were collected. Presence, pattern and distribution of LGE were assessed. Predictors of LV LGE were evaluated with uni- and multivariable logistic regression.
Results
LGE was present in 7/34 athletes (21%), mostly non-ischemic and located in infero/inferolateral segments; two showed a junctional pattern and one had transmural ischemic LGE. At univariable analysis, age ≥35 years and uncommon PVBs morphology (right bundle branch block [RBBB]-like pattern with intermediate/superior axis) were associated with LGE. In the multivariable model, only PVBs morphology remained independently associated (p = 0.04). Neither PVB burden nor complexity (couplets/NSVT) predicted LGE. Nevertheless, the PVB burden on 24-hour 12-lead Holter monitoring showed a non-significant trend suggesting that lower PVB burden may be associated with LGE (p = 0.45), consistent with previous studies.
Conclusions
In athletes evaluated outside tertiary centers, a simple ECG phenotype—RBBB-like PVBs with an intermediate/superior axis—identifies individuals at higher probability of LV scar and supports selective referral for CMR. In our cohort, LV non-ischemic scar was found predominantly in athletes older than 35 years, reinforcing age as a meaningful contextual marker in real-world practice. These findings endorse a tiered diagnostic workflow (ECG morphology, Holter, exercise testing) to target advanced imaging toward athletes with the highest pre-test probability of concealed myocardial disease. The clinical significance of isolated non-ischemic LGE is not always clear; therefore, when first-line tests are reassuring, incidental LGE on CMR may add uncertainty rather than actionable guidance.Baseline characteristicsPredictors of LV LGE