Limitations
This was a single centre study. Further validation of the prediction parameters by other centres would increase the robustness and clinical utility of these findings. The number of individuals included in the study is relatively low. However, this is comparable to similar studies using ECG-morphology based parameters.
In our cohort, the patients had relatively high mean ectopic burdens (16% for RVOT SOO, 24% for LVOT SOO). Consequently, we cannot be sure if these two parameters would work consistently in patients with significantly lower burdens of ectopy, as a low ectopic burden overall would be likely to increase the chance of any hour having < 50 VE. Nonetheless, it is encouraging that some patients in our cohort did have VE burdens as low as 0.5% and the prediction parameters were still diagnostic. Furthermore, despite there being a trend towards lower ectopic burdens in the RVOT cohort, which may bias towards having a single hour with < 50 VE, total VE burden was included in the multivariate model and was not found to be independently associated with SOO.
Finally, the patient population we used to derive these novel parameters had robust myocardial function with a mean LVEF of 53.7% and 57.9% for the RVOT and LVOT VE cohorts, respectively. In the setting of VE-induced cardiomyopathy with significant LV dysfunction, neuro-hormonal regulatory changes might alter the behaviour of the VEs, limiting the applicability of these parameters. This would require further assessment.