Discussion
This study evaluated the correlation between the occurrence and morphology of VAs and LGE in patients with apparently normal hearts. The results demonstrated that LGE is more common in patients with VAs than those without VAs, especially in those with polymorphic VAs and ST-segment depression. Moreover, polymorphic VAs and ST-segment depression were associated with a 25.24-fold increase and an 8.83-fold increase in the occurrence of LV-LGE, respectively. The frequency and burden of VAs have no impact on the presence of LV-LGE. LV-LGE extent tended to increase in patients with polymorphic VAs.
Our study demonstrated that patients with VAs were more prone to show LGE which was similar to results seen in a previous study. Ghannam et al.18 reported that one-quarter of the patients with frequent PVCs were found to have myocardial scars. The presence of myocardial fibrosis represents an increased risk of death.1, 19 Moreover, a previous study reported that patients with PVCs had an odds ratio (OR) of 1.72 for the combined end-points of all-cause mortality, cardiovascular mortality, and sudden cardiac death compared with those without PVCs.20 This could be one reason why LGE was more common in patients with VAs. The present analysis showed that VAs frequencies and burdens have no impact on the presence of LGE. However, Adabag et al.14 reported that myocardial fibrosis, detected on CMR imaging in patients with hypertrophic cardiomyopathy, was associated with an increased frequency of ventricular tachyarrhythmias.14 That study showed that patients with LGE had a 7-fold higher risk of NSVT than patients without LGE. In our study, the constituent ratio of NSVT was 26.3% (5/19) versus 25.0% (8/32) in patients with and without LGE, respectively. A similar constituent ratio between the two groups could have been due to selection bias because of the small number of enrolled patients. Further studies with larger sample sizes are needed to clarify the association between LGE and NVST. Moreover, the lower VAs burdens of our study could also have contributed to the deviating results. Regarding VAs morphology and LGE occurrence, the results of this study showed that patients with polymorphisms and ST-segment depression were more likely to have LGE. A previous study21 reported that polymorphic PVCs (OR=4.25) were significantly associated with the presence of LGE, consistent with this study. Moreover, Niemann et al.22 reported LGE was eliminated when ST-segment alterations were absent. Our study also showed that patients with ST-segment depression had an 8.83-fold higher risk of having LV-LGE. In terms of the extent of LGE between subgroups, inconsistent results have been presented in previous studies.14, 18, 23Adabag et al.14 demonstrated that LGE was similar in patients with hypertrophic cardiomyopathy with and without PVCs, while Ghannam et al.18 reported that patients with inducible VT had larger fibrotic volumes compared with non-inducible patients. Similarly, our study showed no association between the extent of LV-LGE and the presence or burdens of VAs. Nevertheless, patients with polymorphic VAs tended to have increased LV-LGE extent. Most LV-LGE (20/23, 87.0%) were distributed in the midmyocardium, in this study. A previous study1 showed that LGE was also present in the midmyocardium and were associated with increased risks for sudden cardiac death, indicating that these patients should be followed up closely. In the multivariable analysis, polymorphisms and ST-segment depression were associated with an increased incidence of LV-LGE. Muser et al.21 reported that having multiple VAs morphologies is a risk factor for the presence of concealed myocardial structural abnormalities. Since they used CMR imaging and looked at LGE, the conclusions of that study support the results of our study. Some limitations should be acknowledged in this study. First, only a few cases had LGE (23/78) that could lead to biased results. In future studies, we plan to enroll a greater number of patients to strengthen the results. Second, due to the particularity of CMR, the study only focused on the influence of VAs on LV-LGE regardless of the influence on right ventricular LGE. Third, in morphologic terms, the study only analyzed the origins and ST-segments. Additional studies evaluating more ECG details, such as QRS complexes, amplitudes, and durations, are needed. Finally, the present study lacked follow-up regarding the effects of having LGE on patient prognosis. Our team will perform follow-up studies to further understand the influence of fibrosis in patients without known structural heart disease.