Author + information
- Zi Ye, MM,
- Yan Lai, MM,
- Gary S. Mintz, MD,
- Yian Yao, MM,
- Jiani Tang, MM,
- Yu Luo, MD,
- Jiming Li, MD,
- Yunkai Wang, MD,
- Junbo Ge, MD and
- Xuebo Liu, MD∗ ()
- ↵∗Shanghai East Hospital, Tongji University School of Medicine, Jimo Road 150, Shanghai 200120, China
Myocardial bridging (MB) is characterized by epicardial coronary artery tunneling through the myocardium, with angiographic “milking” and an intravascular ultrasound (IVUS) “half-moon” echolucent (1). Optical coherence tomography (OCT), a light-based technique, can provide unprecedented in vivo imaging of coronary vessel wall structure, especially of intima and plaque composition, with a high resolution of 10 μm. So far, there are no data on visualization of MB using OCT.
From November 2013 to July 2014, we prospectively identified 36 patients with angiographically “milking” of the left anterior descending artery (LAD) consistent with the diagnosis of MB. Both OCT and IVUS imaging were performed after patients gave informed consent. OCT was also performed at the most compressed site manually after automatic pullback. In all patients and in all MB segments corresponding to OCT and IVUS imaging, a heterogeneous, generally signal-poor fusiform area with a sharply delineated border closely surrounding the adventitia of the tunneled artery was detected with no gap between the fusiform area and the adventitia (Figure 1). In all patients, the fusiform area in the MB segment was continuous in the longitudinal OCT view, but the morphology changed depending on lumen compression. Most (33 of 36, 92%) MB segments showed a single fusiform area, whereas 3 showed 2 fusiform areas.
MB length, compression ratio (calculated as: [(1 − systolic external elastic membrane–cross-sectional area (CSA) divided by diastolic elastic membrane (EEM)-CSA) × 100%]), remodeling index (calculated as [diastolic external elastic membrane—CSA divided by the reference EEM-CSA]), and the thickness of “half-moon” echolucent area in the IVUS images and the thickness, length, and arc of the fusiform area in the OCT images were measured and calculated. The fusiform length was 15.6 ± 6.5 mm (range 10 to 18 mm) and appeared to depend on the MB length and heart rate. The thickness and arc of the fusiform zone (during diastole) measured 0.42 ± 0.11 mm (range 0.25 to 0.69 mm) and 176.3 ± 53.1° (range 106° to 287.5°), respectively. In addition, OCT detected atherosclerosis proximal to the MB segment in all patients, but only once in an MB segment in a patient with intimal thickening. The following data are normally distributed, and we used Pearson correlation in the correlation analysis. The fusiform thickness had no correlation with that of the IVUS-detectable echolucent band (R = −0.029; p = 0.934). Notably, the correlation between the compression ratio and the arc of the heterogeneous fusiform area at the most compressed tunneled artery was significant (R = −0.49; p = 0.003).
Histological examination using a pig heart model confirmed that the fusiform, heterogeneous signal-poor area was mainly composed of connective tissue and was located at the junction of the MB myocardium and epicardium.
The main findings of the present OCT study were as follows. 1) OCT detected a heterogeneous signal-poor fusiform area with well-delineated borders indicating the existence of the tunneled artery through the myocardium, but this was not the same as the echolucent band found by IVUS (“half-moon phenomenon”) that directly represented the myocardium surrounding the artery. 2) The fusiform area identified by OCT was more easily seen during cardiac diastole. 3) The severity of systolic MB compression was attenuated with an increase in the size of the fusiform arc. We documented these features as well as the histological correlations in a porcine model. According to previous histological evidence, the OCT fusiform area mainly consisted of loose connective tissue in the periarterial space beneath the MB.
Despite its higher resolution, OCT may not be the optimal imaging modality to detect a MB, mainly because of its limited penetration and rapid OCT fiber pullback and image acquisition (20 mm/s vs. 0.5 mm/s in IVUS). For this reason, we also performed OCT manually with the lens stationary at the MB segment.
In conclusion, in patients with MB documented angiographically and by IVUS, OCT detected a sharp border and heterogeneous, signal-poor fusiform area indicative of arterial tunneling through the myocardium that was distinct from the echolucent muscle band found on IVUS.
Please note: The authors have reported that they have no relationships relevant to the contents of this paper to disclose. Drs. Ye and Lai contributed equally to this work.
- American College of Cardiology Foundation