Level of Perivascular Inflammation Is Significantly Lower Around the Left Internal Mammary Artery Than Around Native Coronary Arteries

Abstract

Background: The left internal mammary artery (LIMA) is protected from developing atherosclerosis. Perivascular inflammation, which is closely associated with atherosclerosis, can be measured by perivascular adipose tissue attenuation on computed tomography angiography. Whether the absence of atherosclerosis in LIMA is related to the lower level of perivascular inflammation is unknown. This study was performed to compare the level of perivascular inflammation between LIMA in situ and native coronary arteries in patients with coronary artery disease.

Methods and results: A total of 573 patients who underwent both computed tomography angiography and optical coherence tomography imaging were included. The level of perivascular adipose tissue attenuation between LIMA in situ and coronary arteries was compared. Perivascular adipose tissue attenuation around LIMA in situ was significantly lower around the 3 coronary arteries (-82.9 [-87.3 to -78.0] versus -70.8 [-75.9 to -65.9]; P<0.001), irrespective of the level of pericoronary inflammation or the number of vulnerable features on optical coherence tomography. When patients were divided into high and low pericoronary inflammation groups, those in the high inflammation group had more target vessel failure (hazard ratio, 2.97 [95% CI, 1.16-7.59]; P=0.017).

Conclusions: The current study demonstrated that perivascular adipose tissue attenuation was significantly lower around LIMA in situ than around native coronary arteries. The lower level of perivascular inflammation may be related to the low prevalence of atherosclerosis in LIMA.

Registration: URL: https://www.clinicaltrials.gov; Unique Identifier: NCT04523194.

Keywords: computed tomography angiography; left internal mammary artery; optical coherence tomography; perivascular inflammation.

Figure 1. Study flowchart.

Between January 2011 and October 2022, 291 patients presenting with non–ST‐segment–elevation acute coronary syndrome (230 with non–ST‐segment–elevation myocardial infarction, 61 with unstable angina pectoris) and 355 with stable angina pectoris underwent both computed tomography angiography and optical coherence tomography imaging before intervention. Among patients with NSTE‐ACS, 2 were excluded for coronary spasm, 1 for myocardial infarction with nonobstructive coronary arteries, 1 for spontaneous coronary artery dissection, 5 for previous coronary artery bypass grafting, 3 for culprit lesions located in the left main coronary artery, 5 for culprit lesions located at side branches, 2 for in‐stent restenosis, 3 for no pre–percutaneous coronary intervention OCT images, 4 for hypoplasia of the left internal mammary artery, and 9 for poor image quality. Among patients with SAP, 6 were excluded for previous CABG, 1 for culprit lesion located in the LMCA, 9 for in‐stent restenosis, 5 for no pre‐PCI OCT images, 11 for hypoplasia of LIMA, and 6 for poor image quality. Thus, 256 patients with NSTE‐ACS and 317 patients with SAP were included in the final analysis. CABG indicates coronary artery bypass; LIMA, left internal mammary artery; LMCA left main coronary artery; MINOCA, myocardial infarction with nonobstructive coronary arteries; NSTE‐ACE, Non−segment‐elevation acute coronary syndrome; OCT, optical coherence tomography; PCI, percutaneous coronary intervention; SAP, stable angina pectoris; and SCAD, spontaneous coronary artery dissection.

Figure 2. PVAT attenuation around the LIMA in situ and native coronary arteries.

Values are median (25th–75th percentile). PVAT attenuation around LIMA in situ was significantly lower than the mean PVAT attenuation around the 3 coronary arteries (−82.9 [−87.3 to −78.0] vs −70.8 [−75.9 to −65.9]; P<0.001). LIMA indicates left internal mammary artery; and PVAT, perivascular adipose tissue.

Figure 3. PVAT attenuation around the LIMA in situ and native coronary arteries in patients with high and low levels of pericoronary inflammation.

Values are median (25th–75th percentile). Patients in the high pericoronary inflammation group were defined as those with mean PVAT attenuation around the 3 coronary arteries ≥−70.1 Hounsfield units (HUs), and those in the low group were defined as those with mean PVAT attenuation around the 3 coronary arteries <−70.1 HUs. PVAT attenuation around LIMA in situ was significantly lower than mean PVAT attenuation around the 3 coronary arteries in the high group (−82.9 [−86.9 to −78.0] vs −65.3 [−67.9 to −62.5]; P<0.001). In the low group, the same pattern was shown (−83.1 [−87.7 to −78.0] vs −75.3 [−78.9 to −72.3]; P<0.001). PVAT attenuation around LIMA in situ was not significantly different between high and low groups (−82.9 [−86.9 to −78.0] vs −83.1 [−87.7 to −78.0]; P=0.516). LIMA indicates left internal mammary artery; and PVAT, perivascular adipose tissue.

Figure 4. PVAT attenuation around the LIMA in situ and native coronary arteries in patients with vulnerable plaque features.

Values are median (25th–75th percentile). Patients were divided into 2 groups based on the number of vulnerable coronary plaque features as follows: group 1, from 0 to 3 features; and group 2, from 4 to 6 features. PVAT attenuation around LIMA in situ was significantly lower than mean PVAT attenuation around the 3 coronary arteries in both groups (group 1: −83.2 [−87.6 to −78.1] vs −71.6 [−76.9 to −66.9], P<0.001; group 2: −82.3 [−87.1 to −77.7] vs −69.7 [−74.8 to −64.8], P<0.001). Mean PVAT attenuation around the 3 coronary arteries was significantly lower in group 1 than in group 2 (−71.6 [−76.9 to −66.9] vs −69.7 [−74.8 to −64.8]; P<0.001). There was no significant difference in PVAT around LIMA in situ between these 2 groups (−83.2 [−87.6 to −78.1] vs −82.3 [−87.1 to −77.7]; P=0.255). LIMA indicates left internal mammary artery; and PVAT, perivascular adipose tissue.

Figure 5. The 2‐year cumulative incidence rate of TVF in patients with high and low levels of pericoronary inflammation.

TVF was defined as a composite of cardiac death, target vessel myocardial infarction, and unplanned ischemia‐driven target vessel revascularization. Patients in the high pericoronary inflammation group experienced more TVF than those in the low group (hazard ratio, 2.97 [95% CI, 1.16–7.59]; P=0.017). TVF indicates target vessel failure.