Drug- and distance-dependent peri-stent perivascular adipose tissue changes after implantation of drug-eluting stents
J Yin, Y Yu, W F Guo, C Yan, X Y Zhang, Y L Song, C G Li, D J Wu, M S Zeng, J B GeAbstract
Background
Drug-eluting stent (DES) implantation induces local vascular injury and inflammation, and drug elution may reach surrounding perivascular adipose tissue (PVAT). PVAT inflammation is linked to atherosclerotic activity and can be assessed non-invasively on coronary CT angiography (CCTA) using the fat attenuation index (FAI) and PVAT volume, but the longitudinal impact of DES on peri-stent PVAT remodelling remains unclear.
Purpose
To assess 1-year changes in peri-stent PVAT FAI and volume after DES implantation, explore distance-to-stent gradients, evaluate drug-related differences, and contextualize peri-stent findings using non-target lesion (NTL) perivascular fat attenuation on serial CCTA.
Methods
Single-centre retrospective cohort (2015-2022) of patients with baseline CCTA before DES implantation and 1-year follow-up CCTA. Peri-stent PVAT FAI (HU) and volume were quantified along the stent (in-stent) and an extended peri-stent segment (stent+/-5 mm), using two radial definitions: within one vessel diameter and within half a vessel diameter from the stent surface. Change in FAI was compared across DES drugs; multivariable linear regression adjusted for age, sex, smoking, hypertension, diabetes, LDL and stent length. Lesion-specific perivascular fat attenuation around NTLs was quantified on baseline and follow-up CCTA as an internal comparator of background coronary inflammatory status.
Results
576 patients (64.06+/-9.84 years; 72.05% male) had paired CCTA. Peri-stent FAI increased at 1 year for both in-stent and peri-stent segment PVAT across radial definitions (all P<0.0001). The increase was larger closer to the stent: in-stent change in FAI +17.17 HU within half diameter versus +8.13 HU within one diameter; peri-stent segment change in FAI +15.66 versus +7.17 HU, consistent with a proximity-dependent peri-stent gradient. PVAT volume decreased in parallel (e.g., in-stent: 1812+/-1228 to 772+/-571 mm3 within one diameter; P<0.0001). Change in FAI differed by eluted drug (in-stent, half-diameter median change [IQR]: everolimus 19.48 [11.81-27.93] HU, rapamycin 15.83 [7.35-24.98], zotarolimus 12.09 [2.79-21.38]; P<0.0001). After adjustment, everolimus was associated with larger change in FAI (beta 6.46, 95% CI 3.81-9.11; P<0.0001) and zotarolimus with smaller change in FAI (beta -6.11, 95% CI -9.14 to -3.08; P=0.0001). In contrast, NTL perivascular fat attenuation decreased over 1 year on serial CCTA (P<0.0001).
Conclusion(s)
DES implantation was associated with peri-stent PVAT remodelling characterised by increased FAI and reduced PVAT volume at 1 year. The FAI response showed a spatial gradient and drug dependence, while background NTL perivascular fat attenuation moved in the opposite direction, supporting peri-stent FAI as a marker of a local stent-tissue inflammatory signature rather than a global coronary inflammatory trend.