The Effects of Nicotine on Re-endothelialization, Inflammation, and Neoatherosclerosis After Drug-Eluting Stent Implantation in a Porcine Model

Author's summary

The effects of tobacco consumption on in-stent microstructural changes after percutaneous coronary intervention with drug-eluting stent placement remain unclear. Our study demonstrated that nicotine, a crucial constituent of cigarette smoke, inhibited re-endothelialization and promoted inflammation and neointimal hyperplasia after stent implantation in a porcine model. Exposure to very low amounts of nicotine induced these in-stent atherosclerotic changes. Our preclinical study highlights that smoking after drug-eluting stent implantation may adversely affect neointima formation.

Abstract

Background and Objectives

Cigarette smoking is a major risk factor for atherosclerosis. Nicotine, a crucial constituent of tobacco, contributes to atherosclerosis development and progression. However, evidence of the association between nicotine and neointima formation is limited. We aimed to evaluate whether nicotine enhances neointimal hyperplasia in the native epicardial coronary arteries of pigs after percutaneous coronary intervention (PCI) with drug-eluting stents (DES).

Methods

After coronary angiography (CAG) and quantitative coronary angiography (QCA), we implanted 20 DES into 20 pigs allocated to 2 groups: no-nicotine (n=10) and nicotine (n=10) groups. Post-PCI CAG and QCA were performed immediately. Follow-up CAG, QCA, optical coherence tomography (OCT), and histopathological analyses were performed 2 months post-PCI.

Results

Despite intergroup similarities in the baseline QCA findings, OCT analysis showed that the nicotine group had a smaller mean stent and lumen areas, a larger mean neointimal area, greater percent area stenosis, and higher peri-strut fibrin and inflammation scores than the no-nicotine group. In immunofluorescence analysis, the nicotine group displayed higher expression of CD68 and α-smooth muscle actin but lower CD31 expression than the no-nicotine group.

Conclusions

Nicotine inhibited re-endothelialization and promoted inflammation and NIH after PCI with DES in a porcine model.

Graphical Abstract

INTRODUCTION

Tobacco use is one of the most crucial modifiable risk factors for cardiovascular disorders, such as coronary artery disease (CAD).¹⁾ Despite cumulative scientific data on the effect of smoking on clinical outcomes among patients with CAD,¹⁾ the mechanisms whereby tobacco use after percutaneous coronary intervention (PCI) with drug-eluting stents (DES) influence in-stent microstructural changes remain unclear. In one study, tobacco use did not increase neointimal thickness or area after DES implantation.²⁾ In contrast, continued smoking was associated with increased rates of neointimal hyperplasia (NIH) and strut malapposition in another study.³⁾ High-resolution optical coherence tomography (OCT) was used in both studies to provide accurate intracoronary images to evaluate the in-stent microstructure after stenting. However, the results should be interpreted cautiously due to the retrospective observational nature of these studies.

Nicotine, a pyridine chiral alkaloid, is a crucial constituent of tobacco leaves (Nicotiana tabacum L.). The harmful effects of smoking are mainly associated with the numerous compounds in tobacco smoke.⁴⁾ Nicotine has a detrimental effect by contributing to atherosclerosis development and progression.⁵⁾ Therefore, nicotine is a potential major contributor to CAD development in smokers,⁶⁾ and it may be associated with microstructural changes, such as in-stent restenosis (ISR), after interventional angioplasty.

In this study, we aimed to evaluate the effect of nicotine on in-stent microstructural changes after DES implantation in a porcine coronary model using coronary angiography (CAG), quantitative coronary angiography (QCA), OCT, and histopathological analyses.

METHODS

Ethical statement

The study protocol and all animal care procedures conformed to the Guide for the Care and Use of Laboratory Animals and the ARRIVE Guidelines for Reporting Animal Research. The Institutional Animal Care and Use Committee of Chonnam National University Hospital ratified this study (Approval No. CNUHIACUC-22027).

Study design

This animal study was an 8-week prospective observational trial conducted at Chonnam National University Hospital. We included 20 Yorkshire × Landrace F1 crossbred castrated male pigs weighing 15–20 kg, supplied by SB Farm Co. Ltd. (Naju, Korea). The experimental protocol is illustrated in Figure 1. A total of 20 commercial DES (SYNERGY™ XD; Boston Scientific, Marlborough, MA, USA) were implanted; all 20 male pigs underwent one DES implantation each at either the left anterior descending coronary artery or the right coronary artery. The subjects were allocated into no-nicotine and nicotine groups with 10 stents each. The nicotine group, which simulated patients who continued smoking after stenting (i.e., persistent smokers), received continuous nicotine administration after stenting. The no-nicotine group, which simulated nonsmokers, did not receive continuous nicotine administration. The total daily dose of nicotine was 0.5 mg/kg/day, as described in an animal study conducted by Kim et al.⁷⁾ All subjects were followed up for two months (10 pigs in each group).

Figure 1. Flowchart of the animal experiment.

CAG = coronary angiography; LAD = left anterior descending coronary artery; OCT = optical coherence tomography; QCA = quantitative coronary angiography; RCA = right coronary artery.

Experimental process

Information regarding the experimental process is detailed in Supplementary Data 1.

Osmotic minipump implantation

This animal experiment was designed based on previous animal studies that used osmotic minipumps for continuous nicotine administration.⁸⁾ We employed a continuous nicotine-delivery method using an Alzet osmotic minipump (model 2ML4; Durect Corporation, Cupertino, CA, USA) comprising a core containing a drug material, an osmogen, and a semi-permeable membrane coating for drug delivery to ensure controlled, continuous drug delivery to experimental animals.⁸⁾ The osmotic minipump was filled with an appropriate nicotine dose and primed with sterile saline for at least 4 hours before implantation. After DES implantation and successful completion of post-stenting evaluations, the pump was subsequently implanted subcutaneously at the incision site near the carotid artery, where the 6-French sheath was inserted (Supplementary Figure 1A-C). After successful pump implantation, the incision site was closed with a surgical suture (Supplementary Figure 1D-F). We planned to implant the osmotic minipumps twice, once on day 0 and once on day 28, because the nominal duration for the Alzet osmotic minipump was 28 days (4 weeks) (Supplementary Figure 1G).

In vitro experiments for behaviors and patterns of nicotine release

A pilot in vitro study was conducted to determine the blood concentration of nicotine in a porcine model conducive to the implantation of an osmotic minipump filled with nicotine. The in vitro study methods are provided in Supplementary Data 2.

Quantitative coronary angiogram analysis

We evaluated all coronary segments intended for stenting using QCA by 2 experienced examiners blinded to each other’s visual assessments. Using a Computer-Aided Analysis and Synthesis QCA Workstation device (Pie Medical Imaging, Maastricht, The Netherlands), we measured the maximum, minimum, and mean vessel diameters before and after stenting and during the follow-up interval. The minimum vessel diameter is the smallest diameter of the lumen, while the maximum vessel diameter is the largest. The in-stent diameter stenosis was calculated as 1 − (Follow-up In-stent Minimum Vessel Diameter Within the Stented Segment/Follow-up Reference Vessel Diameter Within the Stented Segment) × 100. In-stent late lumen loss was calculated as Post-stenting Minimum Vessel Diameter Within the Stented Segment − Follow-up In-stent Minimum Vessel Diameter Within the Stented Segment (Supplementary Figure 2).

Optical coherence tomography

We performed all OCT studies using an OCT Frequency Domain system (dragonfly; Light Lab Imaging Inc., Westford, MA, USA) to produce high-resolution, cross-sectional intracoronary images. At follow-up, OCT was used to identify stent optimization during the post-stenting period and microstructural changes, including neointimal responses. Contrast flushing was achieved using a Luer-Lock purge syringe filled with a radiocontrast medium (Omnihexol 350; Korea United Pharm Inc., Seoul, Korea) to avoid light backscattering. After advancing the 6-French coronary guide catheter into the ostium of the desired vessel and engaging the coronary guidewire in the coronary segment distal to the stented site, the OCT catheter was crossed along this guidewire and positioned where it was intended to be scanned. Under the cine angiography, an automated contrast injection device injected a total bolus of 20–30 mL contrast agent into the coronary artery at a pressure of 300 psi/s. Approximately 50-mm intracoronary OCT images were acquired with an anatomic pullback of 20 mm/s.

The OCT images were analyzed at 1-mm longitudinal intervals. For each OCT image, the stent and lumen areas were measured quantitatively. The neointimal area was calculated by subtracting the luminal area from the stent area. The percentage stenosis is the luminal area divided by the stent area. We reported the mean values of each OCT parameter for all the OCT-based quantitative analyses.

Histopathological analysis

We analyzed all histopathological slices using a calibrated microscope, a digital video imaging device, and special software (Visus 2000 Visual Image Analysis System; IMT Tech., San Diego, CA, USA). The specimen processing methods for histopathological analysis are detailed in Supplementary Data 3.

The degree of arterial injury, inflammatory infiltration, and fibrin deposition were assessed using a semiquantitative analysis.⁹⁾ Arterial injury at each strut site was evaluated based on the anatomical structures penetrated by each strut, with numeric values assigned from 0 to 3: 0 for no injury, 1 for a break in the internal elastic membrane, 2 for media perforation, and 3 for perforation of the external elastic membrane extending to the adventitia (most severe injury). The average injury score for each segment was determined by dividing the sum of the injury scores by the total number of struts in the examined section. Inflammatory infiltration was graded as follows: 0 for no inflammatory cells around the strut, 1 for light, non-circumferential lymphohistiocytic infiltrates, 2 for localized, non-circumferential moderate to dense cellular aggregates around the strut, and 3 for circumferential dense lymphohistiocytic cell infiltration. Fibrin deposition was graded as follows: 0 for no fibrin around the strut, 1 for moderate fibrin deposition involving less than one-quarter of the artery circumference, 2 for either moderate fibrin deposition involving over one-quarter of the artery circumference or severe fibrin deposition involving less than one-quarter of the artery circumference, and 3 for severe fibrin deposition involving over one-quarter of the artery circumference. The inflammation and fibrin scores for each cross-section were calculated similarly to the injury score.⁹⁾

Next, the specimens were fluorescently labeled against interest proteins, including the cluster of differentiation 31 (CD31), CD68, and alpha-smooth muscle actin (α-SMA), to assess the role of nicotine in neointima formation after stenting. The fluorescent staining (CD31, CD68, and α-SMA) methods are summarized in Supplementary Data 4.

Statistical analysis

The differences in outcomes between both groups (i.e., no-nicotine group vs. nicotine group) were evaluated through statistical analysis. All continuous values were described as means with standard deviations or medians with interquartile ranges and were analyzed using the independent 2-sample Student’s t-test, Mann-Whitney U test, and one-way analysis of variance. The cut-off p-value for statistical significance was set at p<0.05. All statistical analyses were conducted using SPSS version 25.0 (SPSS Inc., Chicago, IL, USA). The datasets used and analyzed during the current study are available from the corresponding author upon reasonable request.

RESULTS

Behaviors and patterns of nicotine release

Supplementary Figure 3 illustrates the pattern and behavior of nicotine release. After the implantation of an osmotic minipump filled with nicotine, the blood concentration of nicotine was estimated to be 3.410, 1.425, 1.904, 2.074, 3.807, and 0 ng/mL during sustained nicotine release from the osmotic minipump after 2, 4, 14, 21, 27, and 30 days, respectively.

Pre- and post-procedural quantitative coronary angiography assessments

We successfully implanted 20 DES in all the study subjects and included them in our statistical analysis. Table 1 lists the baseline characteristics of the QCA results before and after stent implantation, which showed no significant intergroup differences.

Table 1. Pre- and postprocedural QCA assessments of porcine coronary arteries.

Variables		No nicotine group (n=10)	Nicotine group (n=10)	p value
QCA before stenting (mm)
	Maximum	2.92 (2.73–3.09)	3.01 (2.76–3.27)	0.796
	Minimum	2.16 (1.93–2.38)	1.85 (1.76–2.12)	0.075
	Mean	2.51 (2.46–2.72)	2.50 (2.32–2.67)	0.684
QCA after stenting (mm)
	Stent diameter	3.00 (3.00–3.50)	3.00 (2.81–3.50)	0.911
	Stent length	16.00 (16.00–19.00)	17.50 (16.00–23.00)	0.315
	Maximum	3.30 (3.17–3.68)	3.28 (3.04–3.56)	0.492
	Minimum	2.64 (2.27–2.83)	2.57 (2.45–2.61)	0.869
	Mean	2.97 (2.87–3.39)	2.96 (2.85–3.17)	0.782

Values are presented as medians with interquartile range for continuous values.

QCA = quantitative coronary angiography.

Quantitative coronary angiography and optical coherence tomography assessments 8 weeks after drug-eluting stents implantation

Follow-up evaluations, including CAG, QCA, OCT, and histopathological analyses, were performed 8 weeks after stent implantation (Table 2). In QCA analysis (Item A of Table 2), the nicotine group tended to show significantly lower maximum (3.12 [3.03–3.29] vs. 2.81 [2.63–3.02], p=0.030), minimum (2.15 [1.97–2.43] vs. 1.65 [1.49–1.75], p=0.012), and mean lumen diameter (2.64 [2.57–2.87] vs. 2.26 [2.10–2.55], p=0.018) and significantly greater in-stent late lumen loss (0.49 [0.08–0.77] vs. 0.94 [0.67–1.01], p=0.027) than the no-nicotine group. In OCT analysis (Item A of Table 2, Supplementary Figure 4), the nicotine group had a lower mean stent (7.55±1.88 vs. 7.23±1.35, p<0.001) and lumen areas (5.92±1.74 vs. 4.74±1.72, p<0.001) but higher mean neointimal area (1.63±0.94 vs. 2.49±1.19, p<0.001) and percent stenosis (21.78±11.93 vs. 35.07±15.98, p<0.001 in the cross-section-based analysis, 19.05 [13.71–27.42] vs. 34.04 [23.04–44.92], p=0.019 in the stent-based analysis) than the no-nicotine group. The nicotine group had a higher percentage of percent stenosis >40% than the no-nicotine group (12.3% vs. 37.0%, p<0.001 in the cross-section-based analysis, 0.0% vs. 30.0%, p=0.067 in the stent-based analysis).

Table 2. QCA and OCT assessments and morphometric measurements of porcine coronary arteries at 8 weeks after stenting.

Variables			No nicotine group (n=10)	Nicotine group (n=10)	p value
A. QCA and OCT assessment of porcine coronary arteries at 8 weeks after stenting
	QCA at 2-month follow-up
		Maximum (mm)	3.12 (3.03–3.29)	2.81 (2.63–3.02)	0.030
		Minimum (mm)	2.15 (1.97–2.43)	1.65 (1.49–1.75)	0.012
		Mean (mm)	2.64 (2.57–2.87)	2.26 (2.10–2.55)	0.018
		In-stent late lumen loss (mm)	0.49 (0.08–0.77)	0.94 (0.67–1.01)	0.027
		In-stent diameter stenosis (%)	21.63 (18.01–25.60)	27.37 (19.02–32.15)	0.124
	OCT at 2-month follow-up
		Total cross-sections	1,659	1,852
		Number of cross-sections per stent	165.90±42.94	185.20±44.83	0.339
	Cross-section-based analysis		(n=1,659)	(n=1,852)
		Mean stent area (mm2)	7.55±1.88	7.23±1.35	0.002
		Mean lumen area (mm2)	5.92±1.74	4.74±1.72	<0.001
		Mean neointimal area (mm2)	1.63±0.94	2.49±1.19	<0.001
		Percent stenosis (%)	21.78±11.93	35.07±15.98	<0.001
		Percent stenosis >40% (%)	204 (12.3)	686 (37.0)	<0.001
	Stent-based analysis		(n=10)	(n=10)
		Percent stenosis (%)	19.05 (13.71–27.42)	34.04 (23.04–44.92)	0.019
		Percent stenosis >40% (%)	0 (0.0)	3 (30.0)	0.067
B. The morphometric measurements of porcine coronary arteries at 8 weeks after stenting
	Histopathological assessment		(n=63)	(n=66)
		IEL (mm2)	5.81 (4.39–7.79)	5.47 (4.51–6.92)	0.736
		Lumen area (mm2)	4.36 (3.17–5.33)	3.35 (2.89–4.58)	0.023
		Neointima area (mm2)	1.29 (0.60–2.33)	1.85 (1.26–2.65)	0.004
		Percent stenosis (%)	22.33 (13.84–33.45)	33.33 (22.12–44.59)	0.001
		Injury score	1.00 (1.00–1.00)	1.00 (1.00–1.10)	0.284
		Inflammation score	0.40 (0.20–0.50)	0.50 (0.38–0.80)	<0.001
		Fibrin score	1.00 (1.00–2.00)	2.00 (1.00–2.00)	0.006
	Immunofluorescence analysis
		CD68 expression density (µm2/mm2)	10.00 (10.00–68.49)	10.00 (10.00–72.49)	0.030
			(n=38)	(n=50)
		α-SMA length (µm)	121.10 (80.18–187.27)	165.69 (124.64–308.20)	0.001
			(n=40)	(n=49)
		CD31 expression density (µm2/mm2)	1.81 (0.42–6.41)	0.36 (0.14–1.49)	<0.001
			(n=49)	(n=50)

Values are presented as percentages (numbers) for categorical values and as means ± standard deviations or medians with interquartile range for continuous values.

α-SMA = alpha-smooth muscle actin; CD31 = cluster of differentiation 31; CD68 = cluster of differentiation 68; IEL = internal elastic lamina; OCT = optical coherence tomography; QCA = quantitative coronary angiography.

Histopathological assessments 8 weeks after drug-eluting stent implantation

In histopathological evaluation (Item B of Table 2, Figures 2 and 3), the nicotine group had a significantly larger neointima area (1.29 [0.60–2.33] vs. 1.85 [1.26–2.65], p=0.004) and significantly greater percent stenosis (22.33 [13.84–33.45] vs. 33.33 [22.12–44.59], p=0.001) with higher fibrin (1.00 [1.00–2.00] vs. 2.00 [1.00–2.00], p=0.006) and inflammation (0.40±0.19 vs. 0.60±0.28, p<0.001) scores compared to the no-nicotine group. As shown in the representative photomicrographs of the section-morphometric analysis, the nicotine group demonstrated higher fibrin deposition around the stent strut than the no-nicotine group (Figure 2H, whitish arrows).

Figure 2. Representative images from sections-morphometric analysis at 8 weeks after stent implantation. (A-D) Nicotine group. (E-H) No-nicotine group. The nicotine group demonstrates higher fibrin deposition around the stent strut than the no-nicotine group (whitish arrows).

H&E = hematoxylin and eosin.

Figure 3. The morphometric measurements of porcine coronary arteries at 8 weeks after stenting. Compared to the no-nicotine group, the nicotine group shows a significantly larger neointima area and significantly greater percent stenosis with higher fibrin and inflammation scores.

IEL = internal elastic lamina.

In immunofluorescence analysis, the nicotine group displayed higher expression of CD68 (Item B of Table 2, Figure 4, white arrows) and α-SMA than the no-nicotine group (Item B of Table 2, Figure 5). CD31 expression was lower in the nicotine group than in the no-nicotine group (Item B of Table 2, Figure 6, whitish arrows).

Figure 4. Representative images of immunofluorescence of CD68. The nicotine group displays a higher number of CD68-positive reactive cells (whitish arrows) than the no-nicotine group.

CD68 = cluster of differentiation 68; DAPI = 4′,6 diamidino-2-phenylindole.

Figure 5. Representative images of immunofluorescence of α-SMA. The nicotine group displays a higher number of α-SMA-positive reactive cells than the no-nicotine group.

α-SMA = alpha-smooth muscle actin; DAPI = 4′,6 diamidino-2-phenylindole.

Figure 6. Representative images of immunofluorescence of CD31. CD31 expression is lower in the nicotine group than in the no-nicotine group (whitish arrows).

CD31 = cluster of differentiation 31; DAPI = 4′,6 diamidino-2-phenylindole.

DISCUSSION

This study examined the effects of nicotine on neointimal formation after DES implantation in a porcine model. At the 2-month follow-up, nicotine administration induced a significant increase in the intimal thickness. At baseline, the nicotine and no-nicotine groups had statistically comparable QCA results. Most of the OCT and histopathological parameters differed notably between the two groups, demonstrating that the nicotine group exhibited greater NIH scores than the no-nicotine group.

Despite multi-directional innovations and advances in PCI technologies, some major clinical issues associated with PCI failure include ISR, stent thrombosis, and neoatherosclerosis. ISR develops through the migration and proliferation of vascular smooth muscle cells (VSMCs) in the synthetic phenotype,¹⁰⁾ and the accumulation and oxidation of lipid particles in the intima,¹¹⁾ thereby forming foam cells, which induce NIH.¹²⁾ Relative to the use of bare-metal stents, the increased utilization of DES considerably reduces the incidence of NIH.¹³⁾ However, hypersensitivity, local inflammation, and delayed arterial healing have emerged as other main obstacles to neointima formation in the development of ISR in DES.¹⁴⁾ The antiproliferative agent in DES inhibits VSMC proliferation and attenuates NIH, and also tends to interfere with the re-endothelialization of the in-stent lumen, leading to adverse cardiac events.¹⁵⁾ In addition, some studies have demonstrated that local inflammation due to hypersensitivity to metal platforms, polymers, or drugs may cause ISR.¹⁶⁾

Tobacco use has traditionally been recognized as a crucial, modifiable risk factor for atherosclerosis and many cardiovascular disorders. Accumulating evidence suggests an association between smoking or smoking cessation and atherosclerosis. Hisamatsu et al.¹⁷⁾ demonstrated that tobacco use is strongly associated with atherosclerosis in multiple vascular beds. Notably, several observational cohort-based studies have demonstrated the long-term effect of persistent smoking on the outcomes in patients with CAD who had undergone PCI. A Chinese study emphasized that poor adherence to smoking cessation is a predictor of 3-year all-cause mortality and adverse cardiovascular events.¹⁾ According to the 5-year follow-up data from the SYNTAX trial, smoking influenced the adverse outcomes of patients treated with revascularization using either PCI or coronary artery bypass grafting.¹⁸⁾ Notably, some Asia-Pacific nationwide observational cohort studies have highlighted the harmful effects of smoking on clinical outcomes in patients with myocardial infarction.¹⁹⁾

Notwithstanding the cumulative evidence of the association between smoking status and cardiovascular outcomes detailed above, the influence of tobacco use on the in-stent microstructure after stenting remains debatable. According to an observational registry-based study by Yonetsu et al.,²⁰⁾ smoking is a significant predictor of neoatherosclerosis. An OCT-based comparative study found that current smokers had a higher degree of strut coverage than nonsmokers; however, these results were due to the higher incidence of NIH with a heterogeneous pattern.²⁾ This study showed no significant difference between the groups regarding neointimal thickness and mean neointimal area. In another OCT-based study, persistent smoking enhanced NIH progression and led to stent strut malapposition, whereas smoking cessation promoted post-stenting vascular healing.³⁾ These results suggest that tobacco use promotes NIH, enhances strut coverage in unstable and irregular patterns, and contributes to strut malapposition. Nonetheless, all these results were based on retrospective observational studies, and prospective studies on this topic are lacking.

In our prospective animal study, nicotine promoted NIH after stenting, with greater neointimal area and percent stenosis. The QCA results demonstrated that the nicotine group had significantly lower maximum, minimum, and mean lumen diameter and greater in-stent late lumen loss than the no-nicotine group. Also, the OCT results indicated that the nicotine group had lower mean stent and lumen areas but higher mean neointimal area and percent stenosis than the no-nicotine group. In the cross-section-based analysis, the nicotine group has a significantly higher incidence of percent stenosis >40% than the no-nicotine group. The difference between the groups was not statistically significant in the stent-based analysis (p=0.067); however, it should be noted that the incidence was 0.0% in the no-nicotine group compared to 30.0% in the nicotine group. Even if this finding did not reach statistical significance, it suggests the absence of significant restenosis in the no-nicotine group, indicating that nicotine exposure may contribute to higher rates of considerable restenosis and that avoiding nicotine may have a protective effect on stent patency.

Nicotine, one of the most abundant chemicals in tobacco, stimulates endothelial nicotinic acetylcholine receptors and modulates VSMC function, leading to their transformation into the synthetic type.²¹⁾ Nicotine induces VSMC migration and proliferation from the media to the intimal layer of the arterial wall,²²⁾ where they transform into foam cells with lipid and cholesterol accumulation in the intima and media, which contribute to the developing atherosclerotic plaques and the alteration of their stability.²³⁾ Thus, nicotine appears to promote atherosclerosis development. Our immunofluorescence data demonstrated that the nicotine group showed higher α-SMA expression than the no-nicotine group. Given that α-SMA is the definitive marker of the activated cardiac fibroblast population, which includes VSMCs,⁷⁾ this higher expression may reflect a heightened NIH induction. Though speculative, exophytic proliferation of VSMCs within the media layer may explain why the nicotine group had a lower mean stent area than the no-nicotine group.

Meanwhile, the effects of nicotine on NIH also appear to involve several other mechanisms, including vascular inflammation and impaired re-endothelialization. Histopathological analysis revealed higher levels of peri-strut fibrin deposition and inflammation in the nicotine group compared to the no-nicotine group, indicating nicotine’s proinflammatory effects on the vascular wall.⁵⁾ Immunofluorescence results further support the hypothesis of inflammation-induced neointimal growth. CD68 is highly expressed in human cells of macrophage lineages²⁴⁾; its higher distribution in the nicotine group may reflect an elevated inflammatory status around the strut tissues, reinforcing nicotine’s proinflammatory effects.⁵⁾ The nicotine group demonstrated lower CD31 expression than the no-nicotine group. These findings do not provide conclusive evidence regarding re-endothelialization. However, they suggest that nicotine may interfere with arterial healing, as CD31 is a reliable marker for the re-endothelialization of stented lesions.²⁵⁾ Endothelial progenitor cells (EPCs) maintain endothelial homeostasis.²⁶⁾ Nicotine can activate telomerase in EPCs through the PI3K/Akt pathway, thereby reducing their senescence²⁷⁾; however, our results indicate paradoxical effects. These beneficial effects are limited to short-term exposure,²⁸⁾ and long-term exposure to nicotine may reduce EPC count and impair their function, counteracting the short-term benefits.²⁶⁾ These detrimental effects align with the well-established concept that smoking impairs endothelial function,²⁸⁾ supporting our observations.

The dosage and route of nicotine administration in our experiment were considerably different from those observed with traditional tobacco products, including cigarettes, cigars, and pipes. For example, during cigarette smoking, nicotine is inhaled, not directly administered by intra-arterial injection. Other smoking devices, such as cigars, mini-cigars, and pipes, also involve nicotine inhalation. According to a study conducted by Benowitz et al.,⁶⁾ nicotine is rapidly absorbed from cigarette smoke, after which it enters the arterial circulation and is rapidly distributed to the body tissues. Furthermore, smokeless tobacco is associated with a more considerable amount of overall nicotine exposure because of prolonged absorption compared to cigarette smoking.⁶⁾ In summary, all tobacco products, including cigarette smoke and smokeless tobacco, appear to have adverse cardiovascular effects despite the indirect route of nicotine intake.

Meanwhile, our in vitro experiments showed that nicotine was continuously released at blood concentrations of <5 ng/mL. Considering the results of previous studies, these concentrations are much lower than those with tobacco smoking or chewing nicotine gum²⁹⁾ and rather similar to or lower than those with electronic cigarettes.²⁹⁾ Notably, the implantable infusion utilized in our study allows for continuous nicotine delivery, unlike traditional smoking methods previously discussed. Our experimental approach resulted in relatively low blood nicotine concentrations but ensured sustained nicotine exposure. This sustained exposure could potentially contribute to cumulative vascular nicotine toxicity, although this theory remains speculative.

At the 1-month follow-up, we performed a comparative analysis between groups to determine how early the intravascular effects of nicotine occur after DES implantation. The results of this analysis are detailed in Supplementary Table 1. Despite group-by-group differences in the mean stent, lumen, and neointima areas, both groups showed comparable percent stenosis, with zero proportion of percent stenosis >40%. These findings may be driven by the relatively short time required for phenotypic changes to occur in VSMCs. According to an animal experiment by Hahn et al.,³⁰⁾ this change occurred continuously over one month and up to 3 months after stenting. Thus, the duration of 4 weeks appears to be too short to evaluate nicotine-induced NIH.

Despite the detrimental cardiovascular effects of tobacco smoking, consensus regarding the association between nicotine and in-stent microstructural changes post-stenting is lacking. Our analysis is clinically important for several reasons. Most importantly, unlike previous studies, our animal experiments prospectively identified the effects of nicotine on re-endothelialization, inflammation, and neointima formation. Indeed, our study affirms the association between NIH and atherosclerosis, consistent with previous basic, translational, and clinical studies.⁷⁾

Our study is the first prospective observational trial demonstrating that nicotine inhibits re-endothelialization and promotes inflammation and neoatherosclerosis after DES implantation. In particular, our analysis underscores that even extremely low but continuous nicotine exposure may accelerate these pathological processes in the in-stent areas. Despite potential differences in the route of nicotine intake between our experiment and traditional tobacco products, this preclinical research suggests that persistent tobacco smoking may lead to ISR among patients who undergo PCI, thereby worsening their clinical outcomes.

This study had some limitations. First, an inherent limitation of this study is that the stent was implanted in the normal porcine coronary artery in an oversized manner, a scenario that is different from atherosclerotic coronary restenosis in humans. Second, selecting the artery to be stented was not based on web-based randomization, but on the suitability of QCA results for implanting the stent. Moreover, the limited resources available for animal studies were considered. Third, although two analysts were blinded by the use of anatomical naming when performing the OCT assessment, differences in the stent platform might have affected the analysis results. Furthermore, information on intra- or inter-observer variability was not included. Fourth, we controlled the order of procedures and measurements; however, we did not consider potential confounders such as complete blood count, systemic inflammatory markers, or cholesterol levels. In other words, we did not collect blood samples from the pigs or check systemic inflammatory markers or cholesterol levels during our experiment. Fifth, the follow-up interval was limited to 2 months, which is too short to observe nicotine-induced in-stent microstructural changes. Therefore, further long-term prospective and randomized controlled experimental studies are required to elucidate the effects of nicotine administration on microstructural changes in in-stent areas.

Nicotine, the main component of cigarette smoke, seems to inhibit re-endothelialization and exert proinflammatory effects on the vascular wall, thereby promoting NIH after stenting and contributing to ISR progression. Results from our preclinical study indicate that persistent tobacco smoking may lead to ISR among patients undergoing stent implantation, which worsens the clinical outcomes in such patients.

SUPPLEMENTARY MATERIALS

Supplementary Data 1

Details of the experimental process

Supplementary Data 2

In vitro experiments for behaviors and patterns of nicotine release

Supplementary Data 3

Specimen processing for histopathological analysis

Supplementary Data 4

Fluorescent staining methods (CD31, CD68, and α-SMA)

Supplementary Table 1

Morphometric measurements of porcine coronary arteries at 4 weeks after stenting

Supplementary Figure 1

Osmotic minipump implantation. (A-C) After stenting and post-stenting evaluations, the Alzet osmotic minipump (Model 2ML4; Durect Corporation, Cupertino, CA, USA) is implanted subcutaneously at the incision site near the carotid artery where the 6-French sheath is inserted. (D-F) After the pump implantation, the incision site is closed with the surgical suture. (G) Given that the nominal duration for the Alzet osmotic minipump is about 28 days (4 weeks), these osmotic minipumps are implanted again 28 days after stenting.

Supplementary Figure 2

Schematic diagrams of the quantitative coronary angiogram analysis.

Supplementary Figure 3

Behaviors and patterns of nicotine release.

Supplementary Figure 4

Representative optical coherence tomography images of the stented vessels for visualization of the porcine coronary model. (A) The no-nicotine group at 8-week follow-up. (B) The nicotine group at 8-week follow-up.