B lymphocytes and hyperglycemia synergistically exacerbate coronary in-stent restenosis

Abstract

Background

Lymphocytes are tightly associated with coronary artery disease (CAD). However, there is a lack of clinical evidence that lymphocyte subsets are associated with in-stent restenosis (ISR). The aim of this clinical research was to elucidate the association between these.

Methods

A total of 812 patients were enrolled in the study, and they were categorized into the non-ISR group (N = 575) and the ISR group (N = 237). For each patient, the counts and percentage of CD4⁺T cell, CD8⁺T cell, NK cell and B cell in plasma were determined, as well asimmunoglobulin A (IgA), IgG and IgM.

Results

Compared with non-ISR group, ISR group had higher levels of B-cell% (P = 0.011), B-cell (P = 0.010), non-HDL (P = 0.006), and higher proportion of diabetes (P = 0.002). B cell% (OR 1.033, 95 %CI:1.002–1.065, P = 0.039), diabetes (OR 1.556, 95 %CI:1.134–2.134, P = 0.006), non high density lipoprotein cholesterol (non-HDL-C) (OR 1.210, 95 %CI:1.026–1.428, P = 0.024) were independent risk factors for ISR. The RCS curves showes both B cell% (P for nonlinear = 0.415; P for overall = 0.035) and non-HDL-C (P for nonlinear = 0.168; P for overall = 0.024) showed a linear upward trend consistent with ISR. Subgroup analyses suggested that high B cell% (B cell%≥12) promotes the hazardous effects of diabetes on ISR (P for interaction = 0.042), while chronic hyperglycaemia (HbA1c%≥6.6) exacerbates the ISR promoting effects of B cells(P for interaction = 0.04).

Conclusion

B cells, non-HDL-C, and diabetes are closely associated with the development of ISR, and high B cell% and Hyperglycemia reciprocally promote ISR.

1. Introduction

In recent years, the epidemiology of coronary artery disease (CAD) has become younger and more widespread [1]. At the same time, with the popularity of percutaneous coronary intervention (PCI) technology, the number of patients undergoing drug-eluting stent implantation is increasing globally each year [2]. However, despite advances in stent technology and antiplatelet therapy, coronary in-stent restenosis (ISR) remains an important issue limiting the clinical safety and efficacy of coronary drug-eluting stents [3]. Previous studies have shown that the cumulative incidence of ISR in coronary drug-eluting stent (DES) at 5 years after PCI is as high as 10 % in real life, which increases the difficulty of re-procedure and leads to poorer quality of life and long-term prognosis, and has become an urgent clinical challenge in the field of cardiovascular intervention [4].

The pathogenesis of ISR primarily involves neointimal hyperplasia triggered by endothelial cell injury and subsequent inflammation, leading to fibroblast proliferation [5]. Stent implantation causes mechanical damage to the vascular endothelium, exposing collagen beneath the endothelium and activating growth factors and other mediators, which drive vascular smooth muscle cells (VSMCs) to migrate toward the intima and differentiate into a synthetic phenotype, leading to abnormal proliferation and the secretion of large amounts of extracellular matrix, resulting in pathological thickening of the neointima [4,6]. Additionally, the stent, as a foreign body, induces a persistent inflammatory response, with mononuclear macrophages and lymphocytes infiltrating and releasing pro-inflammatory factors such as IL-6 and TNF-α, further stimulating VSMCs proliferation and delaying endothelial repair, thereby becoming a sustained driving factor for restenosis [7]. In addition to the nature of the lesion, surgical procedure, and type of DES, many traditional CAD risk factors have been associated with the development of ISR, including high levels of blood glucose and lipids [8]. Current studies related to ISR are more limited to different drug coatings, such as rapamycin and paclitaxel, which are non-specific cytotoxic drugs that can reduce the incidence of stent failure events by inhibiting neointimal hyperplasia [9]. However, the application of cytotoxic drugs also affects other vessel wall cells which influences function [10]. Besides, at the edges of the stent is also a high prevalence area of ISR due to not be covered by the drug coating [11]. Thus, stent drug coating does not completely solve the problem of ISR occurrence. Therefore, in order to better prevent this major complication, we urgently need some effective predictive indicators and precise therapeutic targets [12].

There is growing evidence that inflammatory responses are associated with CAD. For example, interleukin-6 (IL-6), a biomarker of vascular inflammation, is an important risk factor for cardiovascular disease and plays a key role in the development of atherosclerosis (As) [13]. However, most of these studies have been limited to the effects of intrinsic immunity on cardiovascular disease. Intrinsic immunity have characteristics of rapid responsiveness and nonspecificity, may have the limitation of instability in serving as a predictor, which may change significantly with the change of body status, while nonspecific immunotherapy is prone to serious complications such as infections, tumours [14]. While adaptive immunity possesses the characteristics of specificity and stability to get great clinical potential for prediction and prevention against intrinsic immunity [15].

Lymphocyte usually plays a role in adaptive immunity of the organism, usually proliferates and differentiates more than 3–4 days after antigen exposure, and its survival time is 3 months or even lifelong [16]. Based on antigenic manifestations, lymphocytes are classified as CD3⁺T-cell, CD20⁺B-cell and the CD56⁺NK-cell. T-cell play a role in adaptive immunity by facilitating inflammatory responses, delivering antigens, and direct killing [17]. B-cell play a role in adaptive immunity by generating antibodies to bind antigens, and facilitating inflammatory responses [18]. NK-cell can participate in antigen presentation and direct killing [19]. Yu et al. [20] have found that a high frequency of CD8⁺ T cells was associated with short-term cardiovascular mortality in AMI patients. Zheng et al. [21] found that functionally deficient B cells fail to differentiate into plasma cells, which greatly reduced immunoglobulin production and accumulation in atherosclerotic lesions, which significantly reducing development of As. In contrast to other inflammatory factors such as C-reaction protein (CRP), cytokines, lymphocytes are relatively stable indicators do not change drastically as a result of the inflammatory state caused by acute diseases [22].

Current studies have demonstrated the involvement of lymphocytes in As, but the role of each subpopulation in each stage is unknown [23]. We speculate that adaptive immunity likewise plays an important role in the development of ISR. Through serological analysis of patients suffering from coronary ISR, we hope to reveal the potential role of adaptive immunity in the development and progression of coronary ISR, and provide a more accurate and effective basis for prevention and treatment. The aim of this study is to explore the potential association of lymphocyte subsets with ISR, combine traditional factors to explore predictive value, and to search for future therapeutic targets.

2. Methods

2.1. Study population

This study successively included 891 patients who underwent prior stenting were screened for coronary angiography (CAG) at Beijing Anzhen Hospital. Exclusion criteria included AMI, severe heart failure, severe hepatic and renal dysfunction, active inflammation, autoimmune disease, and 79 patients were excluded. These 812 patients were divided into non-ISR group (N = 575) and ISR group (N = 237) according to the ISR diagnostic criteria. The study was registered by Beijing Anzhen Hospital, in accordance with the Declaration of Helsinki, and written informed consent was obtained from all study patients.

2.2. Clincal information and lymphocyte subsets

In this study, venous blood was drawn from all participants on a 12-h fasting period after admission to the hospital. General information included gender, age, body mass index (BMI), hypertension, diabetes, hyperlipidaemia, drinking and smoking. In addition, medication history included aspirin, clopidogrel, ticagrelor, and statins. Clinical information included lymphocyte, monophage, ultrasensitive troponin (hsTnI), B-type brain natriuretic peptide (BNP), aspartate aminotransferase (AST), glomerular filtration rate (eGFR), total triglycerides (TG), total cholesterol (TC), high-density lipoprotein cholesterol (HDL-C), low-density lipoprotein cholesterol (LDL-C), and non-HDL-C, glycated haemoglobin percentage (HbA1c%), glucose (GLU). These blood biochemical parameters were measured using a biochemical analyser (Hitachi-7600, Tokyo, Japan) and a chemiluminescent immunoassay analyser (Abbott-i2000SR, Illinois, USA). Lymphocyte subsets included CD3⁺T cell counts and percentage of CD3⁺T cell, CD4⁺T cell, CD8⁺T cell, B cell and NK cell, and CD4⁺T cell/CD8⁺T cell ratios (Th/Ts). The above data were analysed using the Multitest™ IMK kit (BD Biosciences, San Jose, USA) and a BD FACSCanto™ flow cytometer. Both performed simultaneously by investigators who were unaware of the sample allocation.

2.3. CAG and Optical coherence tomography (OCT)

Each patient's coronary imaging was analysed by two independent and experienced interventional cardiologists. When the two professionals agree, the opinion is taken directly, a third expert evaluates the results when the two professionals disagree. OCT is often integrated into CAG, and each patient's OCT results are evaluated by two specialised technologists. In case of disagreement between the two specialists, the results are reviewed by a third imaging specialist.

2.4. Difination

Diagnostic criteria for ISR were ≥50 % luminal diameter stenosis in the stent or adjacent vessels (within 5 mm of the stent edge) as confirmed by CAG or endoluminal imaging [12]. The prevalence of diabetes is defined as the fulfilment of any of the following criteria: a previous diagnosis of diabetes mellitus, the use of glucose-lowering medication, and a blood glucose level that meets the diagnostic criteria for diabetes during hospitalisation (a fasting glucose of more than or equal to 7.0 mmol/L or a random glucose of more than 11.1 mmo/L) [24].

2.5. Statistical analysis

Categorical variables were expressed as n (%) based on the continuous distribution of the variables, and normally and non-normally distributed data for continuous variables were expressed as mean ± standard deviation and median (25th-75th percentile), respectively. Categorical variables were analysed using the chi-square test, normality of continuous variables was tested using the Shapiro-Wilk test, satisfying continuous normally distributed data was tested using the independent t-test, and non-normally distributed data was tested using the Mann-Whitney U test. Screened statistically different variables as well as traditionally related variables were subjected to one-way logistic regression analysis and multifactorial logistic regression analysis, and those that were statistically significant (p < 0.05) were further corrected. To investigate the linear association of non-HDL versus B-cell % with ISR, we used a five-segment restricted cubic bar chart (RCS) analysis. Finally, we performed separate subgroup analyses of B cell% and diabetes according to the underlying clinical characteristics to elucidate the interaction between B cell% and diabetes in terms of important clinical characteristics affecting ISR outcomes. We also crossed subgroup analyses of these two variables to explore whether they interacted with each other to influence ISR outcomes. Statistical analyses were performed using R4.3.0, and results were considered statistically significant if two-tailed P < 0.05.

3. Results

3.1. Biochemical indicators and lymphocyte subsets in the study population

Table 1 summarises the demographic data and clinical information on the participants. Comparison of the two groups (non-ISR and ISR) showed that the ISR group had a higher percentage of diabetes (P = 0.002), and higher levels of LYMPH% (P = 0.045), TG (P = 0.017), TC (P = 0.004), LDL-C (P = 0.044), GLU (P < 0.001), HbA1c% (P = 0.007), AG (P = 0.016), nonHDL (P = 0.006), sdLDL (P = 0.021), BNP (P = 0.023). There were no statistically significant differences in age, gender, smoking, and BMI. Besides, Table 1 shows the results of lymphocyte subsets assays for each group. Compared to the non-ISR group, the ISR group had: higher levels of B cell% (P = 0.011) and B cell (P = 0.010), and lower levels of CD8⁺ T cell% (P = 0.042).

Table 1.

Baseline of clinical characters and lymphocyte subset in each group.

	Overall (N = 812)	non-ISR (N = 575)	ISR (N = 237)	P-value
Age (years)	61 [54, 68]	61 [53, 69]	61 [55, 67]	0.734
Male, n(%)	656 (80.8)	467 (81.2)	189 (79.7)	0.700
Smoke, n(%)	334 (41.1)	236 (41.0)	98 (41.4)	0.998
BMI (kg/m2)	25.82 [23.53, 28.09]	25.76 [23.52, 27.88]	26.09 [23.98, 28.61]	0.125
Hypertension, n(%)	436 (53.7)	308 (53.6)	128 (54.0)	0.970
Diabetes, n(%)	327 (40.3)	211 (36.7)	211 (36.7)	0.002
Medications
Aspirin, n(%)	644 (79.3)	455 (79.1)	189 (79.7)	0.919
Clopidogrel, n(%)	337 (41.5)	247 (43.0)	90 (38.0)	0.218
Tigagrelor, n(%)	244 (30.0)	161 (28.0)	83 (35.0)	0.057
Biochemical parameters
LYMPH% (%)	23.1 [18.7, 28.3]	22.4 [18.6, 27.5]	24.6 [20.0, 29.9]	0.045
MONO% (%)	5.7 [4.8, 6.7]	5.7 [4.7, 6.7]	5.8 [4.8, 6.8]	0.542
AST (U/L)	20.0 [16.0, 24.8]	20.0 [16.0, 25.0]	19.0 [15.0, 24.0]	0.383
EGFR (ml/min)	91 [77, 100]	91 [78, 101]	90 [77, 99]	0.430
TG (mmol/L)	1.48 [1.05, 2.09]	1.43 [1.04, 1.99]	1.58 [1.12, 2.31]	0.017
TC (mmol/L)	3.57 [3.06, 4.15]	3.48 [3.00, 4.10]	3.75 [3.24, 4.30]	0.004
HDL-C (mmol/L)	1.07 [0.90, 1.27]	1.08 [0.90, 1.27]	1.05 [0.90, 1.26]	0.607
LDL-C (mmol/L)	1.81 [1.41, 2.29]	1.78 [1.37, 2.25]	1.90 [1.46, 2.33]	0.044
GLU (mmol/L)	6.24 [5.23, 8.59]	6.04 [5.16, 8.27]	6.66 [5.59, 9.44]	<0.001
Hba1c% (%)	6.3 [5.8, 7.2]	6.2 [5.8, 7.1]	6.5 [5.9, 7.4]	0.007
hsTnI (pg/ml)	4.80 [3.00, 8.90]	4.60 [2.90, 8.80]	4.95 [3.20, 9.72]	0.224
AG (mmol/L)	17.65 [15.80, 19.70]	17.40 [15.62, 19.50]	18.10 [16.10, 20.00]	0.016
nonHDL (mmol/L)	2.44 [1.96, 3.05]	2.40 [1.94, 2.99]	2.63 [2.10, 3.13]	0.006
sdLDL (mmol/L)	0.60 [0.42, 0.86]	0.58 [0.41, 0.84]	0.67 [0.49, 0.91]	0.021
BNP (pg/ml)	38 [19, 81]	35 [19, 76]	44 [22, 92]	0.023
Lymphocyte subsets
CD3+T cell% (%)	71.4 [65.2, 76.5]	71.8 [65.4, 76.7]	70.4 [64.2, 76.0]	0.120
CD3+T cell (/μL)	1067 [854, 1378]	1064 [836, 1373]	1074 [870, 1391]	0.541
CD4+T cell% (%)	42.2 [36.4, 48.2]	42.2 [36.6, 47.8]	42.3 [36.1, 49.2]	0.711
CD4+T cell (/μL)	646 [490, 825]	642 [489, 822]	661 [494, 829]	0.301
CD8+T cell% (%)	25.5 [20.2, 31.3]	25.9 [20.5, 32.0]	24.2 [19.6, 30.7]	0.042
CD8+T cell (/μL)	382 [279, 527]	394 [278, 537]	374 [284, 503]	0.492
Th/Ts	1.64 [1.23, 2.26]	1.61 [1.21, 2.24]	1.72 [1.25, 2.35]	0.070
NK cell% (%)	14.4 [9.8, 20.3]	14.6 [9.8, 20.0]	14.1 [9.7, 21.4]	0.909
NK cell (/μL)	211 [145, 309]	206 [142, 309]	222 [151, 313]	0.442
B cell% (%)	11.9 [8.7, 15.6]	11.5 [8.4, 15.1]	12.7 [9.5, 16.5]	0.011
B cell (/μL)	175 [117, 256]	173 [111, 247]	186 [137, 266]	0.010

Values are expressed as percentages, mean ± SD, or median (25th−75th percentile). Bold values signifies the statistical significant comparisons between groups.

3.2. Logistic regression analyses of ISR

As shown in Table 2, we first performed univariate logistic regression analyses of ISR for statistically different variables, including age, sex, BMI, diabetes, B cell%, non-HDL-C and AG. Subsequent multivariate logistic regression analyses showed that B cell%, diabetes, and non-HDL-C were statistically different in predicting ISR. After adjusting for age, sex, BMI and AG, B cell% (OR 1.033,95 %CI:1.002–1.065,P = 0.039), diabetes (OR 1.556,95 %CI:1.134–2.134,P = 0.006), non-HDL-C (OR 1.210,95 %CI. 1.026–1.428, P = 0.024) were independent risk factors for ISR, respectively.

Table 2.

Logistic regression analysis to predict ISR.

Variables	Univariate logistic regression		Multivariate logistic regression		Adjusted model
	OR (95 % CI)	P-value	OR (95 % CI)	P-value	OR (95 % CI)	P-value
BMI	1.034 (0.988–1.081)	0.151	1.032 (0.983–1.084)	0.202
Male	0.911 (0.626–1.340)	0.629	1.024 (0.660–1.589)	0.914
Age	1.005 (0.990–1.020)	0.541	1.015 (0.997–1.035)	0.107
B cell%	1.038 (1.008–1.069)	0.013	1.037 (1.002–1.072)	0.036	1.033 (1.002–1.065)	0.039
Diabetes	1.654 (1.218–2.247)	0.001	1.501 (1.076–2.094)	0.017	1.556 (1.134–2.134)	0.006
nonHDL	1.205 (1.026–1.417)	0.023	1.221 (1.025–1.454)	0.026	1.210 (1.026–1.428)	0.024
AG	1.064 (1.008–1.124)	0.025	1.044 (0.985–1.107)	0.143

Bold values signify the statistical significance in the logistic regression analysis to predict ISR.

3.3. Linear relationship between the variable of interest and ISR

To clarify whether there were decreasing intervals and inflection points in the linear variables affecting ISR outcomes. As shown in Fig. 1, we analysed the effect of B cell% and non-HDL-C as continuous variables on the occurrence of ISR using a five-node restricted cubic bar chart (RCS). In the RCS curves, there was a linear upward trend in both B cell% (P for nonlinear = 0.415; P for overall = 0.035) and non-HDL-C (P for nonlinear = 0.168; P for overall = 0.024), which was in line with the ISR.

Fig. 1.

RCS model showing the associations of B cell% and nonHDL with ISR

(A) The relationship between B cell% and probability of ISR. (B) The relationship between non-HDL and probability of ISR.

3.4. Subgroup analyse of B cell% and diabetes in predicting ISR

As shown in Fig. 2, subgroup analyses of B cell% and diabetes were performed separately according to important clinical characteristics, and cross-tabulation analyses of these two variables were deliberately performed. It was seen that there was an interaction between HbA1c and B-cell % affecting ISR outcome (p = 0.04), with the two subgroups HbA1c%≥6.6 (OR 1.07, 95 %CI: 1.02–1.12, P = 0.003) and HbA1c% (OR 1.00, 95 %CI: 0.96–1.05, P = 0.936). Meanwhile, there was an interaction effect of B cell % on diabetes affecting ISR outcome (P = 0.042) for the two subgroups B cell%≥12 (OR 2.23, 95 %CI: 1.46–3.42, P < 0.001) and B cell %<12 (OR 1.16, 95 %CI: 0.73–1.85, P = 0.521). Thus, we found that B cell% ≥12 deepened the promotional effect of diabetes on ISR, HbA1c% ≥6.6 exacerbated the adverse effect of B cell% on ISR.

Fig. 2.

Subgroup analysis of variables with statistically significant difference between non-ISR and ISR group.

Interaction of Each Relevant Variable on the Proportion of B Lymphocytes and diabetes affecting ISR Occurrence.

4. Discussion

The number of people undergoing coronary stenting is increasing and trending younger, and the total coronary stenting duration is increasing, which is leading to a yearly increase in ISR [4]. The claim of how to maintain long-term patency of coronary stents is becoming increasingly urgent, yet we still lack effective means of control [12]. Previous long-term glucose-lipid-lowering therapies cannot fully address the residual risk of ISR [25]. It is gradually recognised that As is a chronic inflammatory disease, but knowledge of the immune-inflammatory response is more limited to intrinsic immunity [26]. The non-specific characteristics of intrinsic immunity have led to the possibility that anti-inflammatory therapies may cause serious side effects, such as tumours and infections. Based on the specificity and precision of adaptive immunity, the relationship between adaptive immunity and As has been explored in recent years [17]. Therefore, we hypothesised that adaptive immunity plays an equally important role in the pathological process of ISR. The aim of our study was to explore the potential role of adaptive immunity in ISR and to try to find predictors of ISR in combination with traditional risk factors, and to explore new therapeutic targets.

In this study, we compared the biochemical indices and lymphocyte subpopulations between ISR and non-ISR groups, after correcting for other relevant factors, we found that high levels of B cell%, non-HDL-C and diabetes were independent risk factors for ISR. Five-section restricted cubic bar chart (RCS) analysis suggested that both non-HDL-C and B cell% showed a linear upward trend consistent with ISR. Next, we subgrouped the effects of B cell% and non-HDL-C on ISR, and we found that high levels of B cell% (B cell% ≥12) and high levels of glycated haemoglobin (HbA1c% ≥6.6) reciprocally promoted the development of ISR.

With the intensive research on the inflammatory mechanisms of CAD, chronic inflammatory states in vivo have been shown to promote the progression of As. Xu et al. [27] showed a strong association between SIRI and SII levels and ISR, finding that both patients with higher SII and higher SIRI were more prone to ISR. SIRI and SII can manifest a systemic inflammatory stress state, which can lead to vascular endothelial dysfunction, which in turn is more likely to activate the internal inflammatory cascade in plaques [28]. Zhou et al. [29] found that implantation of sirolimus-eluting stents in coronary artery stenosis reduced the inflammatory chemokine MCP-1, and stronger inflammation suppression showed better angiographic results. This is strong evidence that the inflammatory response within coronary stents accelerates stent life loss and promotes the development of ISR. In contrast to conventional studies on nonspecific immune inflammatory response, in our study we found that B cells play an important role in ISR, and patients in the ISR group showed higher B cells% with statistically significant differences. In recent years, researchers have also found that B cells are important influences in CAD and are widely involved in the pathological process of CAD [21,30]. Kyaw et al. [31] concluded that early B cell activation and autoantibody production are central factors in accelerating the progression of As after myocardial infarction, and they transplanted purified spleen B cells from one week after myocardial infarction into As mice. They found that IgG accumulation in As plaques of these mice increased and accelerated As, and that specific blockade of B cells significantly slowed the progression of As. Although As and stent endothelial hyperplasia showed different characteristics, they still shares the main mechanisms of neointimal hyperplasia and plaque instability. B cells secrete inflammatory mediators and chemokines, which further promote inflammatory cell infiltration, induce structural disintegration of the fibrous cap, and accelerate the formation of unstable rings [32]. At the same time, B cells differentiate mature plasma cells also secrete immunoglobulin, which combines with antigenic substances within the plaque to form antigen-antibody complexes, which are deposited in the plaque, inducing the proliferation of inflammatory cells and the amplification of the inflammatory response within the plaque, exacerbating the As plaque load and instability [31].

Many previous studies on ISR have pointed out that diabetes promotes the development of ISR. Santos-Pardo et al. [8], in a large cohort study, followed patients undergoing PCI. They stratified patients according to glycated haemoglobin and compared the incidence of ISR in patients under different strata. The risk of ISR was found to increase significantly with increasing glycated haemoglobin, with a risk ratio of 1.02 when the glycated haemoglobin level was ≤5.5 %, and 1.33 when the glycated haemoglobin level was increased to 10.1 %. Our study also confirms that diabetes is an independent risk factor for the development of ISR. These studies consistently show that chronic hyperglycaemia is an important risk factor for ISR. Chronic hyperglycaemia severely affects the cardiovascular system, which may be related to the induction of prothrombotic and pre-inflammatory states by the chronic hyperglycaemic state, in addition, hyperglycaemia promotes the process of As in conjunction with dyslipidaemia and so on [33]. Chronic hyperglycaemia inhibits endothelial nitric oxide synthase (eNOS) activity, which reduces nitric oxide production, leading to vasoconstriction, platelet aggregation and endothelial cell dysfunction, thus increasing LDL-C deposition under the vascular endothelium and participating in the chronic inflammatory response [34]. In addition, hyperglycaemia is usually accompanied by insulin resistance, leading to enhanced lipolysis and increased free fatty acids, which promotes hepatic synthesis of non-HDL-C, exacerbates dyslipidaemia and promotes ISR.

Hyperlipidaemia has also been identified as an independent risk factor for the development of ISR. Mahmoud et al. [35] found an independent association between high lipoprotein A (Lpa) values and ISR by following the probability of ISR in patients with different Lp(a) levels. Zheng et al. [36] explored the possibility of residual lipoproteins (RLPs) levels in post-PCI ISR in diabetic patients and found that RLPs were an independent risk factor for ISR. In our study, we found that the probability of ISR increased significantly with the increase of non-HDL-C. Non-HDL-C is the component of TC minus HDL-C, which includes cardiovascularly harmful lipoproteins such as LDL-C, VLDL, and Lp(a) [37]. Non-HDL-C is a traditional risk factor for CAD, after oxidation they cross the vascular endothelium, they are phagocytosed by macrophages, triggering local inflammation and necrotic atherosclerotic deposition [38]. Based on the complexity of the cardiovascular risk proteome and the variability of its members, the assessment of cardiovascular risk for individual lipoproteins may in some cases be strongly influenced by the condition of the organism, whereas a discussion of the relationship between the harmful lipoproteins as a whole and the ISR can help to make a more stable assessment [39].

To further explore the relationship between traditional metabolic factors and immune-related factors, we performed subgroup analyses of B cell% and diabetes. We found that high levels of B cell% increased the probability of ISR in diabetic patients, and high levels of HbA1c% also exacerbated the promotion of ISR by B cells. Our cross-subgroup analysis of diabetes and B cell% also confirmed the synergistic promotion of ISR by chronic hyperglycaemia and humoral immunity. Therefore, we hypothesised that hyperglycaemia promote endothelial dysfunction, when oxidised lipoproteins are more likely to cross the damaged endothelium and accumulate under the endothelium, where they are phagocytosed by macrophages and deliver antigens to the B cells, which initiates and exacerbates the processes that lead to localised adaptive immunity, promoting an increased plaque load of As, which in turn contributes to the development of ISR [18,40].

Although lymphocytes, which are produced 3–4 days after exposure to antigen, have a lifespan of several months or even a lifetime and do not undergo dramatic changes in response to changes in the body's environment [16]. Nevertheless, we chose to exclude patients with AMI because our study focused on changes in metabolic markers of inflammation in chronic ISR. The study of chronic pathological processes also provides solid clinical evidence for long-term care, prompting patients to maintain coronary stent longevity by tightly controlling glycaemic lipids and in vivo inflammation over time. In addition, we uniquely explored the association of adaptive immunity with ISR and correlated traditional metabolic factors with adaptive immunity, and also provided biological targets for future precision immunotherapy.

This study also has some limitations. Firstly, a larger multi-centre study is needed to generalise the findings. Secondly, we conducted only a cross-sectional study, and we were unable to follow up further details of the patients' primary lesions, surgical procedures, and so on, due to long-term record-keeping problems. Prospective cohort studies and trials with detailed records are needed for further refinement.

5. Conclusions

In this study, we found that B cell%, diabetes, and non-HDL-C were found to be independent risk factors for ISR. Subgroup analyses showed that high levels of B cell% and high levels of HbA1c% synergistically promoted the development of ISR. This provides a stable biomarker for pre-PCI assessment and a new target for precision immunotherapy for ISR, as well as reliable clinical evidence for long-term glucose-lowering and lipid-lowering anti-inflammatory therapy in post-PCI community. This study will establish a more detailed multicentre follow-up cohort and will also experimentally explore the potential pathological mechanisms of adaptive immunity in ISR.

CRediT authorship contribution statement

Duo Yang: Writing – review & editing, Writing – original draft, Data curation. Siyao Ni: Writing – original draft, Software, Data curation. Sheng Liu: Writing – review & editing, Methodology, Data curation. Chenyang wang: Writing – review & editing, Supervision, Software, Project administration, Conceptualization. Kexin Yang: Writing – review & editing, Investigation. Ludan Bi: Investigation, Data curation, Conceptualization. Zhijian Yue: Funding acquisition, Formal analysis, Data curation. Liwei Hang: Resources, Funding acquisition. Ming Zhang: Writing – review & editing, Resources, Project administration, Methodology, Formal analysis. Hai Gao: Writing – review & editing, Resources, Methodology, Investigation, Funding acquisition, Formal analysis.

Handling Editor: Dr D Levy