DNA was extracted from blood samples using the GeneMatrix Quick Blood DNA Purification Kit (EUR_X, Poland) according to the manufacturer’s instructions. High-quality DNA was extracted from all samples, with an average yield of 51.7 μg (range 5–460 μg).

The EGF gene was genotyped using the PCR-RFLP (Polymerase Chain Reaction—Restriction Fragment Length Polymorphism) method, whereas the PDGFB, VEGF-A, TGFB1 and bFGF genes were genotyped using 5' exonuclease TaqMan genotyping assays.

Continuous and categorical variables were compared between the ISR and no-ISR groups using U-Mann-Whitney and Fisher’s exact tests, respectively. Continuous variables are presented as median and interquartile ranges. Categorical variables are presented as percentages. The Chi-square test was used to determine genotype consistency with the Hardy Weinberg Equilibrium (HWE). The associations between restenosis risk and the analysed polymorphisms were analysed using the following genetic models: dominant, recessive and log-additive for PDGFB, EGF, TGFB1 and VEGF-A and dominant only for bFGF due to the low number of minor homozygotes. Similar to the methodology used by us [25] and others [27,28], these models were used for patients (n = 265) and adjusted for clinical variables that reached a p-value of <0.3 when comparing patients with and without significant ISR (Table 2). Additionally, following the correction of angiographic data that reached p<0.3 in the group comparison (Table 3), the effect of the considered polymorphisms on LLL (a continuous variable) was analysed. p<0.05 was accepted as the threshold of statistical significance. Analyses were completed using NCSS (Number Crunching Statistical Systems, Kayasville USA) software [29] and with the SNPassoc package in the R environment [30].

The study was approved by the Ethics Committee of the Silesian Medical Chamber, Katowice, Poland. All patients gave their signed consent to participation.

The ISR patients and no-ISR patients were similar in terms of age, sex, and comorbidities, with the exception of a more frequent history of myocardial infarction in the ISR group (Table 2). The treatment recommended on discharge was also similar in both groups. Post-stenting lesions in which angiographically significant ISR developed most frequently were complex lesions, according to the AHA/ACC classification [31]. Lesions in which ISR developed were also longer and more frequently located ostially, and the procedure itself was more frequently complicated by dissection. Despite more frequent ostial location, the vessels in which significant ISR developed were smaller in diameter than lesions in which significant ISR did not develop after stenting (Table 3).

The observed frequencies of polymorphisms conformed to HWE for the PDGFB, bFGF, VEGF-A and TGFB1 genes, but a deviation from HWE was observed for the distribution of the polymorphism of the EGF gene (p = 0.007). Angiographically significant ISR was significantly less frequent in individuals with the A/A genotype of rs180047 (TGFB1) versus individuals with the A/G and G/G genotypes (Table 4). Restenosis frequency did not differ for individual genotypes of other analysed polymorphisms. The frequencies of rarer alleles of the analysed polymorphisms were similar to those observed in European populations (Table 4).

The multivariable analysis that adjusted for clinical variables indicated that the A/A genotype of rs1800470 was related to a lower risk of significant ISR in the dominant model. Rs2285094 rs4444903, rs308395, and rs699947 were not associated with a risk of angiographically significant ISR (Table 5).

In the multivariable analysis, LLL was significantly lower in patients with the A/A genotype of rs1800470 (TGFB1) versus those with the A/G and G/G genotypes and higher in patients with the A/A genotype of the VEGF-A polymorphism versus the A/C and C/C genotypes. The C/C genotype of rs2285094 (PDGFB) was associated with higher lumen loss compared to C/T heterozygotes and T/T homozygotes (Table 6).

The role of TGF-β1 in restenosis appears to be confirmed, but its underlying mechanism has not been fully explained due to the complex role of TGF-β1. TGF-β1 has an antiproliferative effect because it inhibits cell proliferation at the G1 stage [33]; however, at levels exceeding 1–2 fg/cell, TGF-β1 may stimulate proliferation of SMCs, fibroblasts and chondrocytes [11]. TGF-β1 is secreted locally by arterial wall fibroblasts, where it has a paracrine effect on SMCs, endothelial cells and macrophages migrating to a site injured after PCI. Additionally, platelets participating in parietal thrombus formation after an arterial wall injury release significant quantities of TGF-β1 [34]. TGF-β1 is also found in blood serum, and its levels may also depend on genetic factors. Some researchers have associated the C alleles with higher TGF-β1 levels [35–37], whereas other authors postulated higher levels in T/T homozygotes [38].

However, the role of TGF-β1 in restenosis is not a simple function of its serum levels. In studies conducted at our centre, we did not observe a relationship between TGF-β1 serum levels and restenosis history, recurring restenosis or first restenosis [39,40]. In the first paper on the role of rs1800470, Fragoso et al. demonstrated that, in the Mestizo population, the T allele of rs1800470 is associated with a higher risk of restenosis [14]. Our results for the Polish population indicate that the A allele (the T allele on the complementary strand) is associated with a lower risk of restenosis. However, aside from different ethnic backgrounds, our study group differs from that of Fragoso et al. in the type of implanted stent. In Fragoso et al., patients were implanted with either an antimitotic stent or a metal stent, which may affect the results because the restenosis mechanisms for these two types of stents differ [41]. Ethnicity, the type of implanted stents, and the fact that our population comprised patients with SCAD, whereas the population studied by Fragoso et al. also included patients with acute coronary syndromes, could account for the discrepancy in the relationship between the T and C alleles of rs1800470 and restenosis in the Polish and Mestizo populations [14]. These differences, although difficult to explain, are common in terms of the analysis of the already mentioned relationship between TGFB1 polymorphisms and TGF-β1 serum levels and between rs1800470 and other cardiovascular diseases. For example, in a population of Italian and German patients, the C/C genotype was associated with a higher risk of myocardial infarction [42,43]. However, Yokota et al. and Cruz et al. demonstrated that, in Japanese and Mexican populations, respectively, the T allele is associated with a higher risk of myocardial infarction [35,44]. In an European population, T allele carriers with rheumatoid arthritis more frequently suffered from hypertension [45]. One study implied a higher risk of IgA nephropathy development in T allele carriers [46], whereas another study did not confirm this relationship [47]. It would be difficult to postulate a mechanism for a role of rs1800470 in restenosis intensification due to the large number of signalling pathways and processes controlled by TGF-β1. However, the role of this polymorphism may depend on other genetic, environmental, or patient-dependent (e.g., severity of coronary artery atherosclerosis) factors, which could also explain the different between the results obtained by Fragoso et al. and our results concerning the relationship between restenosis risk and the T allele of rs1800470.

PDGFB is a growth factor for cells of mesenchymal origin, including SMCs, and supports their migration to the lumen of an injured vessel. PDGFB is released from platelets forming a parietal thrombus and from activated macrophages migrating to an injury in the arterial wall [48,49]. PDGFB receptor kinase inhibitors inhibit neointima formation in an experimental model [48]. Rs2285094 is located in an intron near a splice site. This polymorphism has been associated with a risk of scleroderma development [17]. No relationship was observed between the risk of type 1 diabetes and IgA nephropathy development and the rs2285094 genotype [15,16]. In our analysis, we observed higher LLL in the C/C homozygote group, even after correction for angiographic variables, suggesting that this result is not an artefact.

Damage to the layer of SMCs results in the release of preformed growth factors, including bFGF and EGF [50]. Under experimental conditions, mitogen activity in the medium in which SMCs were immersed decreased by 34% following administration of anti-EGF antibodies [50]. EGF may influence restenosis primarily through SMC migration to the vessel lumen because an EGF receptor antagonist inhibited SMC migration but not proliferation [51]. A relationship between rs4444903 and a risk of development and malignancy of numerous neoplastic diseases was demonstrated [20,52,53], thus confirming its “functionality”. Our study is the first to examine the relationship between this polymorphism and restenosis, but no relationship between rs4444903 and the intensity of this process was identified.

bFGF is a growth factor influencing both smooth muscle and endothelial cells. Rs308395 is located within the bFGF gene promoter region; thus, it may influence the binding of transcription factors and bFGF expression [18,19]. A relationship was observed between this polymorphism and the prognosis of patients with non-Hodgkin lymphoma. Patients with the C/C genotype were more frequently diagnosed with an aggressive form of this disease compared to G allele carriers [19]. Experimental studies have demonstrated that bFGF levels increase in the first hours after balloon angioplasty or stent implantation, which may be related to intensified inflammatory response [54], whereas SMC proliferation after arterial wall injury is inhibited by antibodies directed against bFGF [55]. However, in dogs undergoing iliac artery angioplasty, bFGF infusion restored endothelium-dependent and -independent relaxation of the arterial wall and reduced LLL [56]. As bFGF stimulates endothelial cell proliferation, thereby accelerating stent endothelialisation, bFGF was proposed as a possible substance for stent coating [57]. Similar studies are in progress for VEGF-A, which also stimulates endothelial cell proliferation [58]. Furthermore, higher VEGF-A levels after PCI are related to restenosis in stents coated with an antimitotic drug [5,59]. Rs699947 has been widely studied in autoimmune and neoplastic diseases. Lower levels of this cytokine are observed in individuals with Kawasaki disease and A/A and A/C genotypes [60]. Additionally, a higher risk of colon cancer and osteosarcoma was identified in carriers of the A allele [61]. Relationships have also been identified between VEGF-A polymorphisms and the development of collateral circulation in individuals with coronary artery disease [23], the risk of coronary artery disease in the general population [62], and myocardial infarction in patients with rheumatoid arthritis [63]. Although VEGF-A stimulates stent re-endothelialisation and SMC migration and rs699947 has been previously associated with coronary artery disease, our study is the first to examine this polymorphism in terms of restenosis and identify an association with LLL.

The strengths of this study include the first (with the exception of the TGFB1 gene) polymorphism analysis of genes encoding growth factor proteins with confirmed roles in restenosis. Polymorphisms of the TGFB1, VEGF-A, bFGF and EGF genes are functional polymorphisms that affect the levels of these growth factors and are associated with numerous, mainly neoplastic, diseases. Another strength of this study is the fact that the number of patients enrolled was sufficient to detect, with a power of 80%, the differences in LLL in a vessel below 0.05 mm, assuming a standard deviation of 0.1 mm under the dominant or recessive model for each polymorphism studied. However, the size of the study group was sufficient to detect a 15% or greater difference in the incidence of significant angiographic restenosis for the dominant/recessive model for all examined polymorphisms which was a limitation of the study. Limitations of this study include also different time to next coronary angiography for patients enrolled, which resulted both from the time of the onset of symptoms and from the various waiting period for an elective procedure, ethnically homogenous study population and small number of homozygotes for the rs308395 polymorphism of the bFGF gene.