The primary endpoint was the occurrence of major adverse cardiovascular events (MACEs), defined as a composite of cardiac death, target vessel revascularization, ischemic stroke, and hospitalization for heart failure during 12 months of follow-up post-PCI. Secondary endpoints included individual components of the composite outcome. Patients were stratified into two groups based on clinical outcomes: those experiencing MACEs were classified into the poor prognosis group, whereas those who remained event-free constituted the good prognosis group. All reported events were adjudicated by an independent clinical event committee comprising two cardiologists who reviewed the source documentation.

This retrospective cohort study was conducted at Lanzhou Second People’s Hospital to investigate the relationship between inflammatory and angiogenic biomarkers and clinical outcomes in patients with AMI undergoing PCI. The study protocol adhered to the principles of the Declaration of Helsinki and was approved by the Institutional Review Board of Lanzhou Second People’s Hospital. Written informed consent was obtained from all participants or their legal representatives.

We reviewed all consecutive patients diagnosed with AMI who underwent PCI at our institution between January 2023 and December 2024. Patients meeting inclusion criteria and without exclusion criteria were enrolled. For comparison, healthy individuals who underwent routine physical examination at our institution during the same period were recruited as controls; these individuals were frequency-matched to AMI patients by age and sex distribution and had no history of cardiovascular disease, diabetes, chronic inflammatory conditions, or malignancy. The patient enrollment process is illustrated in Figure 1.

The sample size was calculated based on detecting a clinically meaningful difference in biomarker levels between prognostic groups. Based on preliminary data and prior literature (18, 19), we estimated that patients with poor prognosis would have biomarker levels approximately 30% higher than those with good prognosis, with an expected standard deviation of 40% of the mean. Using a two-sided α of 0.05, power of 80%, and an anticipated MACE rate of approximately 20–25%, we determined that a minimum of 150 patients would be required. To account for potential data incompleteness, we aimed to enroll at least 170 patients. This sample size would also provide adequate statistical power for logistic regression analysis with up to 4 predictor variables, following the rule of at least 10 events per variable.

Patients underwent systematic follow-up for 12 months post-PCI through structured telephone interviews conducted by trained research nurses blinded to baseline biomarker levels. For patients reporting potential events, medical records were obtained from treating institutions for event adjudication.

All interventional procedures were performed by experienced operators with >100 PCI procedures annually following standard institutional protocols. Vascular access was obtained through either the radial or femoral artery, based on operator preference and patient anatomy. After local anesthesia and sterile preparation with iodophor, arterial puncture was performed using the modified Seldinger technique.

Under fluoroscopic guidance, guidewires were advanced to the coronary ostium, followed by selective coronary angiography using contrast medium to identify culprit lesions. For lesions with significant stenosis, balloon angioplasty was performed to predilate the stenotic segment. Subsequently, drug-eluting stents were deployed and post-dilated when necessary to achieve optimal apposition. Procedural success was defined as residual stenosis of <20% with TIMI flow grade 3. Following stent implantation, all equipment was withdrawn, and hemostasis was achieved using either manual compression or closure devices, according to the institutional protocol.

All patients received dual antiplatelet therapy with aspirin (100 mg daily), clopidogrel (75 mg daily), or ticagrelor (90 mg twice daily), which was initiated before the procedure and continued for at least 12 months. Additional medical therapies included statins, beta-blockers, angiotensin-converting enzyme inhibitors, or angiotensin receptor blockers, as clinically indicated.

Comprehensive baseline characteristics were systematically collected from medical records, including demographic variables (age, sex, body mass index), cardiovascular risk factors (smoking history, alcohol consumption, hypertension), and clinical severity assessed using the Killip classification at presentation. Hypertension was defined as a systolic blood pressure ≥140 mmHg, diastolic blood pressure ≥90 mmHg, or current use of antihypertensive medications. A smoking history was defined as current smoking or cessation in the past year.

Venous blood samples (3 mL) were obtained from the antecubital vein using a standard venipuncture technique with minimal tourniquet application to avoid hemolysis. Samples were collected within 24 h of admission and again at 48–72 h post-PCI in serum separator tubes. After clotting at room temperature for 30 min, samples were centrifuged at 3000 rpm for 10 min at 4 °C. The separated serum was immediately processed or stored at −80 °C until batch analysis.

Serum hs-CRP concentrations were quantified using immunoturbidimetric assays (reported in mg/L) with intra-assay and inter-assay coefficients of variation of <5 and <8%, respectively. IL-18 concentrations were measured using enzyme-linked immunosorbent assay (reported in pg./mL) with similar precision. Vascular endothelial growth factor levels were determined using an enzyme-linked immunosorbent assay (Human VEGF Quantikine ELISA Kit, R&D Systems; reported in pg./mL) with a detection range of 31.2–2000 pg./mL. All assays were performed in duplicate by laboratory personnel who were blinded to clinical outcomes.

Statistical analyses were performed using SPSS software (version 24.0; IBM Corporation, Armonk, NY, USA). Continuous variables are expressed as mean ± standard deviation and were compared using an independent samples t-test after confirming normal distribution with the Kolmogorov–Smirnov test. Categorical variables are presented as frequencies and percentages, and comparisons were performed using the chi-square test or Fisher’s exact test when the expected cell counts were <5.

Univariable logistic regression analysis was first performed to assess the association between each candidate predictor variable and poor prognosis. Variables with p < 0.10 in univariable analysis were considered for inclusion in the multivariable model. Given that traditional clinical characteristics showed no significant associations with outcome in our cohort, the final multivariable logistic regression model included only the three biomarkers (IL-18, hs-CRP, and VEGF). Results were reported as odds ratios (ORs) with 95% confidence intervals (CIs).

Receiver operating characteristic (ROC) curves were constructed to assess the discriminatory capacity of individual biomarkers for predicting adverse outcomes, with area under the curve (AUC) calculations and optimal cutoff values determined using Youden’s index. Pairwise comparisons of AUC values between biomarkers were performed using the DeLong test to evaluate whether discriminatory performance differed significantly among the three markers. Statistical significance was defined as p < 0.05 for all analyses.

A total of 203 consecutive patients with AMI who underwent PCI during the study period were screened. After applying exclusion criteria, 170 patients were included in the final analysis (Figure 1). For comparison, 100 healthy individuals were enrolled as controls. The observation group comprised 97 males and 73 females with a mean age of 62.53 ± 9.97 years (range: 40–69 years). The control group included 95 males and 75 females with a mean age of 62.48 ± 9.91 years (range: 41–70 years). During the 12-month follow-up, 40 patients (23.5%) experienced MACEs and were classified into the poor prognosis group, while 130 patients (76.5%) remained event-free and constituted the good prognosis group.

Patients with AMI demonstrated substantially elevated serum concentrations of all three biomarkers compared to healthy controls (Figure 2; all p < 0.01). These findings confirm the activation of both inflammatory and angiogenic pathways in the acute phase of AMI and support the biological plausibility of these biomarkers as potential prognostic indicators.

Post-procedural assessment revealed significant reductions in all measured parameters compared with pre-PCI values (Figure 3; all p < 0.01). The reduction in IL-18 and hs-CRP levels indicates attenuation of the systemic inflammatory response following successful revascularization. Similarly, the decline in VEGF concentrations reflects decreased hypoxia-driven angiogenic signaling following restoration of coronary blood flow and myocardial perfusion.

Patients who subsequently experienced MACEs demonstrated significantly higher pre-procedural concentrations of all three biomarkers compared to those who remained event-free (Figure 4; all p < 0.01). This finding suggests that greater inflammatory activation and more pronounced ischemia-driven angiogenic responses at baseline identify patients at higher risk for adverse outcomes despite successful revascularization.

Baseline demographic and clinical characteristics showed remarkable homogeneity between patients with favorable and unfavorable outcomes (Table 1). No significant differences were observed in the distribution of traditional cardiovascular risk factors between prognostic groups, including sex, age, BMI, smoking history, alcohol consumption, hypertension, or Killip classification (all p > 0.05). This finding underscores the potential incremental value of biomarker assessment for risk stratification beyond traditional clinical parameters.

Univariable logistic regression analysis was performed to identify potential predictors of poor prognosis (Table 2). Traditional clinical variables including age, sex, BMI, smoking history, alcohol consumption, hypertension, and Killip classification were not significantly associated with adverse outcomes (all p > 0.10). In contrast, all three biomarkers showed significant univariable associations with poor prognosis: IL-18 (OR = 28.142, 95% CI: 1.524–519.615, p = 0.025), hs-CRP (OR = 2.456, 95% CI: 1.142–5.281, p = 0.021), and VEGF (OR = 1.758, 95% CI: 1.362–2.269, p < 0.001).

Multivariable logistic regression analysis confirmed that all three biomarkers remained independent predictors of poor prognosis after mutual adjustment (Table 3). IL-18 showed the strongest association (OR = 26.075 per pg./mL increase, 95% CI: 1.376–494.285, p < 0.001), followed by hs-CRP (OR = 2.284 per mg/L increase, 95% CI: 1.023–5.099, p < 0.001) and VEGF (OR = 1.643 per pg./mL increase, 95% CI: 1.251–2.158, p < 0.001).

ROC curve analysis demonstrated that all three biomarkers had robust discriminatory capacity for identifying patients at risk for adverse outcomes (Figure 5). hs-CRP showed the highest predictive accuracy (AUC = 0.838, 95% CI: 0.774–0.902), followed by IL-18 (AUC = 0.803, 95% CI: 0.732–0.874) and VEGF (AUC = 0.800, 95% CI: 0.727–0.873); all p < 0.001 compared to the null hypothesis of AUC = 0.50.

Pairwise comparisons of AUC values using the DeLong test revealed no statistically significant differences between biomarkers: hs-CRP vs. IL-18 (p = 0.421), hs-CRP vs. VEGF (p = 0.385), and IL-18 vs. VEGF (p = 0.943) (Table 4). These results indicate that while hs-CRP demonstrated numerically superior discriminatory performance, the predictive accuracy of all three biomarkers was statistically comparable.