Patients with STEMI admitted to the Beijing Hospital Catheterization Room for primary PCI from January 2020 to December 2022 were prospectively selected. This study was authorized by the institutional ethics committee and carried out in accordance with the principles of the Declaration of Helsinki. STEMI was defined as persistent chest pain lasting for at least half an hour, accompanied by ST-segment elevation of more than 1 mm in 2 or more adjacent leads. Primary PCI was conducted using standard procedures, and the selection of additional interventions (such as manual thrombectomy or glycoprotein IIb/IIIa inhibitors), while stent placement techniques were determined by the treating operator. All patients were treated with a loading dose of aspirin 100–300 mg and clopidogrel 300 mg or ticagrelor 180 mg. Anticoagulation therapy was administered during the primary PCI procedure with weight-adjusted unfractionated heparin or bivalirudin. An automated injector was used during the coronary angiogram procedure. The choice of postprocedural anticoagulation, including low molecular weight heparin and fondaparinux, was at the discretion of the operator and according to the thrombus burden and the risk of stent thrombosis. Multivessel disease (MVD) was defined as stenosis $\ge$75% of the diameter in at least two major epicardial arteries or their main branches. Left main (LM) disease was defined as left main stenosis $\ge$50% of the diameter. The $\ge$75% and $\ge$50% cutoffs were determined to identify significant stenosis based on visual assessment by at least two experienced operators. The success of the primary PCI was defined as the restoration of final thrombolysis in MI (TIMI) grade 3 or the residual stenosis of IRA $\le$20% with stent implantation. The left ventricular ejection fraction (LVEF) was obtained using echocardiography before discharge. The exclusion criteria were as follows: complications of cardiogenic shock, failed primary PCI, poor coronary angiography images, and insufficient angiography view of the IRA and reference vessel.

CaIMR analysis was performed using the FlashAngio system (Rainmed Ltd., Suzhou, China, Fig. 1). First, a three-dimensional mesh was reconstructed in the target artery using two coronary angiographic projections without overlapping and separated by a minimum 30° angle. Second, aortic pressure was measured using a Flash pressure transducer. Third, several parameters were estimated using computational pressure–fluid dynamics, as previously verified [10]. Hyperemic Pa (Pa${}_{\text{hyp}}$) signifies the maximal hyperemic mean aortic pressure, calculated by averaging the pressure waves over three cardiac cycles using a mathematical formula described previously [8, 10]. Hyperemic Pd (Pd${}_{\text{hyp}}$) is the mean distal coronary pressure during maximal hyperemia, calculated using the Navier–Stokes equation. The computational fluid dynamics method was performed to calculate the pressure drop ($\mathrm{\Delta }$P${}_{\text{hyp}}$) across the meshed coronary arteries, spanning from the inlet to the distal coronary artery.

Fig. 1.

Representative cases of STEMI with caIMR measurement after primary PCI. PCI, percutaneous coronary intervention; LAD, left anterior descending artery; LCX, left circumflex; caIMR, coronary angiography-derived index of microcirculatory resistance; STEMI, ST-segment elevation myocardial infarction; caFFR, coronary angiography-derived index of microcirculatory resistance.

(1)$$P_{d_{hyp}}(\text{unit: }mmHg)=P_{a_{hyp}}-\mathrm{\Delta }{P}_{hyp}$$

L is a nondimensional constant used to represent the distance measured from the inlet to the distal artery. V${}_{\text{diastole}}$ (unit: mm/s) represents the average blood flow velocity during diastole, derived via the TIMI frame count method. K is a constant (K = 2.1), K$\cdot$V${}_{\text{diastole}}$ represented the maximum hyperemic flow velocity [11, 12]. Finally, the caIMR was calculated as follows:

(2)$$\text{caIMR}=P_{d_{\text{hyp}}}\frac{L}{K\cdot V_{\text{diastole}}}$$

CaIMR was calculated after finalizing the PCI in the IRAs or reference vessels. Reference vessels were designated as nonchronic total occlusion vessels. In patients with severe coronary stenosis, Yong’s formula was used to adjust the caIMR, which accounted for the potential influence of the collateral flow-induced wedge pressure on the caIMR in the presence of substantial epicardial stenosis [13]. Two independent operators were blinded to the clinical information of the patients when performing the measurements. An agreement was reached by consensus when inconsistencies occurred.

The prespecified primary endpoint was the occurrence of major adverse cardiovascular events (MACEs) at 3 months and 1 year. MACEs were defined as a composite of cardiac death, target vessel revascularization, rehospitalization due to congestive heart failure (CHF), myocardial infarction (MI), or angina. Follow-up was performed through clinic visits, medical record reviews, and phone contact. Survival free of MACEs = (total amount of patients – number of patients with MACEs/total amount of patients) $\times$ 100%.

Continuous variables and categorical variables were expressed as mean $\pm$ standard deviation or median (interquartile range), as appropriate. Categorical variables are expressed as numbers (percentages). Categorical variables were compared using the chi-squared or Fisher’s exact test. According to distributions, the differences among continuous variables were tested using Student’s t-test or the Mann–Whitney rank sum test. Multivariate analysis was performed using a Cox regression model with a stepwise algorithm and expressed as hazard ratios (HRs) with 95% confidence intervals (CIs), to investigate the independent determinants of the primary outcome. Variables related to the outcome of interest were considered as candidate predictors for multivariate analysis based on clinical consideration and demonstrated a p value of 0.05 in the univariate analysis. Survival curves for MACE-free outcomes were generated using the Kaplan–Meier method, and group differences were evaluated through the log-rank test. A two-tailed p value of 0.05 was considered statistically significant for all tests. All statistical analyses were performed using R version 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria) and Statistical Package for the Social Sciences version 26.0 (IBM Corporation, Armonk, NY, USA).

A total of 194 patients with STEMI who underwent primary PCI were enrolled during the study period. A total of 37 patients were excluded: 2 presented with cardiogenic shock, 5 underwent failed primary PCI, and 30 had poor coronary angiography images and an insufficient angiography view for measurement. Thus, 157 patients were included in the final analysis (Fig. 2). The mean age of the study population was 62.8 $\pm$ 14.3 years. Patients were divided into two groups according to the caIMR measurement in the IRA with a cutoff value of 40: the caIMR $\le$40 group (70%) and the caIMR $>$40 group (30%). The detailed clinical characteristics of the patients are summarized in Table 1. Serum peak troponin I (24.4 $\pm$ 5.1 vs. 19.4 $\pm$ 10.2 ng/mL, p = 0.002) and creatine kinase-myocardial band (CK-MB) (233.7 $\pm$ 128.7 vs. 157.0 $\pm$ 109.9 ng/mL, p 0.001) levels were significantly higher in the caIMR $>$40 group than in the caIMR $\le$40 group. There were no significant differences in any other clinical characteristics between the groups.

Fig. 2.

Study flowchart. PCI, percutaneous coronary intervention; STEMI, ST-segment elevation myocardial infarction; IRA, infarct-related artery; caIMR, coronary angiography-derived index of microcirculatory resistance.

The angiographic and procedural characteristics are presented in Table 2. The door-to-balloon time was comparable between the groups. The left anterior descending artery (54.1%) was the primary culprit vessel, followed by the right coronary artery (30.6%). The proportion of patients with MVD or LM disease was similar in both groups and presented a similar distribution in the IRA. More patients received thrombus aspiration (31.8% vs. 57.4%, p = 0.005) and tirofiban administration (8.2% vs. 25.5%, p = 0.008) in the caIMR $>$40 group than in the caIMR $\le$40 group. The proportions of drug-coated balloon use and drug-eluting stent implantation were similar between the two groups. All patients in the caIMR $\le$40 group achieved a TIMI grade 3 post-primary PCI, whereas the proportion of patients with the restoration of final TIMI grade 3 was 91.5% in the caIMR $>$40 group (p = 0.002). The caIMR in the IRA was significantly higher than in the reference vessel (32.9 $\pm$ 15.8 vs. 27.4 $\pm$ 11.1, p 0.001) (Fig. 3). However, the caIMR in the reference vessel for the caIMR $>$40 group was greater than for the caIMR $\le$40 group (30.9 $\pm$ 11.3 vs. 25.9 $\pm$ 10.7, p = 0.009) (Fig. 4).

Fig. 3.

Paired boxplot between the caIMR in the IRA and reference vessel. caIMR, coronary angiography-derived index of microcirculatory resistance; IRA, infarct-related artery.

Fig. 4.

Distribution of caIMR in the reference vessel according to the caIMR in the IRA. caIMR, coronary angiography-derived index of microcirculatory resistance; IRA, infarct-related artery.

The caIMR in the IRA or the reference vessel was comparable between patients with and without MACEs at 3 months and 1 year (Fig. 5). Table 2 displays the clinical outcomes observed at the 3-month and 1-year follow-ups. The caIMR $>$40 group had higher incidence rates of MACEs at the 3-month (25.5% vs. 8.3%, p = 0.009) and 1-year (29.8% vs. 13.9%, p = 0.04) follow-ups than the caIMR $\le$40 group, which were mainly driven by a higher rate of rehospitalization due to CHF, MI, or angina. Fig. 6 illustrates the MACE-free survival curves at 3 months and 1 year according to whether the caIMR in the IRA was $>$40.

Fig. 5.

CaIMR in the IRA and reference vessel according to MACEs. (A) CaIMR in the reference vessel according to MACEs at 3 months. (B) CaIMR in the IRA according to MACEs at 3 months. (C) CaIMR in the reference vessel according to MACEs at 1 year. (D) CaIMR in the IRA according to MACEs at 1 year. MACEs, major adverse cardiac events; caIMR, coronary angiography-derived index of microcirculatory resistance; IRA, infarct-related artery.

Fig. 6.

Kaplan–Meier curves for MACE-free survival according to the caIMR in the IRA. (A) Curve for MACE-free survival at 3 months. (B) Curve for MACE-free survival at 1 year. caIMR, coronary angiography-derived index of microcirculatory resistance; IRA, infarct-related artery; MACE, major adverse cardiac event; HR, hazard ratio; CI, confidence interval.

Candidate predictors found in the univariate analysis included caIMR in the IRA, caIMR in the reference vessel, caIMR in the IRA of $>$40, age, female sex, hypertension, diabetes mellitus, dyslipidemia, current smoking status, estimated glomerular filtration rate (eGFR), LVEF before discharge 50%, MVD, door-to-balloon time, thrombus aspiration, and final TIMI flow grade 3. The final variables entered into the Cox regression model were caIMR in IRA of $>$40, age, female sex, hypertension, eGFR, and LVEF before discharge 50%. Tables 3,4 show the multivariate predictors of MACEs at 3 months and 1 year in patients with STEMI who underwent primary PCI. Multivariate Cox regression analysis revealed that a caIMR in the IRA of $>$40 was an independent predictor of MACEs in patients with STEMI who underwent primary PCI at 3 months (hazard ratio (HR): 3.459, 95% CI: 1.363–8.779, p = 0.009) and 1 year (HR: 2.384, 95% CI: 1.100–5.166, p = 0.03). Hypertension was also identified as an independent predictor of MACEs at 1 year (HR: 4.026; 95% CI: 1.144–14.162, p = 0.03).

There are some limitations in this study. Firstly, this study was a single-center observational study with a small sample size, and its findings were not robust due to the absence of a control group. However, we measured the caIMR in the reference vessels to perform a self-control analysis. The limited number of events could lead to overfitting in the multivariable Cox survival analysis model. Therefore, a prospective, randomized trial with larger populations is necessary. Secondly, more than 50% of patients with multivessel disease were included in this study, and it was difficult to achieve complete revascularization in cases of primary PCI. Although we performed multivariate analysis to control for the potential confounding impact of multivessel disease, incomplete revascularization has been associated with an unfavorable prognosis [29]. Thirdly, although it has been reported that the severity of epicardial stenosis does not influence coronary microcirculatory resistance, the caIMR measurement was pressure-dependent, which was closely related to Pd${}_{\text{hyp}}$ and V${}_{\text{diastole}}$. As previously described in the Methods section, severe stenosis in the target vessels may result in low blood pressure during the peri-procedure, potentially affecting the velocity and pressure in the distal vessel. However, we used a corrected IMR following the Yong formula in cases with severe coronary stenosis to reveal the actual coronary microvascular function. Fourthly, in the present study, we used a caIMR of $>$40 after primary PCI to reflect the severe microcirculatory impairment. A cutoff value of $>$40 was referenced from a relevant study; this was a pressure‒temperature wire-derived value [4, 5]. Whether a wire-derived IMR cutoff value of $>$40 can be translated into caIMR for primary PCI deserves further study. Fifthly, there were some uninterpretable studies, meaning further studies are warranted to explore the potential explanation.