Association between segmental noninvasive longitudinal strain and quantitative microvascular perfusion in ST-segment elevation myocardial infarction: implications for clinical outcomes

Abstract

Objective

This study aims to investigate the relationship between segmental longitudinal strain (LS) and quantitative microvascular perfusion (qMVP) in patients with ST-segment elevation myocardial infarction (STEMI), and to explore the prognostic value of the two indicators after STEMI.

Methods

The retrospective study enrolled 61 patients who underwent primary percutaneous coronary intervention (pPCI) for first STEMI. Microvascular perfusion (MVP) and qMVP were analyzed by myocardial contrast echocardiography (MCE), and segmental LS was analyzed by two-dimensional speckle tracking echocardiography (2D-STE). Myocardial wall perfusion was qualitatively assessed visually. Quantitative myocardial perfusion parameters were analyzed using an 18-segment model. The correlation between segmental LS and qMVP was assessed. The prognostic value of segmental LS and qMVP for major cardiac adverse events were evaluated.

Results

Among the 314 segments with abnormal wall motion, 44 showed normal microvascular perfusion (nMVP), 87 showed delayed microvascular perfusion (dMVP), and 183 exhibited microvascular obstruction (MVO). Segmental LS was correlated with segmental wall motion (WM) and qMVP. At 12-month follow-up, 19 patients experienced cardiac events. NT-proBNP, regional LS (rLS), and regional qMVP (r-qMVP) were associated with cardiac events. The area under curve (AUC) of combination of rLS and r-qMVP was bigger than single indicator for identifying prognostic value (P < 0.001).

Conclusion

Segmental LS indices are correlated with qMVP within the infarct zone following reperfused STEMI. Both rLS and r-qMVP are sensitive to myocardial damage and provide prognostic value for clinical events after STEMI. And the combination of rLS and r-qMVP shows improved predictive ability compared to a single indicator.

Graphical Abstract

Introduction

In recent decades, reperfusion therapies such as primary percutaneous coronary intervention (pPCI) and thrombolysis have significantly improved the prognosis of patients with myocardial infarction (MI) [1]. However, the mortality rate among MI patients remains alarmingly high [2, 3]. In fact, coronary artery occlusion lasting more than 40 min can lead to irreversible myocardial damage from the endocardium to the epicardium [4,5,6]. Timely reperfusion is the optimal strategy to reduce myocardial necrosis and salvage ischemic myocardium [7]. However, reopening the culprit vessel does not necessarily mean restoring adequate myocardial, as a significant portion of ischemic myocardium may still undergo necrosis in patients with myocardial infarction even after early reperfusion and restoration of vessel patency, which is clinically manifested as min- or no-reflow phenomena [8, 9]. These patients often exhibit coronary microcirculatory dysfunction. Myocardial contrast echocardiography (MCE) is a radiation-free technique that can assess microvascular perfusion (MVP) in real time, thus providing rapid information on myocardial viability [10]. Qualitative MVP has been applied in various clinical conditions, including heart failure, cardiomyopathy, and coronary heart disease [11,12,13]. Previous studies have shown that quantitative segmental analysis of myocardial perfusion can help distinguish between acute myocardial infarction (AMI) and stress cardiomyopathy (SCMP) [14]. However, few studies have explored the role of quantitative MCE to predict long-term cardiac events following myocardial reperfusion in the setting of an acute ST-segment elevation myocardial infarction (STEMI) [15]. Global longitudinal strain (GLS) obtained by two-dimensional speckle tracking echocardiography (2D-STE) is considered to be an effective parameter for quantifying left ventricular function, more sensitive than left ventricular ejection fraction (LVEF) assessed by 2D echocardiography (2DE), and their roles in major ischemic heart disease (IHD) has been previously reported [16]. Related studies have shown that GLS can predict the possibility of recurrent cardiovascular events in AMI patients treated with pPCI within one year. Additionally, it can stratify the risk levels of AMI patient prognosis and has been confirmed to have better predictive ability than LVEF and the wall motion score index (WMSI) for left ventricular global longitudinal strain and strain rate [17]. However, no study has explored the correlation between segmental longitudinal strain (LS) and qMVP in STEMI patients and whether LS and qMVP are associated with cardiac events. Furthermore, whether the combination of segmental LS and qMVP increased extra value over either index alone has not been reported. Therefore, the objective of this study is to evaluate the relation between sLS and qMVP within STEMI patients after pPCI, and their value in predicting prognosis.

Methods

Patient population

A total of 92 patients diagnosed with STEMI treated with pPCI were included from November 2022 to January 2024 at Gansu Provincial People’s Hospital, and all included patients voluntarily signed informed consent. The flow chart was shown in Fig. 1. All patients underwent 2D-STE and MCE within 72 h after pPCI. Patients with a history of MI, coronary artery bypass grafting (n = 4), atrial fibrillation, congenital heart disease, bundle branch block (n = 2), myocardial and pericardial disease history (n = 2), and moderate to severe valvular heart disease (n = 6) were excluded. Patients who did not undergo MCE examination within 72 h after PCI (n = 8) were also excluded. Additionally, patients with poor quality of strain measurement and MCE images (n = 6), and those lost to follow-up (n = 3) were excluded. All segments of 61 patients were analyzed using by both qMVP and LS, and all patients underwent cardiac event follow-up. Timely telephone follow-up was conducted on STEMI patients to record the occurrence of clinical endpoint events, including all-cause mortality, hospitalization due to heart failure, hospitalization for recurrent angina, and non fatal myocardial infarction. At the beginning of each month, telephone follow-up will be conducted on discharged patients, with an error of no more than one week before and after. Prior to emergency pPCI, they were administered loading doses of aspirin 300 mg, a P2Y12 inhibitor (clopidogrel 600 mg, or ticagrelor 180 mg), as well as 100 IU/kg of intravenous heparin (maximum dose 5000IU), and all patients received medication treatment in accordance with the contemporary STEMI patient diagnosis and treatment guidelines of the Chinese Medical Association Cardiology Branch after surgery. The study was approved by the local ethics committees.

Fig. 1

Flow chart depicting patient inclusion and exclusion criteria. STEMI, ST-elevation myocardial infarction.PCI, percutaneous coronary intervention; MCE, myocardial contrast echocardiography

Echocardiography analysis

Standard 2D grayscale, Doppler images, and MCE were acquired using an EPIQ 7 C imaging system (Philips, Andover, MA, USA) equipped with S5-1 transducers. Echocardiographic data were analyzed offline using QLAB11.8 software (Philips Vingmed Ultrasound). Echocardiography parameters and data measurement methods were analyzed according to the American Society of Echocardiography guidelines.

Each subject was operated on by two senior physicians who were blinded to the STEMI results. The left ventricular end-diastolic volume (LVEDV), left ventricular end-systolic volume (LVESV) and left ventricular ejection fraction (LVEF) were measured and calculated using the modified Simpson’s biplane method in the standard left ventricular apical four- and two-chamber views. Use an 18-segment model to visually evaluate regional wall motion, with each segment rated as follows: 1 = normal, 2 = hypokinetic, 3 = akinesia, 4 = dyskinetic, and 5 = aneurysmal change. Global WMSI was calculated as the average of the scores assessed for each segment. (WMSI = sum of the scores of all evaluated segments / number of the evaluated segments). Mitral E and A peak velocities (cm/s) were measured using colour Doppler flow imaging and the E/A ratio was calculated. Pulsed tissue Doppler was used to measure the peak velocities of the septal and lateral walls of the mitral annulus in early diastole (e’), and the E/e’ ratio was calculated. The e’ velocity was calculated from the average of the lateral and septal values. The left atrial volume (LAV) was measured using a biplane area-length method in apical 4- and 2-chamber views immediately before mitral valve opening. The Right atrial volume (RAV) was measured using the same method in the right ventricular-focused 4-chamber cardiac section images. Tricuspid annular plane systolic excursion (TAPSE) was assessed using M-mode echocardiography, defined as the displacement of the tricuspid annulus from end-diastole to end-systole. Additionally, Tissue Doppler Imaging (TDI) was employed to measure the tricuspid annular velocities, capturing peak systolic (TDI S), early diastolic (TDI E), and late diastolic (TDI A) velocities.

Strain analysis

The standard LV apical 4-, 3- and 2-chamber views were acquired respectively to clearly display the morphological structure and endocardial contour of the LV and continuously acquire images of at least three cardiac cycles at frame rates between 50 and 80 frames/s (mean 63.1 ± 4.4 frames/s). The images were stored and backed up in DICOM format for offline analysis by the software. Import the stored dynamic images into QLAB11.8 software for analysis, enter aCMQ mode, the end of systole was defined as the first frame when the aortic valve was closed in the apical long-axis view. At end-systole, the endocardial border was manually delineated and adjusted the region of interest to encompass the entire myocardium, then the software automatically tracked the myocardium. Each apical view was divided into 6 segments. The strain curves of the 18 segments of the LV were automatically analyzed by the software, thereby obtaining the longitudinal peak systolic strain (LPSS) for each segment. All strain values were reported as absolute values, in accordance with the recommendations. Each segment’s LPSS is defined as segmental LS (sLS), and culprit regional LS (rLS) was obtained as the average of the segments belonging to each culprit territory. The GLS was then calculated as the average of the LS values of the 18 segments at end systole (Fig. 2A and B).

Fig. 2

(A). Left ventricular longitudinal strain curve of a patient with STEMI. (Ba). A bullseye chart of a patient with RCA stenosis, GLS=-16.6%; (Bb). A bullseye chart of a patient with LAD obstruction stenosis, GLS=-6.1%; (Bc). A bullseye chart of a patient with three branch lesion, GLS=-4.1%, and abnormal movement in certain segments. (C) From left to right, they are examples of quantitative nMVP, dMVP, and MVO

Myocardial contrast echocardiography

The contrast agent used for MCE was Perfluoropropane-albumin Microspheres (Lizhuo Pharma). This diluted agent was administered as 0.5 to 1 ml slow bolus injections was given via the left forearm median cubital vein.

During contrast infusion, a flash of ultrasound with a high mechanical index was administered to destroy microbubbles in the sector, and then the replenishment of the microbubbles was semiquantitatively observed. Normal MVP (nMVP) is defined as complete supplementation of contrast agent in the myocardium segment within 4 s after high mechanical index pulse = 1. Delayed MVP (dMVP) was defined as at > 4 s, perfusion defects were still observed, but complete replenishment was seen within 10 s = 2. Microvascular obstruction (MVO) was defined as absent contrast perfusion = 3 (Fig. 3). The MCE perfusion score index (PSI) was determined by summing the perfusion scores of all segments and dividing by the total number of evaluable segments. Segments were deemed interpretable by MCE if they were fully visualized from the epicardium to the endocardium. Divide the segments into 3 groups based on the MVP level: group nMVP, segments with nMVP; group dMVP, segments with dMVP; and group MVO, segments with MVO. To investigate the relationship between segmental MVP and cardiac events, the segmental MVP of the culprit region was evaluated and defined as regional MVP (rMVP). The MVP of the culprit area was divided into two groups based on the degree of MVP: group regional MVO, ≥ 2 MVO segments; Group regional non-MVO, < 2MVO segments.

Fig. 3

Example of nMVP, dMVP, and MVO. Demonstration of nMVP (A), dMVP (B), and MVO (C) on real-time MCE after PCI of STEMI. Images from left to right are before the high mechanical index (MI) impulse, immediately after the high MI impulse, and at plateau intensity. Demonstrable contrast defect located at the red arrows

Quantification of resting MCE images was performed by using dedicated QLAB software (Philips Medical Systems, Bothell, WA, USA) in 18 myocardial segments that consisted of 6 segments in apical 4-, 2-, and 3-chamber views. The quantified large segmental regions of interest (ROI) was placed on end-systolic frames, starting from the frame immediately after the flash pulse and including all end systolic frames in subsequent cardiac cycles. Measure the average contrast intensity of each ROI were performed using automatic motion compensation mode to overcome artifacts that may be caused by periodic cardiac motion. After checking the motion compensation if necessary, the ROI position should be finely adjusted for each frame. Time-intensity curves was automatically calculated and subsequently fitted by software to the following monoexponential function: y = A × (1 – e^–βt), in which A is the peak plateau signal intensity reflecting the microvascular cross-sectional area or myocardial blood volume, and β represents the rate of rise in signal intensity from baseline to the plateau that reflects the myocardial blood velocity. Multiplying A by β (Axβ) represents the myocardial blood flow (MBF) (Fig. 2C). At the segmental level, the respective values were noted as sA, sβ, and segmental myocardial blood flow (sAxβ). The values of culprit regional A, β and Axβ were obtained as the average of the segments belonging to each culprit territory, which were used for event-free survival analysis and labeled as rA, rβ, and regional myocardial blood flow rAxβ.

Statistical analysis

Statistical analyses were performed with IBM SPSS statistics version 25 (IBM Corp., Armonk, NY) for Windows software in a compatible computer. Normally distributed continuous data are reported as mean ± SD, whereas variables that were nonnormally distributed are expressed as the median (25th, 75th percentile) and interquartile range, and categorical variables are expressed as frequencies and percentages. Use A one-way analysis of variance with Bonferronni correction or Kruskal-Wallis test (in the case of a nonnormal distribution) to compare the segmental myocardial blood flow (sAxβ) and segmental LS (sLS) in segments with different degrees of MVP. Study the correlation between segment LS and MVP, using Pearson’s correlation coefficient when the distribution of measurement data is normal, and Spearman’s correlation coefficient when the distribution is non normal. The comparison of count data is conducted using the chi square test or Fisher’s exact probability method. When comparing the regional Axβ and LS in patients with and without cardiac events, the independent sample t test was used for inter-group comparison if the distribution of measurement data was normal and the Mann-Whitney U test was performed if it was non-normal. Use the chi-square test for categorical variables. Use receiver operating characteristic (ROC) curve analysis and Youden index to assess the optimal cutoff points of LS and Ax β in predicting of MVO and cardiac events. The comparison of ROC curves adopts the DeLong method. Kaplan-Meier analysis was used to analyze the survival rate and cumulative event rate of STEMI patients, and the incidence of clinical composite endpoint events was predicted based on the optimal cutoff values of segmental LS and MVP for predicting major clinical adverse events. Intra- and inter-observer agreements regarding sAxβ and sLS were assessed by the Bland-Altman plots in 10 randomly chosen patients. The difference was considered statistically significant when two-sided P < 0.05.

Results

Clinical characteristics

Sixty-one patients with STEMI fulfilled the inclusion criteria. Thirty patients were excluded because of a history of myocardial infarction, coronary artery bypass grafting (n = 4), patients with atrial fibrillation, congenital heart disease, bundle branch block (n = 2), myocardial and pericardial disease history (n = 2), moderate to severe valvular heart disease (n = 6), poor quality of strain measurement and MCE images (n = 6) or patients lost to follow up (n = 3), or those who did not undergo MCE examination within 72 h after PCI (n = 8). The clinical characteristics of 61 patients population are listed in Table 1. The mean age was 57.5 ± 12.28 years, and 58 patients (95.08%) were male. Coronary intervention was performed in all 61 patients, including intervention in all coronary arteries with the identified culprit lesions [25 (36.4%) left anterior descending artery (LAD), 30 (50%) right coronary artery (RCA), and 6 (13.6%) left circumflex artery (LCX)]. All patients were reported with post-procedural successful thrombolysis in myocardial infarction (TIMI) flow 3. Blood pressure and heart rate were measured at the first echocardiographic examination. Maximum values of brain natriuretic peptide (BNP) and high-sensitivity troponin I were recorded during the first hospital stay. The median time from symptom onset to vascular recanalization was 6.5 h. Most patients had single-vessel disease (50.9%), and all patients received 1 (or more) drug-eluting stents to the culprit lesion.

Table 1 Clinical characteristics of the study population (N = 61)

Echocardiography characteristics

The Echocardiography characteristics of these patient population are listed in Table 2. Resting regional wall motion abnormalities after STEMI with coronary intervention were reported in all 61 patients, with a mean WMSI of 1.59 ± 0.55, LVEDV (ml) of 108.67 ± 24.07, LVESV (ml) of 51.53 ± 18.86, resting LVEF (%) of 51.53 ± 18.86, LVGLS of 13.58 ± 1.1, PSI of 1.49 ± 0.31, rMVP of 1.77 ± 0.42, rLS of 12.83 ± 0.99, and regional MBF (rAxβ) of 1.62 ± 0.62.

Table 2 Echocardiography characteristics of the study population (N = 61)

Segmental echocardiography variables

MVP and MW could be measured in all 61 patients at the acute phase. Among 314 segments with acute phase functional abnormalities, 44 (14.0%) exhibited nMVP (group nMVP), 87 (27.7%) exhibited dMVP (group dMVP), and 183 segments (58.28%) exhibited MVO (group MVO), and quantitative analysis was possible in all 314 segments. While MVP is theoretically expected to exhibit normal myocardial function, our study identified wall motion abnormalities in 44 segments (14.0%) with normal MVP. This is likely attributable to misinterpretation of contrast agent perfusion or wall motion, myocardial stunning, or abnormal collateral blood flow. Compared with group nMVP and group dMVP, group MVO had lower sLS (14.57 ± 0.95 vs. 13.68 ± 1.08 and 11.61 ± 1.58, P < 0.001), sA (6.18 ± 1.72 dB vs. 3.8 ± 1.38 dB and 1.53 ± 0.7 dB, P < 0.001), sβ (0.6 ± 0.23 s^− 1 vs. 0.4 ± 0.24 s^− 1 and 0.17 ± 0.14 s^− 1, P < 0.001), and sAxβ (3.93 ± 2.16 dB/s vs. 1.67 ± 1.28 dB/s and 0.33 ± 0.44 dB/s, P < 0.001) and higher sWM (2.18 ± 0.39 vs. 2.33 ± 0.47 and 3.59 ± 1.15, P < 0.001), as shown in Table 3.

Table 3 Segment strain and myocardial perfusion characteristics data among the three different MVP groups

Regression analysis of MVP

In the Univariate logistic regression analysis, sWM, sLS, sA, sβ and sAxβ were independently associated with MVO (sWM OR = 12.313; 95% CI, 5.359–28.292; P < 0.001; sLS OR = 0.103; 95% CI, 0.059–0.179; P < 0.001; sA OR = 0.063; 95% CI, 0.037–0.110; P < 0.001; sβ OR = 6.485E-6; 95% CI, 5.093E-7–8.259E-5; P < 0.001; sAxβ OR = 0.037; 95% CI, 0.019–0.072; P < 0.001). Spearman correlation analysis was performed on the indicators with statistical differences (P < 0.05) in the univariate analysis, and it was found that sA, sβ and sAxβ were highly correlated (correlation coefficient r = 0.955, r = 0.973), while the others were not highly correlated. Therefore, sA and sβ were eliminated, and the sWM, sLS, and sAxβ were included for multivariate logistic regression analysis, and the results show that only sA × β values have diagnostic value (P < 0.001), which shown in Table 4. In a comparative analysis of ROC curves, the results of area under the curve (AUC) showed that the accuracy of the sA × β value to identify MVO was higher than sWM, sLS, as shown in Fig. 4.

Table 4 Uni-and multivariate logistic regression analysis of LV segmental MVP

The results of Spearman’s correlation analysis showed that all quantitative parameters of MCE were negatively correlated with the sLS values, including the sA value (r = − 0.862, P < 0.001), sβ value (r = − 0.772, P < 0.001) and sA × β value (r = − 0.539, P < 0.001, Table 5; Fig. 5).

Table 5 Correlation analysis between segment strain and myocardial perfusion

Fig. 4

Comparison of ROC curves of sWM, sAxβ, and sLS to identify segmental MVP was done by the DeLong method

Fig. 5

Scatter plot of the correlation analysis between the quantitative parameters of MCE and sLS. A. The relationship between the A value and sLS value, r = 0.8624. B. The relationship between the β value and sLS value, r = 0.7725. C. The relationship between the A × β value and IMR value, r = 0.8393

Our study included 25 patients with culprit vessel location in the LAD territory (41.0%), 6 patients in the LCX territory (9.8%), and 30 patients in the RCA territory (49.2%). Most patients revascularized in the LAD had significantly lower values of GLS and a higher WMSI and PSI than those of patients corresponding to the LCX and RCA territories (P < 0.05). Patients with culprit lesions in the LAD territory had a significantly higher MVO rate than that of patients corresponding to the non-LAD territory (P < 0.001). Patients with culprit lesions in the LCX territory had lower S’ and TAPSE. The parameters are presented in Table 6.

Table 6 Comparison of clinical and echocardiographic characteristics data among the three culprit vessel groups

Event-free survival analysis

Over a median follow-up of 12.0 (interquartile range, 9–14) months, 19 patients (31.1%) had admission for congestive angina pectoris (n = 9), congestive heart failure (n = 5), recurrent MI (n = 3) and death(n = 2). The results of the parameter comparison between the two are shown in Table 7. Compared with patients with cardiac events at follow-up, patients without cardiac events had significantly higher values of rMVP (1.94 ± 0.47 vs. 1.69 ± 0.38, P < 0.05), WMSI (1.92 ± 0.67 vs. 1.45 ± 0.42, P < 0.05), E/e’(12.02 ± 5.46 vs. 9.27 ± 2.64, P < 0.05), and NT-proBNP (1954.18 ± 1769.86 vs. 686.32 ± 917.07, P < 0.05) and lower values of rAxβ (1.37 ± 0.67 vs. 1.74 ± 0.57, P < 0.05), rLS (12.2 ± 1.21 vs. 13.12 ± 0.73, P < 0.05) and LVGLS (12.95 ± 1.29 vs. 13.86 ± 0.87, P < 0.05).

Table 7 Comparison of clinical and echocardiographic characteristics data between the MACEs group and the No-MACEs groups

Table 8 shows that in univariate Cox proportional hazard analyses, NT-proBNP, with diabetes, LVEF, E/e’, WMSI, PSI, LVGLS, rLS, rMVP and rAxβ is associated with major adverse events in STEMI patients (all P < 0.05). The influencing factors of major adverse clinical events were included as covariates in the multivariable Cox regression analysis. The results showed that rAxβ and rLS were independent predictors of major adverse clinical events after STEMI.

Table 8 Uni-and multivariate Cox proportional hazard models for cardiac events

ROC curve analysis was used to evaluate the accuracy of rLS and rAxβ in predicting cardiac events (Fig. 6). The optimal cutoff value of rLS and rAxβ for predicting the occurrence of major adverse events were −12.97% (sensitivity: 84.21%, specificity: 76.29%) and 1.623 (sensitivity: 89.47%, specificity: 69.05%), respectively. And the AUC of rLS combined with rAxβ in predicting major adverse events increased slightly. Then, the optimal cutoff values of rLS and rAxβ for cardiac events were used to dichotomize the patient population in Kaplan-Meier analysis. Based on this analysis, the results demonstrated a significant reduction in event-free survival for patients with an rLS > -12.97% when compared with patients with an rLS < -12.97% (log-rank χ 2 = 20.14, P < 0.001), rAxβ < 1.623 compared with rAxβ > 1.623 (log-rank χ 2 = 17.57, P < 0.001). (all P < 0.05; Fig. 7)

Fig. 6

Comparison of ROC curves of rLS, rAx β, and rLS combined with rAx β for predicting major adverse clinical events was done by the DeLong method

Fig. 7

Kaplan–Meier Survival Curve for the prediction of cardiac event

Intra‑and inter‑observer variability

Segmental LS and Axβ were assessed in 10 patients by two independent observers and were also repeatedly evaluated by the same observer on the same day. The intraobserver ICC was 0.919 (95% CI, 0.863–0.952), 0.977 (95% CI, 0.962–0.987) for sLS and sAxβ, respectively. The ICC between the two observers was 0.904 (95% CI, 0.837–0.943), 0.964 (95% CI, 0.939–0.978) for sLS and sAxβ, respectively. The Bland-Altman plots of segmental LS and Axβ are shown in Fig. 8.

Fig. 8

The Bland-Altman plots of segmental LS and Axβ. A. intra- and inter-Bland-Altman plots of sLS; B. intra- and inter-Bland-Altman plots of sAxβ

Discussion

The present study demonstrates good clinical feasibility of STE-based LS and MCE-based qMVP for detection of early global and regional myocardial dysfunction caused by abnormal myocardial perfusion in symptomatic STEMI patients. The prognostic of rLS and r-qMVP in patients with STEMI after pPCI, as follows: (1) reductions in rLS and r-qMVP are both strongly correlated to cardiac events; (2) and the ability of the combination of rLS and r-qMVP to predict clinical outcomes is superior to that of a single indicator. Furthermore, numerous studies have shown that even after successful pPCI treatment in patients with STEMI, there remains a significant risk of slow or no reflow phenomena, which was primarily related to abnormal myocardial perfusion at the microcirculatory level [18,19,20]. The presence of no reflow in STEMI patients contributes to LV remodeling and worse outcomes such as increased mortality and heart failure [21]. The present study suggests that, despite achieving TIMI grade 3 blood flow following successful pPCI in the culprit vessels, segmental LS, qMVP and MW abnormalities were frequently found within 72 h after PCI. Compared with segments of normal MVP and dMVP, segments with MVO showed more reduced qMVP, sWM and sLS.

In STEMI patients, myocardial ischemia is a strong predictor of ventricular remodeling or cardiovascular death. However, neither coronary angiography (CAG) nor routine noninvasive echocardiographic testing is sufficient to detect early myocardial ischemia and then guide treatment [22]. At the early stage of myocardial ischemia, subendocardial myocardial fibers, which are characterized by longitudinal motion, are primarily affected. This results in attenuated deformation, which is measurable by LS. The objective of pPCI to reopen the narrowed or even occluded coronaries in patients with STEMI is to restore normal myocardial perfusion, which is crucial for maintaining myocardial viability [23, 24]. Despite successful revascularization of the infarct-related artery (IRA), extensive research indicates a persistently high incidence of MVO and coronary microvascular dysfunction (CMD) [25]. Myocardial blood flow reserve reflects the coronary circulation’s maximal ability to dilate, thereby increasing blood flow in response to heightened myocardial metabolic demands, such as those occurring during ischemic events [26]. MCE is a novel ultrasound technique that has been widely adopted in clinical practice in recent years. Previous studies have demonstrated that myocardial perfusion assessments using MCE are in close agreement with myocardial blood flow measurements obtained by single-photon emission computed tomography (SPECT) and positron emission tomography (PET). Moreover, the quantitative diagnostic capability of MCE is comparable to, or even exceeds, that of SPECT [23, 27]. Lyu, Wei-Yang et al. [20] undertook a study to investigate the relationship between MCE parameters and microcirculatory resistance (IMR) in patients who underwent primary PCI. Their findings highlighted a significant correlation between the A value, β value, and A × β value of MCE with the IMR index, suggesting that MCE can serve as a non-invasive method to assess microcirculatory health in these patients. 2D STE-based LS and MCE-based qMVP indices, as advanced parameters, are more sensitive than conventional echocardiographic parameters in quantitative assessment of myocardial function and microvascular perfusion, which can recognize subclinical LV dysfunction and have been identified in multiple fields, such as chronic coronary syndrome and cardiac hypertrophy [15, 28,29,30].

MVO is a critical factor affecting the prognosis of patients undergoing PCI after STEMI. Understanding the predictors of MVO can significantly aid in the stratification and management of these patients. In a pivotal study, Xie et al. [18] reported that the incidence of MVO ranged from 35 to 39% in patients who were successfully treated with primary PCI, as detected by MCE. Their analysis identified the location of the STEMI, specifically whether it involved the LAD versus non-LAD territories, and a pre-PCI TIMI coronary flow grade of zero as independent predictors of MVO. Consistent with Xie et al.’s findings, our study observed a similar incidence of MVO at 40.6%. Importantly, we also found that patients with culprit lesions in the LAD territory exhibited a higher rate of MVO compared to those with lesions in non-LAD territories. This correlation suggests that the anatomical features and the hemodynamic impact of the LAD territory in STEMI might predispose to more severe microvascular damage, reinforcing the need for targeted therapeutic strategies in such cases. These findings underscore the importance of identifying patients at higher risk of MVO based on lesion location and initial coronary flow characteristics for more tailored intervention approaches. Further investigation into adjunctive therapies aimed at reducing MVO incidence in high-risk groups, such as those with LAD involvement, may offer improved clinical outcomes.

Global strain and microvascular perfusion patterns in STEMI

AMI is marked by regional myocardial damage which leads to impairments in both systolic and diastolic functions, ultimately affecting the heart’s overall performance and increasing the risk of cardiac events [31, 32]. For several decades, researchers have concentrated on understanding the pathophysiology and prognosis of LV systolic dysfunction following AMI and have shown that the reversibility of myocardial perfusion injury after infarction is a critical determinant of adverse cardiovascular events [33, 34]. Ventricular function measured by LVGLS was lower in our major adverse cardiovascular events (MACEs) group than in our no-MACEs group. STE can distinguish abnormal myocardial deformation and detect subclinical LV dysfunction, and its role of STE in STEMI has already been demonstrated. Our previous research concluded that both GLS and peak atrial longitudinal strain (PALS) independently predict adverse left ventricular remodeling and clinical outcomes following STEMI [35]. Ersbøll, Mads et al.’s study found that GLS is a strong predictor of cardiovascular adverse events in patients with acute myocardial infarction. Patients with reduced GLS values were more likely to experience adverse outcomes [36].

Studies have shown that promptly addressing microcirculatory disturbances can significantly improve outcomes in patients who have experienced myocardial infarction [37, 38]. Therefore, the assessment of MVP in patients with STEMI following pPCI is of crucial importance for treatment strategies and prognostic evaluation. Abdelmoneim, Sahar S., et al. [15] conducted a comprehensive study demonstrating that both qualitative and quantitative MCE performed within 24 h of STEMI in patients undergoing acute revascularization procedures can effectively predict future cardiac events. This approach allows for a nuanced understanding of microvascular health, which is crucial for patient prognosis and management. In contrast, the study by Siyao Sun et al. [29] found that MVP was not significantly associated with cardiac events. However, this finding may be influenced by a small sample size and the combined analysis of weak MVP (dMVP, moderate-risk group) with normal MVP (nMVP) patients in the outcome analysis, leading to aβerror. This methodological limitation may have diluted the potential association between MVP and cardiac events, suggesting that further research with larger cohorts and refined stratification is necessary to clarify these relationships. Our findings align with those of Siyao Sun et al., while the differences with the study by Abdelmoneim, Sahar S. et al. highlight the complexity of MVP assessment and its impact on clinical outcomes. This underscores the need for robust study designs and adequate sample sizes to accurately assess the prognostic value of MVP in STEMI patients. Future research should aim to elucidate the impact of different MVP levels on cardiac events to enhance risk stratification and optimize patient care.

Segmental strain and microvascular perfusion patterns in STEMI

Ito Het al.’s shown that evaluating microvascular perfusion after pPCI is crucial for achieving better functional and clinical outcomes in patients with acute myocardial infarction [39]. However, in a study investigating the risk factors for left ventricular thrombus (LVT) as a complication of myocardial infarction (MI), it was concluded that regional longitudinal strain impairment is a more sensitive risk factor for LVT formation [40]. In the realm of cardiomyopathies characterized by increased left ventricular (LV) wall thickness, such as hypertrophic cardiomyopathy (HCM) and cardiac amyloidosis, distinct patterns of regional LV longitudinal strain have been observed [41]. As noted by previous studies, cardiac amyloidosis typically exhibits superior longitudinal strain values in the apical region relative to the base, whereas apical HCM demonstrates an inverse pattern. These findings are pivotal for elucidating the underlying pathophysiological mechanisms and informing clinical management strategies [42]. Myocardial infarction, similar to diseases such as HCM and cardiac amyloidosis, may present with regional LV systolic function that is more indicative than global systolic function, which is typically assessed by evaluating the presence of wall motion abnormalities [43, 44]. Hence, the purpose of our study was to investigate the association of regional quantitative myocardial perfusion with regional longitudinal strain with adverse cardiovascular events. Our study indicates that rAxβ and rLS are independent predictors of cardiovascular adverse events in STEMI patients, with AUC of 0.8246 and 0.8365, respectively, and the cutoff values are 1.263 for rAxβ and − 12.97% for rLS. When used together, the combined area under the curve slightly increases to 0.8697. Compared with global parameters of LV function, rAxβ and rLS have shown high accuracy in distinguishing transmural myocardial infarction. Studies indicate that both global and regional longitudinal strain can detect myocardial dysfunction at an early stage and may serve as suitable targets for preventive strategies. This supports the contribution of GLS to the clinical assessment of heart failure with preserved ejection fraction (HFpEF) in patients with coronary artery disease and hypertension [45]. However, its application has not been extended to other clinical practices. Thus, our study is the first to test the usefulness of combining regional LS and quantitative MVP to predict MACEs in STEMI, and our results support the usefulness of rLS and rAxβ in predicting cardiac events. Nevertheless, as highlighted by recent research, the clinical application of regional longitudinal strain assessment remains limited. This is primarily due to the significant variability in measurements, which poses a challenge to its widespread adoption despite its potential for precise myocardial damage identification [46].

Limitations

This study has several limitations. First, the small sample size and single-center design may limit the generalizability of the findings. Most patients were unable to undergo MCE within 72 h, and the study population was restricted to low-risk patients experiencing their first STEMI, introducing potential selection bias. Second, the study lacked data from other imaging modalities, such as delayed enhancement magnetic resonance imaging and SPECT myocardial perfusion imaging, which could provide additional insights into myocardial tissue characterization. Third, follow-up was limited to cardiovascular events, with no assessment of resting left ventricular remodeling and function. Fourth, the use of bolus injection for the ultrasound-enhancing agent may have introduced variability in myocardial perfusion imaging, despite strict control of dose and injection speed. Future studies should consider constant infusion methods to improve imaging stability. Fifth, the lack of detailed analysis of replenishment dynamics in myocardial perfusion limited the ability to distinguish between normal perfusion and late replenishment, which reflects impaired microcirculatory function. Finally, the inclusion of both single-vessel disease and multivessel disease patients without subgroup analysis may have introduced heterogeneity into the results. Future multicenter, prospective studies with larger sample sizes and broader patient profiles are needed to address these limitations.

Conclusion

Segmental LS indices are correlated with qMVP within the infarct zone following reperfused STEMI. In patients with STEMI in any location treated with pPCI, both rLS and r-qMVP are more sensitive to myocardial damage and provide prognostic value for clinical events after STEMI. And the combination of rLS and r-qMVP shows improved predictive ability compared to a single indicator, as supported by statistical analysis.