Introduction

In the course of the past 30 years, the mortality rate during the acute phase of ST-elevation myocardial infarction (STEMI) has declined steadily and appeared now to have reached a plateau at lower values when compared with those in the pre-reperfusion era [1]. However, the success of emergency coronary reperfusion therapy in STEMI is usually limited by tissue perfusion failure [2], which has been shown to increase the risk of future cardiovascular events [3]. Several published trials have provided evidence that MVO is the best predictor for prognostic value among all CMR parameters, including clinical scores, left ventricular ejection fraction, and infarct size [4,5,6]. Microvascular injury after acute myocardial infarction affects local T1 value [7]. However, in patients with acute myocardial infarction (AMI), myocardial tissue injury and cardiac remodeling extend beyond the region supplied by the culprit artery; they also affect the remote, non-infarcted myocardium [8].

Several studies have reported remote myocardium alterations in patients with AMI, both in animal and clinical studies [9,10,11,12]. A study by Carrick et al. [13] showed that in remote zone native T1 mapping, early changes may be associated with the occurrence of early post-infarction remodeling, and in a secondary analysis, adverse outcome. However, it was not evaluated whether independent and incremental prognostic information was provided by the remote zone native T1 values over established CMR markers of infarct severity. Reinstadler et al. [14] reported that in addition to clinical risk factors and traditional CMR outcome markers, independent and incremental prognostic information was provided by remote zone alterations by native T1 mapping. The reproducibility of native T1 mapping is excellent with significant regional differences [15]. Thus, T1 of remote zone alterations may become a novel therapeutic target and a useful parameter for optimized risk stratification. Previous studies mainly focused on remote myocardium alterations in acute phase and do not evaluate the impact of MVO on remote myocardium alterations [14, 16]. Consequently, the promising role of dynamic remote myocardium alterations with MVO for the prediction of hard clinical events, and especially its potential incremental prognostic value compared with other markers of infarct severity, remains uncertain.

We designed a longitudinal clinical study in which patients with STEMI successfully treated by primary angioplasty were prospectively recruited, and a CMR was performed within the first week post-reperfusion, on day 30, and at 6 months. The impact of the dynamic change on post-MI CMR measures of remote myocardium was evaluated in reperfused MI by performing T1 mapping and a comprehensive serial CMR imaging study. Our study aimed to investigate the correlations between T1 mapping in remote myocardium and microvascular impairment and to comprehensively assess the value of T1 mapping as a prognostic indicator in patients with STEMI treated by primary percutaneous coronary intervention (PPCI).

Materials and methods

Study population and clinical endpoints

This was a prospective observational investigator-led study conducted at the Renji Hospital between June 2015 and December 2018. Patients with STEMI, who underwent PPCI within 12 h of symptom onset, were prospectively enrolled in our study.

Our study protocol was approved by the institutional ethics committee and was also in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants. Fifty age-matched normal participants were recruited as the control group to acquire the normal reference range. Exclusion criteria for participants included the commonly accepted contraindications to CMR, such as the usage of devices (implantable cardioverter-defibrillators, pacemakers, and cerebral aneurysm clips), non-ischemic cardiomyopathy (amyloidosis, cardiomyopathy due to iron deposition, valvular heart disease, evidence of inflammatory processes or Anderson–Fabry disease, and so on), coronary artery bypass grafting, previous AMI, severe claustrophobia, estimated glomerular filtration rate < 30 ml/min/1.73 m2, and/or significant arrhythmias. The clinical endpoint was major adverse cardiac events (MACE) composite of death, myocardial reinfarction, malignant arrhythmia, and hospitalization for heart failure within 3 years.

CMR imaging

CMR examinations were performed within 1 week, 30 days, and 6 months after STEMI with a 3.0 T scanner (Ingenia, Philips, Best, The Netherlands). The parameters of CMR sequences (SSFP, steady-state free-precession; T2WI-STIR, T2-weighted short-tau triple inversion recovery; native T1 mapping; T2* mapping; and first pass perfusion and LGE) are described in Additional file 1: Table S1.

Analysis of CMR images

Cardiac MR images were analyzed by two radiologists from our team with at least 5 years’ experience (Lian-Ming Wu and Bing-Hua Chen, with over 10 and 7 years) in CMR diagnostic imaging. The LV functional parameters (volumes, mass, and function) and morphology were qualified and quantified with commercial software Cvi42 Version 5.11.3 (Circle Cardiovascular Imaging Inc., Calgary, Canada). We defined remote myocardium as myocardium without evidence of infarction, edema, or enhancement [13]. The ROI of T1 mapping was placed in the territory of non-culprit vessel, which located in the myocardial segment 180°from the infarcted territory [17]. ROI in the remote myocardium was taken in the same place across the 3 scans.

Transmurality was estimated by using the centerline chord method as previous study proposed during the acute stage [18]. Microvascular obstruction (MVO) refers to a dark hypointense core within the regions of hyperenhancement on LGE during the first week after reperfusion. Intramyocardial hemorrhage (IMH) is defined as a hypointense core within the myocardial infarction zone on T2* mapping [17, 19]. Special care should be taken to have sufficient distance from adjacent tissue, such as the lungs or blood, to avoid partial volume artifacts [20]. The infarct volume fraction was quantified in LGE with full-width at half-maximum (FWHM) technique [21].

Statistical analysis

Statistical analysis was performed using SPSS Statistics version 23.0, (IBM SPSS Inc., Chicago, Illinois, USA) and MedCalc Software version 11.4.2.0 (MedCalc Software, Ostend, Belgium). Normality of variable distribution was determined by the Kolmogorov–Smirnov test and qualitative inspection of Q–Q plots. Continuous variables with normal distribution, expressed as mean ± SD, were compared using the independent t test of two samples. Non-normally distributed variables, expressed as median and inter-quartile ranges, were compared with Kruskal–Wallis test. Frequency (percentage) for categorical data of different groups was compared using the chi-square test or Fisher exact test. The optimal cutoff values of remote myocardial T1 for the prediction of MACE were calculated by the Youden index value. MACE-free survival was described by the Kaplan–Meier method. To identify predictors of MACE, we performed univariable and multivariable Cox regression analyses. To determine independent associations with MACE during follow-up (adjusted hazard ratio [95% CI]), we performed multivariable analysis with the forward selection (likelihood ratio) modeling. Five separate models were performed to ensure statistical robustness of the Cox regression analysis. C-statistics were used to compare different parameters for predicting MACE. A 2-tailed p < 0.05 was considered statistically significant.

Results

Study population

Of 152 patients with STEMI included in our prospective observational study, 135 patients (mean age 60.72 years; 12.70% female, median follow-up of 510 days) underwent the standardized CMR protocol at least twice (Fig. 1). Overall, 86 patients with MVO and 49 patients without MVO were followed up for three years. Participants were dichotomized according to whether they were complicated with MVO (86 patients with MVO).

Fig. 1
figure 1

Study flowchart. A total of 135 participants were available at the end of our present analysis. Fifty normal participants were recruited as the control group

Full size image

Baseline characteristics

The patients’ baseline characteristics are summarized in Table 1. Patients with MVO tended to have higher peaks of CK-MB (p = 0.006), cTnI (p < 0.001), BNP (p = 0.020), and CRP (p < 0.001) and were more likely to have MACE (25.60% vs. 8.20%, p = 0.013). The obstruction of culprit vessels occurred more often in the proximal segments in patients with MVO than without MVO (p = 0.002). TIMI flow grade was lower in patients with MVO during pre (p = 0.030)- and post-PCI (p = 0.040).

Table 1 Patients baseline characteristics
Full size table

CMR findings

Patients with MVO tended to have lower LVEF on three occasions: within 1 week (p = 0.001), 30 days (p = 0.001), and 6 months (p = 0.003). They were also more likely to have higher LGE volume on three occasions: within 1 week (p < 0.001), 30 days (p < 0.001), and 6 months (p = 0.002) (Table 2). Transmural infarction (p < 0.001), pericardial effusion (p < 0.001), and IMH (p < 0.001) occurred more often in patients with MVO.

Table 2 Left ventricular function and tissue characteristics of CMR in dynamic evolution
Full size table

Native T1 values of remote myocardium in patients with and without MVO changed from 1 week to 6 months after MI dynamically. The native T1 value of remote myocardium in the first week was higher than those of 1 month. Compensatory thickening of the basal left ventricular septum was greater in patients with MVO than those without MVO (Fig. 2). At 1 week and 30 days, remote myocardium T1 values of group with MVO were higher than those without MVO (p = 0.030 and p = 0.001, respectively), while differences were not significant at 6 months (p = 0.09). In patients with and without MVO, remote T1 values were lowest at 30 days and highest at 6 months (Fig. 3A). In patients with and without MACE, remote T1 values were lowest at 30 days (Fig. 3B). Patients with MACE tended to have higher native T1 values (Additional file 1: Table S2).

Fig. 2
figure 2

Native T1 values of remote myocardium in patients with and without MVO changed from 1 week to 6 months after MI dynamically. The native T1 value of remote myocardium in the first week was higher than those of 1 month. Inflammation of remote myocardium increased in the acute phase while diminished gradually during follow-up, and fibrosis of remote myocardium increased gradually. Native T1 values were determined by myocardial fibrosis in the chronic stage. The first patient had mural thrombosis in the apical during the first week. Compensatory thickening of the basal left ventricular septum was greater in patients with MVO than those without MVO

Full size image
Fig. 3
figure 3

The box-plots (25th percentile, median, and 75th percentile) represent dynamic changes in native T1 values of remote myocardium on three occasions. In patients with and without MVO, remote T1 values were lowest at 30 days and highest at 6 months (A). In patients with and without MACE, remote T1 values were lowest at 30 days (B)

Full size image

Endpoints and clinical outcome

During a median (510 days) follow-up, a total of 26 (19.26%) MACE (death, n = 2 [1.48%]; myocardial reinfarction, n = 4 [2.96%]; ventricular tachycardia, n = 1 [0.74%]; and hospitalization for heart failure, n = 19 [14.07%]) were observed.

In the univariable Cox regression analysis (Table 3), patients with the following characteristics were significantly associated with MACE: higher Killip class (HR 2.47, 95%CI 1.53–4, p < 0.001), higher cTnImax (HR 1.01, 95%CI 1.001–1.02, p = 0.030), higher CRPmax (HR 1.02, 95%CI 1.01–1.03, p = 0.004), lower TIMI flow grade of post-PCI (HR 0.37, 95%CI 0.20–0.70, p = 0.002), lower LVEF1w (HR 0.93, 95%CI 0.90–0.97, p < 0.001), lower LVEF6M (HR 0.95, 95%CI 0.92–0.99, p = 0.013), higher LGE1w (HR 1.10, 95%CI 1.06–1.14, p < 0.001), higher LGE30D (HR 1.12, 95%CI 1.08–1.17, p < 0.001), higher LGE6M (HR 1.05, 95%CI 1.01–1.09, p < 0.001), higher remote native1w T1 (HR 1.02, 95%CI 1.004–1.03, p = 0.009), higher remote native30D T1 (HR 1.04, 95%CI 1.03–1.05, p < 0.001), higher remote native6M T1 (HR 1.02, 95%CI 1.01–1.03, p = 0.004), higher frequency of transmural infarction (HR 5.26, 95%CI 1.24–22.27, p = 0.024), IMH (HR 3.57, 95%CI 1.07–11.93, p = 0.038), and MVO (HR 2.95, 95%CI 1.01–8.56, p = 0.047).

Table 3 Clinical and CMR predictors of MACE in univariable cox regression analysis for patients with and without MVO
Full size table

For univariable Cox regression analysis, p ≤ 0.1 was included into stepwise multivariable Cox regression analysis. Native1w T1 (HR 1.03, 95%CI 1.01–1.04, p = 0.002), Native30D T1 (HR 1.05, 95%CI 1.03–1.07, p < 0.001), and LGE (HR 1.10, 95%CI 1.05–1.15, p < 0.001) were joint independent predictors of MACE during mid-term follow-up for all patients with STEMI after PPCI (Table 4). In multivariable cox regression analysis of 86 patients with MVO, native30D T1 (HR 1.05, 95%CI 1.04–1.07, p < 0.001) and LGE (HR 1.10, 95%CI 1.05–1.15, p < 0.001) were joint independent predictors of MACE (Additional file 1: Table S4). Native T1 of remote myocardium surpasses predictive value of MVO or LGE in C-statistics of all patients (p < 0.001). And native30D T1 value was stronger independent predictor than native1w T1 in C-statistics of patients with MVO (p < 0.001) (Fig. 4).

Table 4 Clinical and CMR predictors of MACE in multivariable cox regression analysis for all patients
Full size table
Fig. 4
figure 4

Kaplan–Meier curve for the MACE-free survival rate of three groups. a MVO present versus absent (p = 0.037). b Native T11w > 1261 ms versus ≤ 1261 ms (p < 0.001); c Native T130D > 1245 ms versus ≤ 1245 ms (p < 0.001)

Full size image

Discussion

Our comprehensive, longitudinal CMR investigation demonstrated that T1 mapping of remote myocardium changed during periods of approximately 1 week, 30 days, and 6 months. Patients with MVO following reperfused STEMI showed a substantial correlation with changes in T1 mapping of the remote myocardium. The principal findings included: (1) The severity of microvascular impairment after reperfusion is correlated with the evolution of native T1 values in remote myocardium. Patients who suffered from severe microvascular impairment display higher T1 values, which is consistent with fibrosis expansion. (2) Native1w T1 value, native30D T1 value, and LGE were joint independent predictors of MACE during mid-term follow-up for patients with STEMI after PPCI. (3) In reperfused STEMI patients with MVO, native30D T1 and LGE were joint independent predictors of MACE.

Accurate noninvasive quantification and detection of remote myocardium alteration in reperfused STEMI patients are scientifically and clinically invaluable for their usefulness in prognosis prediction. Carrick et al. [13] found that early inflammation post-MI is associated with native T1 mapping. Several studies have confirmed that the native T1 of remote myocardium in the acute stage provided independent prognostic information for cardiac recovery and MACE [13, 14, 16]. However, none of the former studies revealed differences in native T1 values between STEMI patients with and without MVO. MVO after STEMI has a trend of dynamic change, which appears in the first week and disappears by day 30. MBF of infarcted myocardium significantly increased at 6 months compared with 1 to 3 days after STEMI [22]. Our findings unequivocally demonstrate that the dynamic changes in native T1 mapping from the acute stage to approximately 30 days, and approximately 6 months.

Local and systemic inflammatory responses were triggered due to ischemia and reperfusion in the acute stage after STEMI. Aggregation of neutrophil and platelets will further obliterate the microvascular lumen, which will then induce further microvascular impairment, ultimately leading to the production of vasoconstrictors and inflammatory mediators [23]. Acute STEMI will trigger a spectrum of alterations in the remote vessels, including upregulation of the platelet-endothelial adhesion from endothelial-associated von Willebrand factor multimers and endothelial inflammatory adhesion molecules [24]. The activation of leukocyte infiltration and proinflammatory pathways is related to the response of the remote myocardium [25]. The remodeling process post-infarction has been increasingly recognized as a cause of inflammatory response [26]. Native T1 values were determined by myocardial water content and cellularity [27]. Increased T1 of myocardium with inflammatory cell infiltration was confirmed by histopathology in patients with the features of acute rejection who underwent cardiac transplant [28]. Therefore, myocardial edema and hypercellularity, due to the inflammation during the acute phase post-STEMI, are the reasons why there has been an increase in the remote myocardial native T1 in the acute stage [13, 14]. Most patients with MVO also had IMH implicated severe microvascular impairment. Persistent crystallized iron deposition from MVO complicated with IMH is directly associated with left ventricular adverse remodeling in the chronic stage after MI, infarct resorption, and proinflammatory burden [29].

Inflammation of remote myocardium increased in the acute phase while diminished gradually during follow-up, and fibrosis of remote myocardium increased gradually. Native T1 values were determined by myocardial fibrosis in the chronic stage [30]. These may be the main reasons for the bimodal behavior of the T1 value in the remote myocardium. Diffuse myocardial fibrosis can cause the increase in native T1 [11]. In the chronic phase, prolonged T1 values correlate with the severity of myocardial interstitial fibrosis [31], which has been advocated as the sensitive marker to differentiate between fibrosis and healthy myocardium [32].

Microvascular dilation and increased blood volume may cause T1 values to increase after myocardial ischemia [33]. Another key explanation for the higher T1 values at first week, compared to values at one month, may be that myocardial blood flow of remote regions increased after MI, reflecting the hyperdynamic performance of remote myocardium [34]. Coronary microvascular injury and obstruction occur in approximately half of patients after successful primary PCI and is associated with worse outcomes compared to those without microvascular injury [3]. Severe microvascular impairment will cause a stronger systemic response, including both the infarcted and non-infarcted myocardium. Compensatory enhancement of systolic function in remote myocardium leads to compensatory cellular hypertrophy, hyperkinesis after STEMI, and cellular dysfunction [11].

Our study has several limitations. First, our sample size was small. However, a total of 341 CMR scans were performed (within 1 week = 135, 30 days = 134, 6 months = 72) on 135 patients, with each patient undergoing at least two scans. Thus, it is necessary to confirm our findings in a large-scale study. Nevertheless, our data—baseline characteristics, CMR results on infarct severity, and incidence of MACE—are comparable with other recent CMR studies [4]. In light of the fact that patients were scanned successively, we kept our CMR protocol as short as possible. Instead, both myocardium edema and hemorrhage were assessed by T2W-STIR imaging, a sequence validated and used in many published papers for these purposes. Patients with anterior STEMI, and also patients with inferolateral wall STEMI, were included in our clinical study. This may possibly induce magnetic-field non-homogeneity associated with the inferolateral wall. Additionally, care should be taken when extrapolating results to lateral MI locations, especially when attempting to adequately visualize the phenomenon, as signal loss attributable to through-plane cardiac motion may occur [35]. Wall stress may affect the pathophysiology of the remote zone in STEMI. Future studies are needed to examine wall stress and determine the clinical significance of native T1 changes in the longer-term follow-up.

To summarize, our study is the first longitudinal pathophysiological study of remote myocardium that evaluated mid-term outcomes through CMR. Native T1 mapping could detect myocardial abnormalities in remote myocardium that may be neglected by conventional LGE. The early detection of diffuse myocardial pathological alterations in the remote myocardium among survivors of STEMI allows for a more timely and individualized treatment and disease-specific therapy. The assessment of dynamic evolution of remote myocardium and the adjudication of outcome events are other strengths of the study.

Conclusion

The evolution of native T1 in remote myocardium associated with the extent of microvascular impairment after reperfusion injury. Native1w T1 value, native30D T1 value, and LGE were joint independent predictors of MACE during mid-term follow-up for all patients with STEMI after PPCI. In reperfused STEMI patients with MVO, native30D T1 and LGE were joint independent predictors of MACE. These findings may provide insight into the assessment of LV remodeling and prognosis after myocardial infarction.