Prognostic value of cardiac magnetic resonance pulmonary transit time and myocardial strain across different stages of ST-elevation myocardial infarction

Abstract

Background

This study evaluated the prognostic value of pulmonary transit time (PTT) and myocardial strain in ST-elevation myocardial infarction (STEMI) patients stratified by the Canadian Cardiovascular Society (CCS) classification.

Material and methods

A total of 521 STEMI patients underwent 3.0-T cardiac magnetic resonance (CMR) imaging between January 2017 and December 2025. PTT was measured as the number of cardiac cycles for contrast to travel from the right to left ventricle during first-pass perfusion. Global myocardial strain (longitudinal, circumferential, radial) was assessed via CMR feature tracking. Cox regression analyses were used to evaluate associations of PTT, normalized PTT (nPTT), wall motion score index (WMSI), and global longitudinal strain (GLS) with major adverse cardiac events (MACE: death, re-infarction, heart failure).

Results

Of 409 analyzed patients, higher CCS stages were associated with prolonged PTT, greater myocardial injury, and elevated pulmonary blood volume index (PBVI), WMSI, and microvascular obstruction (MVO). Over six months, myocardial strain improved in CCS1–3, while GLS worsened in CCS4. Multivariable analysis identified PBVI, nPTT, PTT, GLS, and WMSI as independent predictors of MACE. Kaplan-Meier analysis showed significantly lower MACE-free survival in CCS4 patients (P < 0.01).

Conclusion

Elevated PTT correlates with severe myocardial injury and adverse outcomes. PBVI, nPTT, PTT, GLS, and WMSI are independent prognostic indicators, providing valuable risk stratification across CCS stages.

Graphical Abstract

Highlights

• •Increased PTT/nPTT linked to severe myocardial impairment and poor prognosis.
• •PBVI, PTT, nPTT, and GLS predict MACE.
• •CCS4 stage reduces MACE-free survival.

What’s new

This study investigates the potential of pulmonary transit time (PTT) and myocardial strain as prognostic tools in ST-elevation myocardial infarction (STEMI) patients. Findings reveal a positive correlation between longer PTT values and the severity of myocardial injury, as categorized by the Canadian Cardiovascular Society AMI classification (CCS-AMI) system. Additionally, PTT-related parameters (Pulmonary blood volume index, normalized pulmonary transit time), wall motion score index (WMSI), and global longitudinal strain (GLS) emerged as independent predictors of mortality, offering valuable prognostic insights across different stages of CCS. Notably, patients in the most severe CCS stage (CCS-4) exhibited significantly lower major adverse cardiac events-free survival, highlighting the importance of these imaging parameters in identifying high-risk patients and guiding clinical management.

1. Introduction

Recent research has definitively shown significant heterogeneity in pathogenesis and outcomes of acute myocardial infarctions (AMI). For example, the vascular damage occurring in reperfused AMIs can contribute to a second wave of myocardial necrosis, leading to an even larger infarct size than that observed at reperfusion [1]. Recently, the Canadian Society of Cardiovascular AMI Classification (CCS-AMI) described four distinct stages of myocardial injury to capture the progressive injury experienced by the myocardium following ischemia and reperfusion [2]. Specifically, CCS-AMI stages reperfused MI into increasing the severity of myocardial damage. In AMI management, clinical attention is acutely focused on the coronary arteries and the timeliness of reperfusion. Despite recent advances in reperfusion therapy, patients surviving AMI are at a risk of near- and long-term complications, both of which can be traced back to the burden of myocardial damage during the acute event [3].

The analysis of cardiopulmonary surrogate markers might assist in risk assessment and forecasting for individuals with AMI [4], [5]. Despite the limited sensitivity and poor prognostic value of physical examinations, cardiac catheterization is the gold standard for assessing hemodynamic congestion, even though it is invasive [6]. The use of imaging modalities to evaluate pulmonary blood flow as a marker of cardiopulmonary hemodynamic congestion has been previously described in pulmonary and cardiac conditions [7], [8], [9]. Pulmonary transit time (PTT) is defined as the time span needed for a bolus of blood or contrast medium to be pumped from the right ventricle (RV) through the pulmonary vasculature to reach the left ventricle (LV). The systolic function of the RV, resistance in the pulmonary vasculature, and LV diastolic dysfunction influence PTT, which is thus considered a quantitative marker of cardiopulmonary status. PTT and its derived parameters are already used as markers of diastolic dysfunction in different cardiac insufficiency scenarios and correlate with many clinical parameters. Therefore, PTT may be useful cardiac magnetic resonance (CMR)-associated tools for diastolic dysfunction in AMI patients and may be used to refine the diagnosis of the disease. Furthermore, various studies have shown that myocardial strain can further investigate the association between diastolic dysfunction and myocardial motion in patients with ST-segment elevation myocardial infarction (STEMI). This knowledge has rapidly moved into the field of CMR using feature tracking (FT). Longitudinal strain (LS) measured by CMR-FT has been proposed as a standard for the accurate quantification of myocardial wall motion in various scenarios [10], [11].

A disrupted PTT has been associated with several conditions, such as chronic heart failure, congenital heart disease, and cardiac surgeries [12], [13], [14]. In a large-scale study of patients with known or suspected coronary artery disease referred for perfusion CMR imaging, researchers discovered a connection between resting PTT and nonfatal major adverse cardiovascular events (MACE), regardless of traditional factors [15]. Meanwhile, CMR-FT has been shown to provide valuable information to characterize motion in distinct stages of the infarcted myocardium, and to monitor its dynamics and structural implications following infarction [16]. Currently, there are limited CMR data on PTT and strain, both shortly after STEMI and in the follow-up period. Furthermore, the potential of PTT or LS to predict MACE after STEMI remains unclear.

In this context, our research focused on (1) understanding the dynamics of PTT versus the healing process of infarcts within the first 6 months after different stages of CCS-STEMI, and (2) the prognostic value of PTT-related parameters and myocardial strain in predicting adverse clinical events after STEMI.

2. Methods

2.1. Study population

This prospective cohort study enrolled consecutive patients with STEMI treated at XuanWu hospital Capital Medical University, from January 2017 to December 2025. The inclusion criteria were as follows [17]: (i) age ≥ 18 years; (ii) type I ST-elevation MI ≤ 7 days before study participation; (iii) coronary angiography performed during index hospitalization; and (iv) only included those with new-onset STEMI. Individuals meeting the following criteria were excluded from the study: those with a history of allergy or hypersensitivity to contrast media, contraindications to CMR, a prior history of cardiac arrest, previous percutaneous coronary intervention (PCI) or coronary artery bypass grafting, known congestive heart failure, end-stage kidney disease or hepatic failure, recent stroke, coagulopathy, or pregnancy, and pulmonary micro- or macro-circulation such as chronic lung disease.

The primary endpoint of the study was the occurrence of MACE, defined as a composite of all-cause death, myocardial re-infarction, and re-hospitalization for new congestive heart failure. Re-infarction was defined according to the redefined ESC/ACC committee criteria [18]. Coronary perfusion before and after PCI was assessed according to the grade flow of thrombolysis in myocardial infarction (TIMI) [19]. Trained personnel, unaware of the initial data, conducted telephone interviews using a standardized questionnaire to inquire about clinical endpoints at 6, 12, and up to 60 months. Afterwards, all declared endpoints were meticulously verified by examining medical records.

The research adhered to the Helsinki guidelines and received approval from the XuanWu hospital Capital Medical University ethics committee (approval number: D2014044). The authors confirmed that patient consent forms were obtained for this study.

2.2. CMR acquisitions

CMR was performed on a whole-body 3.0-T MR scanner (Magnetom Verio; Siemens, Erlangen, Germany) with patients in the supine position. All patients in the study were treated with PCI, and the CMR examination was performed within 1 week of recanalization. All images were acquired at the end-expiratory position of the breath hold with electrocardiogram (ECG) gating. Scout images were used to localize the heart, which was then carefully shimmed. Subsequently, T2 weight-short TI inversion recovery (T2-STIR), post contrast cines, T2* mapping, early gadolinium enhancement (EGE), and late gadolinium enhancement (LGE) images were acquired in the short-axis view, covering the left ventricle from the base to the apex with no gap between slices. Scan parameters for T2-STIR were: repetition time (TR)= 1985 ms, echo time (TE)= 70 ms, flip angle= 180°, slice thickness= 8 mm, image resolution= 1.3 mm× 1.3 mm; Scan parameters cine were: repetition time= 2.84 ms, temporal resolution= 39.2 ms, echo time= 1.25 ms, flip angle= 50°, slice thickness= 8 mm, image resolution= 1.4 mm× 1.8 mm; Scan parameters for EGE and LGE were: repetition time= 904 ms, echo time= 1.98 ms, flip angle= 20°, slice thickness= 8 mm, image resolution= 1.4 mm× 1.9 mm; Scan parameters for T2* mapping were: repetition time= 968 ms, the eight echo times = 2.1, 4.3, 6.4, 8.6, 10.7, 12.8, 15, 17.1, ms, flip angle= 18°, slice thickness= 8 mm, image resolution= 1.6 mm× 1.6 mm. Our detailed imaging protocol, including imaging parameters, has been published earlier [20]. Simply put, 2-chamber, 4-chamber, and short-axis cine imaging stacks were gathered using a breath-hold technique with retrospective ECG-triggered steady-state free-precession imaging. Late gadolinium enhancement imaging was conducted 15 min after an intravenous injection of 0.2 mmol/kg of a gadolinium-based contrast agent (Magnevist, Bayer Healthcare) by an ECG-triggered phase-sensitive inversion recovery sequence to assess infarct size as well as the presence and extent of microvascular obstruction.

For PTT assessment, a first-pass resting-state perfusion sequence was added to the protocol, using a cardiac-gated single-shot 2D saturation recovery gradient echo sequence (true fast imaging with steady-state free precession). The typical imaging parameters were as follows: TR= 134 ms, TE= 1.12 ms, inversion time (TI) = 90 ms, flip angle 15 °, slice thickness= 8 mm, image resolution 2.2 mm × 1.8 mm. Image acquisition was started simultaneously with contrast media injection, and the resting-state perfusion weight imaging (rs-PWI) sequence includes a one-slice four-chamber view and three-slice short-axis images.

2.3. Post-processing and interpretation

The LV and RV volumes and LV mass were assessed on short-axis cine stacks (8–10 slices) using CVI⁴² professional cardiac post-processing software (version 5.12.1 [1688] Circle Cardiovascular Imaging, Calgary, Canada). The papillary muscles were included in the blood pool. Cardiac output was calculated as the product of LV stroke volume and heart rate. Infarct size was evaluated using the recently described + 5 SD technique, using the mean signal intensity of the remote myocardium opposing the infarcted segment as a reference [21]. Regional LV function included reporting a wall motion score index (WMSI). A 5-point wall motion scoring system was used to determine wall motion abnormalities, with derivation of an indexed WMSI specific to the region of infarction, according to the AHA 16-segment model. The WMSI is calculated by summing the individual segmental scores and dividing by the total number of segments assessed. The formula is:

$$\text{WMSI}=\frac{\sum \text{Segment Scores}}{\text{Number of Segments Assessed}}$$ (1)

PTT time-intensity curves were created on Syngo.via imaging software (VB19B, Siemens Healthineers), as used in clinical routine. Circular ROIs were positioned in the basal RV and LV blood pool in the short-axis view of rs-PWI (Fig. 1). Attention was given to prevent partial volume effects at the papillary muscles or myocardial wall, and ROIs were then extended to all images in the stack. The average signal intensity within the ROI was plotted as a function of time, resulting in an indicator-dilution curve. PTT was calculated as the number of cardiac cycles required for a bolus of the contrast agent to move from the RV to LV, whereas normalized PTT (nPTT) represents the PTT interval corrected for the heart rate. Patients’ heart rates at rest were derived from the CMR scans. nPTT was calculated according to Bazett’s formula:

$$\mathit{nPTT}=\frac{\mathit{PTT}}{\surd {\mathit{RR}}_{\mathrm{interval}}}$$ (2)

Fig. 1.

Schematic diagram of pulmonary transit time measurement. (A) Dynamic first-pass perfusion imaging of a basal short-axis slice showing the right ventricular (RV) and left ventricular (LV) cavities. (B and C) The anatomy diagram and matched figure show region of interest (ROI) placement within the basis of the right (green circle) and left (red circle) ventricles. (D) Schematic gadolinium (Gd) time-signal intensity curves in the RV and LV cavities. (E) Schematic region of a real perfusion defect. Abbreviations: LV = left ventricular; RV = right ventricular.

Pulmonary blood volume index (PBVI) was calculated using the volumetric cardiac output (mL/min) as:

$$\mathit{PBVI}=\frac{\mathit{CO}}{60}\times \mathit{PTT}\times \frac{1}{\mathit{BSA}}$$ (3)

where $\mathit{CO}$ represents the cardiac output, and $\mathit{BSA}$ represents the body surface area.

Late enhancement (10–15 min after contrast agent administration) allows the visualization of the areas of irreversible myocardial injury or necrosis during the stages of contrast agent administration. CMR protocol provides insight into the presence and full extent of microvascular obstruction (MVO) and intramyocardial hemorrhage (IMH) postreperfusion, as microvessels in the infarcted area become compressed due to a combination of factors, including edema, inflammation, and intramyocardial hemorrhage. MVO has been defined as the inability to reperfuse a previously ischemic region by EGE (Fig. 2), and IMH has been delineated by T2* mapping. The CCS classification extends the wavefront paradigm and formulates a new categorization of myocardial ischemia and reperfusion injury into four sequential and progressive stages, initiating with myocardial edema, followed by cardiomyocyte necrosis, then MVO, and finally, myocardial hemorrhage (Fig. 3). CCS Stage 1: No or minimal damage to the heart muscle; Stage 2: Damage to the heart muscle and no injury to small blood vessels in the heart; Stage 3: Damage to the heart muscle and blockage of small blood vessels in the heart; Stage 4: Damage to the heart muscle, blockage and rupture of small blood vessels resulting in bleeding into the heart muscle. The injury accumulated at each stage adds to and builds on the injury of the previous stage, following a sequence of escalating severity, which translates into worsening patient prognosis with each stage.

Fig. 2.

Cardiovascular MRI showing an anterior myocardial infarction with a central zone of MVO in CCS grade 3. T2-STIR, EGE, and LGE (short-axis and 2-chamber view) sequentially showing the location and extent (white or red arrows) of myocardial infarction with MVO. Abbreviations: CMR = cardiovascular magnetic resonance; EGE = early gadolinium enhancement; LGE = late gadolinium enhancement; MVO = microvascular obstruction; MI = myocardial infarction; T2-STIR = T2 weight-short TI inversion recovery.

Fig. 3.

The Canadian Cardiovascular Society classification of ST-elevated myocardial infarction. Overview of the most recent Canadian Cardiovascular Society classification of ST-elevated myocardial infarction. The tissue changes and clinical characteristics are displayed for each stage. T2-STIR sequence is used to delineate myocardial edema (white arrows), while LGE delineates myocardial infarction (red arrows). Abbreviations: CCS = Canadian Cardiovascular Society; CE-SSFP = contrast-enhanced steady-state free precession; LGE = late gadolinium enhancement; STEMI = ST-elevated myocardial infarction; T2-STIR = T2 weight-short TI inversion recovery.

In the myocardial strain module, CVI⁴² software semi-automatically tracks and calculates the global radial, circumferential, and LS peaks of the LV, generating a time-strain curve [22].

All CMR images were evaluated by experienced observers who were unaware of the patients’ clinical details and prior examination outcomes during follow-up assessments. A random subset of 20 patients was chosen to assess both inter-observer and intra-observer variability.

2.4. Statistical analysis

All statistical analyses were conducted utilizing MedCalc software (Version 15.6.1; MedCalc Software bvba, Ostend, Belgium). The Kolmogorov-Smirnov test was employed to assess the normality of data distribution. Continuous variables exhibiting a normal distribution are presented as mean ± standard deviation, whereas categorical variables are reported as frequencies or percentages. Variables with a non-normal distribution are expressed as medians accompanied by interquartile ranges. Multiple groups were compared using ANOVA or a nonparametric equivalent. Differences in continuous variables between two groups were assessed using the Mann-Whitney U test, whereas differences in categorical proportions were analyzed via the chi-square test or Fisher’s exact test, depending on suitability. Inter- and intra-observer reliability were evaluated by calculating intra-class correlation coefficients (ICCs) using a two-way mixed-effects model for single measurements. To identify significant and independent predictors of MACE, univariable and multivariable Cox proportional hazards regression analyses were conducted. Variables demonstrating a P-value of less than 0.10 in univariable analyses were subsequently incorporated into multivariable models. MACE-free survival was estimated using the Kaplan-Meier method, with group differences assessed by the log-rank test. Statistical significance was defined as P < 0.05, and Bonferroni correction was applied to adjust for multiple pairwise comparisons.

3. Results

3.1. Patient characteristics

A total of 409 STEMI patients were included in the final analysis, of whom 285 completed clinical follow-up, and 177 underwent a second CMR scan. The patients received drug treatment according to guideline-directed medical therapy (GDMT). A detailed flowchart of the study is presented in Fig. 4. Among the different CCS stages identified by the CMR, CCS2 had the largest number of patients (n = 202), with a mean age of 59.52 ± 12.89 years; 175 (86.7%) were male. The patient baseline characteristics showed no significant differences between groups (P > 0.05). Culprit vessel analysis of four groups of patients showed that LAD accounted for the highest proportion (≥50%). Post-intervention TIMI flow showed that only six patients (1.5%) remained in grade 1. No significant difference was observed in the laboratory indexes of myocardial infarction, culprit lesion, or TIMI grades pre- and post-interventional. Detailed baseline patient characteristics are presented in Table 1.

Fig. 4.

Study flow chart. Abbreviations: MACE = major adverse cardiac events; PTT = pulmonary transit time.

Table 1.

Comparison of patient baseline characteristics.

Characteristics	CCS Stage 1 (n= 66)	CCS Stage 2 (n= 202)	CCS Stage 3 (n= 105)	CCS Stage 4 (n= 36)	p-Value
Age, years	56.00 ± 13.53	59.52 ± 12.89	60.27 ± 13.37	53.86 ± 7.40	0.587
Male, n (%)	58 (87.9)	175 (86.7)	86 (81.9)	24 (66.7)	0.796
Body mass index, kg/m2	24.10 ± 2.51	25.03 ± 2.84	26.75 ± 2.77	25.43 ± 2.96	0.708
Heart rate, beat/min	68.32 ± 8.45	67.33 ± 9.63	65.37 ± 10.21	65.83 ± 8.77	0.756
Time to CMRI (days)	5.1 ± 0.9	5.2 ± 0.8	4.7 ± 0.6	5 [3], [7]	0.561
Cardiovascular risk factors, n (%)
Diabetes mellitus	10 (15.2)	32 (15.8)	17 (16.2)	5 (13.9)	0.644
Hyperlipidemia	37 (56.1)	121 (59.9)	61 (58.1)	20 (55.6)	0.168
Current smoker	39 (59.1)	125 (61.9)	64 (60.9)	22 (61.1)	0.377
Hypertension	40 (60.6)	124 (61.4)	62 (59.0)	18 (50.0)	0.076
Positive family history	17 (25.8)	57 (28.2)	30 (28.6)	12 (30.6)	0.184
Laboratory
TNI, ng/mL	1.74 ± 0.15	1.81 ± 0.19	1.69 ± 0.17	1.92 ± 0.20	0.688
BNP, pg/mL	475.41 ± 105.37	445.20 ± 99.24	417.35 ± 105.21	498.34 ± 89.77	0.075
Culprit lesion, n (%)					0.057
LAD	34 (51.5)	106 (52.5)	54 (51.4)	18 (50.0)
LCX	8 (12.1)	27 (13.4)	13 (12.4)	5 (13.9)
RCA	24 (36.4)	69 (34.2)	38 (36.2)	13 (36.1)
Revascularization
Pre-interventional TIMI flow, n (%)					0.271
0	45 (68.2)	132 (65.3)	63 (60.0)	18 (50.0)
1	10 (15.2)	27 (13.4)	24 (22.9)	15 (41.7)
2	10 (15.2)	39 (19.3)	18 (17.1)	3 (8.3)
3	1 (1.5)	4 (2.0)	0 (0)	0 (0)
Post-interventional TIMI flow, n (%)					0.502
1	0 (0)	2 (1.0)	2 (2.0)	2 (5.6)
2	6 (9.1)	22 (10.9)	13 (12.3)	5 (13.9)
3	60 (90.9)	178 (88.1)	90 (85.7)	29 (80.6)

-Note: Baseline characteristics of the overall population and in groups. Values are either n (%) or median (IQR).

-Abbreviations: CCS = Canadian Cardiovascular Society, BNP = Brain natriuretic peptide, TNI = Cardiac troponin I, LAD = Left anterior descending artery, LCX = Left circumflex artery, RCA = Right coronary artery, TIMI = Thrombolysis in myocardial infarction.

3.2. Cardiac MRI parameters

CMR studies were performed within 1 week following myocardial infarction. Left ventricular ejection fraction gradually decreased with the increase in myocardial injury (from CCS1-4); on the contrary, left ventricular end-diastolic volume and end-systolic volume gradually increased, and the difference between the four groups was statistically significant (P < 0.05). The mean WMSI value were 1.21 ± 0.13 in CCS1, 1.57 ± 0.18 in CCS2, 1.86 ± 0.25 in CCS3, and 2.22 ± 0.16 in CCS4, respectively (P < 0.05). We found that, as the CCS stage increases, cardiac involvement became more significant and the PTT and nPTT became much longer. The specific values of PTT were as follows: CCS1: 8.21 ± 1.98 vs CCS2: 8.66 ± 2.43 vs CCS3: 9.85 ± 2.72 vs CCS4: 10.33 ± 2.98 (P < 0.05); likewise, nPTT were as follows: CCS1: 9.95 ± 2.41 vs CCS2: 10.58 ± 2.67 vs CCS3: 12.22 ± 2.77 vs CCS4: 12.74 ± 3.07 (P < 0.05). Significant differences at baseline were found PBVI in the CCS1-4 group, myocardial infarct size in the CCS2-3 group, and microvascular obstruction in the CCS3-4 group (P < 0.05).

The median WMSI values were 1.86 ± 0.25 in CCS3 and 2.22 ± 0.16 in CCS4 at baseline, compared with 1.65 ± 0.17 and 1.98 ± 0.20, respectively, at follow-up (P < 0.05). Follow-up PTT and nPTT was slightly shortened from baseline at CCS1-3 but prolonged at CCS4. No significance difference was observed in PBVI parameters at CCS1-2 follow-up visits compared with baseline; however, CCS3-4 showed significant changes (CCS3: 316.52 ± 59.58 vs. 372.70 ± 58.35; CCS4: 346.12 ± 70.09 vs. 290.58 ± 60.22; P < 0.05). The myocardial damage decreased after follow-up. Table 2 presents a summary of the imaging parameters at baseline and follow-up CMR in the overall population (Table 2 and Fig. 5).

Table 2.

Comparison of patient CMR imaging parameters.

Characteristics	CCS Stage 1 (n= 66)	CCS Stage 2 (n= 202)	CCS Stage 3 (n= 105)	CCS Stage 4 (n= 36)	p-Value
CMR parameters at baseline
LV ejection fraction, %	49.46 ± 14.87	44.85 ± 11.47	40.47 ± 11.69	30.12 ± 10.96	＜0.001
LV end-diastolic volume, mL	113.11 ± 32.45	127.60 ± 35.39	147.07 ± 43.54	157.64 ± 36.18	＜0.05
LV end-systolic volume, mL	47.73(38.33,65.45)	65.79(53.80,82.90)	78.75(63.24,102.98)	106.54(85.94,111.41)	＜0.001
LV stroke volume, mL	52.83 ± 16.34	56.73 ± 19.76	57.61 ± 18.40	45.49 ± 14.57	0.328
Cardiac output, L/min/m2	3.70 ± 1.21	3.83 ± 1.15	4.02 ± 1.38	3.59 ± 1.10	0.689
RV ejection fraction, %	46.01 ± 11.38	46.88 ± 10.30	46.76 ± 7.91	41.48 ± 17.04	0.598
RV end-diastolic volume, mL	119.45 ± 29.29	123.30 ± 31.34	128.53 ± 32.03	131.11 ± 42.81	0.608
RV end-systolic volume, mL	56.47(47.17, 79.28)	65.75 ± 21.93	66.74 ± 20.69	69.93(46.89, 80.83)	0.347
RV stroke volume, mL	54.11 ± 16.57	57.54 ± 18.61	60.06 ± 17.89	48.50 ± 16.29	0.311
WMSI	1.21 ± 0.13	1.57 ± 0.18	1.86 ± 0.25	2.22 ± 0.16	＜0.05
PTT	8.21 ± 1.98	8.66 ± 2.43	9.85 ± 2.72	10.33 ± 2.98	＜0.05
nPTT	9.95 ± 2.41	10.58 ± 2.67	12.22 ± 2.77	12.74 ± 3.07	＜0.05
PBVI, mL/m2	237.21 ± 55.30	289.01 ± 73.69	372.70 ± 58.35	290.58 ± 60.22	＜0.05
Infarct damage in the acute phase, % of LVMM	-	14.33 ± 1.76	19.82 ± 2.14	18.69 ± 1.98	＜0.05
Microvascular obstruction, % of LVMM	-	-	9.58 ± 1.69	11.15 ± 1.56	＜0.05
CMR parameters at follow-up	(n= 29)	(n= 86)	(n= 44)	(n= 18)
LV ejection fraction, %	57.80(50.88, 64.73)*	44.89 ± 7.12	40.21 ± 11.87	31.64 ± 9.88	＜0.001
LV end-diastolic volume, mL	119.98(98.71,141.25)	127.89(108.27,138.60)	114.38 ± 35.44*	101.91 ± 28.21*	＜0.05
LV end-systolic volume, mL	49.16(38.49, 59.82)	64.47(57.56, 82.98)	74.31 ± 19.28	77.83 ± 21.30*	＜0.01
LV stroke volume, mL	70.83(50.22, 91.43)	59.71 ± 20.83	40.86 ± 17.98	26.76(22.28,37.15)	＜0.001
Cardiac index, L/min/m2	4.80(3.20, 6.39)	3.92 ± 1.03	4.13 ± 1.25	4.46 ± 1.22	0.577
RV ejection fraction, %	55.46(47.29, 63.63)	50.26 ± 6.03	56.36 ± 7.12	47.26 ± 5.88	0.423
RV end-diastolic volume, mL	114.33(104.24,124.43)	125.38 ± 24.54	129.24 ± 27.96	132.89 ± 35.01	0.755
RV end-systolic volume, mL	51.10(45.25,54.95)	61.90 ± 11.61	64.49 ± 15.22	70.08 ± 20.09	0.451
RV stroke volume, mL	64.23(49.30,79.17)	63.49 ± 16.25	65.20 ± 14.82	62.88 ± 19.11	0.345
WMSI	1.20 ± 0.11	1.48 ± 0.15	1.65 ± 0.17*	1.98 ± 0.20*	＜0.05
PTT	7.85 ± 2.01	8.28 ± 1.98	9.37 ± 2.58	11.34 ± 2.70	＜0.05
nPTT	9.12 ± 2.32	10.21 ± 2.42	11.62 ± 3.02	13.98 ± 3.32	＜0.05
PBVI, mL/m2	242.02 ± 57.67	262.85 ± 64.37	316.52 ± 59.58*	346.12 ± 70.09*	＜0.05
Infarct necrosis extent, % of LVMM	-	10.28 ± 1.73*	16.04 ± 1.50	17.37 ± 1.21	＜0.05

-Note: Imaging parameters of the overall population at baseline and 6-month follow-up as well as in groups. Values are either n (%) or median (IQR). Statistically significant p values are printed in bold. * Represents significant difference between baseline and follow-up data.

-Abbreviations: CMR = Cardiac magnetic resonance, LV = Left ventricular, LVMM = Left ventricular myocardial mass, PBVI = Pulmonary blood volume index, PTT = Pulmonary transit time, nPTT = normalized pulmonary transit time, RV = Right ventricular, WMSI = wall motion score index.

Fig. 5.

Comparison of Patient Imaging Parameters. Imaging parameters of the population were taken at baseline and a 6-month follow-up, and in groups of CCS classification. * Represents significant difference between baseline and follow-up data. Abbreviations: LVEF = left ventricular ejection fraction; LVEDV = left ventricular end-diastolic volume; LVESV = left ventricular end-systolic volume; PBVI = pulmonary blood volume index; PTT = pulmonary transit time, WMSI = wall motion score index.

3.3. Strain parameters

At baseline, the global radial strain of CCS grades 3 and 4 was almost half that of CCS grades 1 and 2 (CCS3: 14.6 ± 3.2 and CCS4: 12.3 ± 1.8 versus CCS1: 28.7 ± 3.9 and CCS2: 25.1 ± 4.1; P < 0.05 for both). All global circumferential strain values were significantly reduced in patients from the CCS1 to CCS4 stages (P < 0.05). Similar significant differences were identified in the global longitudinal strain (GLS; P < 0.05). During the 6-month follow-up, myocardial strain improved to different degrees with the recovery of myocardial injury, but the GLS (baseline: −9.9 ± 1.6 versus follow-up: −8.6 [−8.4, −10.0]) of CCS4 still decreased significantly (Table 3).

Table 3.

Global strain parameters analysis.

Characteristics	CCS Stage 1 (n= 66)	CCS Stage 2 (n= 202)	CCS Stage 3 (n= 105)	CCS Stage 4 (n= 36)	p-Value
Baseline peak strain (%)
radial	28.7 ± 3.9	25.1 ± 4.1	14.6 ± 3.2	12.3 ± 1.8	< 0.05
circumferential	−18.5 ± 2.6	−18.8 ± 2.1	−15.0 ± 2.1	−13.7 ± 2.0	< 0.05
longitudinal	−13.8 ± 2.2	−13.7 ± 1.9	−10.9 ± 1.8	−9.9 ± 1.6	< 0.05
Follow-up peak strain (%)	(n= 29)	(n= 86)	(n= 44)	(n= 18)
radial	27.5 ± 2.8	26.6 ± 3.8	12.8 ± 2.7	12.0 (8.5,13.0)	< 0.05
circumferential	−20.6 ± 3.1	−16.5 ± 2.6	−13.9 ± 1.9	−10.8 (-9.3, −12.4)*	< 0.05
longitudinal	−13.5 ± 2.7	−11.0 ± 1.7*	−8.8 ± 2.0*	−8.6 (-8.4, −10.0)*	< 0.05

-Note: Imaging parameters of the overall population at baseline and 6-month follow-up as well as in groups. Values are either n (%) or median (IQR). Statistically significant p values are printed in bold. * Represents significant difference between baseline and follow-up data.

Two CMR examinations were used to assess myocardial recovery through changes in WMSI and GLS. From CCS1 to CCS4, WMSI decreased by 0.8%, 6%, 11.3%, and 10.8%, while GLS increased by of 2.2%, 19.7%, 19.3%, and 13.1%, respectively.

3.4. Prognostic value of PTT, PBVI, and GLS

During a median follow-up time of 39 (IQR: 24–60) months, 32 patients (18.1%) suffered from MACE (6 deaths, 12 re-infarctions, and 18 heart failure events). In a multivariable model including PBVI, nPTT, PTT, GLS and WMSI (hazard ratio [HR]: 1.48, 1.45, 1.43, 1.36 and 1.20; P = 0.004, 0.013, 0.015, 0.009 and 0.012) significantly predicted MACE. According to the Kaplan-Meier analysis, patients in the CCS4 stage showed a significantly lower MACE-free survival (P < 0.05) than those with other stages (Table 4 and Fig. 6).

Table 4.

Cox regression.

Variable at baseline	Univariable analysis		Multivariable analysis
	Hazard ratio (95% CI)	p-Value	Hazard ratio (95% CI)	p-Value
LV ejection fraction, %	1.02 (0.87–1.19)	0.027
LV end-diastolic volume, mL	0.97 (0.90–1.05)	0.129
LV end-systolic volume, mL	1.05 (1.00–1.02)	0.039
WMSI	1.21 (1.15–1.37)	0.018	1.20 (1.07–1.35)	0.012
PTT	1.45 (1.08–2.15)	0.008	1.43 (1.08–2.15)	0.015
nPTT	1.48 (1.09–2.22)	0.007	1.45 (1.09–2.20)	0.013
PBVI, mL/m2	1.50 (1.21–1.98)	< 0.001	1.48 (1.18–1.85)	0.004
Infarct necrosis extent, % of LVMM	1.05 (1.01–1.09)	0.009
Microvascular obstruction, % of LVMM	1.15 (1.03–1.27)	0.011
Global longitudinal strain, %	1.36 (0.89–2.06)	0.006	1.36 (0.87–2.06)	0.009

-Note. Cox logistic regression models for prediction of major adverse cardiac events. Statistically significant p values are printed in bold.

-Abbreviations: CI = Confidence interval, LV = Left ventricular, LVMM = Left ventricular myocardial mass, PBVI = Pulmonary blood volume index, PTT = Pulmonary transit time, nPTT = normalized pulmonary transit time, WMSI = wall motion score index.

Fig. 6.

Kaplan–Meier survival curve according to CCS classification. Patients with CCS4 had a significantly higher incidence of the composite outcome of all-cause death, myocardial re-infarction, and re-hospitalization for new congestive heart failure than the other groups. Abbreviations: CCS = Canadian Cardiovascular Society; MACE = major adverse cardiac events.

3.5. Inter- and intra-observer variability

A subset of 40 patients underwent re-analysis to assess intra- and inter-observer feasibility and reproducibility. LVEF, LVEDV, LVESV, WMSI, GLS, PTT, and PBVI showed excellent intra- and inter-observer agreement: 0.86, 0.90, 0.89, 0.82, 0.85, 0.88, and 0.87, respectively.

4. Discussion

PTT-related parameters, WMSI, and myocardial strain demonstrated a good diagnostic value for assessing cardiac involvement in patients with CCS-STEMI. Our primary findings were as follows: 1) Higher PTT and nPTT values were associated with more severe cardiac involvement across CCS stages. 2) PBVI, nPTT, PTT, WMSI, and GLS were independent predictors of mortality, providing additive value across different stages of CCS. 3) Patients in the CCS4 stage showed a significantly lower MACE-free survival.

4.1. Relationship between PTT and cardiac involvement in CCS stage 1–4

PTT and nPTT are rapidly and easily obtainable, robust, non-invasive biomarkers of hemodynamics that can be obtained by routine perfusion CMR [23]. Therefore, we assessed the use of PTT and nPTT as a novel tool to quantitatively evaluate patients at different stages of CCS. PTT is affected by RV systolic function, the pulmonary vasculature, and LV diastolic function. In this study, PTT was specifically defined as the number of cardiac cycles between the peak signals in the RV and LV cavities. This definition not only accounts for the impact of RV function and pulmonary circulation but also incorporates, to some extent, the overall function of the LV. PTT is altered in various diseases, including HF [24], [25], pulmonary hypertension [23], [26], and chronic lung disease [27] and correlates with cardiac function. Of note, in this analysis, we find that the higher the PTT and nPTT, the more significant the cardiac involvement in patients with CCS stage. Pamminger et al. previously investigated PTT in a murine model, employing optoacoustic imaging to measure transit time and CMR imaging to evaluate infarct damage and LVEF [28]. Their study demonstrated that transit time was prolonged in relation to infarct damage in the ischemic mouse model compared to non-infarcted controls, a finding that is consistent with that of baseline PTT prolongation in the present study. However, with time, the infarct damage of CCS1-3 gradually decreased, probably due to the overestimation of the actual infarct damage in the acute-phase CMR examination (myocardial edema included). However, the myocardium involvement of CCS4 was serious, and the myocardial edema did not resolve due to subendocardial hemorrhage, resulting in further prolongation of PTT. Additionally, not all patients with STEMI will complete the stress perfusion scan; the automatic quantification of PTT in rs-PWI may therefore be helpful as a tool to guide further diagnostics. In a two-center study of 985 patients, they deploy a fully automated method of PTT estimation using perfusion mapping with CMR and show that both stress and rest PTT are independently associated with adverse cardiovascular outcomes. In their patient cohort, there is no clear incremental prognostic value of stress PTT, over its evaluation during rest. Consequently, from a technical perspective, pulmonary transit time (PTT or nPTT) can be readily integrated into standard clinical CMR examinations. The time-intensity curve necessary for PTT calculation is readily obtainable through commercially available software, eliminating the need for specialized proprietary programs. Thus, the inclusion of PTT assessment within routine clinical CMR protocols is both practical and feasible.

4.2. CCS stages 1–4: Wavefronts of ischemia and reperfusion injury

The CCS classification of STEMI on which this article was based is the first classification to define MI stage based on underlying histopathology [2]. It is important to recognize that these four stages of tissue injury are not discrete or isolated; rather, they are interdependent and sequential, representing the progressive nature of myocardial damage that transpires during acute ischemia and reperfusion injury resulting from epicardial coronary atherothrombosis. Reperfusion injury initially manifests in the sub-endocardial region and subsequently progresses toward the epicardium, with the severity of tissue damage increasing over time. Within this framework, MVO and hemorrhage may be considered subsequent phases of reperfusion injury. Prompt administration of reperfusion therapy is critical, as it can arrest the injury at an early stage and inhibit advancement to more severe stages of damage. Thus far, only few CMR studies have investigated PTT in patients with CCS-STEMI; as such, its link to clinical outcome is unclear. PTT is known to be affected by preload, blood volumes, global ventricular function, and pulmonary microcirculation; as such, quantifying cardiopulmonary hemodynamics may be useful. The median PTT and nPTT in our study cohort of CCS1 and 2 patients was 8.21 (9.95) and 8.66 (10.58), respectively, which is marginally lower than reported in previous studies (median PTT of 9 s in heart failure patients and mean PTT of 9.6 ± 1.6 s after the Ross procedure) [5], [10]. This is due to different calculation methods. Recently, it has been demonstrated that PTT can be derived automatically from CMR data by deep learning and is, therefore, easy to obtain [15]; we agree this finding. However, at CCS3 and CCS4, the PTT and nPTT increased to 9.85 (12.22) and 10.33 (12.74), respectively. Previous studies have confirmed that the best prognostic differentiation of PTT for adverse outcomes in patients with STEMI is 9.6 s, which is nearly consistent with the CCS3 myocardial injury severity grade obtained in this study, further supporting the cutoff value of 9–10 s to reasonably define the transit time of PTT prolongation [29].

4.3. PTT-related parameters, WMSI, and myocardial strain in CCS-STEMI and their prognostic value

There is significant evidence to suggest that, in addition to assessing static cardiac function indexes, such as LVEF, PTT-related parameters have utility in risk stratification in evaluating the dynamic and temporal dimensions of cardiac function and tissue characteristics. For example, PBVI could be useful for the early detection, grading, and monitoring of hemodynamic congestion in patients with LV dysfunction, particularly during the so-called “vulnerable phase.” Previous studies have evaluated the prognostic value of PBVI in heart failure in patients referred for perfusion CMR or after the Ross procedure [8], [15], [30]. Our study corroborates and extends preliminary findings supporting the feasibility of measuring PBVI using rs-PWI [31], [32], and further shows the ability of PBVI to differentiate between the stages of diastolic dysfunction [33] and predict prognosis [7]. Thus, PBVI in the CCS3 stage remains a watershed in our cohort, while PBVI was an independent predictor of MACE.

In addition to static and perfusion-based parameters, changes in regional wall motion and myocardial deformation over time provide valuable insight into post-infarction myocardial recovery. In our cohort, serial CMR demonstrated that from CCS1 to CCS4, WMSI decreased progressively (0.8%, 6%, 11.3%, and 10.8%, respectively) and GLS improved (2.2%, 19.7%, 19.3%, and 13.1%, respectively), indicating graded functional recovery with advancing CCS stages. These findings are consistent with prior studies showing that myocardial strain parameters significantly improve on follow-up imaging after acute MI and correlate with recovery of LV function, underscoring the dynamic nature of myocardial healing beyond single-timepoint measures such as LVEF alone. For example, in a CMR-feature tracking study of STEMI patients, global strain indices including GLS showed significant improvement at 4-month follow-up and were independently associated with clinical outcomes, supporting their role in tracking functional recovery and risk stratification over time. Morever, regional wall motion scoring myocardial strain in acute anterior and non-anterior wall myocardial infarction [34]. Moreover, regional wall motion scoring has been previously validated as a powerful predictor of post-MI outcomes and reflects the extent of segmental dysfunction not fully captured by global metrics. Together, our longitudinal data on WMSI and GLS reinforce the notion that integrating temporal functional parameters with conventional and PTT-derived indices enhances characterization of myocardial viability and prognostic grating temporal functional parameters with conventional and PTT-derived indices enhances characterization of myocardial viability and prognostic stratification in chronic coronary syndrome.

Myocardial strain imaging techniques developed over the past decades have repeatedly shown that GLS is a motion indicator for the sensitive detection of myocardial injury [35]. In the evaluation of functional myocardial performance parameters and tissue characterization, a fundamental debate persists regarding the optimal method for assessing myocardial viability. Integrating regional functional and morphological CMR parameters offers a more comprehensive characterization of injury associated with chronic coronary syndrome CCS-STEMI. This approach surpasses the informational scope provided by isolated assessments of injury metrics (such as IS, MVO, or myocardial salvage) or functional measures (including LVEF, WMSI or GLS). Although global strain values or IS indirectly reflect the compensatory capacity of the post-infarction myocardium, analyses of functional parameters in remote myocardial regions offer a more precise means of distinguishing patients at an elevated risk from those more likely to recover successfully following the acute event [36], [37]. In our research, PBVI, nPTT, PTT, WMSI and GLS were identified as independent predictors of mortality, providing additive value in patients at different stages of CCS, and patients in the CCS4 stage showed a significantly lower MACE-free survival.

5. Limitations

This study has several limitations. The primary limitation is the relatively small number of patients enrolled, which may reduce the prognostic power of the study and limit the generalizability of the findings. Additionally, this is a single-center, CMR-focused study, without incorporating other clinical variables, which may further limit the applicability of the results in broader, multicenter settings. Some patients were also lost to follow-up during the COVID-19 outbreak. Furthermore, the accuracy of PBVI measurement may be limited by the uncertain relationship between contrast bolus concentration and signal intensity on CMR images used for PTT measurements. The cutoff values for PTT, nPTT, PBVI, and GLS were derived from the CCS-STEMI cohort and may not be generalizable; larger cohorts are required to validate their sensitivity and specificity. Finally, we did not compare PTT and PBVI values with those from invasive hemodynamic measurements or segmental strain analysis.

6. Conclusions

The assessment of PTT-related parameters and myocardial strain using contrast-enhanced CMR is a novel imaging application for quantitatively evaluating cardiopulmonary function and myocardial motion. In our CCS-STEMI cohort, increased PTT was associated with more severe myocardial injury and worse cardiovascular outcome. PBVI, PTT, nPTT, WMSI, and GLS were independent predictors of mortality, providing additional prognostic value across different stages of CCS. Finally, patients in the CCS4 stage showed a significantly lower MACE-free survival.

Funding

This study was supported by National Natural Science Foundation of China (62172288); Xuanwu Hospital of Capital Medical University Clinical research Project (LCYJ202410); Research Topics of Education and Teaching Reform in Capital Medical University (2025JYY133).

CRediT authorship contribution statement

Zheng Sun: Writing – review & editing, Writing – original draft, Funding acquisition, Formal analysis, Data curation, Conceptualization. Zi-Xuan Zhang: Data curation. Ying-Ying Hu: Writing – original draft, Methodology, Formal analysis, Data curation. Yan-Ling Wang: Resources, Investigation. Zhi Liu: Supervision, Resources, Funding acquisition. Shuang Li: Formal analysis. Guang-Yan Huo: Methodology. Nan Chen: Writing – review & editing, Supervision, Software, Investigation, Conceptualization. Jie Lu: Supervision, Resources, Project administration.

Central illustration

CMR PTT and Myocardial Strain in STEMI Prognosis

Declaration of Generative AI and AI-assisted technologies in the writing process

The authors declare that they do not use any AI tools in scientific writing in this paper.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data availability

Data will be made available on request.