Abstract
Background
Acute rejection (AR) remains a major obstacle to success after heart transplantation (HT). Impaired microcirculatory function is linked to a higher risk of AR. The value of cardiac magnetic resonance (CMR) resting first‑pass perfusion imaging for evaluating microcirculatory function to predict subsequent AR after HT remains unclear. We aimed to evaluate the prognostic value of CMR resting first‑pass perfusion parameters measured early after HT for AR.
Methods
This single-center retrospective study analyzed data from 79 HT recipients, including 21 with AR [HT (AR+)] and 58 without AR [HT (AR−)], and 32 healthy volunteers. CMR examination was performed at 3 months after HT to assess myocardial perfusion. Resting first-pass perfusion parameters, including time to peak myocardial signal intensity (TTM), Upslope, and maximum signal intensity (MaxSI) were analyzed. Univariable and multivariable Cox regression analyses were performed to identify predictors of AR within 2 years after HT. Incremental prognostic value was evaluated using C-index, continuous net reclassification index (NRI), and integrated discrimination improvement (IDI). P value <0.05 was considered statistically significant.
Results
During a median follow‑up of 24 months, 21 HT recipients (26.6%) experienced AR. Patients with AR showed significantly longer TTM, lower Upslope and MaxSI at 3 months after HT compared with non‑AR patients and controls (all P<0.001). TTM [adjusted hazard ratio (aHR): 1.07 (1.04, 1.10), P<0.001], Upslope [aHR: 0.78 (0.71, 0.86), P<0.001], and MaxSI [aHR: 0.97 (0.95, 0.98), P<0.001] were independent predictors of AR after adjusting for age and body mass index (BMI). Adding TTM, Upslope, and MaxSI individually to the clinical model (age + BMI) significantly improved discriminant and reclassification ability for risk of AR (C‑index: 0.803, 0.889, 0.869 vs. 0.707; continuous NRI: 0.755, 1.049, 1.025; IDI: 0.139, 0.343, 0.340; all P<0.05), with Upslope demonstrating the highest incremental value.
Conclusions
CMR resting first‑pass perfusion parameters acquired early after HT that reflect microcirculatory dysfunction are independent prognostic markers of AR within 2 years and provide incremental prognostic value over clinical variables to guide personalized post‑transplant management.
Keywords: Heart transplantation (HT), acute rejection (AR), microcirculatory dysfunction, cardiac magnetic resonance (CMR), first-pass perfusion imaging
Highlight box.
Key findings
• Patients who developed acute rejection (AR) within 2 years after heart transplantation (HT) exhibited significantly impaired microcirculatory function as assessed by cardiac magnetic resonance (CMR) resting first-pass perfusion at 3 months post-transplantation. First‑pass perfusion parameters (time to peak myocardial signal intensity, Upslope, maximum signal intensity) are independent predictors of subsequent AR after HT and provide incremental prognostic value over clinical variables.
What is known and what is new?
• Microcirculatory dysfunction is closely related to AR. Invasive coronary physiology (such as index of microcirculatory resistance) can predict AR after HT.
• CMR first-pass perfusion parameters, acquired non-invasively early after HT, independently predict subsequent AR and add prognostic value over clinical variables.
What is the implication, and what should change now?
• CMR resting first-pass perfusion imaging may help identify high-risk patients for post-transplant AR and guide personalized treatment strategies. Future multicenter studies with biopsy-confirmed AR are needed to validate these findings.
Introduction
Heart transplantation (HT) is considered an effective treatment option for patients with end-stage heart failure (1). Although advances in immunosuppression and complication control have markedly enhanced heart transplant survival, acute rejection (AR) remains the primary obstacle to success (2,3). AR occurs in 13–30% of HT recipients within the first year after transplantation (2). Approximately 44% of transplant recipients require treatment for AR within 5 years after HT (4). Furthermore, recipients experiencing acute rejection carry an elevated risk of subsequent cardiac allograft vasculopathy, graft failure, and mortality (5). Therefore, greater emphasis should be placed on the early detection and prediction of AR.
The current gold standard tool for AR is endomyocardial biopsy (EMB) (6). However, EMB has inherent limitations, including its invasive nature, potential complications (such as tricuspid regurgitation, hemorrhage, and cardiac perforation), and interobserver variability in pathological interpretation. Furthermore, pathological diagnoses are mostly made after myocardial cell injury (7,8). Detecting AR before the onset of cardiomyocyte damage is of critical importance. Microvascular dysfunction resulting from immune responses may serve as either a phenotypic manifestation or a precursor of acute rejection. The index of microcirculatory resistance (IMR), obtained through invasive coronary angiography, is a measurable and reliable indicator of microcirculatory dysfunction. It has been shown to enable early prediction of acute rejection occurring within the subsequent 1 to 2 years (9-11). Furthermore, impaired microcirculatory function, as assessed by the microvascular resistance reserve (MRR), has been associated with an increased risk of death or acute cellular rejection (12). Nevertheless, these approaches remain invasive and are not readily applicable in routine clinical practice.
Cardiac magnetic resonance (CMR) first-pass perfusion imaging is a non-invasive technique for assessing myocardial perfusion with high spatial and temporal resolution, and has been shown to reflect myocardial microvascular dysfunction in a variety of disorders (13-16). Nevertheless, it remains unclear whether resting first-pass perfusion early after HT enables prediction of subsequent AR.
Therefore, the aim of this study was to investigate the prognostic value of myocardial microcirculatory dysfunction assessed by resting first‑pass perfusion imaging for AR after HT. We present this article in accordance with the STROBE reporting checklist (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-0215/rc).
Methods
Study design and population
This study was a retrospective analysis based on a prospective cohort. The protocol was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (No. KY2023-1007-01) and was in accordance with the Declaration of Helsinki and its subsequent amendments. The requirement for written informed consent was waived due to the retrospective nature of the study.
A total of 159 patients who underwent HT at the Guangdong Provincial People’s Hospital between February 2020 and January 2023 were enrolled. The exclusion criteria were as follows: (I) age <18 years; (II) death during the perioperative period or before discharge after transplantation; (III) acute rejection occurring prior to CMR examination; (IV) death due to causes other than acute rejection during follow-up; (V) comorbid diabetes mellitus (given the independent effect of diabetic microvascular disease on myocardial perfusion) (17,18); (VI) severe renal insufficiency (estimated glomerular filtration rate (eGFR) <30 mL/min/1.73 m2; (VII) contraindications to CMR examination; (VIII) poor image quality for analysis; (IX) refusal to undergo CMR examination. The patient flow diagram and details of exclusions are shown in Figure 1. In addition, a control group of 32 healthy individuals (20 males, 12 females) was selected from the database of healthy volunteers, and the same CMR examination was performed. All healthy volunteers had no cardiovascular risk factors (such as hypertension, hyperglycemia, and hyperlipidemia) and showed normal findings on clinical symptoms, echocardiography, and electrocardiography.
Figure 1.
Inclusion and exclusion flowchart for this study. AR, acute rejection; CMR, cardiac magnetic resonance; eGFR, estimated glomerular filtration rate; HT, heart transplantation; HT (AR−), heart transplantation recipients without acute rejection; HT (AR+), heart transplantation recipients with acute rejection.
As a result, 79 HT recipients and 32 healthy controls were included in the current analyses. The median interval between HT and acquisition of CMR examinations was 3.1 months (Q1–Q3, 2.8–3.4 months).
Immunosuppressive therapy and surveillance acute rejection
All patients received standard immunoinduction and maintenance suppression therapy. For induction immunosuppressive therapy, basiliximab is routinely used 2 hours before transplantation and 4 days after transplantation. Maintaining immunosuppression is based on calcineurin inhibitors (tacrolimus), antimetabolic drugs (azathioprine or mycophenolate mofetil), and prednisone. Prednisone is gradually reduced in dosage in the first year. According to the clinical status or regimen, some patients use mammalian target rapamycin inhibitors (sirolimus). Carefully monitor the therapeutic level of immunosuppressants and related side effects, and adjust the dosage accordingly. Post-transplant AR surveillance in Guangdong Provincial People’s Hospital (our center) followed a multi-modality strategy, including: (I) monthly routine monitoring of ultrasound electrocardiogram, electrocardiogram, and laboratory tests (NT-proBNP, hsTNT, blood routine, liver and kidney function, etc.) was conducted after the operation; (II) CMR examinations were performed at 3, 6, 12 and 24 months after the operation to monitor for the occurrence of AR; (III) selective EMB based on clinical indication when AR was highly suspected, taking into account patient consent and clinical status. In our center, EMB was performed for patients with clinical suspicion of AR (manifested as abnormal clinical symptoms, laboratory biomarkers, electrocardiographic findings, or imaging results) rather than routine surveillance, mainly due to: (I) it is an invasive procedure associated with potential procedural complications; (II) interobserver variability; (III) substantial cumulative costs from repeated testing. Nevertheless, due to its invasive nature, the majority of patients still refuse to undergo EMB when AR is clinically suspected. In addition, several patients presented absolute contraindications such as coagulation disorders and unfavorable cardiac anatomical structures, which further limited EMB performance.
According to the International Society for Heart and Lung Transplantation (ISHLT) consensus (8,19), the pathological diagnosis of AR was defined as acute cellular rejection (ACR; grade ≥ 2R) and/or antibodymediated rejection (AMR). For patients who did not undergo EMB, we adopted rigorous clinical diagnostic criteria based on the ISHLT consensus guidelines and previously published studies (20-24): (I) during follow-up, new‑onset heart failure signs and symptoms; (II) symptoms resolved after adjustment of the antirejection regimen; (III) one major criterion or two minor criteria were met among assessments including cardiac structure and function, electrocardiographic abnormalities, and clinical symptoms/signs (Table S1 provides the detailed diagnostic criteria). Fulfillment of any one of the above criteria I–III establishes the diagnosis of clinical AR.
Patient follow-up and outcome
The primary endpoint was AR confirmed by pathological diagnosis on EMB or clinically diagnosed within 2 years after HT. Patients were followed up from the time of HT until the occurrence of the endpoint event or the cutoff date of January 2025, with median follow-up of 24 months [interquartile range (IQR), 20.5–24 months].
CMR scanning protocol
CMR images were obtained in all the participants on a 3.0 T scanner (Ingenia, Philips Healthcare, The Netherlands), which had 32 coil elements. CMR images included: cine images, resting first-pass myocardial perfusion imaging, and late gadolinium enhancement (LGE) imaging. Detailed imaging protocols and sequence parameters are provided in Appendix 1.
CMR image analysis
All CMR imaging analyses were performed using CVI42 software (Circle Cardiovascular Imaging, Calgary, Alberta, Canada) by two experienced radiologists blinded to all clinical data. The endocardial and epicardial contours of each left ventricular layer were semiautomatically delineated on short‑axis images at end‑diastole and end‑systole, with the papillary muscles and regulatory zones excluded from the analysis, followed by manual correction when necessary. Conventional cardiac structural and functional parameters were obtained using workstation analysis. Left ventricular (LV) CMR functional parameters included left ventricular end-diastolic volume index (LV EDVi), left ventricular end-systolic volume index (LV ESVi), stroke volume index (SVi), cardiac index (CI), LV mass index at end-diastole (LV MI) and left ventricular ejection fraction (LV EF). Right ventricular (RV) CMR functional parameters included right ventricular end-diastolic volume index (RV EDVi), right ventricular end-systolic volume index (RV ESVi), and right ventricular ejection fraction (RV EF).
For the evaluation of myocardial microcirculatory perfusion, endocardial and epicardial contours at basal, midventricular, and apical short-axis slices were manually delineated on resting first-pass perfusion images, with a region of interest (ROI) placed in the LV blood pool. The time-signal intensity curve was subsequently generated, and semi-quantitative LV perfusion parameters were automatically computed for all 16 myocardial segments. Global myocardial perfusion indices were derived by averaging segmental values across the 16 myocardial regions per subject. Resting first-pass perfusion parameters included time to maximum signal intensity (TTM), maximum signal intensity (MaxSI), and Upslope, which were obtained from the myocardial time-signal intensity (Figure 2). All parameters reflected myocardial perfusion, which was indirectly associated with microvascular function (25).
Figure 2.
CMR resting first-pass perfusion image analysis. (A) The CMR first-pass perfusion image of the midventricular left ventricular short-axis slice and (B) the outline of the myocardium and blood pool in image analysis; (C) the time-signal curve derived from postprocessing software: the orange curve is the time signal curve of the blood pool, and the other color curves are those of the myocardium. The first-pass perfusion parameters including TTM (a), Upslope (b), and MaxSI (c). CMR, cardiac magnetic resonance; MaxSI, maximum signal intensity; TTM, time to peak myocardial signal intensity.
LGE analysis included both qualitative and quantitative assessments. For qualitative analysis, LGE was considered present when it was visualized on both short‑axis and orthogonal long‑axis views of the left ventricle using PSIR sequences. For quantitative analysis, the PSIR images were first imported into the LGE analysis module of the CVI42 post‑processing software. Endocardial and epicardial contours were delineated on short‑axis slices of the left ventricle. The full width at half maximum (FWHM) method was adopted for quantitative LGE analysis. The automatically identified LGE regions were inspected slice by slice to ensure accuracy. LGE extent was defined as the ratio of LGE mass to left ventricular mass.
Statistical analysis
Normality of continuous variables was assessed with the Shapiro–Wilk test. Continuous variables were reported as mean ± standard deviation (SD) for normal distributions or as median (IQR) for non‑normal distributions. Depending on the normality of the data, continuous variables were compared using the independent sample t-test or Mann-Whitney U test for two‑group comparisons, and one‑way (ANOVA) or Kruskal-Wallis test for multi‑group comparisons. Categorical variables were summarized as frequencies (percentages) and compared using the χ2 test or Fisher’s exact test, as appropriate.
Univariable and multivariable Cox regression analyses were used to identify predictors of acute rejection in patients after HT. Variables with a P value <0.05 in the univariable analysis were entered into the multivariable Cox regression model. To avoid overfitting in the multivariable analysis, model development followed the 5-10‑events‑pervariable rule (5–10 times the number of events) (26). Accordingly, multiple separate Cox proportional hazards models were constructed to determine independent predictors and prediction models for AR. Specifically, each CMR resting first‑pass perfusion parameter (TTM, Upslope, and MaxSI) was added individually to a clinical model [age + body mass index (BMI)] to build several predictive models. The discriminative and reclassification abilities of the different models were evaluated using the C‑index, continuous net reclassification improvement (NRI), and relative integrated discrimination improvement (IDI). Time-dependent receiver operating characteristic (t-ROC) curves were used to determine the optimal cutoff values of perfusion parameters for risk stratification at clinically relevant time points (24 months post-transplant, based on the median follow-up duration of 24 months in this cohort). Patients were stratified according to the cutoff values derived from the ROC analysis, and Kaplan-Meier survival analysis was performed. Survival curves were compared using the log‑rank test.
Inter- and intra-observer reliability were assessed by two blinded observers who evaluated CMR first‑pass perfusion images from 20 random patients after HT. Reliability was quantified using the intraclass correlation coefficient (ICC) with corresponding 95% confidence interval (CI).
A two‑sided P value <0.05 was considered statistically significant. All statistical analyses were performed using SPSS version 26.0 (IBM Corporation, Armonk, NY, USA) and R software version 4.5.0 (R Foundation for Statistical Computing). Statistical graphs were generated using GraphPad Prism version 10.1.0 (GraphPad Software, San Diego, CA, USA).
Results
Population characteristics
The baseline clinical characteristics of the study population are presented in Table 1. A total of 79 HT recipients and 32 healthy controls were finally enrolled in this study. The mean age of the HT recipients was 42.3±11.7 years, and 73.4% (58/79) were male. Dilated cardiomyopathy was the most common indication for transplantation, accounting for 63.3% (50/79). Regarding donor characteristics, the mean donor age was 33.5±12.7 years, 79.7% (63/79) of donors were male, and 70.9% (56/79) had donor-recipient sex match. All HT recipients received triple immunosuppressive therapy consisting of tacrolimus, prednisolone, and mycophenolate mofetil, and 20.3% (16/79) of recipients were additionally treated with sirolimus.
Table 1. Baseline clinical characteristics.
| Variables | HT (n=79) | HT (AR−) (n=58) | HT (AR+) (n=21) | Control (n=32) | P value |
|---|---|---|---|---|---|
| Age (years) | 42.3±11.7 | 44.2±11.7† | 36.9±13.5‡ | 37.2±14.5 | 0.03* |
| Male | 58 (73.4) | 43 (74.1) | 15 (71.4) | 20 (62.5) | 0.81 |
| BMI (kg/m2) | 20.8±3.5 | 20.3±3.6† | 22.2±2.9‡ | 23.1±2.5 | 0.02* |
| Heart rate (bpm) | 88.1±11.8 | 88.0±12.1† | 88.2±11.0† | 73.1±11.6 | 0.94 |
| Hypertension | 11 (13.9) | 7 (12.1) | 4 (19.0) | – | 0.47 |
| Hyperlipidemia | 7 (8.9) | 5 (8.6) | 2 (9.5) | – | >0.99 |
| Indication for transplant | |||||
| Dilated cardiomyopathy | 50 (63.3) | 38 (65.5) | 12 (57.1) | – | 0.50 |
| Ischemic heart disease | 13 (16.5) | 7 (12.1) | 6 (28.6) | – | 0.10 |
| Others | 16 (20.3) | 13 (22.4) | 3 (14.3) | – | 0.54 |
| Immunosuppression | |||||
| Tacrolimus | 79 (100.0) | 21 (100.0) | 21 (100.0) | – | – |
| Prednisolone | 79 (100.0) | 58 (100.0) | 21 (100.0) | – | – |
| Mycophenolate mofetil | 79 (100.0) | 58 (100.0) | 21 (100.0) | – | – |
| Sirolimus | 16 (20.3) | 10 (17.2) | 6 (28.6) | 0.34 | |
| Donor characteristics | |||||
| Donor age (years) | 33.5±12.7 | 30.9±13.8 | 36.4±11.3 | – | 0.21 |
| Donor sex, male | 63 (79.7) | 47 (81.0) | 16 (76.2) | – | 0.75 |
| Donor/recipient sex match | 56 (70.9) | 42 (72.4) | 14 (66.7) | – | 0.62 |
| Time of ischemia donor heart (min) | 198.5 [187.2, 203.0] | 198.9 [185.3, 204.1] | 200.4 [195.1, 202.5] | – | 0.55 |
| Laboratory data | |||||
| NT-proBNP (pg/mL) | 387.0 [258.9, 755.6] | 357.6 [255.8, 684.9] | 665.2 [301.6, 1,195.0] | – | 0.09 |
| hsTnT (pg/mL) | 22 [15.1, 44.3] | 20.9 [14.1, 39.5] | 25.0 [16.2, 63.9] | – | 0.29 |
| Creatinine (μmol/L) | 91.0 [75.4, 113.8] | 89.1 [72.2, 112.9] | 93.0 [77.0, 114.0] | – | 0.52 |
| eGFR (mL/min/1.73 m2) | 67.7±25.5 | 69.5±26.4 | 63.0±22.6 | 0.32 |
Values are presented as mean ± standard deviation, number (%), or median [Q1, Q3]. *, statistical significance. †, HT (AR+)/HT (AR−) vs. controls (P<0.05); ‡, HT (AR+) vs. HT(AR−) (P<0.05); P values represent comparisons between HT (AR+) and HT (AR−). AR, acute rejection; BMI, body mass index; eGFR, estimated glomerular filtration rate; HT, heart transplantation; HT (AR−), heart transplantation recipients without acute rejection; HT (AR+), heart transplantation recipients with acute rejection; hsTNT, high-sensitivity troponin T; NT-proBNP, N-terminal pro–B-type natriuretic peptide.
During a median follow-up of 24 months (IQR, 20.5–24.0 months) after transplantation, 21 patients experienced AR with a cumulative incidence of 26.6% (8 cases were confirmed by EMB, and 13 were clinically diagnosed). All AR patients were successfully treated. Within 2 years after primary transplantation, no patient required retransplantation.
Among the HT recipients with AR [HT (AR+)], HT recipients without AR [HT (AR−)], and control groups, the average age was significantly higher in the HT (AR−) group than in the other two groups (all P<0.05). BMI was significantly lower in the HT (AR−) group than in the other two groups (all P<0.05). Heart rate was significantly higher in both the HT (AR+) and HT (AR−) groups than in the control group (all P<0.001).
Comparison of CMR-derived parameters among the three groups
As demonstrated in Table 2, for conventional CMR parameters, LV EDVi, LV ESVi, LV SVi, RV EDVi, and RV ESVi were significantly lower in both the HT (AR−) and HT (AR+) groups than in the control group (all P<0.05), and LV MI was significantly higher than in the control group. There were no significant differences in these parameters between the HT (AR−) and HT (AR+) groups (all P>0.05). The HT (AR+) group had a higher prevalence of LGE than the HT (AR−) group, although the difference was not statistically significant (28.6% vs. 19.0%, P=0.37). For first-pass perfusion parameters (Figure 3), TTM was significantly higher [33.34 (28.02, 49.50) vs. 26.00 (23.27, 28.76), 24.97 (23.34, 26.75), all P<0.001], whereas Upslope [24.11 (20.29, 27.68) vs. 32.90 (28.94, 37.89), 33.60 (29.44, 38.82), all P<0.001] and MaxSI (265.17±42.11 vs. 331.74±45.21, 344.31±44.22, all P<0.001) were significantly lower in the HT (AR+) group than in the HT (AR−) and control groups. There were no significant differences in TTM, Upslope, or MaxSI between the HT (AR−) and control groups. The representative CMR first-pass perfusion images in a control, an HT patient without AR, and an HT patient with AR are illustrated in Figure 4.
Table 2. Comparison of CMR findings among healthy controls and HT recipients with/without AR.
| Variables | HT (n=79) | HT (AR−) (n=58) | HT (AR+) (n=21) | Control (n=32) | P value |
|---|---|---|---|---|---|
| Conventional CMR parameters | |||||
| LV EDVi (mL/m2) | 62.3±11.4 | 62.9±11.5† | 60.7±11.2† | 75.5±13.0 | 0.47 |
| LV ESVi (mL/m2) | 24.1 [21.0, 28.7] | 24.2 [21.0, 29.0]† | 23.6 [18.5, 28.8]† | 26.4 [24.9, 34.7] | 0.42 |
| LV EF (%) | 61.7±5.6 | 61.4±4.8 | 60.5±7.4 | 62.2±3.9 | 0.40 |
| LV SVi (ml/m2) | 37.0±7.3 | 37.1±6.6† | 36.7±9.0† | 47.2±8.3 | 0.86 |
| LV CI (L/min/m2) | 3.2±0.7 | 3.3±0.6 | 3.2±0.9 | 3.5±0.9 | 0.64 |
| LV MI (g/m2) | 50.7±8.7 | 49.7±8.2† | 53.3±9.2† | 43.8±6.6 | 0.09 |
| RV EDVi (mL/m2) | 63.1±11.8 | 63.5±12.1† | 62.1±11.2† | 79.0±15.6 | 0.66 |
| RV ESVi (mL/m2) | 28.9±7.6 | 28.8±7.7† | 29.3±7.4† | 35.9±8.1 | 0.80 |
| RV EF (%) | 54.8±6.5 | 55.4±6.7 | 53.3±6.0 | 54.6±4.9 | 0.20 |
| LGE | 17 (21.5) | 11 (19.0) | 6 (28.6) | 0 | 0.37 |
| LGE extent (%) | 0 [0, 1.3] | 0 [0, 0.5] | 0 [0, 1.7] | 0 | 0.54 |
| First-pass perfusion parameters | |||||
| TTM (s) | 27.00 [24.05, 29.97] | 26.00 [23.27, 28.76] | 33.34 [28.02, 49.50]†‡ | 24.97 [23.34, 26.75] | <0.001* |
| Upslope | 30.25 [25.72, 35.14] | 32.90 [28.94, 37.89] | 24.11 [20.29, 27.78]†‡ | 33.60 [29.44, 38.82] | <0.001* |
| MaxSI | 314.04±53.14 | 331.74±45.21 | 265.17±42.11†‡ | 344.31±44.22 | <0.001* |
Values are presented as mean ± standard deviation, number (%), or median [Q1, Q3]. *, statistical significance. †, HT (AR+)/HT (AR−) vs. controls (P<0.05); ‡, HT (AR+) vs. HT(AR−) (P<0.05); P values represent comparisons between HT (AR+) and HT (AR−). AR, acute rejection; CMR, cardiac magnetic resonance; HT, heart transplantation; HT (AR−), heart transplantation recipients without acute rejection; HT (AR+), heart transplantation recipients with acute rejection; LGE, late gadolinium enhancement; LV CI, left ventricular cardiac index; LV EDVi, left ventricular end-diastolic volume index; LV EF, left ventricular ejection fraction; LV ESVi, left ventricular end-systolic volume index; LV MI, left ventricular myocardial mass index; LV SVi, left ventricular stroke volume index; MaxSI, maximum signal intensity; RV EDVi, right ventricular end-diastolic volume index; RV EF, right ventricular ejection fraction; RV ESVi, right ventricular end-systolic volume index; TTM, time to peak myocardial signal intensity.
Figure 3.
Comparative analysis of perfusion parameters between HT (AR−), HT (AR+), and control groups. (A) TTM. (B) Upslope. (C) MaxSI. ns, P>0.05. HT (AR−), heart transplantation recipients without acute rejection; HT (AR+), heart transplantation recipients with acute rejection; MaxSI, maximum signal intensity; TTM, time to peak myocardial signal intensity.
Figure 4.
Representative first-pass perfusion images (first column), time-signal intensity curves (second column) from the mid-left ventricular slice. HT (AR−), heart transplantation recipients without acute rejection; HT (AR+), heart transplantation recipients with acute rejection; MaxSI, maximum signal intensity; TTM, time to peak myocardial signal intensity.
Analysis of risk factors for AR after HT
Univariate predictors of AR after HT are presented in Table S2. In univariate Cox regression analysis, clinical variables (age and BMI) and first-pass perfusion parameters (TTM, Upslope, MaxSI) were associated with AR. In the multivariable Cox regression analysis, to avoid model overfitting, each first‑pass perfusion parameter was added separately to the clinical variables (age + BMI) to construct multiple prediction models (Table 3). The multivariable Cox regression results showed that after adjusting for age and BMI, TTM [adjusted hazard ratio (aHR): 1.07 (1.04, 1.10), P<0.001], Upslope [aHR: 0.78 (0.71, 0.86), P<0.001], and MaxSI [aHR: 0.97 (0.95, 0.98), P<0.001] were independent predictors of post-transplant AR.
Table 3. Univariate and multivariable Cox regression analyses for predicting AR after HT.
| Variables | Univariate Cox | Model 1 (age + BMI) | Model 2 (age + BMI + TTM) | Model 3 (age + BMI + Upslope) | Model 4 (age + BMI + MaxSI) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| HR (95% CI) | P value | aHR (95% CI) | P value | aHR (95% CI) | P value | aHR (95% CI) | P value | aHR (95% CI) | P value | |||||
| Age | 0.96 (0.93, 0.99) | 0.03* | 0.96 (0.93, 0.99) | 0.02* | 0.97 (0.94, 1.01) | 0.17 | 0.97 (0.93, 1.00) | 0.08 | 0.97 (0.93, 1.00) | 0.07 | ||||
| BMI | 1.14 (1.01, 1.29) | 0.03* | 1.15 (1.02, 1.29) | 0.02* | 1.10 (0.97, 1.25) | 0.13 | 1.17 (1.02, 1.36) | 0.03* | 1.19 (1.04, 1.37) | 0.01* | ||||
| TTM | 1.08 (1.05, 1.11) | <0.001* | 1.07 (1.04, 1.10) | <0.001* | ||||||||||
| Upslope | 0.79 (0.72, 0.86) | <0.001* | 0.78 (0.71, 0.86) | <0.001* | ||||||||||
| MaxSI | 0.97 (0.96, 0.98) | <0.001* | 0.97 (0.95, 0.98) | <0.001* | ||||||||||
*, statistical significance. aHR, adjusted hazard ratio; AR, acute rejection; BMI, body mass index; CI, confidence interval; HT, heart transplantation; MaxSI, maximum signal intensity; TTM, time to peak myocardial signal intensity.
Prognostic value of first-pass perfusion parameters
T-ROC curve analysis was performed to assess the prognostic value of TTM, Upslope, and MaxSI for AR at the 24-month landmark time point (corresponding to the median follow-up duration of this cohort), and used to determine the optimal cutoff value for risk stratification at clinically meaningful landmark time points. (Figure S1). The optimal cutoff values were determined using the Youden index (Table 4), and Kaplan‑Meier survival curves were constructed to evaluate the risk stratification efficacy of these parameters for AR (Figure 5). The results demonstrated that the area under the curve (AUC) of TTM, Upslope, and MaxSI for predicting AR were 0.796 (95% CI: 0.670, 0.923), 0.884 (95% CI: 0.811, 0.958), and 0.829 (95% CI: 0.725, 0.933), respectively. Although the AUC of Upslope was the highest among the three parameters, pairwise DeLong comparisons showed no statistically significant differences between these AUC (Upslope vs. TTM, Upslope vs. MaxSI, all pairwise P>0.05). The optimal cutoffs for TTM, Upslope, and MaxSI were 29.40 s, 28.56, and 310.61, with corresponding sensitivities of 71.4%, 85.7%, and 90.5%, specificities of 82.8%, 77.6%, and 69.0%, PPV of 60.0%, 58.1%, and 51.4%, NPV of 88.9%, 93.7%, and 95.3%, respectively. Survival analysis showed that patients with TTM ≥29.40 s, Upslope ≤28.56, and MaxSI ≤310.61 had a significantly higher incidence of AR within 2 years after HT (P<0.001).
Table 4. Optimal cutoff values of first-pass perfusion parameters for predicting of AR in 79 HT recipients.
| Parameters | AUC (95% CI) | Cutoff value | Sensitivity (%) | Specificity (%) | PPV (%) | NPV (%) | P value |
|---|---|---|---|---|---|---|---|
| TTM (s) | 0.796 (0.670, 0.923) | 29.40 | 71.4 | 82.8 | 60.0 | 88.9 | <0.001* |
| Upslope | 0.884 (0.811, 0.958) | 28.56 | 85.7 | 77.6 | 58.1 | 93.7 | <0.001* |
| MaxSI | 0.829 (0.725, 0.933) | 310.61 | 90.5 | 69.0 | 51.4 | 95.3 | <0.001* |
*, statistical significance. AR, acute rejection; AUC, area under the curve; CI, confidence interval; HT, heart transplantation; MaxSI, maximum signal intensity; NPV, negative predictive value; PPV, positive predictive value; TTM, time to peak myocardial signal intensity.
Figure 5.
Kaplan‑Meier survival curves for AR stratified by optimal cutoffs of first-pass perfusion parameters. (A) TTM. (B) Upslope. (C) MaxSI. The cumulative incidence of AR was significantly higher in patients with TTM ≥29.40 s, Upslope ≤28.56, and MaxSI ≤310.61 (all log‑rank P<0.001). Numbers at risk are shown below each curve. AR, acute rejection; CI, confidence interval; HR, hazard ratio; MaxSI, maximum signal intensity; TTM, time to maximum signal intensity.
Incremental prognostic value of first-pass perfusion parameters for risk of AR
To evaluate the clinical usefulness of first-pass perfusion parameters in the prediction of AR, discriminant and reclassification abilities of prediction models were compared (Table 5). Relative to model 1 (clinical variables only), adding TTM, Upslope, and MaxSI individually significantly increased discriminant and reclassification ability for the prediction of AR (model 1 vs. model 2: C index 0.707 vs. 0.803, NRI 0.755, IDI 0.139, P=0.004, 0.002, 0.024; model 1 vs. model 3: C index 0.707 vs. 0.889, NRI 1.049, IDI 0.343, all P<0.001; model 1 vs. model 4: C index 0.707 vs. 0.869, NRI 1.025, IDI 0.340, P=0.005, <0.001, <0.001). Compared with model 2 (Model 1 + TTM), adding Upslope and MaxSI separately to the clinical variables yielded superior predictive performance (model 2 vs. model 3: C index 0.803 vs. 0.889, NRI 0.989, IDI 0.201, P=0.049, <0.001, <0.001; model 2 vs. model 4: NRI 0.962, IDI 0.171, P<0.001, =0.02). Model 4 (Model 1 + MaxSI) had inferior reclassification ability compared with Model 3 (Model 1 + Upslope), but the difference was not statistically significant (NRI −0.498, P=0.10).
Table 5. Incremental prognostic value of CMR first-pass perfusion parameters for AR after HT.
| Model | C-index | Continuous NRI | IDI | |||||
|---|---|---|---|---|---|---|---|---|
| Median (95% CI) | P value | Median (95% CI) | P value | Median (95% CI) | P value | |||
| Model 1 as a reference | ||||||||
| Model 1: clinical variables (age + BMI) | 0.707 (0.596, 0.817) | Reference | Reference | Reference | Reference | Reference | ||
| Model 2 (Model 1 + TTM) | 0.803 (0.707, 0.898) | 0.004* | 0.755 (0.285, 1.226) | 0.002* | 0.139 (0.018, 0.260) | 0.02 | ||
| Model 3 (Model 1 + Upslope) | 0.889 (0.826, 0.942) | <0.001* | 1.049 (0.613, 1.485) | <0.001* | 0.343 (0.209, 0.471) | <0.001* | ||
| Model 4 (Model 1 + MaxSI) | 0.869 (0.804, 0.933) | 0.005* | 1.025 (0.548, 1.370) | <0.001* | 0.340 (0.183, 0.503) | <0.001* | ||
| Model 2 as a reference | ||||||||
| Model 2 (Model 1 + TTM) | 0.803 (0.707, 0.898) | Reference | Reference | Reference | Reference | Reference | ||
| Model 3 (Model 1 + Upslope) | 0.889 (0.826, 0.942) | 0.049 | 0.989 (0.541, 1.436) | <0.001* | 0.201 (0.109, 0.294) | <0.001* | ||
| Model 4 (Model 1 + MaxSI) | 0.869 (0.804, 0.933) | 0.18 | 0.962 (0.466, 1.458) | <0.001* | 0.171 (0.024, 0.318) | 0.02 | ||
| Model 3 as a reference | ||||||||
| Model 3 (Model 1 + Upslope) | 0.889 (0.826, 0.942) | Reference | Reference | Reference | Reference | Reference | ||
| Model 4 (Model 1 + MaxSI) | 0.869 (0.804, 0.933) | 0.34 | −0.498 (−1.081, −0.086) | 0.10 | 0.002 (−0.089, 0.094) | 0.96 | ||
*, statistical significance. The time point for continuous NRI and IDI was set at 24 months after heart transplantation. AR, acute rejection; BMI, body mass index; CI, confidence interval; CMR, cardiac magnetic resonance; HT, heart transplantation; MaxSI, maximum signal intensity; NRI, net reclassification index; IDI, integrated discrimination improvement; TTM, time to peak myocardial signal intensity.
Inter- and intraobserver consistency tests
The inter- and intraobserver agreement were examined for first-pass myocardial perfusion parameters (as depicted in Table S3). The intra-observer ICCs for TTM, Upslope, and MaxSI were 0.895, 0.945, and 0.957, respectively. The inter-observer ICCs for the same parameters were 0.881, 0.930, and 0.963, respectively, indicating excellent reproducibility.
Discussion
This study investigated the prognostic association between myocardial microvascular dysfunction assessed by CMR resting first-pass perfusion parameters in the early period after HT with the risk of AR. First, patients with AR showed significant prolonged TTM and decreased Upslope/MaxSI compared with patients without AR and the control group at 3 months after HT, reflecting impaired microvascular function. Second, TTM, Upslope, and MaxSI measured early after HT are independent predictors of subsequent AR within 2 years. Third, adding perfusion parameters to the prediction model with clinical variables significantly increased discriminant and reclassification ability for the risk of AR, among which Upslope achieved the optimal predictive performance.
Although immunemediated rejection is the key initiating process, multiple studies have established associations among acute rejection, myocyte damage, microscopic architectural distortion, microvasculopathy, and CAV (5,27-29). Recent research has highlighted the critical role of microcirculatory dysfunction in the pathophysiology of AR. Microcirculatory dysfunction not only participates in core pathological processes such as endothelial injury and inflammatory cell infiltration, but may also serve as a phenotypic manifestation or a preceding factor of AR, and could potentially act as an early diagnostic biomarker (10,30-33). Previous studies have evaluated the potential association between microcirculatory dysfunction and adverse outcomes (including AR) after HT (30,34). Hiemann et al. (35) demonstrated that microvasculopathy detected on biopsy samples independently predicted mortality even in the absence of epicardial CAV, and that histologic evidence of stenotic microvasculopathy was found in 43% of endomyocardial biopsy specimens, indicating that not all pathologically proven microvasculopathy leads to clinically significant graft failure or clinical deterioration. Therefore, assessing microcirculatory dysfunction in HT recipients may provide additional information for better patient risk stratification and subsequent modification of treatment strategies, thereby potentially improving patient outcomes.
Several previous studies have confirmed that the index of IMR and MRR derived from invasive coronary angiography enable early prediction of acute rejection (9-12), but their invasive nature limits routine clinical application. Most studies using noninvasive imaging in heart transplant recipients have focused on stress CMR perfusion or the detection of CAV. Erbel et al. (36). reported that myocardial perfusion reserve index (MPRI) derived from CMR stress perfusion imaging enables early detection of transplant microvascular disease before the onset of CAV. CMR resting first pass perfusion imaging has been validated for assessing myocardial microcirculatory dysfunction in various conditions, including diabetes mellitus, Takotsubo syndrome, chemotherapy induced myocardial injury, and exercise induced myocardial injury in athletes (13-16). However, its application after HT has been less explored.
In this study, resting first pass perfusion imaging was used to non-invasively assess myocardial microcirculatory status. We demonstrated that patients who subsequently developed AR already exhibited significantly prolonged TTM and reduced Upslope and MaxSI at three months after HT, indicating early microcirculatory injury characterized by delayed contrast filling, decreased perfusion rate, and reduced perfusion volume in the transplanted heart. TTM, Upslope, and MaxSI were independently associated with AR, which is consistent with previous invasive studies. Conventional cardiac functional parameters (such as LVEF and RVEF) showed no significant differences between the AR and the non-AR group, further emphasizing that perfusion abnormalities reflect subclinical microcirculatory dysfunction rather than overt systolic failure. LGE as a standalone marker has limited diagnostic utility for acute rejection, with studies reporting specificity as low as 31–36% (37) and a prevalence of non-vascular LGE up to 49–71% in OHT recipients (38). Furthermore, LGE more commonly appears in the chronic phase—LGE-positive patients had longer time since transplantation (8.3 vs. 5.7 years, P=0.041) with no significant difference in rejection history (39). These findings suggest LGE predominantly reflects chronic myocardial fibrosis, while resting first-pass perfusion parameters reflect functional microvascular status that may become abnormal early in immune-mediated injury. Consistent with this, LGE did not differ between AR+ and AR− groups in our cohort (28.6% vs. 19.0%, P=0.37), whereas perfusion parameters provided independent prognostic value, highlighting the unique advantage of functional perfusion imaging over structural LGE for early AR prediction.
Therefore, first-pass perfusion imaging may detect early immune injury before irreversible myocardial damage occurs, providing a window of opportunity for timely intervention. We performed time-dependent ROC analysis to determine optimal cutoff values, which facilitate translation of continuous variables into actionable risk stratification tools in clinical practice. These cutoffs are intended to inform prognosis and risk stratification, not to serve as a diagnostic test for AR. Notably, the optimal cutoff values we identified (TTM ≥29.40 s, Upslope ≤28.56, MaxSI ≤310.61) showed high negative predictive values (88.9%, 93.7%, and 95.3%), suggesting that normal resting perfusion parameters may identify a subgroup of patients at relatively lower risk for future acute rejection. This observation, if validated in larger prospective cohorts, could potentially inform more personalized surveillance strategies. However, the role of this approach in ruling out AR and reducing the need for invasive biopsies requires further investigation in specifically designed prospective trials with systematic biopsy confirmation. Moreover, first‑pass perfusion imaging depends on multiple factors such as scanner settings, contrast injection protocol, heart rate, and post‑processing methods, which imposes strict restrictions on the applicability of this cutoff value.
Various risk factors for AR have been proposed, including age, sex, ethnicity, circulating anti-human leukocyte antigen antibodies, induction therapy, human leukocyte antigen mismatch, and genetic polymorphisms (40-42). Younger patients exhibit more active immune responses and carry a relatively higher risk of rejection. In contrast, patients with obesity often present with chronic inflammation, insulin resistance, and abnormal adipokines, which can exacerbate endothelial injury and immune-inflammatory responses, thereby increasing the risk of AR (43). In the present study, younger age and higher BMI were associated with an increased risk of acute rejection, consistent with previous findings. Furthermore, after adjusting for age and BMI, TTM, Upslope, and MaxSI remained independent predictors of AR. Incremental value analysis further demonstrated that adding any perfusion parameter to the clinical model (age + BMI) significantly improved the C-index, continuous NRI, and IDI, with Upslope showing the most prominent incremental performance. Upslope represents the maximum rate of increase in signal intensity during myocardial first-pass contrast enhancement, directly reflecting the amount of contrast agent entering the myocardium per unit time; thus, it may serve as the most sensitive early marker of microcirculatory function. In summary, resting first-pass perfusion parameters provide key complementary information for clinical risk assessment and may help optimize individualized immunosuppressive regimens and follow-up monitoring strategies.
Limitations
First, this was a single-center study with a relatively small sample size and limited number of positive events (n=21), which may carry a risk of overfitting in predictive modeling. We mitigated the risk of model overfitting by including only 2–3 variables in each predictive model and adhering to the 5–10 events per variable principle. Second, EMB was not performed in all patients; the majority of AR cases (13 out of 21, 61.9%) were diagnosed clinically, while only 8 cases were biopsy-confirmed. Although our diagnostic criteria were consistent with ISHLT standards and previous studies, this reliance on clinical diagnosis may introduce misclassification bias. Our findings need to be verified in a larger cohort with a higher proportion of biopsy-proven AR. Third, we did not perform invasive coronary microvascular assessments such as IMR for direct comparison with CMR perfusion parameters. Fourth, only resting perfusion was evaluated; stress perfusion or MPRI was not available and may provide additional pathophysiological information. Fifth, patients who experienced AR before the 3-month CMR examination and those who died perioperatively were excluded. Thus, the study population consisted only of recipients who survived long enough to undergo CMR and remained event-free until that time, potentially introducing survivor bias. Consequently, we could not evaluate the relationship between ultra-early AR and CMR perfusion parameters. This selection bias may have led to an underestimation of the association between microvascular dysfunction and rejection risk. Sixth, we excluded patients with diabetes and severe renal dysfunction to reduce confounding; thus, our findings may not be directly generalizable to these high-risk populations. Seventh, data available regarding other important risk factors for acute rejection, such as panel reactive antibody (PRA) levels, such as PRA, an important immunological risk factor for AR, were not available for analysis in this study due to an ongoing independent research project at the cardiovascular surgery department of our hospital involving it. The absence of PRA may introduce residual confounding, as it is possible that the observed CMR perfusion abnormalities partly reflect pre-existing immunologic sensitization rather than purely microvascular dysfunction. Future prospective studies with systematic PRA assessment are needed to validate the independent prognostic value of CMR perfusion parameters beyond humoral immune risk. Finally, different imaging devices may use different sequences, acquisition parameters, and reconstruction algorithms. Similarly, different post‑processing software may also employ different algorithms. These factors may lead to discrepancies in the resulting values. Therefore, the findings of this study are only applicable to the specific imaging device and post-processing software used in our research.
Conclusions
In conclusion, coronary microcirculatory dysfunction assessed by CMR first-pass perfusion imaging in the early period after HT is significantly associated with the risk of AR. TTM, Upslope, and MaxSI are independent predictors of subsequent AR and provide significant incremental prognostic value over clinical variables (age + BMI). Among these, Upslope demonstrates the highest predictive performance. This non‑invasive approach may help identify high-risk patients and guide personalized immunosuppression and surveillance after HT.
Supplementary
The article’s supplementary files as
Acknowledgments
None.
Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Guangdong Provincial People’s Hospital (No. KY2023-1007-01). Informed consent was not required because this is a retrospective study.
Footnotes
Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-0215/rc
Funding: This work was supported by the National Natural Science Foundation of China (grant Nos. 82502289 and 82371903), Guangzhou Clinical High-tech, Major, and Distinctive Technology Projects (grant No. 2023P-TS43), Basic and Applied Basic Research Foundation of Guangdong Province (grant No. 2024A1515012087), Science and Technology Projects in Guangzhou (grant No. 2025A04J4778), Medical Scientific Research Foundation of Guangdong Province (grant No. A2025146), Guangdong Provincial Key Laboratory of Artificial Intelligence in Medical Image Analysis Application (grant No. 2022B1212010011), and National Key Research and Development Program of China (grant No. 2023YFC2414206).
Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-0215/coif). The authors have no conflicts of interest to declare.
Data Sharing Statement
Available at https://cdt.amegroups.com/article/view/10.21037/cdt-2026-0215/dss
References
- 1.Chambers DC, Perch M, Zuckermann A, et al. The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: Thirty-eighth adult lung transplantation report - 2021; Focus on recipient characteristics. J Heart Lung Transplant 2021;40:1060-72. 10.1016/j.healun.2021.07.021 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Khush KK, Cherikh WS, Chambers DC, et al. The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: Thirty-sixth adult heart transplantation report - 2019; focus theme: Donor and recipient size match. J Heart Lung Transplant 2019;38:1056-66. 10.1016/j.healun.2019.08.004 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Anthony C, Imran M, Pouliopoulos J, et al. Cardiovascular Magnetic Resonance for Rejection Surveillance After Cardiac Transplantation. Circulation 2022;145:1811-24. 10.1161/CIRCULATIONAHA.121.057006 [DOI] [PubMed] [Google Scholar]
- 4.Christie JD, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: Twenty-eighth Adult Lung and Heart-Lung Transplant Report--2011. J Heart Lung Transplant 2011;30:1104-22. 10.1016/j.healun.2011.08.004 [DOI] [PubMed] [Google Scholar]
- 5.Raichlin E, Edwards BS, Kremers WK, et al. Acute cellular rejection and the subsequent development of allograft vasculopathy after cardiac transplantation. J Heart Lung Transplant 2009;28:320-7. 10.1016/j.healun.2009.01.006 [DOI] [PubMed] [Google Scholar]
- 6.Velleca A, Shullo MA, Dhital K, et al. The International Society for Heart and Lung Transplantation (ISHLT) guidelines for the care of heart transplant recipients. J Heart Lung Transplant 2023;42:e1-e141. 10.1016/j.healun.2022.10.015 [DOI] [PubMed] [Google Scholar]
- 7.Porcari A, Baggio C, Fabris E, et al. Endomyocardial biopsy in the clinical context: current indications and challenging scenarios. Heart Fail Rev 2023;28:123-35. 10.1007/s10741-022-10247-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Stewart S, Winters GL, Fishbein MC, et al. Revision of the 1990 working formulation for the standardization of nomenclature in the diagnosis of heart rejection. J Heart Lung Transplant 2005;24:1710-20. 10.1016/j.healun.2005.03.019 [DOI] [PubMed] [Google Scholar]
- 9.Lee JM, Choi KH, Choi JO, et al. Coronary Microcirculatory Dysfunction and Acute Cellular Rejection After Heart Transplantation. Circulation 2021;144:1459-72. 10.1161/CIRCULATIONAHA.121.056158 [DOI] [PubMed] [Google Scholar]
- 10.Ahn JM, Zimmermann FM, Gullestad L, et al. Microcirculatory Resistance Predicts Allograft Rejection and Cardiac Events After Heart Transplantation. J Am Coll Cardiol 2021;78:2425-35. 10.1016/j.jacc.2021.10.009 [DOI] [PubMed] [Google Scholar]
- 11.Okada K, Honda Y, Luikart H, et al. Early invasive assessment of the coronary microcirculation predicts subsequent acute rejection after heart transplantation. Int J Cardiol 2019;290:27-32. 10.1016/j.ijcard.2019.04.018 [DOI] [PubMed] [Google Scholar]
- 12.Ha J, Lee SH, Choi KH, et al. Microvascular Resistance Reserve and Prognosis After Heart Transplantation. JACC Cardiovasc Interv 2025;18:439-52. 10.1016/j.jcin.2024.11.011 [DOI] [PubMed] [Google Scholar]
- 13.Yang MX, Li QL, Wang DQ, et al. Myocardial microvascular function assessed by CMR first-pass perfusion in patients treated with chemotherapy for gynecologic malignancies. Eur Radiol 2022;32:6850-8. 10.1007/s00330-022-08823-2 [DOI] [PubMed] [Google Scholar]
- 14.Cau R, Pitzalis C, Pisu F, et al. Myocardial microvascular function assessed by cardiovascular magnetic resonance first-pass perfusion in patients with Takotsubo syndrome. Eur Radiol 2025;35:3917-26. 10.1007/s00330-024-11340-z [DOI] [PubMed] [Google Scholar]
- 15.Jiang L, Shi K, Guo YK, et al. The additive effects of obesity on myocardial microcirculation in diabetic individuals: a cardiac magnetic resonance first-pass perfusion study. Cardiovasc Diabetol 2020;19:52. 10.1186/s12933-020-01028-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Liu Q, Li C, Qi WY, et al. Assessment of Myocardial Microvascular Function in Athletes Using Resting Cardiac Magnetic Resonance First-Pass Perfusion. J Magn Reson Imaging 2026;63:1348-62. 10.1002/jmri.70235 [DOI] [PubMed] [Google Scholar]
- 17.Gondi KT, Hammer Y, Yosef M, et al. Longitudinal Change and Predictors of Myocardial Flow Reserve by Positron Emission Tomography for the Evaluation of Cardiac Allograft Vasculopathy Following Heart Transplantation. J Card Fail 2024;30:915-25. 10.1016/j.cardfail.2023.09.013 [DOI] [PubMed] [Google Scholar]
- 18.Liu X, Yang ZG, Gao Y, et al. Left ventricular subclinical myocardial dysfunction in uncomplicated type 2 diabetes mellitus is associated with impaired myocardial perfusion: a contrast-enhanced cardiovascular magnetic resonance study. Cardiovasc Diabetol 2018;17:139. 10.1186/s12933-018-0782-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Berry GJ, Burke MM, Andersen C, et al. The 2013 International Society for Heart and Lung Transplantation Working Formulation for the standardization of nomenclature in the pathologic diagnosis of antibody-mediated rejection in heart transplantation. J Heart Lung Transplant 2013;32:1147-62. 10.1016/j.healun.2013.08.011 [DOI] [PubMed] [Google Scholar]
- 20.Mehta A, DeFilippis EM, Stehlik J, et al. Contemporary Antibody-Mediated Rejection in Heart Transplantation: JACC: Heart Failure Position Statement. JACC Heart Fail 2025;13:102614. 10.1016/j.jchf.2025.102614 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Butler CR, Savu A, Bakal JA, et al. Correlation of cardiovascular magnetic resonance imaging findings and endomyocardial biopsy results in patients undergoing screening for heart transplant rejection. J Heart Lung Transplant 2015;34:643-50. 10.1016/j.healun.2014.12.020 [DOI] [PubMed] [Google Scholar]
- 22.Asimacopoulos EP, Garbern JC, Gauvreau K, et al. Association of Clinical Rejection Versus Rejection on Protocol Biopsy With Cardiac Allograft Vasculopathy in Pediatric Heart Transplant Recipients. Transplantation 2020;104:e31-7. 10.1097/TP.0000000000002983 [DOI] [PubMed] [Google Scholar]
- 23.Harrington JK, Richmond ME, Woldu KL, et al. Serial Changes in Right Ventricular Systolic Function Among Rejection-Free Children and Young Adults After Heart Transplantation. J Am Soc Echocardiogr 2019;32:1027-1035.e2. 10.1016/j.echo.2019.04.413 [DOI] [PubMed] [Google Scholar]
- 24.Peyster EG, Wang C, Ishola F, et al. In Situ Immune Profiling of Heart Transplant Biopsies Improves Diagnostic Accuracy and Rejection Risk Stratification. JACC Basic Transl Sci 2020;5:328-40. 10.1016/j.jacbts.2020.01.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Thomson LE, Wei J, Agarwal M, et al. Cardiac magnetic resonance myocardial perfusion reserve index is reduced in women with coronary microvascular dysfunction. A National Heart, Lung, and Blood Institute-sponsored study from the Women’s Ischemia Syndrome Evaluation. Circ Cardiovasc Imaging 2015;8:e002481. 10.1161/CIRCIMAGING.114.002481 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Riley RD, Ensor J, Snell KIE, et al. Calculating the sample size required for developing a clinical prediction model. BMJ 2020;368:m441. 10.1136/bmj.m441 [DOI] [PubMed] [Google Scholar]
- 27.Tan CD, Baldwin WM, 3rd, Rodriguez ER. Update on cardiac transplantation pathology. Arch Pathol Lab Med 2007;131:1169-91. 10.5858/2007-131-1169-UOCTP [DOI] [PubMed] [Google Scholar]
- 28.Vecchiati A, Tellatin S, Angelini A, et al. Coronary microvasculopathy in heart transplantation: Consequences and therapeutic implications. World J Transplant 2014;4:93-101. 10.5500/wjt.v4.i2.93 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Pighi M, Gratta A, Marin F, et al. "Cardiac allograft vasculopathy: Pathogenesis, diagnosis and therapy". Transplant Rev (Orlando) 2020;34:100569. 10.1016/j.trre.2020.100569 [DOI] [PubMed] [Google Scholar]
- 30.Haddad F, Khazanie P, Deuse T, et al. Clinical and functional correlates of early microvascular dysfunction after heart transplantation. Circ Heart Fail 2012;5:759-68. 10.1161/CIRCHEARTFAILURE.111.962787 [DOI] [PubMed] [Google Scholar]
- 31.Iwańczyk S, Woźniak P, Smukowska-Gorynia A, et al. Microcirculatory Disease in Patients after Heart Transplantation. J Clin Med 2023;12:3838. 10.3390/jcm12113838 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Shahandeh N, Song J, Saito K, et al. Invasive Coronary Physiology in Heart Transplant Recipients: State-of-the-Art Review. J Soc Cardiovasc Angiogr Interv 2023;2:100627. 10.1016/j.jscai.2023.100627 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Chih S, Chong AY, Džavík V, et al. Fibrotic Plaque and Microvascular Dysfunction Predict Early Cardiac Allograft Vasculopathy Progression After Heart Transplantation: The Early Post Transplant Cardiac Allograft Vasculopathy Study. Circ Heart Fail 2023;16:e010173. 10.1161/CIRCHEARTFAILURE.122.010173 [DOI] [PubMed] [Google Scholar]
- 34.Yang HM, Khush K, Luikart H, et al. Invasive Assessment of Coronary Physiology Predicts Late Mortality After Heart Transplantation. Circulation 2016;133:1945-50. 10.1161/CIRCULATIONAHA.115.018741 [DOI] [PubMed] [Google Scholar]
- 35.Hiemann NE, Wellnhofer E, Knosalla C, et al. Prognostic impact of microvasculopathy on survival after heart transplantation: evidence from 9713 endomyocardial biopsies. Circulation 2007;116:1274-82. 10.1161/CIRCULATIONAHA.106.647149 [DOI] [PubMed] [Google Scholar]
- 36.Erbel C, Mukhammadaminova N, Gleissner CA, et al. Myocardial Perfusion Reserve and Strain-Encoded CMR for Evaluation of Cardiac Allograft Microvasculopathy. JACC Cardiovasc Imaging 2016;9:255-66. 10.1016/j.jcmg.2015.10.012 [DOI] [PubMed] [Google Scholar]
- 37.Krieghoff C, Barten MJ, Hildebrand L, et al. Assessment of sub-clinical acute cellular rejection after heart transplantation: comparison of cardiac magnetic resonance imaging and endomyocardial biopsy. Eur Radiol 2014;24:2360-71. 10.1007/s00330-014-3246-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Quinn S, Catania R, Appadurai V, et al. Cardiac MRI in Heart Transplantation: Approaches and Clinical Insights. Radiographics 2025;45:e240142. 10.1148/rg.240142 [DOI] [PubMed] [Google Scholar]
- 39.Lawson AA, Watanabe K, Griffin L, et al. Late-gadolinium enhancement is common in older pediatric heart transplant recipients and is associated with lower ejection fraction. J Cardiovasc Magn Reson 2023;25:61. 10.1186/s12968-023-00971-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Kobashigawa JA, Kirklin JK, Naftel DC, et al. Pretransplantation risk factors for acute rejection after heart transplantation: a multiinstitutional study. The Transplant Cardiologists Research Database Group. J Heart Lung Transplant 1993;12:355-66. [PubMed] [Google Scholar]
- 41.Kilic A, Weiss ES, Allen JG, et al. Simple score to assess the risk of rejection after orthotopic heart transplantation. Circulation 2012;125:3013-21. 10.1161/CIRCULATIONAHA.111.066431 [DOI] [PubMed] [Google Scholar]
- 42.Jarcho J, Naftel DC, Shroyer TW, et al. Influence of HLA mismatch on rejection after heart transplantation: a multiinstitutional study. The Cardiac Transplant Research Database Group. J Heart Lung Transplant 1994;13:583-95; discussion 595-6. [PubMed] [Google Scholar]
- 43.Kubo SH, Naftel DC, Mills RM, Jr, et al. Risk factors for late recurrent rejection after heart transplantation: a multiinstitutional, multivariable analysis. Cardiac Transplant Research Database Group. J Heart Lung Transplant 1995;14:409-18. [PubMed] [Google Scholar]