Abstract
Distinct histopathologic changes occur in acute cellular rejection (ACR), antibody mediated rejection (AMR), and biopsy negative rejection (BNR). Cardiovascular magnetic resonance (CMR)-based myocardial tissue characterization can be used to quantify these changes. We assessed T1, T2 and extracellular volume fraction (ECV) by CMR in patients with sub-types of rejection. T1, T2 and ECV were quantified at the mid-ventricular level and compared between patients with and without rejection. The association between quantitative tissue characteristics and the combined outcome of death, re-transplantation, heart failure hospitalization, or myocardial infarction was evaluated with a Cox-proportional hazards model. In 46 patients, mean age 53.3 ± 13.7 years, 71.7% male, at a median of 7.4 years from transplant, average myocardial T1 was increased in BNR compared to no rejection (1057 vs 1012 msec, p=0.006). Average myocardial T2 was elevated in all types of rejection, p<0.05. In a cox-proportional hazards model, higher T2 values were associated with an increase in the combined clinical outcome (adjusted HR 1.21, 95% CI 1.06 – 1.37, p=0.004) after adjusting for left ventricular mass index. Myocardial tissue characteristics are abnormal in all subtypes of rejection and abnormal T2 quantified by CMR provides additional prognostic value.
Keywords: cardiac transplant, transplant rejection, cardiac MRI, non-invasive imaging
INTRODUCTION
Heart transplantation is the most effective therapy for end-stage heart failure, with more than 5000 transplants performed annually 1. The incidence of graft rejection within the first year following transplant is approximately 10%, decreasing with increasing time post-transplant 2. Additionally, rejection is a leading cause of death within the first year of cardiac transplant 3. Diagnosis of rejection has traditionally relied on endomyocardial biopsy (EMB); however tissue characterization with cardiovascular magnetic resonance (CMR) is increasingly being used to complement histopathologic diagnosis 4.
CMR offers a unique advantage over other noninvasive imaging modalities by allowing quantitative tissue characterization with T1 and T2 relaxation times. Graft rejection is characterized by myocardial infiltration of inflammatory cells which results in edema, expansion of the extracellular space and myocyte death 5. These histopathologic changes are associated with alteration in myocardial T1 and T2 which can be quantified with multiparametric CMR mapping. Extracellular volume (ECV) expansion can be quantified by calculating ECV from pre and post-contrast T1 measurements 6. Previous studies have shown that myocardial T2 is elevated in patients with graft rejection when combining all rejection types- acute cellular rejection (ACR), antibody mediated rejection (AMR) and biopsy-negative rejection (BNR) 7,8. However, the number of patients with rejection included in these studies has been small such that differences between subtypes of rejection could not be analyzed 7,8. There is less robust data evaluating the association between myocardial T1, extracellular volume (ECV) or late gadolinium enhancement (LGE)-a marker of cell death and scarring- and graft rejection. Furthermore, the association between quantitative CMR tissue characterization and clinical outcomes has not been characterized in post-transplant patients evaluated for graft rejection. We sought to evaluate differences in T1, T2, ECV, and LGE between patients with ACR, AMR, BNR and no rejection, and to determine if multiparametric tissue characterization could be used for prognostication.
MATERIALS AND METHODS
Study Population:
Cardiac transplant patients who underwent CMR between January 1, 2007 and June 30, 2018 at Cedars-Sinai Medical Center were included in the study. Scans were included regardless of the time from heart transplantation. Patients in whom myocardial tissue characterization imaging (either T1 or T2 mapping) was not performed were excluded (n=27). Studies were performed on a clinical basis to evaluate graft dysfunction (n=73) or to evaluate suspected cardiac mass (n=6). Patients with T1 imaging performed with the modified look-locker inversion recovery (MOLLI) sequence (n=6) were excluded.
Data Considerations
Patient information including, age, sex, medical history (hypertension, diabetes, dyslipidemia, smoking status, peripheral vascular disease), and medications (aspirin, statin, angiotensin converting enzyme inhibitors, angiotensin receptor blockers, calcium channel blockers, and beta-blockers) were obtained from electronic medical records. Chronic kidney disease (CKD) was defined as estimated glomerular filtration rate <30 ml/kg/min 9. Transplant variables including, transplant date, cardiomyopathy etiology, donor age, donor gender, cytomegalovirus (CMV) status, sensitization status, rejection history, biopsy results, history of CMV infection, history of cardiac allograft vasculopathy, and follow-up for death, re-transplantation, and major adverse cardiovascular events (MACE) including heart failure admission, and myocardial infarction or revascularization were recorded. CMV mismatch was defined as CMV donor positive, recipient negative. The biopsy closest to CMR study date was used to classify patients as ACR (defined as grade 2 or 3 cellular rejection, or treated grade 1 cellular rejection), AMR (defined as biopsy with AMR grade 2 or AMR grade 1 with clinical impression of AMR) and BNR (defined as allograft dysfunction in the presence of a normal biopsy, after exclusion of other causes such as cardiac allograft vasculopathy) or no rejection 10,11. All patients with BNR had invasive coronary angiography performed (mean 61 days prior to CMR), with CAV grade 1 identified in one patient and grade 2 identified in three patients. In all patients the degree of LV dysfunction was out of proportion to the degree of CAV. The indication for CMR in patients without rejection included cardiac or pericardial mass (n=6), unexplained dyspnea (n=11), pre-operative assessment (n=3), new-onset atrial fibrillation (n=2), abnormal echocardiogram (n=2) or syncope (n=2). Patients without biopsy were categorized according to clinical interpretation at the time of CMR (n=9, all without rejection). Patients without biopsies were referred for CMR for cardiac or pericardial mass assessment (n=6) or pre-operative assessment (n=3). The remaining seventeen patients without rejection underwent clinically indicated biopsy, demonstrating 0R or 1R which was not treated.
Image Acquisition and Analysis
CMR was performed on a 1.5-Tesla whole-body scanner (MAGNETOM Avanto®, Siemens Healthcare, Erlangen, Germany) with the patient in a supine position using an 8-element phased-array radiofrequency coil with breath-holding and cardiac gating. Cine images of the LV in short and long axes were acquired using a steady-state free precession sequence (SSFP, TR 2.4–3.9, TE 0.9–2.0, slice thickness 8 mm). T1 mapping was performed using the shortened MOLLI (shMOLLI) sequence (n=44) both pre and post contrast administration12. T2 mapping was performed as previously described13. T1 mapping was integrated into clinical workflow prior to adoption of T2 mapping for routine clinical use; as a result, in 8 patients T2 mapping data was unavailable. Additionally, in one patient only T2 mapping was performed during modified scan protocols with limited acquisitions. LGE images (segmented k-space inversion recovery sequence, TR 3.4–5.0, TE 1.1–1.5, TI 200–300, slice thickness 8 mm) were acquired throughout the entire LV starting 10 min, after administration of 0.1–0.2 mmol/kg of gadolinium diethylenetriaminepentaacetic acid (Gd-DTPA, Magnevist®, Schering AG, Germany). The inversion time was set to null the signal of normal myocardium after Gd-DTPA and was adjusted during the scan as necessary. ECV was calculated as previously described14.
Cine images were analyzed using Medis Suite (Vitrea Workstation, Vital Images, Minnetonka, USA). Semi-automated and manual tracing of endocardial and epicardial borders from short-axis images was performed to calculate left ventricular (LV) and right ventricular (RV) end-diastolic volumes (LVEDV and RVEDV), LV and RV end-systolic volumes (LVESV and RVESV), LV mass, and LV and RV ejection fractions (LVEF and RVEF). Pre-contrast T1 and T2 as well as post-contrast T1 maps were generated and analyzed using Siemens workstation, with regions of interest drawn in 6-segments on a short axis slice at the mid-ventricular level excluding regions with LGE 8,15. Average global values, as well as peak segment values of T1 and T2 were considered in analyses. A case example of a patients with ACR is shown in Figure 1. LGE sequences were analyzed semi-quantitatively as previously described.16 Analysis of LGE was performed visually by defining the areas of hyperenhancement in all myocardial segments as seen on either long or short-axis slices and quantified using a full width, half maximum method.17
Figure 1:
Case of a patient with ACR. The patient had elevated T1 (Panel A; average 1105 msec, peak 1193 msec) and T2 (Panel B; average 52.5 msec, peak 56.5 msec) values. Short-axis stead-state free precession image shown in Panel C. The patient had transmural inferolateral late-gadolinium enhancement and patchy epicardial enhancement, shown in panel D.
Outcomes:
The primary outcome was a difference in T1, T2, calculated ECV and LGE extent in patients with ACR, AMR, or BNR compared to patients without rejection. We assessed the ability of CMR tissue characteristics to differentiate patients with and without rejection. The secondary outcome was the association between quantitative tissue characteristics and a combined clinical outcome including death, re-transplantation or MACE which was defined as heart failure hospitalization, non-fatal myocardial infarction or revascularization.
Analysis
Continuous variables are summarized as mean (standard deviation) if normally distributed and compared using a Student’s t-test. Continuous variables which are not normally distributed are summarized as median (interquartile range [IQR]) and compared using a Mann-Whitney U test. Normality was assessed using the Shapiro-Wilk test. Categorical variables are summarized as number (proportion) and compared using a Fisher’s Exact test or chi-square test as appropriate.
The primary outcome, difference in CMR tissue characteristics, was assessed using the Wilcoxon rank-sum test using patients without rejection as the reference category. The ability to differentiate patients with rejection was assessed using receiver-operating characteristic (ROC) curves. Area under the curve (AUC) was compared using the method suggested by Delong et al. The secondary clinical outcomes of association between CMR tissue characteristics and the combined outcome of death, re-transplantation, or MACE was assessed using Cox-proportional Hazards. The proportional hazards assumption was assessed with Schoenfeld residuals and was found to be valid in all analyses. Lastly, we assessed for associations between myocardial tissue characteristics and reduced LVEF.
All statistical tests were two-sided, with a p-value < 0.05 considered significant. All analyses were performed using Stata version 13 (StataCorp, College Station, Texas). The study was approved by the investigational review board at Cedars-Sinai Medical Center (CR00013886).
RESULTS
Patient Population
Baseline population characteristics of the 46 patients, mean age 53.3 ± 13.7 years and 71.7% male, included in the analysis are outlined in Table 1. Five patients were classified as having ACR, 3 with AMR, 12 with BNR, and 26 as having no evidence of rejection. Biopsies occurred a median of 2 days before CMR (IQR −1 to 10). The majority of patients with rejection were started on therapy prior to CMR (n=16, 80.0%). There were no significant differences in patient or donor age, BMI, cardiovascular disease risk factors chronic kidney disease or history of immunosuppressive treatment between patients with ACR, AMR, or BNR compared to patients without rejection. Interval between transplantation and CMR studies was shorter in patients with ACR compared to patients without rejection (median 1.3 vs. 10.0 years, p=0.016).
Table 1:
Baseline Population characteristics
| No Rejection n=26 |
ACR n=5 |
AMR n=3 |
BNR n=12 |
|
|---|---|---|---|---|
| Age (years), mean ± SD | 54.2 ± 14.1 | 56.0 ± 13.3 | 33.7 ± 14.2 * | 55.2 ± 10.1 |
| Male, n(%) | 18 (69.2) | 4 (80.0) | 2 (66.7) | 9 (75.0) |
| Years post-transplant, median (IQR) | 10.0 (5.3 – 13.5) | 1.3 (0.3 – 1.6) * | 1.6 (0.9 – 9.3) | 5.3 (2.5 – 8.7) |
| Donor age (years), mean | 31.7 ± 10.9 | 32.6 ± 7.4 | 31.5 ± 4.9 | 38.4 ± 12.1 |
| Male donor, n(%) | 20 (76.9) | 3 (60.0) | 3 (100.0) | 8 (66.7) |
| Body mass index (kg/m2), mean ± SD | 26.1 ±5.3 | 25.1 ± 4.4 | 31.2 ± 4.4 | 6.1 ± 6.6 |
| CMV mismatch, n(%) | 4 (15.4) | 1 (20.0) | 1 (33.3) | 2 (16.7) |
| Sensitized pre-transplant, n (%) | 8 (30.8) | 1 (20.0) | 1 (33.3) | 5 (41.7) |
| Hypertension, n (%) | 12 )46.2) | 2 (40.0) | 1 (33.3) | 7 (58.3) |
| Diabetes, n (%) | 8 (30.8) | 1 (20.0) | 0 (0.0) | 7 (58.3) |
| Dyslipidemia, n (%) | 9 (34.6) | 0 (0.0) | 0 (0.0) | 6 (50.0) |
| Chronic Kidney Disease, n (%) | 2 (7.7) | 1 (20.0) | 0 (0.0) | 3 (25.0) |
| Tacrolimus, n (%) | 21 (80.8) | 5 (100.0) | 3 (100.0) | 11 (91.7) |
| mTOR inhibitor, n(%) | 14 (53.9) | 2(40.0) | 2 (66.7) | 9 (75.0) |
| History of CAV, n(%) | 4 (15.4) | 0 (0.0) | 1 (33.3) | 3 (25.0) |
| History of treated rejection, n(%) | 5 (19.2) | 0 (0.0) | 2 (66.7) | 1 (8.3) |
Summary of baseline population characteristics in patients with and without rejection.
- value significantly different (p<0.05) compared to patients without a history of rejection. ACR – acute cellular rejection, AMR – antibody mediated rejection, BNR – biopsy-negative rejection, IQR – interquartile range.
CMR characteristics of patients included in this study are shown in Table 2. The overall patient cohort had mildly reduced LVEF (mean 41.0%) and RVEF (mean 41.3%). Average pre-contrast T1 was higher in patients with BNR compared to patients without rejection (1057 vs 1012, p=0.006). Maximal T1 values were significantly higher in all patients with rejection compared to patients without rejection. ECV was higher in patients with BNR compared to patients without rejection (37.1% vs 30.7%, p=0.043). Average T2 relaxation was increased in all three subtypes of rejection (ACR: 53.4 ± 3.1 msec, p=0.001; AMR 55.2 ± 2.8 msec, p=0.006; BNR 51.8 ± 2.4 msec, p<0.001), compared to patients without rejection (47.0 ± 1.7msec), with similar findings for peak T2 values. There was a trend towards higher prevalence of LGE in patients with AMR (100.0% vs. 34.6%, p=0.060)
Table 2:
CMR Characteristics:
| No Rejection n=26 |
ACR n=5 |
AMR n=3 |
BNR n=12 |
|
|---|---|---|---|---|
| LVEDVI, median (IQR) | 75.7 (60.4 – 96.2) | 71.1 (64.8 – 75.9) | 73.1 (70.9 – 94.1) | 68.6 (55.2 – 75.0) |
| LVEF (%), median (IQR) | 47.4 (29.9 – 53.0) | 43 (32.8 – 51.5) | 45.0 (28.6 – 49.0) | 33.4 (22.1 – 38.0) |
| RVEDVI, median (IQR) | 72.1 (64.9 – 88.7) | 78.8 (69.6 – 91.1) | 73.6 (59.3 – 96.4) | 59.7 (49.0 – 67.5)* |
| RVEF (%), median (IQR) | 38.2 (32.5 – 54.1) | 44.0 (36.3 – 61.0) | 35.1 (31.0 – 61.4) | 41.5 (30.5 – 47.5) |
| LVMI (g/m2), mean ± SD | 59.7 (50.9 – 66.7) | 58.8 (53.7 – 73.8) | 69.7 (67.5 – 71.3) | 59.7 (54.0 – 69.5) |
| T1 mapping performed (44/46) | 26 (100.0) | 5 (100.0) | 3 (100.0) | 11 (91.7) |
| Average Pre-contrast T1 (msec), mean ± SD | 1012 ± 40 | 1040 ± 32 | 1044 ± 18 | 1057 ± 47 * |
| Peak Pre-contrast T1 (msec), mean ± SD | 1052 ± 46 | 1108 ± 21.6 | 1122 ± 21 | 1108 ± 75 |
| T2 mapping performed | 20 (76.9) | 5 (100.0) | 3 (100.0) | 10 (83.3) |
| Average T2 (msec), mean ± SD | 47.0 ± 1.7 | 53.4 ± 3.1* | 55.2 ± 2.8* | 51.8 ± 2.4* |
| Peak T2 (msec), mean ± SD | 49.8 ± 2.5 | 58.2 ± 6.3* | 58.9 ± 3.3* | 55.8 ± 3.6* |
| Post-contrast T1 mapping performed | 15 (57.7) | 4 (80.0) | 1 (33.3) | 7 (58.3) |
| ECV estimate | 30.7 (26.5 – 34.8) | 30.7 (28.7 – 32.4) | 32.0 | 37.1 (31.2 – 41.4)* |
| LGE present, n(%) | 9 of 26 (34.6) | 1 of 5 (20.0) | 3 of 3 (100.0) | 7 of 12 (58.3) |
| LGE volume (g), median (IQR) in patients with LGE | 11.0 (8.5 – 21.6) | 6.3 | 13.4 (5.7 – 44.4) | 6.9 (4.2 – 18.5) |
Summary of CMR characteristics in patients with and without rejection.
- value significantly different (p<0.05) compared to patients without a history of rejection. ACR – acute cellular rejection, AMR – antibody mediated rejection, BNR – biopsy-negative rejection, IQR – interquartile range.
Differentiating Rejection from absence of rejection
Receiver operating characteristic (ROC) curves for discriminating patients with any type of rejection are shown in Figure 2. Among patients in whom both T1 and T2 relaxation was quantified, the area under the ROC curve (AUC) for average T2 was 0.982 (95% CI 0.952 – 1.000) and was greater than the AUC for average T1 (0.862, 95% CI 0.745 – 0.978, p=0.047). Findings were similar for peak T2 (AUC 0.954, 95% CI 0.897 – 1.000) and peak T1 (AUC 0.821, 95% CI 0.680 – 0.961, p=0.053). Calculated ECV demonstrated moderate discrimination of patients with rejection AUC 0.688 (95% CI 0.466 – 0.908). Presence of late gadolinium enhancement (AUC 0.601, 95% CI 0.456 – 0.748) and RVEDVI had poor discrimination (AUC 0.631, 95% CI 0.463 – 0.798).
Figure 2:
Receiver Operating Characteristic Curves for Identifying Patients with Rejection. AUC for T2 was significantly larger compared to the AUC for T1 (p=0.047) in patients with both measurements.
Clinical Outcomes
During follow-up, 15 patients died or underwent re-transplantation at a median of 0.6 years and 13 patients experienced at least one MACE event at a median of 0.4 years. The first MACE event was: CHF admission in 11 (76.2%), and non-fatal myocardial infarction or revascularization in 2 (19.0%). Table 3 outlines the unadjusted analyses assessing associations with clinical outcomes. BNR was associated with an increase in death or re-transplantation (unadjusted hazard ratio [HR] 3.22, 95% CI 1.11 – 9.37, p=0.032) whereas, increased LV mass index (LVMI) was associated with a reduction in the combined outcome of death, re-transplantation, or MACE (unadjusted HR 0.96, 95% CI 0.92 – 1.00, p=0.029). After adjusting for LVMI, higher average T2 was associated with an increase in death, re-transplantation or MACE (adjusted HR 1.18, 95% CI 1.05 – 1.33, p=0.006). There was a similar association with peak T2 in adjusted analyses (adjusted HR 1.15, 95% CI 1.05 – 1.25, p=0.002). T1, ECV, and LGE were not associated with an increase in clinical outcomes.
Table 3.
UnivariableCox Proportional Hazards Analysis for Clinical Outcomes
| Death or Re-transplant | Death, Re-transplant or MACE | |||
|---|---|---|---|---|
| Hazard Ratio (95% CI) | p-value | Hazard Ratio (95% CI) | p-value | |
| LVEDVI | 0.98 (0.95 – 1.00) | 0.143 | 0.98 (0.96 – 1.01) | 0.196 |
| LVEF (%) | 1.01 (0.97 – 1.04) | 0.737 | 1.00 (0.97 – 1.03) | 0.943 |
| RVEDVI | 1.00 (0.97 – 1.02) | 0.861 | 0.99 (0.97 – 1.01) | 0.435 |
| RVEF (%) | 1.00 (0.97 – 1.04) | 0.835 | 1.01 (0.98 – 1.04) | 0.598 |
| LVMI (g/m2) | 0.98 (0.94 – 1.01) | 0.241 | 0.96 (0.92 – 1.00) | 0.029 |
| Average Pre-contrast T1 (msec) | 1.00 (0.99 – 1.01) | 0.652 | 1.01 (1.00 – 1.02) | 0.063 |
| Peak Pre-contrast T1 (msec) | 1.00 (0.99 – 1.01) | 0.496 | 1.01 (1.00 – 1.01) | 0.165 |
| Average T2 (msec) | 1.06 (0.93 – 1.21) | 0.388 | 1.10 (0.98 – 1.23) | 0.109 |
| Peak T2 (msec) | 1.08 (0.98 – 1.19) | 0.136 | 1.09 (0.99 – 1.19) | 0.078 |
| ECV estimate | 1.07 (0.99 – 1.16) | 0.087 | 1.04 (0.96 – 1.12) | 0.332 |
| LGE present | 0.62 (0.21 – 1.82) | 0.384 | 0.94 (0.38 – 2.36) | 0.901 |
| LGE volume | 0.99 (0.92 – 1.06) | 0.758 | 1.02 (0.97 – 1.06) | 0.462 |
| Diagnosis (Normal reference) | ||||
| Acute cellular | 0.78 (0.10 – 6.33) | 0.815 | 2.31 (0.64 – 8.38) | 0.201 |
| Antibody-mediated | -- | -- | -- | -- |
| Biopsy-negative | 3.22 (1.11 – 9.37) | 0.032 | 1.73 (0.59 – 5.03) | 0.316 |
| History of CAV | 1.41 (0.40 – 5.02) | 0.595 | 1.24 (0.36 – 4.32) | 0.735 |
Univariable analysis of associations with clinical outcomes. ECV – extracellular volume, LGE – late gadolinium enhancement, LVEDVI – left ventricular end-diastolic volume index, LVEF – left ventricular ejection fraction, LVMI – left ventricular mass index, MACE – major adverse cardiovascular event, RVEF – right ventricular ejection fraction, RVEDVI – right ventricular end-diastolic volume index.
Increased average T1 was associated with reduced LVEF (unadjusted odds ratio [OR] 1.02, 95% CI 1.00 – 1.04, p=0.012). However, neither elevated average T2 (unadjusted OR 1.13, p=0.139) nor elevated ECV (unadjusted OR 1.02, p=0.707) were associated with reduced LVEF.
DISCUSSION
CMR with quantitative tissue characterization is increasingly being used for the assessment of patients with suspected graft rejection. In this study we found that average myocardial T2 is increased in all subtypes of rejection. We further report that average T1 is increased in patients with BNR and peak T1 is increased in all forms of rejection compared to patients without rejection. Additionally, we demonstrated that elevated average and peak T2 are associated with an increase in adverse outcomes. To our knowledge, this is the first study to show the association between abnormal T2 and adverse events in heart transplant patients with suspected graft rejection. Our results expand on the existing literature demonstrating the utility of CMR tissue characterization in detecting rejection, by showing that significant differences in T1 and T2 values exist for each sub-type of rejection. However, in this exploratory study the population size was not large enough to describe differences between rejection sub-types.
Previous CMR studies of heart transplant patients have described abnormalities in T1, T2 and calculated ECV in patients with clinical signs and symptoms of rejection 7,8,18–20. However, these studies were limited by the small sample size of the overall patient population and the limited number of patients within each rejection subtype4,7,8,18. T2 relaxation has previously been shown to be elevated in patients with rejection 7. However, only 8 out of 74 patients who were evaluated in that study had evidence of rejection. Therefore, evaluation of differences in T1, T2 and ECV between rejection subtypes was not feasible 7. Butler et al. showed that elevated T2 relaxation time and abnormal right ventricular end-diastolic volume index were independent predictors of abnormal findings in endomyocardial biopsy 18. An increase in T2 relaxation was also reported by Vermes et al., in a group of 7 patients (6 with ACR and 1 with AMR) with rejection compared to patients without rejection. Our results are consistent with these studies and demonstrate that T2 is increased in patients with rejection compared to those without 7,18. However, our results are novel in that we report significant differences in T1, T2 and ECV among patients with BNR, a group where CMR may be particularly useful and which to our knowledge has not been previously described.
An additional unique finding of our work is the association between an increase in T2 values and an increase in the risk of adverse cardiac outcomes. Similar associations have been demonstrated in a small cohort of patients with suspected acute myocarditis.21 It should be noted that some patients in our cohort may have been classified as having BNR instead of no rejection due to abnormal T2 values, since the clinical diagnosis of rejection may have been influenced by CMR findings. However, if this was true one would expect that patients with BNR may have better outcomes, which was the opposite of what was observed in our cohort. Our results are similar to a previous study of patients with unexplained graft failure (BNR) who had lower 12-month survival compared to patients with graft dysfunction related to AMR or ACR 22.
In contrast to prior imaging studies that yielded conflicting results regarding the association between abnormal T1 and graft rejection 8,20,23, we observed significant increases in average myocardial T1 in patients with ACR and BNR compared to patients without rejection. T1 allowed discrimination of patients with rejection, but had lower discrimination compared to T2 values. Additionally, we showed that peak T1 values are significantly higher in all sub-types of rejection. Increased T1 relaxation has been demonstrated in myocarditis which is characterized by inflammation and myocardial edema 24. It is likely that increased T1 in transplant rejection is related to similar histologic changes. Previous studies have also demonstrated an association between ECV and rejection. In a study of 26 patients post-transplant, increase in ECV was found to be present in patients with rejection19. A similar increase in ECV was also reported in a smaller cohort of patients by Vermes et al. 8. Although we observed an increase in ECV in patients with rejection, it was limited to patients with BNR. It is important to note that patients in our study cohort presented with clinical signs of rejection at a much later time than the patients included in previous studies. Additionally, mean ECV in our overall post-transplant patient cohort was lower than patients with rejection studied previously 19. Our study shows that abnormal T1 and T2 may be useful as diagnostic markers of rejection and may be valuable especially in patients with BNR.
Our study has several important limitations. Although our cohort consisted of more patients than previously reported CMR studies of patients with signs and symptoms of graft rejection, this is a single-center retrospective study of a small group of patients and included few patients classified as having AMR. Therefore, our findings may be more applicable to patients with BNR and ACR compared to those with AMR. A small number of patients included in our study were imaged within the first year post-transplant, where CMR may have decreased diagnostic accuracy 25. Furthermore, 40% of patients with AMR, ACR and BNR did not undergo post-contrast T1 measurement and ECV could not be quantified. This limits our ability to draw definitive conclusions about the significance of ECV in this cohort. As mentioned previously, clinicians were not blinded to CMR results which may have influenced clinical decision making regarding the diagnosis of rejection especially among patients with BNR or ACR with biopsies demonstrating mild (1R) cellular rejection. This may affect the association between CMR characteristics and a diagnosis of rejection. Additionally, the majority of patients with rejection were started on therapy prior to CMR, which may have decreased differences between groups. We studied relatively stable patients referred for clinically indicated CMR, which likely led to a higher proportion of patients with BNR. Most patients without rejection were referred for CMR due to clinical concern regarding graft function and may not reflect healthy, asymptomatic post-transplant patients. However, this reflects the real-world application of CMR where it is often used as a discriminating test in complex but stable patients. Lastly, our study was not sufficiently powered to allow differentiation of CMR characteristics between sub-types of rejection and larger studies are required.
CONCLUSIONS
We demonstrated that myocardial tissue characteristics are abnormal in all sub-types of rejection and can be used to distinguish between patients with and without transplant rejection. An increase in T2 relaxation is associated with increased likelihood of adverse cardiovascular outcomes including death and re-transplantation. Further investigation is required to confirm our findings and to better delineate the association between tissue characteristics and response to therapy.
ACKNOWLEDGEMENTS
The work was supported in part by the Dr. Miriam and Sheldon Adelson Medical Research Foundation. Dr. Miller receives funding support from the Arthur J E Child Fellowship grant. Dr. Tamarappoo is supported by R56HL131871 from the NHLBI.
ABBREVIATIONS:
- ACR
acute cellular rejection
- AMR
antibody mediated rejection
- BNR
biopsy negative rejection
- CKD
chronic kidney disease
- CMR
cardiovascular magnetic resonance
- CMV
cytomegalovirus
- ECV
extracellular volume
- EDV
end-diastolic volume
- EF
ejection fraction
- ESV
end-systolic volume
- HR
hazard ratio
- ISHLT
international society of heart and lung transplantation
- LGE
late gadolinium enhancement
- LV
left ventricle
- MOLLI
modified look-locker inversion
- RV
right ventricle
Footnotes
DISCLOSURES
The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.
DATA AVAILABILITY
The data that support the findings of this study are available from the corresponding author upon reasonable request.
REFERENCES
- 1.Khush KK, Cherikh WS, Chambers DC, et al. The International Thoracic Organ Transplant Registry of the International Society for Heart and Lung Transplantation: Thirty-fifth Adult Heart Transplantation Report; Focus Theme: Multiorgan Transplantation. J Heart Lung Transplant 2018;37(10):1155–1168. [DOI] [PubMed] [Google Scholar]
- 2.Christie JD, Edwards LB, Kucheryavaya AY, et al. The Registry of the International Society for Heart and Lung Transplantation: twenty-seventh official adult lung and heart-lung transplant report--2010. J Heart Lung Transplant 2010;29(10):1104–1118. [DOI] [PubMed] [Google Scholar]
- 3.Gradek WQ, D’Amico C, Smith AL, Vega D, Book WM. Routine surveillance endomyocardial biopsy continues to detect significant rejection late after heart transplantation. J Heart Lung Transplant 2001;20(5):497–502. [DOI] [PubMed] [Google Scholar]
- 4.Crespo-Leiro MG, Barge-Caballero G, Couto-Mallon D. Noninvasive monitoring of acute and chronic rejection in heart transplantation. Curr Opin Cardiol 2017;32(3):308–315. [DOI] [PubMed] [Google Scholar]
- 5.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(11):1710–1720. [DOI] [PubMed] [Google Scholar]
- 6.Flett AS, Hayward MP, Ashworth MT, et al. Equilibrium contrast cardiovascular magnetic resonance for the measurement of diffuse myocardial fibrosis: preliminary validation in humans. Circulation 2010;122(2):138–144. [DOI] [PubMed] [Google Scholar]
- 7.Usman AA, Taimen K, Wasielewski M, et al. Cardiac magnetic resonance T2 mapping in the monitoring and follow-up of acute cardiac transplant rejection: a pilot study. Circ Cardiovasc Imaging 2012;5(6):782–790. [DOI] [PubMed] [Google Scholar]
- 8.Vermes E, Pantaleon C, Auvet A, et al. Cardiovascular magnetic resonance in heart transplant patients: diagnostic value of quantitative tissue markers: T2 mapping and extracellular volume fraction, for acute rejection diagnosis. J Cardiovasc Magn Reson 2018;20(1):59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Levey AS, Eckardt KU, Tsukamoto Y, et al. Definition and classification of chronic kidney disease: a position statement from Kidney Disease: Improving Global Outcomes (KDIGO). Kidney Int 2005;67(6):2089–2100. [DOI] [PubMed] [Google Scholar]
- 10.Fishbein MC, Kobashigawa J. Biopsy-negative cardiac transplant rejection: etiology, diagnosis, and therapy. Curr Opin Cardiol 2004;19(2):166–169. [DOI] [PubMed] [Google Scholar]
- 11.Bruneval P, Angelini A, Miller D, et al. The XIIIth Banff Conference on Allograft Pathology: The Banff 2015 Heart Meeting Report: Improving Antibody-Mediated Rejection Diagnostics: Strengths, Unmet Needs, and Future Directions. Am J Transplant 2017;17(1):42–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Piechnik SK, Ferreira VM, Dall’Armellina E, et al. Shortened Modified Look-Locker Inversion recovery (ShMOLLI) for clinical myocardial T1-mapping at 1.5 and 3 T within a 9 heartbeat breathhold. J Cardiovasc Magn Reson 2010;12(1):69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Giri S, Chung YC, Merchant A, et al. T2 quantification for improved detection of myocardial edema. J Cardiovasc Magn Reson 2009;11:56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Sado DM, Flett AS, Banypersad SM, et al. Cardiovascular magnetic resonance measurement of myocardial extracellular volume in health and disease. Heart 2012;98(19):1436–1441. [DOI] [PubMed] [Google Scholar]
- 15.Kellman P, Hansen MS. T1-mapping in the heart: accuracy and precision. J Cardiovasc Magn Reson 2014;16(1):2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Doesch C, Huck S, Bohm CK, et al. Visual estimation of the extent of myocardial hyperenhancement on late gadolinium-enhanced CMR in patients with hypertrophic cardiomyopathy. Magn Reson Imaging July 2010;28(6):812–819. [DOI] [PubMed] [Google Scholar]
- 17.Gräni C, Eichhorn C, Bière L, et al. Comparison of myocardial fibrosis quantification methods by cardiovascular magnetic resonance imaging for risk stratification of patients with suspected myocarditis. J Cardiovasc Magn Reson 2019;21(1):14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.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(5):643–650. [DOI] [PubMed] [Google Scholar]
- 19.Coelho-Filho OR, Shah R, Lavagnoli CFR, et al. Myocardial tissue remodeling after orthotopic heart transplantation: a pilot cardiac magnetic resonance study. Int J Cardiovasc Imaging 2018;34(1):15–24. [DOI] [PubMed] [Google Scholar]
- 20.Nishimura T, Sada M, Sasaki H, et al. Identification of cardiac rejection with magnetic resonance imaging in heterotopic heart transplantation model. Heart Vessel 1987;3(3):135–140. [DOI] [PubMed] [Google Scholar]
- 21.Spieker M, Haberkorn S, Gastl M, et al. Abnormal T2 mapping cardiovascular magnetic resonance correlates with adverse clinical outcome in patients with suspected acute myocarditis. J Cardiovasc Magn Reson 2017;19(1):38. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Shahzad K, Aziz QA, Leva JP, et al. New-onset graft dysfunction after heart transplantation--incidence and mechanism-related outcomes. J Heart Lung Transplant 2011;30(2):194–203. [DOI] [PubMed] [Google Scholar]
- 23.Aherne T, Tscholakoff D, Finkbeiner W, et al. Magnetic resonance imaging of cardiac transplants: the evaluation of rejection of cardiac allografts with and without immunosuppression. Circulation 1986;74(1):145–156. [DOI] [PubMed] [Google Scholar]
- 24.Ferreira VM, Piechnik SK, Dall’Armellina E, et al. Native T1-mapping detects the location, extent and patterns of acute myocarditis without the need for gadolinium contrast agents. J Cardiovasc Magn Reson 2014;16:36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Miller CA, Naish JH, Shaw SM, et al. Multiparametric cardiovascular magnetic resonance surveillance of acute cardiac allograft rejection and characterisation of transplantation-associated myocardial injury: a pilot study. J Cardiovasc Magn Reson 2014;16(1):52. [DOI] [PMC free article] [PubMed] [Google Scholar]