Author + information
- Received March 13, 2015
- Revision received October 2, 2015
- Accepted October 7, 2015
- Published online March 1, 2016.
- Christian Erbel, MDa,
- Nodira Mukhammadaminovaa,
- Christian A. Gleissner, MDa,
- Nael F. Osman, PhDb,
- Nina P. Hofmann, MDa,
- Christian Steuera,
- Mohammadreza Akhavanpoor, MDa,
- Susanne Wangler, MDa,
- Sultan Celik, MDa,
- Andreas O. Doesch, MDa,
- Andreas Voss, MDc,
- Sebastian J. Buss, MDa,
- Philipp A. Schnabel, MDd,
- Hugo A. Katus, MDa and
- Grigorios Korosoglou, MDa,∗ ()
- aDepartment of Cardiology, University Hospital Heidelberg, Heidelberg, Germany
- bMyocardial Solutions, Inc., Morrisville, North Carolina
- cInstitute of Psychology, University of Heidelberg, Heidelberg, Germany
- dInstitute for Pathology, University Hospital Heidelberg, Heidelberg, Germany
- ↵∗Reprint requests and correspondence:
Dr. Grigorios Korosoglou, University of Heidelberg, Department of Cardiology, Im Neuenheimer Feld 410, Heidelberg 69120, Germany.
Objectives This study sought to evaluate myocardial perfusion reserve index (MPRI) and diastolic strain rate, both assessed by cardiac magnetic resonance (CMR) as a noninvasive tool for the detection of microvasculopathy.
Background Long-term survival of cardiac allograft recipients is limited primarily by cancer and cardiac allograft vasculopathy (CAV). Besides epicardial CAV, diagnosed by coronary angiography, stenotic microvasculopathy was found to be an additional independent risk factor for survival after heart transplantation.
Methods Sixty-three consecutive heart transplant recipients who underwent CMR, coronary angiography, and myocardial biopsy were enrolled. Stenotic vasculopathy in microvessels was considered in myocardial biopsies by immunohistochemistry and CAV was graded during coronary angiography according to International Society of Heart and Lung Transplantation criteria. In addition, by CMR microvasculopathy was assessed by myocardial perfusion reserve during pharmacologic hyperemia with adenosine and strain-encoded magnetic resonance using a modified spatial modulation of magnetization tagging pulse sequence in all patients.
Results Decreasing MPRI and diastolic strain rates were observed in patients with decreasing microvessel luminal radius to wall thickness ratio and decreasing capillary density (r = 0.45 and r = 0.61 for MPRI and r = 0.50 and r = 0.38 for diastolic strain rate, respectively; p < 0.005 for all). Using multivariable analysis, both MPRI and diastolic strain rate were robust predictors of stenotic microvasculopathy, independent of age, organ age, and CAV by International Society of Heart and Lung Transplantation criteria (hazard ratio: 0.07, p = 0.006 for MPRI; hazard ratio: 0.91, p = 0.002 for diastolic strain rate). Patients without stenotic microvasculopathy in the presence of no or mild CAV (n = 36) exhibited significantly higher median survival free of events, compared with patients with stenotic microvasculopathy in the presence of no or mild CAV (n = 18; p = 0.04 by log rank).
Conclusions CMR represents a valuable noninvasive diagnostic tool, which may be used for the early detection of transplant microvasculopathy before the manifestation of CAV during surveillance coronary angiographic procedures.
- capillary density
- cardiac magnetic resonance
- heart transplantation
- microvessel lumen to wall thickness
- myocardial perfusion reserve index
- strain-encoded CMR
Long-term graft survival after heart transplantation is still limited by cardiac allograft vasculopathy (CAV) (1). Every second heart transplant (HTx) recipient develops CAV within 10 years after transplantation, regardless of age.
CAV affects epicardial blood vessels and microvasculature in cardiac allograft recipients, but the pathophysiology of the 2 vascular regions is different. CAV causes concentric intimal hyperplasia in epicardial coronary arteries (2,3), whereas medial thickening is seen in intramyocardial blood vessels (4).
Microvasculopathy can occur in parallel with epicardial coronary artery stenosis, but most cardiac allograft recipients with a microvascular dysfunction have no evidence of impaired epicardial physiology (5). Both microvasculopathy and macrovasculopathy result in reduced myocardial perfusion and function, which leads to increased morbidity and mortality rates (6). More importantly, the prognostic impact for long-term survival of transplanted patients with microvasculopathy is independent of that of macrovasculopathy (4). Thus, early detection of CAV is an important clinical goal in this context. Although CAV of epicardial blood vessels is a common part of the post-transplant routine follow-up visits, screening for microvasculopathy is still missing in a routine setting.
In the present study, we sought to investigate whether myocardial perfusion reserve index (MPRI) and myocardial strain, both assessed by cardiac magnetic resonance (CMR), are related to histological parameters of microvascular integrity, including capillary density and luminal radius to wall thickness in 10- to 20-μm diameter microvessels, assessed by myocardial biopsy.
Study population and separation into patient subgroups
Sixty-three consecutive HTx recipients underwent routine coronary angiography, myocardial biopsy, and CMR within 4 weeks from cardiac catheterization. The International Society of Heart and Lung Transplantation (ISHLT) criteria for acute cellular rejection were applied for each endomyocardial biopsy. At the time of the study, all HTx recipients were in stable condition, without clinical signs of heart failure (New York Heart Association functional class >II, ejection fraction <40%), unstable angina, or acute/ongoing rejection, diagnosed by histology (ISHLT class >IA) (2). Patients received a standard double immunosuppressive therapy (mycophenolate mofetil and a calcineurin inhibitor-cyclosporine A or tacrolimus) (7). Thirty-six patients (57% of the present cohort) were included in previous reports, where the ability of angiographic myocardial blush grade was investigated to predict cardiac outcomes in transplant recipients (8). According to the results of myocardial biopsy and coronary angiography, patients were separated as follows: Group A, including patients with no or mild CAV by angiography and without stenotic microvasculopathy; Group B, including patients with no or mild CAV by angiography but evidence of stenotic microvasculopathy; and Group C, including patients with both manifest stenotic macrovasculopathy and microvasculopathy, which were subsequently excluded for correlative and receiver operating characteristic (ROC) analysis.
All procedures complied with the Declaration of Helsinki and were approved by our local ethics committee, and all patients gave written informed consent.
HTx recipients were examined in a clinical 1.5-T whole-body magnetic resonance (MR) scanner Achieva system (Philips Medical Systems, Best, the Netherlands). This is part of our institutional post-transplant protocol performed in HTx recipients. Patients were asked to refrain from caffeine intake 12 h before testing. A standardized imaging protocol was used, aiming at the assessment of baseline parameters of left ventricle (LV), such as LV diameters, septal and lateral wall thickness, and ejection fraction using cine imaging. Myocardial perfusion reserve was assessed during pharmacologic hyperemia with adenosine in all patients.
Myocardial perfusion imaging
A 3-slice turbo field echo-echo-planar imaging sequence was used as described previously (9). Stress perfusion imaging was performed using a continuous intravenous infusion of 140 μg/kg/min adenosine. Three heartbeats after initiation of the sequence a bolus of 0.04 mmol/kg bodyweight gadolinium-DTPA (Magnevist, Berlex, New Jersey) was injected over an antecubital vein at a rate of 5 ml/s flushed with 20 ml 0.9% sodium chloride.
Semiquantification of myocardial perfusion was conducted in 3 LV short-axis slices using View Forum software (Philips Medical Systems). Manual contouring of endomyocardial and epimyocardial borders was assessed on the image with the brightest contrast enhancement in LV and an automated algorithm was used to match contours in the remaining images of the slice. The myocardium was divided into 6 equiangular segments per slice and spatially averaged signal intensity (SI) values were used to plot SI curves over time of the whole myocardial circumference and for a region of interest in the center of the LV blood pool. The mean SI before contrast agent injection was subtracted from post-contrast data, and the upslope of the resulting SI time curves was determined by using a linear fit, based on the least squares regression line. With this approach, a mean MPRI was calculated by dividing the corrected upslope at pharmacologic hyperemia through the corrected upslope at rest in 16 myocardial segments, which served as a semiquantitative estimate of the global perfusion reserve in each patient (10).
Strain-encoded MR (SENC) was performed using a modified spatial modulation of magnetization tagging pulse sequence. Technical details with this sequence are described elsewhere (9,11). Briefly, in our study SENC images of 3 short-axis planes were acquired with 10-mm slice thickness. Typical parameters were as follows: field of view = 350 × 350 mm2, matrix size = 80 × 80, flip angle = 30°, repetition time/echo time = 25/0.9 ms, and acquired voxel size = 4.4 × 4.4 × 10 mm3. For SENC images, a temporal resolution of 25 ms was used. The total scan duration was 10 to 14 s.
Quantification of myocardial strain and strain rate was conducted using Diagnosoft software Version 1.06 (Diagnosoft Inc., Palo Alto, California). Because with SENC the tagging modulation gradient is applied in slice selection direction, short axis strain-encoded images were used for quantification of longitudinal myocardial strain and strain rate (12). For each segment, temporal course of regional myocardial strain was registered throughout the cardiac cycle, and quadratic interpolation was used to calculate: 1) the peak systolic strain (S expressed in %); 2) the peak systolic strain rate during systole (expressed in 1/s); and 3) the mean diastolic strain rate over the duration from peak systole to 50% of the diastole (mean early diastolic strain rate, expressed in 1/s) (13). Normal values for strain and strain rate during systole and diastole, respectively, were assessed in control subjects.
Cardiac catheterization and assessment of CAV
Coronary angiograms were performed in a standardized fashion under 200 μg of intracoronary nitroglycerin. Coronary vessels were classified by ISHLT nomenclature for CAV (14) as follows: no CAV (grade 0), no detectable angiographic lesion; mild CAV (grade 1), angiographic left main <50%, or primary vessel with maximum lesion of <70%, or any branch stenosis <70% (including diffuse narrowing) without allograft dysfunction; moderate CAV (grade 2), angiographic left main >50%, a single primary vessel >70%, or isolated branch stenosis >70% in branches of 2 systems, without allograft dysfunction; severe CAV (grade 3), angiographic left main >50%, or 2 or more primary vessels with >70% stenosis, or isolated branch stenosis >70% in all 3 systems; or CAV grade 1 or 2 with allograft dysfunction (defined as LV ejection fraction <45% in the presence of regional wall motion abnormalities).
Biopsy harvest procedures and histological tissue preparations
Endomyocardial biopsies were acquired during cardiac catheterization of the LV from all patients after heart transplantation as part of routine institutional post-transplant care. Experienced cardiac pathologists reviewed and graded the biopsies according to the nomenclature by the ISHLT, through the Standardized Grading System of 1990 (2) and 2005 (15). Biopsy specimens were used for cryosections of 5 μm for immunohistochemistry staining.
CD34, elastica–van Gieson, and hematoxylin-eosin staining
Immunohistochemistry was prepared as described previously (16). Briefly, after fixation in acetone primary antibody CD34 (eBioscience, Frankfurt am Main, Germany) was used. Isotype- and concentration-matched mouse control mAb served as negative control subjects. Biotinylated, polyclonal swine anti-goat-mouse-rabbit immunoglobulin (Dako, Hamburg, Germany) was used as a secondary reagent (1/200, 20 min at room temperature). For hematoxylin-eosin staining, cryosections were incubated with hematoxylin for 90 s and with 0.1% eosin for 3 min. For elastica–van Gieson staining, cryosections were incubated with celestin blue, hematoxylin, and Curtis stain for 5 min, respectively.
Quantitative assessment and interpretation of histological parameters
Assessment of microvasculopathy was performed by light microscopy (×200 to 400, Nikon-widefield-automated-microscope-Ni-E and Nikon-DS-Ri1-color-CCD-camera; Nikon, Duesseldorf, Germany) and analyzed by using imaging software/image processing and analysis software (Nikon-NIS-Elements-AR) at the Nikon-Imaging-Center at the University of Heidelberg. Through this method, overlapping was automatically avoided. The ratio of luminal radius to wall thickness was quantified in each microvessel (10 to 20 μm) as described by Hiemann et al. (4). Wall thickening was considered as stenotic if the ratio of luminal radius to wall thickness was <1 (4). Stenotic microvasculopathy was diagnosed if there was evidence of microvascular stenosis (luminal radius to wall thickness ratio <1) caused by either endothelial or wall thickening in at least 1 blood vessel per field of view. Five to 12 arterioles per field were investigated. Such patients have been reported to exhibit a significantly higher rate of lethal cardiac events, compared with patients without stenotic microvasculopathy independent of the presence of CAV by coronary angiography (4). For capillary density, an independent influence factor on impaired microcirculatory hemodynamics (17), the number of capillaries per square millimeter endomyocardial tissue were counted.
Study nurses unaware of CMR and coronary angiography results contacted each subject or an immediate family member and the date of this contact was used to calculate the follow-up time duration. Outcome analysis included the combined endpoint of cardiac death (death caused by intractable heart failure or myocardial infarction or sudden death caused by infarction or severe arrhythmia) and acute coronary syndrome. All patients were followed for at least 6 months and clinical follow-up data were obtained. No patients were lost during the follow-up period.
Analysis was performed using the commercially available software MedCalc9.3 (MedCalc software, Mariakerke, Belgium). Continuous variables were expressed as mean ± SD and categorical variables as proportions. Group differences between continuous variables were tested using analysis of variance with Bonferroni adjustment for multiple comparisons. Differences between ordinal variables were tested using the exact Mann-Whitney test, and differences between nominal variables were assessed using the Fisher exact test. All tests were 2-tailed. Correlations were analyzed by linear regression method. ROC was used to determine the value of MPRI and diastolic strain rate for the detection of stenotic microvasculopathy by pre-defined histological criteria (4). Furthermore, the association between clinical parameters (i.e., age, organ age), angiographic parameters (CAV by ISHLT), MPRI, and myocardial strain was investigated using logistic regression models and multivariable procedures. Kaplan-Meier curves were generated for the estimation of the cardiac events during the follow-up duration. Serial CMR examinations were analyzed using paired Student t test (2-tailed). Intraobserver and interobserver variability for semiquantification analysis of MPRI and myocardial strain were calculated by repeated analysis of 30 randomly selected patients by 2 observers with more than 5 years’ experience in cardiovascular imaging who were blinded to clinical and CMR data (N.P.H. and G.K.). The readings were separated by 8 weeks to minimize recall bias. Differences were considered statistically significant at p < 0.05.
Demographic and baseline CMR parameters
Demographic and clinical data of HTx recipients (n = 63) are illustrated in Table 1. Overall, 36 patients showed no or mild macrovascular and no microvascular disease (Group A), 18 had microvasculopathy in the presence of no or mild macrovascular disease (Group B), whereas 9 patients exhibited stenotic microvasculopathy and more than mild CAV by ISHTL criteria (CAV grade 2 or 3) and were excluded from subsequent correlative analysis (Group C). Demographic characteristics, baseline CMR parameters, and medications did not significantly differ among the 3 patient groups.
Correlation of histopathologic results and myocardial perfusion and strain
Increasing MPRI and diastolic strain rates were observed with increasing microvessel luminal radius to wall thickness ratio and increasing capillary density after exclusion of patients with macrovascular disease (Figures 1A to 1D).
Patients with stenotic microvasculopathy with no or only mild CAV (Group B) exhibited significantly lower MPRI and diastolic strain rate compared with Group A (without microvasculopathy), but similar to those observed in Group C (with manifest macrovascular CAV) (Figures 2A and 2B).
In addition, ROC analysis demonstrated that cutoff values of MPRI = 1.75 and early diastolic strain rate = 75/s demonstrated sensitivities of 61% and specificities of 78% and 86%, respectively, for the prediction of stenotic microvasculopathy by predefined histological criteria, after exclusion of patients with stenotic macrovasculopathy (Figures 3A and 3B).
Figures 4A to 4H illustrate a patient with normal microvessels (luminal radius to wall thickness >1), high capillary density of ∼889/mm2 by histology, and only mild CAV by invasive angiography. Corresponding CMR measures demonstrated an MPRI of 2.6 and global systolic strain and diastolic strain rate of -21% and 95/s, respectively. Another patient (Figures 4I to 4P), however, showed stenotic microvasculopathy (luminal radius to wall thickness ratio <1) and low capillary density (∼286/mm2) by histology, but only mild CAV by angiographic criteria. Corresponding CMR measures, however, exhibited markedly reduced MPRI of 1.3, preserved systolic strain of -18% and reduced early diastolic strain rate of 55/s, strongly indicative of stenotic microvasculopathy.
Using multivariable analysis, both MPRI and diastolic early strain rate were robust predictors of stenotic microvasculopathy, independent of age and organ age, after exclusion of patients with stenotic macrovasculopathy from analysis (Table 2).
During a mean follow-up duration of 3.1 ± 1.4 years there were 10 cardiac events, including 5 cardiac deaths, 2 nonfatal infarction, and 3 urgent coronary revascularizations (caused by unstable angina or symptomatic heart failure). Event rates for the 3 patient groups are illustrated in Figure 5. Pairwise comparison of Kaplan-Meier curves showed that Groups B and C did not significantly differ (p = 0.75 by log rank), whereas significant differences were noted between Group A and Group B (p = 0.04 by log rank) and between Group A and Group C (p = 0.005 by log rank).
In addition, serial CMR (n = 30) at 3.7 ± 1.1 years of follow-up showed significantly reduced MPRI only in patients with stenotic microvasculopathy (1.3 ± 0.18 during follow-up vs. 1.6 ± 0.36 at baseline CMR; p = 0.04) (Online Figure 1), whereas MPRI remained unchanged in patients without stenotic microvasculopathy. ROC analysis further showed that initially assessed microvessel luminal radius to wall thickness ratio was strongly associated with follow-up MPRI (area under the curve = 0.98; SE = 0.02; p < 0.001) (Online Figure 1).
Furthermore, logistic regression demonstrated that both MPRI and diastolic early strain rate were predictive of cardiac events independent of age and organ age (MPRI p = 0.04 and mean diastolic strain rate [1/s] p = 0.006) (Table 3).
The results of our study indicate that patients with microvasculopathy exhibit reduced MPRI and diastolic strain rates compared with those without microvasculopathy. In addition, patients with histological and/or CMR evidence of microvasculopathy exhibit significantly worse cardiac outcomes compared with those without microvasculopathy. Thus, the assessment of myocardial perfusion and strain by CMR seems to be sensitive in this context and may therefore be useful surrogate markers for adverse clinical outcomes in the absence of angiographically evident CAV. Finally, semiquantification of myocardial perfusion reserve, systolic strain, and diastolic strain rate is feasible by CMR in HTx recipients with high reproducibility.
Present results and clinical implications
Besides epicardial changes, which are routinely checked, microvascular disease has recently become the focus of scientific interest in HTx recipients (4,5,18). Such microvascular changes seem to occur independent of the presence or absence of epicardial disease in cardiac allograft recipients (19). In this regard, it was postulated that the underlying pathogenic mechanisms might be different (19).
Microvasculopathy includes structural changes (i.e., vascular remodeling with reduced lumen-to-wall ratio) and functional changes (i.e., vasoconstriction governed by neurohumoral factors and endothelial dysfunction) (20). The importance of microvascular endothelial changes was initially studied by Hollenberg et al. (21), who demonstrated that epicardial and microvascular endothelial dysfunction result in a higher grade of CAV development, ischemic events, and death. In the same line, Hiemann et al. (4) showed that stenotic microvasculopathy is a relevant prognostic factor for long-term survival independent of epicardial disease. In contrast, Kubrich et al. (22) found no association between microvascular endothelial changes and outcome. The underlying reasons might include different patient numbers (21,22); different study endpoints; and different methods used for the identification of microvascular disease, such as invasive coronary flow velocity reserve and index of microcirculatory resistance versus CMR and histopathology.
Previous work of others and our group implicated (contrast-enhanced) MR imaging as a well-suited diagnostic tool for the detection of CAV (7,23). Hussain et al. (23) postulated late gadolinium enhancement by CMR as a valuable tool to noninvasively detect coronary vessel intimal wall thickening. Our group further demonstrated that contrast-enhanced CMR identifies silent myocardial infarction myocardial fibrosis already in apparently event-free cardiac allograft recipients with absent or mild angiographic CAV (7). Furthermore, we and others previously demonstrated that assessment of myocardial perfusion reserve during pharmacological hyperemia and estimation of baseline myocardial deformation using SENC are valuable parameters for the early detection of CAV in HTx recipients (9,24,25). Moreover, Miller et al. (25) were recently able to identify epicardial and microvascular CAV by CMR myocardial perfusion reserve, outperforming conventional coronary angiography in terms of sensitivity and specificity. In the present study, we could confirm that reduced CMR perfusion reserve is associated with microvasculopathy. We further extended the previously published CMR data by showing that reduced myocardial perfusion reserve and impaired myocardial relaxation (i.e., diminished diastolic strain rates) by SENC are associated with histological markers of CAV, including stenotic microvasculopathy in 10- to 20-μm diameter coronary microvessels and reduced capillary density in the myocardium of HTx recipients. Importantly, reduced myocardial perfusion reserve and impaired myocardial relaxation by SENC both seem to represent early markers of microvascular disease because they seem to be substantially altered in patients with stenotic microvasculopathy and no or mild evidence for macrovasculopathy by angiographic criteria. Interestingly, the weak association between luminal radius and wall thickness in microvessels with capillary density may indicate different pathophysiologic mechanisms of microvascular and capillary CAV, which warrants further investigation in future experimental studies. The lack of association between microvascular and macrovascular CAV, however, is in agreement with previous studies (4). In conclusion, early detection of microvasculopathy by CMR could have a prognostic impact for long-term survival of cardiac allograft recipients. According to our results, we propose a possible diagnostic algorithm for the risk stratification of HTx recipients, illustrated in Figure 6.
Technical and methodological aspects with CAV detection
Surveillance radiograph coronary angiography has been used over the past 3 decades for the detection of the epicardial CAV. However, even stenotic macrovasculopathy may be difficult to detect using radiograph angiography because of its concentric and diffuse nature (26). Therefore, numerous studies proposed the use of invasive methods to enhance the sensitivity to detect epicardial CAV, such as intravascular ultrasound (IVUS) or fractional flow reserve as well as microvasculopathy by coronary flow velocity reserve and index of microcirculatory resistance measurements (22,27–30). However, even if angiography is combined with IVUS and the other techniques, involvement of intramural coronary vessels remains undetected, despite its adverse prognostic implications for HTx recipients even in the absence of stenotic macrovasculopathy (4). Moreover, IVUS and the other methods are costly and invasive procedures and necessitate the additional insertion of a coronary guidewire and microcatheter into the coronary arteries and may therefore be associated with further risk for patients.
Coronary angiography was conducted without the use of IVUS. Therefore, intimal thickening in the presence of angiographically normal coronary arteries may have remained undetected. In addition, some complications driven by CAV, such as myocardial infarction and death, may have occurred in the context of normal or mildly abnormal angiographic findings and in the absence of grade 2 or 3 by ISHTL criteria, which represent advanced disease. Furthermore, invasive coronary flow velocity reserve and microvascular resistance measures, which were recently shown to be closely related to MPRI by CMR (25), were not yet available during our study. However, measurement of the myocardial perfusion reserve by CMR is a clinically established procedure, which was performed in all our patients and has been previously validated in several clinical and experimental settings. Moreover, our multivariate analysis should be interpreted with caution because of the limited number of cardiac allograft recipients with microvasculopathy.
Reduced myocardial perfusion reserve during pharmacological hyperemia and impaired relaxation by SENC are both markers of stenotic microvasculopathy in 10- to 20-μm diameter coronary microvessels and of reduced capillary density in the myocardium of HTx recipients. Both these CMR markers are significantly reduced in patients with isolated stenotic microvasculopathy by histology and exhibit reduced median survival free of events. Thus, CMR may aid the risk stratification of HTX recipients by identifying those with microvasculopathy.
COMPETENCY IN MEDICAL KNOWLEDGE: Besides epicardial cardiac allograft vasculopathy stenotic microvasculopathy is an additional independent risk factor for poor survival post heart transplantation. Myocardial perfusion reserve index and diastolic strain rate by CMR has been used as a potent noninvasive tool for the detection of microvasculopathy.
TRANSLATIONAL OUTLOOK: Although coronary angiography and IVUS provide reliable information for the detection of epicardial cardiac allograft vasculopathy, CMR is the only noninvasive technique that allows for precise detection of transplant microvasculopathy. A close relationship exists between CMR measures and cardiac histology. Patients with no or mild epicardial vasculopathy, but with microvasculopathy by CMR, were reclassified as patients at high risk for future events. This could be confirmed by the lower rate of median survival free of cardiovascular complications in such patients during follow-up. Whether intensive treatment of stenotic microvasculopathy would favorably influence the poor prognosis of such cardiac allograft recipients needs to be investigated in future prospective, randomized trials.
The authors thank magnetic resonance technicians Angela Stöcker-Wochele, Birgit Hoerig, and Sina Fenker for the high-quality cardiac stress magnetic resonance examinations, and Nadine Wambsganss for excellent technical assistance.
For a supplemental figure, please see the online version of this article.
Dr. Osman is a founder and shareholder in Diagnosoft Inc., the software used for the analysis of the acquired strain-encoded magnetic resonance images. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- cardiac allograft vasculopathy
- cardiac magnetic resonance
- heart transplant
- International Society of Heart and Lung Transplantation
- intravascular ultrasound
- left ventricle
- magnetic resonance
- myocardial perfusion reserve index
- receiver operating characteristic
- strain-encoded MR
- signal intensity
- Received March 13, 2015.
- Revision received October 2, 2015.
- Accepted October 7, 2015.
- American College of Cardiology Foundation
- Hiemann N.E.,
- Wellnhofer E.,
- Knosalla C.,
- et al.
- Haddad F.,
- Khazanie P.,
- Deuse T.,
- et al.
- Tuzcu E.M.,
- Kapadia S.R.,
- Sachar R.,
- et al.
- Steen H.,
- Merten C.,
- Refle S.,
- et al.
- Nagel E.,
- Klein C.,
- Paetsch I.,
- et al.
- Korosoglou G.,
- Lossnitzer D.,
- Schellberg D.,
- et al.
- Erbel C.,
- Chen L.,
- Bea F.,
- et al.
- Kaufmann P.A.,
- Camici P.G.
- Hollenberg S.M.,
- Klein L.W.,
- Parrillo J.E.,
- et al.
- Hussain T.,
- Fenton M.,
- Peel S.A.,
- et al.
- Muehling O.M.,
- Wilke N.M.,
- Panse P.,
- et al.
- Miller C.A.,
- Sarma J.,
- Naish J.H.,
- et al.
- Al-Saadi N.,
- Nagel E.,
- Gross M.,
- et al.
- Ng M.K.,
- Yeung A.C.,
- Fearon W.F.
- St Goar F.G.,
- Pinto F.J.,
- Alderman E.L.,
- et al.