Article Text

Download PDFPDF
Original article
Changes in left and right ventricular function of donor hearts during the first year after heart transplantation
  1. Sorel Goland1,
  2. Robert J Siegel2,
  3. Kevin Burton2,
  4. Michele A De Robertis2,
  5. Asim Rafique2,
  6. Ernst Schwarz2,
  7. Kaveh Zivari2,
  8. James Mirocha2,
  9. Alfredo Trento2,
  10. Lawrence S C Czer2
  1. 1Heart Institute, Kaplan Medical Center, Rehovot, Israel
  2. 2Division of Cardiology, Department of Cardiothoracic Surgery, Cardiac Non-Invasive Laboratory, Cedars-Sinai Medical Center, Los Angles, California, USA
  1. Correspondence to Sorel Goland, Heart Institute, Kaplan Medical Center, Rehovot 76100, Israel; sorelgoland{at} or sorel_g{at}


Objective Expected values of tissue Doppler imaging (TDI) velocities and myocardial performance index (MPI) after heart transplantation (HTx) have not been evaluated. This study assessed left and right ventricular (LV and RV) structure and function during the first year after HTx using these indexes.

Methods and results Echocardiography including MPI and TDI systolic (S′), early (E′) and late (A′) diastolic velocities of RV and LV were performed in 20 donors (mean age 35±13 years) and serially in 20 recipients (mean age 59±9 years) during the first year after HTx. Increase in LV mass occurred at 7 days, with normalisation at 3 months (p<0.001). An increase in MPI (p<0.001) and a decrease in E', S' velocities on TDI occurred at week 1 with gradual improvement during the first year (p<0.001). Normalisation of LV and RV MPI occurred at 6 months (p<0.001) and LV TDI velocities at 1 year (p<0.001). TDI velocities of both ventricles, however, at 1 year remained lower than at baseline. No patient had greater than grade IA rejection during the follow-up. No significant change was found in myocyte size within the first year. However, there was a 3.3-fold increase in fibrosis.

Conclusions This study is the first to identify the normal changes of TDI and MPI of both ventricles during the first year after HTx. An increase in LV mass and impairment of bi-ventricular systolic and diastolic function occur early after HTx with gradual improvement during the first year. No significant changes in myocyte size were observed, but there was a substantial increase in fibrosis.

  • Donor
  • echocardiography (three-dimensional)
  • function
  • heart failure
  • heart transplant
  • heart transplantation
  • heart transplant pathology
  • left ventricle
  • right ventricular
  • transplantation

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

Heart transplantation (HTx) provides definitive therapy for patients with end-stage congestive heart failure. It is an effective treatment that results in improved survival in this population. However, allograft dysfunction remains a serious problem after heart transplantation. Right ventricular dysfunction is a cause of more than 50% of all early cardiac complications.1 In addition, left ventricular (LV) diastolic dysfunction has been reported early in the post-HTx period.2–7 Several attempts to describe early changes in LV wall thickness and diastolic function have been made using conventional echocardiography and heart catheterisation.8 However, there is little information on mechanisms responsible for these changes. Different results on myocyte hypertrophy and myocardial fibrosis have been reported during the early and late post-HTx follow-up.9–11 Non-invasive evaluation of right ventricular (RV) function using echocardiography is limited due to geometric problems. In addition, the assessment of LV diastolic function using conventional echo indexes may be inaccurate due to changes in preload.

Tissue Doppler imaging (TDI) velocities are relatively independent of changes in ventricular loading conditions and have been shown to have advantages in quantitative assessment of LV and RV function.12 13 TDI of the mitral annulus appears to be reliable to exclude severe rejection in adults and children.14–16 Reduction in tricuspid and mitral annular TDI velocities post-HTx has been reported in children.17 In addition, myocardial performance index (MPI) has been introduced as a reproducible non-geometric echo measurement of combined systolic and diastolic performance18 and was used for post-HTx evaluation.19–21

TDI velocities and MPI have been reported to predict cellular rejection, but the normal post-HTx time course of these values in adults have not been evaluated.14 15 19 21 Moreover, no data on the changes in LV and RV function of the donor's heart following heart transplantation have been published to date. Therefore, the purpose of our study was to assess the normal changes of LV and RV function that occur in donor hearts during the first year after transplantation using TDI and MPI, and to evaluate the morphological change that may correlate with these findings.


Patient population

The study was conducted from January 2004 to December 2007 with approval of the Institutional Review Board and patient informed consent.

Twenty consecutive heart donors with an available complete baseline echocardiogram and 20 recipients who were prospectively followed for 1 year by serial echocardiograms within 24 h of biopsy at days 7 and 3, 6, and 12 months post-HTx were included in this study. To check the hypothesis whether changes on the myocardial cellular level during the first year after HTx might contribute to changes in echo parameters, assessment of myocyte size and fibrosis on biopsy specimens was done.

The myocardial preservation, surgical transplant technique, post-transplant surveillance protocol and pathology assessment has been described elsewhere.22 The post-HTx regimen included induction therapy consisting of Thymoglobulin for 5–7 days. Maintenance immunosuppressive therapy included cyclosporine or tacrolimus, azathioprine or mycophenolate mofetil and prednisone. The detailed induction and immunosuppression therapies were published recently.23


A complete two-dimensional echo Doppler study was performed from the standard views using commercially available ultrasound systems (HDI-5000; Philips Medical Systems, Bothell, Washington, USA) with a 3.5-MHz transducer and Philips iE-33 (Philips Medical Systems, Andover, Massachusetts, USA) machine with S3 probe. Images were digitised, and off-line measurements were performed with the VERICIS Echo Review application (Camtronics Medical Systems Inc., Hartland, Wisconsin, USA). All measurements were performed according to the guidelines of the American Society of Echocardiography.24 The LV mass (LVM) was estimated by using the anatomically validated formula of Devereux. The LVM index was calculated by dividing the LVM by the body surface area. LV internal dimensions and wall thickness were measured from two-dimensional echocardiographic tracings in the parasternal long axis view.

From the apical four-chamber view, pulse-wave Doppler recordings of the mitral inflow were acquired with the sample volume placed at the tips of the mitral valve leaflets. The following parameters were measured by pulse-wave Doppler: peak velocities of early (E) and late (A) diastolic filling, and deceleration time. The ratio of early diastolic to late diastolic mitral inflow velocities was calculated (E/A). Additional time interval measurements were performed using Doppler recordings: isovolumic relaxation time (IVRT), isovolumic contraction time (IVCT) and LV ejection time. Pulsed-wave TDI was performed using spectral pulsed Doppler signal filters. On the apical four-chamber view, a 2-mm pulsed Doppler sample volume was placed at the level of the lateral corner of the tricuspid and mitral annuli. Gain and filter settings were adjusted to optimise the image. High temporal resolution (>100 frames/s) and a sweep speed set to 100 mm/s were used. Analysis was performed for early (E′) and late diastolic velocity (A′) and systolic velocity (S′). TDI of the RV and LV was not available in all donors, and was obtained from healthy age, sex and body mass index (BMI)-matched controls. The MPI was calculated as the sum of isovolumic relaxation and isovolumic contraction time divided by LV ejection time,18 (figure 1e B, supplement) and by pulsed-wave TDI for RV25 (figure 1e A, supplement). The echo reader was blinded to time intervals and biopsy results. In addition, two blinded observers independently measured the E′ (TDI) and calculated the MPI in 10 patients.

Myocyte size and fibrosis assessment

Specimen preparation, microscopy and image analysis are described in details in the Appendix (supplement). Biopsy specimens were obtained from the right ventricular septum at 1 week and 3, 6 and 12 months following heart transplant and embedded in paraffin. Specimens were stained with H&E or Gomori's trichrome. Digital images were acquired on a bright field microscope using Image Pro Plus (Media Cybernetics, Silver Spring, Maryland, USA).

Myocyte size was measured 7 days and 1 year following heart transplantation by drawing a line across the cell in longitudinal sections or around the cell in cross sections. Areas of interstitial fibrosis were identified by computer analysis of coloured pixels and expressed relative to the entire stained area of the specimen. Thin sections of biopsies were stained to distinguish between healthy myocardium and areas of fibrotic collagen (figure 3e A and B, supplement). We compared the variation in measurement of myocyte size between two observers (interobserver) and at two different times for the same observer (intraobserver). Measurements were made on 7–30 cells in one biopsy section from each of five patients. Each observer was blinded to the measurements of the other. Fibrosis was compared in a similar manner. Measurements were made on one biopsy section from each of five patients. The interobserver variation for myocyte size was 0.6% (0.2 μm), and the intraobserver variation was −0.5% (−0.1 μm). The intraobserver variation for fibrosis was 7% (0.5% fibrosis), and the interobserver variation was 1% (−0.3% fibrosis).

Statistical analysis

Continuous variables were summarised as mean±SD and categorical variables as percentages and frequencies. Within-subject comparisons of numerical variables across the Echo times were assessed by one-way repeated measures analysis of variance (ANOVA) with five levels (pre HTx, 7 days post HTx, 3 months post HTx, 6 months post HTx, and 12 months post HTx). Normality at each time was assessed using the Kolmogorov–Smirnov and Shapiro–Wilk tests. After a significant ANOVA, pre-HTx and post-HTx measures were compared using least squares means and associated p values. Paired comparisons of myocyte size between time-points were made using measurements of cells with the same orientation. Changes in myocyte size and fibrosis per cent were analysed using paired t-tests (normality satisfied) or the Wilcoxon signed rank test (normality questionable).

Statistical analysis was performed using SPSS version 13.0 and SAS version 9.1. A p-value of <0.05 was considered significant.


Patient population

The clinical pre and post-HTx characteristics of 20 recipients (mean age 59±10 years) and their donors (mean age 35±12 years) are presented in table 1. The majority of donors (80%) were male, and of non-Caucasian ethnicity (60%). Demographics of donors and controls were similar except for a larger proportion of Caucasian among controls compared with donors (table 1e, supplement). Sixty-five per cent of the recipients were male and most were ethnically Caucasian (75%). Nearly half of the recipients were on inotropic support (45%). A history of diabetes and hypertension was identified in 36% and ischaemic cardiomyopathy was the indication for HTx in 40% of patients. The mean ischaemic time was 137±5 min; all of them received induction therapy with thymoglobulin followed by standard immunosuppression protocol. Out of all of the biopsies on the 20 patients, 23 episodes of rejection 1A occurred (11 patients had at least one episode of 1A rejection). No patient had greater than grade 1R (grade 1A) rejection during the follow-up. At 1-year coronary angiography no patient had obstructive coronary artery disease. The mean systolic and diastolic blood pressures were 132.4±2.0 and 79.2±0.7 mm Hg at 1 year. Of note, the vast majority of the patients received aggressive antihypertensive therapy either with ACE inhibitors or angiotensin receptor blockers (85%) and/or calcium-channel blockers (60%).

Table 1

Donor and recipient clinical characteristics


Overall, 96 echocardiograms were analysed (out of 100, four were not available for analysis) (20 donors/controls and 76 recipients). The echocardiographic data of donor hearts pre-explantation and at 7 days, 3, 6 and 12 months after HTx are presented in table 2. RV TDI velocities of age, gender and BMI-matched controls were used as pre-HTx values for 16 donors and LV TDI velocities for 10.

Table 2

Changes in echocardiographic measurement during the first year after heart transplantation

Significant time-course changes in LV mass (LVM) as well as index (LVM/BSA) were observed (p<0.001). An increase in LV wall thickness (IVS: 10.0±1.5 vs 11.5±1.2, p=0.008) and LVM index (84.1±19.4 vs 106.2±17.7 g/m2, p=0.003) occurred early post-HTx (7 days), with return to normal values during follow-up (figure 1). At 1 year no significant increase in LV wall thickness with development of LVH was obtained (p=0.15). Normal LV ejection fraction was evident during follow-up. However, significant changes in diastolic indexes, E (p<0.001) and A velocities (p<0.001) and E/A ratio (p=0.002) were observed.

Figure 1

Changes in left ventricular (LV) mass index within the first year after heart transplantation. p-Values reflect comparisons between LV mass index pre and post-heart transplantation (HTx) at each follow-up time interval.

Changes in MPI

An increase in mean MPI of LV (0.29±0.06 vs 0.56±0.17, p<0.001) and RV (0.30±0.09 vs 0.65±0.16, p<0.001) was obtained at 7 days. Of note, recovery of LV MPI occurred earlier than of RV with gradual normalisation (figure 2) within 1 year.

Figure 2

Changes in myocardial performance index within the first year after heart transplantation. p-Values reflect comparisons between myocardial performance index (MPI) of LV (left ventricle) and RV (right ventricle) pre- (donor/matched controls) and post-heart transplantation (HTx) at each follow-up time interval.

Changes in TDI velocities

Changes in TDI velocities are presented in figure 3. TDI systolic and diastolic velocities of mitral (10.7±2.3 vs 7.4±1.1, p<0.001 and 13.5±2.8 vs 7.3±1.2, p<0.001, respectively) and tricuspid annuli (11.6±1.4 vs 7.2±1.8, p<0.001 and 11.5±1.6 vs 6.1±1.7, p<0.001, respectively) were significantly reduced at 1 week compared with pre-HTx/controls (figure 3). The LV E/E′ ratio was also elevated early after HTx (6.1±2.5 vs 11.0±2.5, p<0.001) followed by a decrease (table 2). TDI velocities improved over time and the RV improvement was slower than LV. LV systolic (S′) and diastolic (E′) velocities improved within 1 year but did not recover completely (10.7±2.3 vs 9.5±1.1, p=0.03 and 13.5±2.8 vs 11.5±2.6, p=0.01, between pre-transplant/controls and 1 year, respectively). Moreover, E′ and S′ velocities of RV remained significantly lower than in pre-transplant/controls (8.9±2.0 vs 11.5±1.6, p<0.001 and 9.9±2.2 vs 11.6±1.4, p<0.001).

Figure 3

Changes in systolic and diastolic tissue Doppler velocities of both ventricles within the first year after heart transplantation. p Values reflect comparisons between systolic (S′) and early diastolic (E′) tissue Doppler velocities of LV (left ventricle) and RV (right ventricle) pre (donor/matched controls) and post-heart transplantation (HTx) at each follow-up time interval.

In addition to standard echo variables, no differences were observed when compared LV systolic (E′) and diastolic (S′) velocities (13.5±2.9 vs 13.6±3.0, p=0.9 and 10.5±28 vs 10.9±1.5, p=0.8) and E′ and S′ velocities of RV (11.6±2.0 vs 11.5±1.2, p=1.0 and 11.7±1.2 vs 11.4±1.6, p=0.7) between pre-transplant donors and matched controls (table 1e, supplement). The interobserver and intraobserver variability for E′ was 5%±2% and 4%±2%; and for MPI was 6%±5% and 4%±4%.

Changes in myocyte size

The myocytes ranged in size approximately 15 and 30 μm, averaging about 20 μm at 7 days and at 1 year after transplantation (table 3). The diameters at the two time points were compared for each of 20 patients although for a given patient the change in diameter could vary greatly (minimum −57%, maximum +56%). The mean change for all patients was not significantly different from zero (−1.4±1.4 μm, or −3±6%). In addition, a poor correlation was observed when differences (7 days to 1 year) in myocyte size were compared with differences in LVM for each patient (r=0.6).

Table 3

Quantitation of myocyte diameter measurements

Changes in fibrosis

Fibrosis increased 1.6-fold between 7 days and 6 months following transplantation, 2.1-fold between 6 and 12 months, and thus the total increase was 3.3-fold over 1 year (figure 4). In addition, paired comparisons of the changes from 1 week to 6 months (p=0.01), from 6 to 12 months (p=0.007) and from 1 week to 1 year (p=0.001) were significant. The changes over all time points were highly significant (p<0.001).

Figure 4

Change in fibrosis within the first year after heart transplantation. Mean±SEM at each time point for number of patients as indicated.


This is the first study to identify the expected values of TDI velocities and MPI of both the LV and RV for heart transplant recipients during the first year after surgery in the absence of clinically significant rejection. The main findings of this study are: (1) impairment of bi-ventricular systolic and diastolic function assessed by TDI and MPI occurs early after HTx with gradual improvement during the first year; (2) increase in LVM occurred very early with normalisation at 3 months, with no significant LV hypertrophy development at 1 year follow-up; (3) substantial increase in fibrosis within the first year after HTx was found with no significant changes in myocyte sizes.

Normal changes in MPI and TDI velocities during the first year after HTx

Allograft dysfunction is one of the most serious post-HTx problems. Therefore, non-invasive assessment is important to potentially detect and prevent complications. RV dysfunction is a major complication early after HTx.28 It is well recognised, however, that standard two-dimensional echo assessment of RV function is limited due to its asymmetric shape and TDI has been found useful for RV function assessment.26 29 30 In addition, a number of studies demonstrated abnormalities in LV function that occurred even in the absence of rejection early post-HTx.2–4 Post-HTx LV diastolic dysfunction has been reported in asymptomatic patients as well as in patients with heart failure.2 3

Restrictive mitral inflow pattern is seen after HTx in recipients with and without cellular rejection.27 28 Unfortunately, Doppler mitral inflow patterns have been found unreliable to differentiate normal and abnormal diastolic function in the post-HTx setting due to changes in preload as well as due to the variable age of donors.27 29 TDI that is relatively independent of loading conditions has been introduced for better assessment of LV function.12 13 This technique has been used mostly to identify complications such as rejection or graft incompetence after HTx. Stengel et al14 reported the usefulness of high late TDI diastolic velocities to exclude severe rejection. Fyfe et al30 showed that a progressive decrease in tricuspid annular systolic (S′) and diastolic (E′) velocities was able to identify children with post HTx pre-terminal graft failure.

In this study, we report the normal expected changes in TDI velocities early after surgery and also compared the post-HTx values to pre-transplant/controls measurements. We found a significant decrease in early diastolic (E′) and systolic (S′) velocities of both ventricles at 7 days in the absence of rejection with gradual improvement within the first year, remaining, however, lower compared with pre-transplant/controls, suggesting residual impairment of the allograft function at follow-up.

The reduction of TDI velocities of both ventricles in the first week can be explained by a number of factors such as inadequate myocardial preservation and ischaemic injury, the effect of cardiopulmonary bypass as well as the patient's immune response. All of these factors can result in myocardial oedema manifested by increased wall thickness and impaired bi-ventricular systolic and diastolic function. Elevated pulmonary vascular resistance pre-HTx is an additional factor that may contribute to early RV dysfunction, as the donor's heart has not previously been exposed to increased pulmonary artery systolic pressure.

The lack of complete recovery of systolic and diastolic velocities, which occurs to a greater degree with the RV may reflect some residual allograft dysfunction that may be related to the increase in percentage fibrosis found on the post-transplant myocardial biopsy specimens.

MPI has been shown to be useful to identify rejection in the post-HTx population,19 21 but the normal longitudinal changes early after HTx have not been reported. In our study we found an increase in mean MPI of the LV and RV at 1 week after surgery with further reduction and normalisation. However, normalisation of LV MPI occurs earlier than the RV, indicating that the RV may need a longer time to recover.

Changes in LVM and myocardial structure within the first year after HTx

We found a significant increase in LV wall thickness and LVM and index in the first week after surgery without histological evidence of changes in myocyte size or increased fibrosis on biopsies. There was a normalisation of LVM within the first year suggesting interstitial tissue oedema as a possible explanation for the early increase in LVM. Similar to our study, others have found significant increases in LV wall thickness compared with pre-explantation values in spite of the absence of rejection, myocyte hypertrophy or interstitial fibrosis.2 3 8 Pereira et al8 have hypothesised that transient increases in LVM as well as impaired diastolic dysfunction are caused by myocardial oedema.

We did not find a significant difference between wall thickness and LVM at pre-explantation and at 1 year indicating that LVH did not develop within the first post-HTx year. The absence of a significant increase in LVM at follow-up is supported by a retrospective study by Armstrong et al10 showing that for individual mean values at various time points a statistical increase in mean diameter occurred only at 2 years and remained unchanged at all subsequent time points.

In the current study the ischaemic time was short, there were no patients with donor/recipient mismatch, all patients received induction therapy and no patient had rejection grade higher than 1A; moreover, only a half (11/20) had one episode of 1A. The mean blood pressures at follow-up were well controlled and all patients received aggressive antihypertensive treatment. Therefore, controlling the main factors that promote hypertrophy in the post-HTx setting appear to have a favourable effect preventing early LVH and extensive myocyte hypertrophy. We recently supported this finding with a large retrospective study on 427 patients reporting low rates of post-HTx LVH (32–38%) at 5 years. Moreover, patients received donor hearts with LVH existing showed regression of LVH at follow-up.22


Although our study is the first longitudinal prospective study to look at changes in cardiac structure, function and morphology early after HTx, it has limitations. First, we had a relatively small number of patients, but they were studied in a very systematic manner using echo acquisition within 24 h of biopsy. Second, TDI measurements are angle dependent, but all efforts were made to align the pulsed-wave beam vertically with the sample site. Third, we did not have data on TDI on all donor hearts before explanation, but age, gender and BMI-matched controls were used. Moreover, the pre-transplant systolic and diastolic TDI velocities of the donors were similar to those of controls.


Our study identifies the normal values and changes in structure and function that occur normally in a donor heart after heart transplantation in the absence of significant cellular rejection. An increase in LVM and a decrease in TDI velocity and MPI very early after surgery occur regardless of change in myocyte size and may be related to interstitial myocardial oedema. Normalisation of LVM as well as improvement of bi-ventricular function with some residual allograft dysfunction normally occurs at 1 year.


The authors gratefully thank Dr Daniel Luthringer for his expertise and guidance during the analysis of the myocardial biopsies.



  • See Editorial, p 1634

  • Funding This work was supported in part by a grant from the Rav-Noy Foundation and the Save-A-Heart Foundation, Cedars-Sinai Medical Center, Los Angeles, California, USA.

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval This study was conducted with the approval of the Cedars-Sinai Medical Center, Los Angles, California, USA.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles