Diagnostic and Prognostic Value of a TDI-Derived Systolic Wall Motion Analysis as a Screening Modality for Allograft Rejection after Heart Transplantation

Background: Despite the risk for complications, allograft surveillance after orthotopic heart transplantation (OHT) is performed by cardiac catheterization and biopsies. We investigated the diagnostic and prognostic value of a TDI-derived systolic wall motion analysis of the posterobasal wall of the left ventricle (Sm) as a screening modality in OHT aftercare. Methods: We examined data of 210 eligible patients who underwent OHT between 2010 and 2020. Forty-four patients who had died within the initial hospital stay were excluded. For 166 patients, baseline and follow-up data were analyzed. The mean age at OHT was 46.2 (±11.4) years; 76.5% were male. Results: Within the observational period, 22 (13.3%) patients died. In total, 170 episodes of acute cellular or humoral rejections occurred (84 ISHLT1R; 13 ISHLT2R; 8 ISHLT3R; 65 AMR), and 29 catheterizations revealed cardiac allograft vasculopathy (5 CAV1; 4 CAV2; 20 CAV3). Individual Sm radial/longitudinal remained stable within the follow-up period (11.5 ± 2.2 cm/s; 10.9 ± 2.1 cm/s). Patients with acute rejections and CAV3 showed significant Sm radial/longitudinal reductions (AMR1: 1.6 ± 1.9 cm/s, confidence interval (CI) 0.77–0.243, p < 0.001; 1.8 ± 2.0 cm/s, CI 0.92–0.267, p < 0.001. ISHLT1R: 1.7 ± 1.8 cm/s, CI 1.32–2.08, p < 0.001; 2.0 ± 1.6 cm/s, CI 1.66–2.34, p < 0.001. CAV3: 1.3 ± 2.5 cm/s, CI 0.23–2.43, p < 0.017; 1.4 ± 2.8 cm/s, CI 0.21–2.66, p < 0.021). Lower Sm was associated with a threefold increase in all-cause mortality (hazard ratio (HR) 3.24, CI 1.2–8.76, p = 0.020; HR 2.92, CI 1.19–7.18, p = 0.019). Overall, Sm-triggered surveillance led to 0.75 invasive diagnostics per patient post-OHT year. Conclusions: Sm remained stable in the post-OHT course. Reductions indicated ISHLT1R, AMR1 and CAV3 and were associated with higher all-cause mortality. Sm-triggered surveillance may be referred to as a safe, high-yield screening modality in OHT aftercare.


Introduction
Acute rejection (AR) and cardiac allograft vasculopathy (CAV) are leading causes of mortality after orthotopic heart transplantation (OHT) [1]. Approximately one-quarter of patients experience at least one AR within the first year; in one-third of patients, CAV occurs within 5 years after OHT [2,3]. The early diagnosis and treatment of these complications are crucial in the followup care of heart transplant recipients. Since the symptoms and clinical signs of AR and CAV are often unspecific and typically occur late, routine surveillance by endomyocardial biopsies (EMB) and angiographies are recommended every six to twelve months if the renal function is not severely impaired [4]. Procedures are accompanied by a risk for acute complications and for the development of tricuspid valve regurgitation after repetitive catheter passages [5]. Furthermore, invasive follow-ups consume hospital resources and are of limited acceptance by the patients, leading caregivers to keep the frequency of invasive diagnostics as low as possible. However, this results in a period of uncertainty between examinations, which might be particularly challenging in patients with persisting donorspecific human leukocyte antigene (HLA) antibodies or false negative EMB. Furthermore, routine surveillances of asymptomatic patients may reveal clinically insignificant, lowgrade rejections, and lead to immunosuppressant overtreatment with a concomitant risk for infections and malignancies [6].
Despite advances in noninvasive cardiac imaging, a reliable, easily accessible screening modality for follow-up care after OHT has not yet been widely introduced in clinical practice. In our institution, echocardiographic AR and CAV surveillance via pulsed-wave tissue Doppler imaging (PW-TDI) has been routinely performed and used for navigation of invasive diagnostics for more than 15 years.
The aim of this retrospective analysis was to investigate the long-term course, diagnostic reliability, and prognostic value of systolic wall motion analysis after OHT.

Materials and Methods
The files of 438 patients who had undergone heart transplantation at our institution between 2006 and 2020 were retrospectively analyzed. Due to technical restrictions of the image data storage before 2010, only patients who had received a transplantation between 2010 and 2020 were further analyzed for eligibility. Recipients aged under 18 years at OHT and patients who had died within the transplant hospital stay were excluded from final analysis ( Figure 1). AR is referred to as T-cell-mediated acute cellular rejection (ACR), antibody mediated rejection (AMR), or both. AR grading was defined according to the classification of The International Society for Heart and Lung Transplantation (ISHLT) [7,8]. CAV affects both epicardial and intramural coronary arteries, and is characterized by diffuse intimal proliferation with successive lumen narrowing and microvascular dysfunction [9,10]. The grading was defined by the degree of stenosis, number and localization of lesions, and graft function, as described by the ISHLT [11]. The baseline characteristics of 166 donors and recipients, including age, sex, body mass index, cause of heart failure, concomitant diseases; intra-and postoperative data, including ischemic time, immunosuppressive treatment, and intensive care unit stay; and Life 2021, 11, 1206 3 of 11 echocardiographic follow-up data from more than 1100 visits were collected in an electronic database. The study conforms to the principles outlined in the Declaration of Helsinki and was approved by the local ethics committee (EA2/169/19).
AR is referred to as T-cell-mediated acute cellular rejection (ACR), antibody mediated rejection (AMR), or both. AR grading was defined according to the classification of The International Society for Heart and Lung Transplantation (ISHLT) [7,8]. CAV affects both epicardial and intramural coronary arteries, and is characterized by diffuse intimal proliferation with successive lumen narrowing and microvascular dysfunction [9,10]. The grading was defined by the degree of stenosis, number and localization of lesions, and graft function, as described by the ISHLT [11].
Systolic wall motion peak velocities (Sm) were measured by PW-TDI at the subendocardial, posterobasal segment of the left ventricle at the level of the mitral leaflet tips. Radial wall motion was measured in parasternal long axis view and longitudinal wall motion in apical 3-chamber view [12]. In routine echocardiography, measurements of the Sm during or after extrasystoles are excluded from analysis. Since transducer angulation might result in a detection of lower velocities, the highest systolic wall motion velocity from at least three beats was selected in clinical routine ( Figure 2). AR is referred to as T-cell-mediated acute cellular rejection (ACR), antibody mediated rejection (AMR), or both. AR grading was defined according to the classification of The International Society for Heart and Lung Transplantation (ISHLT) [7,8]. CAV affects both epicardial and intramural coronary arteries, and is characterized by diffuse intimal proliferation with successive lumen narrowing and microvascular dysfunction [9,10]. The grading was defined by the degree of stenosis, number and localization of lesions, and graft function, as described by the ISHLT [11].
Systolic wall motion peak velocities (Sm) were measured by PW-TDI at the subendocardial, posterobasal segment of the left ventricle at the level of the mitral leaflet tips. Radial wall motion was measured in parasternal long axis view and longitudinal wall motion in apical 3-chamber view [12]. In routine echocardiography, measurements of the Sm during or after extrasystoles are excluded from analysis. Since transducer angulation might result in a detection of lower velocities, the highest systolic wall motion velocity from at least three beats was selected in clinical routine ( Figure 2). Clinical, electrocardiographic, and echocardiographic signs of graft failure were routinely evaluated every three to six months in asymptomatic patients in OHT aftercare in our institution. If Sm values decreased by >10% compared to individual previous values, invasive diagnostics, including coronary angiography and biopsies, were performed. Clinical, electrocardiographic, and echocardiographic signs of graft failure were routinely evaluated every three to six months in asymptomatic patients in OHT aftercare in our institution. If Sm values decreased by >10% compared to individual previous values, invasive diagnostics, including coronary angiography and biopsies, were performed.
Continuous data are summarized as mean and standard deviation (SD) or, in the case of skewed data, as median and interquartile range (IQR). Comparisons between groups were analyzed using t-tests for paired samples. Due to the small samples, t-tests were not calculated for the following groups: Sm reduction in ACR ISHLT 2R and 3Rm and ISHLT CAV1 and CAV2. Frequencies and percentages are reported for categorical data. The association of echocardiographic parameters with all-cause mortality was tested in a time-varying covariate Cox regression model. For patients who survived the first year after transplantation, conditional survival rates were analyzed using the Kaplan-Meier method. SPSS 25 and R 4.0 were used for statistical analysis.

Patient Characteristics
Between 2010 and 2020, 166 adult patients received OHT and survived the transplant hospital stay. The mean age of recipients was 46.2 (±11.4) years at OHT; donors were aged 40.8 (±14.1) years. Overall, 127 (76.5%) recipients and 107 (64.5%) donors were male. Most patients (88.0%) underwent heart transplantation in a highly urgent status on the waiting list of Eurotransplant with terminal heart failure, mainly due to dilated cardiomyopathy (68.7%) or ischemic heart disease (21.1%). Half of the patients (50.6%) were on long-term

Mortality, AR and CAV
Of all patients, 13.3% (n = 22) died in a mean of 4.1 (±2.4) years after OHT. Figure 4 presents the survival rates of patients who survived the first post-transplant year. Causes of death were graft failure due to AR and CAV in 6 (27.3%) and 4 (18.2%) patients, respectively. Five (22.7%) patients died due to an infection (necrotizing pancreatitis, mediastinitis, intracranial abscess, sepsis of other origins) and 2 (9.1%) patients due to lung cancer. In 4 (18.2%) patients, the cause of death was unknown.

Mortality, AR and CAV
Of all patients, 13.3% (n = 22) died in a mean of 4.1 (±2.4) years after OHT. Figure 4 presents the survival rates of patients who survived the first post-transplant year. Causes of death were graft failure due to AR and CAV in 6 (27.3%) and 4 (18.2%) patients, respectively. Five (22.7%) patients died due to an infection (necrotizing pancreatitis, mediastinitis, intracranial abscess, sepsis of other origins) and 2 (9.1%) patients due to lung cancer. In 4 (18.2%) patients, the cause of death was unknown.

Sm Course after OHT
Individual Sm values remained stable from discharge after OHT over the follow-up period. Radial and longitudinal Sm were 11.5 ± 2.2 and 10.9 ± 2.1 cm/s, respectively ( Figure 5).

Sm Course after OHT
Individual Sm values remained stable from discharge after OHT over the follow-up period. Radial and longitudinal Sm were 11.5 ± 2.2 and 10.9 ± 2.1 cm/s, respectively ( Figure  5).

Prognostic Value of Sm
In a time-varying covariate Cox regression model, lower radial or longitudinal Sm values at any time were associated with higher all-cause mortality per cm/s decrease (hazard ratio (HR) 0.77, CI 0.63-0.93, p = 0.008; HR 0.7, CI 0.58-0.95, p = 0.016; HR, respectively), as presented in Table 4. A radial Sm of less than 9 cm/s and a longitudinal Sm of less than 10 cm/s were associated with an approximately threefold increased risk of mortality (HR 3.24, CI 1.2-8.76, p = 0.020; HR 2.92, CI 1.19-7.18, p = 0.019; HR, respectively).

Discussion
Our analysis shows that the PW-TDI-derived systolic wall motion of the posterobasal segment of the left ventricle remains stable for at least 10 years after OHT. A decrease in Sm may indicate AR or CAV and is a predictor of mortality.
After Dandel et al. described the high diagnostic value of Sm measurements for predicting ACR and CAV, and its usefulness for the scheduling of invasive diagnostics in a prospective clinical trial in 2001, our institutional routine post-transplant care was adapted with a shift from routine surveillance to a clinical-echocardiographic navigation of EMB and angiography [12][13][14]. The present work represents a long-term follow-up of our experiences following this novel approach. In our recent analysis, low-grade ACR and high-grade CAV were accompanied by a significant reduction in Sm. This complements previously published work, where medium-and high-grade ACR (ISHLT ≥ 2R), and lowgrade or angiographically invisible CAV were reliably indicated by a Sm decrease [12][13][14]. Additionally, we first describe here a significant Sm reduction in patients with AR caused by AMR (which was defined by the ISHLT in 2005) [7].
Moreover, in our analysis, we observed a stable long-term Sm course, which supports the usefulness of Sm surveillance over the post-OHT period. A one-time detection of radial or longitudinal Sm of <9 cm/s or < 10 cm/s, respectively, was associated with a threefold increased risk of all-cause mortality, underlining the prognostic value of the Sm measurements.
Our patients received 0.75 EMBs and/or angiographies per patient post OHT year. Patients who had been transplanted for at least 5 years (n = 77, mean 8.4 ±1.7 years post OHT) had a median of 3 (IQR 2-6) invasive diagnostic procedures during follow-up, which is an extremely low frequency compared to the guideline recommendations or to symptomtriggered biopsy protocols reported in the literature [4,[15][16][17]. In 170 (35.4%) invasive procedures, EMBs or angiographies confirmed suspected AR or CAV, so that Sm-navigated invasive diagnostics can be referred to as a high-yield screening modality for post-OHT AR or CAV [18].
Our institutional conditional survival rates were comparable to the rates reported for de novo or post long-term circulatory support transplant recipients in large American databases or European transplant centers within the Eurotransplant or Scandiatransplant region [19][20][21][22].
LV-EF and strain analysis, as other echocardiographic parameters for LV myocardial function, have been described as predictors of graft failure after OHT. LV-EF decreases in advanced AR. But several pitfalls limit the accuracy of LV-EF measurements: (i) the endocardium border must be clearly visible in every segment of the apical 4-and 2-chamber view, (ii) apical foreshortening due to transducer angulation or atypical views leads to measurement errors, and (iii) the eye-balling technique for LV-EF estimation in cases of insufficient imaging quality is highly interobserver-dependent [23]. Due to these limitations, small changes in LV-EF, as reported for our patients with ISHLT1R AR, are not reliably detectable. Speckle-tracking-based global longitudinal strain (GLS) analyses and segmental longitudinal strain reliably indicate AR and CAV in OHT cohorts. Compared to TDI, this method is less angle-and interobserver-dependent. However, a major disadvantage is that high imaging quality is required; therefore, GLS is only applicable in selected patients [24][25][26].
Overall, we postulate clinical-Sm-triggered OHT follow-up care as a safe, noninvasive, reliable, and potentially cost-effective strategy for heart transplant recipients.

Limitations
This analysis has certain limitations: The investigation was performed retrospectively in a single center and the samples of high-grade ACR and low-grade CAV were too small to perform statistical testing. Furthermore, we examined the role of echocardiographic surveillance in detecting histological rejection; the occurrence of symptoms and the need for treatment were not taken into account.