Echocardiography is among the non-invasive techniques for prediction of heart transplant rejection [12, 15]. Reduction in LV mass and thickness as well as reduced systolic function and pericardial effusion are significant echocardiographic findings in heart transplant rejection, but these findings are not predictors due to a late unset [16]. In contrast to systolic dysfunction, diastolic dysfunction appears earlier in heart transplant rejection. However, because of the difference in the definition of diastolic function parameters, the use of these parameters have shown different accuracy in diagnosis of heart transplant rejection [17, 18]. Furthermore, diastolic function might be affected by conditions including hypertension and allograft vasculopathy, which make the interpretation of the findings difficult [12, 16, 19]. Among the new measurements, the strain assessment by TDI has shown promising findings in early detection of heart transplant rejection [20–23].
In this study heart transplant rejection was observed in 32.8% of patients. This is in line with the findings of a previous study in Iran, which reported a 34.4% rejection rate in heart transplant patients [8]. In contrast, the rate of rejection was lower in the reports of Europe (25%) and the United Stated (7.8%) [9, 24]. One reason for the observed difference might be due to the younger age of patients in Iran compared to developed countries, as previously demonstrated that younger age was associated with higher risk for transplant rejection [18]. Another reason might be related to the differences in sample size and the number of EMBs performed based on different protocols.
This study demonstrated that E was the only significant factor that was higher in non-rejection group compared with the rejection group, the finding that showed in previous study, as well [18]. Unalike the findings of our study, Sun et al. revealed a significant difference in IVRT between rejection and non-rejection groups, but they also proved that IVRT might not be a good predictor for rejection based on the small sensitivity (53%). They reported a restrictive mitral inflow pattern among patients with rejection [18]. This is in contrast with the study by Haghighi that reported no significant difference in TDI parameters between rejection and non-rejection groups [14]. It was stated that a restrictive pattern might be a specific finding in heart transplant rejection but it might not have adequate sensitivity [14]. In another study, an association was found between posterior wall thickness and ACR of any grade in univariate analysis, but did not remain after multivariate analysis. They also found some associations of ACR with mitral inflow velocities determined by conventional pulsed Doppler, only in univariate analysis, that disappeared after adjusting by other parameters [25].
Another possible reason for such differences might be due to the method used for heart transplant. It was previously shown that heart transplantation using bicaval method might result in improved atrial function, while older techniques might lead to varying A velocity and thus increasing the E/A ratio which can resemble a restrictive pattern even in patients with no rejection [26]. However, all studies did not report any significant differences regarding E/A ratio between rejection and non-rejection groups, which is in line with the findings of the current study [12, 27].
Dandel et al. found that peak systolic and early diastolic peak wall motion velocity obtained at the basal LV posterior wall were significantly associated with rejection [17]. Puleo et al. reported similar results only for diastolic velocities [28]. Mankad et al. described that a peak-to-peak mitral annular velocity (the same s′ + e′ parameter) > 13.5 mm/s, determined by color-coded tissue Doppler, had 93% sensitivity, 71% specificity, and 98% negative predictive value for detecting rejection, defined as grade IB or above of the former classification of the International Society of Heart and Lung Transplantation (ISHLT) [29]. Logistic regression also revealed that E was significantly related to rejection; one-unit increase in E was associated with 78.07 times increased risk for developing heart transplant rejection. Although this was significant, the wide CI for OR indicates that E may not be a sensitive predictor for heart transplant rejection. One reason for the various observations regarding the association of E or E/A velocity and the risk of heart transplant rejection might be due to the patchy nature of cellular rejection which could result in different effects on flow and wall motion. It was also shown that E/A ratio remained unchanged after heart transplantation regardless of the normalization of diastolic function, which might also interfere with the interpretation of the observed patterns in heart transplant rejection [13]. In another study on heart transplant patients (723 echocardiographic assessments), significant variations were observed in DT, EF, E and E/A [30]. In the current study, the mean E velocity was lower compared to other studies that might have resulted in lower E/A values leading to a non-significant reduction in E/A ratio in the rejection group.
The findings of the current study revealed that septal Ts was significantly higher in non-rejection group compared to the rejection one. The relation between septal Ts and rejection was also documented in the current study after performing the logistic regression analysis. The regression analysis revealed that higher septal Ts was associated with 93.1% reduced risk for heart transplant rejection. It was previously reported that heart transplantation may result in abnormal septal motion due to alterations in right ventricular pressure and conduction abnormalities [31]. This finding may either strengthen or weaken the diagnostic properties of septal Ts in detection of heart transplant rejection.
The ROC curve revealed that septal Ts could better detect rejection compared to E. The sensitivity and specificity for septal Ts for detection of rejection were 75% and 69% respectively, while the sensitivity and specificity for E velocity in detection of rejection were 65% and 61% respectively. Combination of septal Ts and E velocity increased the sensitivity to 90%, but reduced the specificity to 42.1%. Therefore, assessment of E velocity may be used in suspicious cases for rejection to double confirm the prediction but may not be useful in ruling out rejection in low suspicion cases based on septal Ts.
One of the limitations of this study was the sample size which prevented us from performing subgroup analyses. Although the primary objective of the study was met, further studies are required to assess the effect of the type of transplant surgery technique and different grades of rejection on the sensitivity and specificity of the echocardiographic parameters. There is also a need for inclusion of older patients as the observed rejection pattern might be different in the elderly.