Abstract
Heart transplantation results in complete denervation of the donor heart with loss of afferent and efferent nerve connections. The majority of patients remain completely denervated during the first 6–12 months following transplantation. Evidence of reinnervation is usually found during the second year after transplantation and involve the myocardial muscle, sinoatrial node, and coronary vessels, but remains incomplete and regionally limited many years post-transplant. Restoration of cardiac innervation can improve exercise capacity as well as blood flow regulation in the coronary arteries, and hence improve quality of life. As yet, there is no evidence that the reinnervation process is associated with the occurrence of allograft-related events or survival.
Cardiac denervation during heart transplantation
The intact heart is innervated by the parasympathetic and sympathetic fibres of the autonomic nervous system. Cardiac transplantation results in transection of the post-ganglionic neural axons innervating the heart. Axonal degeneration develops within days after transplantation and leads to total depletion of cardiac norepinephrine stores and eventual disappearance of nerve terminals in the transplanted tissue, resulting in total cardiac denervation1 (Figure 1).
Cardiac denervation post-heart transplantation. Cross-sectional positron emission tomography images from a patient with recent cardiac transplantation: rubidium-82 (Rb-82) presents homogeneous blood flow throughout the left ventricle, whereas the 11C-hydroxyephedrine images below indicate a marked reduction of tracer retention that is homogeneous throughout the left ventricle. Reproduced with permission from Schwaiger et al.18
Loss of afferent innervation alters cardiovascular homeostasis by impairing the normal vasoregulatory response to changing cardiac filling pressures. Efferent denervation results in the loss of sympathetic and parasympathetic regulation, leading to an increase in resting heart rate and blunting the rapid changes in heart rate and contractility during exercise.2–7
Methods to evaluate post-transplant cardiac reinnervation
Various approaches have been applied to demonstrate subsequent restoration of cardiac catecholamine uptake and storage sites after heart transplantation (HT) in humans.
The evaluation of the sympathetic nervous system of the heart has been limited in the past to invasive procedures to determine catecholamine concentrations. This involved induction of myocardial sympathetic nerve fibre activation with either intra-coronary injection of tyramine (an agent that causes norepinephrine release from intact sympathetic nerve terminals) or with exercise such as handgrip maneuver, with subsequent evaluation of the response by invasive measurement of cardiac spillover of norepinephrine, left ventricular pressure, and coronary blood flow velocity.8–11
More recently, non-invasive methods to delineate sympathetic nerve terminals of the heart and assess cardiac innervation have become possible with the introduction of radiolabeled catecholamines analogues. One option is to use the guanethidine analogue,123I-meta-iodo-benzylguanidine (MIBG), which is taken up by pre-synaptic nerve terminal norepinephrine transporters and retained in norepinephrine storage vesicles. As such, radionuclide images of 123I-MIBG have been employed to demonstrate extent and location of sympathetic nerve reinnervation of the transplanted heart, while washout rates of the tracer may provide an indication for sympathetic nerve function.12,13 Positron emission tomography (PET) imaging with catecholamine analogue 11C-hydroxyephedrine (11C-HED) also has been used to trace and quantitative uptake and storage of norepinephrine in the pre-synaptic adrenergic nerve terminals.14,-–18 A PET tracer capable of imaging cardiac beta adrenergic receptor density also is available, 11C-CGP-12177—which is a beta adrenergic receptor antagonist, and may prove useful in better understanding the physiology of the denervated transplanted heart and associated changes as sympathetic reinnervation develops.19–21 Similarly, PET imaging of parasympathetic (muscarinic) nerve function also has been reported and may prove useful in evaluating extent and time course of parasympathetic reinnervation of the transplanted heart.22,23
Sympathetic reinnervation of the sinus node has been evaluated by the increase in heart rate as a response to maneuvers that increase adrenergic stimulation, like tyramine injection into the coronary artery that perfuses the sinus node or after positional change.2,9,10,17
Initiation of cardiac reinnervation post-transplant
While sympathetic reinnervation of transplanted hearts was documented in animal models in the early 1970s,1,24 data on human hearts became available only after the developments of techniques for accurate in vivo evaluation of the autonomic nervous system.
Most studies have confirmed complete denervation within the first 6–12 months post-transplant (Figure 1). Once initiated, sympathetic reinnervation is progressive, and increases even late after transplantation, up to 15 years after surgery. However, the reinnervation process remains incomplete and regionally limited8–18,25,26 (Figure 2).
Initiation of cardiac reinnervation post-transplant. Short-axis positron emission tomography images obtained in a patient 55 months after cardiac transplantation. Rb-82 images show homogeneous blood flow throughout the left ventricle. 11C-hydroxyephedrine images below indicate tracer retention in the anterior aspects of the left ventricle, increasing from the mid-left ventricle toward the base of the left ventricle (bottom row). Reproduced with permission from Schwaiger et al.14
A retrospective analysis among 77 HT recipients was performed to identify the influence of wide variety of clinical parameters on extent and intensity of sympathetic reinnervation.27 Several parameters were found to be significant with univariate analysis: time post-transplant, recipient and donor’s age at transplant, idiopathic cardiomyopathy, aortic cross clamp time, perioperative aortic complications, and frequency of rejection episodes. With multivariate analysis to identify independent determinants of reinnervation, only donor age, aortic cross clamp time, and frequency of rejection were found to be independently associated.27
Reinnervation process
Myocardial muscle
Most studies that evaluated myocardial adrenergic reinnervation post-HT confirmed a progressive process with a definitive pattern which usually starts during the second year post-transplantation8–18,25,26 (Figure 2).
The first evidence of myocardial sympathetic reinnervation is found in basal parts of the anterior wall and subsequently progresses to more distal parts of the myocardium. In addition to the gradient from base to apex, anterior and septal wall were reinnervated earlier, whereas the lateral wall appears to be involved later. These results suggest that sympathetic nerves are first restored in the territory of the left anterior descending (LAD) artery followed later by the left circumflex territory (Figure 3A).
(A and B) Regional reinnervation according to coronary artery territories. (A) Polar maps showing the level of 11C-hydroxyephedrine retention in one patient with denervation and one with reinnervation post-transplant. The patient with denervation has low levels of 11C-hydroxyephedrine retention in all regions (dark red), whereas the patient with reinnervation has a high level of 11C-hydroxyephedrine retention (yellow) in the basal anteroseptal wall, indicating partial restoration of catecholamine uptake sites. LAD, left anterior descending artery; LCX, left circumflex artery; RCA, right coronary artery. (B) The myocardial sympathetic reinnervation process assessed by the level of 11C-hydroxyephedrine retention (per cent per minute) suggests that sympathetic nerves are restored differently according to the coronary artery territory. Reproduced with permission from Bengel et al.25
In contrast, the territory of the right coronary artery, demonstrates little or no reinnervation until late after transplant at which time there is only modest, at most, evidence of sympathetic nerve terminals at the apex and inferior wall of the left ventricle10,13,15,17,25 (Figure 3B).
Though myocardial sympathetic reinnervation, particularly of the anterior and to a lesser extent lateral left ventricular wall may be substantial over time, parasympathetic nervous system reinnervates involves mainly the atria and to a much lesser extent the left ventricle.28 Indeed, the extent of myocardial parasympathetic reinnervation post-HT remains controversial.8,25,29–31
Myocardial blood flow
Myocardial blood flow post-HT can be quantified by PET imaging with 13N-ammonia or O.21,32 A number of reflexes participate in the control of coronary vascular resistance through activation of the sympathetic or parasympathetic nervous system.16,33–36
Myocardial blood flow regulation varies as a function of time after transplantation.
During the first year after HT basal myocardial blood flow at rest is greater compared with healthy individuals (1.86 ± 1.01 mL/min/g vs. 1.01 ± 0.20 mL/min/g; mean ± SD) with gradual decline starting in the second year post surgery to levels similar to that of healthy controls (1.01 ± 0.20) (1.17 ± 0.73 1–3 years post-transplant and 0.98 ± 0.34 for recipients >3 years post-transplant;37,–39). These differences in basal myocardial blood flow cannot be explained based on differences in heart rate or rate-pressure product at rest, as these parameters were similar among all transplant recipients regardless of the time after transplantation.37
Myocardial blood flow increase in response to adenosine was comparable amongst recipients studied between 1 and 36 months post-transplant and did not differ significantly from that of normal controls37 (Figure 4). In contrast, for recipients over 3 years post-transplant the average increase in myocardial blood flow with adenosine was less, and coronary vascular resistance, which is an index of maximal dilator capacity of the coronary microvasculature, increased compared with that of early transplant patients and healthy controls.37 This finding suggests an immune-mediated mechanism may cause endothelial dysfunction and impairment of microvascular dilator function as time post-transplant increases beyond 3 years (Figure 4).
Myocardial blood flow post-cardiac transplantation. Myocardial blood flow at rest and with adenosine is shown for different patient groups according to the interval after heart transplantation, as well as for a healthy control group. Values are expressed as mean; error bars represent SD. *P < 0.02 for the 1- to 12-month group vs. all other groups. †P < 0.05 for the ≥37-month group vs. the 1- to 12-month and 13- to 34-month groups, and P = 0.06 vs. controls. Reproduced with permission from Kushwaha et al.27
Myofibrillar lysis correlated strongly with reduced maximal dilator capacity in response to adenosine (<3 mL/min/g) across all transplant recipients regardless of time interval from transplantation. It appears likely that an immune mediated mechanism causes both the myofibrillar injury and the impairment of microvascular dilator function37 (Figure 5).
Correlation between myofibrillar lysis and myocardial blood flow post-transplant. Individual data values of maximal myocardial blood flow with adenosine for patients after heart transplantation according to the presence (POS) or absence (NEG) of myofibrillar lysis on biopsy . Reproduced with permission from Kushwaha et al.27
Baseline blood flow in transplant recipients are similar in all the coronary territories despite differences in sympathetic reinnervation, suggesting that resting coronary flow is not substantially affected by adrenergic influences.16 In response to sympathetic nerve stimulation (by a cold pressor test) blood flow increased significantly in all coronary territories, but the magnitude of the increase in flow was higher in the territory of the LAD artery, which is the territory with maximal sympathetic reinnervation.16 This finding suggests that restoration of sympathetic reinnervation can improve blood flow regulation, and that the increase in coronary flow in response to sympathetic stimulation correlates with the magnitude of regional stores of norepinephrine in cardiac sympathetic nerve terminals.16 Several potential mechanisms could explain coronary vasodilatation after sympathetic nerve terminals activation. One of them is that release of norepinephrine results in direct activation of beta-2 adrenergic receptors on smooth muscles and both beta 1 (NO linked33) and beta 2 receptors on endothelial cells33,36 in the vessels wall in addition to increasing myocardial contractility (primarily beta 1 effect) and hence myocardial oxygen demand. The later mechanism likely is the more important one in the human heart in which coronary tone is primarily (though not exclusively) regulated in response to metabolic demand. Another potential mechanism is more complex. Thus, alpha-1 and alpha-2 adrenergic stimulation generally results in constriction of coronary microvessels (alpha-1 of those with diameter of 100–300 µ; alpha-2 of arterioles with diameter of <100 µ).34 However, it also has been shown that alpha-2 adrenergic stimulation may enhance endothelial cell nitric oxide release by a kinin synthesis related mechanism; since the response may be blocked either by eNOS inhibition directly (l-NAME) or by a serine protease inhibitor which blocks production of kallikrein.40 Accordingly, depending on the status of endothelial cell function norepinephrine stimulation of alpha-1 and alpha-2 adrenergic receptors may result either in net coronary microvascular constriction (e.g. impaired endothelial cell function >3 years post-transplant37) or dilation (e.g. intact endothelial cell function 1–3 years post-transplant37). Future studies are clearly warranted to investigate in more detail both the time course of these responses post-transplant and the extent to which adrenergic receptor sub type density and interactions with endothelial cell function play a role in modulating overall responses to adrenergic stimulation in the transplanted heart.
Activation of vagal efferent fibres to the normal human heart results in vasodilation due to the release of acetylcholine and activation of muscarinic receptors.35 Further, it has been found that activation of vagal reflexes, which usually result in cholinergic coronary vasodilation, is blocked by inhibition of nitric oxide synthesis. Moreover, brief exercise training up-regulates nitric oxide production by coronary blood vessels and so potentiates cholinergic coronary vasodilation in the normal human heart.35 There is little information concerning the extent to which these findings apply to the post-transplant heart (and when) and represents an important area for future research. The ability of PET to make quantitative measurements of myocardial blood flow and function as well as autonomic receptor density and distribution make it a very attractive modality for conducting such studies.
Sino-atrial node
Transection of the autonomic nerve fibres during HT results in parasympathetic denervation and loss of the suppression of the sino-atrial (SA) node, leading to a persistent increase in resting heart rate. The sympathetic denervation contributes to a delay in exercise or stress-induced augmentation of SA node automaticity, resulting in diminished maximal heart rate response with exercise, which is primarily dependent on increase in plasma catecholamine concentration.2–7,9,26
These patients usually present an elevated resting heart rate and a minimal increase in heart rate during the first few minutes of exercise. The maximal heart rate is reached in the recovery period rather than at peak exercise, returning slowly to the resting values. This may reflect the delayed humoral catecholamine release. In recovery heart rate falls slowly as plasma catecholamines are metabolized.
Sympathetic reinnervation of the SA node has been shown by measurement of the change in heart rate after exercise or after injection of tyramine into the coronary artery that perfuses the sinus node. This process is also progressive and directly correlated to the period of time since transplant surgery. Tyramine injection had a significant effect in recipients more than 1 year post-cardiac transplant as opposed to no change in patients within 4 months of transplant.2,9,10,17
The major clinical consequence of sinus node sympathetic reinnervation after HT is partial restoration of a normal heart rate response to exercise which corresponded to the extent of reinnervation as detected by heart rate response to intracoronary tyramine injection.2,29
There was no direct correlation between sympathetic reinnervation of the sinus node and of the left ventricle after HT, demonstrating the regionally heterogeneity of sympathetic reinnervation as some patients had isolated left ventricular without SA reinnervation and vice versa.10,13
Parasympathetic reinnervation of the SA node was evaluated by respiratory sinus arrhythmia, which is mainly caused by vagal innervation of the sinus node. Due to parasympathetic denervation post-HT there is fixed heart rate without heart rate variation within at least 12 months post-transplant. An increase in heart rate variability post-transplant suggest vagal reinnervation,29–31 although other studies failed to show any evidence of parasympathetic reinnervation in the majority of patients up to 8 years post-transplant.25 Additional studies with a PET muscarinic receptor tracer may be helpful in shedding new light on the issue.
Effect of reinnervation on exercise capacity
Exercise capacity and the maximal heart rate response to exercise are lower in HT recipients than in normal controls due to the absence of atrial and SA node innervation. Most of the increase in cardiac output during physical activity in transplant recipients is achieved through an increase of stroke volume, which physiologically is quite limited in untrained subjects. Typically, the heart rate continues to rise after cessation of exercise as a result of delayed humoral catecholamine release and then return slowly to the resting values.2–7
Transplant recipients with evidence of restoration of sympathetic innervation had better exercise performance compare to denervated recipients due to better chronotropic and inotropic response.25 Overall exercise time was significantly longer in reinnervated patients with significantly greater increase of heart rate above baseline, and peak heart rate attained during exercise compared with denervated patients.16,25,26
Multiple studies have demonstrated the benefit of exercise training after HT by improving peak oxygen uptake, peak heart rate, and chronotropic response, and high-intensity, interval based aerobic exercise has been documented to have superior positive effect compared with moderate exercise.41–43 The fact that the improvement in exercise capacity is lost after a few months without training, may suggest that the physiological mechanisms for improvement are primarily peripheral and not through cardiac remodeling.41
There is also evidence that the reinnervation process was significantly correlated with an increase in ejection fraction (EF) during exercise.25,26 There was no change in EF at rest among transplant patients and control group, but during peak exercise the global EF was significantly lower among denervated recipients compared with reinnervated mainly due to improvement in the anteroseptal regional EF in these groups. This regional change in EF corresponds to the area of the LAD artery where reinnervation process begins. The inferoapical region remains denervated in most transplant recipients, and its regional peak EF was lower in all transplant patients compared with the control group16,25,26 (Figure 6).
Correlation between reinnervation process and increase in ejection fraction during exercise. T bars indicate standard deviations. Reproduced with permission from Bengel et al.25
Effect of cardiac reinnervation on allograft vasculopathy
After HT, cardiac allograft vasculopathy (CAV) is the major limitation to recipient long-term survival.44,45 Cardiac allograft vasculopathy results from an initial injury to the allograft endothelium causing chronic inflammatory state, and develops in as many as 50% of HT patients within 5 years post-transplantation.46 Early diagnosis of CAV in humans is limited by the lack of clinical symptoms for ischemia in the denervated allograft, and surveillance angiography is performed annually to detect coronary involvement early.
There is almost no data on the relation between the extent of the reinnervation process and the development of CAV. It is possible that in many transplant recipients, the development of CAV is occurring simultaneously with the reinnervation process and that the release of NO may be affected, not only by adrenergic influences (as discussed above in the section on myocardial blood flow) but also by impairment of endothelial cell function.47 Equally important may be the effect of CAV on the reinnervation process. In that regard, Estorch et al. evaluated the extent of reinnervation and vasculopathy in 31 HT recipients.48 They reported that patients with established angiographic CAV had less reinnervation than those without CAV as assessed by 123I-MIBG imaging.
The improvement in imaging modalities to detect early stages of CAV and to evaluate reinnervation post-transplant may facilitate additional studies to explore this interaction further.
Effect of beta adrenergic blockade in cardiac transplant receipts
Several studies have shown detrimental effects of beta-blockade on exercise capacity post-HT.
Beta blockade reversed the significant differences in exercise capacity between reinnervated and denervated recipients due to attenuated increase in heart rate, peak heart rate, and EF during exercise among reinnervated recipients.26,49
The fact that attenuating effects of beta blockade were more pronounced in reinnervated then in denervated patients suggest that beneficial effects of reinnervation and restoring sympathetic nerve terminals, are mediated via beta-adreno receptors.26 As noted above, this is an excellent area for further research with PET and beta adrenergic receptor tracer. Thus, in patients with ischemic cardiomyopathy preliminary data indicate that ‘mismatch’ between beta receptor density (11C-CGP-12177) and norepinephrine transport function (11C-HED) may portend a worse prognosis due to propensity to sudden cardiac death.50 The extent to which myocardial beta receptor density and norepinephrine transport function are matched has not been studied in the transplanted human heart and would provide important information not only on left ventricular function but also risk for ventricular arrhythmia particularly in both the early and very late time periods following transplantation.
Effect of reinnervation on outcome
The reinnervation process improves exercise capacity in HT recipients and provides them better quality of life. Nevertheless, survival analysis did not reveal a significant association between reinnervation process and occurrence of allograft-related events or survival.27
Summary
Heart transplantation results in complete denervation of the donor heart with loss of afferent and efferent nerve connections and the loss of sympathetic and parasympathetic regulation. The development of non-invasive methods using radiolabeled catecholamines analogues (123I-MIBG and 11C-HED) enables us to better understand the cardiac reinnervation process. The majority of patients remain completely denervated during the first 6–12 months following transplantation, while the myocardial reinnervation process is usually found during the second year post-transplant and involves the myocardial muscle, SA node, and coronary vessels. This process appears to be progressive, but remains incomplete and regionally limited many years after transplantation (Figure 7).
From denervation to reinnervation post-cardiac transplantation.
As yet, there is little and controversial data regarding parasympathetic reinnervation process, as well as the adrenergic receptor sub type density. The ongoing developments in imaging modalities facilitate additional studies on the interaction between reinnervation process and endothelial cell function, and its role in modulating blood flow responses and development of CAV, which is a major factor effecting survival post-HT.
Conflict of interest: none declared.
