Imaging of cardiac sympathetic activity in heart failure: Not out of the woods yet

Currently, 5.7 million people are living with heart failure (HF) in the U.S., and given the aging of the population, the prevalence of HF continues to grow.1 In addition, HF is the leading cause of hospitalizations in people over the age of 65, thereby posing an increasing problem for global healthcare systems.2 In view of this burden, risk-based, personalized therapeutic strategies and a cost-effective management for patients with HF are needed. However, despite notable improvements in medical therapy, outcomes for HF patients are still unacceptable and precise risk stratification in HF remains challenging.2 In fact, various prognostic markers of mortality and morbidity of HF have been identified; however, their clinical applicability is limited. Similarly, tools contributing to the appropriate selection of therapeutic strategies or monitoring their efficacy are lacking. There is accumulating evidence that imaging of cardiac sympathetic activity by ¹²³I-metaiodobenzylguanidine (mIBG), a sympathetic neurotransmitter radionuclide analog (Figure 1), may help to direct therapy and clinical decision-making in HF patients. Indeed, given that dysfunction of the sympathetic nervous system is one of the key features of worsening systolic HF, pharmacological and non-pharmacological targeting of sympathetic dysfunction has emerged as a promising treatment strategy in HF.2 Accordingly, ¹²³I-mIBG imaging has been shown to have an incremental prognostic value in predicting disease progression, arrhythmic events, and cardiac death in HF patients beyond that provided by traditional functional and neurohormonal markers.3 However, despite its prognostic importance and its FDA approval for cardiac applications in 2013, the exact role of cardiac neuronal imaging in diagnosis and management of HF is still under debate, and therefore, ¹²³I-mIBG scintigraphy is currently mostly applied as a research tool.

Figure 1

Schematic description of norepinephrine and ¹²³I-Metaiodobenzylguanidine (mIBG) kinetics in sympathetic nerve ending. Norepinephrine is synthesized in presynaptic neurons from tyrosine by dopamine β-monooxygenase (DMO) via L-DOPA (tyrosine hydroxylase, TH) and dopamine (DOPA decarboxylase, DCB), then transported into presynaptic neuronal terminal vesicles via vesicular monoamine transporter (VMAT), and finally released into the synaptic cleft via exocytosis/diffusion. Only a small percentage of norepinephrine binds to postsynaptic α1-, β1-, and β2-adrenoceptors for downstream effects, while the majority of norepinephrine is reabsorbed into the presynaptic neuron via uptake-1 transporter. Most of the norepinephrine is stored again in vesicles while 5% to 10% of norepinephrine is metabolized to dihydroxyphenylglycol (DHPG) by monoamine oxidase (MAO) and released into the circulation. In addition, norepinephrine can be absorbed in non-neuronal tissue (e.g., cardiomyocytes) via the uptake-2 mechanism. In these tissues, norepinephrine is metabolized to DHPG (via MAO) and then to 3-methoxy-4-hydroxyphenylglycol (MHPG) by catechol-O-methyl transferase (COMT). ¹²³I-mIBG is chemically modified from guanethidine which is an analog of norepinephrine. ¹²³I-mIBG uses the same transporters (mainly uptake-1) and storage mechanisms like norepinephrine. Unlike norepinephrine, ¹²³I-mIBG is not metabolized by MAO and COMT and does not bind to postsynaptic adrenergic receptors

In this issue of the Journal of Nuclear Cardiology, Valborgland et al. evaluate the effect of aerobic exercise training on cardiac sympathetic activity in twenty-three HF patients receiving optimal medical therapy including β-blockers. All patients were submitted to planar ¹²³I-mIBG scintigraphy at baseline and after 12 weeks of exercise training. The authors found that different types of exercise training did not significantly change cardiac neuronal uptake as quantified by Heart-Mediastinum Ratio (HMR) in these patients. Given the strong and positive association between regular aerobic exercise and improved outcomes4,5 as well as the restoration of resting cardiac sympathetic activity following exercise in HF,6 these results are surprising. Importantly, one of the major advances in HF treatment during the past decade was the implementation of β-adrenergic-blocking agents in therapeutic strategies, which has substantially improved clinical outcomes in these patients.7,8 This beneficial effect seems to be driven by an improvement of cardiac sympathetic dysfunction.9 Valborgland et al. now provide evidence against a synergistic effect of exercise training and contemporary pharmacotherapy on sympathetic activation in HF by demonstrating that the hemodynamic and metabolic adaptations induced by 12 weeks of exercise training do not further improve cardiac sympathetic dysfunction in HF patients already on β-blocking therapy. As stated by the authors, these data contrast with an earlier report demonstrating a favorable effect of exercise training on cardiac neuronal dysfunction in HF patients not on β-blocking therapy.6 In addition, it has been shown that exercise training reduces muscle sympathetic nerve activity (MSNA) even in HF patients receiving β-blocker treatment.10 Of note, however, in contrast to quantification of sympathetic activity in individual organs, MSNA does not provide information on regional sympathetic outflow. Given that the magnitude of sympathetic activation in HF is target-organ specific, results may substantially differ. Importantly, while no changes in HMR or tracer washout rate (WR) were detected, Valborgland et al. observed a significant improvement in 6-minute walk test, peak oxygen uptake, and maximal work-load following exercise, indicating that the exercise regimen applied in their study further enhanced functional capacity in these patients. These intriguing results leave us with the question why there is a discrepancy between changes in functional capacity and cardiac innervation imaging in the response to exercise training? And might pathophysiological or technical factors account for these findings?

The benefits of exercise training in chronic HF have been extensively studied and comprise, besides reducing sympathetic outflow, an increase in renal blood flow as well as improved endothelial and muscle function, reduction of reactive oxygen species (ROS), circulating cytokines, angiotensin II and aldosterone levels, enhanced baro- and chemoreflex sensitivity, and an increase in vagal tone (Figure 2).11,12 While there is clear evidence that exercise-induced adaptions in HF are strongly interrelated, the relative contributions of these variables to improved outcomes in HF patients are not well understood. Indeed, a downregulation of sympathetic neuronal activity via central angiotensin II receptors and central antioxidant mechanisms has been shown, and studies have demonstrated an improvement of cardiac sympathetic dysfunction by angiotensin-converting enzyme (ACE) inhibitors and aldosterone, which do not directly affect myocardial adrenergic functions (Figure 2).13,14,15,16 However, a cause-effect relationship between these variables is yet to be established. Following this line of reasoning, contemporary pharmacotherapy including inhibitors of the renin-angiotensin-aldosterone system (RAAS) and β-blockers may substantially change the relative impact of different pathways and organ systems on exercise-induced adaptions in HF and may, as demonstrated by Valborgland et al., result in a greater exercise-induced effect on peripheral circulatory systems as compared to cardiac neuronal activation. Thus, although these data should be approached cautiously because of the small sample size, the most important contribution of Valborgland et al. is the demonstration that restoration of autonomic function following exercise training might substantially differ in HF patients with and without maximal pharmacotherapy. In this regard, it must be emphasized that the best response variable for the assessment of the efficacy of a treatment in HF patients is still under debate.17 Thus, knowledge about the exact organ system and specific pathways mediating the benefits of a therapeutic intervention in HF will lay the foundation for precision medicine and improved clinical endpoints in HF.

Figure 2

Effects of exercise training in systolic heart failure. N-terminal pro b-type natriuretic peptide. CNS central nervous system, LV left ventricular, TNF-α tumor necrosis factor-α, IL interleukin, iNOS inducible nitric oxide synthase, nNOS neuronal nitric oxide synthase, NT-proBNP N-terminal pro b-type natriuretic peptide, NO nitric oxide, GABA _A R γ-aminobutyric acid A receptor, Ang II angiotensin II, MitoVD mitochondrial volume density, RSNA renal sympathetic nerve activity

The merit of the study by Valborgland et al. also lies in the fact that current limitations of ¹²³I-mIBG imaging have been depicted. In fact, only few studies have assessed whether ¹²³I-mIBG imaging might help to decide who might benefit from a particular therapy.18,19 In addition, there are no clinical data using cardiac MIBG imaging for drug intervention in HF patients. Thus, the effect of ¹²³I-mIBG scintigraphy on patient’s management and selection of therapy is currently unknown. Given the high benefit to risk/cost ratio of conventional medical therapies, ¹²³I-mIBG imaging is currently unlikely to preclude their use. The present study by Valborgland et al. further supports this notion by indicating that the assessment of functional parameters might be a more sensitive test for a treatment response than cardiac neuronal imaging. This leads us to the ultimate question whether there is a role for ¹²³I-mIBG imaging in selecting an appropriate therapeutic strategy in HF patients. Noteworthy, despite improved therapeutic options in HF, the risk reduction rate of current HF drugs is still limited and drug intolerance and adverse effects are common. In addition, non-responders to therapeutic interventions in HF have been identified in previous studies.20 Finally, wide variations in HF disease phenotypes and the abundance of comorbidities in HF patients undoubtedly necessitate better-informed decision-making and a personalized disease management. Thus, given its ability to trace alterations of cardiac sympathetic nerve function caused by drug interventions,18 ¹²³I-mIBG imaging might have a clinical value in optimizing therapy in specific subpopulations such as patients not responding to routine therapeutic strategies or in patients in whom intolerable side effects prohibit adequate pharmacological dosing. In this regard, cardiac neuroimaging has been attributed a potential role in selecting patients likely to benefit from device therapy, given its value as a negative predictor for life-threatening arrhythmias.3 However, optimal cutoff values to identify these patients are lacking; therefore large-scale, interventional studies are needed to strengthen these findings.

Valborgland et al. conclude the discussion of their study by wondering whether a significant effect of exercise training on cardiac sympathetic activity would have been detected in a more deconditioned HF population such as patients classified in New York Heart Association (NYHA) class III or IV. Another reason for their neutral findings might be that the level of exercise applied in the present study was insufficient to induce significant changes in cardiac sympathetic activation. In fact, as acknowledged by the authors, previous studies have applied longer treatment durations when assessing the effect of exercise training on HF outcomes.6 Alternatively, pertaining to the small sample size of the study, individual variations might have had a significant confounding influence on cardiac sympathoexcitatory measures. Indeed, comorbidities such as obesity, sleep apnea, hypertension, myocardial ischemia, and diabetes, each of which have been shown to elevate the set point for sympathetic outflow, are common in HF patients.21,22 Thus, sympathetic tone differs widely among patients with HF and large inter-individual variations of myocardial ¹²³I-mIBG uptake have also been reported in normal subjects.23 These variations seem to be the result of both, methodological and (patho)physiological factors and need to be taken into account when interpreting ¹²³I-mIBG images.24 However, technical refinements in ¹²³I-mIBG imaging such as iterative reconstruction, volumetric analysis techniques as well as automated quantitation have been introduced in recent years. In addition, the need to standardize myocardial ¹²³I-mIBG scintigraphy has recently been emphasized.25 These current developments will certainly help to overcome the above limitations and to pave the way for a clinical implementation of ¹²³I-mIBG imaging.

Pertaining to a wider clinical use of cardiac ¹²³I-mIBG imaging, it will also be indispensable to consider the influence of sex and age on cardiac sympathetic activity in HF. The study of Valborgland et al. could not address this important issue, given the low number of females (n = 1) included in the study. Notably, sexual dimorphism in autonomic nervous control of the cardiovascular system has previously been demonstrated and has gained increasing attention in the context of Takotsubo cardiomyopathy.26 The fact that women with ischaemic HF exhibit increased cardiac-specific sympathetic activation and have worse outcomes than men, once affected, implies that this sexual dimorphism in cardiac autonomic physiology is pathogenetic.27 Of note, sex-specific differences in the predictive power of peak VO₂ measures in HF have also been described and lower cut-off values for women with HF have been suggested.28,29 In addition, several peculiarities with regard to sympathetic activation in older people need to be considered, such as a higher sympathetic nervous activity at baseline and in response to stress as compared to younger subjects.30 Thus, before ¹²³I-mIBG imaging can become a valuable addition to clinical practice in the treatment of HF, sufficiently powered studies are necessary to determine whether gender- and age-specific cutoff values are needed.

In summary, Valborgland et al. take another step forward in providing novel and important information indicating that the impact of a therapeutic intervention on autonomic dysfunction in HF might be influenced by concomitant medication. While their results require confirmation in additional studies where the above-mentioned limitations should be addressed, they also point out that understanding the multifactorial nature of cardiac autonomic modulation will continue to provide pathophysiologic and therapeutic insights for HF management. Larger clinical trials will be required to test whether ¹²³I-mIBG imaging might indeed be of value in better selecting patients who may benefit from additional treatment. Such an individualized treatment approach will become ever more important in the upcoming years, given the aging of the population and the increasing burden of HF. Therefore, specific subgroups, such as elderly and females, should also be investigated in more detail. As a priority for future research, standardization of myocardial ¹²³I-mIBG scintigraphy should be achieved in order to improve its clinical utility. Thus, for the time being, we should withhold the final word on the clinical utility of cardiac ¹²³I-mIBG imaging and await further data linking treatment responses with ¹²³I-mIBG images and, ideally, with future cardiovascular events.