With an increasing life expectancy throughout the world, patients undergoing surgery will be inherently older and exhibit a greater prevalence of (congestive) heart failure (CHF). This increased prevalence of heart failure (HF) may give rise to significant increases in perioperative complications and mortality [1, 2]. Conclusions drawn from previous findings showed that opioids elicit cardioprotective effects against myocardial ischemic events in vitro (conditioning) [3]. As such, a high-dose opioid-based anesthesia technique has been advocated to reduce adverse cardiac events during anesthesia [4].

Recent experimental studies have detected all three opioid receptors in normal rat and human cardiac tissue [5‐8]. Besides a colocalization with sympathetic, parasympathetic, and sensory neurons, opioid receptors seem to be coexpressed with important structures of the myocardial excitation–contraction coupling and mitochondria in the left ventricle [5‐7, 9, 10]. Moreover, an activation of this intrinsic cardiac opioid system in volume-overloaded rat hearts was detected, reflected by an upregulation of delta- and kappa-opioid receptor expression and their respective endogenous ligand peptide precursors [5, 7]. In this context, acute infusion of kappa-opioid receptor agonists acutely provoked an augmented negative inotropic and lusitropic response in the failing ex vivo perfused hamster heart [11]. In addition, short-term systemic delta-opioid receptor inhibition increased cardiac output and improved left ventricular performance in dogs with CHF [12].

The question now arises whether the intrinsic cardiac opioid system may possess direct cardiodepressant effects. This proof-of-concept study is aimed at investigating whether the cardiac opioid system may influence cardiac function and neurohumoral parameters in volume-overloaded rat hearts as one experimental model of heart failure. We hypothesized that the persistent inhibition of the endogenous opioid tone by chronic treatment with the opioid receptor antagonist naltrexone in rats with aortocaval fistula (ACF)-induced volume overload will improve cardiac contractility and attenuate neurohumoral activation.

This study shows MOR, DOR, and KOR expressions in LV cardiomyocytes of ACF rats colocalizing with the expression of voltage-gated L-type Ca²⁺-channel Cav1.2 as part of the excitation–contraction coupling in the LV. In rats with ACF-induced volume overload, chronic administration of the opioid antagonist naltrexone was associated with an improved LV function as evidenced by decreases in CVP and LVEDP, increases in cardiac contractility, and decreases in rBNP-45 and angiotensin-2 plasma levels. In parallel, chronic naltrexone treatment led to a significant decrease of the ACF-induced increased expression of MOR, DOR, and KOR mRNAs and to a further significant increase in the already elevated mRNA of the endogenous opioid peptide precursors POMC and PENK, but not PDYN. These results possibly reflect a reduction in the cardiodepressive effects of the opioid system due to opioid receptor blockade and a compensatory prevalence of the sympathetic stress on the heart.

Our findings of opioid receptor and peptide mRNA as well as protein in the left ventricular myocardium are in line with several previous publications in rats which have demonstrated MOR, DOR, and KOR by mRNA detection, western blot, immunohistochemistry, and radiolabeled ligand binding [5‐7]. They are consistent with studies in humans in which MOR, DOR, and KOR were identified in myocardial tissue that was obtained during autopsy after sudden death [8] and with PET imaging studies in human volunteers at the Johns Hopkins Hospital demonstrating cardiac MOR and DOR [25]. Interestingly, all three opioid receptors (MOR, DOR, and KOR) as well as their respective endogenous ligands (derived from the precursors POMC, PENK, and PDYN) were shown to be upregulated during the stressful conditions of ACF-induced left ventricular volume overload [5‐7]. This animal model is characterized by significantly elevated CVP and LVEDP, a shift to higher volumes in the pressure–volume loops without any overlap, a significant decrease in the ejection fraction by almost 40%, and a significantly reduced maximum rate of pressure decay and prolonged time of tau depicting the transition from eccentric hypertrophy with preserved cardiac function to severe biventricular dilatation with decompensated heart failure [13]. This was further corroborated by showing clear biochemical, immunohistochemical, and electron microscopical evidence for extended myocardial apoptosis in this animal model [26]. Following naltrexone treatment of ACF, rats’ myocardial expression of MOR, DOR, and KOR mRNAs was significantly downregulated—even beyond control values—which might be related to the previously demonstrated enhanced opioid peptide levels during volume overload-induced heart failure [5, 7]. On the other side, opioid peptides derived from POMC and PENK targeting predominantly MOR and DOR, while dynorphins derived from PDYN having the highest affinity for KOR appear to be differentially regulated. Naltrexone increased the opioid peptide precursor levels POMC and PENK above those in the ACF/Veh group, while reducing PDYN levels below control values. While in the spinal cord or central nervous system, dynorphin follows an opposite functional role compared to ß-endorphin and met-enkepahlin [27]; nothing has been reported so far for the heart.

Due to their presumed neutral hemodynamic effects, high-dose opioids are widely used to provide analgesia in high-risk patients undergoing anesthesia [28, 29]. Opioids also seem to protect the heart from ischemia/reperfusion (IR) injury, and are, thus, a cornerstone of cardiac and high-risk non-cardiac anesthesia [30]. Interestingly, especially amongst elderly patients (age > 75 years) and patients with concomitant congestive heart failure (CHF); intravenous morphine increased in-hospital mortality when administered in acute coronary syndromes [31]. In addition, patients receiving morphine were more likely to develop CHF. In this context the question of potential opioidergic adverse effects on cardiac function arises.

Recent animal studies provided evidence for an intrinsic cardiac opioid system which seemed to be upregulated—receptor and endogenous ligand—during volume overload [5, 7]. In this context, short-term infusion of the opioid receptor antagonist naloxone increased arterial pressure, cardiac contractile function, and organ blood flow by acting on DOR in conscious dogs with pacing-induced CHF [12]. In line with this, Bolte et al. were able to provoke an augmented negative inotropic and lusitropic response in the failing ex vivo perfused hamster heart by administering selective agonists for KOR and DOR [11]. Opposite results were obtained by a short-term high-dose infusion of non-selective beta-endorphin in patients with mild to moderate CHF which improved LVEF, reduced systemic vascular resistance, and blunted the neurohormonal activation [32].

Extending the aforementioned findings, we demonstrated that chronic administration of naltrexone was able to improve cardiac function in volume-overloaded hearts in anesthetized rats. In previous work, morphine sulfate was shown to decrease heart rate and cardiac output in a dose-related fashion in healthy rat hearts, and naloxone attenuated these negative cardiovascular effects in ex vivo perfused healthy rat hearts [33]. This is somewhat in contrast to a previous observation in conscious dogs with right-sided CHF, in which a 6-week oral administration of naltrexone showed no effects on resting cardiac function [34]. An improvement of LV contractility was only detectable after beta-adrenergic isoprenaline stimulation.

Cardiac opioid receptors functionally and physically cross-talk with beta-adrenergic receptors via multiple hierarchical mechanisms, including heterodimerization of these receptors, counterbalance of functional opposing G protein signaling, and interface at downstream signaling events [35]. Moreover, opioid receptors such as KOR have been described to form heterodimers with the apelin receptors (APJ) and the AngII/AT1R system which are known regulators of the cardiovascular system [36‐38]. The heterodimerization of APJ and ATR1 reduces the physiological effects of angII, and heterodimerization of APJ and KORs modulates cardiac contractility and intensifies the blood pressure-lowering effect of apelin. Therefore, the interaction between the apelin/APJ, opioid/OPRs, and ang II/ATR1 systems and the heterodimerization of their receptors during volume overload conditions, under which apelin levels are reduced, may contribute to the cardiovascular regulation in ACF-induced HF [36‐38].

Liang et al. demonstrated already in 1987 that an upregulated opiate system may accompany CHF-induced sympathetic activation [39]. Effects of opioid receptor inhibition may, thus, be mediated via compensatory stimulation of the sympathetic nervous system or—more likely—by a reduced inhibition of the sympathetic nervous system. In the latter one, it is very intriguing that the cardiac actions of beta-adrenergic receptor (beta-AR) stimulation are attenuated by activation of the opioid receptor [40]. However, both states of increased sympathetic activity would hypothetically amplify the so-called vicious circle in cardiac remodeling resulting in increased workload reflected by increased BNP concentrations. We were able to demonstrate a significant reduction of the elevated rBNP-45 and angiotensin-2 levels in ACF rats treated with naltrexone. The RAAS, as part of the neurohumoral activation in heart failure, is also known to play a key role in fluid retention and cardiac remodeling [41]. The reduction in angiotensin-2 plasma levels might reflect an attenutated sympathetic activation in volume-overloaded rats with continuous naltrexone treatment. Sympathetic activation is known to induce inflammatory and apoptotic processes and is associated with adverse cardiac events [42, 43]. Inhibition of the RAAS significantly reduces cardiac fibrosis [44]. These findings might suggestthat reduced adrenergic activation might be associated with improved cardiac outcome.

Several limitations have to be considered. Firstly, these findings cannot directly be transfered to a clinical situation, since our experimental model does not represent the multimorbidity and causality of cardiac compromised patients. However, this animal model produces a very predictable state of nearly decompensated heart failure with a dilatative cardiomyopathy within 28 days. The Wistar rats showed overt signs of decompensation, e.g., ascites, strained breathing, decreased mobility, and sudden arrhythmia. Intriguingly, organ/body weight indices of the liver and kidney were not significantly altered in ACF rats which may be due to the large AV fistula itself that affects organ perfusion preventing it from increasing organ weight. Nonetheless, both organs showed overt histopathological changes in the liver [45] and kidney [46], as has been previously published. Secondly, this study was conducted to assess the indirect adverse opioidergic effects in cardiac volume overload. Therefore, the cardiac opioid system was blocked with an opioid receptor antagonist in a proof-of-concept design. Application of mu-, kappa-, and delta-opioid agonist would have been of benefit, but there already exists clinical data strongly supporting a negative influence of mu-opioid agonist [31]. In addition, naltrexone and its major metabolite, 6-β-natrexol, are known to mainly bind to mu-opioid receptors and to a lesser extent to kappa- and delta-opioid receptors (affinity (Ki) mu 1.0 nM, kappa 3.9 nM, delta 149 nM [47, 48]. We have not included a control group treated with naltrexone; however, naltrexone is known to lack any intrinsic activity at opioid receptors [49, 50]. The inhibition of the cardiac opioid system might be accompanied by centrally mediated effects. Thus, in future studies, effects of opioid receptor subtype selective agonists and antagonists with/without the ability to cross the blood brain barrier on cardiac function have to be investigated in an experimental setting. Thirdly, opioids can profoundly affect the respiratory system. The opioidergic effects on the respiratory system could likely be another significant consideration to explain the alteration of cardiac function. However, within the CRUSADE study, it has been demonstrated that acute and unique morphine administration during myocardial infarction resulted in an increased incidence of heart failure thereafter [31]. Unfortunately, we did not measure the breathing rate in all rats. But we can state that from a clinical point of view these rats appeared clinically inapparent during hemodynamic measurements. In this context, all measurements were performed under tiletamine/zolazepam anesthesia as this combination has to be found to elicit the least hemodynamic effect [18]. Thus, different anesthetic regimens might affect our findings. Our findings also have to be confirmed in other experimental models, e.g., myocardial infarction or pressure overload. Finally, the study was only conducted in male Wistar rats. Female Wistar rats are prone to a more difficult ACF induction, and standardization has not been established yet. Therefore, a comprehensive analysis of this nature amongst female Wistar rats has yet to be conducted.

In conclusion, the results of this experimental study give first-hand evidence of a cardiodepressant effect of the intrinsic cardiac opioid system during chronic volume overload. Thus, future studies have to address clinical opioid effects in perioperative patients with cardiovascular diseases.