Muscle loss during the ageing process

Narici and Maffulli calculated that total lean mass, which approximately equals the total skeletal muscle mass but is easier to measure, declines by about 18 % in men and by 27 % in women from the second to the eighth decade of life⁽ Reference Narici and Maffulli ^{10 )}. The development of sarcopenia has been implicated in several detrimental effects that are commonly encountered in the elderly⁽ Reference Dirks and Leeuwenburgh ^{29 )}: (i) increased incidence of falls that can lead to injury; (ii) impaired immune function and capability to fight off infection as a consequence of reduced protein storage in skeletal muscle; and (iii) development of disability, frailty and ultimately death⁽ Reference Wanger, Foster and Nguyen ^{25 ,} Reference Malmstrom, Miller and Herning ^{30 )}. In fact, it has been estimated that the annual healthcare cost in the USA due to sarcopenia-related morbidity reaches 18·5 billion US dollar⁽ Reference Janssen, Samson and Meeuwsen ^{31 ,} Reference Janssen, Baumgartner and Ross ^{32 )}. A fundamental difference, however, exists between disuse atrophy and ageing-associated sarcopenia, because only sarcopenia is associated with a decrease in muscle fibre size and number. Atrophy, on the other hand, is only associated with a decrease in muscle fibre size, but not number. Many different mechanisms are involved in these processes. To understand them, knowledge of the basics of skeletal muscle structure is pivotal. Indeed, muscle consists of slow twitch type I and fast twitch type II muscle fibres. Whilst type I fibres are more efficient at using oxygen for continuous, extended muscle contractions, type II fibres possess less mitochondria and preferably use anaerobic metabolism to create energy⁽ Reference von Haehling, Steinbeck and Doehner ^{14 ,} Reference Muiller-Hocker ^{33 )}.

The first description of different muscle fibre types dates back to 1873, when Louis Ranvier described red and white muscles in rabbits and rays as either tonic-slow or tetanic-fast, respectively⁽ Reference Ranvier ^{34 )}. Red muscles, now known to be primarily composed of type I fibres, possess high oxidative capacity due to high mitochondrial density, high myoglobin content and a high number of capillaries⁽ Reference Hultman ^{35 )}. Their metabolic profile is adjusted to their long twitch contraction time and their low force production, which also makes them resistant to fatigue. In addition, type I fibres show low activity of ATPase, creatine kinase and glycolytic enzymes. Their activating motor neurons are small⁽ Reference Hultman ^{35 )}. In this context, it is important to acknowledge that mitochondria produce about 90 % of ATP required for cellular functioning during oxidative phosphorylation⁽ Reference Dirks and Leeuwenburgh ^{29 )}. Type II fibres are different in many ways. Not only are they activated by larger motoneurons, which means that the level of excitation will determine the pool of neurons and thus the fibre type that will contract, but they also possess high ATPase, creatine kinase and anaerobic glycolysis activity.  Further, type II fibres are differentiated into type IIA and IIB fibres, which have different oxidative capacities⁽ Reference Hultman ^{35 )}. The fibre composition of any muscle can be subject to change in response to appropriate stimuli; for example, reduced muscular activity will yield slow-to-fast fibre transformation⁽ Reference Daugaard and Richter ^{36 )}. The opposite transformation appears during exercise training. During the ageing process, fast type II are more prone to atrophy than type I fibres, and indeed, cross sectional area of type II fibres has been shown to be reduced by 26 % in subjects aged 80 years compared with subjects who were aged 20 years⁽ Reference Lexell, Taylor and Sjöström ^{37 )}. No such reduction was reported for slow type I fibres. One of the reasons for this change is the preferential denervation and loss of fast motor units⁽ Reference Lexell ^{38 )}. The number of motor units remains fairly constant up to age 60 years, but declines thereafter at a rate of 3 % annually⁽ Reference Campbell, McComas and Petito ^{39 )}, amounting to 60 % loss at age 80 years⁽ Reference Narici and Maffulli ^{10 )}. Altogether it appears that up to the late seventies, type II fibres are predominantly affected, but from approximately age 80 years onwards, both types of fibres are lost⁽ Reference Narici and Maffulli ^{10 )}.

Several other factors play roles in declining muscle mass and strength (Fig. 1). The infiltration of skeletal muscle by fat and connctive tissue is another major contributor⁽ Reference Borkan, Hults and Gerzof ^{40 )}. Overend et al. ⁽ Reference Overend, Cunningham and Paterson ^{41 )} described more than 20 years ago that, even though total thigh cross-sectional area is not different between young and elderly men, older men have 13 % lower total muscle plus bone area, 26 % lower quadriceps and 18 % lower hamstring area. Using computed tomography imaging, Taaffe et al. ⁽ Reference Taaffe, Henwood and Nalls ^{42 )} recently described that cessation of resistance exercise in trained older persons increases the fatty infiltration of muscle, while resumption of exercise decreases it. This effect adds to the development of mitochondrial dysfunction that is evident in the elderly and possibly a consequence of oxidative damage by reactive oxygen species⁽ Reference Dirks and Leeuwenburgh ^{29 ,} Reference Cadenas and Davies ^{43 )}; reduced energy production may also play a role in the development of myocyte apoptosis discussed later⁽ Reference Pollack and Leeuwenburgh ^{44 )}. In fact, animal models have shown that the ATP content in aged muscle may be 50 % lower than that of young counterparts⁽ Reference Drew, Phaneuf and Dirks ^{45 )}.

Fig. 1.

(Colour online) The wasting continuum in heart failure (HF). Mitochondrial dysfunction, infiltration by fat and connective tissue, muscle fibre transformation, denervation of fast type fibres and myocyte apoptosis all play a role in muscle wasting (sarcopenia) that develops early in HF. Overactivity of the ubiquitin–proteasome system leads to myofibril degeneration. Cachexia is less prevalent than muscle wasting. It is associated with muscle and adipose tissue wasting; osteoporosis affects up to 50 % of patients but has been studied least well in HF.

An anabolic/catabolic imbalance is another important player in the development of sarcopenia. From this perspective, muscle loss may be a consequence of reduced muscle anabolism, increased muscle catabolism or both⁽ Reference von Haehling, Steinbeck and Doehner ^{14 )}. Such imbalance may yield histological changes in the muscles’ ultrastructure, finally leading to increased degradation of myofibrils and myocyte apoptosis. Indeed, TNF has been shown to be able to induce myocyte apoptosis⁽ Reference Narici and Maffulli ^{10 )}, but is also able to stimulate myofibril degradation via the ubiquitin–proteasome pathway⁽ Reference von Haehling, Genth-Zotz and Anker ^{46 )}. Therefore, the balance between anabolic players such as testosterone, growth hormone or insulin-like growth factor-1 and the catabolic factors TNF, IL-1β, interferon-γ, myostatin or glucocorticoids is of major importance in this regard⁽ Reference von Haehling, Steinbeck and Doehner ^{14 )}. Lack of anabolic growth factors inside of skeletal muscles and reduced physical activity have also been found to be involved in an increased rate of apoptosis⁽ Reference Barton-Davis, Shoturma and Sweeney ^{47 )}. Interestingly, apoptosis activity has been shown to inversely relate to muscle weight, suggesting a causative relationship⁽ Reference Dirks and Leeuwenburgh ^{29 )}. The predominant mechanism of myofibril degradation involves the adenosine triphosphate-dependent ubiquitin–proteasome pathway, which is responsible for the degradation of myofibril proteins from the intracellular compartment. However, it appears that an overactivity of this pathway does not play a major role in the development of sarcopenia in healthy elderly subjects, whereas it does play a prominent role in patients with chronic inflammatory syndromes such as HF⁽ Reference Narici and Maffulli ^{10 ,} Reference Murton, Constantin and Greenhaff ^{48 )}.

Muscle loss in heart failure

Peripheral loss of skeletal muscle tissue is a general finding in patients with HF and occurs early in the course of the disease⁽ Reference Loncar, Fülster and von Haehling ^{49 )}. Muscle mass is the major determinant of both resting energy expenditure⁽ Reference Tacke, Ebner and Boschmann ^{50 )} and exercise capacity⁽ Reference Cicoira, Zanolla and Franceschini ^{51 )}. The muscle hypothesis⁽ Reference Coats ^{52 )} holds that many factors are involved in the development of reduced peak oxygen uptake (peak VO₂) in HF. Besides physical inactivity, the muscle itself is affected by ultrastructural abnormalities, alterations in mitochondrial structure and function, oxidative stress and a shift in fibre distribution⁽ Reference von Haehling, Steinbeck and Doehner ^{14 ,} Reference Sullivan, Green and Cobb ^{53 )}. All these factors play a role not only in the reduced exercise capapcity in patients with HF, but also in the development of sarcopenia. Among male non-cachectic patients with HF, lean tissue in the legs was reduced by 9·1 % compared with healthy male control subjects of similar age⁽ Reference Anker, Ponikowski and Clark ^{21 )}. The thirty-six patients with HF in the present study were (mean (sem)) age 58·9 (1·3) years, weighed 83·8 (2·0) kg, their mean New York Heart Association class was 2·7 (0·1), and their mean left ventricular ejection fraction was 25 (2)%. Study participants’ total lean tissue was 57·4 (1·0) kg⁽ Reference Anker, Ponikowski and Clark ^{21 )}. However, patients in the present study were not investigated for the presence or absence of sarcopenia. A later study from my group in a significantly larger cohort of patients from the studies investigating co-morbidities aggravating heart failure, conducted among 200 male and female patients with stable ambulatory HF (mean (sd), age 66·9 (10·4) years, weight 86·7 (16·9) kg, New York Heart Association class 2·3 (0·5), left ventricular ejection fraction 38·9 (13·5) %) found that the patients’ mean lean mass was 54·6 (10·8) kg and thus similar to that seen in the study published one and half decades earlier⁽ Reference Fülster, Tacke and Sandek ^{20 )}. The prevalence of sarcopenia in the present study was 19·5 %, and thus too high to be simply attributed to advanced age, since these patients had, on average, not reached 70 years. Sarcopenia, however, was associated with lower hand grip strength, lower quadriceps strength, lower gait speed and 6-min walk distance as well as lower left ventricular ejection fraction and peak oxygen consumption as assessed using spiroergometry⁽ Reference Fülster, Tacke and Sandek ^{20 )}.

Previous studies had studied skeletal muscle mass and came to different conclusions, because no strigent cut-off criteria for sarcopenia were used. Rather, the respective authors described skeletal muscle mass reduction per se. Mancini et al. ⁽ Reference Mancini, Walter and Reichek ^{54 )}, for example, found that 68 % of their fifteen patients with chronic HF had reduced skeletal muscle mass compared with healthy controls, as evidenced by a decrease in the creatinine: height ratio and/or a reduced upper arm circumference of <5 % of normal. Using MRI, the authors also demonstrated that the patients’ calf muscle volume was reduced, whereas fat mass was largely preserved⁽ Reference Mancini, Walter and Reichek ^{54 )}. Drexler et al. ⁽ Reference Drexler, Riede and Münzel ^{55 )} studied fifty-seven patients with chronic HF and eighteen healthy control subjects and found that patients with HF had by 20 % reduced density of mitochondria in their skeletal muscle. Imprtantly, the fibre-type distribution was shifted towards type II fibres in patients with HF and their capillary length density of skeletal muscle was also reduced⁽ Reference Drexler, Riede and Münzel ^{55 )}. Most importantly, the density of mitochondria was strongly associated with the patients’ peak oxygen consumption⁽ Reference Drexler, Riede and Münzel ^{55 )}. Using quadriceps biopsies from nine patients with severe chronic HF, Lipkin et al. ⁽ Reference Lipkin, Jones and Round ^{56 )} found increased acid phosphatase, increased interstitial cellularity, excess intracellular lipid accumulation, atrophy of both type I and II fibres and variation in size with hypertrophy and atrophy of fibres. Completing the pathophysiological portfolio of sarcopenia development in HF, increased skeletal myocyte apoptosis has been observed in affected patients⁽ Reference Vescovo and Dalla Libera ^{57 )}. Adams et al. ⁽ Reference Adams, Jiang and Yu ^{58 )} studied thirty-four patients with mild-to-moderate chronic HF and detected apoptosis in skeletal muscle biopsies from the vastus lateralis muscle in sixteen of these patients (47 %). Patients with skeletal muscle apoptosis showed significantly lower peak oxygen consumption as assessed by spiroergometry (12·0 (sd 3·7) v. 18·2 (sd 4·4) ml/kg per min, P = 0·0005). Lower peak VO₂ was also described when Harrington et al. ⁽ Reference Harrington, Anker and Chua ^{59 )} studied 100 patients with chronic HF and compared them with thirty-one healthy controls (18 (sd 0·6) v. 33·3 (sd 1·4) ml/min per kg, P < 0·0001). Computed tomography-measured cross-sectional area in the quadriceps at mid-thigh and in the total leg were lower in patients with HF than in controls (both P < 0·05). Overall, patients were weaker than the control subjects as assessed by quadriceps maximal isometric strength (P < 0·005), even after correcting strength for the cross-sectional area⁽ Reference Sullivan, Green and Cobb ^{53 )}.

Just as in patients with ageing-associated sarcopenia, mitochondria-released cytochrome c has been implicated in the development of muscle wasting in HF⁽ Reference Vescovo and Dalla Libera ^{57 )}. Cytochrome c release into the cytosol is a consequence of the intracellular balance of the regulatory proteins Bax (pro-apoptotic) and Bcl-2 (anti-apoptotic). Cytochrome c release in turn activates caspase-9, which initiates a cascade of events that leads to apoptosis⁽ Reference Dirks and Leeuwenburgh ^{29 )}. Gielen et al. ⁽ Reference Gielen, Sandri and Kozarez ^{60 )} have recently shown that MuRF-1 expression in vastus lateralis muscle biopsies was elevated among sixty patients with chronic HF when compared with sixty healthy controls of similar age, suggesting clinically relevant activation of the ubiquitin–proteasome system, for which MuRF-1 acts as an important transcription factor. Subjects were randomised to 4 weeks supervised endurance exercise training or a control group. Exercise training reduced the expression of MuRF-1 mRNA by more than 30 % in a repeat biopsy at 4 weeks follow-up⁽ Reference von Haehling, Steinbeck and Doehner ^{14 ,} Reference Anker, Ponikowski and Clark ^{21 )}. Other local factors appear to play important roles in HF as well as evidenced by reduced expression of local skeletal muscle expression of the anabolic insulin-like growth factor-1⁽ Reference Hambrecht, Schulze and Gielen ^{61 )}. As reviewed in detail elsewhere⁽ Reference von Haehling, Steinbeck and Doehner ^{14 )}, insulin-like growth factor-1 is a key regulator of protein kinase B, also known as Akt, whose downstream target is called mammalian target of rapamycin, a pivotal regulator of translation initiation and overall muscle size. It appears that basal protein kinase B phosphorylation is decreased in patients with HF compared with age- and physical-activity-matched controls⁽ Reference Toth, Ward and van der Velden ^{62 )}. Pro-inflammatory cytokines such as TNF do not only play a role in the induction of proteasome-activity and apoptosis, but they can also induce a state of growth hormone resistance⁽ Reference Cicoira, Kalra and Anker ^{63 )}. Indeed, many pro-inflammatory mediators have been shown to be overexpressed in patients with HF⁽ Reference Anker and von Haehling ^{64 )} and their levels are associated with poor survival⁽ Reference Ferrari, Bachetti and Confortini ^{65 ,} Reference Rauchhaus, Doehner and Francis ^{66 )}. Ventricular assist device implantation in patients with advanced HF has been shown to enhance skeletal muscle expression of insulin-like growth factor-1, to increase protein kinase B phosphorylation, and to increase the fibre cross-sectional area⁽ Reference Khawaja, Chokshi and Ji ^{67 )}.

Muscle mass, muscle strength and musle wasting

It is important to understand that muscle wasting and fat wasting follow different pathophysiological pathways. Therefore, sarcopenia and cachexia may require different therapeutic approaches, even though these can show overlap to some extent. The only clinically meaningful therapeutic approach to muscle wasting in HF that has sufficient evidence is exercise training, because it is able to reduce oxidative stress, pro-inflammatory cytokine production, tissue expression of myostatin and the overall activity of the ubiquitin–proteasome system⁽ Reference Gielen, Sandri and Kozarez ^{60 ,} Reference Lenk, Erbs and Höllriegel ^{75 –} Reference Smart and Steele ^{77 )}. Exercise training is also being advocated for patients with cancer in order to improve their exercise capacity and quality of life⁽ Reference Gould, Lahart and Carmichael ^{78 ,} Reference Sasso, Eves and Christensen ^{79 )}. The present guidelines for HF management by the European Society of Cardiology state that ‘it is recommended that regular aerobic exercise is encouraged in patients with HF to improve functional capacity and symptoms⁽ Reference McMurray, Adamopoulos and Anker ^{80 )}. This is a class I recommendation with a level of evidence grade A as at least one randomised controlled study has proven its efficacy⁽ Reference O'Connor, Whellan and Lee ^{81 )}. Effects of exercise training include improved quality of life and exercise capacity and ultimately beneficial effects on mortality and hospitalisations⁽ Reference Piepoli, Conraads and Corrà ^{82 )}. Detailed recommendations can be found in a consensus statement issued by the European Society of Cardiology⁽ Reference Piepoli, Conraads and Corrà ^{82 )}.

A number of treatment approaches have been pursued for the treatment of muscle wasting and in order to increase muscle strength and exercise capacity in patients with HF⁽ Reference Ebner, Elsner and Springer ^{83 ,} Reference Anker, von Haehling and Springer ^{84 )}. No study has so far specifically investigated therapeutic effects in sarcopenic patients with HF, very few have studied cardiac cachexia. Interventions include essential amino acid supplementation in order to provide elements for the growth of skeletal muscle⁽ Reference Wakabayashi and Sakuma ^{85 )}, recombinant human growth hormone, synthetic ghrelin, testosterone replacement therapy, β-blockers, fish oil and β-receptor agonists⁽ Reference Ebner, Elsner and Springer ^{83 )}. Included patient cohorts, however, rarely exceeded fifty patients. Growth hormone application has seen tremendous trial acitivty in the past, but clinical benefit could not be demonstrated⁽ Reference Osterziel, Strohm and Schuler ^{86 )}. β-Receptor agonists have been tested in the management of muscle wasting⁽ Reference Ryall and Lynch ^{87 –} Reference Toledo, Springer and Busquets ^{89 )}, for example fenoterol or clenbuterol, but despite its pathophysiological attractiveness, this approach may be dangerous in patients with HF. Nutritional administration of essential amino acids appears to be promising, as an investigator-blinded, randomised study in thirty-eight patients with HF has shown that improvements in peak VO₂ and 6-min walk distance are possible with this approach within 2 months of treatment⁽ Reference Aquilani, Opasich and Gualco ^{90 )}, but larger studies are required to confirm these results. Banerjee et al. ⁽ Reference Banerjee, Caulfield and Crowe ^{91 )} used another interesting approach. testing the effects of electrical stimulation using a 6-week training programme during which patients were exposed to electrical stimulation of the major leg muscles for a minimum of 1 h, for 5 d each week. Peak VO₂ increased from 19·5 (sd 3·5) to 21·2 (sd 5·1) ml/kg/min (P < 0·05), walking distance from 415 (sd 57) to 455 (sd 55) m (P < 0·005) and quadriceps strength increased from 378 (sd 110) to 405 (sd 109) N (P < 0·005). No change in BMI was noted⁽ Reference Banerjee, Caulfield and Crowe ^{91 )}.

Testosterone administration has been associated with improvements in exercise capacity, and, indeed, testosterone deficiency is a phenomenon that is frequently observed in patients with HF⁽ Reference Volterrani, Rosano and Iellamo ^{92 )}. Reduced serum levels are associated with muscle myopathy⁽ Reference Josiak, Jankowska and Piepoli ^{93 )} as well as poor survival⁽ Reference Jankowska, Biel and Majda ^{94 )}. Caminiti et al. ⁽ Reference Caminiti, Volterrani and Iellamo ^{95 )} studied seventy male patients with stable chronic HF, who were randomly assigned to reiceive either intramuscular injection of testosterone every 6 weeks or placebo (n 35 per group). Baseline peak VO₂ was directly related to the serum testosterone level and improved significantly after 12 weeks of treatment from 13·4 ± 4·4 to 16·3 ± 1·7 ml/kg/min (P < 0·05). Likewise, testosterone-treated patients’ 6-min walk distance increased from 387 (sd 121) to 473 (sd 138) m (P < 0·05). Body weight increased from 63·5 (sd 13·7) to 66·8 (sd 11·4) kg (P < 0·05). No such change was noted in the placebo group⁽ Reference Caminiti, Volterrani and Iellamo ^{95 )}. A similar study from the same group of workers was published 1 year later, performed in thirty-six elderly female patients with chronic HF of ischaemic aetiology, twenty-four of whom were randomised to a testosterone transdermal patch or to placebo⁽ Reference Iellamo, Volterrani and Caminiti ^{96 )}. After 6 months treatment, their peak VO₂ had increased from 10·5 (sd 1·0) to 13·2 (sd 1·8) ml/kg/min and their 6-min walk distance from 261 (sd 52) to 357 (sd 43) m (both P < 0·05). Altogether, testosterone therapy was well tolerated; only one patient left the study for generalised purigo. Virilisation was not noted in any of the patients⁽ Reference Iellamo, Volterrani and Caminiti ^{96 )}. A novel class of drugs, selective androgen receptor modulators, has recently received tremendous research endeavour for their ability to increase muscle mass in clinical studies; however, their effects on musle strength or exercise capacity are not yet convincing⁽ Reference Takagi, Horie and Fujita ^{97 –} Reference Crawford, Dalton and Hancock ^{101 )}. Studies in HF are not available.

Cachexia

Treatment studies to tackle cachexia and thus body rather than muscle wasting are different and require different approaches. Strategies include nutritional advice, appetite stimulants such as megestrol acetate, cyproheptadine and dronabinol anabolic steroids, ghrelin, or myostatin antibodies⁽ Reference von Haehling, Lainscak and Springer ^{102 )}. However, most experience stems from cancer cachexia studies, and studies into the treatment of cardiac cachexia remain a rarity. This is surprising, considering an estimated prevalence of 2 % for HF in European countries alone. Even though 80 % of these patients are at risk to develop cardiac cachexia, approximately only 10 % are in fact cachectic⁽ Reference von Haehling and Anker ^{103 )}. But even this comparatively low prevalence of cachexia translates into approximately 1·2 million patients with cardiac cachexia in Europe⁽ Reference von Haehling and Anker ^{103 )}. One small study evaluated the anabolic steroid oxymetholone, administered for 3 months at a dose of 5–10 mg daily to patients with idiopathic dilated cardiomyopathy; however, these patients were not cachectic. Treatment led to significant decreases in left ventricular end-diastolic and end-systolic diameters and decreased left ventricular mass. Rozentryt et al. studied the effects of a nutritional intervention of 6 weeks duration in twenty-nine patients with cardiac cachexia⁽ Reference Rozentryt, von Haehling and Lainscak ^{104 )}. A diet containing 2512kJ (600 kcal)/d with relatively high fat content increased body weight during the treatment and the subsequent 12-week follow-up period and had anti-inflammatory effects in that it reduced serum levels of TNF. Patients’ quality of life of patients improved for the 6 weeks of the period of the nutritional intervention, but decreased somewhat in the follow-up period⁽ Reference Rozentryt, von Haehling and Lainscak ^{104 )}.

Ghrelin is a peptide hormone that was originally identified in 1999⁽ Reference Kojima, Hosoda and Date ^{105 )} and is mainly produced in the fundus region of the stomach⁽ Reference Date, Kojima and Hosoda ^{106 )}, but also in other organs⁽ Reference von Haehling, Lainscak and Springer ^{102 )}. Ghrelin induces the release of growth hormone from the pituitary gland thereby regulating appetite, but it also has anti-inflammatory properties. Several studies in cancer have shown promising results⁽ Reference Blum, Oberholzer and de Wolf-Linder ^{107 )}. In patients with HF, ghrelin plasma levels are elevated in cachectic HF compared with non-cachectic patients (237 (sd 18) v. 147 (sd 10) fmol/ml, P < 0·001)⁽ Reference Nagaya, Uematsu and Kojima ^{108 )}. A small, uncontrolled study of intravenous infusion of ghrelin in ten patients with cardiac cachexia, predominantly patients in New York Heart Association class III, showed promising cardiovascular results⁽ Reference Nagaya, Moriya and Yasumura ^{109 )}. Patients received ghrelin at a dose of 2 µg/kg body weight for 30 min twice daily for 3 weeks, thereby inducing a 25-fold increase in growth hormone serum levels and an increase in food intake lean body mass as well as an increase in left ventricular ejection fraction (from 27 (sd 2) to 31 (sd 2) %, P < 0·05)⁽ Reference Nagaya, Moriya and Yasumura ^{109 )}. Similar results regarding heart function, body weight and body composition have been described in animal models of HF⁽ Reference Nagaya, Uematsu and Kojima ^{110 )}. Some ghrelin analogues are in development and have shown promising early results in animal models of HF⁽ Reference Palus, von Haehling and Doehner ^{111 )}, reducing, for example, the expression of myostatin in skeletal muscle⁽ Reference Lenk, Palus and Schur ^{112 ,} Reference Chen, Guillory and Zhang ^{113 )}. Therapeutic effects included increases in lean mass as well as in fat mass⁽ Reference Palus, Schur and Akashi ^{114 )}. Ghrelin receptor agonists such as anamorelin are currently also undergoing clinical testing and can achieve increases in body weight as well as appetite, but studies have so far focused on cancer cachexia, and large-scale studies are still missing⁽ Reference Garcia, Friend and Allen ^{115 –} Reference Pietra, Takeda and Tazawa-Ogata ^{118 )}. Besides, several myostatin antibodies are currently in development and are being tested in clinical trials. No studies in HF are currently available, but animal models have shown that, for example, REGN1033, a fully human anti-myostatin antagonist antibody, increased fibre size, muscle mass and force production in young mice by approximately 20 %, and it improved physical performance outcomes in combination with treadmill exercise in 2-year-old mice⁽ Reference Bauerlein, Pangilinan and Salzler ^{119 )}.

Espindolol (MT-102) is another substance that has recently finished a phase II clinical trial, showing good efficacy in patients with cancer cachexia. A total of eighty-seven patients were randomly assigned to espindolol, an anabolic/catabolic transforming agent that has three potential pharmacological targets in cachexia, namely reduced catabolism via non-selective β-blockade, reduced fatigue and thermogenesis through central 5-hydroxytryptamine 1a antagonism and increased anabolism through partial β₂ receptor agonism or placebo⁽ Reference Stewart Coats, Srinivasan and Surendran ^{120 )}. After 16 weeks treatment, patients on espindolol showed beneficial effects on body weight and hand grip strength, i.e. the slope of weight change in the high-dose group of espindolol demonstrated a positive slope compared with a negative slope in the placebo group (P < 0·0001)⁽ Reference Stewart Coats, Fuang and Prabbash ^{121 )}.