are suppressed less by leptin administration in older than young rats. This suggests that ageing may be accompanied by leptin resistance, which would tend to increase food intake. The impact of human ageing on the effects of fasting on leptin levels and of leptin administration has not been reported.52
Ghrelin
Ghrelin stimulates feeding and growth hormone release. It is present in the hypothalamus, but the main site of production is the gastric mucosa.30 Circulating ghrelin concentrations increase with fasting and diet‐induced weight loss in obese subjects and are elevated in underweight, undernourished young and older subjects. In contrast, circulating concentrations decrease after food ingestion, particularly fat and carbohydrate, and are reduced in obese people. These changes are consistent with compensatory responses to, rather than causes of, these altered nutritional states. It therefore seems unlikely that reduced ghrelin activity contributes significantly to anorexia and weight loss in markedly undernourished older subjects. Nevertheless, the effects of ageing and undernutrition on sensitivity to ghrelin have not been reported, and ghrelin resistance may occur in these states. In support of this, older subjects are less sensitive to the growth hormone (GH)‐releasing effects of intravenous ghrelin (i.v. ghrelin) than young adults.53 The effect of healthy ageing on circulating ghrelin concentrations has not yet been clarified. A possible rationale for a decline in ghrelin levels with age, particularly in men, is the positive association between circulating testosterone and ghrelin concentrations and the increase in plasma ghrelin concentrations that occurs in hypogonadal men in response to testosterone therapy.54 As normal ageing is accompanied by reductions in circulating androgen levels (see the following section), this might have the effect of reducing ghrelin concentrations and thus food intake. One study found a rise in plasma ghrelin concentrations with increasing age, but there was no relationship with age per se when a multivariate analysis was performed, and the study did not include subjects older than 64.55 Two small studies have reported circulating ghrelin concentrations 20%56 and 35%57 lower in healthy older adults (69–87 and 67–91 years, respectively) than young adults, the latter reduction being statistically significant. However, increasing body fat, as indicated by BMI, is associated with decreasing ghrelin concentrations, and the older subjects had higher BMIs than the young subjects in both studies. Neither study included detailed body composition analysis, so the lower ghrelin levels in older subjects may have been because of differences in body composition. Another study found no difference in fasting and postprandial serum ghrelin concentrations between healthy older (mean age 78) and young adults. On balance, the effect of healthy ageing on circulating ghrelin concentrations has not yet been clarified but is probably minimal.58
Insulin
Human ageing tends to be associated with increased fasting and postprandial circulating insulin concentrations.59 Increased insulin activity could, therefore, be a cause of reduced food intake in older people. However, the evidence for a satiating role of insulin is limited. Suppression of food intake by insulin has only been demonstrated in animals and has required central insulin administration at high doses or high‐dose, prolonged peripheral administration. Short‐term, peripheral, euglycaemic insulin infusions have been shown not to affect appetite or food intake in humans.60 Moreover, age‐associated increases in insulin concentrations are due mainly to insulin resistance resulting from increased adiposity and only to a small extent to ageing itself. It seems unlikely that insulin contributes substantially, if at all, to the anorexia of ageing.
Testosterone and other androgens
Circulating androgen concentrations decline with ageing. This may contribute to the development of sarcopenia and the decrease in functional status that occurs with ageing (reviewed by Bhasin61). Whereas androgen replacement therapy (ART) is advocated for men with marked androgen deficiency, there is no consensus for the use of ART in elderly men with less severe ageing‐related declines in androgen concentrations or in elderly women. Studies of androgen replacement have been performed in healthy, older men with androgen deficiency, but although benefits have been seen in muscle mass and, in some cases, strength, as yet there is insufficient evidence regarding improvements in functional status (reviewed by Morley62). Several studies have suggested that there may be benefits from treating older men, particularly if frail, with testosterone therapy. Amory et al.63 gave older men with a mean total testosterone within the normal range 600 mg i.m. testosterone weekly for four weeks before elective knee replacement surgery and found significant increases in the ability to stand postoperatively and trends to improvements in walking and stair climbing, compared with placebo‐treated men. Bakhshi et al.64 gave older men in a rehabilitation programme with low‐normal testosterone levels 100 mgi.m. testosterone or placebo weekly and found significant increases in grip strength and the Function Independence Measure after testosterone but not placebo. Chapman et al.65 showed a reduced rate of hospitalization over one year in older men and women either undernourished or at risk of undernutrition given a combination of oral testosterone and a nutritional supplement compared with an untreated group. Srinivas‐Shankar et al.66 treated older, frail men with low circulating testosterone concentrations with transdermal testosterone or placebo for six months and found improvements in muscle strength and physical function, the latter effects confined to older (≥75 years) and more frail men. These results are not conclusive but justify further studies of testosterone treatment in frail older people.
In women, serum concentrations of testosterone and the adrenal androgens gradually and progressively decline from the decade preceding menopause. Even if testosterone therapy does not increase food intake in older, undernourished people, it may provide functional benefits by treating the associated sarcopenia.
Cytokines
Age‐associated increases in the production and/or effect of satiating cytokines may contribute to the anorexia of ageing67. Cytokines are secreted in response to significant stress, often because of malignancy or infection. Circulating concentrations of the cytokines interleukin 1 (IL‐1), interleukin 6 (IL‐6), and TNF‐α are increased in cachectic patients with cancer or AIDS. They act to decrease food intake and reduce body weight via a number of central and peripheral pathways. Blockade of these cytokines – for example, of TNF‐α in mice with TNF‐producing sarcomas – significantly attenuates weight loss in high‐stress conditions associated with cachexia. Ageing itself may be a form of stress. It is associated with stress‐like changes in circulating hormonal patterns, increased cortisol and catecholamines, and decreased sex hormones and growth hormone. Increased cortisol and catecholamine levels, in turn, stimulate the release of IL‐6 and TNF‐α, whereas sex hormones inhibit IL‐6. Interleukin 1 and IL‐6 levels are elevated in older people with cachexia, whereas plasma IL‐6 concentrations apparently increase as a function of normal ageing and correlate inversely with levels of functional ability in elderly people. Higher circulating levels of CRP and cytokine receptors also appear to be associated independently with physical dysfunction and disability.68 Increased cytokine levels due to the stress of ageing per se or the amplified stressful effects of other pathologies may explain some of the declines in appetite and body weight that occur in many older people.30
Chronic disease and anorexia in older adults
Common medical conditions in the elderly, such as gastrointestinal disease, malabsorption syndromes, acute and chronic infection, and hyper‐metabolism (i.e. hyperthyroidism), often cause anorexia, micronutrient deficiencies, and increased energy requirements. Cancer and rheumatoid arthritis, which produce anorectic effects by releasing cytokines, are also common in older adults. Protein‐energy malnutrition is particularly likely to develop in the presence of other ‘pathological’ factors, many of which become more common with increased age. The majority are at least partly responsive to treatment, so recognition is important.
Depression, often associated with bereavement and the deterioration of social networks, is a common psychological problem in older people, present in 2–10%