|Home | About | Journals | Submit | Contact Us | Français|
UNINTENTIONAL WEIGHT LOSS, or the involuntary decline in total body weight over time, is common among elderly people who live at home. Weight loss in elderly people can have a deleterious effect on the ability to function and on quality of life and is associated with an increase in mortality over a 12-month period. A variety of physical, psychological and social conditions, along with age-related changes, can lead to weight loss, but there may be no identifiable cause in up to one-quarter of patients. We review the incidence and prevalence of weight loss in elderly patients, its impact on morbidity and mortality, the common causes of unintentional weight loss and a clinical approach to diagnosis. Screening tools to detect malnutrition are highlighted, and nonpharmacologic and pharmacologic strategies to minimize or reverse weight loss in older adults are discussed.
Unintentional weight loss is the involuntary decline in total body weight over time. In clinical practice, it is encountered in up to 8% of all adult outpatients1 and 27% of frail people 65 years and older.2 Weight loss is an important risk factor in elderly patients. It is associated with increased mortality, which can range from 9% to as high as 38% within 1 to 2.5 years after weight loss has occurred.1,3,4 Frail elderly people,5 people with low baseline body weight,5,6,7 and elderly patients recently admitted to hospital are particularly susceptible to increased mortality.8,9 Weight loss is also associated with an increased risk of in-hospital complications,10,11 a decline in activities of daily living or physical function,12,13 higher rates of admission to an institution2,8 and poorer quality of life.14
Unintentional weight loss may reflect disease severity (e.g., in patients with advanced heart disease, lung disease or malignant disease) or undiagnosed illness. Weight loss of 4%–5% or more of body weight within 1 year, or 10% or more over 5–10 years or longer, is associated with increased mortality or morbidity or both (Table 1). This association has also been seen in a number of epidemiologic and clinical studies that adjusted for comorbidity,5,7,13 disability,5,13 smoking,5,7 alcohol use5 or level of physical activity13 and that excluded deaths within the first few years of weight loss to exclude undiagnosed illness.7 In frail elderly populations, even small weight loss (e.g., 1 kg,5 or 3% of body weight18) may be significant. Voluntary weight loss among elderly patients is also associated with increased risk of death17 and of hip fracture,19 which highlights the importance of maintaining weight with age.
Prevalence estimates of weight loss among elderly people vary tremendously. The results of epidemiologic studies have shown that most elderly patients maintain weight over a reasonably long period of 5–10 years.13,20,21 Nevertheless, about 15%–20% experience weight loss — defined in these studies as a loss of either 5 kg or more or 5% of usual body weight over 5–10 years — with little difference between sexes.13,15,20 This prevalence estimate rises to 27% in high-risk populations, such as free-living frail elderly people receiving community services.2 Increasing age,13,22 disability,9,13 coexisting medical illnesses,19,22 previous admission to hospital,13 low education level,22 presence of cognitive impairment,23 smoking,13,19,22 loss of a spouse13,22 and low baseline body weight9,22 have been associated with a higher likelihood of weight loss. The proportion of elderly people who experience rapid (within 6 months), severe (≥ 7.5% of baseline body weight) and unexplained weight loss is only 0.45%.24
The incidence of unintentional weight loss in clinical studies involving adults seeking health care varies from 1.3% to 8%, depending on the setting and definition of weight loss.1,3,4,25 There is also a difference in rates between clinical and epidemiologic studies, probably because most patients with weight loss present within a year of onset of their clinical symptoms.
In general, causes of weight loss in elderly people are similar to those in middle-aged people and can be classified as organic (e.g., neoplastic, nonneoplastic and age-related changes), psychological (e.g., depression, dementia, anxiety disorders) or nonmedical (e.g., socioeconomic conditions) (Box 1). Up to one-quarter of all cases have no identifiable cause, despite extensive investigation.1,3,24 People with no known cause of weight loss generally have a better prognosis than people with known causes, particularly when the cause is neoplastic.1,29
Often a combination of factors will lead to weight loss in elderly people, particularly frail people 75 years or older. Many of these factors are not traditional medical diseases (Box 2). For example, patients with dementia or late-life psychotic disorders may become paranoid and suspicious that the food being served to them is poisoned.32 Elderly people with dementia and habitual wandering expend significant energy in pacing. As well, some common illnesses may cause weight loss (e.g., gallstones may lead to chronic nausea and decreased appetite or avoidance of high-energy, fatty foods).
Several important age-associated physiologic changes predispose the elderly person to weight loss, such as declining chemosensory function (smell and taste),33,34 reduced efficiency of chewing,35 slowed gastric emptying36 and alterations to the neuroendocrine axis (including changes in levels of leptin, cholecystokinin, neuropeptide Y and other hormones and peptides).37,38 These changes are associated with early satiety and a decline in both appetite and the hedonistic appreciation of food, and collectively they contribute to the “anorexia of aging.” Other evidence also suggests that, compared with healthy younger adults, elderly people are less able to adapt to periods of over- and undereating and less likely to return to their usual body weight after such periods,39 which makes them more susceptible to weight change. The importance of medications in contributing to weight loss cannot be overstated, since many elderly people take medications, mostly for chronic conditions (Table 2).
Various mechanisms have been suggested to explain the association between weight loss and adverse outcomes. Weight loss exacerbates the loss of fat-free mass (sarcopenia) associated with aging,42 which leads to functional decline and fractures.43,44 Many elderly patients with unintentional weight loss are experiencing concomitant malnutrition45 and thereby have cachexia.46 Cachexia is associated with a disproportionate loss of skeletal muscle rather than of body fat and is generally defined as a profound and marked state of constitutional disorder, general ill health and malnutrition. A decline of even 10% of skeletal muscle mass may be associated with a decline in physical function (e.g., decreased exercise tolerance or difficulty performing activities of daily living).47 In addition, cachexia is associated with a systemic inflammatory response, increased cytokine concentrations and impaired immunity, all of which are thought to contribute to adverse outcomes, including early death.48
It is important to establish up front the presence of weight loss. A significant proportion of elderly people with documented weight loss may not complain about losing weight or, less commonly, may mistakenly attribute weight loss to successful diet or lifestyle modifications.1,24 Furthermore, disturbed eating behaviours and body image (e.g., anorexia tardive)32 among some elderly people may lead them to regard weight loss as desirable and therefore nonreportable. Conversely, up to half of people who claim to have lost weight have no documented evidence of weight loss.1 If it is not possible to measure weight directly, a change in clothing size, corroboration of weight loss by a relative or friend, or a numerical estimate of weight loss provided by the patient are suggestive enough of true weight loss.1,3,29 A careful history may elicit localizing symptoms (e.g., changes in defecation frequently imply involvement of the gastrointestinal tract) that may guide further investigations in almost half of patients.4,29 All elderly patients with weight loss should undergo screening for dementia and depression30 by using instruments such as the Mini-Mental Status Examination49 and the Geriatric Depression Scale50 respectively. Specific features on physical examination, such as cachexia, lymphadenopathy or palpable masses, may suggest a physical cause of weight loss (e.g., malignant disease).1,4 However, the diagnostic utility of the medical history and physical examination in identifying the cause of weight loss have not been adequately evaluated.
Although few studies have systematically evaluated the utility of screening investigations for weight loss, the most useful noninvasive procedures appear to include a complete blood count, tests of liver enzyme levels (including alkaline phosphatase and bilirubin), measurement of lactate dehydrogenase level, and chest radiography.1,4,25,29 Patients with iron-deficiency anemia or symptoms likely to originate in the gastrointestinal tract, and patients with elevated liver enzyme levels on initial screening, should undergo investigation of their gastrointestinal tract (either endoscopy or upper gastrointestinal series) or an abdominal ultrasound, respectively.1,4,24,29
Three scoring systems have been developed to help clinicians identify which patient with weight loss is likely to have a physical1,4 or malignant51 cause as opposed to a psychological or unknown cause. None of these scoring systems has been validated in independent populations presenting with weight loss.1,4,51
When weight loss is apparent in the elderly patient with no evidence of an organic disorder, primary malnutrition (i.e., resulting from inadequate food intake) must be considered as a contributor.
In general, elderly people are at increased risk of malnutrition because of insufficient food intake (quantity) rather than inappropriate selection of food (quality). Two screening tools, ENS52,53 (www.dietitians.ca/seniors/content/other/clsc_overview.asp) and SCREEN54,55 (www.dietitians.ca/seniors/index.asp), have been developed and validated in Canada to identify community-dwelling elderly people who are at risk of malnutrition. Two other assessment tools, the Mini Nutritional Assessment (www.mna-elderly.com) and the Nutrition Screening Initiative (www.aafp.org/x16081.xml), which was created by the American Academy of Family Physicians with the American Dietetic Association, are also freely available to be used for nutritional assessment of elderly patients.
The first priority in managing weight loss is to systematically identify and treat the underlying causes (Fig. 1). Treatment of unintentional weight loss often requires enabling access to good nutrition, and several important nonpharmacologic strategies can be implemented to prevent or treat malnutrition and enhance food intake (Table 3). Factors such as poverty, poor dental health, difficulty in chewing or swallowing, vision or hearing loss, arthritis, stress (e.g., illness or death of a loved one) and unhappiness, which are associated with poor diet quality, should be targeted.52,80 It is therefore prudent to involve a dietitian and a social worker to assist with assessment and management, particularly in cases where an obvious organic cause has not been identified. A physiotherapist may help patients increase their amount of exercise, to thereby stimulate appetite and increase energy intake and muscle mass.68,74,75,76
The use of oral nutritional supplements, such as high-energy drinks, as a means of reversing weight loss and increasing food intake may sometimes, but not always, reverse weight loss.66,69,81,82 Counselling and encouraging patients to consume supplements in addition to their usual food intake rather than as a replacement of that intake is essential, since weight gain is confined to those who actually increase their energy intake.69,82 Advising patients to consume supplements between meals, rather than with the meal, may help minimize appetite suppression and facilitate increased overall intake.71 Although supplement use has been associated with short-term weight gain and improvements in biochemical, anthropometric and quality-of-life parameters in a number of trials, long-term beneficial effects on health, ability to function and survival in undernourished elderly people are yet to be consistently demonstrated.66,83 A systematic review showed a reduction in mortality among elderly patients who received protein-energy supplements, irrespective of whether they had weight loss.70
Many elderly people consume too little food to meet their nutritional needs,84,85 which puts them at risk of vitamin and mineral deficiency. A broad-spectrum vitamin and mineral supplement should be considered for people at risk of malnutrition or where improvements in food intake are not observed.85
In our experience, the evidence supporting any pharmacologic agent for the treatment of weight loss is limited to mostly small, uncontrolled studies, and benefits are generally restricted to a small gain in weight without evidence of decreased morbidity and mortality or improved function and quality of life. Most of these agents have significant side effects, particularly in frail elderly people, which limits their usefulness. Various pharmacologic agents, including orexigenic (appetite-stimulating) and anabolic medications, have been used to improve appetite or cause weight gain in subjects with weight loss. Only 4 have been studied in randomized trials (Table 4).
The synthetic progestational agent megestrol acetate is best associated with weight gain in well-designed, randomized trials in populations of patients with malignant disease or HIV infection.90,91 Evidence for its use with elderly people is limited.89,91,92,93 Ornithine oxoglutarate led to weight gain in one randomized trial but has not been studied in other trials.86 There are no randomized trials of either cyproheptadine or dronabinol in elderly people with weight loss, although dronabinol has been studied in one trial involving patients with dementia who were refusing food.88 Both medications are associated with significant side effects, particularly central nervous system toxicity.88,94,95
Among anabolic agents, a 4-week randomized trial of human growth hormone in 20 undernourished elderly people demonstrated slightly faster weight gain and improved walking time in those receiving the hormone. After 4 weeks, between-group differences in weight were no longer statistically significant.87 Use of human growth hormone in other settings has been associated with increased mortality.96 Several small clinical studies or cross-over trials of androgenic agents have not shown that they lead to weight gain.95,97 Other pharmacologic approaches, such as anticytokine therapies, antileptin therapies and anti-inflammatory medications, are being investigated.95,98
Unintentional weight loss is common in elderly people and is associated with significant adverse health outcomes, increased mortality and progressive disability. The differential diagnosis is broad, ranging from reduced food intake to organic causes to psychological disorders. Medications may also contribute to weight loss, as may social or economic factors. Up to 1 in 4 elderly people with unintentional weight loss will have no obvious medical cause. In others, a limited set of initial symptom-oriented investigations may reveal the underlying causes. A variety of nonpharmacologic interventions may improve energy intake and lead to weight gain, whereas the role for pharmacotherapy remains limited.
This article has been peer reviewed.
Contributors: Shabbir Alibhai contributed to the article design, performed the primary systematic literature review, summarized the key studies, drafted significant portions of the paper and revised the article. Carol Greenwood and Hélène Payette contributed to the article design, performed supplementary literature reviews, drafted significant portions of the paper and critically revised the article for important intellectual content. All of the authors approved the final version of the paper.
Competing interests: None declared.
Correspondence to: Dr. Shabbir M.H. Alibhai, University Health Network, Rm. ES 9–407, 200 Elizabeth St., Toronto ON M5G 2C4; fax 416 595-5826; firstname.lastname@example.org