To the Editor: In the elderly, anorexia could be a direct consequence of many chronic diseases, but it was not until 1988 that Morley and Silver defined the syndrome “anorexia of aging” as a physiopathological decline in food intake with aging that predisposes older persons to develop protein-energy malnutrition when certain psychosocial factors and diseases are concomitant.1 This decline has also been seen in disease-free ambulant older persons.2 The relationship between anorexia of aging and malnutrition has been largely reviewed,1,3,4 but the prognostic relevance of this symptom on mortality independent from biochemical indices of nutrition has never been studied. This preliminary study analyzes whether anorexia of aging could be a predictor of mortality, independent of indices of malnutrition, illness severity, comorbidity, and cognitive and functional status. All consecutive patients discharged from a Geriatric Evaluation and Rehabilitation Unit (Richiedei Medical Center, Palazzolo s/O—Brescia, Italy) were considered. Patients underwent a multidimensional assessment to evaluate cognitive, mood, functional, nutritional, and somatic health status. Particular attention was paid to anorexia, which was detected by asking relatives upon admission, “During the last three months, was food intake decreased for whatever the cause?” from the Mini Nutritional Assessment (MNA).5 This sample question was submitted separately to 20 pairs of caregivers to evaluate its validity. The first caregiver's answer was compared with that of the second one, and interrater reliability was analyzed using the Spearman correlation coefficient (rho=0.79, P<.001). Moreover, a week after the admission, the first relative was asked again, and the answer was compared with the previous one (test-retest) (rho=0.91, P<.001). Because current criteria for diagnosis of anorexia of aging do not exist, to study anorexia of aging (not related to disease), patients with the highest severity of all diseases (Individual Disease Severity Index6 (IDS IV)) or with moderate severity (IDS III) of conditions directly or indirectly associated with dyspepsia or poor food intake (cancer, cardiopulmonary, gastrointestinal, hepatobiliary, and renal diseases) were excluded. After discharge, data on mortality were collected by telephoning the family or the Mortality Register. Patients were observed for a median of 10.5 months (25th–75th percentiles=6.0–15.5 months). The final sample consisted of 316 patients; 50 (15.8%) were affected by anorexia. In Table 1, the characteristics of the selected population are shown; patients with anorexia of aging are older and have a greater risk of malnutrition (MNA), and the percentage of subjects with weight loss is higher. According to cognitive and functional status, the two groups were not different. Taking into account some biochemical indices, anorectic patients had a lower level of serum albumin and transferrin, but total cholesterol and ferritin were similar. The prevalence of somatic chronic diseases was analogous in the groups. Mortality of patients with anorexia was higher than patients without this symptom (16.0% vs 6.4%, P<.05). The relationship between anorexia and mortality was analyzed using the model of Cox regression analysis, including age, Mini-Mental State Examination, depression (Geriatric Depression Scale score ≥9/15), Barthel index, gastrointestinal and hepatobiliary diseases, Burden of Disease7 (sum of IDS score of 16 chronic diseases), MNA, weight loss (≥3 kg/3 months), serum albumin, and transferrin, which are possible mediators of the link between poor food intake and survival. Anorexia (hazard ratio (HR)=2.9 95% confidence interval (CI)=1.1–7.4) and Barthel index (HR=1.1 95% CI=1.0–1.2) were found to be independent predictors of mortality. The risk of mortality was three times higher in the anorectic subjects, after adjusting for possible causes or the direct consequence of poor food intake. The advantage of investigating food intake reduction is supported by the fact that it is a simple self-reported or caregiver-reported symptom. Recently, clinicians have been urged to regularly evaluate nutritional risks using questionnaire such as the MNA,1 and the need for simple, reliable, and valid tools, based on self-report and applicable in a variety of settings has been underlined.8 A clinical evaluation of nutritional risk in the multidimensional assessment may promote therapeutic interventions by identifying and treating underlying diseases, increasing the quality of nursing feeding assistance, and planning early nutrition rehabilitation and education programs for caregivers,3,9,10 although the efficacy of these interventions is controversial. Perhaps the detection of early risk factors of malnutrition, such as anorexia, could increase the probability of success of nutritional interventions in target elderly subjects. In conclusion, we found that anorexia (not directly associated with chronic diseases) is an easy way to identify subjects particularly likely to have higher mortality independent of malnutrition.
No takes yet. Share an insight, caveat, or question.
Cornali et al. (2005) studied this question.