Evidence that albumin is not a suitable marker of body composition-related nutritional status in elderly patients.

OBJECTIVE: Serum albumin has long been used in clinical practice as a marker of protein-energy undernutrition, but very few studies have focused on its relation with dual-energy X-ray absorptiometry-assessed lean mass measurements, which is the current reference method in routine for body composition-related nutritional status. Serum albumin concentration is also affected by non-nutrition-related factors, and there is published evidence on the relation between serum albumin concentration and morbidity/mortality in the elderly. This study was designed to examine the relationship between serum albumin and lean mass and nutrition-related risk of morbidity/mortality in geriatric patients. Our objective was to clarify whether serum albumin in geriatric patients is a marker of body composition-related nutritional status, risk of morbidity/mortality, neither, or both.

METHODS: This prospective study enrolled 125 elderly patients hospitalized in a rehabilitation unit [83.8 (SD 7.7) y]. Subjects were evaluated for serum albumin concentration and nutritional status at admission [body mass index, lean mass, appendicular skeletal muscle mass index, and body cell mass index (calculated as the ratio of the mass to the square of the height), evaluated by dual-energy X-ray absorptiometry combined with bioelectrical impedance analysis]. Outcome scores were assessed 6 mo later, taking into account complications (pressure ulcers and/or infections) and 6-mo mortality.

RESULTS: Serum albumin concentration was not correlated with the lean mass, appendicular skeletal muscle mass, or body cell mass indexes. Serum albumin concentration was, however, correlated with outcome score (r = 0.22, P = 0.02).

CONCLUSION: This study clearly demonstrates that albumin is not suitable as a marker of body composition in elderly patients.