Biomarkers are used to assess nutritional intakes and nutritional status because they are believed to provide more accurate estimates of intake/bioavailability of specific nutrients, especially where these are combined (as in most meals) or cooked to variable extent. There are also concerns about the validity of food composition tables when foods are known to vary over time (as with industrialized food processing) or cultivated in different soils or climatic conditions.
Although some researchers think of biomarkers as potential gold standards of dietary intake, there are good reasons to accept such a view only with thoughtful reservations (Weinstein et al., 2004). Subjects are well known to vary in metabolic pathways for reasons of sex, genetics, and previous nutrient exposures (Arab et al., 2003). These differences include age-related variations in absorption, the impact of coingestion of foods that may inhibit the absorption of others (e.g., green vegetables contain oxalic acid, which impairs the absorption of calcium). Likewise, renal excretion is subject to tight controls designed to remove unwanted metabolic products and excess nutrients but to retain needed nutrients. Urinary analyses, for these reasons, can be misleading as they indicate the net outcome of these diverse processes (including saturation) and are not directly correlated with intakes.
Biomarkers may be measured in a wide range of tissues that include plasma, serum, red blood cells, white blood cells, feces, urine, hair, nail, buccal cells, and a number of measures specifically designed to assess overall antioxidant capacity (Polidori et al., 2001), metabolic state (e.g., exhaled air), or the extent of DNA damage (Potischman, 2003a, 2003b). Some tissues seem especially relevant to aging studies (e.g., lipid content of red blood cell membranes; Hulbert et al., 2004), but it is best to plan to collect diverse types of samples before drawing any general conclusions. This is true in most biomarker studies; whenever possible, ensure that the chosen marker is examined or tested in at least two biological systems (Ilich et al., 2003). Before deciding upon any biomarker, it is always useful to discuss the choice with the laboratory where the tests will be conducted. This advice often includes appraisal of recent analytical developments, the competing interests of high reliability/validity of a specific measure when set against convenience and compliance. There are also helpful critiques of specific biomarkers, the principles of which can easily be applied to another biomarker (e.g., Ness et al., 1999). It is also essential to discuss all aspects of sampling, separation, labeling, storage, and how laboratory data are to be linked to the research database. Decisions on these and related matters should be recorded and circulated as a clear written statement of protocol.
Body composition studies in old people provide useful measures that discriminate between health and disease states (Fuller et al., 1996). This is a highly specialized research technique, requiring great care in its application and interpretation. Available methods include bioelectric impedance studies (Baumgartner, 2000; Aghdassi et al., 2001).
In nutritional gerontology, the issue of differences in cognitive ability in the assessment of dietary history/ nutritional status is rarely investigated. Studies are needed that investigate the advantages of biomarkers over retrospective recall in the presence/absence of cognitive impairment (Morris et al., 2002). Likewise, in the investigation of the role of nutrients in aging and age-related diseases, it is useful to place firmly among the aims of a project the needs of clinicians who might make use of the findings. For example, in clinical practice, it is commonplace to identify the concentration of vitamin B12/folate as being suboptimal and predictive of disease. It is uncommon for these values to be considered in the context of a relevant metabolite such as homo-cysteine, largely because uncertainties remain about the value of such measures and their availability. Nevertheless, these matters do interest clinicians, and the value of measuring a number of parameters linked to a pathophysiologic process is appealing and worth bearing in mind when choosing what might be measured.
Resources available include a series of papers published in 2003 as a supplement to The Journal of Nutrition (''Biomarkers of Nutritional Exposures and Nutritional Status'').
Was this article helpful?