et al., 1974). Adapting the work of Lossy et al. (1951), Sauberlich and colleagues (1974) suggested a reference value of 1.4 mg or more for the normal 4-hour urinary excretion of riboflavin after a 5-mg load.

The change in slope of urinary excretion of riboflavin after a load test is an especially useful status assessment. Caution in interpretation is needed because the size of the test dose, method of administration, and method of calculating amount recovered have varied among studies. Moreover, the break point may reflect not only tissue saturation but also renal threshold and solubility (compartment) effects (Sauberlich et al., 1974). Urinary riboflavin has been shown to increase under conditions causing negative nitrogen balance and with the administration of antibiotics and certain psychotropic drugs (e.g., phenothiazine) (McCormick, 1994).

In an early study (Horwitt et al., 1949), indices of carbohydrate metabolism (lactic and pyruvic acid concentrations) were measured in riboflavin-depleted subjects after a short period of exercise. Because no changes were observed, these do not appear to be promising indicators of riboflavin status.

Riboflavin status (low as assessed by EGRAC) has been related to certain site-specific cancers (e.g., esophageal) in areas of China (Merrill et al., 1991). However, randomized nutrition intervention trials in Linxian, China, indicated that a riboflavin and niacin combination, given for about 5 years, did not reduce total or cancer mortality (Blot et al., 1995).

Although lens opacities in humans have been associated with high glutathione reductase activity (with FAD) (Leske et al., 1995), evidence is insufficient for considering the use of risk of cataract as the basis for setting the Estimated Average Requirement (EAR).

Overall, greatest credence is given to status assessments that use more than one indicator, because the response variables indicate somewhat different aspects of riboflavin status. Many investigators have obtained data concurrently on several indicators of riboflavin status. Sauberlich and colleagues (1972) noted that, for the most