Only a small percentage of dietary manganese is absorbed. Absorbed manganese is excreted very rapidly into the gut via bile (Britton and Cotzias, 1966; Davis et al., 1993). Most estimates of absorption have been based on whole body retention curves at approximately 10 to 20 days after dosing with 54Mn. Using this method, Finley and coworkers (1994) estimated absorption from a test meal containing 1 mg manganese to be 1.35 ± 0.51 percent (standard deviation [SD]) for men and 3.55 ± 2.11 percent (SD) for women. From a test meal containing 0.3 to 0.34 mg of manganese, Davidsson and coworkers (1988) found the retention of 54Mn to be 5.0 ± 3.1 percent (SD) 10 days after administration to young adult women. Turnover of orally administered 54Mn was much more rapid after oral administration than after intravenous administration (Davidsson et al., 1989b; Sandstrom et al., 1986). Furthermore, absorption of manganese after 30 weeks of supplementation was 30 to 50 percent lower than had been observed in nonsupplemented subjects (Sandstrom et al., 1990). Some studies indicate that manganese is absorbed through an active transport mechanism (Garcia-Aranda et al., 1983), but passive diffusion has been suggested on the basis of studies indicating that manganese absorption occurs by a nonsaturable process (Bell et al., 1989).
Manganese is taken up from the blood by the liver and transported to extrahepatic tissues by transferrin (Davidsson et al., 1989c) and possibly α2-macroglobulin (Rabin et al., 1993) and albumin (Davis et al., 1992). Manganese inhibited iron absorption, both from a solution and from a hamburger meal (Rossander-Hulten et al., 1991). 54Mn has a longer half-life in men than in women (Finley et al., 1994). A significant negative association between manganese absorption and plasma ferritin concentrations has recently been reported (Finley, 1999). Serum ferritin concentrations differ in men and women (Appendix Table G-3); therefore, a major factor in establishing manganese requirements may be gender.
Manganese is excreted primarily in feces. Urinary excretion of manganese is low and has not been found to be sensitive to dietary manganese intake (Davis and Greger, 1992). Urinary excretion in a balance study of five healthy men varied from 0.04 to 0.14 percent of their intake, and absolute amounts in the urine decreased during the depletion phases of the study (Freeland-Graves et al., 1988). Therefore, potential risk for manganese toxicity is highest when bile excretion is low, such as in the neonate or in liver disease (Hauser et al., 1994). Plasma manganese concentrations can be-