Magnesium

BACKGROUND

Magnesium is a cofactor for more than 300 enzyme systems (Wacker & Parisi 1968) and is involved in both aerobic and anaerobic energy generation and in glycolysis, either directly as an enzyme activator or as part of the Mg-ATP complex. Magnesium is required for mitochondria to carry out oxidative phosphorylation. It plays a role in regulating potassium fluxes and in the metabolism of calcium (Al-Ghamdi et al 1994, Classen 1984, Waterlow 1992, ). The human body contains about 760 mg of magnesium at birth and 25 g in adulthood (Forbes 1987, Schroeder et al 1969, Widdowson et al 1951). Just over half the body's magnesium is found in bone, where it forms a surface constituent of the hydroxyapatite mineral component, and a further third is found in muscles and soft tissues (Heaton 1976, Webster 1987). The intracellular concentration is about ten times that of the extracellular fluid.

Magnesium is widely distributed in the food supply in both plant and animal foods. Most green vegetables, legumes, peas, beans and nuts are rich in magnesium, as are some shellfish and spices. Most unrefined cereals are reasonable sources, but highly refined flours, tubers, fruits, oils and fats contribute little. Between 50% and 90% of magnesium in breast milk or infant formula is absorbed (Lonnerdal, 1995, 1997). In adults on conventional diets, the efficiency of absorption varies greatly with magnesium content (Seelig 1982, Spencer et al 1980) ranging from 25% on high magnesium diets in one study to 75% on low magnesium diets (Schwartz et al 1984). The homeostatic capacity of the body to adapt to a wide range of intakes is thus high (Abrams et al 1997, Sojka et al 1997).

Magnesium is absorbed in the duodenum and ileum by both active and passive processes (Greger et al 1981). High fibre intakes (40–50 g/day) lower magnesium absorption, probably because of the magnesium-binding action of the phytate phosphorus associated with the fibre (Kelsay et al 1979, McCance & Widdowson 1942a,b). There is no consistent evidence that moderate increases in calcium, iron or manganese affect magnesium balance (Abrams et al 1997, Andon et al 1996, Lonnerdal 1995, Sojka et al 1997). However, high intakes of zinc at 142 mg/day reduce absorption (Spencer et al 1994b). Protein may also influence magnesium absorption. When protein intake is less than 30 g/day (Hunt & Schofield 1969), magnesium absorption decreases. When protein intake is greater than 94 g/day, renal magnesium excretion may increase (Mahalko et al 1983), although adaptation may occur.

The kidney plays a central role in magnesium homeostasis through active reabsorption that is influenced by the sodium load in the tubules and possibly acid-base balance (Quarme & Disks 1986). High dietary calcium intake (about 2,600 mg/day) with high sodium intake enhances magnesium output (Greger et al 1981), contributing to a shift to negative magnesium balance (Kesteloot & Joosens 1990, Quarme et al 1986).

Pathological effects of primary nutritional deficiency of magnesium occur only rarely in humans, unless low intakes are accompanied by prolonged diarrhoea or excessive urinary loss. The body is generally protected by the lability of serum magnesium. Most of the early signs of deficiency are neurologic or neuromuscular defects (Shils 1969, 1988) that may develop with time into anorexia, nausea, muscular weakness, lethargy, weight loss, hyper-irritability, hyper-excitability, muscular spasms, tetany and finally convulsions.

Hypocalcaemia also occurs in moderate to severe magnesium deficiency. Some studies have indicated that low magnesium status may be a risk for postmenopausal osteoporosis (Abraham & Grewal 1990, Reginster et al 1989, Sojka & Weaver 1995, Stendig-Lindberg et al 1993, Tucker et al 1995, Yano et al 1985), however others have not confirmed the link between low magnesium and risk of osteoporosis (Angus et al 1988, Freudenheim et al 1986). Sub-optimal magnesium status may be a factor in the aetiology of coronary heart disease and hypertension, but evidence is relatively sparse (Elwood 1994). Magnesium depletion has been shown to cause insulin resistance and impaired insulin secretion (Paolissa et al 1990), and magnesium supplements have been reported to improve glucose tolerance and insulin response in the elderly (Paolissa et al 1989, 1992).

Indicators used for estimating magnesium requirements have included serum magnesium, plasma ionised magnesium, intracellular magnesium, magnesium balance, estimates of tissue accretion in growth, magnesium tolerance tests and epidemiologic studies including meta-analysis. However, serum magnesium has not been properly validated as a reliable indicator of body magnesium status (Gartside & Glueck 1995). Plasma ionised magnesium may be an improvement on serum magnesium but requires further evaluation and the validity evidence for intracellular magnesium is limited. Magnesium balance is problematic if not carried out under close supervision, as magnesium in water can confound results, a factor that precluded the use of many early studies conducted in free-living situations or current studies where intakes were calculated, not analysed.

Accurate estimates of tissue accretion during growth throughout childhood are dependent on more extensive information about whole body mineral retention than are currently available, although there is some information for specific ages from cadaver data (Fomon & Nelson 1993, Koo & Tsang 1997). The magnesium tolerance test is an invasive procedure based on renal excretion of parenterally administered magnesium load. It is considered accurate for adults but not infants and children (Gullestad et al 1992, Ryzen et al 1985). The test requires normal renal handling and may be unreliable in diabetics or drug or alcohol users. It may also be affected by ageing of kidney tissue (Gullestad et al 1994). Epidemiological studies with meta-analysis may indicate relationships between magnesium intake and health outcomes.