insulin infusion, but during the second hour of insulin infusion, plasma potassium concentration continued to decline at the lowest insulin doses but began to rise at the highest insulin dose levels. This finding suggests the presence of a regulatory mechanism that influences insulin-mediated alterations in plasma potassium. The effect was not influenced by β-adrenergic blockade or aging (Minaker and Rowe, 1982; Rowe et al., 1992). These studies suggest that during aging, hormonal regulation of extrarenal potassium homeostasis remains normal.

Summary. In summary, for children, the AI was extrapolated from the adult AI based on energy intake. Older adults consume less energy than younger adults; however, because of the increased risk of elevated blood pressure with aging, the potassium need may be greater, and is thus not adjusted down for older adults. Because of the lack of evidence to suggest that the requirement for potassium differs in apparently normal, healthy older adults and the elderly compared with that of younger individuals, the AI is set at the same level of intake as for young adults.

Still, the AI does not apply to individuals with medical conditions or who are taking drugs that impair potassium excretion because of the potential for serious adverse effects on the heart from hyperkalemia (see later section, “Special Considerations”). Older individuals more commonly have such conditions or take such drugs and hence are at greater risk of hyperkalemia.

Accretion. There is little information on body potassium stores during pregnancy. The few available estimates range from cumula-