It is known that the intake of aluminum-containing antacids is related to increases in serum concentration and the urinary output of aluminum, and that silicon intake as orthosilicic acid may facilitate renal excretion of aluminum. However, how either the amount of antacid ingested or the therapy period affects these and the subsequent interrelationship of aluminum and silicon has not been fully investigated. In addition, factors that affect the solubility of aluminum hydroxide need further clarification. A study of 122 patients with dyspepsia who were receiving aluminum hydroxide therapy and 144 healthy controls confirmed that, in the patients, serum and urine aluminum (0.39 0.21 and 2.02 2.0 lmol/L, mean 1 SD, respectively) were slightly but significantly increased (P < 0.05) compared with the controls (0.21 0.13 and 1.55 0.89). However, there was no evidence of any association between these changes and the amount ingested or the therapy period. Urine silicon although increased in the patients (716 488 lmol/L) compared with the controls (472 333 lmol/L) was not aected by the amount or time of therapy but was just signi®cantly correlated with the urine (r = +0.19, P = 0.03) and serum aluminum (r = +0.13, P = 0.05) in the combined groups. Thus, the renal excretion of silicon and aluminum may in some part be interdependent. Studies into factors that aect the solubility of aluminum from aluminum hydroxide indicated that at a pH > 3.5 the solubility of aluminum reduced signi®cantly. However, addition of 5 or 20 mmol/L phosphate had no signi®cant eect. With a synthetic food, solubility of aluminum was markedly increased even at the elevated pH 5.5. Thus, the pH, as well as the chemical complexity of the mixture, is an important factor in aecting solubilization of aluminum. The potential of increased exposure to aluminum could be dependent on co-ingestion with food components and may explain the increased risk to renal failure patients on aluminum-containing phosphate binders. J. Trace Elem. Exp. Med. 15:9À19,