Multiple forms of alkaline phosphatase in plasma of hemodialysis patients.

We used quantitative assays to measure the activity of the bone, liver, and intestinal forms of alkaline phosphatase in plasma in 75 patients with endstage chronic renal failure undergoing hemodialysis. The results were correlated with radiological and other biochemical indices of bone disease and with biochemical indices of liver disease. The total activity of alkaline phosphatase in plasma increased in 28 patients. In 10 of these patients, nine of whom had increased activity of gamma-glutamyltransferase in plasma, the increase in total activity of alkaline phosphatase was from the liver isoenzyme alone (nine patients) or from the liver and bone isoenzymes together (one patient). Intestinal alkaline phosphatase in plasma, although greater than 23 U/L in eight patients, was solely responsible for the increase in total alkaline phosphatase in one patient (who had normal gamma-glutamyltransferase). Bone alkaline phosphatase in plasma was increased in 25 patients, seven of whom had normal total alkaline phosphatase, and was closely correlated (r = 0.78) with osteocalcin concentration in plasma, which was increased in a much greater proportion of patients (99%). Both total and bone alkaline phosphatase were correlated with parathyrin in plasma (r = 0.46 and 0.50, respectively) and with osteocalcin (r = 0.60 and 0.78, respectively). Osteocalcin and bone alkaline phosphatase, but not parathyrin, decreased with age, implying that the skeletal response to parathyrin may be age dependent. In patients with increased total alkaline phosphatase undergoing hemodialysis, the concurrent measurement of gamma-glutamyltransferase may help identify whether the enzyme increase originates from the liver or bone, but this approach wrongly identified the source of the increase in three of 28 patients. Therefore, we recommend a separate measurement of the bone isoenzyme of alkaline phosphatase.