Bone biopsies were studied in 73 patients to determine if a two-site radioimmunometric assay for serum bone alkaline phosphatase (BAP), total serum alkaline phosphatase (ALP), serum intact parathyroid hormone (iPTH), hand X-rays, regional bone mineral density (BMD) measurements and parathyroid enlargement detected by ultrasonography could accurately predict renal osteodystrophy. In the patients studied 57 had hyperparathyroid bone disease, 4 mixed renal osteodystrophy, 3 adynamic bone disease, 1 osteomalacia and 8 normal histology. Serum BAP, ALP and iPTH correlated positively with mineral apposition rate, osteoblastic, osteoid and eroded surface. In the diagnosis of hyperparathyroid bone disease serum iPTH was the most sensitive investigation, detecting 81% of patients at a level > 100 pg/ml but with a specificity of only 66%. Serum BAP was more sensitive, 70% at a level of > 10 ng/ml, than serum total ALP, 30% at a level of 300 IU/l, with similar specificities, 92 and 100%, respectively. Ultrasound detection of an enlarged parathyroid gland had a sensitivity of 64% and a specificity of 100% for the diagnosis of hyperparathyroid bone disease. Hand X-rays had a poor sensitivity, 47%, but a high specificity, 92%, for the detection of hyperparathyroid bone disease. The majority of patients had regional BMD values within the normal reference range and this test was of poor discriminatory value. The non-invasive markers were unable to distinguish between patients with low turnover, mild hyperparathyroidism and patients with normal histology. In conclusion the measurement of serum iPTH is a useful screening tool for the detection of hyperparathyroid bone disease which can be confirmed by the finding of a raised serum BAP or parathyroid enlargement. For definitive diagnosis, however, the gold standard remains bone biopsy and at present one cannot recommend any non-invasive method as an adequate substitute.