In dialysis centers using reverse osmosis-treated water but not restricting A1(OH)3 administration, a high prevalence of histological aluminum bone disease has been reported. To assess whether this is also the case in our center where A1(OH)3 intake has always been restricted and even completely given up after 1980 thanks to high doses of CaCO3, we reviewed 42 bone biopsies performed between 1975 and 1985 in patients dialyzed for a mean duration of 56 months. Seventeen of these patients had been dialyzed before 1978 with softened water moderately contamined by aluminum, 15 had always been dialyzed with reverse osmosis-treated water and 10 had been exclusively treated by hemofiltration. The prevalence of aluminum bone disease in the whole population was 9.5% (4 patients) and consisted only of adynamic bone disease, osteomalacia being totally absent. When the patients dialyzed with aluminum-contaminated water were excluded as well as 1 diabetic patient who had taken A1(OH)3 for 1.5 years the prevalence of aluminum bone disease was null in this population. When the whole population is considered the prevalence of the other types of bone disease was 76% for osteitis fibrosa and 14.5% for a non-aluminic adynamic bone disease (6 cases). These latter cases differed from the osteitis fibrosa group only by a relative hypoparathyroidism not explained by higher plasma concentrations and higher oral cumulative doses of calcium, magnesium and aluminum or by lower plasma concentrations of phosphate and bicarbonate. None had previous parathyroidectomy, one had an unsuccessful transplantation and one was diabetic. Iron overload was excluded by negative Perls staining.(ABSTRACT TRUNCATED AT 250 WORDS)