There is an association between bone aluminum (Al) accumulation and dialysis-associated osteomalacia (OM). To study whether Al is pathogenic in OM, quantitative bone histomorphometry was done in six dogs before (Bx 1) and after (Bx 2) 3 to 5 weeks of intravenous Al administration (1 mg Al /kg/day). Bone Al was determined by histochemical and chemical methods. The percent osteoid rose from 2.8 +/- 0.8 to 7.0 +/- 4.3% (mean +/- SD), P less than 0.05, and osteoid width increased from 5.7 +/- 0.6 to 8.0 +/- 1.2 mu, P less than 0.01, after Al. Bone Al rose from 1.3 +/- 1.6 to 94.0 +/- 19.0 mg/kg after Al, and the severity of OM, expressed as either percent forming surface or percent osteoid, correlated with bone Al measured histochemically and expressed as either percent surface or percent area of trabecular bone staining for Al (r = 0.85 - 0.90, P less than 0.01). Poor tetracycline uptake (six dogs), which indicates impaired mineralization, and little or no separation of tetracycline labels (four dogs) were noted at Bx 2; thus, bone apposition and formation rates were below the limits of detection. Resorptive surface did not change but trabecular volume, expressed as percent of tissue volume, fell from 22.1 +/- 3.0 to 17.1 +/- 1.4%, P less than 0.05. Serum levels of 1,25(OH)2D fell from 26.8 +/- 9.1 to 4.5 +/- 5.5 pg/ml after 17 days of Al; serum 25(OH)D levels were unchanged. These data indicate that Al can cause OM and that its severity correlates with the bone Al content.2 +