Subclinical hypervitaminosis A causes fragile bones in rats.

Excessive intake of vitamin A has been associated with an increased risk of hip fracture in humans. This finding has raised the question of whether long-term intake of relatively moderate doses ("subclinical" hypervitaminosis A) contributes to fracture risk. Although it has been known for more than half a century that toxic doses of vitamin A lead to spontaneous fractures in rats, the lowest intake that induces adverse effects is not known, and the result of exposure to excessive doses that do not cause general toxicity has been rarely investigated. In this study, mature female rats were fed a standard diet with 12 IU vitamin A/g pellet (control, C), or standard diet supplemented with either 120 IU ("10 x C") or 600 IU ("50 x C") vitamin A/g pellet for 12 weeks. Fifteen animals were included in each group. The supplemented diets correspond to a vitamin A intake of approximately 1800 IU/day and 9000 IU/day, respectively. The latter dose is about one third of that previously reported to cause skeletal lesions. At the end of the study, serum retinyl esters were elevated 4- (p < 0.01) and 20-fold (p < 0.001) and the total amount of liver retinoid had increased 3- (p < 0.001) and 7-fold (p < 0.001) in the 10 x C and 50 x C group, respectively. The animals showed no clinical signs of general toxicity, and there were no significant bone changes in the 10 x C group. However, in the 50 x C group, a characteristic thinning of the cortex (cortical area -6.5% [p < 0.001]) and reduction of the diameter of the long bones were evident (bone cross-sectional area -7.2% [p < 0.01] at the midshaft and -11.0% [p < 0.01] at the metaphysis), as measured by peripheral quantitative computed tomography. In agreement with these data and a decreased polar strength strain index (-14.0%, p < 0.01), the three-point bending breaking force of the femur was reduced by 10.3% (p < 0.01) in the 50 x C group. These data indicate that the negative skeletal effects appear at a subchronic vitamin A intake of somewhere between 10 and 50 times the standard diet. This level is considerably lower than previously reported. Our results suggest that long-term ingestion of modest excesses of vitamin A may contribute to fracture risk. Copyright 2002 by Elsevier Science Inc.