Alcohol‐induced Osteonecrosis – dose and duration effects

Abstract

We read with great interest the paper by Ikemura et al. published in the August issue of the International Journal of Experimental Pathology (Ikemura et al. 2011). In their study, the authors used a model of alcohol-induced osteonecrosis in rabbits to evaluate the morphological changes in bone marrow fat cells and the changes in the serum lipid levels. Animals received either 15 ml/kg/day (LDA: low-dose alcohol) or 30 ml/kg/day (HDA: high-dose alcohol) of a solution containing 15% ethanol for intragastrically for 4 weeks, whereas a control group received a physiological saline solution. Results were observed at 6 weeks after the beginning of the treatment (i.e. 2 weeks after withdrawal). As these results are difficult to reconcile with aspects of current understanding of alcohol effects on bone metabolism, we would like to comment on their protocol and their results in perspective. Regarding the quantity of alcohol used to induce osteonecrosis, the animals thus ingested either 2.25 g/kg/day (LDA) or 4.5 g/kg/day (HDA) of alcohol. In comparison, Wang et al. (2008) used a dose of 9.2 g/kg/day (i.e. more than twice the amount used in the HDA group) to induce objectively recognizable alcohol-induced osteonecrosis in rabbits, with concomitant marked marrow fat cell hypertrophy and proliferation, thinner and sparse trabeculae, diminished hematopoiesis and increased empty osteocyte lacunae (Wang et al. 2008). In the same manner, Broulik et al.(2010) used a dose of 7.22 g/kg/day of ethanol (corresponding to a consumption of 1 l wine or 2.5 l of 12 ° beer in male adults) to induce bone metabolism disturbances in rats (Broulik et al. 2010). In animal models there may be a cause and effect relationship between alcohol consumption and osteonecrosis of the femoral head (Hirota et al. 1993), but for osteonecrosis to develop in humans, the alcohol exposure threshold is approximately 150 l of 100% ethanol, at a consumption rate of 400 ml or more of absolute ethanol weekly (Cruess 1986; Jones 1994). Therefore, we believe that it is difficult to compare rabbit, rat and human studies with regard to the quantity of alcohol used to induce bone disturbances. Bearing this in mind, we suggest that the lack of osteonecrotic lesions in both LDA and HDA groups might be linked to the fact that the doses of alcohol ingested over 4 weeks was too low a dose of alcohol. We believe that the authors need to justify and discuss their choice of dosage. In addition, the short duration of alcohol treatment (4 weeks) is surprising. The authors do not give an explanation for this duration of alcohol treatment. In comparison, alcohol was administered intragastrically to rabbits for one to 6 months, and changes in lipid metabolism were observed only after 2–3 months of treatment (Wang et al. 2003). In another study, mice were treated with alcohol daily for up to 10 months and significant modifications of lipid metabolism appeared at 6 months, with significant increase in largest fat cell size detectable at 10 months (Wang et al. 2008). Finally, animals were treated with alcohol for 4 weeks, and the results were observed at 6 weeks after the beginning of the treatment, therefore after 2 weeks of withdrawal. The authors do not discuss this aspect of the protocol and its relevance in the aetiology of induced osteonecrosis. A possible explanation is to mimic the clinically relevant symptoms of alcohol withdrawal (Schuckit 2009). However, enlightenment about the impact of this withdrawal on osteonecrosis development would be welcome. Thus, without affecting the reported results, we suggest that the findings are difficult to interpret based upon current understanding of alcohol effects on bone metabolism and that there are protocol-related issues, including dose, duration and eventual withdrawal, which might account for the observations reported.