Obstructive sleep apnoea and bone health

To the Editor:

We thank Liguori et al. [1] for sharing their interesting study about patients suffering from obstructive sleep apnoea (OSA) and their tendency towards lower bone mineral density (BMD) in the lumbar spine and femur compared with control subjects matched for age, body mass index and physical activity. The researchers defined “osteopenia” as a T-score value < −1 sd and “osteoporosis” as a T-score value < −2.5 sd. They also suggested that OSA could be detrimental to BMD, resulting in osteopenia and osteoporosis. However, there are some concerns with this study.

First, the mean±sd age of the subjects included in this study was 51.17±11.82 years in the OSA population and 51.10±11.68 years in the control group, which means that many of the subjects were <50 years old. The author classified these subjects according to T-score, which is the BMD value compared with a healthy subject of the same sex in who is at peak BMD. However, according to the recent consensus of the International Society for Clinical Densitometry [2], when researching males <50 years old, a Z-score should be used, representing a value that can be compared with those of subjects matched for age and sex. A Z-score of −2.0 or lower is defined as “below the expected range for age”. We suggest that the authors should divide the subjects into two groups, older than and younger than 50 years of age, and use T- or Z-scores in the different age groups, respectively. In addition, the authors should not use “osteopenia” or “osteoporosis” to define subjects <50 years old.

Moreover, fracture is the most serious consequence of osteoporosis. BMD might be used to predict fractures but the sensitivity is very low. To better understand the effect of OSA on bone health, fracture rate might be a more useful marker. A previous study showed that hypoxia during sleep might be a risk predictive factor for falls and fractures in elderly men [3]. The result of Liguori et al. [1] suggests that OSA might be a risk factor for predicting fractures, in addition to low BMD. However, this hypothesis should be confirmed using a clinical study.