Abstracts

Rationale: The prevalence of low bone mineral density (BMD) is greater in persons with epilepsy (PWE) than in the general population. Prior research has suggested this is due to low vitamin D levels (VitD). However, no large population-based studies have looked to determine if VitD levels correlate with low BMD. VitD levels and BMD were retrospectively reviewed in PWE to determine if there is a correlation between low VitD levels and low BMD. Methods: Total serum 25-hydroxy-vitamin D (D2 plus D3: 25-OHD) was measured by liquid chromatography-tandem mass spectroscopy. Using the Marshfield Clinic electronic medical record we retrospectively compared 25-OHD in PWE with a large unselected dataset of 25-OHD results performed at Marshfield Laboratories (control group) and within the cohort of PWE sorted by BMD Z-scores. BMD was performed by dual X-ray absorptiometry (DXA: GE-Lunar, Madison, WI) and reported by lowest Z-score of the lumbar spine, total hip, femoral neck or 1/3rd radius in accordance with ISCD standards. Subjects were excluded for age < 25 years and anatomical abnormalities that rendered DXA uninterpretable. Results: (Figure 1, Table 1): In PWE (n =480), VitD levels were significantly lower when compared to the general population (n = 19, 895). Twenty-one patients (4.4%) had severe VitD deficiency (<10 ng/ml CI 95%, 2.7-6.6%), 126 (26.3%) had moderate deficiency (<20 ng/ml, CI 95%, 22.4-61.1%) and 271 (56.6%) had mildly reduced VitD (<30 ng/ml, CI 95%, 52.0-61.1%). The results indicate that our PWE have significantly lower VitD levels than the general population. In patients with both VitD and BMD studies (n =351), 119 (33.9%) had Z scores < or = to -2 (osteoporosis), 119 (33.9%) had scores between -2 and -1 (osteopenia) and 113 (32.2%) had scores > or = to -1 (normal). Of those PWE with osteoporosis, the mean VitD level was 27.3 (± 11.7). The osteopenic groups mean VitD level was 30.9 (± 13.8). The mean VitD level in PWE with normal BMD was 28.6 (± 12.7). The reduction in bone mineral density was not correlated with VitD levels (p > 0.05). Conclusions: This is the first large, population-based study to assess the correlation between 25-OHD and BMD Z-scores in PWE. While 25-OHD in PWE are significantly lower than in our controls, our population of PWE have better 25-OHD than generally reported in other studies (only 6% of our patients are defined as severe deficiency). We found no correlation between BMD Z-score and 25-OHD. Issues related to vitamin D replacement, potential causes of bone loss in PWE and risk factors for low BMD in PWE will be discussed.