FULL-LENGTH ORIGINAL RESEARCH
The association between BsmI polymorphism and bone mineral density in young patients with epilepsy who are taking phenytoin
Address correspondence to Kanitpong Phabphal, Neurology Unit, Department of Medicine, Faculty of Medicine, Prince of Songkla University, Hat Yai, Songkhla 90110, Thailand. E-mail: firstname.lastname@example.org
Purpose: This study sought to determine the association between BsmI polymorphism and bone mineral density, 25-hydroxyvitamin D, and parathyroid hormone levels in patients with epilepsy.
Methods: We recruited ambulatory young adults with epilepsy who were taking phenytoin. Data regarding demographics, basic laboratory studies, history of clinical epilepsy, parathyroid hormone, and vitamin D levels, as well as BsmI polymorphism of the vitamin D receptor (VDR) gene, were obtained. The bone mineral density (BMD) of the lumbar spine and left femur were measured using dual-energy x-ray absorptiometry.
Key Findings: Ninety-four patients were included in the study. BsmI polymorphism had a statistically significant lower T-score of the lumbar spine and left femoral neck than patients with wild-type VDR gene (p < 0.01 and p < 0.01, respectively). In addition, patients with BsmI polymorphism had a statistically significant lower z-score of the lumbar spine and left femoral neck than patients with wild-type VDR gene (p < 0.01 and p < 0.01, respectively). The 25-hydroxyvitamin D level in patients with wild-type genes was higher than in epileptic patients with BsmI polymorphism (p < 0.01 and p < 0.01, respectively). Parathyroid hormone level in patients with wild-type VDR gene or patients having BsmI polymorphism was not correlated with BMD at either site.
Significance: In patients with epilepsy taking phenytoin, having BsmI polymorphism was associated with lower BMD.
Epilepsy is a chronic condition that may affect individuals for years. The long-term use of antiepileptic drugs (AEDs) in patients who have seizures has been associated with alterations in bone mineral density (BMD). In comparison with the general population, patients with epilepsy have a six times greater risk of fracture, and the reduction in their bone density can be detected as early as 1–5 years after commencing treatment. The fracture risk of the epileptic population is approximately twice that of the general population, independent of seizure-related falls, whereas 35% of the fractures may be related to seizures (Petty et al., 2007). It is important to note that the fractures are associated with increased rates of morbidity and mortality (Germaine-Smith et al., 2011).
Previous studies on the bone health of patients with epilepsy have reported a significant variability in the prevalence of low BMD, ranging from 26% to 75% depending on ethnicity and/or country of origin; this may be due to genetic variability (Farhat et al., 2002; Pack et al., 2003; Lado et al., 2008; Phabphal et al., 2009). Moreover, the epileptic treatment is characterized by unpredictable efficacy and optimal dose response, and especially adverse drug reactions in individual patients. Treatment with phenytoin (PHT) was reported to be affected by genetic variability in drug metabolism >25 years ago (Löscher et al., 2009). For instance, reduced metabolism of PHT in people with certain cytochrome P450 (CYP)2C9 alleles was reported by Mamiya et al. (1998), and HLA-B*1502 was found to be associated with Stevens-Johnson syndrome/toxic epidermal necrosis induced by PHT (Hung et al., 2010). The principle of pharmacogenetics in the field of epilepsy is to test how the patient’s genotype might affect variation in response to AEDs among individuals—especially regarding the most important adverse reactions.
Vitamin D plays a central role in bone biology and mineral homeostasis. Vitamin D, by interacting with its receptors, plays an important role in calcium homeostasis by regulating bone cell growth and differentiation, intestinal calcium absorption, and parathyroid secretion. According to a report by Morrison et al. (1992), BsmI polymorphism has been associated with bone turnover, and bone density or fracture. This finding was confirmed by meta-analysis (Cooper & Umbach, 1996). Nevertheless, the studies of BsmI polymorphism of the vitamin D receptor (VDR) gene in patients with epilepsy who are taking PHT are limited. Only two studies have focused on BsmI polymorphism in patients with epilepsy (Tsukahara et al., 2002; Lambrinoudaki et al., 2011). However, the evidence pertaining to this issue is limited by confounding factors such as including several AEDs that affect BMD differently. Moreover, all of these studies consisted of a single measurement of BMD in the lumbar spine and small sample sizes.
This study was conducted with two objectives, affording us the opportunity to, first, determine the prevalence of low BMD in patients with a wild-type VDR gene and BsmI polymorphism and second, to evaluate the association between BsmI polymorphism and BMD, parathyroid hormone activity, and 25-hydroxyvitamin D level in 94 young among ambulatory adults with epilepsy who were taking PHT.
We performed a cross-sectional study on young ambulatory Thai adults with epilepsy who were living in Songkhla province or a neighboring province who had been treated with PHT monotherapy for at least 2 years and attended the general medical and neurologic clinics of Songklanagarind Hospital between October 2010 and March 2012. The inclusion criteria were the following: a Thai national with epilepsy living in Songkhla province or a neighboring province; age 15–50 years; using PHT for at least 2 years before enrollment; with regular menstruation (for women); with a stable weight status (over the previous 6 months); on a stable dosage of PHT (in the previous 9 months because one bone remodeling cycle takes approximately 3 months; thus, three remodeling cycles would have been completed); with no chronic medical illness other than epilepsy; taking no medication except antiepileptic(s); with an active daily life (performing activities of daily living without assistance); with no history of amenorrhea, hysterectomy, or oophorectomy; not consuming alcohol or tobacco; and with an adequate calcium intake (>800 mg/day). Patients taking other AEDs before inclusion, but who had stopped taking them at least 2 years before enrollment were also included. The exclusion criteria were the following: pregnancy; being unable to read and converse in the Thai language; having a significant disability such as mental retardation, ataxia, paresis, or other motor disability, learning disability, language disorder, hearing or visual disability, psychosis or psychiatric disease; having a significant medical disorder other than epilepsy known to affect bone metabolism such as hepatic, hematologic, rheumatologic, renal or gastrointestinal disorders, hyperparathyroidism, hyperthyroidism, hypogonadism, osteogenesis imperfecta, and previous fracture in the past year; and taking medications and/or supplements known to affect bone turnover such as glucocorticoids, bisphosphonates, thiazides, anticoagulants, gonadotropin-releasing hormone (GnRH) analogs, vitamin D or A, and steroids. Subjects who had a family history of osteoporosis were also excluded from the study. All of the patients completed a questionnaire covering the areas of age, duration of treatment, occupation, dietary intake (previous 7 days), and medication. All of them completed the questionnaire in one session. The variables from the patients were rechecked with the medical records and, in some cases, by contacting their physicians. The ethics review committee of the Prince of Songkla University Faculty of Medicine approved the study, and informed consent was sought and obtained from all of the patients.
The weight and height were measured and the body mass index (BMI) was calculated—body weight (kg)/height (m)2. A BMI in the range of 18.5–23.0 was considered normal, higher than 23.0 as overweight, and lower than 18.5 as underweight, in accordance with the 2000 World Health Organization (WHO) recommendations. To determine the baseline data, the weight and height were measured and blood samples were taken for albumin, calcium, and phosphate level determination on a HITACHI 971 automatic analyzer during the sample-collection hospital visit. The 25-hydroxyvitamin D and parathyroid hormone levels were determined and the BMD of the lumbar spine (L1–L4) and left femur were measured by dual-energy x-ray absorptiometry (DEXA).
Serum 25-hydroxyvitamin D
The serum 25-hydroxyvitamin D level was measured in duplicate by the high-performance liquid chromatography (HPLC) method with UV detection (Chromsystem CLC 200: Munchen, Germany). The intraassay variability was <1.48% within a concentration range of 64.74–68.70 ng/ml (normal range 30–80 ng/ml). Vitamin D deficiency was defined as a 25-hydroxyvitamin D level of <20 ng/ml and insufficiency as a level of 20–29 ng/ml (Holic, 2007).
Serum parathyroid hormone
The intact parathyroid hormone level was measured in duplicate by electrochemiluminescence immunoassay (ECLIA; Molecular Analytics E170, Mannheim, Germany). The detection limits were 1.20–5,000 pg/ml and the intraassay variability was <0.6% within a concentration range of 52.37–53.64 pg/ml (normal range 15–65 pg/ml).
The BMDs of the lumbar spine (L1–L4) and left femur were measured by DEXA (DXP MD. Software version: 4.6; Lunar Corporation, Madison, WI, U.S.A.) and reported as a T-score (the difference in standard deviation units between the measured bone density value and the peak bone density in the normal reference Japanese population as supplied by the manufacturer because no Thai reference was available) and the z-score (standard deviation [SD] from age-sex-specific score in the reference population). The areas selected for the determination of BMD were the neck of the left femur and the lumbar (L1–L4) vertebrae. The WHO defines normal BMD as a T-score >−1.0, osteopenia as a T-score between >−2.5, and <−1.0, and osteoporosis as a T-score ≤−2.5. All of the patients who had measurements taken at both sites and again at the same sites on the second visit were included in the analysis.
Vitamin D receptor gene. The genomic DNA was isolated from leukocyte nuclides using the QIAamp DNA Mini Kit (Qiagen, Hilden, Germany). The real-time polymerase chain reaction (PCR) probe was performed in a strip tube with a reaction volume of 20 μl, containing 2 μl of SsoFast Probes Supermix “BioRad” (Hercules, CA, U.S.A.) (ready to use reactionmix) 0.15 μm of the sense, antisense and probe primers for VDR and 150 ng/μl of genomic DNA. The primer and probe sequences for real-time use were VDR sense 5′GAGCCCAGTTCACGCAAGAG3′, VDR antisense 5′GGGGGGATTCTGAGGAACTAGATA3′, VDR 6-FAM-5′ACAGGCCTGCGCATTCCCAAT3′-TAMRA for wild-type, and VDR HEX-5′ACAGGCCTGCACATTCCCAAT3′-TAMRA for mutation (Kim et al., 2001).
The real-time PCR probe amplification to detect VDR was performed using the CFX 96 Connect Realtime-PCR (Bio-Rad Laboratories) with an initial denaturation at 95.0°C for 2 min, followed by 35 cycles of denaturation at 95.0°C for 10 s, 35 cycles of annealing at 60.0°C for 30 s, and final extension at 95.0°C for 10 s. The analysis was performed using the CFX Manager Software 2.1 (Life Science Research, Hercules, CA, U.S.A.).
The real-time PCR HRM (polymerase chain reaction high-resolution melting) was performed in a strip tube with a reaction volume of 20 μl, containing 2 μl of SsoFast Evagreen Supermix “BioRad” (ready to use reactionmix), 0.15 μm of the forward and reverse primers, and 100 ng/μl of genomic DNA. The forward and reverse primer sequences used were 5′TAGGGGGGATTCTGAGGAACTA3′ and 5′AGTTTTGTACCCTGCCCGC3′, respectively. The amplification to detect VDR was performed using CFX 96 Connect Realtime-PCR (Bio-Rad Laboratories, Hercules, CA, U.S.A.) with an initial denaturation at 95.0°C for 3 min, followed by 40 cycles of denaturation at 95.0°C for 10 s, annealing at 60.0°C for 15 s, and a final extension at 95.0°C for 10 s, at a Melt Curve between 75.0 and 95.0°C and increments of 0.2°C per 10 s. The HRM analysis employed the Precision Melt Analysis software 1.1. (Life Science Research) Eight microliter aliquots of the real-time PCR HRM product were digested with 5 U BsmI restriction enzymes at 65°C for 1 h (New England BioLabs, Ipswich, MA, U.S.A.), detected using 12% polyacrylamide gel electrophoresis run at 120 V for 2 h and 30 min, and confirmed by the sequence ABI 3130 Genetic Analyzer (3130 Collection Software, Life Technologies, Grand Island, NY, U.S.A.).
The polymorphism was defined as in the study by Lambrinoudaki et al., 2011, BB (absence of restriction site on both alleles), Bb (heterozygous), and bb (presence of restriction on both alleles). The trough serum level for PHT was also examined.
Statistical analysis. The characteristics of the study patients were described in terms of mean and standard deviation for continuous variables, and number and percentage for categorical variables. Comparisons of continuous variables between two subgroups of subjects were performed using a two-tailed t-test. Correlation between 25-hydroxyvitamin D and BMD parameters in group with BsmI polymorphism were explored using Pearson product movement correlation coefficient. Statistical analyses were performed using Stata version 7.0 (Stata Statistical Software: version 7.0; College Station, TX, U.S.A.).
Basic demographic and clinical characteristics of the patients are shown in Table 1. Ninety-six patients were invited to participate in the study. Two patients declined because of personal reasons other than bone disorder. We enrolled 94 patients in the study. Ninety patients had never taken AEDs other than phenytoin. One patient had febrile convulsion and had taken a short course of benzodiazepine (<7 days per course), but had stopped taking benzodiazepine 10 years previously. Three patients had taken carbamazepine but developed serious skin reaction after 1 month and had stopped the medication for at least 6 years previously. All of them continued taking PHT for treatment of epilepsy. The mean age ± SD of all patients was 37.4 ± 9.4 years (range 15–50 years); 43 were male and 51 female. The 27 patients with BsmI polymorphism (homozygotes, heterozygotes) had a mean age ± SD of 39.7 ± 8.4 years (range 15–50 years); 12 were male and 15 female. The 67 patients who had wild-type VDR gene had a mean age ± SD of 36.4 ± 9.7 years (range 15–50 years); 31 were male and 36 female. The mean duration of PHT ± SD at the time of the study was 14.8 ± 11.8 years and 14.7 ± 10.3 years in epileptic patients with BsmI polymorphism and wild-type VDR gene (p = 0.98), respectively. In addition, there was no statistically significant difference in mean duration of epilepsy at the time of the study in patients with BsmI polymorphism (14.9 ± 11.7 years) and wild-type VDR gene (15.1 ± 10.3 years) (p = 0.92). The mean ± SD of 25-hydroxyvitamin D level was 34.1 ± 10.6 and 39.7 ± 9.9 ng/ml in patients with BsmI polymorphism and wild-type VDR gene, respectively (p = 0.02). The mean ± SD of parathyroid hormone level was 38.6 ± 13.9 and 41.5 ± 16.9 ng/ml in patients with BsmI polymorphism and wild-type VDR gene, respectively (p = 0.43). In addition, we found that patients with epilepsy who had BsmI polymorphism and wild-type VDR gene did not differ significantly in serum PHT level. The other basic demographics of our population are shown in Table 2.
Table 1. Baseline demographic, bone mineral density, and blood chemistry findings of the study population
|Age|| || |
| 15–25 years||10||2|
| 26–35 years||24||7|
| 36–45 years||21||9|
| 46–50 years||12||9|
|Sex|| || |
|Duration|| || |
| 2–5 years||12||8|
| >5–10 years||14||3|
| >10–20 years||26||7|
| >20 years||15||9|
|Duration of phenytoin treatment|| || |
| 2–5 years||12||8|
| >5–10 years||14||3|
| >10–20 years||26||7|
| >20 years||15||9|
|Phosphate level (mg/dl)|| || |
|Calcium level(mg/dl)|| || |
|25-Hydroxyvitamin D level (ng/ml)|| || |
|Parathyroid hormone level (ng/ml)|| || |
|Femoral neck T-score|| || |
| −2.5 to −1||5||5|
|Lumbar spine T-score|| || |
| −2.5 to −1||9||9|
| > −1||58||18|
Table 2. Clinical characteristics, mean levels of biochemical parameters, parathyroid hormone, and 25-hydroxyvitamin D levels, and BMD according to the genotype of the vitamin D receptor
|Age (years)||36.4 ± 9.7||34.1–38.8||39.7 ± 8.4||36.3–43||0.13|
|Duration of epilepsy (years)||15.1 ± 10.3||12.5–17.6||14.9 ± 11.7||10.2–19.5||0.92|
|Duration of phenytoin (years)||14.7 ± 10.3||12.2–17.2||14.8 ± 11.8||10.1–19.5||0.98|
|25-Hydroxyvitamin D (ng/ml)||39.7 ± 9.9||37.29–42.13||34.1 ± 10.6||29.89–38.23||0.02|
|Parathyroid hormone (ng/ml)||41.5 ± 16.9||37.40–45.65||38.6 ± 13.9||33.13–44.11||0.43|
|Phosphate (mg/dl)||3.35 ± 0.56||3.21–3.49||3.35 ± 0.49||3.16–3.54||0.98|
|Calcium (mg/dl)||9.12 ± 0.38||9.02–9.21||9.04 ± 0.38||8.89–9.19||0.36|
|Femoral neck z-score||0.25 ± 0.95||0.02–0.48||−0.43 ± 0.15||−0.74–0.12||<0.01|
|Lumbar spine z-score||0.43 ± 1.09||0.17–0.70||−0.36 ± 0.84||−0.69 to −0.02||<0.01|
|Femoral neck T-score||0.22 ± 1.17||−0.07–0.50||−0.57 ± 0.92||−0.93 to −0.21||<0.01|
|Lumbar spine T-score||0.56 ± 1.08||0.29–0.82||−0.56 ± 0.92||−0.92 to −0.19||<0.01|
Bone mineral density
The distribution of the T-scores and z-scores at the lumbar spine and left femur are shown in Tables 1 and 2. The mean T-score ± SD and mean ± SD at the lumbar spine in patients with wild-type VDR gene were 0.56 ± 1.08 and 0.43 ± 1.09, respectively. Compared with −0.56 ± 0.92 and −0.36 ± 0.84, respectively, in patients with BsmI polymorphism (p < 0.01 in both comparisons). Concerning the left femoral neck, the patients with wild-type VDR gene had a mean T-score ± SD and a mean z-score ± SD of 0.22 ± 1.17 and 0.25 ± 0.95, respectively. Compared with −0.57 ± 0.92 and −0.43 ± 0.15, respectively, in patients with BsmI polymorphism (p < 0.01 in both comparisons). All of the BMD scores in patients with BsmI polymorphism were significantly lower than those of patients with wild-type VDR gene (Table 2). Osteopenia at either site was detected in 15% of patients with wild-type VDR gene and 37% of patients with BsmI polymorphism.
The mean ± SD of phosphate level in patients with BsmI polymorphism (3.4 ± 0.5) did not show a statistically significant difference compared with that of patients with wild-type VDR gene (3.4 ± 0.6) (p = 0.98). Likewise, the mean ± SD of calcium level in patients with BsmI polymorphism (9.0 ± 0.4) was not statistically different from that of patients with wild-type VDR gene (9.1 ± 0.4) (p = 0.36). Hypocalcemia was observed in 29.8% and 37% of epileptic patients with wild-type VDR gene and BsmI polymorphism, respectively.
The mean ± SD of serum 25-hydroxyvitamin D levels in patient with wild-type VDR gene was 39.7 ± 9.9 ng/ml. No patient had a level of ≤20 ng/ml, which would have indicated a 25-hydroxyvitamin D deficiency. Twelve (17.9%) of the 67 patients had levels of >20–29 ng/ml, suggesting 25-hydroxyvitamin D insufficiency. Moreover, the mean ± SD of serum 25-hydroxyvitamin D levels in patients who had BsmI polymorphism was 34.1 ± 10.6 ng/ml. Two (7.4%) of the 27 patients had levels of ≤20 ng/ml, indicating 25-hydroxyvitamin D deficiency. Four (14.8%) of the 27 patients had levels of >20–29 ng/ml suggesting 25-hydroxyvitamin D insufficiency. The 25-hydroxyvitamin D level in patients with wild-type VDR gene was higher than that in epileptic patients with BsmI polymorphism (p = 0.02). The 25-hydroxyvitamin D level in the combined group of patients with either wild-type VDR gene or BsmI polymorphism was not correlated with BMD at either site.
Parathyroid levels in the two groups were similar. The mean ± SD of serum parathyroid hormone levels in patients with wild-type VDR gene was 41.5 ± 16.9 ng/ml. Five (7.5%) of the 67 patients had hyperparathyroidism, but there were no patients with hypoparathyroidism. The mean ± SD of serum 25-hydroxyvitamin D levels in patients having BsmI polymorphism was 38.6 ± 13.9 ng/ml. Two (7.4%) of the 27 patients had hyperparathyroidism, but there were no patients with hypoparathyroidism. The parathyroid hormone level in the combined group of patients with either wild-type VDR gene or BsmI polymorphism was not correlated with BMD at either site.
The results of this study among young adults with epilepsy who were receiving phenytoin confirm that epileptic patients who had BsmI polymorphism have lower BMD than epileptic patients who have wild-type VDR gene. Epileptic patients with BsmI polymorphism were more likely to have BMD level below the expected range for age. The study also showed that epileptic patients with BsmI polymorphism had lower serum 25-hydroxyvitamin D level than epileptic patients with wild-type VDR gene. Moreover, serum parathyroid hormone level, serum calcium level, and serum phosphate level showed no statistically significant difference between epileptic patients with BsmI polymorphism and epileptic patients with wild-type VDR gene.
In 1992, Morrison et al. described for the first time a relationship between BsmI polymorphism and osteocalcin levels. In that work, the presence of the wild-type VDR gene was related to a higher BMD in the normal population and in twin pairs. In 1994, Morrison reported that BsmI polymorphisms were associated with lower BMD in postmenopausal women (Morrison et al., 1994). A great deal of attention was focused on this relationship, and these VDR polymorphisms have been the most extensively studied genetic markers. The frequency of the BsmI polymorphisms was found to be as high as 17% among Caucasians. In our study, we found a frequency of BsmI polymorphism of 28.7% in patients with epilepsy. By contrast, a previous study found a frequency of BsmI polymorphism of 23.7% in patients with systemic lupus erythematosus and only 17% in healthy subjects (Chaimuangraj et al., 2006).
The association between BMD and BsmI polymorphism has been the subject of many studies. Recent reports have identified BsmI polymorphism to be associated with reduced BMD and osteoporosis among postmenopausal woman (Chen et al., 2001), young adults (Kung et al., 1998), and prepubertal individuals (Saiz et al., 1997). In addition, BsmI polymorphism has been associated with a low BMD, and tends to be associated with an increased risk of fractures, independent of BMD level (Langdahl et al., 2000; Garnero et al., 2005). Of interest, a recent report by Creatsa et al. (2011) indicated that individuals with BsmI polymorphism may more responsive to oral alendronate than patients with wild-type gene. To date, only two studies have been conducted to evaluate the association between the BsmI polymorphism and BMD in patients with epilepsy. Tsukahara et al. (2002) examined the allelic variation in the VDR gene among 18 Japanese children with primary epilepsy who had taken various long-term AEDs. This study did not reveal any statistically significant association between the genotype by BsmI restriction fragment length polymorphism and low BMD. Lambrinoudaki et al. (2011) conducted a study to evaluate 73 long-term users of monotherapy AED and found a statistically significant association between BMD and the BsmI polymorphism. However, both of these studies had some limitations. First, they included epileptic patients taking CYP-inducer AEDs, CYP-inhibitor AEDs, or neither. Secondly, both studies had small sample sizes. Thirdly, BMD was measured at only the lumbar spine. To solve the problem in the design and analysis of these previous studies, it is necessary that: (1) the sample be large enough to have sufficient statistical power; (2) the sample population be a homogenous example of epileptic patients taking a single AED; and (3) the study design allow for the evaluation of BMD at femur and lumbar sites. The measurements of the regional skeleton such as the lumbar spine or the femur are often done clinically because of their importance in predicting fracture risk in later life. Our study was conducted to evaluate the BsmI polymorphism in 94 epileptic patients taking PHT as monotherapy and to measure the z-score and T-score at both the lumbar spine and femoral neck. The results of this study support the existence of a significant association between both T-score and z-score of BMD at both the lumbar spine and the femoral neck. Vitamin D is an important hormone in bone biology and mineral homeostasis. Its actions are mediated through the active metabolism of 1,25-hydroxyvitamin D3. Most of the biologic activities of 1,25-hydroxyvitamin D3 are mediated by a high-affinity receptor that acts as a ligand-activated transcription factor. The major steps involved in the control of gene transcription by the VDR include ligand binding, heterodimerization with the retinoid X receptor (RXR), binding of the heterodimer to vitamin D response elements (VDREs), and recruitment of other nuclear proteins into the transcriptional preinitiation complex (Valdivielso & Fernandez, 2006). Chromosome 12cen-q12 is the location of the VDR-encoding gene, in which at least 22 unique mutations have been reported. The BsmI polymorphism occurs in the intron between exons VIII and IX in the 3′ region of the hVDR gene and is in linkage with two of the other polymorphic sites (ApaI and TaqI). The mechanism underlying the association of BsmI VDR polymorphism and bone metabolism is uncertain. A possible mechanism may be the BsmI VDR polymorphism affecting calcium metabolism, and/or osteoblast and osteoclast function. In previous studies, AEDs, especially CYP enzyme inducer, have been shown to induce bone loss via the alteration of osteoclast differentiation and function (Dziak et al., 1988; Koide et al., 2009). Vitamin D receptor polymorphism, when present, can alter bone metabolism and, therefore, affect the susceptibility to the osteopenic effects of AEDs.
Studies on the association between BsmI polymorphism and 25-hydroxyvitamin D or parathyroid hormone are limited and also conflicting. Two studies have found that the BsmI polymorphisms are associated with differences in the 25-hydroxyvitamin; the baseline 1,25-hydroxyvitamin D levels were higher in premenopausal woman with the BsmI polymorphism than in those with the wild-type, which may have important implications in relation to the pathophysiology of osteoporosis (Howard et al., 1995; Laaksonen et al., 2002). Lambrinoudaki et al. (2011) reported that in an epileptic population, the BsmI polymorphism was associated with lower levels of serum 25-hydroxyvitamin D. This association was confirmed in our study. In addition, the effect of the BsmI polymorphism on PTH has been investigated in patients with chronic renal failure or dialysis. The findings of these studies were in the same direction; BsmI polymorphism may affect circulating levels of PTH as patients with wild-type VDR gene had a significantly lower serum PTH level (Fernández et al., 1997; Tagliabue et al., 1999). These findings were not supported by our study.
Nevertheless, the present study has some limitations. The study assessed the calcium intake via quantitative information regarding certain calcium sufficient food and estimated calcium intake by manual calculation. We obtained the information on calcium intake via an interview based on the patient’s recall of calcium intake over the previous 7 days. The dietary intake estimated from the recall technique over a short period of time may not be a good representation of usual or long-term calcium intake. Secondly, we included a few patients who had taken other AEDs before inclusion. However, one bone remodeling cycle takes approximately 3 months and it has been supported that BMD may return to normal after three remodeling cycles. Therefore, part treatment with other AED, but stopped at 6 years previously should not have had an impaired on BMD in our study. Thirdly, concerning the serum PHT levels and BMD, Lau et al. (1995) found that PHT biphasically increases [3H]-thymidine incorporation, cell number, alkaline phosphatase activity, and collagen synthesis in culture human bone cells, indicating that PHT stimulates both growth and differentiation. In theory, there is no reason to suppose that BsmI polymorphism should affect AED metabolism as the gene plays no role in the metabolism of PHT. However, we did not find any association between BsmI and serum PHT levels. In the future, we plan to evaluate whether gene polymorphisms have any effects on PHT metabolism and subsequent serum PHT concentration.
The strengths of this study were that it comprised a large epileptic population of young adult epileptic patients, allowing us to assess the association between VDR gene polymorphism and BMD, 25-hydroxyvitamin D, and parathyroid hormone level. Secondly, we had a homogenous sample population (epileptic patients taking a CYP-inducer AED). Thirdly, the measurement of 25-hydroxyvitamin D and parathyroid hormone as well as that of the VDR gene polymorphism and BMD used modern techniques and appropriate measurement sites (femur and lumbar spine). Furthermore, we excluded patients with potential confounding factors that are known to affect BMD such as low calcium intake, previous fracture 1 year before inclusion, smoking, alcohol consumption, abnormal menstruation, use of drugs including hormonal drugs, and recent weight change. Therefore, our study was able to evaluate the true association between VDR gene polymorphism and BMD in young adult patients with epilepsy.
In conclusion, to our knowledge, this is the first study to investigate the association between BsmI polymorphism and BMD, parathyroid hormone, and 25-hydroxyvitamin D level among young patients with epilepsy who are taking CYP-inducer AED, PHT as monotherapy. Our study confirmed an association between BMD and the BsmI polymorphism. The presence of BsmI polymorphism in patients with epilepsy was associated with BMD below the expected range for age. Therefore, patients with epilepsy taking CYP-inducer AEDs should have their BMDs monitored. In a future study, we will evaluate which AED medication is associated with the least reduction of BMD in outpatients with long-term use of AEDs and with BsmI polymorphism.
This study was supported by grants for scientific research from the Faculty of Medicine, Prince of Songkla University, Thailand. We thank Ms. Anongtip Sa and Ms. Walailuk Jitpiboon for their assistance with the data collection and analysis, and Ms. Wanwisa Maneechay for her assistance with VDR gene analysis. Also we wish to thank Mr. Edmond Subashi for English-language editing of the manuscript.
None of author has any conflict of interest to disclosure. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.