Antiseizure medications and bone disease

INTRODUCTION — Epilepsy is a chronic condition that affects over two million people in the United States, approximately 1 percent of the population [1]. Antiseizure medications (ASMs) remain the mainstay of treatment for epilepsy. In addition, these agents now have numerous other indications, including the treatment of migraine headaches, bipolar disorder, and chronic pain.

Both epilepsy and ASMs are associated with adverse effects on bone health. Persons with epilepsy treated with ASMs have increased rates of bone loss and abnormalities in bone and mineral metabolism [2,3]. These adverse effects may contribute to an increased risk of fracture [4-7].

This topic will review the association between ASMs and disorders of bone and mineral metabolism, including osteomalacia/rickets and osteoporosis or low bone mass. In addition, the screening, treatment, and prevention of ASM-related bone disease will be discussed.

Osteopenia, osteoporosis, and bone fractures are also a concern for children with intractable epilepsy maintained on ketogenic diets. This topic is reviewed separately. (See "Ketogenic dietary therapies for the treatment of epilepsy", section on 'Common adverse effects'.)

METABOLIC BONE DISEASE — The first reports linking antiseizure medication (ASM) therapy to skeletal disease were published in the late 1960s [8]. Affected patients had florid bone disease with clinical, biochemical, and histologic abnormalities consistent with rickets and osteomalacia. However, these early reports included mostly institutionalized patients.

Studies of community-dwelling, ambulatory patients with epilepsy who receive ASM therapy generally describe more subtle radiographic and biochemical abnormalities, including decreased bone mineral density (BMD), abnormal vitamin D metabolism, and biochemical evidence of increased bone remodeling activity [9-11]. Each of these abnormalities is associated with decreased bone strength and increased risk of fractures compared with individuals who do not have epilepsy and do not take ASMs. Patients with epilepsy have an even higher risk for fracture, owing to the increased risk of falls and physical trauma associated with the seizures themselves, the neurologic disease that underlies the epilepsy (stroke, cerebral palsy), and the effects of certain ASMs that impair gait stability. Thus, the identification of individuals with epilepsy who are predisposed to fracture because of decreased bone strength is an important aspect of their management.

Osteoporosis — Osteoporosis is defined as a skeletal disorder characterized by decreased bone strength leading to an increased risk of fracture. Bone strength represents the integration of the amount of bone present (bone mass or density) and the quality of that bone [12]. Although fractures are the only clinical symptom of osteoporosis, osteoporosis may be diagnosed radiographically before fractures occur by measuring the amount of bone present (bone mineral density or BMD). (See "Clinical manifestations, diagnosis, and evaluation of osteoporosis in postmenopausal women" and "Clinical manifestations, diagnosis, and evaluation of osteoporosis in men".)

Osteoporosis, whether manifested by low BMD or fractures, is most common in postmenopausal women and older men, in whom the effects of aging (estrogen and testosterone deficiency) cause bone to be lost more rapidly than it can be restored. This leads to deterioration of the bone microarchitecture and associated reductions in strength.

However, osteoporosis can also develop as a result of a host of underlying medical conditions and medications, in which case it is termed secondary osteoporosis. Concomitantly occurring secondary causes of osteoporosis can increase the severity of osteoporosis in postmenopausal women and older men and can cause osteoporosis in young men, women, and children. ASMs are one of the classes of medications associated with secondary osteoporosis (table 1) [13]. In a murine study of untreated chronic temporal lobe epilepsy, there was no evidence of overt bone abnormalities due to epilepsy itself [14]. (See "Clinical manifestations, diagnosis, and evaluation of osteoporosis in postmenopausal women" and "Etiology of osteoporosis in men" and "Epidemiology and etiology of premenopausal osteoporosis".)

Fracture — Fracture is the most important manifestation of osteoporosis. Several studies have reported that fracture rates are higher in patients with epilepsy than in control populations, and among patients with epilepsy, fractures are more common in postmenopausal women and older men [4-7,15-20].

In a report from the Women's Health Initiative (WHI) study, which included 1385 users of ASMs and 137,282 nonusers aged 50 to 79 years followed for a mean of 7.7 years, use of ASMs was associated with a significantly increased risk of total fractures (annualized percentage 3.35 versus 2.10 percent, hazard ratio [HR] 1.44, 95% CI 1.30-1.61) and site-specific fractures, including hip fracture (0.29 versus 0.15 percent, HR 1.51, 95% CI 1.05-2.17) and clinical vertebral fracture (0.48 versus 0.22 percent, HR 1.60, 95% CI 1.20-2.12) [21]. Both the number of ASMs used in a patient (more than one ASM compared with one) and type (enzyme inducing versus non-enzyme inducing) were significantly associated with fracture risk. (See 'Effect of ASM type' below.)

Other factors associated with higher fracture risk in adults with epilepsy in some [16,22], but not all [23,24], studies include older age, female sex, greater severity of epilepsy, and duration of ASM use. Children and young adults may also have an increased risk of fractures compared with controls [25]. For example, a case-control study of 23 pairs of siblings aged 5 to 18 years who were within two years of age and who were ASM exposure-discordant found that the siblings taking ASMs reported more fractures (15 fractures in 8 subjects treated with ASMs compared with 4 fractures in 3 controls) and had reductions in tibial volumetric BMD and lower limb muscle force compared with their matched controls [26]. Another study found an age effect among new ASM users with increased nontraumatic fracture incidence in a younger age group peaking at 11 to 13 years then decreasing with older age group [27].

Several observations indicate that the increase in fracture rate is related to seizure-related injuries and the effects of certain ASMs that impair gait stability, as well as the adverse effects of ASMs on bone strength:

●Patients with generalized tonic-clonic seizures are at higher risk for fracture than patients with partial seizures, suggesting that seizure-related injuries contribute [24].

●The observation that epilepsy was associated with an even higher relative risk (RR) of hip versus forearm fracture (5.3 versus 1.7) suggests a possible preponderance of falls occurring in patients unable to use their hand to break the fall, a situation that might occur during a seizure [10,16,28].

●In the WHI study, there was a significant association between ASM use and the risk of falls (HR 1.6, 95% CI 1.5-1.7) [21].

●In a longitudinal study of balance measures and lower limb strength in 26 sibling pairs, chronic ASM users had poorer balance compared with nonusers [29]. There was no difference in quadriceps muscle strength.

Bone mineral density — Some ASMs alter bone and mineral metabolism. Several studies have reported low BMD at multiple sites in adult patients receiving ASMs, with measurements generally ranging between 10 and 16 percent below controls [9,19,30-34]. Men and women appear to be affected similarly [34]. A meta-analysis of 12 studies measuring BMD in patients with epilepsy found an overall significant deficit of BMD in both hip and spine with mean Z-score deviations of -0.56 and -0.38, respectively [10]. (See "Overview of dual-energy x-ray absorptiometry", section on 'DXA technology'.)

In contrast, however, a cross-sectional study of premenopausal women found BMD to be normal despite ASM monotherapy for an average of 8 to 13 years [35]. In children, reduced BMD at axial and appendicular sites has been described but not as consistently as in older adults [32,36-45]. Increasing age is associated with decreased BMD among patients using ASMs, perhaps contributing to these discrepancies [30,33]. Other potential confounders include body mass index (BMI), number and type of ASMs, and duration of treatment [32,46]. (See 'Effect of ASM type' below.)

Most [36,37,44,47], but not all [30], cross-sectional studies demonstrate that duration of ASM treatment is associated with the degree of BMD loss. In some prospective studies, progressive bone loss occurred over time [9,33,48]. As examples:

●In a prospective study of 81 men aged 25 to 54 years, followed for 12 to 29 months, there was a 1.8 percent annualized loss of BMD by dual-energy x-ray absorptiometry (DXA), much higher than would be expected in men of this age [33].

●In a cohort of older women followed prospectively for an average of 4.4 years, the rate of bone loss at the total hip (after adjusting for confounders) was -0.70 percent per year in 4094 non-ASM users, -0.87 percent per year in 61 intermittent ASM users, and -1.16 percent per year in 41 continuous ASM users [48].

In contrast, in the subset of ASM users (84 women) in the WHI study who had BMD measured prospectively, there was no significant difference in three-year change in BMD of the hip, spine, or total body between users and nonusers of ASMs [21]. The small number of women who had BMD measured, combined with the fact that many ASM users were taking hormone therapy, may have reduced the ability of the study to show an association between ASM use and change in BMD. Alternatively, these observations may indicate that the increase in fracture rate may be due to seizure-related injuries and the effects of certain ASMs that impair gait stability rather than the adverse effects of ASMs on bone strength alone.

Markers of bone turnover — Markers of bone formation and resorption, measured in the serum and urine, may be elevated in adults with epilepsy receiving ASMs. Elevated levels have been reported during long-term ASM therapy [31,40,48,49], and increases have been reported to occur after initiation of ASM therapy [31,35,40,48-51]. Elevated bone turnover markers are considered to reflect increased bone remodeling activity, are associated with higher rates of bone loss, and are independent predictors of fracture. (See "Use of biochemical markers of bone turnover in osteoporosis".)

Findings in adults with epilepsy include:

●Increased serum alkaline phosphatase activity in adults receiving some ASMs [49,50,52-54]. In those studies that measured isoenzymes of alkaline phosphatase, the increase in total alkaline phosphatase activity appeared to be related to increases in the bone isoenzyme fraction [35,55,56]. Phenytoin has been associated with the most consistent and marked elevations of alkaline phosphatase. However, increases have also been reported with carbamazepine. (See 'Effect of ASM type' below.)

●High serum levels of other bone formation markers. High levels of osteocalcin and C-terminal extension peptide of type I procollagen (PICP) have been described in persons receiving ASMs [31,40,49-51].

●Elevated markers of bone resorption. C-terminal telopeptide of human type I collagen (CTX) and cross-linked N-terminal telopeptides of type I collagen (NTX) are also elevated in patients treated with ASMs [31,49-51].

While bone turnover markers are not used to diagnose osteoporosis because values overlap substantially in normal subjects and patients with osteoporosis, the collective data suggest that bone remodeling activity is higher in patients taking ASMs. (See "Use of biochemical markers of bone turnover in osteoporosis".)

Osteomalacia and rickets — Osteomalacia is a disorder of bone mineralization in which the deposition of calcium and phosphate into newly formed bone matrix (osteoid) at sites of bone remodeling or periosteal or endosteal apposition is either insufficient or completely absent. On histologic examination, thickened layers of unmineralized osteoid cover a higher than normal proportion of the bone surface.

Clinically, patients with osteomalacia complain of bone pain, which is particularly prominent in the lower extremities and is exacerbated by weightbearing. They may also notice muscle weakness, particularly of the upper arms and thighs. Physical examination often reveals tenderness to pressure over the anterior tibiae, an antalgic gait, and proximal muscle weakness. Although there are many causes of osteomalacia, when osteomalacia occurs in association with ASM treatment, it is generally associated with profound vitamin D deficiency. Serum calcium and phosphorous and serum 25-hydroxyvitamin D (25[OH]D) may be low, and parathyroid hormone (PTH) levels and total alkaline phosphatase activity may be elevated. (See "Epidemiology and etiology of osteomalacia".)

Rickets, a failure of or delay in mineralization of cartilage at growth plates, occurs only in children and results in profusion of disorganized, nonmineralized, degenerating cartilage and consequent widening of the epiphyseal plates with flaring or cupping and irregularity of the epiphyseal-metaphyseal junctions. If untreated, the disorder may progress to include bowing of the lower extremities. In adults, joint enlargement due to the profusion of cartilage and ultimate formation of excess but undermineralized bone is sometimes evident [57]. (See "Overview of rickets in children".)

As noted above, early reports of bone disease in patients with epilepsy describe osteomalacia [8,58,59]. However, these reports primarily included institutionalized patients, and in ambulatory, community-dwelling persons with epilepsy, osteomalacia is rarely seen today.

Bone biopsy studies in patients taking ASMs found normal osteoid seam width and mineralization rates that were consistently normal or increased [60,61]. These results are not consistent with osteomalacia and suggest that bone disease associated with ASM use is a disorder of increased remodeling (consistent with bone turnover markers data) rather than abnormal mineralization, resulting in osteoporosis rather than osteomalacia. Similarly, while rickets has been reported in older studies of children treated with ASMs, most of the subjects were institutionalized, and the findings likely reflect the additive effects of ASM and lack of sun exposure on vitamin D metabolism [62]. More recent studies of ambulatory, community-dwelling children taking ASMs do not report clinical rickets [36-39].

EFFECT OF ASM TYPE — The relationship between antiseizure medication (ASM) type and fracture risk remains uncertain. In a systematic review of 13 observational studies in patients with epilepsy that compared treatment with ASMs that induce the cytochrome P450 system versus ASMs that do not induce, five studies showed decreased bone mineral density (BMD) in users of ASMs that induce the cytochrome P450 system, whereas five studies showed no effect on BMD [63]. Two studies showed an increased risk of fracture in patients treated with enzyme-inducing ASMs, whereas one study showed no difference in fracture risk. In the largest study, a prospective study of over 63,000 patients with epilepsy, there was an increased risk of hip fracture with use of enzyme-inducing compared with non-enzyme-inducing ASMs for both men (hazard ratio [HR] 1.53, 95% CI 1.10-2.12) and women (HR 1.49, 95% CI 1.15-1.94) [64].

Induction of the cytochrome P450 system by ASMs leads to increased catabolism of vitamin D to inactive metabolites and a subsequent rise in parathyroid hormone (PTH), which increases the mobilization of bone calcium stores and subsequent bone turnover [52,65-67]. However, this mechanism does not account for the studies that have shown accelerated bone turnover or bone loss independent of vitamin D deficiency [32,33,35,49-51,61].

Other mechanisms that have been implicated in the pathogenesis of ASM-induced bone disease include:

●A cytochrome P450-induced increase in metabolism of sex steroids, resulting in lower estrogen levels [68]

●A direct inhibitory effect of phenytoin on intestinal absorption of calcium [69,70]

●Hyperparathyroidism with normal serum 25-hydroxyvitamin D (25[OH]D) level [49]

●Direct effects of phenytoin to stimulate osteoclastic bone resorption [71]

●Direct effects of phenytoin and carbamazepine to inhibit proliferation of human osteoblast-like cells at concentrations equivalent to therapeutic doses for the treatment of epilepsy [47]

●Vitamin K deficiency [72]

●Calcitonin deficiency [73]

●Elevated homocysteine levels [74]

●Suppression of long bone growth by valproate by inhibiting cartilage formation and accelerating ossification of the growth plate [75]

●Genetic factors [76,77]

Enzyme inducing — The ASMs most commonly associated with altered bone and mineral metabolism and decreased bone density are those that induce the cytochrome P450 enzyme system [35-37,48-53]. The ASMs that induce the cytochrome P450 enzyme system are phenytoin, primidone, carbamazepine, and phenobarbital (table 2) [31,35,37,48,49,52,53]. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Induction of the cytochrome P450 system by ASMs leads to increased catabolism of vitamin D to inactive metabolites [52,65-67]. Decreased availability of active vitamin D metabolites leads, in turn, to decreased gastrointestinal absorption of calcium, hypocalcemia, and a rise in circulating PTH. PTH increases the mobilization of bone calcium stores and subsequent bone turnover. This mechanism is the one most often purported to underlie the bone disease associated with ASM use.

Laboratory findings in patients taking these drugs often, but not consistently, include hypocalcemia, hypophosphatemia, reduced serum 25(OH)D levels, elevated PTH levels, and elevated markers of bone formation and resorption. All of these have been described in adults taking enzyme-inducing ASMs [52,53,55,56,78-82].

Basic studies have evaluated the effect of these ASMs on the expression of specific cytochrome P450 isoenzymes involved in vitamin D metabolism [83,84]. Phenobarbital, phenytoin, and carbamazepine are among a class of drugs known as xenobiotics. Xenobiotics activate a nuclear receptor known as either the steroid and xenobiotic receptor (SXR) or pregnane X receptor (PXR):

●One study found that xenobiotics upregulate 25(OH)D3-24-hydroxylase (CYP24) in the kidney through activation of PXR [84]. This enzyme catalyzes the conversion of 25(OH)D to its inactive metabolite (24,25-dihydroxyvitamin).

●Other investigators found that xenobiotic activation of PXR did not upregulate CYP24 but did increase expression of a different isoenzyme, CYP3A4, in the liver and small intestine [83]. This enzyme converts vitamin D to more polar inactive metabolites. They also found that xenobiotic activation of PXR represses CYP24 expression in the liver and intestine, suggesting a dual role in mediating vitamin D metabolism.

Studies that have compared ASM regimens often find that phenytoin is associated with the most marked increases in bone turnover markers and decreases in BMD [31,35,48,49,85]. In a prospective study of premenopausal women with epilepsy receiving a single ASM, treatment with phenytoin for one year was associated with a 2.6 percent decrease in femoral neck BMD, whereas treatment with other ASMs (carbamazepine, lamotrigine, valproate) did not have such an adverse effect on BMD [85]. In a prospective cohort study of 90 young Thai adults, patients who stopped or switched phenytoin to levetiracetam had a significant increase in BMD at multiple sites as well as an increase in 25(OH)D concentrations [86]. In contrast, patients who continued phenytoin had a significant decrease in BMD at multiple sites and a decrease in 25(OH)D concentrations. Animal studies suggest that phenytoin may also have direct toxicity on bone [47,71].

Carbamazepine appears to be less commonly associated with bone disease than phenytoin. While hypocalcemia, hypophosphatemia, decreased vitamin D metabolites, and elevated markers of bone turnover have all been reported with carbamazepine, these findings have not been observed as consistently [11,36,41,42,50,51]. One short-term (10-week) study of normal, young men found that carbamazepine was not associated with any change in markers of bone turnover [87]. In contrast, significant increases in markers of bone turnover were documented in children and adolescents after initiation of carbamazepine treatment for epilepsy [50,51].

Bone mass has been reported to be normal or decreased in children and young adults treated with long-term carbamazepine therapy [32,42,49,88,89]. In a prospective study of premenopausal women with epilepsy receiving a single ASM, treatment with carbamazepine was not associated with any changes in markers of bone turnover or BMD [85]. However, fracture risk has been reported to be increased in adults treated with carbamazepine [5].

In a meta-analysis of 87 studies evaluating bone health indices in carbamazepine-treated persons compared with controls, vitamin D levels as well as calcium were significantly lower and alkaline phosphatase was higher in carbamazepine-treated persons with epilepsy. Interestingly, there was no difference in BMD [90].

Non-enzyme inducing

Valproate — In contrast to phenytoin, primidone, carbamazepine, and phenobarbital, valproate is an inhibitor of the cytochrome P450 enzyme system. (See "Antiseizure medications: Mechanism of action, pharmacology, and adverse effects".)

Although studies published in the 1980s did not find any association with bone disease [52,78], subsequent studies have consistently reported bone loss, abnormal biochemical indices of bone and mineral metabolism, and higher fracture rates with valproate in both children and adults [5,31,36,39,40]. Both dose and duration of valproate therapy have been associated with increased rates of bone loss and higher incidence of fracture [5,40].

Three meta-analyses support negative effects of valproate on children and adults with epilepsy [91-93]. BMD reductions and increased serum alkaline phosphatase were seen in children and adults. Reduced vitamin D and calcium levels and increased PTH occurred in children and not adults with epilepsy.

In cell-culture studies, valproate has been shown to suppress longitudinal bone growth by inhibiting cartilage formation and accelerating ossification of the growth plate [75]. Similarly, valproate-treated rats have decreased bone volume fraction and decreased apparent density [94]. Perhaps consistent with this effect, short stature has been reported in children treated with valproate [39]. In the same study, low BMD, short stature, and elevated markers of bone turnover were more pronounced in children treated with valproate in combination with lamotrigine [39]. In this study, physical inactivity was identified as a possible confounder.

Other ASMs — Few studies have evaluated the effect of newer antiseizure medications (ASMs) on bone and mineral metabolism and BMD:

●Lamotrigine – Lamotrigine is increasingly used to treat epilepsy. A study of premenopausal women on ASM monotherapy (phenytoin, carbamazepine, valproate, or lamotrigine) found no significant effect of lamotrigine on BMD, calciotropic hormones, or bone turnover markers [35,85]. Although treatment duration was shorter for patients on lamotrigine, there was no association between duration of therapy and BMD or any marker of bone and mineral metabolism. Similarly, 32 adults with newly diagnosed epilepsy treated with lamotrigine had no changes in markers of bone turnover after two years [95] and eight Korean men and women did not have any changes in BMD after six months of treatment with lamotrigine [89]. In contrast, children treated with lamotrigine were significantly shorter and had lower BMD and higher markers of bone turnover compared with children not receiving ASMs [39]. These findings were most pronounced in children treated with both lamotrigine and valproate.

●Oxcarbazepine – A large case-control study found that oxcarbazepine (and clonazepam) were among the ASMs associated with small increases in the relative risk (RR) of fracture after controlling for epilepsy diagnosis [5].

A study of 45 patients found reduced serum 25(OH)D and slightly elevated osteocalcin levels among patients taking oxcarbazepine [11]. A retrospective claims-based study found that initiating oxcarbazepine among 4 to 13 year olds was associated with an elevated risk of fragility fracture [96]. Among children with new-onset focal epilepsy treated with oxcarbazepine, serum calcium was reduced and parathyroid hormone levels were increased [97].

●Levetiracetam – A study of 61 patients (mean age 31 years) initiating levetiracetam found no adverse effect on BMD or in biochemical markers (25[OH]D, calcium, phosphorus, PTH, markers of bone remodeling) after 14 months of treatment [98]. Among 25 children with new-onset focal epilepsy treated with levetiracetam, there was no effect of levetiracetam on bone metabolism [97]. Similarly, 47 adults with newly diagnosed epilepsy treated with levetiracetam had no changes in markers of bone turnover after two years [95]. In another study, there was improvement in BMD two years after switching from phenytoin to levetiracetam [86]. A retrospective claims-based study found that initiating levetiracetam among 4 to 13 year olds did not increase the risk of fragility fracture [96].

●Topiramate and zonisamide – Topiramate and zonisamide are carbonic anhydrase inhibitors. As carbonic anhydrase inhibitors are potent inhibitors of osteoclastic bone resorption, one might expect lower rates of bone loss to be associated with their use [99]. Limited data suggest no effect on BMD. As an example, in a cross-sectional study in premenopausal women with epilepsy, there were no differences in BMD Z-scores or vitamin D metabolites in women taking topiramate compared with control women and compared with women taking carbamazepine or valproic acid [100]. Topiramate was associated with lower PTH, bicarbonate, and calcium concentrations as well as higher biochemical markers of bone turnover compared with the other groups. In a prospective study in 59 adults with new onset epilepsy, there were no changes in BMD or biochemical markers of bone turnover after 13 months of treatment with zonisamide [101].

SCREENING — Osteoporosis screening is typically recommended in postmenopausal women and older men with risk factors for fracture (table 3) (see "Screening for osteoporosis in postmenopausal women and men"). For patients with epilepsy, we consider the risk factors for bone disease related to antiseizure medication (ASM) use in addition to more general risk factors for osteoporosis and fracture.

Clinical risk factor assessment — Risk factors for bone disease in ASM users include high-dose, multidrug regimens; long-term therapy; inadequate intakes of vitamin D; limited sunlight exposure; chronic illness; old age; institutionalization; low physical activity; adjuvant therapy to induce chronic metabolic acidosis (acetazolamide or ketogenic diets); and concomitant therapy with other drugs that induce hepatic enzymes (rifampin, glutethimide).

Risk factors for osteoporosis and fractures include female sex, postmenopausal status, old age, poor balance, tobacco, low body mass index (BMI), and low dietary calcium and vitamin D intake (table 3). ASMs are an accepted secondary cause of osteoporosis. (See "Osteoporotic fracture risk assessment".)

Bone mineral density — It is reasonable to recommend measuring bone mineral density (BMD) in patients with epilepsy who have risk factors for osteoporosis and risk factors for ASM-related bone disease, as well as in patients who have experienced a fragility fracture. Persons treated with long-term (>5 years) enzyme-inducing ASMs or valproate are likely at risk and therefore should be screened.

In a person with a normal baseline BMD, the frequency of testing should depend upon the age and sex of the patient. More frequent testing (every two years) is appropriate in a postmenopausal woman or older man and less frequent in a premenopausal woman or young man. Because multiple risk factors further increase risk of fracture, earlier and more frequent screening is prudent in such patients (table 1) [6].

Laboratory evaluation — In patients with epilepsy who are taking ASMs, we typically measure serum calcium, phosphate, and 25-hydroxyvitamin D (25[OH]D) levels [46,102-104].

TREATMENT AND PREVENTION — Few studies have rigorously evaluated strategies for prevention and treatment of bone disease associated with antiseizure medications (ASMs).

Lifestyle measures — Patients with epilepsy who are taking ASMs should adopt the same lifestyle measures that are recommended for all patients at risk for osteoporosis and fracture. These include regular weightbearing exercise, smoking cessation, limiting alcohol intake, and preventing falls. In one study, however, adults with epilepsy did not engage in higher level of osteoprotective behaviors and were less physically active when compared with persons without epilepsy [105]. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women", section on 'Lifestyle measures to reduce bone loss' and "Treatment of osteoporosis in men", section on 'Lifestyle measures'.)

Calcium and vitamin D — As a general rule, patients should achieve the reference calcium intake for their age group (table 4) through diet and supplements (if needed) and receive supplemental vitamin D (400 to 800 international units/day).

Because studies have shown that much higher doses (up to 4000 international units) may be necessary to normalize serum levels of 25-hydroxyvitamin D (25[OH]D) in some patients, we typically measure serum 25(OH)D levels to determine whether vitamin D intake is sufficient to maintain levels above 30 ng/mL [106-108]. This is particularly important in the case of older adults or institutionalized patients who may have limited access to sunshine.

There are limited data on the optimal doses of calcium and vitamin D for patients with epilepsy who are taking ASMs. In a study that included both institutionalized and ambulatory individuals receiving ASMs who had low baseline serum 25(OH)D levels, vitamin D supplementation was titrated to achieve a normal serum 25(OH)D concentration [109]. All achieved normal 25(OH)D levels over a period of 12 to 15 months. Dose requirements ranged from 400 to 4000 international units daily.

In two randomized trials in adults and children given either low-dose vitamin D (400 international units/day) or high-dose vitamin D (4000 international units/day in adults and 2000 international units/day in children), there were no significant differences in bone mineral density (BMD) at one year between the adult groups, although BMD had increased from baseline in the high-dose group only [106]. In children, both dose groups had comparable increases in BMD after one year.

In a systematic review of nine studies evaluating vitamin D supplementation in adults with epilepsy, vitamin D treatment increased serum calcium in three of eight studies, decreased alkaline phosphatase in six of eight studies, and decreased parathyroid hormone (PTH) in two of four studies [110]. All six studies that investigated BMD had significant findings; however, likely due to varying methodologies employed by the studies, the impact of vitamin D on BMD was not conclusive.

Pharmacologic therapy — In the absence of adequate data specifically targeting osteoporosis in individuals treated with ASMs, treatment recommendations should follow other guidelines, such as for postmenopausal women and men (see "Overview of the management of low bone mass and osteoporosis in postmenopausal women" and "Treatment of osteoporosis in men"). However, it is of particular importance to make certain that individuals receiving ASMs are vitamin D replete before initiating therapy for osteoporosis.

In a two-year, double-blind trial evaluating risedronate or placebo (all patients received calcium and vitamin D supplementation) in 80 male veterans with epilepsy treated with either phenytoin, phenobarbital, carbamazepine, or valproate, there was a greater increase in lumbar spine BMD in the risedronate group (0.065 versus 0.016 g/cm² in the placebo group) [111]. There were fewer new vertebral fractures in the risedronate group (none versus five in the placebo group).

SUMMARY AND RECOMMENDATIONS

●Fracture risk – Patients with epilepsy are at higher risk for fracture because of a higher risk for falls and physical trauma related to seizures themselves, the neurologic disease that underlies the epilepsy (stroke, cerebral palsy), and the effects of certain antiseizure medications (ASMs) that impair gait stability and alter bone and mineral metabolism. (See 'Fracture' above.)

●Effect of ASM type on fracture risk and bone loss – The relationship between ASM type and fracture risk remains uncertain. However, ASMs that induce the hepatic cytochrome P450 enzyme system (phenytoin, phenobarbital, primidone, carbamazepine) are those most commonly implicated as affecting bone and mineral metabolism (table 2). The most commonly implicated pathogenetic mechanism is via accelerated catabolism of vitamin D. In addition, there is growing evidence to suggest that valproate, a cytochrome P450 enzyme inhibitor, also affects bone and mineral metabolism. There are limited data available on the skeletal effects of newer ASMs, such as lamotrigine. (See 'Effect of ASM type' above.)

Several additional mechanisms for ASM-associated bone loss have been proposed, including direct effects on intestinal calcium absorption and also on osteoblasts and osteoclasts. It is likely that more than one mechanism contributes to ASM-associated bone disease and that these are, to some extent, specific to the particular ASM. (See 'Effect of ASM type' above.)

●Screening – For individuals with a history of prolonged use (>5 years) of ASMs, especially those using enzyme-inducing ASMs or valproate, those with other risk factors for ASM-induced bone disease (high-dose, multidrug regimens; low vitamin D intake; limited sunlight exposure; chronically ill, older, or institutionalized patients; low physical activity levels; exposure to drugs that induce chronic metabolic acidosis; and concomitant therapy with other drugs that induce hepatic enzymes), and those with history of fragility fracture or additional risk factors for osteoporosis, we suggest bone mineral density (BMD) testing (Grade 2C). (See 'Screening' above.)

●Calcium and vitamin D intake – For patients receiving ASMs, we suggest calcium and vitamin D supplementation (Grade 2C). Although the optimal intake (diet plus supplement) has not been clearly established in this patient population, the age group-based reference intake for calcium (table 4) and 800 international units of vitamin D daily are the typical doses used. However, patients receiving ASMs, particularly enzyme-inducing ASMs, may require higher doses of vitamin D. We typically measure serum 25-hydroxyvitamin D (25[OH]D) levels in such patients to determine whether vitamin D intake is sufficient to maintain levels within the normal range. (See 'Calcium and vitamin D' above and "Calcium and vitamin D supplementation in osteoporosis" and "Vitamin D deficiency in adults: Definition, clinical manifestations, and treatment", section on 'Vitamin D replacement'.)

●Pharmacologic therapy – The pharmacologic treatment of osteoporosis and fracture in patients taking ASMs is the same as the treatment of osteoporosis in individuals not on ASM therapy. (See "Overview of the management of low bone mass and osteoporosis in postmenopausal women" and "Treatment of osteoporosis in men" and "Evaluation and treatment of premenopausal osteoporosis".)