- Patient Information:
Details with Side Effects
Valproic acid dissociates to the valproate ion in the gastrointestinal tract. The mechanisms by which valproate exerts its therapeutic effects have not been established. It has been suggested that its activity in epilepsy is related to increased brain concentrations of gamma-aminobutyric acid (GABA).
A single-dose randomized crossover study compared Stavzor 500-mg strength capsules to 500-mg Depakote delayed-release tablets. These studies demonstrated that the 2 products had similar plasma concentration-time profiles under fasted conditions in terms of valproic acid, although the median Tmax occurred earlier with STAVZOR (2.0 hrs versus 3.5 hrs). Co-administration with food increased the Tmax of Stavzor (2.0 hrs without food and approximately 4.8 hours with food), and resulted in a 23% decrease in Cmax of valproic acid, although there was no change in systemic exposure (AUC).
Although the rate of valproate ion absorption may vary with the conditions of use (eg, fasting or postprandial), these differences should be of minor clinical importance under the steady-state conditions achieved in chronic use in the treatment of epilepsy. However, it is possible that differences among the various valproate products in Tmax and Cmax could be important upon initiation of treatment. For example, in single dose studies, the effect of feeding had an influence on the rate of absorption of the capsule (increase in Tmax from 2.3 to 6.1 hours). While the absorption rate from the GI tract and fluctuation in valproate plasma concentrations vary with dosing regimen, the efficacy of valproate as an anticonvulsant in chronic use is unlikely to be affected. Experience employing dosing regimens from once-a-day to 4-times-a-day, as well as studies in primate epilepsy models involving constant rate infusion, indicates that total daily systemic bioavailability (extent of absorption) is the primary determinant of seizure control and that differences in the ratios of plasma peak to trough concentrations are inconsequential from a practical clinical standpoint. Whether or not rate of absorption influences the efficacy of valproate as an antimanic or antimigraine agent is unknown. Co-administration of oral valproate products with food should cause no clinical problems in the management of patients with epilepsy [see DOSAGE AND ADMINISTRATION].
An in vitro study evaluating dissolution of valproic acid showed earlier dissolution in the presence of ethanol than in the absence of ethanol. This has not been studied in humans. However, there is a potential for an earlier Tmax and therefore a higher Cmax when valproic acid is given with alcohol.
Any changes in dosage administration, or the addition or discontinuance of concomitant drugs, should ordinarily be accompanied by close monitoring of clinical status and valproate plasma concentrations.
The plasma protein binding of valproate is concentration dependent and the free fraction increases from approximately 10% at 40 mcg/mL to 18.5% at 130 mcg/mL. Protein binding of valproate is reduced in the elderly, in patients with chronic hepatic diseases, in patients with renal impairment, and in the presence of other drugs (eg, aspirin). Conversely, valproate may displace certain protein-bound drugs (eg, phenytoin, carbamazepine, warfarin, and tolbutamide) [See DRUG INTERACTIONS for more detailed information on the pharmacokinetic interactions of valproate with other drugs].
Valproate concentrations in cerebrospinal fluid (CSF) approximate unbound concentrations in plasma (about 10% of total concentration).
Valproate is metabolized almost entirely by the liver. In adult patients on monotherapy, 30-50% of an administered dose appears in urine as a glucuronide conjugate. Mitochondrial β-oxidation is the other major metabolic pathway, typically accounting for over 40% of the dose. Usually, less than 15-20% of the dose is eliminated by other oxidative mechanisms. Less than 3% of an administered dose is excreted unchanged in urine.
The relationship between dose and total valproate concentration is nonlinear; concentration does not increase proportionally with the dose, but rather, increases to a lesser extent due to saturable plasma protein binding. The kinetics of unbound drug are linear.
Mean plasma clearance and volume of distribution for total valproate are 0.56 L/hr/1.73 m² and 11 L/1.73 m², respectively. Mean plasma clearance and volume of distribution for free valproate are 4.6 L/hr/1.73 m² and 92 L/1.73 m². Mean terminal half-life for valproate monotherapy ranged from 9 to 16 hours following oral dosing regimens of 250 to 1000 mg.
The estimates cited apply primarily to patients who are not taking drugs that affect hepatic metabolizing enzyme systems. For example, patients taking enzyme-inducing antiepileptic drugs (carbamazepine, phenytoin, and phenobarbital) will clear valproate more rapidly. Because of these changes in valproate clearance, monitoring of antiepileptic concentrations should be intensified whenever concomitant antiepileptics are introduced or withdrawn.
Effect of Age
Children within the first 2 months of life have a markedly decreased ability to eliminate valproate compared to older children and adults. This is a result of reduced clearance (perhaps due to delay in development of glucuronosyltransferase and other enzyme systems involved in valproate elimination) as well as increased volume of distribution (in part due to decreased plasma protein binding). For example, in one study, the half-life in children under 10 days ranged from 10 to 67 hours compared to a range of 7 to 13 hours in children greater than 2 months.
Pediatric patients (ie, between 3 months and 10 years) have 50% higher clearances expressed on weight (ie, mL/min/kg) than do adults. Over the age of 10 years, children have pharmacokinetic parameters that approximate those of adults.
The capacity of elderly patients (age range: 68 to 89 years) to eliminate valproate has been shown to be reduced compared to younger adults (age range: 22 to 26). Intrinsic clearance is reduced by 39%; the free fraction is increased by 44%. Accordingly, the initial dosage should be reduced in the elderly [see DOSAGE AND ADMINISTRATION].
Effect of Gender
There are no differences in the body surface area adjusted unbound clearance between males and females (4.8±0.17 and 4.7±0.07 L/hr per 1.73 m², respectively).
Effect of Race
The effects of race on the kinetics of valproate have not been studied.
Effect of Disease
Liver disease impairs the capacity to eliminate valproate. In one study, the clearance of free valproate was decreased by 50% in 7 patients with cirrhosis and by 16% in 4 patients with acute hepatitis, compared with 6 healthy subjects. In that study, the half-life of valproate was increased from 12 to 18 hours. Liver disease is also associated with decreased albumin concentrations and larger unbound fractions (2-to 2.6-fold increase) of valproate. Accordingly, monitoring of total concentrations may be misleading since free concentrations may be substantially elevated in patients with hepatic disease whereas total concentrations may appear to be normal [See BOXED WARNING, CONTRAINDICATIONS, WARNINGS AND PRECAUTIONS].
A slight reduction (27%) in the unbound clearance of valproate has been reported in patients with renal failure (creatinine clearance < 10 mL/minute); however, hemodialysis typically reduces valproate concentrations by about 20%. Therefore, no dosage adjustment appears to be necessary in patients with renal failure. Protein binding in these patients is substantially reduced; thus, monitoring total concentrations may be misleading.
Plasma Levels and Clinical Effect
The relationship between plasma concentration and clinical response is not well documented. One contributing factor is the nonlinear, concentration dependent protein binding of valproate which affects the clearance of the drug. Thus, monitoring of total serum valproate cannot provide a reliable index of the bioactive valproate species.
For example, because the plasma protein binding of valproate is concentration dependent, the free fraction increases from approximately 10% at 40 mcg/mL to 18.5% at 130 mcg/mL. Higher than expected free fractions occur in the elderly, in hyperlipidemic patients, and in patients with hepatic and renal diseases.
The therapeutic range in epilepsy is commonly considered to be 50 to 100 mcg/mL of total valproate, although some patients may be controlled with lower or higher plasma concentrations.
The effectiveness of valproate for the treatment of acute mania was demonstrated in two 3-week, placebo controlled, parallel group studies.
(1) Study 1: The first study enrolled adult patients who met DSM-III-R criteria for bipolar disorder and who were hospitalized for acute mania. In addition, they had a history of failing to respond to or not tolerating previous lithium carbonate treatment. Valproate was initiated at a dose of 250 mg TID and adjusted to achieve serum valproate concentrations in a range of 50-100 mcg/mL by day 7. Mean valproate doses for completers in this study were 1118, 1525, and 2402 mg/day at Days 7, 14, and 21, respectively. Patients were assessed on the Young Mania Rating Scale (YMRS; score ranges from 0-60), an augmented Brief Psychiatric Rating Scale (BPRS-A), and the Global Assessment Scale (GAS). Baseline scores and change from baseline in the Week 3 endpoint (last-observation-carried-forward[LOCF]) analysis were as follows:
|YMRS Total Score|
|Group||Baseline1||BL to Wk32||Difference3|
|BPRS–A Total Score|
|Group||Baseline1||BL to Wk32||Difference3|
|Group||Baseline1||BL to Wk32||Difference3|
|1 Mean score at baseline
2Change from baseline to Week 3 (LOCF)
3Difference in change from baseline to Week 3 endpoint (LOCF) between valproate and placebo
Valproate was statistically significantly superior to placebo on all three measures of outcome.
(2) Study 2: The second study enrolled adult patients who met Research Diagnostic Criteria for manic disorder and who were hospitalized for acute mania. Valproate was initiated at a dose of 250 mg TID and adjusted within a dose range of 750-2500 mg/day to achieve serum valproate concentrations in a range of 40-150 mcg/mL. Mean valproate doses for completers in this study were 1116, 1683, and 2006 mg/day at Days 7, 14, and 21, respectively. Study 2 also included a lithium group for which lithium doses for completers were 1312, 1869, and 1984 mg/day at Days 7, 14, and 21, respectively. Patients were assessed on the Manic Rating Scale (MRS; score ranges from 11-63), and the primary outcome measures were the total MRS score, and scores for 2 subscales of the MRS, i.e., the Manic Syndrome Scale (MSS) and the Behavior and Ideation Scale (BIS). Baseline scores and change from baseline in the Week 3 endpoint (LOCF) analysis were as follows:
|MRS Total Score|
|Group||Baseline1||BL to Day 212||Difference3|
|MSS Total Score|
|Group||Baseline1||BL to Day 212||Difference3|
|BIS Total Score|
|Group||Baseline1||BL to Day 212||Difference3|
|1Mean score at baseline
2Change from baseline to Day 21 (LOCF)
3Difference in change from baseline to Day 21 endpoint (LOCF) between valproate and placebo and lithium and placebo
Valproate was statistically significantly superior to placebo on all three measures of outcome. An exploratory analysis for age and gender effects on outcome did not suggest any differential responsiveness on the basis of age or gender.
A comparison of the percentage of patients showing ≥ 30% reduction in the symptom score from baseline in each treatment group, separated by study, is shown in Figure 1.
Figure 1: Percentage of
Patients Achieving ≥ 30% Reduction in Symptom Score From Baseline
The efficacy of valproate in reducing the incidence of complex partial seizures (CPS) that occur in isolation or in association with other seizure types was established in 2 controlled trials.
In one, multiclinic, placebo-controlled study employing an add-on design, (adjunctive therapy) 144 patients who continued to suffer 8 or more CPS per 8 weeks during an 8-week period of monotherapy with doses of either carbamazepine or phenytoin sufficient to assure plasma concentrations within the “therapeutic range” were randomized to receive, in addition to their original antiepilepsy drug (AED), either valproate or placebo. Randomized patients were to be followed for a total of 16 weeks. The following Table presents the findings.
Adjunctive Therapy Study
Median Incidence of CPS per 8 Weeks
|Add-On Treatment||Number of Patients||Baseline Incidence||Experimental Incidence|
|*Reduction from baseline statistically significantly greater for valproate than placebo at p ≤ 0.05 level.|
Figure 2 presents the proportion of patients (X axis) whose percentage reduction from baseline in complex partial seizure rates was at least as great as that indicated on the Y axis in the adjunctive therapy study. A positive percent reduction indicates an improvement (i.e., a decrease in seizure frequency), while a negative percent reduction indicates worsening. Thus, in a display of this type, the curve for an effective treatment is shifted to the left of the curve for placebo. This Figure shows that the proportion of patients achieving any particular level of improvement was consistently higher for valproate than for placebo. For example, 45% of patients treated with valproate had a ≥ 50% reduction in complex partial seizure rate compared to 23% of patients treated with placebo.
The second study assessed the capacity of valproate to reduce the incidence of CPS when administered as the sole AED. The study compared the incidence of CPS among patients randomized to either a high-or low-dose treatment arm. Patients qualified for entry into the randomized comparison phase of this study only if 1) they continued to experience 2 or more CPS per 4 weeks during an 8-to 12-weeklong period of monotherapy with adequate doses of an AED (ie, phenytoin, carbamazepine, phenobarbital, or primidone) and 2) they made a successful transition over a 2-week interval to valproate. Patients entering the randomized phase were then brought to their assigned target dose, gradually tapered off their concomitant AED and followed for an interval as long as 22 weeks. Less than 50% of the patients randomized, however, completed the study. In patients converted to valproate monotherapy, the mean total valproate concentrations during monotherapy were 71 and 123 mcg/mL in the low-dose and high-dose groups, respectively.
The following Table presents the findings for all patients randomized who had at least one post-randomization assessment.
Monotherapy Study Median
Incidence of CPS per 8 Weeks
|Treatment||Number of Patients||Baseline Incidence||Randomized Phase Incidence|
|*Reduction from baseline statistically significantly greater for high dose than low dose at p ≤ 0.05 level.|
Figure 3 presents the proportion of patients (X axis) whose percentage reduction from baseline in complex partial seizure rates was at least as great as that indicated on the Y axis in the monotherapy study. A positive percent reduction indicates an improvement (i.e., a decrease in seizure frequency), while a negative percent reduction indicates worsening. Thus, in a display of this type, the curve for a more effective treatment is shifted to the left of the curve for a less effective treatment. This Figure shows that the proportion of patients achieving any particular level of reduction was consistently higher for high dose valproate than for low dose valproate. For example, when switching from carbamazepine, phenytoin, phenobarbital, or primidone monotherapy to high-dose valproate monotherapy, 63% of patients experienced no change or a reduction in complex partial seizure rates compared to 54% of patients receiving low-dose valproate.
Both studies employed essentially identical designs and recruited patients with a history of migraine with or without aura (of at least 6 months in duration) who were experiencing at least 2 migraine headaches a month during the 3 months prior to enrollment. Patients with cluster headaches were excluded. Women of childbearing potential were excluded entirely from one study, but were permitted in the other if they were deemed to be practicing an effective method of contraception.
In each study following a 4-week single-blind placebo baseline period, patients were randomized, under double blind conditions, to valproate or placebo for a 12-week treatment phase, comprised of a 4-week dose titration period followed by an 8-week maintenance period. Treatment outcome was assessed on the basis of 4-week migraine headache rates during the treatment phase.
In the first study, a total of 107 patients (24 M, 83 F), ranging in age from 26 to 73 were randomized 2:1, valproate to placebo. Ninety patients completed the 8-week maintenance period. Drug dose titration, using 250-mg tablets, was individualized at the investigator's discretion. Adjustments were guided by actual/sham trough total serum valproate levels in order to maintain the study blind. In patients on valproate doses ranged from 500 to 2500 mg a day. Doses over 500 mg were given in 3 divided doses (TID). The mean dose during the treatment phase was 1087 mg/day resulting in a mean trough total valproate level of 72.5 mcg/mL, with a range of 31 to 133 mcg/mL.
The mean 4-week migraine headache rate during the treatment phase was 5.7 in the placebo group compared to 3.5 in the valproate group (see Figure 2). These rates were significantly different.
In the second study, a total of 176 patients (19 males and 157 females), ranging in age from 17 to 76 years, were randomized equally to one of three valproate dose groups (500, 1000, or 1500 mg/day) or placebo. The treatments were given in 2 divided doses (BID). One hundred thirty-seven patients completed the 8-week maintenance period. Efficacy was to be determined by a comparison of the 4-week migraine headache rate in the combined 1000/1500 mg/day group and placebo group.
The initial dose was 250 mg daily. The regimen was advanced by 250 mg every 4 days (8 days for 500 mg/day group), until the randomized dose was achieved. The mean trough total valproate levels during the treatment phase were 39.6, 62.5, and 72.5 mcg/mL in the valproate 500, 1000, and 1500 mg/day groups, respectively.
The mean 4-week migraine headache rates during the treatment phase, adjusted for differences in baseline rates, were 4.5 in the placebo group, compared to 3.3, 3.0, and 3.3 in the valproate 500, 1000, and 1500 mg/day groups, respectively, based on intent-to-treat results (see Figure 4). Migraine headache rates in the combined valproate 1000/1500 mg group were significantly lower than in the placebo group.
Figure 4: Mean 4-Week
Last reviewed on RxList: 6/28/2013
This monograph has been modified to include the generic and brand name in many instances.
Additional Stavzor Information
Report Problems to the Food and Drug Administration
You are encouraged to report negative side effects of prescription drugs to the FDA. Visit the FDA MedWatch website or call 1-800-FDA-1088.
Get tips on therapy and treatment.