What Is the Best Strategy for Converting from Twice-Daily Divalproex to a Once-Daily Divalproex ER Regimen?

Ronald C. Reed; Sandeep Dutta


Clin Drug Invest. 2004;24(9) 

In This Article


Published data suggest that there is a relationship between plasma VPA concentration and seizure control,[6,7,8,22] with 50 mg/L total VPA concentration being required for effect and a VPA concentration above 110 mg/L resulting in a greater incidence of adverse effects.[22] Our pharmacokinetic simulations predict that the chance for subtherapeutic total VPA concentrations, resulting in breakthrough seizures, or high plasma VPA concentrations, resulting in clinical toxicity, is low following an immediate conversion strategy from delayed-release, enteric-coated divalproex q12h to divalproex ER once daily, as long as this conversion is initiated 12 hours after the last dose of conventional divalproex.

Particular attention should be focused on patients concomitantly taking enzyme-inducing AEDs converting from divalproex q12h to divalproex ER once daily who could be instructed by their clinician to wait 24 hours after taking their regular divalproex q12h dose before starting their once-daily divalproex ER. It is easy to comprehend why physicians may intuitively believe that waiting 24 hours to convert is necessary, since it appears that a patient would be taking too much divalproex within a 24-hour period (taking the first dose of a divalproex q12h regimen combined with the full daily dose of divalproex ER just 12 hours later, i.e. 150% of the total daily divalproex dose taken in the same day). Hence, physicians who instruct patients to follow a delayed conversion strategy may be attempting to limit the total dose burden by spreading the divalproex and/or divalproex ER dose out over a longer period. However, our simulations clearly predict that there will be a rapid decline in plasma total VPA concentrations if such a delayed conversion strategy is used, potentially resulting in clinical consequences for the patient. The decline in plasma VPA was most pronounced for the delayed conversion strategy in induced patients; other conversion strategies tested did not have this magnitude of effect.

It should be noted that the decline in plasma VPA observed with any conversion strategy tested here does not represent any inherent flaw in the conventional enteric-coated or the extended-release divalproex dosage forms; rather, it reflects the short elimination half-life of the VPA molecule itself. A rapid decline in plasma VPA concentration can be expected in many patients, especially following the administration of the conventional divalproex formulation. The rate of decline in plasma VPA is expected to be less for the divalproex ER formulation, because of deliberate physical engineering of the formulation to release drug at an apparent zero-order rate over ~22 hours.[18]

Alternative conversion strategies (as represented in panels b-d of figures 1-3) do not seem to offer any real pharmacokinetic advantage over an immediate, all-at-once dose conversion strategy. In fact, these alternative conversion strategies are unnecessarily complicated and still result in moderate to substantial perturbation in plasma VPA concentrations, leading to subtherapeutic plasma VPA levels. More importantly, patients and/or pharmacists may misunderstand dosing instructions while attempting to follow physicians' prescriptions for a stepwise or mixed conversion strategy, probably leading to incorrect dosing and subsequently adverse clinical consequences. Ultimately, misunderstood instructions can have clinical repercussions: patients experiencing any number and/or an increasing severity of adverse effects from their AEDs are much more likely to be noncompliant with their treatment regimen, hence precipitating breakthrough symptoms.

The immediate conversion strategy is applicable regardless of whether the patient is instructed to take once-daily divalproex ER in the morning or evening. Some physicians may prescribe divalproex ER as a once-daily regimen to be taken in the evening, with the expectation of further reducing the potential for VPA-related adverse effects. Since there is minimal biopharmaceutical difference between divalproex ER administered in the morning or in the evening,[21] the immediate conversion strategy is applicable for once-daily morning or evening administration. Practically speaking, for clinicians who intend to administer divalproex ER once daily in the morning, our pharmacokinetic simulation data indicate that the night-time dose of divalproex q12h should be given as usual, with the entire once-daily divalproex ER dose starting the next morning, i.e. the immediate conversion. For clinicians who intend to administer divalproex ER once daily in the evening, the morning-time dose of divalproex q12h should be given as usual, with the entire once-daily dose of divalproex ER starting the same evening as part of the immediate conversion strategy.

The predicted steady-state plasma VPA Cmin and Cmax concentrations derived in our simulation study are largely dependent on our choice of the most commonly used total daily doses of conventional divalproex used in patients with epilepsy, namely, 1250mg (625mg q12h) in monotherapy and 2500mg (1250mg q12h) in induced polytherapy patients, and even higher total daily doses that are not uncommonly encountered in induced patients, such as 4000mg (2000mg q12h) divalproex. Considering that divalproex ER is ~89% bioavailable,[18] larger daily doses of divalproex ER are given to compensate for and achieve equivalent VPA exposure.[16,17] Thus, the 1500, 3000 and 4500mg divalproex ER doses chosen for our simulation study correspond to total daily doses of 1250mg (i.e. ~18 mg/kg in a 70kg patient), 2500mg (~36 mg/kg) and 4000mg (~57 mg/kg) of conventional, enteric-coated divalproex. Total daily divalproex doses of 18-36 mg/kg are commonly considered the usual effective doses in the treatment of epilepsy in adolescents and adults.[1,2,13,22,23,24,25] Moreover, the higher dose simulations, 2000mg divalproex q12h to 4500mg divalproex ER therapy given daily, that have been performed take into consideration patients who may have an intractable seizure disorder and are likely to be receiving several concomitant enzyme-inducing antiepileptic medications. This higher total daily divalproex dose of up to 60 mg/kg is still within the dosage recommendations of the package insert for divalproex[13] and the published literature.[22,25] In particular, a randomised, double-blind, parallel prospective study in patients with poorly controlled complex partial seizures demonstrated a statistically significant reduction in both partial and secondarily generalised seizures with a high total daily divalproex dose (56 mg/kg/day) as monotherapy, compared with a 'usual dose' (16 mg/kg/day).[22] Overall, we believe our choice of total daily divalproex dosages truly represents a clinically meaningful spectrum of doses that will be encountered in adult patients with epilepsy. This broad range of doses examined under different patient scenarios makes our findings concerning the optimal conversion strategy generalisable. We consistently found across all three patient scenarios tested that conversion from divalproex q12h to once-daily divalproex ER regimen via a delayed, stepwise and/or mixed-conversion strategy yielded no clear advantage, in terms of the pattern of perturbation in plasma VPA concentrations, over an immediate, all-at-once conversion strategy.

It should be recognised that the theoretical plasma VPA simulation profiles generated here apply to young to middle-aged adults, not to paediatric or geriatric patients, since adult VPA pharmacokinetic parameters were used in this model. VPA pharmacokinetic parameters (clearance, volume of distribution, protein binding) change in the very young (<10 years)[26,27] and elderly,[28] and hence age-specific parameters would be needed to accurately assess changes in plasma VPA concentrations over time in this simulation model. The data generated from our simulations and the subsequent recommendations in this study should not be automatically applied to such distinctly different patient populations; further studies in these age groups are required.

Every possible alternative strategy involving a switch from divalproex to divalproex ER could not be analysed; more research using this computer simulation model would be needed to define plasma VPA concentration-time profiles in other scenarios. Further simulations would be required to determine the perturbation in plasma VPA concentrations for any patient converting from conventional divalproex administered every 6 hours or every 8 hours (e.g. patients with epilepsy receiving several concomitant enzyme-inducing AEDs) to divalproex ER administered once daily.

In summary, we examined, via comprehensive pharmacokinetic simulations, the impact of four distinct q12h enteric-coated divalproex to once-daily divalproex ER conversion strategies (namely, immediate, delayed, stepwise and mixed conversion), on perturbations in plasma VPA concentrations within the first 48 hours after conversion. We found that a delayed conversion strategy (waiting 24 hours after the last dose of divalproex q12h to convert to once-daily divalproex ER) produced the largest perturbation in plasma total VPA concentrations. This delayed conversion strategy is likely to have the greatest potential for adverse clinical consequences, by virtue of the rapid decline in plasma total VPA concentrations. Examination of our tabular results and/or mere visual inspection of the plasma VPA concentration-time profiles do not indicate any advantage for the stepwise or mixed conversion strategies when compared with the immediate conversion strategy. Our simulation findings are applicable across a wide range of divalproex total daily doses for both uninduced and induced patients with epilepsy. Clinicians under the mistaken impression that conversion from divalproex q12h to divalproex ER once daily requires small steps, fancy dose conversion strategies, and/or tinkering or tweaking of the divalproex or divalproex ER doses are cautioned that such an approach is not really necessary or advisable. The immediate, all-at-once conversion strategy is convenient and simple for the patient to understand, and is therefore recommended.