Reduced valproate plasma levels possible after ... - Wiley Online Library

3 downloads 0 Views 60KB Size Report
Saraga M, Preisig M, Zullino DF. Reduced valproate plasma levels possible after introduction of efavirenz in a bipolar patient. Bipolar Disord 2006: 8: 415–417.
Copyright ª Blackwell Munksgaard 2006 Bipolar Disorders 2006: 8: 415–417

BIPOLAR DISORDERS

Case Report

Reduced valproate plasma levels possible after introduction of efavirenz in a bipolar patient Saraga M, Preisig M, Zullino DF. Reduced valproate plasma levels possible after introduction of efavirenz in a bipolar patient. Bipolar Disord 2006: 8: 415–417. ª Blackwell Munksgaard, 2006 The case of a bipolar patient with concomitant multidrug addiction is described who presented a decrease in valproate plasma level of more than 50% shortly after antiretroviral therapy was initiated. Despite important dose augmentations of up to 4 g/day, plasma levels were difficult to raise to the target concentration of 50 mg/dL. The possible mechanisms underlying the observed pharmacokinetic changes are discussed: (i) a uridine diphosphate-glucuronosyl transferase induction by efavirenz; and (ii) a plasma protein displacement of valproate by efavirenz with subsequent increased elimination of the free fraction.

Mania is associated with drug abuse and at-risk sexual behaviour and can thus be considered as an indirect risk factor for HIV infection. Indeed, studies have shown an increased prevalence of HIV infection in bipolar patients and vice-versa (1). Antiretroviral drugs are known to interact with the metabolism of many other drugs. Keeping plasma levels of mood stabilizers, such as lithium and valproate, stable is considered an important part of relapse prevention. Typically, plasma levels are controlled regularly in bipolar patients who are on lithium therapy because of the dangerous effects of high plasma levels and because a low plasma level increases the risk for a manic relapse. For patients on valproate, the medical practice is similar, though requiring less frequent controls, because interpersonal valproate plasma levels variation is lower than variations under lithium. Though it is not established that a low valproate plasma level carries a similar risk for manic relapse, it seems to be a likely hypothesis. Introducing HIV therapy in a bipolar patient might carry the risk for either reaching toxic levels

The authors of this paper do not have any commercial associations that might pose a conflict of interest in connection with this manuscript.

Michael Saraga, Martin Preisig and Daniele Fabio Zullino University Department of Adult Psychiatry, Lausanne, Switzerland

Key words: antiviral agents – glucuronidation – interaction – protein binding – valproate Received 24 January 2005, revised and accepted for publication 13 January 2006 Corresponding author: Michael Saraga, De´partement Universitaire de Psychiatrie Adulte, Site de Cery, CH-1008 Prilly, Lausanne, Switzerland. Fax: +41 21 643 62 54; e-mail: [email protected]

of the mood stabilizer, or lowering them and thus induce a new mood episode. A manic episode would be especially preoccupying in an HIV patient as he might put others in danger through sexually at-risk behaviour. Such a manic episode is also more likely to show compliance problems, potentially leading to the emergence of a resistant HIV-strain. We report the case of a young bipolar patient who presented low valproate plasma levels despite a high oral daily dosage, shortly after antiretroviral therapy was initiated. Potential interactions are discussed, focusing mainly on the possible role of efavirenz, the most likely candidate. To our knowledge, it is the first report implicating efavirenz as a potential inducer of valproate metabolism. Case report

Mr A, 30 years old, was diagnosed with bipolar disorder during an obvious manic episode (the psychiatric diagnosis until then was a borderline personality disorder). At that time, he had a 10 year history of cocaine and heroin addiction and was on a methadone maintenance program; he was also HIV positive for many years, but did not meet the indication criteria for antiretroviral chemotherapy.

415

Saraga et al. Table 1. Valproate plasma levels and corresponding daily valproate dose Plasma level Before tritherapy 70 mg/dl With concomitant tritherapy 29 mg/dl 31 mg/dl 33 mg/dl 41 mg/dl 47 mg/dl 52 mg/dl

Dose

1.5 g 1.5 2.0 2.5 3.0 3.5 4.0

g g g g g g

He was started on valproic acid, 1.5 g/day, and the episode resolved. On that dosage, the valproate plasma level was 70 mg/dL. Two months later, Mr A went through a short period of intense IV cocaine consumption and was admitted to a psychiatric hospital. At that time, he did not show any clear-cut manic symptom and was discharged after a few days. During his stay, he developed a subcutaneous infection that required oral antibiotics. Viraemia was over 610,000 copies/ mL, CD4 at 245/mm3 and he was started on antiretroviral chemotherapy: didanosine 400 mg/ day, lamivudine 300 mg/day and efavirenz 600 mg/ day. Three months later, he was again admitted to a psychiatric hospital because of intense cocaine consumption. This time he was clearly in a manic state that lasted for more than 6 weeks. The valproate plasma levels measured during this stay and the corresponding daily valproate dose are shown in Table 1. It is only after an augmentation of the dose to 4 g/day that the patient reached the target level. He was discharged on 4 g/day, still exhibiting mild manic symptoms that resolved over the following weeks. Two months later, his general practitioner reduced the dose to 1.5 g/day, because of diarrhoea. Despite this 60% reduction, 1 month later the plasma level still remained at 52 mg/dL. Discussion

The most striking observations after the introduction of the antiretroviral chemotherapy were the clear-cut decrease in valproate plasma level of more than 50% and the difficulty in increasing plasma levels despite the important dose augmentations (from 1.5 to 4 g/day). The target plasma level was reached only after prescribing double of the recommended dose. These elements suggest a change in the drug metabolism after antiretroviral chemotherapy was started. The fact that a few months later, plasma levels went up

416

on a much lower dose suggests that this change was transient. Efavirenz has the most complicated drug interaction profile of the three antiretroviral agents. It is a CYP3A4 and CYP2B6 substrate, a CYP3A4 inducer and inhibitor and a CYP2C9, CYP2C19, CYP2D6 and CYP1A2 inhibitor (2). In general, concomitant use of efavirenz may lower the plasma levels of CYP3A4 substrates. The effects of didanosine and lamivudine on the main metabolic enzymes do not seem to be clinically relevant. Potential interaction mechanisms for didanosine are related to the buffer agent in the formula and are probably not relevant to the present case. CYP3A4 does not seem to play a major role in valproate metabolism, which implies mainly three paths: b-oxidative metabolism, glucuronidation and to some extent via other cytochromes. Uridine diphosphate-glucuronosyl transferases (UDP) are inducible enzymes. The low plasma levels measured in our patient might thus possibly result from UDP induction by one of the antiretroviral agents added to his regimen. UDP induction has been documented for rifampicin, phenobarbital and other xenobiotics that are UDP substrates (3, 4). Efavirenz is an UDP substrate and may have induced an increase in valproate metabolism in our case as it has a similar effect on CYP3A4. A further possible explanation lies in the protein-binding affinities of valproate and efavirenz (although the clinical relevance of plasma protein binding has been questioned over the past few years) (5). Efavirenz is strongly bound to plasma proteins (6) and might displace valproate from its binding sites, transiently increasing its active, free fraction. The increase in valproate free fraction would result in increased valproate elimination, until a new steady state is reached, with a lower total valproate plasma concentration but a similar free valproate concentration. Measuring the free fraction would solve the question but was not done in our case. Still, the pharmacological and clinical picture (a manic episode with low total valproate plasma level, a very slow clinical response with a very slow increase in the drug level) leads us to believe that valproate metabolism was increased in our patient after efavirenz initiation, resulting in a low valproate plasma level and a manic relapse. Though such an interaction is far from established with this single case report, physicians should check their patientsÕ mood stabilizer plasma levels before and shortly after the introduction of a complex treatment such as antiretroviral therapy, and alter the dosage. A manic episode in an HIV

Valproate – antiretroviral interaction patient might result in dire consequences and prudence here is a cardinal virtue. Clinical follow up, plasma level monitoring and sometimes more specific measures (such as free fraction determination) would contribute to lowering the risk of a relapse with potential disastrous consequences. References 1. Otto-Salaj LL, Stevenson LY. Influence of psychiatric diagnoses and symptoms on HIV risk behavior in adults with serious mental illness. AIDS Read 2001; 11: 197–198. 2. Zapor MJ, Cozza KL, Wynn GH, Wortmann GW, Armstrong SC. Antiretrovirals, part II: focus on

3.

4.

5.

6.

non-protease inhibitor antiretrovirals (NRTIs, NNRTIs, and fusion inhibitors). Psychosomatics 2004; 45: 524–535. Soars MG, Petullo DM, Eckstein JA, Kasper SC, Wrighton SA. An assessment of udp-glucuronosyltransferase induction using primary human hepatocytes. Drug Metab Dispos 2004; 32: 140–148. Tukey RH, Strassburg CP. Human UDP-glucuronosyltransferases: metabolism, expression, and disease. Annu Rev Pharmacol Toxicol 2000; 40: 581–616. Benet LZ, Hoener BA. Changes in plasma protein binding have little clinical relevance. Clin Pharmacol Ther 2002; 71: 115–121. Boffito M, Back DJ, Blaschke TF et al. Protein binding in antiretroviral therapies. AIDS Res Hum Retroviruses 2003; 19: 825–835.

417