You are on page 1of 6

Bipolar Disorders 2012: 14: 206210

2012 John Wiley and Sons A/S


Brief Report

Allopurinol augmentation in the outpatient treatment of bipolar mania: a pilot study

Fan A, Berg A, Bresee C, Glassman LH, Rapaport MH. Allopurinol augmentation in the outpatient treatment of bipolar mania: a pilot study. Bipolar Disord 2012: 14: 206210. 2012 The Authors. Journal compilation 2012 John Wiley & Sons A S. Background: Allopurinol promotes the salvage of purines, possibly increasing endogenous adenosine levels. Recent studies suggest that adenosine has neuroprotective and inhibitory eects. Two previous inpatient trials demonstrated that allopurinol has anti-manic activity. Our objective was to test allopurinol as an adjunct to standard medications in bipolar disorder manic outpatients. Methods: In this double-blind, placebo-controlled trial, 27 subjects who met DSM-IV criteria for bipolar disorder and scored 14 on the Young Mania Rating Scale (YMRS) were randomized to augmentation with allopurinol or placebo for six weeks. The primary ecacy measure was the YMRS. The primary safety measure was the Treatment Emergent Symptom Scale. Results: The eect of allopurinol augmentation in decreasing mean YMRS scores was modest, with an overall eect size of )0.25 (Cohens d). Allopurinol-treated individuals who abstained from caeine (n = 4) had a greater decrease in YMRS scores ()15.3 1.8) than subjects using caeine (n = 5) ()9.6 3.4, p = 0.219), with an eect size of )0.86. Conclusion: In this small outpatient pilot study, allopurinol augmentation did not show a statistically signicant improvement over placebo in attenuating manic symptoms. Subjects with restricted caeine use showed a greater eect size compared to caeine users. This nding may be interpreted as corroborating the hypothesized mechanism of action of allopurinols anti-manic eect in previous studies.

Alexander Fan, Andrew Berg, Catherine Bresee, Lisa Hayley Glassman and Mark Hyman Rapaport
Department of Psychiatry and Behavioral Neurosciences, Cedars-Sinai Medical Center, Los Angeles, CA, USA

doi: 10.1111/j.1399-5618.2012.01001.x Key words: adenosine allopurinol augmentation bipolar caffeine mania Received 10 June 2011, revised and accepted for publication 14 December 2011 Corresponding author: Alexander Fan, M.D. Department of Psychiatry and Behavioral Neurosciences Cedars-Sinai Medical Center 8730 Alden Drive, Room C301 Los Angeles, CA 90048 USA Fax: 310-423-8397 E-mail: Clinical Trials Registration: A double-blind, placebo-controlled augmentation study with allopurinol for treatment resistant mania, identier: NCT00643123.

Despite treatment advances, bipolar disorder is still associated with signicant functional impairment, morbidity, and mortality, as well as an economic burden for society. Although several classes of medications are eective for bipolar mania, these medications have signicant shortcomings, including intolerable side eects, narrow therapeutic indices, the need for regular serum level monitoring, and high monetary costs. Furthermore, a signicant proportion of bipolar disorder patients still fail to respond adequately to these currently accepted medications (1, 2). A growing body of evidence suggests that adenosine is a neurotransmitter with a pervasive

and generally inhibitory eect on neuronal activity (3, 4). Animal studies indicate that endogenous adenosine release has neuroprotective eects in brain hypoxia, ischemia, hypoglycemia, and mechanical cell damage (3). To date, four adenosine receptors (A1, A2A, A2B, and A3) have been characterized and an increasing volume of data indicates that adenosine negatively modulates dopamine activity through A2A receptors (5). Human studies nd that adenosine inuences sleep homeostasis (6). Researchers have proposed that adenosine receptor agonists and antagonists may have therapeutic potential in the treatment of neuropsychiatric disorders (5, 7). However, currently


Allopurinol augmentation in outpatient mania available adenosine agonists have peripheral eects that decrease heart rate, blood pressure, and body temperature, which greatly limit their utility. Allopurinol is a xanthine oxidase inhibitor routinely used to treat gout and hyperuricemia which inhibits the last step of purine degradation. The resulting increased xanthine levels are thought to promote the salvage of purines and increased adenosine levels (8). Stimulation of A2A receptors has been shown to decrease the anity of dopamine D2 agonist binding sites in the rat striatum (9), and this nding has led to clinical studies that demonstrate the antipsychotic and anticonvulsant properties of allopurinol (10, 11). Two augmentation studies have reported that allopurinol also has anti-manic activity in hospitalized bipolar disorder manic patients taking lithium or a combination of lithium and haloperidol (12, 13). The present study is the rst preliminary trial to test allopurinol augmentation in an outpatient sample of manic bipolar disorder patients taking a variety of standard medications for this disorder. We hypothesized that allopurinol augmentation would decrease the severity of mania more than placebo.

This study was approved by the Cedars-Sinai Medical Centers Institutional Review Board and all subjects provided signed informed consent to participate in the research. Subjects met the criteria for bipolar I disorder as conrmed by the Mini International Neuropsychiatric Interview (MINI) (14). All subjects were experiencing a manic or mixed-manic episode and were partial responders to lithium, valproic acid, carbamazepine, or atypical antipsychotic medications. Partial response was dened as a Young Mania Rating Scale (YMRS) score 14 while maintaining a conrmed therapeutic blood level of lithium, carbamazepine, or valproic acid, or a therapeutic dose of an atypical antipsychotic medication. All subjects were medically stable with normal medical histories, physical examinations, electrocardiograms, urine analyses, urine toxicology screens, and renal, hepatic, and thyroid function panels. Serum uric acid levels were conrmed to be within normal range. Medication adherence was conrmed by measurement of serum levels of lithium, valproic acid, and carbamazepine at both baseline and at study completion. Exclusion criteria included: substance abuse within one month prior to screening; mood symptoms severe enough to require hospitalization; current suicidal or homicidal ideation; or active comorbid Axis I psychiatric disorders. Women of childbearing

potential without adequate contraception were excluded. Subjects entered the treatment phase of the study one week after screening to conrm a YMRS score 14 and were randomized in a 1:1 ratio to either allopurinol or placebo augmentation. Subjects continued taking their current psychiatric medications during the study, with the dose of medications held constant for the duration of the study. Allopurinol was given at a xed dose of 300 mg day for the rst week and then 600 mg day for the remainder of the study. Four visits were conducted over the following six weeks. The primary outcome measure was the YMRS. The battery of assessments also included the 17-item version of the Hamilton Depression Rating Scale (HAM-D17) (15), Clinical Global Impression Scale (CGI) (16), Sheehan Disability Scale (SDS) (17), and the Quality of Life Enjoyment and Satisfaction Questionnaire (Q-LES-Q) (18). Side eects were recorded in each study visit and assessed by the Treatment Emergent Symptom Scale (TESS) (16). Daily caeine intake was recorded at baseline visit. Data were analyzed for homogeneity of variance and normal distribution. Comparisons of data across groups were made using a Students t-test for continuous data or an exact chi-square test for categorical data. An analysis of variance (ANOVA) was used where comparisons were to be made across more than two groups. Change from baseline scores were computed for all outcome measures and were analyzed using last observation carried forward (LOCF) for those subjects with at least one data observation after starting the study drug. Statistical analysis was performed using SAS v9.1 software (The SAS Institutes, Cary, NC, USA) and eect sizes were computed using the method of Cohen (19). Data were presented as mean the standard error of the mean (SEM). Given the exploratory nature of this study, an a priori power estimate was not performed.

A total of 40 subjects were screened for eligibility in the study; 13 were excluded and 27 were randomized to either allopurinol (n = 15) or placebo (n = 12). Twelve subjects from the allopurinol group and 11 from the placebo-treated group continued to the treatment phase of the study and provided data for analysis. The eect size in our results for the primary outcome measure (YMRS) was similar to that observed in many augmentation trials. Two subjects from the allopurinol group were removed from the study early


Fan et al.

at weeks 2 and 4 due to a worsening of depressive symptoms requiring medical intervention. There were no statistically meaningful dierences between subjects who terminated early and those who completed the study. At baseline, there were no statistically signicant dierences in age, gender, race, employment status, or psychological measures between the two groups of subjects, although subjects in the allopurinol group had statistically higher TESS scores (t = 3.26, df = 23, p = 0.003) (Table 1). Subjects were taking a variety of standard medications prescribed for bipolar disorder. The most common were second-generation antipsychotic agents (n = 20), anticonvulsants (n = 10), antidepressants (n = 8), lithium (n = 6), benzodiazepines (n = 5), and hypnotics (n = 1). All medications were taken within Food and Drug Administration (FDA)-approved therapeutic dose ranges. After six weeks of treatment, both the allopurinol and placebo groups had lower YMRS scores compared to their mean baseline scores, with allopurinol augmentation causing a slightly greater decrease in mean YMRS score (allopurinol: )11.5 1.6, placebo: )9.9 2.3; t = )0.58, df = 21, p = 0.568) (Fig. 1). The eect size was )0.25. Within the subgroup of subjects who were noncaeine users, we observed a decrease in YMRS scores of )15.3 1.8 with allopurinol treatment (n = 4) but only a )9.6 3.4 decrease with placebo treatment (n = 5) (t = )1.35, df = 7, p = 0.219). This corresponds to an eect size of )0.86. We also observed a greater decrease in
Table 1. Baseline measures by treatment group for all randomized subjects Allopurinol (n = 15) Age, years, mean (SEM) 39.7 (3.3) Female, n (%) 8 (53) Caucasian, n (%) 9 (60) Employed, n (%) 4 (27) Psychiatric assessments, mean (SEM) YMRS 21.0 (1.3) HAM-D17 9.5 (1.6) CGI-Severity 4.1 (0.1) Q-LES-Q 15.3 (2.5) TESS 24.2 (3.7) Placebo (n = 12) 46.6 5 8 4 18.9 6.3 3.9 12.1 11.1 (2.6) (42) (67) (33) (1.1) (1.5) (0.1) (1.8) (1.8)

Fig. 1. Mean standard error of the mean change in Young Mania Rating Scale scores from baseline stratied by caeineuse status. A = allopurinol-treated group; P = placebo-treated group.

TESS scores from baseline for the allopurinoltreated group ()8.42 2.66) than the placebotreated group ()2.73 2.85) (t = )1.46, df = 21, p = 0.159). There were no clinically relevant changes in the secondary measures: (i) HAM-D17 scores (allopurinol: )1.33 3.18, placebo: )2.46 2.08; t = 0.29, df = 21, p = 0.775); (ii) CGI scores (allopurinol: 0.25 0.13, placebo: 0.09 0.09; t = 0.98, df = 21, p = 0.337); (iii) SDS scores (allopurinol: )3.5 2.8, placebo: )4.5 1.8; t = 0.31, df = 21, p = 0.760); or (iv) Q-LES-Q scores (allopurinol: )0.4 5.7, placebo: 1.5 3.4; t = )0.29, df = 20, p = 0.778).

p-value 0.792a 0.547b 0.234b 0.655b 0.251a 0.160a 0.206a 0.403a 0.003a

SEM = standard error of the mean; YMRS = Young Mania Rating Scale; HAM-D17 = 17-item Hamilton Depression Rating Scale; CGI-Severity = Clinical Global Impression Severity Scale; Q-LES-Q = Quality of Life Enjoyment and Satisfaction Questionnaire; TESS = Treatment Emergent Symptom Scale. a p-values are the results of testing across groups via Students t-test. b p-values are the results of testing across groups via the chi-square test.

The results of this preliminary study suggest that allopurinol may have a limited potential as an augmentation treatment for bipolar mania. The results did not show a statistically signicant decrease in mean YMRS scores in the allopurinol-treated group compared to the placebo-treated group. This nding contrasts with two previous augmentation studies that tested allopurinol for bipolar mania (12, 13). These studies had larger sample sizes and were conducted in controlled inpatient settings with more severely manic subjects. Participants in these inpatient studies were restricted from caeine use, monitored more closely for medication compliance, and prevented from taking mood-altering substances of abuse. Our study design did not allow for the control of these factors. For example, our outpatient subjects were less severely manic than those of previous studies, and milder mania has a high spontaneous remission rate, which increases the diculty in detecting the therapeutic eect of a novel treatment.


Allopurinol augmentation in outpatient mania This study was the rst to measure the eect of caeine intake on the therapeutic action of allopurinol in neuropsychiatric disorders. Caeine intake was measured because it is a known adenosine receptor antagonist, giving it the ability to diminish the hypothesized adenosinergic eects of allopurinol. There is evidence that caeine has appreciable antagonistic action on A1 and A2A receptors after drinking as much as a one-cup coee equivalent (19). In a sub-analysis of caeineabstinent subjects compared to subjects who regularly ingest caeinated products, we observed a clinically relevant decrease in symptoms of mania in the caeine-abstinent group. Study subjects were considered as users of caeine if their daily intake exceeded one cup of coee equivalent per day. Although the results of this sub-analysis were not statistically signicant, the results were more consistent with the results of the previous studies showing a greater anti-manic eect in an inpatient setting (12, 13). These data provide indirect support suggesting that allopurinol aects adenosine neurotransmission, as caeine is a known adenosine receptor antagonist. This study was also the rst to measure the eects of allopurinol augmentation on depressive symptoms in bipolar disorder. Although two subjects randomized to the allopurinol augmentation group did develop a worsening of depressive symptoms, the HAM-D17 scores were not clinically or statistically dierent for the two groups. Another intriguing nding was the decreased side eects in the allopurinol group compared to the placebo group. Although this result was not statistically signicant, the decrease in treatmentemergent side eects with augmentation medication is a very uncommon nding in clinical trials for psychiatric medications. Notably, this result was consistent with that of an allopurinol augmentation study for schizophrenia that reported lower extrapyramidal symptoms in the allopurinol group compared to the placebo group (10). The limitations of this study were those common to all pilot outpatient clinical trials. These included small sample size, medication compliance, hidden substance abuse, and study population heterogeneity. However, the more naturalistic design of the present study provided valuable data for considering the potential utility of allopurinol in an inpatient setting versus an outpatient setting. Bipolar disorder patients who are frequently hospitalized may have more severe manias or more treatment-resistant manias which may call for innovative and heroic treatments. Despite the limitations of this study, our results strongly highlight the need for further, larger studies to clarify the eect of allopurinol on adenosine neurotransmission. Since previous studies have shown a protective and dopamine modulatory role of adenosine in the brain, allopurinol may emerge as a safe and cost-eective treatment option for bipolar disorder. Future neuropsychiatric clinical studies on allopurinol or adenosinergic agents will need to consider the eect of caeine intake as a confounding factor. Additionally, future studies should consider the use of validated measures to monitor caeine intake at multiple time points throughout the study.
MHR serves as a consultant to Aectis Pharmaceutics, Methylation Sciences, PAX Pharmaceuticals, Johnson & Johnson, Brain Cells, Inc., and Quintiles. AF, AB, CB, and LHG report no conicts of interest.

1. Gitlin M. Treatment-resistant bipolar disorder. Mol Psychiatry 2006; 11: 227240. 2. Gajwani P. Treatment-refractory bipolar disorder: classication to aid in clinical management. Expert Opin Pharmacother 2009; 10: 19071915. 3. Nayak GH, Prentice HM, Milton SL. Neuroprotective signaling pathways are modulated by adenosine in the anoxia tolerant turtle. J Cereb Blood Flow Metab 2011; 31: 467475. 4. Stone TW, Ceruti S, Abbracchio MP. Adenosine receptors and neurological disease: neuroprotection and neurodegeneration. Handb Exp Pharmacol 2009; 193: 535587. 5. Fuxe K, Marcellino D, Borroto-Escuela DO et al. Adenosine-dopamine interactions in the pathophysiology and treatment of CNS disorders. CNS Neurosci Ther 2010; 16: 1842. 6. Porkka-Heiskanen T, Kalinchuk AV. Adenosine, energy metabolism and sleep homeostasis. Sleep Med Rev 2011; 15: 123135. 7. Shah U, Hodgson R. Recent progress in the discovery of adenosine A(2A) receptor antagonists for the treatment of Parkinsons disease. Curr Opin Drug Discov Devel 2010; 13: 466480. 8. Marro PJ, Mishra OP, Delivoria-Papadopoulos M. Eect of allopurinol on brain adenosine levels during hypoxia in newborn piglets. Brain Res 2006; 1074: 444450. 9. Ferre S, von Euler G, Johansson B, Fredholm BB, Fuxe K. Stimulation of high-anity adenosine A2 receptors decreases the anity of dopamine D2 receptors in rat striatal membranes. Proc Natl Acad Sci USA 1991; 88: 72387241. 10. Akhondzadeh S, Safarcherati A, Amini H. Benecial antipsychotic eects of allopurinol as add-on therapy for schizophrenia: a double blind, randomized and placebo controlled trial. Prog Neuropsychopharmacol Biol Psychiatry 2005; 29: 253259. 11. Togha M, Akhondzadeh S, Motamedi M, Ahmadi B, Razeghi S. Allopurinol as adjunctive therapy in intractable epilepsy: a double-blind and placebo-controlled trial. Arch Med Res 2007; 38: 313316. 12. Akhondzadeh S, Milajerdi MR, Amini H, Tehrani-Doost M. Allopurinol as an adjunct to lithium and haloperidol


Fan et al.
for treatment of patients with acute mania: a double-blind, randomized, placebo-controlled trial. Bipolar Disord 2006; 8: 485489. 13. Machado-Vieira R, Soares JC, Lara DR et al. A doubleblind, randomized, placebo-controlled 4-week study on the ecacy and safety of the purinergic agents allopurinol and dipyridamole adjunctive to lithium in acute bipolar mania. J Clin Psychiatry 2008; 69: 12371245. 14. Sheehan DV, Lecrubier Y, Sheehan KH et al. The MiniInternational Neuropsychiatric Interview (M.I.N.I.): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 1998; 59 (Suppl. 20): 2233. 15. Hamilton M. A rating scale for depression. J Neurol Neurosurg Psychiatry 1960; 23: 5662. 16. Guy W. ECDEU Assessment Manual for Psychopharmacology. Rockville: U.S. Department of Health, Education, and Welfare, Public Health Service, Alcohol, Drug Abuse, and Mental Health Administration, National Institute of Mental Health, Psychopharmacology Research Branch, Division of Extramural Research Programs, 1976. 17. Sheehan DV. The Anxiety Disease. New York: Scribners, 1983. 18. Endicott J, Nee J, Harrison W, Blumenthal R. Quality of Life Enjoyment and Satisfaction Questionnaire: a new measure. Psychopharmacol Bull 1993; 29: 321326. 19. Cohen J. Statistical Power Analysis for the Behavioral Sciences, (2nd ed.). Hillsdale: Lawrence Earlbaum Associates, 1988.


Copyright of Bipolar Disorders is the property of Wiley-Blackwell and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.