Omega 3 Fatty Acids in Bipolar Disorder
 
                                     A Preliminary Double-blind, Placebo-Controlled Trial
 
                                         Andrew L. Stoll, MD; W. Emanuel Severus, MD, PhD; Marlene P. Freeman, MD;
                                     Stephanie Rueter; Holly A. Zboyan; Eli Diamond; Kimberly K. Cress, MD; Lauren B.
                                     Marangell, MD

                                     Background  3 Fatty acids may inhibit neuronal signal transduction pathways in a
                                     manner similar to that of lithium carbonate and valproate, 2 effective treatments for
                                     bipolar disorder. The present study was performed to examine whether 3 fatty acids
                                     also exhibit mood-stabilizing properties in bipolar disorder.

                                     Methods  A 4-month, double-blind, placebo-controlled study, comparing 3 fatty acids
                                     (9.6 g/d) vs placebo (olive oil), in addition to usual treatment, in 30 patients with bipolar
                                     disorder.

                                     Results  A Kaplan-Meier survival analysis of the cohort found that the 3 fatty acid
                                     patient group had a significantly longer period of remission than the placebo group
                                     (P=.002; Mantel-Cox). In addition, for nearly every other outcome measure, the 3
                                     fatty acid group performed better than the placebo group.

                                     Conclusion  3 Fatty acids were well tolerated and improved the short-term course of
                                     illness in this preliminary study of patients with bipolar disorder.

                                     Arch Gen Psychiatry. 1999;56:407-412
 
 

                                     BIPOLAR DISORDER (manic-depressive illness) is a common neuropsychiatric illness
                                     with a high morbidity and mortality.1 Despite available mood-stabilizing drugs, such as
                                     lithium carbonate and valproate, the illness is characterized by high rates of
                                     recurrence.1, 2 Recent research suggests that all of the currently available
                                     mood-stabilizing drugs have inhibitory effects on neuronal signal transduction systems.
                                     These findings have led to the hypothesis that overactive cell-signaling pathways may
                                     be involved in the pathophysiological mechanisms underlying bipolar disorder.3-6 By
                                     using this model of mood stabilizer action based on suppression of neuronal signal
                                     transduction mechanisms, novel mood-stabilizing agents can be rationally developed.
                                     One promising group of compounds is the 3 fatty acids, obtained from marine or plant
                                     sources.7 Among other effects, the ingestion of large amounts of 3 fatty acids is
                                     associated with a general dampening of signal transduction pathways associated with
                                     phosphatidylinositol, arachidonic acid, and other systems.8, 9 Thus, 3 fatty acids may
                                     be useful in conditions such as bipolar disorder, where the pathophysiological process
                                     may involve overactivity of cell signal transduction.

                                     We hypothesized that orally administered 3 fatty acids would exhibit inhibitory effects
                                     on signal transduction mechanisms in human neuronal membranes, and that high-dose
                                      3 fatty acids would be an effective mood stabilizer in bipolar disorder. The goal of this
                                     preliminary study was to assess the subacute mood-stabilizing effects of 3 fatty acids
                                     in patients with unstable bipolar disorder.
 

 

                                     PATIENTS AND METHODS
 
 

                                     OVERVIEW
 
                                     This was a 4-month, parallel-group, placebo-controlled, double-blind pilot study in which
                                     outpatients with bipolar disorder were randomized to receive either 3 fatty acids or
                                     placebo, in addition to their ongoing usual treatment.

                                     PATIENTS
 
                                     Participating subjects were men and women, 18 to 65 years old, who met DSM-IV10
                                     criteria for bipolar disorder (types I or II), and were free of notable medical and
                                     psychiatric comorbidity. The diagnosis of bipolar disorder was established by means of
                                     all available clinical information, including the mood disorder module of the Structured
                                     Clinical Interview for DSM-IV.11 Patients were required to have had at least 1 manic or
                                     hypomanic episode within the past year, because the expected high risk of recurrence
                                     in this subgroup1 enhanced the power of the study to detect a difference between the
                                     2 treatment groups within the study period. Forty percent of the study cohort had
                                     rapid-cycling symptoms, defined as 4 or more mood episodes in the 1 year before
                                     enrollment in the study.12 Patients were permitted to continue with their outpatient
                                     psychiatrist or psychotherapist, but no new psychotherapy treatment was started.
                                     Subjects receiving other medications at study entry continued to receive these
                                     medications at constant dosages, whether or not they were in the therapeutic range.

                                     Table 1 summarizes the demographic and clinical characteristics of the study subjects.
                                     This study was approved by the human studies committees of Brigham and Women's
                                     Hospital, Boston, Mass, and Baylor College of Medicine, Houston, Tex, and all
                                     participating patients gave written informed consent after receiving a full explanation of
                                     the study.

                                     STUDY PROCEDURES
 
                                     During the baseline visit, a detailed psychiatric and medical history was obtained, and
                                     the following standard rating scales were performed: Structured Clinical Interview for
                                     DSM-IV screening questions for current mania and depression, Young Mania Rating
                                     Scale13 (11-item structured interview version), Hamilton Rating Scale for Depression14
                                     (31-item structured interview version), investigator- and patient-rated Clinical Global
                                     Impression scale,15 the Global Assessment Scale,10 and a brief adverse-effect rating
                                     scale. The rating scales were repeated during office visits at weeks 2, 4, 6, 8, 12, and
                                     16. Because of a presumed delay in the therapeutic effects of 3 fatty acids, a priori
                                     criteria mandated that subjects remain in the study for 30 days or more to be included
                                     in the analysis. Identical gelatin capsules containing concentrated 3 fatty acid ethyl
                                     esters or placebo (olive oil ethyl esters) were obtained from the Fish Oil Test Materials
                                     Program, a joint research program of the National Institutes of Health and the National
                                     Marine Fisheries Service. Each capsule of 3 fatty acid concentrate contained 440 mg
                                     of eicosapentanoic acid (C20:5,3) and 240 mg of docosahexanoic acid (C22:6,3),
                                     which was vacuum deodorized and supplemented with tertiary-butylhydroquinone, 0.2
                                     mg/g, and tocopherols, 2 mg/g, as antioxidants. The source of the 3 fatty acids was
                                     menhaden fish body oil concentrate.

                                     Subjects were randomized by the Brigham and Women's Hospital Research Pharmacy
                                     to receive either 3 fatty acid treatment or placebo. The randomization was stratified
                                     according to sex, the presence or absence of concurrent lithium treatment, and the
                                     presence or absence of rapid cycling. Subjects received 7 capsules twice daily, for a
                                     total daily 3 fatty acid dosage of 6.2 g of eicosapentanoic acid and 3.4 g of
                                     docosahexanoic acid. Patients randomized to placebo also received 7 identical
                                     capsules twice daily. A relatively high dosage of eicosapentanoic acid and
                                     docosahexanoic acid was used, because similar doses have been safely and
                                     effectively administered in other disease states. Furthermore, because of the lack of
                                     data regarding the effective dosage of 3 fatty acids in mood disorders, a relatively
                                     high dosage was chosen to avoid a potentially ineffective low dose. Blood levels of 3
                                     fatty acids were not monitored in this trial.

                                     OUTCOME MEASURES
 
                                     The main outcome measure chosen a priori was the duration of time to exit
                                     double-blind treatment because of symptoms of bipolar disorder of sufficient severity to
                                     warrant a change in medication. Specifically, patients ended their participation in the
                                     study and treatment was considered to have change in medication. Specifically,
                                     patients ended their participation in the study and treatment was considered to have
                                     failed if mood symptoms emerged, or continued beyond 30 days in patients who were
                                     not euthymic at baseline. Hence, duration of time in the study represented an overall
                                     measure of treatment efficacy. The two blinded principal investigators (A.L.S. and
                                     L.B.M.), in collaboration with each patient, were responsible for the decision whether to
                                     end a patient's participation in the study. Secondary outcome measures were the
                                     results of the Young Mania Rating Scale, Hamilton Rating Scale for Depression, Clinical
                                     Global Impression, and Global Assessment Scale ratings, before and after treatment.

                                     STATISTICAL ANALYSIS
 
                                     A power calculation was performed before the study to determine the appropriate
                                     sample size. Assuming a large effect size, we calculated that 60 patients (including
                                     dropouts) would be sufficient to demonstrate a difference between the 2 arms at 90%
                                     power with an  .05.

                                     The study was originally intended to include 60 randomized patients, each for 9
                                     months of double-blind treatment. However, an unexpected cessation of production by
                                     the National Marine Fisheries Fish Oil Program led to a shortage of material.
                                     Simultaneously, a preplanned, blinded, interim analysis performed when 20 subjects
                                     had either failed treatment or completed 4 months suggested significant differences
                                     between the groups. The combination of these 2 factors led us to end accrual and
                                     then reanalyze the data after 30 patients had either failed treatment or completed at
                                     least 4 months of follow-up. A standard sequential design would prescribe looking for a
                                     P value of .02 or less to signal significance on the first interim analysis, and a P value
                                     of .04 or less to signal significance on the final analysis. Because of the 2 factors cited
                                     above, the results in this study fall between the interim and final analysis, and the P
                                     value designating significance could be taken conservatively as .015 or liberally as
                                     .042. A Kaplan-Meier "survival" analysis (Mantel-Cox log-rank statistic; df=1) was used
                                     to compare the duration of remission in the 2 groups. The rating scale scores on the
                                     last day of the study for each patient were used as the "final" data points (last
                                     observation carried forward). Categorical variables were analyzed by means of the
                                     Fisher exact test. Continuous variables were examined with the nonparametric
                                     Mann-Whitney test. Statistical significance for the primary outcome measure was set at
                                        <.01 (2 tailed).

                                     Forty-four patients were randomized, but only 30 had evaluable data, based on the a
                                     priori criteria for inclusion. Four subjects dropped out before the 1 month point because
                                     of noncompliance with the study protocol (n=2), gastrointestinal tract side effects (n=1),
                                     or concern over the possibility of receiving placebo (n=1). The remaining 10 subjects
                                     had not yet reached the 4-month end point required for the main outcome measure
                                     when the trial was ended and therefore were not included in the analysis.
 

 

                                     RESULTS
 
 

                                     The results for the 30 patients with evaluable data, as defined above, are presented
                                     herein. There were no significant differences in the demographic and baseline clinical
                                     characteristics of the 3 fatty acid and placebo groups (Table 1). Figure 1 depicts a
                                     Kaplan-Meier survival analysis of the study cohort. The duration of time remaining in
                                     the study was significantly greater in the 3 fatty acid­treated group when compared
                                     with placebo (P=.002; Mantel-Cox, log-rank statistic, 21=9.990). The time to a 50%
                                     rate of ending the study prematurely ("nonresponse") was 65 days for the placebo
                                     group, reflecting the unstable nature of the study population. A post hoc analysis was
                                     also performed for the subgroup of 8 subjects who entered the study while receiving no
                                     other mood-stabilizing drugs. As was observed in the whole study cohort, the 4
                                     subjects who received 3 monotherapy remained in remission for a significantly longer
                                     time than the 4 subjects who received placebo monotherapy (Figure 2; P=.04;
                                     Mantel-Cox). Other post hoc analyses showed that sex, the presence or absence of
                                     rapid cycling, and the type of bipolar disorder (I vs II) did not predict response to 3
                                     fatty acids, although the number of subjects in each cell was small.

                                     Table 1 displays the comparison of the secondary outcome measures between the 3
                                     and placebo groups. For nearly every outcome measure, the 3 fatty acid group
                                     performed better than the placebo group.

                                     Three patients developed side effects of the study drug and were permitted to lower
                                     the dosage to a minimum of 5 capsules twice daily. The most common adverse effect in
                                     both the 3 and olive oil groups was mild gastrointestinal tract distress, generally
                                     characterized by loose stools. Of the patients with adverse effect data at week 4 of the
                                     trial, 8 (62%) of 13 3-treated subjects complained of mild gastrointestinal tract side
                                     effects, whereas 8 (53%) of 15 placebo-treated subjects experienced gastrointestinal
                                     tract side effects (P=.72 by Fisher exact test; 2 subjects with missing data). No other
                                     adverse effects appeared with significant frequency or severity, and overall the patients
                                     tolerated the trial well. No research subjects were hospitalized or developed marked
                                     suicidal ideation or behavior. Demographic and clinical data for each subject are listed
                                     in Table 2.
 

 

                                     COMMENT
 
 

                                      3 Fatty acids used as an adjunctive treatment in bipolar disorder resulted in significant
                                     symptom reduction and a better outcome when compared with placebo in this pilot
                                     study. Improvement was significantly greater in the 3 fatty acid group than the olive oil
                                     control group on almost every assessment measure. The striking difference in relapse
                                     rates and response appeared to be highly clinically significant.

                                     These pilot results are intriguing and suggest that the addition of 3 fatty acids
                                     improved the subacute course of illness in this cohort of patients with bipolar disorder.
                                     The baseline clinical state of the research subjects in this study did not permit an
                                     evaluation of the antimanic effects of 3 fatty acids. Although the study was also not
                                     designed to provide definitive data on antidepressant effects, most of the patients
                                     receiving placebo who were considered treatment failures exhibited depressive
                                     exacerbations or recurrence. The suggestion of antidepressant effects of 3 fatty acids
                                     in this cohort of patients is noteworthy and warrants further study.

                                     Although this was a double-blind, placebo-controlled study, several methodological
                                     factors must be considered. The mixture of bipolar types I and II, varied mood states at
                                     study entry, and varying concomitant medications was a less rigorous design than in
                                     the ideal clinical trial. The variability in the clinical profiles of the study patients was
                                     controlled to some degree by stratifying the randomization for sex, concurrent lithium
                                     treatment, and rapid cycling. It would be ideal, although impossible in a small study,
                                     also to stratify for other variables. However, the randomization did result in a
                                     comparable representation of key variables in the active and control groups, including
                                     concomitant medications and baseline mood state.

                                     A further concern is the potential compromise of the blind. A distinct "fishy" aftertaste
                                     was episodically reported by subjects in both groups, but more often in the 3 group.
                                     When patients were asked to guess their randomization status, 86% of the 3 group
                                     guessed correctly, compared with 63% of the placebo group. Although in some cases
                                     the guess was based on the presence of a fishy aftertaste, in many cases it was based
                                     on the patient's perceived clinical response (or lack thereof in the placebo group).
                                     Correctly guessing a putative active treatment in the presence of a good clinical
                                     response is probably unavoidable. However, the possibility that the 3 group exhibited
                                     a placebo effect must be considered. Future studies to replicate and extend these
                                     findings should consider strategies to improve the blind, such as using a lower dose of
                                      3 fatty acids to reduce the frequency of the fishy aftertaste, or alternatively adding a
                                     small amount of a fishy-tasting substance to the placebo.

                                     If the results of this study are correct, and 3 fatty acids do possess mood-stabilizing
                                     action, then there are tangible implications for our understanding of the
                                     pathophysiological mechanisms of bipolar disorder and for the development of future
                                     treatments. Biochemical studies of human white blood cells show that high-dose
                                     therapy with 3 fatty acids leads to the incorporation of these polyunsaturated
                                     compounds into the membrane phospholipids crucial for cell signaling.8, 16 Increased
                                     concentrations of 3 fatty acids in membrane phospholipids appears to suppress
                                     phosphatidylinositol-associated signal transduction pathways.8, 16 The precise
                                     mechanism of this effect remains unclear. However, the incorporation of the
                                     polyunsaturated 3 fatty acids into the lipid bilayer of the cell membrane alters the
                                     physical and chemical properties of the membrane,17 possibly producing a local
                                     environment in which the membrane phospholipids are more resistant to hydrolysis by
                                     phospholipases. This could result in reduced generation of the second messenger
                                     molecules diacylglycerol and inositol triphosphate, thereby producing less activation of
                                     "downstream" intracellular signaling molecules, such as protein kinase C and calcium
                                     ion (Figure 3).

                                     As in peripheral tissues, the 3 fatty acids are also highly incorporated into neuronal
                                     phospholipids in animal models.18 Thus, it is possible that the 3 fatty acids also inhibit
                                     signal transduction mechanisms in the human central nervous system. Recent work by
                                     several investigators3-6 strongly suggests that the mechanism of action of typical mood
                                     stabilizers, such as lithium and valproate, involves a similar inhibition of postsynaptic
                                     signal transduction processes (Figure 3).

                                     Our results support other data suggesting that the mechanism of action of mood
                                     stabilizers in bipolar disorder is the suppression of aberrant signal transduction
                                     pathways. This is consistent with a model of abnormal signal transduction as the
                                     pathophysiological basis of bipolar disorder. If further studies confirm their efficacy in
                                     bipolar disorder, 3 fatty acids may represent a new class of membrane-active
                                     psychotropic compounds, and may herald the advent of a new class of rationally
                                     designed mood-stabilizing drugs.
 

 
 
                                     Author/Article Information

 
                                     From the Psychopharmacology Unit, Division of Psychiatry, Brigham and Women's
                                     Hospital (Drs Stoll, Severus, and Freeman, Ms Rueter, and Mr Diamond), and
                                     Department of Psychiatry, Harvard Medical School (Drs Stoll and Freeman), Boston,
                                     Mass; Free University of Berlin, Berlin, Germany (Dr Severus); and Department of
                                     Psychiatry, Baylor College of Medicine, Houston, Tex (Ms Zboyan and Drs Cress and
                                     Marangell). Dr Stoll is now with the Psychopharmacology Research Laboratory, McLean
                                     Hospital, Belmont, Mass, and continues with the Department of Psychiatry, Harvard
                                     Medical School.
 
                                     Reprints: Andrew L. Stoll, MD, Psychopharmacology Research Laboratory, McLean
                                     Hospital, 115 Mill St, Belmont, MA 02478 (e-mail: alstoll@mclean.harvard.edu).

                                     Accepted for publication October 2, 1998.

                                     This study was supported in part by a grant from the National Alliance for Research in
                                     Schizophrenia and Depression (NARSAD), Chicago, Ill (Drs Stoll and Marangell).
                                     Capsules of 3 fatty acids and matching olive oil placebo were provided by the Fish Oil
                                     Test Materials Program, a joint research program of the National Institutes of Health,
                                     Bethesda, Md, and the Southeastern Fisheries Science Center, National Marine
                                     Fisheries Service of the National Oceanic and Atmospheric Administration, Charleston,
                                     SC.

                                     Presented in part at the 36th Annual Meeting of the American College of
                                     Neuropsychopharmacology, Waikoloa, Hawaii, December 10, 1997.

                                     We thank John Orav, PhD, for his assistance with the statistical analysis.
 
 

 

                                     REFERENCES
 
 

                                     1.
                                     Goodwin FK, Jamison KR.
                                     Manic Depressive Illness.
                                     Oxford, England: Oxford University Press; 1990.

                                     2.
                                     Gitlin MJ, Swendsen J, Heller TL, Hammen C.
                                     Relapse and impairment in bipolar disorder.
                                     Am J Psychiatry.
                                     1995;152:1635-1640.
                                     MEDLINE

                                     3.
                                     Stoll AL, Severus E.
                                     Mood stabilizers: shared mechanisms of action at post-synaptic signal transduction and
                                     kindling processes.
                                     Harvard Rev Psychiatry.
                                     1996;4:77-89.

                                     4.
                                     Berridge MJ, Downes CP, Hanley MR.
                                     Lithium amplifies agonist-dependent phosphatidylinositol responses in brain and
                                     salivary glands.
                                     Biochem J.
                                     1982;206:587-595.
                                     MEDLINE

                                     5.
                                     Chen G, Manji HK, Hawver DB, Wright CB, Potter WZ.
                                     Chronic sodium valproate selectively decreases protein kinase C a and e in vitro.
                                     J Neurochem.
                                     1994;63:2361-2364.
                                     MEDLINE

                                     6.
                                     Manji HK, Bersudsky Y, Chen G, Belmaker RH, Potter WZ.
                                     Modulation of protein kinase C isozymes and substrates by lithium: the role of
                                     myo-inositol.
                                     Neuropsychopharmacology.
                                     1996;15:370-381.
                                     MEDLINE

                                     7.
                                     Stensby ME.
                                     Nutritional properties of fish oils.
                                     World Rev Nutr Diet.
                                     1969;11:46-105.
                                     MEDLINE

                                     8.
                                     Sperling RI, Benincaso AI, Knoell CT, Larkin JK, Austen KF, Robinson DR.
                                     Dietary omega-3 polyunsaturated fatty acids inhibit phosphoinositide formation and
                                     chemotaxis in neutrophils.
                                     J Clin Invest.
                                     1993;91:651-660.
                                     MEDLINE

                                     9.
                                     Tappia PS, Ladha S, Clark DC, Grimble RF.
                                     The influence of membrane fluidity, TNF receptor binding, cAMP production and
                                     GTPase activity on macrophage cytokine production in rats fed a variety of fat diets.
                                     Mol Cell Biochem.
                                     1997;166:135-143.
                                     MEDLINE

                                     10.
                                     American Psychiatric Association.
                                     Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.
                                     Washington, DC: American Psychiatric Association; 1994.

                                     11.
                                     Spitzer R.
                                     Structured Clinical Interview for DSM-IV.
                                     Washington, DC: American Psychiatric Association; 1994.

                                     12.
                                     Dunner DL, Fieve RR.
                                     Clinical factors in lithium carbonate prophylaxis failure.
                                     Arch Gen Psychiatry.
                                     1974;30:229-233.
                                     MEDLINE

                                     13.
                                     Young RC, Biggs JT, Ziegler VE, Meyer DA.
                                     A rating scale for mania: reliability, validity and sensitivity.
                                     Br J Psychiatry.
                                     1978;133:429-435.
                                     MEDLINE

                                     14.
                                     Williams JB.
                                     A structured interview guide for the Hamilton Depression Rating Scale.
                                     Arch Gen Psychiatry.
                                     1988;45:742-747.
                                     MEDLINE

                                     15.
                                     Guy N.
                                     ECDEU Assessment Manual for Psychopharmacology.
                                     Washington, DC: US Dept of Health, Education, and Welfare; 1976.

                                     16.
                                     Medini L, Colli S, Mosconi C, Tremoli E, Galli C.
                                     Diets rich in n-9, n-6 and n-3 fatty acids differentially affect the generation of inositol
                                     phosphates and of thromboxane by stimulated platelets, in the rabbit.
                                     Biochem Pharmacol.
                                     1990;39:129-133.
                                     MEDLINE

                                     17.
                                     Barton PG, Gunstone FD.
                                     Hydrocarbon chain packing and molecular motion in phospholipid bilayers formed from
                                     unsaturated lecithins.
                                     J Biol Chem.
                                     1975;250:4470-4476.
                                     MEDLINE

                                     18.
                                     Bourre JM, Bonneil M, Clement M, Dumont O, Durand G, Lafont H, Nalbone G, Piciotti
                                     M.
                                     Function of dietary polyunsaturated fatty acids in the nervous system.
                                     Prostaglandins Leukot Essent Fatty Acids.
                                     1993;48:5-15.
                                     MEDLINE
 
 
 
 
 
 
 

 
 

 
 
 
 
 

 

 
                                     © 1999 American Medical Association. All rights reserved.