Omega-3 Fatty Acids: Theory, Clinical Trials and Safety Issues
Omega-3 Fatty Acids: Theory, Clinical Trials and Safety Issues
Since the 1930s, research evidence has indicated that certain essential fatty acids (EFAs) are required for normal human fetal and neonatal development (Uauy et al., 1996). An inadequate supply of these fatty acids during critical developmental periods can result in pathological changes in immune function; degenerative changes in the lungs, liver and kidneys; and abnormalities in central nervous system maturation.
Other researchers have hypothesized that chronic deficiencies in dietary EFAs can result in an increased incidence of several diseases, including multiple sclerosis, arthritis, enteritis, immune system dysfunction, heart disease, cancer, diabetes, schizophrenia and bipolar disorder (Peet et al., 1999; Rudin, 1982, 1981; Stoll, 2001). This article briefly discusses EFAs, and then reviews the uses of omega-3 fatty acids for treating psychiatric disorders, with attention to efficacy and safety issues.
Linoleic acid (LA; 18:2n-6) and α-linolenic acid (ALA; 18:3n-3) are true EFAs. They must be consumed in the diet because humans lack the ability to synthesize them. Food sources of LA include seeds, nuts, grains and legumes; ALA is abundant in cold-water fish (e.g., mackerel, herring, tuna); green leaves of plants; and some seeds, nuts and legumes (flax, canola, walnuts and soy).
Linoleic acid and ALA are metabolically transformed into long-chain polyunsaturated fatty acids (LC-PUFAs) in the liver with the aid of many cofactors, including insulin, zinc and several vitamins. Long-chain fatty acids include the omega-6 fatty acids of dihomo-y-linolenic acid (DGLA; 20:3n-6) and arachidonic acid (AA; 20:4n-6) and the omega-3 fatty acids of eicosapentaenoic acid (EPA; 20:5n-3) and docosahexaenoic acid (DHA; 22:6n-3).
However, various dietary, lifestyle and disease factors (e.g., oxidative damage, viral infection and hormonal changes) can interfere with the synthesis process. Thus, many individuals may be deficient in long-chain fatty acids, despite an availability of the EFA precursors in their diet.
After conversion from LA and ALA in the liver, the LC-PUFAs enter the blood and are incorporated into cell membranes. The fatty acids of cell membrane phospholipids are essential for normal membrane structures, for the functioning of membrane-bound and membrane-associated proteins, and for normal cell-signaling responses.
Rudin (1982, 1981) and others have suggested that present-day refining and food selection patterns have led to widespread deficiencies of omega-3 fatty acids in some industrialized countries, with a consequent increase in the incidence and prevalence of many medical and psychiatric disorders.
According to the phospholipid deficiency hypothesis, there has been an increase in the consumption of plant-derived products high in omega-6 fatty acids concurrent with dietary deficiencies in sources of the omega-3 fatty acids.
The resulting increased ratio of omega-6 to omega-3 fatty acids available for metabolism has important consequences for the balance of opposing processes in normal human physiology, as fatty acids are precursors of prostaglandins, the body's principal regulatory molecules. The omega-6 fatty acids are crucial to synthesis of many cytokines that mediate inflammation, including several interleukins, tumor necrosis factor α (TNF-α) and interferon-y (Maes et al., 1997a, 1997b).
In contrast, diets high in the omega-3 fatty acids are correlated with reduced overall production of these inflammatory cytokines (Caughey et al., 1996). Some researchers suggest that the shift in the omega-6/omega-3 ratio has led to increases in medical and psychiatric disorders mediated by chronic inflammatory changes.
LC-PUFAs and the Brain
Phospholipids comprise approximately 25% of the dry weight of the human brain. Together, arachidonic acid and docosahexaenoic acid account for roughly half of total brain phospholipids. Brain phospholipids are essential for fluidity of nerve cell membranes and provide the physicochemical environment in which cell-membrane-associated proteins are embedded, influencing their tertiary structure and their capacity to function as neurotransmitter receptors.
The fatty acid components of neuron membrane phospholipids also are central to the integrity of cell-signaling systems involving protein kinases and other second messenger systems believed to play an important role in the pathogenesis of many psychiatric disorders.
Additionally, the synthesis and breakdown of phospholipids in the brain is integral to normal growth of axons and dendrites, as well as the formation of new synapses and the pruning of old ones.
Phospholipids in Schizophrenia
Horrobin (1998, 1996) has proposed a "membrane phospholipid" model of schizophrenia. He stated that abnormal metabolism of phospholipids resulting from genetic and environmental factors manifests as a range of symptoms that are classified as schizophrenia.
Several case reports (Puri et al., 2000) and double-blind studies (Laugharne et al., 1996; Mellor et al., 1995; Peet et al., 2001; Peet et al., 1996) have consistently demonstrated sustained improvement of both positive and negative symptoms in patients with chronic schizophrenia consuming certain fatty acids who were not being treated concurrently with conventional antipsychotic medications.
However, a double-blind, placebo-controlled study revealed no differences in response between eicosapentaenoic acid (E-EPA) (3 g/day) and placebo in a group of 87 patients with schizophrenia or schizoaffective disorder who were concurrently receiving antipsychotic medications (Fenton et al., 2001). In contrast to earlier studies, patients in this study were older, had a longer duration of illness and had residual psychotic symptoms.
The membrane phospholipid hypothesis, according to Horrobin, may provide a unifying conceptual framework for not only understanding schizophrenia but also bipolar disorder, dyslexia, schizotypal personality disorder, other schizophrenia-like syndromes and possibly other psychiatric disorders (Horrobin, 1998; Horrobin and Bennett, 1999).
Additionally, his hypothesis may be compatible with the paradigm ascribing the etiology of psychiatric disorders to dysfunction at the level of neurotransmitters and their receptors. According to Horrobin's theory, an inadequate dietary supply of EFAs or metabolic factors interfering with the normal conversion of parent EFAs into DHA or EPA would ultimately restrict the supply of omega-3 fatty acids to the brain for incorporation into nerve cell membranes. This deficiency results in abnormal phospholipid composition and suboptimal functioning of a range of membrane-based neurotransmitter systems.
The membrane phospholipid hypothesis suggests that a spectrum of psychiatric disorders is associated with abnormalities at the level of the neuronal membranes, and that the nature and severity of symptoms are related to the magnitude and type of metabolic errors leading to abnormal phospholipid metabolism. In schizophrenia, for example, there is evidence for an increased rate of docosahexaenoic and arachidonic acid loss from membranes because of enhanced phospholipase A2 activity (Horrobin et al., 1995), whereas in dyslexia, dyspraxia and attention-deficit/hyperactivity disorder (ADHD), there may be a problem in the conversion of linoleic acid and α-linolenic acid into their long-chain derivatives (Richardson, 2001).
Horrobin cites the following findings in support of the membrane phospholipid hypothesis:
- Increased blood levels of an enzyme (phospholipase A2) that is known to remove polyunsaturated fatty acids from phospholipids in nerve cell membranes of patients with schizophrenia (Horrobin, 1996);
- Reduced levels of AA and DHA in red cell membranes of patients with schizophrenia (Horrobin et al., 1994);
- Magnetic resonance imaging data showing relatively increased rates of phospholipid breakdown in the brains of never-medicated patients with schizophrenia;
- Reduced electroretinogram (ERG) response in patients with schizophrenia (an indicator of reduced retinal DHA);
- Studies showing that clozapine (Clozaril) increases red blood cell phospholipid AA and DHA levels;
- The gene for lipoprotein lipase, the enzyme that regulates supply of EFAs to the brain, is on chromosome 8, where evidence points to a gene predisposing to schizophrenia. This enzyme's activity typically diminishes during puberty, a time often associated with the full expression of schizophrenia;
- Children with ADHD have reduced blood concentrations of EFAs.
Omega-3s and Depression
Food preferences influencing EFA consumption may directly relate to observed cross-national differences in depression rates. Countries where fish is a mainstay of the average diet are characterized by significantly lower rates of major depression and postpartum depression (Hibbeln, 2002; 1998). For example, in Japan, where fish consumption is very high, only 0.12% of the population experienced depressed mood in a given year. In contrast, New Zealanders, who consume relatively little fish, reported a 6% annual rate of depression. Recently, Hibbeln (2002) found that higher concentrations of DHA in mothers' milk and greater seafood consumption both predicted lower prevalence rates of postpartum depression. The data suggest that a nearly 50-fold difference in prevalence rates of major postpartum depressive symptoms across countries is associated with omega-3 fatty acid nutritional status.
Well-controlled studies looking at long-chain fatty acids and depression are needed. To date, only one small double-blind study of omega-3 fatty acid and depression has been completed (Nemets et al., 2002). This four-week study compared two groups of adults with relatively uncomplicated unipolar depressive disorder who received a placebo or ethyl ester of E-EPA (2 g/day) while continuing their antidepressant medications. None of the study participants met criteria for refractory depression, and all but one had been successfully treated with conventional antidepressants previously. The 24-item Hamilton Rating Scale for Depression (HAM-D) was administered at baseline and weekly thereafter. Average HAM-D baseline scores were ≥18. By the end of the study, E-EPA-treated patients showed a mean reduction of 12.4 points, compared to a mean reduction of 1.6 points in patients receiving a placebo. Overall, six of the 10 patients assigned to the E-EPA group showed a 50% reduction in HAM-D scores compared to only one of 10 patients in the placebo group. The researchers found that E-EPA had a significant effect on several core depressive symptoms, including feelings of guilt, worthlessness and insomnia.
There were no reports of significant side effects in either group, and only one patient (from the placebo group) dropped out because of worsening depression. The authors noted that the results do not clarify whether E-EPA has an independent antidepressant effect or augments antidepressants via second messenger systems in the manner of lithium. Confirmation of the significance of an antidepressant effect and clarification of the mechanism of action of E-EPA will require a long-term prospective design.
A recent case report (Puri et al., 2002) claimed rapid dramatic improvement in a 21-year-old student with a seven-year history of unremitting depressive symptoms. The patient was actively suicidal and had a Montgomery-Asberg Depression Rating Scale (MADRS) score of 32. He had been refractory to multiple antidepressant trials, including lithium augmentation. Administration of E-EPA (4 g/day) led to rapid improvement and sustained improvement over nine months. The patient's symptoms eventually disappeared. No adverse effects were reported.
Accumulating laboratory evidence suggests a plausible link between the dietary ratio of omega-6 to omega-3 fatty acids and the incidence of depression. Adams et al. (1996) demonstrated a positive correlation between the severity of depression and the ratio of AA (an omega-6) to EPA (an omega-3) in erythrocyte phospholipids.
Additionally, researchers have found evidence that major depression is accompanied by activation of the inflammatory response system as indicated by an increased production of pro-inflammatory cytokines (Kubera et al., 2001; Maes et al., 1996; Mikova et al., 2001). Further, direct administration of the same cytokines into the brain causes dysregulation in serotonin metabolism that is consistent with changes observed in depressed individuals.
Tricyclic antidepressants and selective serotonin reuptake inhibitors are known to suppress release of many pro-inflammatory cytokines by immune cells in the blood, consistent with the view that these drugs may perform a similar therapeutic role in the brain, resulting in improved mood (Kubera et al., 2001).
Considerable indirect evidence and one preliminary double-blind, placebo-controlled trial (Stoll et al., 1999) suggest that omega-3 fatty acids improve both depressive and manic symptoms in patients with bipolar disorder (BD). Omega-3 fatty acids may inhibit the activity of CNS phospholipases, thereby limiting the production of specific prostaglandins (e.g., PGE1) that are known to be associated with mania or depression. It has been postulated that lithium, dopamine antagonists and serotonin-blocking agents are effective in the treatment of mania through a similar mechanism of correcting such "overactivity" in cell membrane signal transduction processes.
In a four-month study, Stoll and colleagues (1999) compared a combination of EPA and DHA (9.6 g/day) versus placebo (olive oil) in 30 patients with BD. While most of the patients received omega-3 fatty acids or placebo along with their usual mood-stabilizing medications (e.g., lithium, divalproex [Depakote], carbamazepine [Tegretol]), eight patients entered the study while receiving no other mood-stabilizing drugs. Significantly, a post-hoc analysis determined that four out of eight patients who took only omega-3 fatty acids remained in remission significantly longer than three patients who received only placebo. Further, among the remaining 22 patients taking mood-stabilizing drugs, those treated with omega-3 fatty acids performed significantly better on all outcome measures than patients in the placebo arm.
Stoll and colleagues received a grant from the National Center for Complementary and Alternative Medicine (2000) to repeat the study on a much larger scale and a more rigorous design in 120 patients (Perry, 2001). The study examined the efficacy of omega-3 fatty acids as a maintenance therapy in patients with bipolar I disorder (BDI) over 12 months. All patients had to be euthymic or have only subsyndromal mood symptoms. Patients were randomized to receive omega-3 fatty acids or placebo in combination with their ongoing mood-stabilizing medication (e.g., lithium, divalproex). The study design also included one group of stable patients who will remain off mood stabilizers while taking omega-3s. Results of this study will help to clarify the role of omega-3 fatty acids in maintenance treatment of BDI and may also provide useful insights about safe and appropriate ways to combine mood-stabilizing agents with omega-3 fatty acids for different patient populations with BD.
A multicenter study (National Institute of Mental Health, 1999) is underway on the effects of omega-3 fatty acids in the treatment of major depression and BD. The patients recruited for this study were randomly assigned to receive E-EPA (6 g/day) or placebo during a 16-week period, followed by an optional eighth-month open trial on the omega-3 fatty acid. All patients continued on their current antidepressant or mood-stabilizing medications. The hypothesis of the study is that EPA acting on some of the same signal transduction mechanisms as mood stabilizers will be beneficial in breakthrough depression, mania and cycling of BD.
Cognitive decline. Emerging evidence suggests that regular intake of omega-3 fatty acids is inversely related to cognitive impairment or rate of cognitive decline in nondemented older males. Kalmijn et al. (1997) compared cognitive impairment scores over a three-year period in two groups of men (69 years to 89 years of age) with different dietary preferences. High intake of foods rich in linoleic acid (omega-6) was associated with higher rates of cognitive decline (Kalmijn, 2000). In contrast, high fish consumption (containing large amounts of omega-3 fatty acids) was inversely correlated with cognitive impairment. The authors inferred that high dietary intake of omega-6 polyunsaturated fatty acids likely contributes to increased oxidative stress in the brain, indirectly promoting atherosclerosis and thrombosis that eventually manifest as declines in cognitive functioning. Conversely, high intake of omega-3-rich foods, especially fish, may reduce oxidative stress and associated atherosclerotic changes in the brain, mitigating factors known to be associated with cognitive decline.
Violent and impulsive behavior. Observational studies (Hibbeln et al., 2000, 1998a, 1998b) suggest that low plasma DHA levels and, therefore, presumably low CNS levels may increase the predisposition of some individuals to violent or impulsive behavior. This effect appears to be larger in certain groups, especially males who develop alcohol dependence before 20 years of age. One proposed explanation is that genetic abnormalities in EFA metabolism result in a higher turnover rate of serotonin in the CNS, associated with higher levels of cerebrospinal fluid hydroxyindolacetic acid (CSF 5-HIAA).
A placebo-controlled, double-blind study (Hamazaki et al., 1996) compared DHA (1.5 g/day to 1.8 g/day) with placebo (soybean oil plus 3% fish oil) in matched cohorts of Japanese students. A significant rate of "aggression against others" was reported in the placebo group at times of peak academic stress whereas students taking DHA did not exhibit increased aggressive behavior.
A recent report of a double-blind, placebo-controlled, randomized trial in 231 adult prisoners showed that prisoners given a vitamin/mineral supplement and an EFA supplement committed an average of 26.3% fewer disciplinary offenses than those receiving placebos (Gesch et al., 2002).
Dyslexia, ADHD and other learning disabilities. Anecdotal reports of improvements in dyslexia with DHA supplementation have long suggested that abnormal phospholipid metabolism is associated with this disorder. Using MRIs, Richardson et al. (1997) found results suggestive of a problem in the synthesis of membrane phospholipids in dyslexia. Double-blind trials (Richardson, 2002; Richardson et al., 2000, 1999) are providing promising evidence that dietary supplementation with fatty acid supplements may be of benefit in managing dyslexia and other disorders.
In a descriptive study, a greater number of behavioral, sleep, learning and health problems were reported in boys with lower plasma phospholipid omega-3 fatty acid levels compared with those with higher levels (Stevens et al., 1996).
Burgess et al. (2000) reported that omega-3 supplementation trials are being conducted in subjects with ADHD who show symptoms of EFA deficiency and lower proportions of EFAs than control subjects. Well-controlled prospective studies -- including those with age-matched girls -- are needed.
A review article suggested that no significant safety issues are associated with consumption of unsaturated fatty acids, including EFAs, as long as these substances do not account for more than 10% of total caloric intake (Eritsland, 2000).
Above this level, there is evidence linking high intake of certain EFAs with reduced glycemic control in type 2 diabetics (Friday et al., 1989; Glauber et al., 1988) and also with a tendency for increased bleeding (Goodnight et al., 1982). There is also evidence of a slight overall increase in liver enzyme activity with large amounts of omega-3 fatty acids, which could affect metabolic clearance of certain medications.
Anecdotal reports of a possible correlation between high doses of omega-3 fatty acids and hypomania or mania have been reported (Stoll et al., 2000; Su et al., 2000). Stoll et al. (2000) reported fewer than 10 cases of apparent omega-3-induced hypomania or mania in a series of more than 300 patients treated with various open-label preparations of flaxseed oil or fish oil. Almost all cases of apparent hypomania induction were associated with flaxseed oil. This effect was first noticed more than 20 years ago (Rudin, 1981) and was also associated with high doses of flaxseed oil, but not fish oil products. This effect is still under investigation.
There has been a case report that prolonged use of a fish oil supplement resulted in hypervitaminosis A (Grubb, 1990). Other reports have suggested an increased incidence of hypertension and stroke (Kenny, 1990) in individuals who consume large amounts of omega-3-containing supplements. Controlled studies have not confirmed the presence or magnitude of these risks.
Some patients complain of gastrointestinal (GI) distress when taking flaxseed oil or fish oil products. Eight of 13 (62%) patients taking omega-3 fatty acids in the Stoll study (Stoll et al., 1999) reported mild GI side effects compared to eight of 15 (53%) placebo-treated subjects. The difference between the groups was not statistically significant, and no other side effects were reported frequently.
Serum and RBC Lipid Analysis
In view of the growing evidence for fatty acid abnormalities in several psychiatric disorders, it is likely that more clinicians will consider treating patients with omega-3 fatty acid preparations. Lipid analysis will have an increasingly important role in differential diagnosis and treatment decisions regarding EFA supplementation for a range of medical and psychiatric disorders. Plasma fatty acid analysis reflects dietary intake, while red cell fatty acid analysis can reveal disturbances in metabolism. In the future, errors of metabolism may be approached through genetic engineering and corrective metabolic nutritional therapies. Conversely, disorders resulting from dietary fatty acid imbalances may be treated with targeted ratios of specific fatty acids. Research into making accurate laboratory determinations of the specific biochemical type and magnitude of dysfunction in the metabolic or dietary pathways of EFAs is underway.