Two double-blind placebo-controlled pilot studies of eicosapentaenoic acid in the treatment of schizophrenia
Introduction
The search for new treatments for schizophrenia has focused mainly on modulating neurotransmitters and their receptors, despite the relative lack of evidence of a primary abnormality in these systems. Novel treatment approaches to schizophrenia are likely to come only on the basis of new hypotheses relating to aetiology. The phospholipid hypothesis of schizophrenia, recently reviewed by Horrobin (1998), has proved to be of considerable heuristic value.
There is substantial evidence of phospholipid and polyunsaturated fatty acid (PUFA) metabolic abnormalities in schizophrenia (Peet et al., 1999). Studies of peripheral tissues, including erythrocytes and skin fibroblasts, have shown reduced levels of phospholipid subtypes (phosphatidylcholine and phosphatidylethanolamine) in schizophrenic patients (Hitzemann et al., 1985, Keshaven et al., 1993, Mahadik et al., 1994). Recent studies using 31P magnetic resonance spectroscopy have shown increased levels of phosphodiesters and decreased levels of phosphomonoesters in prefrontal and temporal brain of drug-naı̈ve schizophrenic patients (Fukuzako et al., 1999, Pettegrew et al., 1991, Stanley et al., 1994). Since phosphomonoesters are utilized in phospholipid synthesis, and phosphodiesters are phospholipid breakdown products, these data have been taken to imply increased phospholipid turnover in the brains of schizophrenic patients. Increased levels of the phospholipid breakdown product lysophosphatidylcholine have also been reported in platelets from schizophrenic patients (Pangerl et al., 1991).
There have been several investigations of the PUFA content of membrane phospholipids in schizophrenia. Although earlier studies produced inconsistent results, more recent investigations have produced a more consistent picture of depleted n-6 and n-3 PUFA in red blood cells (Glen et al., 1994, Peet et al., 1995, Vaddadi et al., 1989, Yao et al., 1994a) and brain (Horrobin et al., 1991, Yao, 1999) of schizophrenic patients.
An associated biochemical abnormality in schizophrenia is elevated activity of phospholipase A2 (PLA2) in plasma (Gattaz et al., 1987) serum (Gattaz et al., 1990, Noponen et al., 1993) and platelets (Gattaz et al., 1995) from drug-free schizophrenic patients, although there was a negative report from Albers et al. (1993). This enzyme releases fatty acids from phospholipids. Ross et al. (1997) pointed out that there are several different species of phospholipase A2, and they found that the calcium-independent phospholipase A2 was elevated in schizophrenic patients whereas calcium-stimulated PLA2 levels were normal. They subsequently showed that this enzyme activity is also increased in pre-temporal cortex (Ross, 1999).
Overall, the evidence suggests that schizophrenia is associated with abnormal phospholipid metabolism, including an increased rate of breakdown possibly due to excess PLA2 activity.
Fatty acids in the diet, both essential and non-essential, have significant effects upon neuronal membrane phospholipid composition (Abedin et al., 1999, Martinsdottir et al., 1998). Therefore, it would be expected that any dietary changes could mitigate or aggravate an underlying abnormality of phospholipid metabolism. There is evidence of a correlation between dietary fat consumption and schizophrenic symptomology.
Christensen and Christensen (1998) reported that the ratio of saturated fatty acids (SFA) to PUFA in the national diet correlated with outcome figures for schizophrenia published by the World Health Organization. Those countries that obtained relatively more of their dietary fat from land animals and birds (mostly SFA) and relatively less of their fat from vegetable, fish and seafood sources (mostly PUFA) had a worse outcome of schizophrenia. This accounted for 97% of the variance in outcome between countries. Similar relationships have been shown within groups of schizophrenic patients. Yao et al. (1994b) showed that the SFA-to-PUFA ratio in red blood cell membranes showed a significant positive relationship with severity of positive schizophrenic symptoms. Mellor et al. (1996) reported that a greater intake of n-3 PUFA and particularly eicosapentaenoic acid (EPA) in the normal daily diet was associated with less severe positive schizophrenic symptoms.
These associations between dietary fat and schizophrenic symptoms do not prove cause and effect, but taken together with the biochemical evidence, they provided us with a strong rationale for intervention studies using PUFA. Previous studies using n-6 PUFA in the treatment of schizophrenia gave mixed results (Vaddadi et al., 1986, Vaddadi et al., 1989, Wolkin et al., 1986). In one previous study using n-3 PUFA, linseed oil (50% alpha-linolenic acid) was given to five schizophrenic patients, with apparent benefit (Rudin, 1981). We chose to investigate n-3 PUFA because that was the strongest dietary correlation found by our group (Mellor et al., 1996).
In an initial supplementation study, 20 schizophrenic patients who were still symptomatic despite best efforts to treat them with antipsychotic drugs, were given 10 g per day of concentrated fish oil (MaxEPA) in addition to their existing antipsychotic medication, which remained unchanged during the 6 week study period. Patients showed a significant improvement in scores on the Positive and Negative Syndrome Scale (PANSS; Kay et al., 1987) and on the Abnormal Involuntary Movement Scale (Kane et al., 1992). There is no known pharmacokinetic interaction between PUFA and antipsychotic drugs, so this appeared to be a true pharmacological effect. Multiple regression analysis showed that improvement in schizophrenic symptoms was importantly related to the increased level of n-3 PUFA measured in red blood cell membranes. Ten grams of MaxEPA per day contains 1.7 g of EPA and 1.1 g of docosahexaenoic acid (DHA). Whilst these are both n-3 PUFA, they have differing metabolic functions. Thus, DHA is primarily a membrane structural component, whereas EPA takes part in eicosanoid synthesis. The two fatty acids can have differing physiological effects (Bates et al., 1993, De Caterine et al., 1994, Gilbert et al., 1999, Willumsen et al., 1996). Therefore, we considered it important to differentiate between the possible effects of EPA and DHA in schizophrenia. A pilot double-blind trial was therefore conducted to compare and EPA enriched oil with a DHA enriched oil and a corn oil placebo.
Section snippets
Subjects and methods
Subjects of the study were outpatients in Sheffield, UK with a DSM IV (American Psychiatric Association, 1994) diagnosis of schizophrenia. All were receiving a stable dosage of antipsychotic medication, and no change in dosage was anticipated during the study. They were required to be still symptomatic, with a PANSS score of at least 40, despite medical treatment that was considered appropriate by the responsible consultant psychiatrist. No patient suffered from significant physical illness or
Subjects and methods
Subjects for the study were 30 DSM IV (American Psychiatric Association, 1994) diagnosed schizophrenic patients on no medication who presented as new or relapsed cases to the psychiatric clinic in Baroda, India. The study, which was part of an ongoing research collaboration, was approved by the local Ethical Committee, which included senior academics and clinicians, and all patients gave informed consent. Prior to starting the double-blind study, a small open-label study had shown marked
Discussion
It is in the nature of pilot studies to be indicative rather than definitive. They are intended to provide pointers to the design of more definitive studies. Indeed, it is unlikely that a placebo controlled trial of, say, chlorpromazine or olanzapine would produce statistically significant results in this patient population and with such small numbers. We were therefore surprised to find the level of statistical significance that emerged from these two studies.
The first pilot study was designed
Acknowledgments
We are grateful to Dr David Horrobin and Laxdale Limited for supplying the materials and some financial support for this work.
References (63)
- et al.
Are fish oils beneficial in the prevention and treatment of coronary artery disease?
Am. J. Clin. Nutr.
(1997) - et al.
Schizophrenia and rheumatoid arthritis: a review
Schizophr. Res.
(1992) - et al.
Increased plasma phospholipase-A2 activity in schizophrenic patients: reduction after neuroleptic therapy
Biol. Psychiatry
(1987) - et al.
Increased platelet phospholipase A2 activity in schizophrenia
Schizophr. Res.
(1995) - et al.
A red cell membrane abnormality in a subgroup of schizophrenic patients: evidence for two diseases
Schizophr. Res.
(1994) - et al.
Fatty acid levels in the brains of schizophrenics and normal controls
Biol. Psychiatry
(1991) The membrane phospholipid hypothesis as a biochemical basis for the neurodevelopmental hypothesis of schizophrenia
Schizophr. Res.
(1998)- et al.
Plasma membrane phospholipid and cholesterol distribution of skin fibroblasts from drug-naı̈ve patients at the onset of psychosis
Schizophr. Res.
(1994) - et al.
Changes in dietary fatty acids alter phospholipid fatty acid composition in selected regions of rat brain
Prog. Neuropsychopharmacol. Biol. Psychiatry
(1998) - et al.
Identification of skin as a major site of prostaglandin D2 release following oral administration of niacin in humans
J. Invest. Dermatol.
(1992)