Evening Primrose Oil in the Treatment of Atopic and Irritant Contact Dermatitis

January 1999; Volume 2: 8-10

By David Schiedermayer, MD

When I examine patients’ hands, I learn about their interaction with the world. Farmers have hypertrophied interosseous muscles and pole-sized wrists. Anxious patients have chewed nails, or worse yet, gnawed-on cuticles. Mechanics have deeply oil-stained fingers. Unlucky or careless laborers have missing digits.

But the main cause of occupational skin disease is irritant contact dermatitis (ICD), which I’ve seen in shoemakers, homemakers, and doctors. Atopic dermatitis (AD) has some pathophysiological similarities to ICD, but often presents without any known external chemical irritation. AD is a genetically determined condition readily worsened by seasonal flare, infections, and food allergies.1

Pathophysiology

In dermatitis, irrespective of cause, the outer layer of the stratum corneum is structurally and functionally altered; its usual impermeable nature is impaired. Substances that irritate it are able to cross into the dermis unimpeded. The theory behind treatment is to restore the impermeable nature of the barrier through restoring some of the fatty acids present in the skin. Gamma linolenic acid (GLA) is one such fatty acid. It is essential for production of prostaglandins, prostacyclins, thromboxanes, and leukotrienes. A deficiency can lead to hyperproliferative changes and scaliness.

Derivation

Many different strains of evening primrose (Oenothera biennis) are found in nature and under cultivation. The seed oil, which is a typical vegetable oil, varies considerably in the concentration of the essential/n6 fatty acid GLA, the active ingredient in evening primrose oil (EPO).

Epogam, one commercially available source of EPO, has seed oil derived from four flower strains that have been selected and bred to yield oil of constant composition.2

Metabolism

In healthy volunteers with no dermatological symptoms who received six capsules of Epogam, there was significant individual variation in absorption and elimination of GLA from the serum. However, on average, the maximal concentration was about 4.5 times higher than the concentration in volunteers not taking Epogam. Other fatty acid metabolites showed no significant changes in serum concentrations. The uptake of GLA from the GI tract was much slower in the morning than in the evening. About 12 hours after administration of six capsules, serum levels returned to baseline.3

Proposed Mechanism of Action

Some authors hypothesize that patients with ICD or AD have defective functioning of the enzyme that desaturates linoleic acid, delta-6-desaturase. Without this enzyme, prostacyclins, thromboxanes, and leukotrienes cannot be produced from this pathway. Figure 1 shows arachidonic acid as the substrate for the formation of prostaglandin E2, and DGLA as the substrate for prostaglandin E1 formation.4

Clinical Studies

A meta-analysis of placebo-controlled studies of the efficacy of Epogam, published by the manufacturer’s research institute, showed that in four controlled parallel designed trials, both patient and doctor scores showed a highly significant improvement over baseline. In the crossover trials, however, doctors’ ratings failed to achieve significance. When the results were related to changes in plasma levels of DGLA and arachidonic acid levels, there was a positive correlation between improvement in clinical score and rise in serum fatty acid levels. The researchers hypothesized that a rise in arachidonic acid as well as DGLA may reflect conservation of arachidonic acid in the lipid fraction and phospholipids of the skin.2 Unfortunately, insufficient detail of prior unpublished studies used in the analysis is provided, and this "meta-analysis" appears incomplete.

When EPO (Epogam) was evaluated in a double-blind, placebo-controlled trial of 39 patients with chronic (greater than one year) stable hand dermatitis, the findings were not as encouraging. Active therapy with 600 mg/d of GLA (twelve 500 mg capsules of Epogam containing 50 mg of GLA) was administered orally to the study group for 16 weeks with another eight weeks of continued observation. Patients were assessed clinically, using a visual analogue scale at four-week intervals. Plasma and red blood cell lipograms, as well as skin biopsies, were taken before therapy, after the 16-week supplementation, and at week 24.

Improvements in clinical parameters were present in both the Epogam and placebo groups, but no statistical difference could be confirmed. Regular follow-up visits made the patients more aware of their condition and clinical guidance could have enhanced the use of emollients and avoidance of allergens. CBC and epidermal biochemistry were normal at baseline. No change in the lipid composition of plasma red cells or epidermis could be detected during the trial. The authors concluded that orally administered GLA for chronic hand dermatitis is not superior to placebo.5

The best study was a double-blind, placebo-controlled, parallel group study done to avoid the methodological and analytic problems of previous studies.6 Patients were randomized to receive EPO, EPO and fish oil, or placebo for 16 weeks. Of 123 subjects recruited, 102 completed the treatment period. No improvement with any active treatment was demonstrated. This study combined n6 (EPO) and n3 essential fatty acids (fish oil), analyzed the two groups together, and still showed no positive effect.

Adverse Effects

Adverse effects included nausea, indigestion, headaches, and softened stools but no serious adverse effect has been reported.7 Weight gain is possible if large doses are taken. One observer warns of a potential risk of inflammation, thrombosis, and immunosuppression from slow accumulation of tissue arachidonate after use of GLA for more than one year.8

Administration of EPO is considered as safe as the administration of corn oil.9 Effects on serum cholesterol, HDL, and LDL fractions are not evaluated in most studies, but should be evaluated to define potentially adverse effects of EPO on the lipid profile. EPO studies in rats, mice, and dogs do not support the hypothesis that linoleic acid initiates or promotes tumor formation.

Current Clinical Therapy

Current therapeutic options for both ICD and AD are limited; the only generally accepted therapy is avoiding irritants for as long as the skin is not completely healed. The benefit of topical corticosteroids is controversial. Steroids are believed to act, at least in part, by reducing the mobilization of arachidonic acid from phospholipid stores and decreasing the availability of free arachidonic acid for conversion to pro-inflammatory metabolites.2 However, clinical trials are notoriously difficult to perform, because of the inherent variability of the clinical state, the subjective nature of the assessment, and a large placebo response.2

Other treatments include emollients, antihistamines, ultraviolet B, psoralen plus ultraviolet A, Chinese herbs, and cyclosporin.

Drug Interactions

EPO should not be used with anticonvulsants because it may lower the seizure threshold.10

Formulations and Dosing

Epogam contains 320 mg of linoleic acid and 40 mg of GLA as well as trace amounts of palmitic, oleic, and stearic acid, and added vitamin E as an antioxidant. Efamol is 72% linoleic acid, 12% oleic acid, and has 45 mg of GLA per capsule. In most studies, subjects took 10-12 capsules per day.

Conclusion

The claim that Epogam produces "a substantial and highly significant clinical improvement" in AD or ICD appears exaggerated. Epogam is no more effective than placebo in a good clinical trial.

Many of the trials to date that suggest an effect have been crossover studies in small number of patients. The long duration of treatment, the variable natural courses of the disease in question, the placebo effect, and publication bias (nearly all trials so far have been sponsored by the same company) suggest the need to take any purported effect with some caution.11 Further rigorous trials of both GLA and EPO are warranted.12

Recommendation

At present, EPO is best regarded only as an optional addition to existing treatment for ICD or AD, and as a dietary fatty acid supplement rather than a medication. It is no more effective than placebo and is also expensive.8 Many unanswered questions remain about the optimal dose and duration of treatment. It does appear to be a relatively harmless substance.

References

1. Meneghini CL, Bonifazi E. Correlation between clinical and immunological findings in atopic dermatitis. Acta Derm Venereol (Stockh) 1985;114:140-142.

2. Morse PF, et al. Meta-analysis of placebo-controlled studies of the efficacy of Epogam in the treatment of atopic eczema: Relationship between plasma essential fatty acid changes and clinical response. Br J Dermatol 1989;121:75-90.

3. Martens-Lobenhoffer JM, Meyer FP. Pharmacokinetic data of gamma-linolenic acid in healthy volunteers after the administration of evening primrose oil (Epogam). Inter J Clin Pharmacol Ther 1998;36: 363-366.

4. Wright S. Atopic dermatitis and essential fatty acids: A biochemical basis for atopy? Acta Derm Venereol (Stockh) 1985;114:143-145.

5. Whitaker DK, et al. Evening primrose oil (Epogam) in the treatment of chronic hand dermatitis: Disappointing therapeutic results. Dermatology 1996;193:115-120.

6. Berth-Jones J, Graham-Brown RAC. Placebo-controlled trial of essential fatty acid supplementation in atopic dermatitis. Lancet 1993;341:1557-1560.

7. Anonymous. Gamolenic acid in atopic eczema: Epogam. Drug Ther Bull 1990;28:69-70.

8. Phinney S. Potential risk of prolonged gamma-linolenic acid use. Ann Intern Med 1994;120:692.

9. Everett DJ, et al. Chronic toxicity studies of Efamol evening primrose oil in rats and dogs. Med Sci Res 1988;16:865-866.

10. Miller LG. Herbal medicinals: Selected clinical considerations focusing on known or potential drug-herb interactions. Arch Intern Med 1998;158:2200-2211.

11 Goonaratna C. Good morning, evening primrose oil. Ceylon Med J 1995;40:132-133.

12. Kleijnen J. Evening primrose oil: Currently used in many conditions with little justification. BMJ 1994;309:824-825.