Complementary and Alternative Medicine for Atopic Dermatitis

An Evidence-Based Review

Brittany L. Vieira; Neil R. Lim; Mary E. Lohman; Peter A. Lio;


Am J Clin Dermatol. 2016;17(6):557-581. 

In This Article

Results and Implications for Future Research

The initial search using medical subject headings (MeSH) and keywords as outlined above yielded a total of 883 articles from the four databases (379 PubMed, 231 Cochrane, 50 EMBASE, 223 GREAT database). Of these articles, 130 RCTs met the inclusion criteria for this review during preliminary analysis of titles and abstracts. Of the 128 full-text articles assessed for eligibility, 70 met the inclusion criteria and were included in the final analysis.

Convergence of multiple large-scale RCTs remains the gold standard for making sound evidence-based recommendations.[7] Accruing such evidence, however, requires significant resources—resources that are notably absent from the realm of alternative medicine. Alternatives lack the funding and organization of large pharmaceutical companies studying new drugs, and also lack the gravitas to warrant government-sponsored research on a large scale. Thus, many of these treatments remain on the fringes, often supported by small, sometimes poorly conceived studies that make analysis difficult. Despite these limitations, we present multiple studies that met our inclusion criteria and, when taken together, argue for consideration of some promising alternatives and for more research across the board.

Acupuncture and Acupressure

Itch is the major symptom of AD, and is central to its pathogenesis. Acupuncture has been shown to exhibit a significant effect on itch in other contexts, such as in an RCT for uremic pruritus, which demonstrated that acupuncture was significantly better at reducing pruritus in refractory cases than a sham acupuncture control (p<0.001).[8]

In two studies of 30 and 20 patients, respectively, Pfab et al. demonstrated that acupuncture significantly reduced allergen-induced itch intensity when compared with placebo and no treatment,[9] and that acupuncture was more effective than the antihistamine cetirizine when used abortively rather than preventatively.[10] While somewhat contrived from a clinical perspective, these studies suggest the possibility of direct modulation of pruritus and inflammation by acupuncture in atopic individuals, a population not previously examined ( Table 1 ).

Lee et al. carried this forward with an investigator-blinded trial of 15 patients with moderate-to-severe AD and found that acupressure (using a small titanium bead to massage an acupoint on the arm 3 times weekly) significantly improved pruritus (p = 0.04), lichenification (p = 0.03) and Investigator Global Assessment (IGA) (p = 0.03) scores compared with control group, though overall Eczema Area and Severity Index (EASI) scores remained unchanged.[11] This study was limited by small patient numbers, lack of placebo, and unmonitored application of acupressure. No adverse effects were reported in any of the studies.

The results of these studies are encouraging, but the numbers are small—thus, larger studies are needed to clarify a repeatable approach, as well as to further elucidate the possible mechanisms of such an effect.

Hypnotherapy, Relaxation, and Massage

Several lines of evidence suggest that stress plays a role in AD.[12] It then follows that stress-reduction techniques may be helpful in managing the disease.

In a study of 44 patients by Sokel et al., both hypnotherapy and biofeedback (a relaxation/awareness technique based on a physiological feedback device that measures galvanic skin resistance) groups exhibited significant reduction in surface damage and lichenification compared with the control group.[13] Neither treatment appeared to be superior to the other, however, suggesting that both techniques may benefit patients similarly through relaxation ( Table 2 ). Results of another study of 33 patients demonstrated that patients who were trained to decrease skin conductance (generally associated with more relaxed states) using biofeedback showed significant improvement in disease severity, while those who were trained to increase skin conductance (generally associated with increased anxiety) showed worsened severity.[14] Of note, this study included only patients with dyshidrotic eczema.

A study of 28 patients over 14 weeks found that cognitive behavioral stress (CBS) management therapy showed no significant improvement in disease severity over the non-treatment group.[15] Patients with other atopic comorbidities were excluded from the study. Of note, CBS did reduce endocrine and psychological stress responses in AD patients, though no skin manifestations were noted during the study period.

A study comparing an educational program versus autogenic training versus cognitive behavioral treatment versus a combination of education and behavioral treatment determined that at 1 year follow-up, all psychological treatments decreased severity of the AD and the amount of topical steroids used more than the standard medical care control, or education alone.[16]

Schachner et al. compared massage therapy to standard topical treatment and found that massage significantly improved redness, scaling, lichenification, excoriation, and pruritus compared with control in 20 children aged 2–8 years.[17] In a trial of 25 patients that examined standard of care with or without progressive muscle relaxation therapy, it was found that progressive muscle relaxation significantly reduced pruritus and loss of sleep when compared with baseline. However, there was no significant difference in EASI scores between treatment and control groups.[18]

In a study by Anderson et al., 16 children with refractory AD were divided into groups that either received massage therapy in conjunction with essential oils (aromatherapy), or massage therapy alone.[19] The authors found no significant difference between the two groups, but both did show improvement following treatment from baseline. However, there was no untreated control. Of concern was that some patients may have become sensitized to the essential oils and actually worsened during the study, possibly negating any positive effects of the oils.

From the limited evidence, it appears that massage, hypnosis, and perhaps other stress-relieving techniques could play at least an adjunctive role in managing AD given their minimal side effects. Sleep disruption continues to be a major problem for many patients, and the notion that something safe and inexpensive like progressive muscle relaxation could be of value is compelling and worthy of more research.[20] Aromatherapy oils do not appear to add any therapeutic value to massage, and may even pose a problem by acting as sensitizers.


Increased transepidermal water loss (TEWL) is one of the key characteristics of AD. Thus, it is not surprising that certain AD interventions involve water and bathing. Balneotherapy (bath treatments) are ancient practices, but emerged as modern treatment modalities in the 1800s in Europe.[21] Although it has been suggested that 'hard water' (water rich in minerals such as magnesium, calcium, and bromide) can provoke AD,[22] in a recent RCT of 336 children, Thomas et al. demonstrated that ion-exchange water softeners provide no additional benefit to traditional AD therapy ( Table 3 ).[23] Togawa also evaluated the efficacy of ion-exchange water softeners in a sample of 12 children in Japan with mild to moderate eczema (EASI ≤20).[24] Their results suggested that ultra-pure soft water does not significantly improve the EASI score; however, patients reported a significant reduction in pruritus and an increase in satisfaction.

In an RCT of 180 AD patients, Heinlin et al. found that synchronous balneophototherapy—a simulation of Dead Sea treatment conditions in terms of mineral bath content and ultraviolet light exposure—significantly reduced AD severity when compared with narrow-band UVB monotherapy (p<0.001), with only mild and comparable adverse effects noted in both groups.[25] Finally, an open RCT of 104 children found that both balneotherapy and topical corticosteroids significantly improved disease severity. Topical corticosteroids proved to be more effective than balneotherapy regarding SCORing Atopic Dermatitis (SCORAD) (p<0.03); however, IGA, Patient's Self Global Assessment (PSGA), Children's Dermatology Life Quality Index (CDLQI), and Family Dermatitis Impact Questionnaire (FDIQ) score improvements were similar. Interestingly, at 4 months, the incidence and duration of relapse were both significantly reduced in patients treated with balneotherapy compared with those treated with topical corticosteroids (p<0.0001).[26]

From the evidence, it is reasonable to conclude that water softeners are likely not a useful adjunct for most patients with AD. Conversely, a large, high-quality trial by Heinlin et al. suggests that the supplementation of hypertonic, magnesium-rich salt baths in addition to phototherapy does confer an additional therapeutic benefit.[25] The corroboration by Farina et al. is helpful here, warranting some enthusiasm for balneotherapy, keeping in mind that the effect itself may be modest.[26]

Herbal Therapy

Herbal therapy encompasses an enormous, complicated, and somewhat confounding area of alternative medicine. While many of these studies fall under the rubric of traditional Chinese medicine (TCM), others appear to focus more directly on the pharmacological aspects of the plants involved. Even those studies claiming to investigate aspects of TCM must be reviewed with caution: while TCM is a beautiful, intricate system of health and disease that has very ancient roots, its many variations and schools make any proclamations about the ''efficacy of TCM'' rather tenuous.[27] In addition, as with conventional medications, herbal therapies may have adverse effects—both intrinsically from the pharmacological effects of the herbs themselves, as well as from extrinsic contaminants like heavy metals that may adulterate some herbal preparations.[28]

There are a number of small- and medium-sized RCTs that show significant improvements in AD with TCM herbal preparations over placebo ( Table 4 ). A systematic review in 2013 examined seven RCTs and concluded that TCM herbal medicine significantly improved AD severity including erythema, pruritus, and sleep over placebo (p<0.0001, p<0.0001, and p<0.00001, respectively) and was well-tolerated, but that the poor quality and heterogeneity of the studies did not allow for meta-analysis or ''valid conclusions''.[29]

Looking at some of these studies more closely, an RCT of 71 patients found that Xiao-Feng-San herbal preparation significantly ameliorated disease severity, itch, and sleep loss compared with placebo.[30]Hwang-Yeon-Hae-Dok-San (TJ-15) and Ou-Ryung-San (TJ-17) are two other oral herbal formulas that have been used to treat AD. Choi et al. determined that treatment with TJ-15 alone and TJ-15 plus TJ-17 reduced symptom severity in a study of 24 patients; however, there was no significant difference between the two groups, and there was no placebo group. The authors reported that both formulations provide safe and effective treatment for patients with the ''dampness-heat pattern'' type of AD.[31]

TCM herbs have also been used in combination with acupuncture. The results of a study of 60 patients with mild-to-severe disease suggested that adjunctive treatment with combined ''flying needle'' acupuncture and TCM herbal medicine may be more effective than herbal medicine alone (p<0.05).[32] Adverse effects were minimal and included nausea, dizziness, vomiting, fatigue, and headaches. This supports the aforementioned conclusion that acupuncture appears to have a small, but measurable effect in treating AD, but unfortunately does not give us much insight into the effect of the herbal medicine given the structure of the trial.

Sheehan and colleagues conducted a double-blinded, placebo-controlled study of TCM, which demonstrated positive effects on AD in both adults and children.[33,34] Using a crossover study design, Sheehan et al. recruited 40 adults and 47 children with refractory AD and demonstrated that a TCM preparation (a mixture of ten plant extracts, referred to as ''Zemaphyte'') significantly improved erythema and surface damage, as well as subjective itching and loss of sleep when compared with placebo. In contrast, results from a subsequent randomized trial of the same mixture conducted by Fung et al. failed to show a significant treatment effect over placebo. However, the authors importantly acknowledge that the latter study's lower dosing, shortened dosing schedules, and distinct patient population could have contributed to these divergent findings given the known differences in drug bioavailability and metabolism.[35]

In addition to the rare—but real—pharmacotoxic effects of some TCM herbs (including herb–drug interactions, variability in contents, and potential contaminants), many of these preparations have also proven to be unpalatable, particularly for children. To address this treatment barrier, Henderson et al. compared two different preparations of Zemaphyte and found that freeze-dried preparations were more palatable and equally as effective at improving erythema and surface damage when compared with traditional preparations. However, this study is limited by a lack of placebo.[36]

In animal models, the orally administered herbal preparation Hochu-ekki-to has demonstrated various immunopharmacological effects, especially anti-allergy activity. Kobayashi et al. conducted a multicenter RCT with 91 patients to evaluate this treatment, and found that Hochu-ekki-to reduced the amount of topical steroids and/or tacrolimus used for AD treatment without aggravating the disease.[37]

There are also studies that did not find significant improvement with oral herbal therapy. Hon et al. found that a formulation containing five herbs based on a commonly used concoction (Flos lonicerae, Herba menthae, Cortex moutan, Rhizoma atractylodis, and Cortex phellodendri) did not significantly decrease disease symptoms or severity compared with placebo.[38] Another study that investigated an oral herbal combination of Eleutherococcus senticosus, Achillea millefolium, and Lamium album found a similar lack of efficacy.[39] Despite finding no improvement in AD symptoms or severity, Hon et al. did find a >30 % improvement in the CDLQI (p = 0.008) and a significant reduction in topical corticosteroid use compared with placebo.[38]

Outside of the TCM rubric, Polypodium leucotomos (PL) extract, an antioxidant derived from a Central American fern, has been used to treat several inflammatory and atopic diseases in addition to its role as a photoprotective agent.[40] In Spain, PL extract has been available for many years as a drug product licensed for treatment of adults and children with AD. In an RCT of 105 patients designed to evaluate whether daily administration of oral PL extract would reduce topical corticosteroid use in children with AD, PL extract significantly reduced oral antihistamine use, but did not significantly reduce corticosteroid use compared with placebo.[41] While the significance here is somewhat marginalized given the fairly minor role of oral antihistamines in the treatment of AD, these results suggest that PL extract may offer some itch relief, and thus may warrant further study.

Beyond oral preparations, topical applications of herbal formulations also exist. One such topical herbal preparation used for treating AD is Hypericum perforatum (St. John's wort extract). This herb contains hyperforin, a compound with suspected anti-inflammatory properties that has traditionally been used to treat burns and other wounds. A randomized, double-blind, placebo-controlled, split-body comparison study of 21 patients with mild-to-moderate AD demonstrated that 1.5 % hyperforin cream was significantly better than vehicle control upon assessment with a modified SCORAD at weeks 1, 2, and 3 (p<0.05).[42] Notably, reduction in skin colonization of Staphylococcus aureus trended towards the hyperforin group (but did not reach significance), and no serious adverse events were reported. Although these are very limited data, these results undoubtedly warrant further study, and remain promising.

In another split-body comparison study, Kamillosan cream (which contains chamomile extract) was tested against 0.5 % hydrocortisone cream and a vehicle placebo. After a 2-week treatment period, Kamillosan cream demonstrated modest superiority over 0.5 % hydrocortisone cream, but showed only a marginal difference against placebo.[43] Results from another RCT of 108 patients suggested that both 1 and 2 % licorice extract significantly reduced erythema, edema, and pruritus compared with placebo, with 2 % extract proving more effective than 1 % extract.[44] On the other hand, a topical ointment containing Mahonia aquifolium, Viola tricolor, and Centella asiatica extract did not significantly reduce symptoms compared with vehicle control in a study of 88 patients with mild-to-moderate AD.[45]

While heterogeneous, these studies are fairly large and numerous by the standards of AD trials, and their collective results make it difficult to dispute that herbal preparations can have real effects in the treatment of AD. However, the complexity of these preparations and their potential risks make this area inscrutable for most practitioners without formal training in herbology. There remains an important asterisk as well: while there is real evidence of positive effect, there are simply too many unanswered questions to warrant routine clinical use of such herbs, and significant further research is needed before widespread clinical adoption can occur.

Topical Oils

Dry skin and impaired barrier function are defining characteristics of AD, so it is not surprising that the topical application of different oils remains a popular alternative therapeutic modality. From a conventional standpoint, the very act of moisturizing is a mainstay of treatment—and there are some very promising plant oils that may confer additional therapeutic value on top of their emollient properties.

For example, sunflower seed oil (Helianthus annuus) contains high levels of linoleic acid, a compound that may improve the skin barrier and yield anti-inflammatory effects.[46] An open, randomized trial of 80 children with mild to moderate AD found that a 2 % sunflower oleodistillate cream exhibited significant improvement in disease severity and quality of life comparable to hydrocortisone butyro-propionate at days 7 and 21 of treatment (p<0.01).[47] When compared with topically applied olive oil, sunflower seed oil was found to preserve stratum corneum integrity and improve skin hydration, while the latter significantly reduced stratum corneum integrity and induced erythema.[48] The results suggest that the use of olive oil in the treatment of dry skin and infant massage should be discouraged, as poor skin barrier is directly linked to greater allergen and irritant penetration, as well as onset and severity of AD ( Table 5 ).[49] More generally, these findings directly challenge the common claim that all natural oils are beneficial to the skin.

Skin colonization by S. aureus is abnormal, but is a well-known feature in patients with AD.[50] While initially thought to be mostly innocuous, more recent studies suggest that the bacteria secrete a toxin that can actually drive the skin disease, thus making anti-staphylococcus measures of rising importance.[51] An RCT with 52 patients compared virgin coconut oil to virgin olive oil and found that while both oils reduced S. aureus colonization and improved disease severity, coconut oil was significantly superior: 95 % of subjects in the coconut oil group were cleared of S. aureus versus 50 % of subjects in the olive oil group (p = 0.004).[52] In another study, including 117 patients, virgin coconut oil was compared with mineral oil. Again, while both groups found significant improvement in clinical (SCORAD) and instrumental (TEWL and skin capacitance) assessments compared with baseline, the virgin coconut oil was significantly superior on all accounts compared with mineral oil.[53] What is particularly compelling here is the dual functionality of coconut oil; having both emollient and anti-bacterial properties makes for a highly appealing therapy in AD.

Evening primrose oil (Oenothera biennis) contains high levels of gamma-linolenic acid (GLA) and omega-6 fatty acids, which may play a role in barrier repair and antiinflammatory pathways.[54] A study comparing topical evening primrose oil to placebo using a split-body comparison in 12 patients demonstrated significant improvement in subjective patient symptoms. Physician-assessed clinical improvement was also noted, but did not reach significance.[55] Similarly, although a different study observed that three emollients enriched with GLA improved symptom severity and TEWL compared with placebo, none of these effects reached statistical significance.[56]

Similarly, borage oil has been examined as a topical treatment for AD, as it also contains high levels of GLA and omega-6 fatty acids. Sixteen children wearing pure organic cotton undershirts coated with borage oil experienced significant improvement in erythema, itch, and TEWL after two weeks of treatment (p = 0.048), while 16 children wearing placebo undershirts demonstrated no such improvement during the study period.[57] Of note, the treatment and control groups were not compared with each other; each group was compared only to itself at baseline.

Finally, 99 children with mild to moderate AD were randomized to receive pale sulfonated shale oil cream (also known as ''ichthyol,'' and related to ''ichthammol''; a mixture of many different compounds obtained by distillation of oil shale) or placebo. After 4 weeks, there were significantly improved AD symptoms in the shale oil cream group compared with vehicle control (p<0.0001).[58]

These studies suggest that topical sunflower seed oil and virgin coconut oil hold promise for the treatment of AD, while olive oil should likely be avoided. Topical pale sulfonated shale oil has one positive study that clearly warrants follow-up, especially as tar preparations are known to have an effect in AD. Topically applied borage oil and evening primrose oil may have some benefits as well, but the somewhat mixed results suggest that the effect is either very small or that it is perhaps only applicable to a subset of patients—if it exists at all.

Oral Oils and Fatty Acids

The concept that a deficiency in essential fatty acids (EFAs) may contribute to the pathogenesis of atopic disorders dates as far back as 1937, when reduced plasma levels were measured in patients with eczema.[59] However, trials of EFA supplementation in patients with AD have since yielded inconsistent results.

GLA has been shown to correct deficiencies in skin lipids and improve inflammation and immunity in AD patients, but clinical studies have been mixed. The role of GLA in AD has been studied since the 1980 s, with commonly used sources including borage oil (containing 23 % GLA) and evening primrose oil (containing 8–10 % GLA). While there is at least some measurable effect with topical application borage and evening primrose oils (as discussed above), the same cannot be said for oral administration—at least, not in aggregate. Indeed, the Cochrane Review published a systematic review of borage and evening primrose oils and concluded that these concoctions are ''not effective treatments for eczema'' when administered orally.[60]

Three prominent studies evaluated the effectiveness of orally ingested borage oil in the treatment of AD, and all reached similarly disappointing conclusions. In the first study, including 160 patients, several clinical symptoms improved in response to borage oil containing high proportions of GLA (minimum 23 %) when compared with placebo. However, overall improvement as measured by corticosteroid use did not reach statistical significance ( Table 6 ).[61] Similarly, in another study, including 151 patients, Takwale et al. determined that borage oil supplements did not provide any significant benefit over placebo in any outcome measure in either adult or pediatric patients.[62] Finally, a study of 31 patients with AD found that while borage oil significantly improved disease symptoms compared with baseline, these results failed to significantly differ from placebo outcomes.[63] These studies collectively suggest that although borage oil is well-tolerated, it does not consistently provide therapeutic benefits when taken orally.

Randomized trials of evening primrose oil have failed to come to a consensus. Bordoni et al. studied 24 patients and found that oral evening primrose oil significantly improved eczema symptoms compared with placebo.[64] Similarly, Schalin-Karrila et al. studied 25 patients and found that while both the evening primrose oil and placebo groups demonstrated improvement in disease severity and grade of inflammation compared with baseline, the evening primrose oil group did show a significantly greater improvement over placebo. Furthermore, the evening primrose oil group also demonstrated a statistically significant improvement in dryness, pruritus, and in the percentage of body involved in disease.[65] Humphreys et al. conducted a study of 58 moderate-to-severe AD patients and showed that evening primrose oil significantly improved erythema and surface damage (but not lichenification) compared with placebo.[66] Finally, in a study of 25 patients, evening primrose oil significantly improved all measured disease symptoms when compared with placebo.[67]

Two relatively large studies also examined treatment with oral evening primrose oil. The first large trial, a double-blinded crossover trial of 154 patients with AD by Bamford et al. found that evening primrose oil did not significantly improve erythema, scale, excoriation, lichenification, or overall severity compared with placebo.[68] Importantly, this study was excluded from some earlier meta-analyses because of allegations that the placebo group and experimental groups were muddled.[69] In a sobering essay, Dr. Williams suggests that important data on evening primrose oil were not made publicly available, making such analyses impossible.[69] Another study of 123 subjects by Berth-Jones and Graham-Brown showed that evening primrose oil with and without fish oil did not significantly improve disease symptoms compared with placebo.[70] Both trials ultimately concluded that GLAcontaining evening primrose oil did not show an effect superior to placebo.

The question of a potential dose-dependent effect of evening primrose oil has been considered. Biagi et al. examined high dose evening primrose oil and found that it significantly reduced disease severity compared with placebo and low dose evening primrose oil.[71] Another study of children and adolescents—too recent for the latest Cochrane Review—also concluded that high dose evening primrose oil significantly improved disease symptoms compared with baseline (p<0.001), while low dose evening primrose oil also improved symptoms, but not to a level reaching statistical significance (p = 0.550).[72] Though these two studies were designed to evaluate a comparative dose-dependent effect, both were limited by lack of a placebo group. Finally, in a study of 39 children and 60 adults, high dose evening primrose oil significantly improved subjective evaluation of itch, scaling, overall severity, and physician-assessed overall severity, while low dose evening primrose oil significantly improved only itch.[73]

In summary, despite the vehemence of the venerable Cochrane Review, more granular evaluations of the evidence on evening primrose oil (as well as the recent doseresponse study) make it difficult to dismiss completely; evening primrose oil may yet be proven useful as an adjunct to the treatment of AD. It is possible, for example, that a particular sub-group may be identified for whom evening primrose oil will make a more consistent and clinically relevant difference. Nevertheless, further work needs to be done to standardize dosing regimens. The promise of potentially unconfirmed benefits, along with the minimal risk of adverse events suggests that oral evening primrose oil probably deserves further study, although this will probably frustrate those who have already fully dismissed it.

In addition to GLA supplementation, a number of studies have investigated the efficacy of fish oil because of its potential anti-inflammatory properties. A randomized trial of 22 patients with moderate-to-severe AD compared infusions of 10 % fish oil (omega-3) to 10 % soybean oil (omega-6) and found that though both treatments showed improvement, change in disease severity score was more pronounced (p<0.05) in the fish oil group than the soybean oil group. Of note, patients who received fish oil treatment experienced greater relapse post-treatment when compared with the soybean oil group, which showed nearly constant long-term improvement in the post-treatment phase.[74] Similarly, after 12 weeks of oral supplements with fish oil, Bjorneboe et al. reported improvement in disease severity compared with parallel oral placebo supplement.[75] However, these positive findings were undermined by the largest well-reported trial, which showed no significant difference between fish oil and placebo.[76]

Other oral EFA supplements have been studied, but results have often been contradictory and weakened by a much smaller body of evidence. Hempseed oil is a rich source of omega-6 and omega-3 polyunsaturated fatty acids, both of which possess anti-inflammatory properties. One study of 20 patients found that dietary hempseed oil significantly improved subjective evaluations of skin dryness and itching, and reduced topical medication usage when compared with an olive oil control.[77] Additionally, docosahexaenoic acid (omega-3 fatty acid) supplementation resulted in significant clinical improvement, while improvement seen with placebo failed to reach significance.[78] In contrast, in a trial comparing linolenic acid (omega-6) versus eicosapentaenoic acid (omega-3) with docosahexaenoic acid (omega-3), only linolenic acid demonstrated significant symptom reduction compared with placebo.[79]

As Bath-Hextall et al. concluded with their 2012 review, while there are some promising signals in these studies, convincingly positive results from larger, well-designed studies are necessary before clinical action can be recommended for oral oil and fatty acid supplementation. Given the reported positive outcomes and the good theoretical basis for their role in dampening the inflammatory pathway, we agree with Bath-Hextall et al. that further studies of fish oil might be worthwhile.[80]

Vitamins and Minerals

Vitamins are a bastion of the alternative medicine movement, and multiple vitamins have been studied as possible therapies in AD. Pyridoxine hydrochloride (vitamin B6) did not significantly improve disease symptoms compared with placebo in 48 patients with moderate-to-severe AD ( Table 7 ); however, this study was limited by its short duration of 4 weeks.[81] Selenium (600 lg) with or without vitamin E (600 IU) was studied in 60 patients, but neither selenium nor the combination improved disease severity compared with placebo.[82] In a study of 50 children with AD, Ewing et al. examined daily administration of oral zinc sulphate and found that it did not significantly improve disease symptoms compared with placebo. Of note, the authors reported that some patients actually seemed to worsen or develop additional eruptions that may have been caused by zinc supplementation.[83]

Vitamin D, on the other hand, has a much more robust—yet somewhat controversial—literature foundation behind its use. In fact, even recent guidelines for AD management disagree over the role of vitamin D supplementation: the American Academy of Dermatology's guidelines note that there are insufficient data to recommend it, while the Joint Task Force on Practice Parameters (representing several US-based allergy and immunology groups) supports its use.[84] Two RCT's by Amestejani et al. and Camargo et al. have indeed found that vitamin D supplementation significantly improved disease symptoms; however, the former study was limited by not comparing vitamin D with placebo, while the latter only examined a specific population of Mongolian children with winter-related AD.[85,86] In contrast, two other well-designed studies found no significant improvement with supplementation.[87,88] Notably, however, a double-blind, placebo-controlled RCT comparing 1600 IU per day of vitamin D with or without 600 IU of vitamin E (tocopherol) found that the combination supplementation resulted in greater improvement of AD severity versus vitamin D alone, as well as versus placebo.[89]

Topical formulations of vitamins have also been studied. A trial of 0.05 % clobetasol with and without 2.5 % zinc sulphate in 47 patients with hand eczema found that while both treatments significantly improved disease symptoms, clobetasol with zinc sulphate was significantly more effective, and was also associated with significantly lower disease recurrence.[90] Topical vitamin B12—thought to attenuate eczema-related symptoms by reducing nitric oxide levels—was found to significantly improve disease symptoms compared with placebo in both pediatric and adult populations (p = 0.01 and p<0.001, respectively).[91,92]

Studies of vitamin and mineral supplementation continue to be mired in controversy in many areas of medicine,[93] and bring with them many limitations, including small sample sizes, relatively short study durations, and the mechanistic reality that their effect sizes will likely be modest. From the available data, it seems reasonable to assign oral vitamin D supplementation and topical vitamin B12 as potential candidates that are worthy of more research. Their low cost, relatively safe side effect profile, and familiarity to patients make the risk–benefit ratio very favorable for vitamins, further underscoring the need for more studies.


There are some alternative interventions that defy easy classification or toe the line between conventional and unconventional therapies. One such example is bioresonance therapy, which purports to use electromagnetic waves to diagnose and treat disease. In a study involving 32 hospitalized Swiss children with AD, bioresonance therapy showed no statistically significant improvement of disease severity compared with sham therapy ( Table 8 ). However, this study is limited in terms of external validity because of its very specific population.[94]

Another trial, involving 117 patients, examined topically applied furfuryl palmitate—an antioxidant compound—in comparison to vehicle control. Though furfuryl palmitate and the vehicle control both showed significant clinical improvement from baseline, the vehicle control was more efficacious overall.[95] Additionally, an interesting trial demonstrated that cyclosporin with glucosamine improved symptoms better than cyclosporine alone—but not to a statistically significant degree.[96]

A randomized, crossover, single-blinded study examined children with cow's milk allergy and AD who were assigned to drink goat's milk or donkey's milk for 6 months. The study found that the donkey's milk group had significantly improved AD symptoms (p<0.03) compared with both the goat's milk group and with baseline.[97] However, multiple adverse effects were reported, including urticaria, angioedema, eczema exacerbation, wheezing, diarrhea, and abdominal pain.

Isolated studies like these are difficult to interpret and, unfortunately, must remain curiosities until further data can be obtained.