REVIEW

Chinese herbal medicine for atopic dermatitis: A systematic review

Hsiewe Ying Tan, BAppSci,a Anthony Lin Zhang, PhD,a DaCan Chen, PhD,b Charlie Changli Xue, PhD,a,b and George Binh Lenon, PhDa Bundoora, Australia, and Guangzhou, China
Background: Atopic dermatitis (AD) is a chronic, itching skin disease, and conventional therapies offer inadequate symptom management. Patients with AD are increasingly turning to Chinese medicine. Objective: We systematically evaluated the clinical evidence of the efcacy and safety of oral Chinese herbal medicine for AD. Methods: Searches were conducted on major electronic databases using the following key words: randomized controlled trials, atopic dermatitis, traditional Chinese medicine, traditional East Asian medicine, herbal medicine, Chinese herbal drugs, medicinal plants, phytotherapy, Kampo medicine, and Korean traditional medicine. The results were screened to include English/Chinese randomized controlled trials. A metaanalysis was conducted on suitable outcome measures. Results: Seven randomized controlled trials were included (1 comparing Chinese herbal medicine and Western medicine with Western medicine alone; 6 comparing Chinese herbal medicine with placebo). Combined Chinese herbal medicine with Western medicine was superior to Western medicine alone. Three placebo controlled trials showed signicant treatment efcacy and 2 showed signicantly reduced concurrent therapy with Chinese herbal medicine. No abnormalities in safety prole or severe adverse events were reported. Limitations: A metaanalysis of all included studies could not be conducted because of study heterogeneity. Conclusions: Chinese herbal medicine signicantly improved symptom severity of AD and was reported as well tolerated. However, the poor quality of studies did not allow for valid conclusions to support its tolerability and routine use. Additional studies addressing the methodologic issues are warranted to determine the therapeutic benet of Chinese herbal medicine for AD. ( J Am Acad Dermatol 10.1016/ j.jaad.2013.01.019.) Key words: allergy; alternative and complementary medicine; atopic dermatitis; Chinese herbal medicine; eczema; systematic review.

topic dermatitis (AD) is a chronic, inammatory skin disease that affects approximately 15% to 30% of children and 2% to 10% of adults.1 It is characterized by severe itching, skin

redness and dryness, weeping, scarring, and lichenification.2,3 AD impacts activities of daily living and increases financial burden of patients and their carers.4 In the United Kingdom, the annual

From the Traditional and Complementary Medicine Research Program,a Health Innovations Research Institute, School of Health Sciences, Royal Melbourne Institute of Technology University, Bundoora Campus, Victoria, Australia, and the Guangdong Provincial Hospital of Chinese Medicine,b Guangzhou, China. Supported by the Guangdong Provincial Academy of Chinese Medical Sciences, Guangdong Provincial Hospital of Chinese Medicine, China. Conflicts of interest: None declared.

Reprint requests: George Binh Lenon, PhD, Traditional and Complementary Medicine Research Program, Health Innovations Research Institute, School of Health Sciences, RMIT University, Bundoora, VIC 3083, Australia. E-mail: george. [email protected]. Published online June 04, 2013. 0190-9622/$36.00 2013 by the American Academy of Dermatology, Inc. http://dx.doi.org/10.1016/j.jaad.2013.01.019

2 Tan et al

J AM ACAD DERMATOL

expenditure on AD was approximately 465 million5; Study selection criteria in the United States, the national direct costs ranged RCTs with the following criteria were included: from $364 million to $3.8 billion USD.6 In Australia, published English or Chinese RCTs; diagnosis of AD the annual personal financial costs ranged from $330 (or atopic eczema) using clinical diagnosis or valito $1255 AUD, and is said to be greater than the dated diagnostic criteria; diagnosis of eczema was management costs of asthma.7 only accepted when referring to children/infants; The exact pathophysiology of AD has not placebo, no treatment or non-Chinese medicine been fully elucidated, and treatments as control interthere is currently no cure. ventions; non-Chinese mediCAPSULE SUMMARY Symptomatic management cine cointerventions were has been the focus of manaccepted only if it was apPrevious reviews on Chinese medicine agement,8 involving the plied to all groups; and $ 1 for atopic dermatitis were inconclusive. long-term use of topical corof the following outcome Clinical trials examining the safety and ticosteroids and calcineurin measures: (1) disease/sympefficacy of Chinese medicine for atopic inhibitors. However, these tom severity scoring and (2) dermatitis have since been conducted treatments have been associquality of life. Concurrent but have yet to be systematically ated with adverse events9,10 therapies, adverse events, reviewed. and the development of drug and safety proles were reA metaanalysis revealed significant tolerance.11 corded as secondary outimprovement in symptom severity and Consequently, patients come measures. Studies quality of life, but the quality of studies frequently seek other therainvolving other forms of was poor. peutic options, such as TCM therapy (eg, acupuncChinese herbal medicine Patients can potentially reduce steroid ture and topical CHM) or (CHM).12 CHM is part of tratherapy use with the help of Chinese dermatitis (eg, neurodermaditional Chinese medicine medicine and subsequently reduce the titis and contact dermatitis) (TCM) therapy, whereby inoccurrence of related side effects. were excluded. quiry of the condition and observation of the patient Data extraction and risk leads to the TCM diagnosis and treatment. Two of bias assessment previously published systematic reviews on CHM Two independent reviewers extracted the data of for AD were inconclusive,8,13 and the current state of included studies onto the Cochrane Skin Group data evidence of CHM treatment of AD remains unextraction form and conducted a risk of bias assessknown.4,13,14 The different pathways of medication ment using the Cochrane Collaborations tool for administration affect treatment actions and indicaassessing risk of bias16 (Fig 1). The metaanalysis was tions; oral and topical CHM are therefore considered conducted in RevMan5. For CHM and Western meddifferent interventions.15 This systematic review aims icine (WM) compared with WM only, the metaanalto evaluate the published randomized controlled ysis was performed on overall clinical scores. For trials (RCTs) on the safety and efficacy of treatment of CHM compared with placebo, the metaanalysis AD with orally administered CHM compared with was conducted on disease or symptom severity placebo, pharmacotherapy, or no treatment. scores, quality of life, and reduction in concurrent treatments.
d d d

METHODS
Protocol This systematic review was not registered with the Cochrane Collaboration but was conducted with reference to the Cochrane Handbook.16 Electronic searches were carried out on the following databases: CINAHL, Cochrane Library, Embase, ProQuest, PubMed, ScienceDirect, Scopus, Web of Science, VIP Database for Chinese Technical Periodicals (CQVIP), and China National Knowledge Infrastructure (CNKI). All databases were searched from their inception to 2011. The search terms were a combination of Medical Subject Heading terms and their synonyms (Table I).

RESULTS
Description of studies The searches yield a total of 1014 articles. After screening titles and abstracts, duplicates, nonEnglish or Chinese articles, animal or immunologic studies, studies not involving AD or CHM, and nonRCTs were excluded. Two hundred eighty-one full articles were retrieved for further evaluation. Seven studies were included for qualitative analysis (Fig 2). One study was excluded from the metaanalysis because of insufficient data. The 6 studies in the metaanalysis included a total of 432 participants with AD.

J AM ACAD DERMATOL

Tan et al 3

Table I. Example for search strategy (in PubMed)

No. Search term Hits*

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Randomized controlled trial (publication 315,419 type) Randomized controlled trials as topic 75,706 (MeSH) Dermatitis, atopic (MeSH) 12,635 Medicine, Chinese traditional (MeSH) 9495 Medicine, East Asian traditional (MeSH) 11,336 Herbal medicine (MeSH) 1,116 Drugs, Chinese herbal (MeSH) 22,232 Plants, medicinal (MeSH) 48,114 Phytotherapy (MeSH) 23,989 Medicine, Kampo (MeSH) 353 Medicine, Korean traditional (MeSH) 115 Terms 1 or 2 385,792 Terms 4 or 5 or 6 or 7 or 8 or 9 or 10 or 11 93,361 Terms 3 and 12 and 13 28

MeSH, Medical Subject Heading. *As of December 19, 2011.

Table II summarizes the studies characteristics. Table III summarizes the diagnoses, interventions, and outcome measures. One study was a single blinded RCT comparing WM with CHM (Jian Pi Shen Shi granules) with WM alone17; 6 studies were placebo controlled, double blinded RCTs (1 used Xiao Feng San,18 3 used the formula from the study by Sheehan et al,19-21 1 applied a Pentaherb formula,22 and 1 used Hochu-ekki-to10). The ingredients of each formula are listed in Table IV. Only 2 RCTs included TCM diagnosis,10,17 and 1 measured quality of life.22 CHM and WM combination compared to WM alone Only one study compared a combination of CHM and WM treatment to WM alone.17 Primary outcome measure. Disease/symptom severity scoring. The overall clinical score was calculated by summing the severity score (0-3) of each symptom (ie, itch, erythema, papules, exudation, erosion, inltration, lichenication, dryness). A signicant difference was shown in the end score (mean difference [MD], 2.56; 95% condence interval [CI], 3.46 to 1.66; P \ .00001), favoring the combination treatment. CHM compared to Placebo Five studies compared CHM with placebo. Two studies compared the formula used by Sheehan et al21 with a placebo of inert plant materials of similar appearance, taste, and smell, but without known benefits to AD20,21; another study compared

Pentaherb with an identical-looking placebo containing corn starch and caramel22; 1 study compared Xiao Feng San with a placebo made of caramel, lactose, and starch18; and the fifth study compared Hochu-ekki-to with placebo, with no details of placebo content.10 The authors of the Pentaherb trial22 provided raw data for analysis. The Hochuekki-to trial presented results in graphs only10; consequently, estimate figures from the graphs were used in analysis. Primary outcome measures. Disease/symptom severity scoring. Three studies reported overall clinical scores1 used the validated Scoring Atopic Dermatitis (SCORAD) instrument,22 another used the scoring system by the Atopic Dermatitis Severity Evaluation Committee of Japanese Dermatological Association,10 and the third study used a standardized scoring system calculating the extent and severity of dermatitis.18 The former 2 studies presented results as means and standard deviations, and the latter study presented the scores as least-square means. Only 1 trial showed significant improvement in overall clinical score by CHM compared to placebo.18 Because of the heterogeneity of instruments, metaanalysis of overall clinical scores could not be conducted. Three studies18,20,21 used a standardized scoring system to calculate the extent and severity of erythema and surface damage. Erythema and surface damage scores were presented as least-squares means, percentage of median change, and geometric mean, respectively. The metaanalysis of erythema scores (standard mean difference, 0.76; 95% confidence interval [CI], 1.05 to 0.47; P \.00001) and surface damage scores (standard mean difference, 1.08; 95% CI, 1.59 to 0.56; P \.0001) favor CHM compared to placebo (Fig 3). One trial reported pruritus score.18 A significant difference was seen in CHM treatment compared to placebo (mean difference [MD], 1.10; 95% CI, 1.59 to 0.61; P \ .0001). In a crossover trial by Sheehan et al,21 out of 31 participants, 14 reported improvement in itching during CHM treatment and 1 during the placebo phase. Cheng reported that CHM signicantly improved sleep scores when compared with placebo (MD, 0.80; 95% CI, 1.12 to 0.48; P \ .00001).18 Two other studies reported sleep improvement.20,21 In the former study, out of 37 participants, 19 had improved sleep during CHM treatment and 3 during placebo treatment; in the latter study, out of 31 participants, 15 experienced improved sleep with CHM and 6 with placebo. Quality of life. One study measured quality of life.22 The metaanalysis found that CHM significantly

4 Tan et al

J AM ACAD DERMATOL

One study reported 2 cases of mild transient gastrointestinal upset during CHM treatment.18 Three studies10,21,22 reported the occurrence of mild or moderate side effects, such as gastrointestinal upset, various dermatoses (including new rash, hives, acne pustulosa, and facial herpes), and dizziness in both the treatment and control groups. One study in children required participants parents to complete a questionnaire relating to adverse effects but did not report the occurrence of any adverse events.20

DISCUSSION
Results from the metaanalysis revealed signicant improvement in disease severity scores by the combination treatment of CHM and WM compared with WM alone (P \ .00001). When compared with placebo, CHM revealed significant improvement in erythema (P \ .00001), surface damage (P \ .0001), pruritus (P \ .0001), sleep scores (P \ .00001), and quality of life (P \ .05). CHM also significantly reduced the need for concurrent pharmacotherapy (P \ .00001). The overall risk of bias assessment found that the quality of studies was poor; therefore, the results from the metaanalysis have to be translated with caution. Only 1 study was judged with low risk in all domains for bias assessment,18 while the other studies had unclear or high risk judgments in $ 1 domains.10,17,19-22 In the study comparing CHM and WM with WM alone, despite being labeled as a single blind trial, there were no details on blinding. Aside from selective reporting and a lack of details regarding random sequence generation and allocation concealment, the main flaw was the lack of intention to treat (ITT) analysis, leading to incomplete outcome data. ITT analysis stresses that any exclusions from analysis would affect the comparability generated through randomization between groups.23 In addition, dropouts may reflect a flaw of the intervention.24 The 4 studies with high risk in this domain had exclusions likely related to the intervention, such as side effects, noncompliance, and the use of prescribed drugs.10,17,20,21 One study that excluded 2 participants at baseline because of family objections18 was considered low risk because the withdrawals were not related to the intervention. Two previous systematic reviews evaluating CHM for AD focused on the clinical evidence of the formula used by Sheehan et al.21 The conflicting evidence with regard to the efficacy of the formula used by Sheehan et al21 was said to be related to the variance in dosage, dropout rates, or racial variability in drug responsiveness.8 In addition, the investigators did not provide details on the chemical properties of the herbal placebo; its similar smell and taste

Fig 1. Summary of risk of bias assessment.

improved Childrens Dermatology Life Quality Index (CDLQI) compared to placebo (MD, 2.50; 95% CI, 4.77 to 0.23; P \ .05). Secondary outcome measures. Concurrent treatment. Kobayashi et al10 measured the effects of CHM on the amount of concurrent topical treatments used by participants during the trial. At the end of the study, the total equivalent amount of topical agents used was significantly lower in the CHM group compared to the placebo group (MD, 24.50; 95% CI, 27.92 to 21.08; P \ .00001). In the study by Hon et al,22 the amount of topical mometasone furoate used at the end of the trial was significantly lower in the CHM group (P = .024) compared to the placebo group.22 Safety prole and adverse events. Out of the included studies, 1 study did not evaluate safety prole and occurrence of adverse events.17 In the other studies, no significant differences in safety profiles were reported. However, there was 1 case of transient elevation in aspartate aminotransferase that was reversed within 8 weeks of treatment cessation.18

J AM ACAD DERMATOL

Tan et al 5

Fig 2. Study selection process. CHM, Chinese herbal medicine; TCM, traditional Chinese medicine.

to the active treatment indicates a possibility of containing similar chemical properties and, subsequently, similar or other pharmacologic actions. From the metaanalysis of the placebo controlled trials, 3 studies favored CHM compared to placebo, 2 of which were studies by Sheehan and Atherton20 and Sheehan et al.21 The third trial used a CHM formula commonly used for atopic dermatitis, Xiao Feng San. However, participants were given CHM granules with daily doses that were significantly higher than the recommended dose.25 Also, the placebo of caramel, lactose, and starch may pose a risk to lactose intolerant individuals. In addition, according to TCM theory, sweet foods such as caramel could aggravate skin conditions.26

In contrast, 2 trials found no signicant difference in clinical scores. The lack of efcacy in the study by Hon et al22 may be related to an inadequate dose of CHM.27 Although the capsule dose calculation was not stated, the Pentaherb formulation consisted of 9 g of raw herbs for patients 5 to 21 years of age.22 Despite the lack of efficacy, the Pentaherb group saw significant improvement in CDLQI, decrease in days of corticosteroid use and amount of mometasone furoate. However, it was unclear if the days of corticosteroid use comprised of corticosteroids other than mometasone furoate, including those of higher potencies. Kobayashi et als study10 only included patients with Kikyo (Qi deficiency) constitution, which did not address the condition as per

6 Tan et al

J AM ACAD DERMATOL

Table II. Characteristics of the included studies

Run-in/treatment/ follow-up/washout period

Author (year)

Age group

Study design

No. of participants

Dropout

Cheng et al (2010) Not specified

Fung et al (1999)

Double blind RCTcomputer generated randomization list by an independent statistician 7-50 years of age Double blind, crossover RCT

TCM, 47; placebo, 24 2 (dropped out at baseline, not included in ITT)

Run-in, not mentioned; treatment, 8 wks; follow-up, 4 wks

TCM, 40; placebo, 40 3 (ITT analysis not mentioned)

TCM, 42; placebo, 43 ITT analysis to 5-21 years of age Double blind include all RCTcomputer participants generated randomization code Run-in, not TCM plus WM, 49; 6 (ITT analysis Huang et al (2004) 3-11 years of age Single blind mentioned; WM, 49 used to analyze RCTsimple overall treatment treatment, 4 wks; randomization follow-up, 3 effect) method (ratio 1:1) months TCM, 43; placebo, 48 7 (excluded from Run-in, not Kobayashi et al 20-40 years of age Double blind analysis) mentioned; (2010) RCTblock treatment, 24 randomization wks; follow-up, not mentioned Sheehan and Children (age Double blind, TCM, 47; placebo, 47 10 (excluded from Run-in, 4 wks; Atherton (1992) group not crossover RCT analysis) treatment, 8 wks; specified) follow-up, not mentioned; washout, 4 wks Sheehan et al 16-65 years of age Double blind, TCM, 40; placebo, 40 9 (excluded from Run-in, not (1992) crossover RCT analysis) mentioned; treatment, 8 wks; follow-up, not mentioned; washout, 4 wks Hon et al (2007)
ITT, Intention to treat; RCT, randomized controlled trial; TCM, traditional Chinese medicine; WM, Western medicine.

Run-in, not mentioned; treatment, 8 wks; follow-up, not mentioned; washout, 4 wks Run-in, 2 wks; treatment, 12 wks; follow-up, 4 wks

TCM theory and may have reduced the treatment efficacy. Nevertheless, there was a significant decrease in the total equivalent amount of topical agents in the CHM group when compared to the placebo group.10 These 2 studies indicate that CHM may function as an adjunct treatment for AD. The CHM in 4 of the studies underwent quality check for potential contaminants (including steroids),18,20-22 while the Hochu-ekki-to was manufactured by a Good Manufacturing Practiceecertified company.10 Three of the 5 studies that had adverse events reported nonsignificant differences between

both groups.10,20,21 However, the Pentaherb group in the study by Hon et al22 had significantly more general practitioner visits without further explanation, making it difficult to determine whether these visits were related to Pentaherb. There was 1 case of transient elevation of aspartate amino transferase that was reversed within 8 weeks of treatment cessation.18 Elevated aspartate amino transferase levels can have multiple causes, including alcohol abuse, medications, such as antihistamines or nonsteroidal antiinflammatory drugs, and certain herbs.28 However, the herbal ingredients of the respective

Table III. Diagnosis, interventions, and outcome measures of included studies

Author (year) WM/TCM diagnosis Severity Treatment interventions Control interventions Outcome measures

J AM ACAD DERMATOL

Cheng et al (2010) AD: Hanifin and Rajka diagnostic criteria Fung et al (1999) AD: Hanifin and Rajka diagnostic criteria

Hon et al (2007)

AD: Hanifin and Rajka diagnostic criteria

Huang et al (2004) AD: UK diagnostic criteria/ SP deficiency

AD: diagnostic criteria according to the Japanese Dermatological Association for AD/Kikyo condition (Qi deficiency) AD: diagnostic criteria not Extensive nonexudative Sheehans formula (decoction): 1-7 Placebo Sheehan and years of age, 2 large plus 2 small Atherton (1992) stated AD, not confined to sachets of herbs per day; 8-13 flexural sites years of age, 3 large plus 3 small sachets; [14 years of age, 4 large plus 4 small sachets (100 mL decoction/day) Sheehan et al AD: Hanifin and Rajka Extensive AD ([20% Sheehans formula (decoction): 4 Placebo (1992) diagnostic criteria BSA involved) large plus 4 small sachets (200 mL decoction/day)

Kobayashi et al (2010)

Xiao Feng San granules (3-7 years Placebo of age, 3 g tid; 8-12 years of age, 6 g tid; [13 years of age, 9 g tid Placebo Moderate to severe AD Sheehans formula (decoction): 7-13 years of age, 2 large plus 2 small sachets of herbs per day; [14 years of age, 3 large plus 3 small sachets Moderate to severe AD Pentaherb capsule 3 capsules bid Placebo (objective SCORAD [15) Moderate AD Jian Pi Shen Shi granules (3-5 years Cyproheptadine tablets 0.25 mg/kg/day tid; of age, 5 g tid; 6-11 years of age, triamcinolone urea 10 g tid plus cyproheptadine cream tablets 0.25 mg/kg/day tid; triamcinolone urea cream Hochu-ekki-to granules 3.25 g bid Placebo

Extensive AD ([20% BSA involved)

Total clinical lesion; erythema score; surface damage score; pruritus score; sleep score Clinical scores for erythema, surface damage, lichenification, and scaling

SCORAD; CDLQI; allergic rhinitis symptoms; concurrent treatment Rajka and Langeland scoring; overall treatment effect; total immunoglobulin E; rate of recurrence 3 months posttrial Skin severity score; dose of topical steroids/tacrolimus used; prominent efficacy rate; aggravated rate Erythema score; surface damage score; preference in treatment; improvement in ability to sleep

Erythema score; surface damage score; improvement in itching, sleep and asthma; preference of treatment

AD, Atopic dermatitis; bid, twice daily; BSA, body surface area; CDLQI, Childrens Dermatology Life Quality Index; SCORAD, Scoring Atopic Dermatitis; SD, spleen deficiency; tid, three times daily.

Tan et al 7

8 Tan et al

J AM ACAD DERMATOL

Table IV. Herbal ingredients used in Chinese herbal medicine formulas of included studies
Sheehans formula Jian Pi Shen Shi granules Pentaherb Hochu-ekki-to Xiao Feng San

Glycyrrhiza uralensis (Gan Cao) Ledebouriella seseloides (Fang Feng) Schizonepeta tenuifolia (Jing Jie) Lophatherum gracile (Dan Zhu Ye)

Wolfiporia extensa (Fu Ling) Codonopsis pilosula (Dang Shen) Atractylodes rhizoma (Bai Zhu) Aurantii nobilis pericarpium (Chen Pi)

Herba menthae (Bo He) Flos lonicerae (Jin Yin Hua) Cortex phellodendri (Huang Bai) Rhizoma atractylodis (Cang Zhu)

Paeonia lactiflora Semen coicis (Yi Yi Ren) Cortex moutan (Bai Shao) (Mu Dan Pi) Rehmannia glutinosa (Sheng Di Huang) Anebia clematidis (Chuan Mu Tong) Dictamnus dasycarpus (Bai Xian Pi) Tribulus terrestris (Ji Li) Potentilla chinensis (Wei Ling Cai)

Glycyrrhizae radix (Gan Cao) Ginseng radix (Ren Shen) Atractylodes rhizoma (Bai Zhu) Aurantii nobilis pericarpium (Chen Pi) Angelicae radix (Dang Gui) Bupleuri radix (Chai Hu) Zizyphi fructus (Da Zao) Astragali radix (Huang Qi) Zingiberis rhizome (Gan Jiang) Cimicifugae rhizome (Sheng Ma)

Glycyrrhiza uralensis (Gan Cao) Saposhnikovia divaricate (Fang Feng) Schizonepeta tenuifolia (Jing Jie) Atractylodes lancea (Cang Zhu) Angelica sinensis (Dang Gui) Rehmannia glutinosa (Sheng Di Huang) Clematis armandii (Chuan Mu Tong) Cryptotympana pustulata (Chan Tui) Linum usitatissimum (Hu Ma Ren) Anemarrhena asphodeloides (Zhi Mu) Gypsum fibrosum (Shi Gao) Sophora flavescens (Ku Shen) Articum lappa (Niu Bang Zi)

Fig 3. Metaanalysis of Chinese herbal medicine compared to placebo studies. CHM, Chinese herbal medicine; CI, confidence interval.

study were not for such effects. No additional details were provided to evaluate the possible relation between the elevation and the trial intervention. This review showed no signicant safety concerns regarding CHM treatment for AD. However, in these studies, CHM was only administered for 4 to 24 weeks under controlled conditions. The included studies applied different methods of medication delivery: 1 study used capsules,22 3 used

granules,10,17,18 and 3 used decoctions.19-21 Decoctions are the most common form of CHM, while granules and capsules are modern forms aimed to be more convenient and palatable.29 It remains unclear if the different methods of CHM delivery influence their treatment effects.29,30 However, studies with significantly better outcomes used higher treatment doses, regardless of the method of delivery.

J AM ACAD DERMATOL

Tan et al 9

According to TCM, AD is caused by a congenitally weak constitution, resulting in a predisposition toward atopic diseases and susceptibility towards external or internal pathogenic factors, such as wind, dampness, and heat.31 Its recurrent and chronic nature can injure Yin and blood and generate dryness.31 However, other dermatologic conditions, such as psoriasis, can share the same TCM diagnosis. Among the included studies, only 2 applied TCM syndrome differentiation, while the others used a standardized CHM formula. Nevertheless, each formula was able to reduce AD symptom severity, improve quality of life, and/or reduce the need for topical corticosteroids. This implies that there are clinical benefits of CHM treatment of AD, and that better treatment efficacy may be seen with TCM syndrome differentiation. There is debate regarding the effects of geographic factors on the active compounds of herbs. However, no details or concerns in this regard were mentioned in any of the studies. Although the herbal ingredients of these formulas exhibit pharmacologic and Chinese medicine actions relevant to the treatment of AD, the pharmacokinetics and pharmacodynamics of formulas may differ from that of individual herbs.32 In addition, in vitro data may not reflect treatment effects in humans.32 The absence of in vitro and in vivo pharmacologic data of a CHM formula deters the full understanding of the drug mechanism of an herbal formula. In order to provide strong evidence of the efficacy and safety of CHM treatment for AD, aside from rigorous RCTs, further laboratory studies on the pharmacokinetics and pharmacodynamics of each formula should be carried out. There are insufcient data to show that CHM treatment in combination with WM is more effective than WM treatment alone. The meta-analysis showed signicant improvement in symptom severity by CHM compared to placebo. CHM was reported as well-tolerated in all the studies and there were no reports of severe adverse events. However, the poor quality and heterogeneity of studies do not allow for valid conclusions. More rigorous RCTs and research on pharmacologic studies of CHM formulas are needed for stronger evidence regarding the efcacy and safety of CHM treatment for AD outside of clinical trial settings.
We would like to thank Iris Wenyu Zhou, PhD, for her assistance in data searching and evaluation.
REFERENCES 1. Bieber T. Atopic dermatitis. N Engl J Med 2008;358:1483-94. 2. Chang C, Keen CL, Gershwin ME. Treatment of eczema. Clin Rev Allergy Immunol 2007;33:204-25.

3. Feingold S, Huang C, Kristal L, Kalish R, Clark RAF. Eczemas. Curr Prob Dermatol 1998;10:41-90. 4. Hoare C, Li Wan Po A, Williams H. Systematic review of treatments for atopic eczema. Health Technol Assess 2000;4: 1-191. 5. Herd RM, Tidman MJ, Prescott RJ, Hunter JAA. The cost of atopic eczema. Br J Dermatol 1996;135:20-3. 6. Mancini AJ, Kaulback K, Chamlin SL. The socioeconomic impact of atopic dermatitis in the United States: a systematic review. Pediatr Dermatol 2008;25:1-6. 7. Su JC, Kemp AS, Varigos GA, Nolan TM. Atopic eczema: its impact on the family and financial cost. Arch Dis Child 1997;76: 159-62. 8. Zhang W, Leonard T, Bath-Hextall Fiona J, Chambers C, Lee C, Humphreys R, et al. Chinese herbal medicine for atopic eczema. Cochrane Database Syst Rev 2004;4:CD002291. 9. Buys LM. Treatment options for atopic dermatitis. Am Fam Physician 2007;75:523-8. 10. Kobayashi H, Ishii M, Takeuchi S, Tanaka Y, Shintani T, Yamatodani A, et al. Efficacy and safety of a traditional herbal medicine, hochu-ekki-to in the long-term management of Kikyo (Delicate Constitution) patients with atopic dermatitis: a 6-month, multicenter, double-blind, randomized, placebo-controlled study. Evid Based Complement Alternat Med 2010;7:367-73. 11. Hsu CJ, Wang LF. Emerging treatment of atopic dermatitis. Clin Rev Allergy Immunol 2007;33:199-203. 12. Bedi MK, Shenefelt PD. Herbal therapy in dermatology. Arch Dermatol 2002;138:232-42. 13. Armstrong NC, Ernst E. The treatment of eczema with Chinese herbs: a systematic review of randomized clinical trials. Br J Clin Pharmacol 1999;48:262-4. 14. Kayne SB. Traditional medicine: a global perspective. London and Chicago: Pharmaceutical Press; 2009. 15. Daukes SH. A consideration of the various methods of drug administration. Postgrad Med J 1929;5:1-7. 16. Higgins JPT, Green S. Cochrane handbook for systematic reviews of interventions version 5.1.0 (updated March 2011). London: The Cochrane Collaboration; 2011. 17. Huang YQ, Chen DC, Mo XM. Clinical observation of the treatment of spleen deficiency childhood atopic dermatitis with Jian Pi Shen Shi granules. Shan Xi Zhong Yi 2004;5: 396-8. 18. Cheng HM, Chiang LC, Jan YM, Chen GW, Li TC. The efficacy and safety of a Chinese herbal product (Xiao-Feng-San) for the treatment of refractory atopic dermatitis: a randomized, double-blind, placebo-controlled trial. Int Arch Allergy Immunol 2011;155:141-8. 19. Fung AYP, Look PCN, Chong LY, But PPH, Wong E. A controlled trial of traditional Chinese herbal medicine in Chinese patients with recalcitrant atopic dermatitis. Int J Dermatol 1999;38: 387-92. 20. Sheehan MP, Atherton DJ. A controlled trial of traditional Chinese medicinal plants in widespread non-exudative atopic eczema. Br J Dermatol 1992;126:179-84. 21. Sheehan MP, Rustin MHA, Atherton DJ, Buckley C, Harris DJ, Brostoff J, et al. Efficacy of traditional Chinese herbal therapy in adult atopic dermatitis. Lancet 1992;340:13-7. 22. Hon KLE, Leung TF, Ng PC, Lam MCA, Kam WYC, Wong KY, et al. Efficacy and tolerability of a Chinese herbal medicine concoction for treatment of atopic dermatitis: a randomized, double-blind, placebo-controlled study. Br J Dermatol 2007; 157:357-63. 23. Newell DJ. Intention-to-treat analysis: implications for quantitative and qualitative research. Int J Epidemiol 1992;21: 837-41.

10 Tan et al

J AM ACAD DERMATOL

24. Montori VM, Guyatt GH. Intention-to-treat principle. CMAJ 2001;165:1339-41. 25. Scheid V, Bensky D, Ellis A, Barolet R. Chinese herbal medicine formulas and strategies. Seattle (WA): Eastland Press; 2009. 26. Yuan CS, Bieber EJ, Brent AB. Traditional Chinese medicine. New York: Informa Healthcare; 2011. 27. Leung TF, Wong KY, Wong CK, Fung KP, Lam CWK, Fok TF, et al. In vitro and clinical immunomodulatory effects of a novel Pentaherbs concoction for atopic dermatitis. Br J Dermatol 2008;158:1216-23. 28. Giboney PT. Mildly elevated liver transaminase levels in the asymptomatic patient. Am Fam Physician 2005;71:1105-10. 29. Luo H, Li Q, Flower A, Lewith G, Liu J. Comparison of effectiveness and safety between granules and decoction

of Chinese herbal medicine: a systematic review of randomized clinical trials. J Ethnopharmacol 2012;140: 555-67. 30. Li SL, Song JZ, Qiao CF, Zhou Y, Xu HX. UPLC-PDA-TOFMS based chemical profiling approach to rapidly evaluate chemical consistency between traditional and dispensing granule decoctions of traditional medicine combinatorial formulae. J Pharm Biomed Anal 2010;52:468-78. 31. Chen D, Lu C. The clinical practice of Chinese medicine: eczema and atopic dermatitis. Beijing: Peoples Medical Publishing House; 2007. 32. Yuan R, Lin Y. Traditional Chinese medicine: an approach to scientific proof and clinical validation. Pharmacol Ther 2000; 86:191-8.