CD003263 Standard

Interventions for vitiligo (Review)

Whitton ME, Pinart M, Batchelor J, Lushey C, Leonardi-Bee J, Gonzlez U

This is a reprint of a Cochrane review, prepared and maintained by The Cochrane Collaboration and published in The Cochrane Library2010, Issue 1

http://www.thecochranelibrary.com

Interventions for vitiligo (Review)

Copyright 2011 The Cochrane Collaboration. Published by John Wiley & Sons, Ltd.

T A B L E O F C O N T E N T S

1HEADER . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1ABSTRACT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2PLAIN LANGUAGE SUMMARY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2BACKGROUND . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6OBJECTIVES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

6METHODS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

9RESULTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Figure 1. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 11

Figure 2. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

25DISCUSSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

28AUTHORS CONCLUSIONS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30ACKNOWLEDGEMENTS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

30REFERENCES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

36CHARACTERISTICS OF STUDIES . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

113DATA AND ANALYSES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

122ADDITIONAL TABLES . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

123WHATS NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

123HISTORY . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

124CONTRIBUTIONS OF AUTHORS . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

125DECLARATIONS OF INTEREST . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

125SOURCES OF SUPPORT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

125DIFFERENCES BETWEEN PROTOCOL AND REVIEW . . . . . . . . . . . . . . . . . . . . .

125INDEX TERMS . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

iInterventions for vitiligo (Review)


[Intervention Review]

Interventions for vitiligo

Maxine E Whitton1, Mariona Pinart2, Jonathan Batchelor3, Clare Lushey4, Jo Leonardi-Bee5, Urb Gonzlez2

1c/o Cochrane Skin Group, The University of Nottingham, Nottingham, UK. 2Department of Dermatology, Research Unit for

Evidence-based Dermatology, Hospital Plat, Barcelona, Spain. 3Department of Dermatology, Addenbrookes Hospital, Cambridge,

UK. 4Centre of Evidence Based Dermatology, The University of Nottingham, Nottingham, UK. 5Division of Epidemiology and Public

Health, The University of Nottingham, Nottingham, UK

Contact address: Maxine EWhitton, c/o Cochrane Skin Group, The University of Nottingham, Room A103, Kings Meadow Campus,

Lenton Lane, Nottingham, NG7 2NR, UK. [email protected].

Editorial group: Cochrane Skin Group.

Publication status and date: Edited (no change to conclusions), published in Issue 12, 2011.

Review content assessed as up-to-date: 3 November 2009.

Citation: Whitton ME, Pinart M, Batchelor J, Lushey C, Leonardi-Bee J, Gonzlez U. Interventions for vitiligo. Cochrane Databaseof Systematic Reviews 2010, Issue 1. Art. No.: CD003263. DOI: 10.1002/14651858.CD003263.pub4.


A B S T R A C T

Background

Around one per cent of the worlds population has vitiligo, a disease which causes white patches on the skin. There are a variety of

treatments available, most of which are unsatisfactory.

Objectives

To assess all interventions used to manage vitiligo.

Search methods

In November 2009 we updated searches of the Cochrane Skin Group Specialised Register, the Cochrane Central Register of Controlled

Trials in The Cochrane Library (Issue 4, 2009), MEDLINE, EMBASE, AMED, PsycINFO, LILACS and ongoing trials databases.

Selection criteria

Randomised controlled trials (RCTs).

Data collection and analysis

At least 2 review authors independently assessed study eligibility and methodological quality, and carried out data extraction. Two of

the 57 included studies could be combined for meta-analysis.

Main results

In this update, 57 trials, including 19 from the original review, were assessed with 3139 participants. Most of the RCTs, which covered a

wide range of interventions, had fewer than 50 participants. All of the studies assessed repigmentation, 6 measured cessation of spread,

and 5 investigated the effect of treatment on quality of life.

Most of the studies assessed combination therapies which generally reported better results. New interventions include monochromatic

excimer light (MEL), Polypodium leucotomos, melanocyte transplantation, oral antioxidants, Chinese zengse pill, and pimecrolimus. Weanalysed the data from 28 studies that met our outcome criteria of improvement in quality of life and greater than 75% repigmentation.

1Interventions for vitiligo (Review)


Fifteen analyses from studies comparing various interventions showed a statistically significant difference between the proportions of

participants achieving more than 75% repigmentation. The majority of analyses showing statistically significant differences were from

studies that assessed combination interventions which generally included some form of light treatment.

Topical preparations, in particular corticosteroids, reported most adverse effects. However, in the combination studies it was difficult

to ascertain which treatment caused these effects. None of the studies was able to demonstrate long-term benefits. Very few studies

were conducted on children or included segmental vitiligo. We found one study of psychological interventions and none evaluating

micropigmentation, depigmentation, or cosmetic camouflage.

Authors conclusions

This review has found some evidence from individual studies to support existing therapies for vitiligo, but the usefulness of the findings

is limited by the different designs and outcome measurements and lack of quality of life measures. There is a need for follow-up studies

to assess permanence of repigmentation as well as high quality randomised trials using standardised measures and which also address

quality of life.

P L A I N L A N G U A G E S U M M A R Y

Treatments for vitiligo (a disease causing loss of skin colour in patches)

Vitiligo is a chronic, unpredictable disease causing a loss of skin colour in patches. People of all ages and from all ethnic backgrounds

can develop the disease. Vitiligo does not cause physical symptoms but because of its unsightly appearance, particularly on dark or

tanned skin it can have considerable impact both psychologically and socially. Many ways of restoring normal colour to the skin have

been tried but improvement is usually short-lived. The causes of vitiligo are not yet clearly understood, so many treatments have been

developed on the basis of limited scientific evidence. There is no cure, and no way of limiting the spread of the disease has so far been

found.

We found twice as many trials (38) as we did in the original review (19) making a total of 57 studies, with 3139 participants overall.

Most of the studies had fewer than 50 participants and few lasted longer than 6 months.

The studies, which covered a wide range of treatments, also used many different ways of assessing the degree of repigmentation. Only

28 of the trials reported quality of life and greater than 75% repigmentation. There were many different designs: some studies looked

at individual patches whereas others compared participants. Most of the trials assessed combination treatments using ultraviolet light

to enhance repigmentation. In general, combination studies reported better results.

None of the trials reported long-term benefit (i.e. sustained repigmentation lasting at least two years). Results from this review should

therefore be treated with caution.

Some studies described adverse effects, in particular those using topical corticosteroids, but in the combination studies it was difficult

to ascertain which treatment caused these effects. There is a great need for an extensive and well planned programme of research to

establish the causes of vitiligo and to find effective ways to manage this disease.

B A C K G R O U N D

Description of the condition

Definition

Vitiligo is a disease in which melanocytes (pigment-producing

cells) are damaged or destroyed resulting in a patchy loss of pig-

ment from areas of skin. It is sometimes referred to as leucoderma

(leuco: white; derma: skin). Vitiligo patches can appear anywhere

on the skin but commonly affected sites include the area around



the orifices, the genitals, or any sun-exposed areas such as the face

and hands. The hair and, rarely, the eyes may also be affected.

The most common form of vitiligo, generalised vitiligo or vitiligo

vulgaris, is symmetrical and may be localised to certain areas but

may also spread to involve the entire body surface. In contrast,

segmental vitiligo only affects one side of the body and usually has

limited progression (Taeb 2007).

Although vitiligo may develop at any age, it is rarely present at

birth when itmay be confusedwith piebaldism (inherited localised

skin pigment loss). Most people develop the disease before the

age of 20 years (Lerner 1971). It is difficult to get a true picture

of the prevalence of vitiligo. The largest epidemiological study of

the prevalence of vitiligo is based on 47,033 inhabitants of the

island of Bornholm in Denmark (Howitz 1977) where vitiligo

affected 0.38% of the population. Although many papers quote

an estimate of 0.5% to 1% prevalence worldwide this estimate

may vary according to cultural and social differences. In countries

where there is more stigma attached to the disease for cultural and

social reasons or because it is more visible because of dark skin

colour, more people with the disease are likely to consult a doctor

than in other countries where this is not the case, thus reported

estimates of prevalence may be high. Figures as high as 8.8% have

been reported in India where stigma associated with the disease is

high (Behl 1972).

Impact

Though neither life-threatening nor symptomatic (except that de-

pigmented patches burn easily when exposed to the sun) vitiligo

can be cosmetically and psychologically devastating (Lerner 1978)

resulting in a lower self-esteem (Papadopoulos 1999), poor body

image (Porter 1979), and difficulties in sexual relationships (Porter

1990).Many people are frightened because they do not knowwhat

is causing it and some may also experience high levels of anxiety

because of the unpredictability of the disease. As a result of the

visible nature of the disease there is a stigma attached to vitiligo

which can make it difficult for those affected to cope with it from

day to day (Schmid-Ott 2007). For people with dark complexions

vitiligo is much more noticeable. Parsads review of the quality of

life of people in India with vitiligo describes their experience of

social isolation and stigmatisation (Parsad 2003a). Some people

with vitiligo experience discrimination in employment, particu-

larly in jobs where they have to deal with the public (Porter 1987).

The effect of vitiligo on the quality of life of those who suffer from

it is often overlooked (Kent 1996) although more recent studies

have started to address this (Ongenae 2005a).

For many people with vitiligo, the disease is episodic, having peri-

ods when it does not change and sometimes periods of aggressive

spreading. The disease can stop spreading spontaneously without

any treatment and may even improve spontaneously, at least par-

tially, for a short time. Loss of pigmentation normally returns,

even after successful treatment in some areas. However, many peo-

ple with the disease would like to at least have stable disease in

the absence of a cure. If this could be achieved they do not have

to keep adjusting to a constantly changing appearance. This is

an aspect of vitiligo which is so distressing. The fear and anxiety

which the unpredictability engenders can be quite great and have

a marked effect on the lives of people with vitiligo. From the clin-

ical perspective, if the disease is stable it is likely to respond better

to treatment, in particular surgical interventions. There are many

anecdotal accounts of people with vitiligo having lesions improve

while new patches are appearing at the same time.

Causes

The cause of this condition is unclear but seems to be dependent

on the interaction of biochemical, genetic, and environmental fac-

tors. There are several hypotheses for the cause of vitiligo, none of

which fully explain the disease. Several factors, including autoim-

mune, biochemical (including oxidative stress), genetic, neuronal,

and environmental, may interact to contribute to its development

(Schallreuter 2008; Spritz 2007;Westerhof 2007). Although there

are few epidemiological studies of vitiligo, it is believed that about

a third of people with vitiligo report close familymembers affected

by vitiligo (Bhatia 1992), suggesting that genetic factors play an

important role in the development of the disease, and this is sup-

ported by several studies (Fain 2003; Spritz 2004). In particu-

lar, the gene NALP-1 predisposes people to vitiligo as well as to

various autoimmune diseases (Jin 2007). However, it also seems

that triggers such as trauma to the skin, hormonal changes and

psychological distress may need to be present for the disease to

become apparent. Some studies support the theory that stress is

implicated in the onset and exacerbation of skin disease in some

individuals (AlAbadie 1994). In one particular study, people with

vitiligo experienced a significantly higher number of stressful life

events than controls, suggesting that stress may be an important

factor in the cause of the disease (Papadopoulos 1998).

Association with other diseases

Vitiligo is sometimes associated with autoimmune diseases such

as pernicious anaemia, thyroid disorders, diabetes mellitus, and

Addisons disease (Rezaei 2007). People with vitiligo have been

found to have antibodies directed against melanocytes and some

develop antibodies to other tissues and either suffer from the above

mentioned autoimmune diseases themselves or have close relatives

that do (Bystryn 1988, Cui 1995, Klaus 1984). These observations

have been used to support the popular hypothesis, which has been

strengthened by recent genetic studies (Jin 2007) that vitiligo is

an autoimmune disease.

Description of the intervention



It is important to stress that as yet there is no cure nor any ef-

fective method of stopping the spread of this disease. Several in-

terventions have been used in the treatment of vitiligo includ-

ing pharmacological interventions (e.g. topical corticosteroids and

immunomodulators); various forms of phototherapy (i.e. ultra-

violet A (UVA), narrow- and broadband ultraviolet B (UVB),

psoralen and UVA (PUVA), excimer laser, and monochromatic

excimer light (MEL)); surgical procedures (grafting, melanocyte

transplantation, micropigmentation); cosmetic measures (depig-

mentation, cosmetic camouflage, fake tan); complementary ther-

apies, and psychotherapy. Many published studies describe com-

bination therapies, usually combining a light source with another

formof treatment in order to enhance repigmentation. This review

assesses all interventions for vitiligo which are described below.

(1) Topical Therapies

1.1 Topical corticosteroids

The theory that vitiligo is an autoimmune disease led to the use of

corticosteroids in its treatment. Topical corticosteroids may arrest

the attack on pigment cells by the immune system and can en-

courage some repigmentation, particularly in dark skinned peo-

ple (Kumari 1984). Intralesional corticosteroids (Goldstein 1992),

orally administered corticosteroids (Kim 1999), and recently, in-

travenous pulsed steroids (Seiter 2000), have also been tried, all

of which can have serious side-effects. The combination of cor-

ticosteroid creams with UVA light is reported to be more effec-

tive than topical corticosteroids alone (Westerhof 1999). Side-ef-

fects of topical corticosteroids include atrophy, telangiectasia, and

striae.

1.2 Intralesional steroids

This intervention is not now used in practice but we found one

old study (Vasistha 1979) in the original review which did not

examine any outcomes of interest for this review update.

1.3 Vitamin D analogues (Calcipotriol and Tacalcitol)

Topical vitamin D analogues such as calcipotriol and tacalcitol

are commonly used to treat other skin diseases such as psoriasis.

When combined with light some investigators have reported that

vitamin D analogues with phototherapy have been beneficial in

treating vitiligo (Ermis 2001; Lu-Yan 2006).

1.4 Calcineurin Inhibitors (Tacrolimus and Pimecrolimus)

Calcineurin inhibitors have an immunomodulatory effect and are

primarily used in the treatment of eczema. They have been sug-

gested to be of some therapeutic value in treating vitiligo, espe-

cially in children or if the disease is of recent origin and affects

the face (Lepe 2003). They are applied to the skin as ointments or

creams and are reputed to have a better safety profile than that of

corticosteroids. Calcineurin inhibitors have been used in combi-

nation with excimer laser (Kawalek 2004) and also with narrow-

band UVB (Mehrabi 2006) to treat vitiligo.

1.5 Khellin

Khellin is a topical preparation whose chemical structure closely

resembles that of psoralen but which appears to be less phototoxic.

It has been used to treat vitiligo in conjunction with sunlight or

UVA (KUVA) (Cestari 2001; Procaccini 1995).

1.6 Pseudocatalase (PC-KUS) and catalase/dismutase

superoxide

This is an experimental treatment based on the discovery that peo-

ple with vitiligo have high levels of hydrogen peroxide (H2O2)

in both affected and unaffected skin, and low levels of the en-

zyme catalase, which inactivates H2O2. This replacement ther-

apy uses an analogue of normal human catalase (called pseudo-

catalase). It is applied to the skin in a moisturising cream base

in conjunction with narrow band UVB to stimulate melanocyte

activity (Schallreuter 2000; Schallreuter 2002). A study reporting

on the use of plant-based topical catalase/dismutase superoxide

(Sanclemente 2008) has also been included in this review.

1.7 Melagenina (human placental extract)

This intervention, derived from human placental extract, is used

topically in conjunction with light, including sunlight and infra-

red light (Souto 1997) to treat vitiligo.

(2) Oral Therapies

These include oral PUVA which is mentioned under light thera-

pies. Polypodium leucotomos, a type of fern of which the extract hasbeen shown to have photoprotective (Caccialanza 2007) and im-

munomodulatory properties in both animal models and humans.

It has been used in conjunction with phototherapy for treating

vitiligo (Middelkamp-Hup 2007; Reyes 2006). Oral levamisole,

an antihelminthic developed to treat intestinal worms also has

immunostimulating properties. It has been tried (Agarwal 2005)

in conjunction with a topical corticosteroid. Oral Ginkgo biloba,a herb with immunomodulatory and anti-oxidant properties has

been used in an attempt to halt the progression of the disease

(Parsad 2003b). Oral minipulses of betamethasone (Rath 2008)

and oral azathioprine (Radmanesh 2006), have both been used

in conjunction with phototherapy to repigment the skin. Some

studies use oral antioxidants (DellAnna 2007), vitamin B12 and

folic acid (Tjioe 2002), and L-phenylalanine (Siddiqui 1994), in

combination with various forms of light therapy to treat vitiligo.



(3) Light therapies

3.1 - 3.4 PUVA and PUVAsol (oral and topical)

PUVA (psoralen and ultraviolet A light) was developed from a tra-

ditional Egyptian treatment using giant hogweed (a photosensi-

tising plant) and sunlight. The most commonly used preparation

is 8-methoxypsoralen (8-MOP) but trimethylpsoralen (TMP) is

also used (Bhatnagar 2007). Many studies use a combination of

light with other interventions, for example surgical techniques

(Van Geel 2004), in order to enhance the speed or degree of repig-

mentation. Modern psoralens developed in the laboratory can be

taken orally, added to bath water (mainly used in the treatment

of psoriasis), or applied to the skin, which is then exposed to ul-

traviolet A light (UVA) (Njoo 1998). In hot countries sunlight is

commonly used instead of UVA: the term used for this treatment

is PUVAsol (Parsad 1998). PUVA is claimed to be particularly

effective in darker skinned individuals (Grimes 1997).

3.5 UVA

UVA has been used as monotherapy (El Mofty 2006; Westerhof

1999) or in conjunction with other interventions (Siddiqui 1994).

3.6 UVB

BB-UVB and NB-UVB

Both broad band (BB-UVB) and narrow band UVB (NB-UVB)

have been used to treat vitiligo, though only a few studies evaluate

BB-UVB (Asawanonda 2008). There has been a recent increase

in the number of published studies on the use of NB-UVB (

Bhatnagar 2007; Yones 2007). An open controlled study by Njoo

suggests that narrow band UVB monotherapy is also safe and

effective for children with vitiligo (Njoo 2000b).

3.7 Lasers

Excimer laser

The 308 nm excimer laser delivers a condensed beam of UVB

light to the skin. The beam can be targeted but its use is limited

to small areas. It is most commonly used in combination with

topical therapies such as tacrolimus (Kawalek 2004), calcipotriol

(Goldinger 2007), or tacalcitol (Lu-Yan 2006).

Helium Neon Laser

HeliumNeon Laser, a gas laser which operates in the red spectrum

at 632.8 nm, is a recent intervention for treating vitiligo, used

as monotherapy but also in combination with tacrolimus, which

looks promising, particularly as it is reported to be effective for

segmental vitiligo which can be difficult to treat by conventional

methods (Wu 2008). There are as yet no published RCTs of this

intervention.

3.8 Other forms of light therapy

Light is an important factor in the proliferation of melanocytes.

It is therefore not surprising that other forms of light, including

sunlight (Rodriguez-Martin 2009) and the special light of the

Dead Sea (climatotherapy) (Schallreuter 2002), have been used to

treat this disease.

Monochromatic excimer light (MEL)

Monochromatic excimer light (MEL) is another form of targeted

UVB light without the laser element therefore it can be used to

treat larger areas of vitiligo (Casacci 2007).

(4) Surgical Interventions

4.1 - 4.2 Suction blister grafts, punch grafts, minigrafts, split

skin grafts

Various types of grafting have been used to treat vitiligo, using

the persons own normal skin (autologous). These include punch

grafting (Khandpur 2005) minigrafting (Navarro 2002), suction

blister grafting (Ozdemir 2002), and split skin grafts (Khandpur

2005). Punch grafting involves taking 1 to 2 mm diameter thick-

ness circular grafts from normal areas of skin and transplanting

them to depigmented areas fromwhich similar-sized circles of skin

have been removed. The suction blister technique involves cre-

ation of blisters on normal skin, followed by transfer of the roof

of the blisters to dermabraded vitiliginous skin. Grafting appears

to work best on stable vitiligo and on segmental forms (Boersma

1995).

4.3 Micropigmentation

Micropigmentation, a formof semi-permanent tattooing, involves

the implantation of minute metabolically inert pigment granules

(e.g. iron oxide, titanium dioxide) into the upper dermis using tat-

tooing machines (Garg 2005). Micropigmentation has been used

on areas resistant to treatment such as the lips and the tips of the

fingers. This technique is best suited to stable vitiligo because of

the risk of triggering another patch of vitiligo at sites of damage,



known as the Koebner phenomenon (Halder 1989). Some pig-

ment may migrate to regional lymph nodes, resulting in fading of

the original tattoo.

4.4 Melanocyte transplantation

Melanocyte transplantation involves the culture of pigment cells

(melanocytes) obtained from the persons unaffected skin and in-

jecting them into blisters created on the depigmented areas or di-

rectly into dermabraded skin (Van Geel 2004). It is also possible

to transplant melanocytes and keratinocytes grown in the same

culture medium (Gokhale 1991).

(5) Depigmentation

When a large proportion of the total body area is affected, depig-

mentation can be achieved using hydroquinone or theQ-switched

ruby laser with topical 4-methoxylphenol to destroy remaining

melanocytes (Njoo 2000c). However, depigmentation therapy re-

quires careful counselling of the participant as it leaves the skin

without protection from sunlight and depigmentation is largely

irreversible. People from dark skinned races may be traumatised

by their complete loss of colour as it may be perceived as a loss of

identity.

(6) Cosmetic camouflage and fake tanning

preparations

Cosmetic camouflage creams can be prescribed, especially when

the face, neck, and hands are affected (Ongenae 2005b). The

creams can give excellent results if applied properly and normally

remain waterproof for up to eight hours. However, there are a

number of problems with cosmetic camouflage: it may be difficult

to find a suitable colour match; the creams tend to rub off on areas

where there is friction such as the cuffs and neckline; most men

are reluctant to apply them; they are not suitable for very young

children, and they have to be applied daily.

Fake tanning preparations containing dihydroxyacetone (DHA)

(Rajatanavin 2008) are suitable for some paler skin types and can

provide a cosmetically acceptable colour for three to four days after

which time the top layer of skin is normally shed. They do not

rub off or wash off although they fade gradually with time. They

are helpful for large exposed areas such as the arms and legs but

can sometimes produce an orange tinge and are more useful in the

summer when normal skin tans. They do not offer sun protection

so people need to be advised to use high factor sunscreens on

affected, exposed areas.

(7) Psychological therapy

Although vitiligo does not affect a persons general health, the

psychosocial effects can be devastating (Ongenae 2005a; Schmid-

Ott 2007). A small non-randomised study suggests that people

with vitiligo can benefit from cognitive behavioural therapy in

terms of coping with the disease from day to day (Papadopoulos

1999).

(8) Complementary therapies

When there is no improvement with conventional therapies

many people with vitiligo may turn to complementary treatments

(Orecchia 2000). These include a human placental extract (mela-

genina) used topically, vitamin supplements including B12, vi-

tamin C, and folic acid (Montes 2006), and Indian or Chinese

herbal medications (Shi 2008). Given the theory that oxidative

stress is involved in the pathogenesis of vitiligo, antioxidants such

asGinkgo biloba extract (Parsad 2003b) and combinations of otherantioxidants have been used in the treatment of vitiligo, some-

times in conjunction with other treatment modalities (DellAnna

2007). Some of these studies have already been mentioned in this

section under oral or topical preparations.

It is worth mentioning that herbal remedies may not have been

subjected to toxicological tests and often the constituents of the

medication are unknown. In a recent systematic review of natural

remedies for vitiligo the authors conclude that published studies

are poor and reporting is flawed but that the use of Ginkgo bilobawarrants further investigation (Szczurko 2008). It is also worth

noting that piperine, an extract from a common herb, black pep-

per, shows evidence of a proliferative effect on mouse melanocytes

(Faas 2008).

Why it is important to do this review

When this systematic review was originally published, there were

few RCTs and no systematic review covering all available interven-

tions for vitiligo. Since the date of the last literature search, new in-

terventions have been trialled and twice as many RCTs published.

Although we found other reviews and one systematic review of

natural products for vitiligo there is still no systematic review of

all available interventions for vitiligo.

This update is important to assess new interventions, to highlight

the gaps in research and the need for better designed and powered

studies, thus informing clinical decisions and future research pri-

orities.

O B J E C T I V E S

To assess the effects of all therapeutic interventions used in the

management of vitiligo.

M E T H O D S



Criteria for considering studies for this review

Types of studies

Randomised controlled trials.

Types of participants

People of all age groups who have any type of vitiligo.

Types of interventions

All types of interventions used in the management of vitiligo in-

cluding topical and oral preparations, various forms of light ther-

apy, surgical techniques, psychological therapy, and unconven-

tional or complementary therapies.

Types of outcome measures

Primary outcomes

(a) Quality of life measured using a validated tool e.g. DLQI (Der-

matology Quality of Life Index), CDLQI (Childrens Dermatol-

ogy Quality of Life Index), Skindex-29.

(b) Percentage of repigmentation (restoration of normal skin

colour) of vitiliginous skin: success rate in terms of more than 75%

repigmentation of individual patches or of total body surface area,

measured by objective means (e.g. photographs, rule of nines).

Secondary outcomes

(c) Cessation of spread of vitiligo or stabilisation of the disease

defined as:

i)No increase in the size of individual vitiligo patchesmeasured ob-

jectively withWoods light, photography, or other objective means

within a period of a) less than one year or b) one year or more

ii) No new lesions appearing, despite no improvement in existing

patches resulting from treatment, within a period of a) less than

one year or b) one year or more

(d) Long-termpermanence of repigmentation resulting from treat-

ment (at least two years follow-up)

(e) Adverse effects

Search methods for identification of studies

Electronic searches

We updated searches in the Cochrane Skin Group

Specialised Register (to 3rd November 2009) using the search

terms: vitiligo OR leucoderma OR leukoderma.

We searched the Cochrane Central Register of Controlled

Trials (CENTRAL) in The Cochrane Library (Issue 4, 2009 )using the search strategy in Appendix 1.

We updated searches of MEDLINE (to 3rd November

2009) using the search strategy in Appendix 2.

We updated searches of EMBASE (to 3rd November 2009)

using the search strategy in Appendix 3.

We updated searches of PsycINFO (to 3rd November

2009) using the search strategy in Appendix 4.

We searched CINAHL (Cumulative Index to Nursing and

Allied Health Literature) (1982 to January 2009) using the

strategy in Appendix 5.

We updated searches of LILACS (Latin American and

Caribbean Health Science Information database) (to 3rd

November 2009) using the strategy in Appendix 6.

We updated searches of AMED (the Allied and

Complementary Medicine database) (to 3rd November 2009)

using the strategy in Appendix 7.

Ongoing Trials

We searched the following Ongoing Trials databases on 3rd

November 2009 using the term vitiligo and its synonyms, leuco-

derma or leukoderma.

The metaRegister of Controlled Trials www.controlled-

trials.com

The U.S. National Institutes of Health ongoing trials

register www.clinicaltrials.gov

The Australian and New Zealand Clinical Trials Registry

www.anzctr.org.au

The World Health Organization International Clinical

Trials Registry platform www.who.int/trialsearch

The Ongoing Skin Trials register on

www.nottingham.ac.uk/ongoingskintrials

Searching other resources

References from published studies and reviews

We checked the bibliographies of the included and excluded stud-

ies and reviews to identify further trials.

Unpublished literature

We found some studies as a result of the searches of the Ongo-

ing Trials Registers which are completed but are not yet pub-

lished. These are listed in the Characteristics of studies awaiting

classification.



Conference proceedings

We searched ISI Web of Knowledge from 2004 to February 2009

using search terms vitiligo and its synonyms, leucoderma, and

leukoderma.

Languages

For all search methods, we included studies in other languages if

they met the criteria and could be translated into English.

Data collection and analysis

Selection of studies

Two of the authors (MW, JB) checked the titles and abstracts

identified in the search and independently assessed the full text

of all studies of possible relevance. Two of three of the authors

(MW, JB, JLB) decided which trials met the inclusion criteria. Any

disagreements were resolved by discussion and a decision made by

consensus.

Data extraction and management

MP extracted data from all the new studies which were also allo-

cated randomly to four of the other authors (JLB, JB, CL, MW)

who independently extracted data. One of three of the authors

checked the data extraction forms for discrepancies (MW, JLB,

UG), ensuring that they did not check their own forms. Where

discrepancies could not be solved by reference to the text of the

studies, differences were resolved by consensus.

Assessment of risk of bias in included studies

At least two authors (MW, JB, MP) independently assessed risk

of bias for the new studies identified in the updated search and

differences were resolved by consensus.

Wemade an assessment of the risk of bias which includes an evalu-

ation of the following components for each included study, using

the criteria outlined in the Cochrane Handbook for Systematic

Reviews of Interventions (Higgins 2008). Details are available in

the Characteristics of included studies tables (Risk of bias) for each

study.

(a) The method of generation of the randomisation sequence;

(b) The method of allocation concealment - it was considered

adequate if the assignment could not be foreseen;

(c) Who was blinded or not blinded (participants, clinicians, out-

come assessors); and

(d) How many participants dropped out of the study overall, and

whether participants were analysed in the groups to which they

were originally randomised (intention-to-treat).

In addition the quality assessment also included:

(e) Baseline assessment of the participants for age, sex, duration

and severity of vitiligo;

(f ) Aims, interventions (including drug doses and duration of

treatment) and outcome measures clearly and objectively defined;

and

(g) Whether or not the assessment of compliance was reported.

Measures of treatment effect

We expressed the results as risk ratio (RR) with 95% confidence

intervals (CI) for dichotomous outcomes, and difference in means

(MD) with 95% CI for continuous outcomes. Dichotomous out-

comes are presented as more than 75% of body surface since these

levels are regarded as clinically important levels of repigmentation.

Where data are presented as other dichotomous categories (that

is, not as 75%), we calculated RRs but data was not presented in

figures.

Unit of analysis issues

Where there weremultiple intervention groups within a trial, pair-

wise comparisons were made of similar active interventions versus

no treatment, placebo, or another active intervention. No cross-

over trials were included in the review. Internally controlled trials

were analysed using appropriate techniques for paired designs (for

example, for continuous outcomes using Wilcoxon Signed Ranks

test or paired t-test; or for dichotomous data using McNemars

test) where available and were not pooled with studies of other

designs. However, where paired data could not be extracted from

the papers, we presented the data using non-paired methods. Non-

randomised controlled studies were excluded from the review.

Dealing with missing data

If participant dropout led to missing data we conducted an inten-

tion-to-treat analysis. Trial authors of studies were contacted to

provide missing statistics such as standard deviations.

Assessment of heterogeneity

Statistical heterogeneity was assessed using I statistic. Data were

synthesised using meta-analysis techniques if I statistic was less

than 80%.

Assessment of reporting biases

Publicationbiaswas not tested since insufficient datawere available

for similar types of interventions.



Data synthesis

For studies with a similar type of active intervention, a meta-anal-

ysis was performed, to calculate a weighted treatment effect across

trials, using a random-effects (DerSimonian and Laird) model.

Where it was not possible to perform ameta-analysis, the datawere

summarised for each trial. We have presented the corresponding

results as P values. If raw data could not be extracted, we extracted

the results from appropriate statistical analyses presented in the

paper and reported these in the review. We considered a P value

< 0.05 as statistically significant. Excluded studies and reasons for

exclusion are described in the table Characteristics of excluded

studies but not discussed further. Data relating to adverse effects

are described qualitatively as reported in each trial.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were not conducted since no substantial het-

erogeneity (I statistic > 50%) existed between studies for the pri-

mary outcome.

Sensitivity analysis

Weplanned to conduct sensitivity analyses to examine the effects of

excluding poor quality studies, defined as those with amoderate or

high risk of bias as described in the Cochrane Handbook (Higgins

2008), however insufficient numbers of studies were included in

the meta-analysis.

R E S U L T S

Description of studies

See: Characteristics of included studies; Characteristics of

excluded studies; Characteristics of studies awaiting classification;

Characteristics of ongoing studies.

Results of the search

In the updated version of this review we retrieved 218 references.

Added to the 19 RCTs included in the past review we found

38 new RCTs from the updated search, making a total of 57

studies assessed for this review with 3139 participants overall.

We excluded 7 RCTs of which 4 correspond to RCTs retrieved

from the new search, which are presented in the Characteristics

of excluded studies table. We found 5 RCTs that are ongoing

trials and 13 that are awaiting assessment, which are listed in

Characteristics of ongoing studies and Characteristics of studies

awaiting classification tables, respectively. If they are published we

will include the ongoing RCTs in future updates of this review.

Included studies

Fifty-seven studiesmet the inclusion criteria for this review.Details

of these studies are provided in the table for Characteristics of

included studies. Summary details are provided below.

Designs

Of the 57 included RCTs, 18 were within-participant, left/

right comparison studies (Anbar 2008; Casacci 2007; Dawid

2006; Ermis 2001; Goldinger 2007; Hamzavi 2004; Kandil

1974; Kawalek 2004; Leone 2006; Lepe 2003; Lim-Ong 2005;

Lu-Yan 2006; Mehrabi 2006; Parsad 1998; Passeron 2004; ;

Sanclemente 2008; Van Geel 2004; Westerhof 1999), 1 was de-

scribed as a left/right comparison study (Procaccini 1995), 3

were within-participant studies (Asawonanda 2008; Hofer 2005;

Sharquie 2005) and the rest were parallel group studies. Ozdemir

2002 was an inter-technique comparison using randomised

grafts on non-symmetrical vitiliginous areas. Asawanonda 2008,

Hofer 2005, and Sharquie 2005 randomly assigned lesions, in

the same body area, to receive different treatments or treat-

ment regimens. All included studies allocated either partici-

pants or bilaterally symmetrical lesions to treatment groups

in a random manner. Placebo-controlled studies were under-

taken by 29 studies (Agarwal 2005; Dawid 2006; Ermis 2001;

Esfandiarpour 2008; Goldinger 2007; Hamzavi 2004; Kandil

1974; Kawalek 2004; Lim-Ong 2005; Lu-Yan 2006;Maldonado

1975; Mehrabi 2006; Middelkamp-Hup 2007; Navarro 2002;

Papadopoulos 2004; Parsad 1998; Parsad 2003b; Pathak 1984;

Procaccini 1995; Reyes 2006; Rodriguez-Martin 2009; Rojas-

Urdaneta 2007; Schallreuter 2002; Sharquie 2005; Siddiqui

1994; Souto 1997; Van Geel 2004; Vasistha 1979; Yones 2007)

and the rest used active controls.

Sample size

The number of participants evaluated in the studies varied from

8 to 596. The majority of studies (38) consisted of a small sam-

ple size (less than 50 participants). Eighteen studies contained a

medium sample size (between 51 to 150 participants) and only

one study (Pathak 1984) involved a large sample size (more than

150 participants).

Population

The type of vitiligo and the extent, distribution, and duration of

lesions varied between studies and within studies. However, the

majority of studies (40) included participants with symmetrical

vitiligo. None of the studies included only participants with seg-

mental vitiligo although some studies (6) included participants

with any type of vitiligo. The remaining studies did not provide

information on the type of vitiligo affecting their participants. Fur-

ther details are provided in the table for Characteristics of included

studies.



The majority of studies included male and female participants

(apart from Arca 2006, which included only male participants)

who were of any age except in the studies by Khalid 1995 (which

was confined to children under 12 years old), Ruiz-Maldonado

1975 (children aged 9 to 10 years old), and Lepe 2003 (children

aged 4 to 17 years old). See Characteristics of included studies for

further details.

Setting

The 57 included studies were undertaken in various continents.

The studies were most commonly carried out in Asia (26) fol-

lowed by Europe (19), with only a small number undertaken in the

Americas (10) and Africa (2). India conducted the largest number

of studies (11) compared to any other country. Two studies were

undertaken across 2 countries (Casacci 2007; DellAnna 2007).

See the Characteristics of included studies tables for further de-

tails.

Interventions

The studies were concerned with a wide range of interventions

including topical treatments, light therapies, oral treatments, sur-

gical methods, and psychological therapies.

The topical treatments assessed included steroids, calcipotriol,

immunomodulators (pimecrolimus and tacrolimus), tacalcitol,

khellin, antioxidant plus mitochondrial stimulating cream, mela-

genina (placental extract), Dead Sea climatotherapy plus pseudo-

catalase, fluorouracil, psoralen tincture, and lactic acid.

The light therapies commonly consisted of psoralen plus ultravi-

olet A light (PUVA), psoralen plus sunlight (PUVAsol), ultravio-

let A light (UVA), broadband ultraviolet B light (BB-UVB), nar-

rowband ultraviolet B light (NB-UVB, including excimer laser),

sunlight and monochromatic light.

The oral treatments investigated included vitamin B12 plus folic

acid, levamisole plus mometasone furoate, an antioxidant pool

(alpha lipoic acid, vitamins C and E, and polyunsaturated fatty

acids), antioxidants plus phenylalanine, ginkgo biloba, azathio-

prine, Chinese zengse pill, L-phenylalanine, and Polypodium leu-cotomos.Finally a small number of studies assessed the efficacy of surgical

methods, such as skin grafts and cell transplants. Only one study

(Papadopoulos 2004) assessed psychological therapies, comparing

cognitive behavioural therapy with person centred therapy.

See the Characteristics of included studies tables and the Effects

of Interventions section for further details.

Outcomes

All studies assessed outcomes as the presence of repigmentation.

No two studies used exactly the samemethod of scoring in relation

to repigmentation; however the majority scored the percentage

repigmentation and placed participants into categories (e.g. 1% to

25%, 26% to 50%, 51% to 75%, > 75%). Many authors made an

attempt to clarify the amount of repigmentation that was deemed

successful and marked complete or excellent, however opinions

varied considerably. For example some authors perceived more

than 50% as excellent, whereas others believed it should be higher

(e.g. more than 75% or more than 90%). An older study (Vasistha

1979) used excellent, good, fair, and no response to report levels of

repigmentation. Ruiz-Maldonado 1975 expressed repigmentation

rates as slight, marked, and clinically cured, and Souto 1997

used improvement, deterioration, and no change to describe

differences in repigmentation. Finally, two studies presented repig-

mentation in millimetres (Navarro 2002; Rojas-Urdaneta 2007).

A small number of studies measured cessation of spread of vi-

tiligo (Agarwal 2005; Barman 2004;Dawid 2006; Lim-Ong2005;

Parsad 2003b; Siddiqui 1994). Five studies investigated improve-

ment in quality of life (Agarwal 2005; Middelkamp-Hup 2007;

Papadopoulos 2004; Sassi 2008; Yones 2007). Middelkamp-Hup

2007 and Sassi 2008 assessed quality of life using Skindex-29,

whereas Papadopoulos 2004 and Yones 2007 used the Dermatol-

ogy Life Quality Index, and Agarwal 2005 used the Dermatology

Life Quality Index and the Childrens Dermatology Life Quality

Index.

Assessment was commonly made visually by the clinician us-

ing photographs taken of lesions. However, some studies used

planimetry (i.e. drawing lesions on transparent papers) to deter-

mine repigmentation and a small number took photographs of

lesions and subjected them to morphometry analysis using Corel

Draw, version 9.0.

The duration of the studies varied widely (three weeks to three

years) and was, in part, related to the outcomes to be observed, for

example initiation of repigmentation or extent of repigmentation.

Within these periods some studies continued treatment until an

optimal response was achieved. The longest study was Pathak

1984, which lasted up to three years, although the placebo group

was not followed up after nine to twelve months.

See the Characteristics of included studies tables and the Effects

of Interventions section for further details.

Excluded studies

We have excluded two RCTs (Babu 2008; Ghosh 1994) because

the authors recruited all consecutive participants with vitiligo, al-

though they claimed that these participants were randomly se-

lected. Moreover, they did not mention whether the participants

were randomly distributed into the study groups nor how. One

study (El Mofty 2006B) reported that participants were randomly

allocated to each study but treatment allocation was not ran-

domised within each study. One RCT (Rondon Lugo 1987) al-

though randomised had insufficient published data to assess the

study. Two other RCTs (El Mofty 2001; Godse 2008) stated that

they were randomised but the methodology revealed that they

were not. The last RCT (El Zawahry 1997) was excluded because



the randomisation method was not confirmed in correspondence

with the author.

Studies awaiting classification

We have found thirteen RCTs (Bakis-Petsoglou 2009; Dearman

2005; Elgoweini 2009; Farajzadeh 2009; Hui-Lan 2009; Lanigan

2004; Linthorst Homan 2006; Martin 2007; Nordal 2005;

Radakovic 2009; Seckin 2007; Stinco 2009; Syed 2006) that

are awaiting assessment (see Characteristics of studies awaiting

classification for more details).

Ongoing studies

We have found five registered ongoing RCTs (Alghamdi 2009;

Hofer 2008; Pandya 2008; Picardo 2007; Sachan 2009). See

Characteristics of ongoing studies for details.

Risk of bias in included studies

With respect to risk of bias in the included studies, we looked at the

following four possible sources of bias: generation of the randomi-

sation sequence; allocation concealment; blinding and losses to

follow-up. We did not look for evidence of selective reporting and

publication bias was also not formally assessed. See Characteristics

of included studies and the Risk of bias table for each study. See

also Figure 1 and Figure 2 which give a graphical summary of the

risk of bias components.

Figure 1. Methodological quality graph: review authors judgements about each methodological quality

item presented as percentages across all included studies.



Figure 2. Methodological quality summary: review authors judgements about each methodological quality

item for each included study.



Allocation

Sequence Generation and Allocation Concealment

The method of generation of the randomisation sequence was

deemed adequate in 32 of the studies. Adequate sequence gen-

eration by computer-generated random list or random number

sequence was described in 20 studies (Agarwal 2005; Akhyani

2005; Anbar 2008; Bhatnagar 2007; Casacci 2007; Cestari

2001; DellAnna 2007; Kumaran 2006; Leone 2006; Lepe 2003;

Lim-Ong 2005; Middelkamp-Hup 2007; Ozdemir 2002; Reyes

2006; Rodriguez-Martin 2009; Sanclemente 2008; Sassi 2008;

Shi 2008; Tegta 2006; Vasistha 1979), 2 of which used permuted

blocks (Lepe 2003;Rodriguez-Martin 2009)while Sassi 2008 used

stratified blocks. The old study by Vasistha was judged in the first

review to have adequate sequence generation following correspon-

dence with the author and this has not been changed despite the

very unequal numbers of participants in each group. Esfandiarpour

2009 used simple randomisation, Asawanonda 2008, Hofer 2005

and Radmanesh 2006 used cards, Hamzavi 2004 and Mehrabi

2006 used coin toss, Navarro 2002 used chips, Passeron 2004 ran-

domised by drawing lots, Van Geel 2004 used a type of lottery; and

Yones 2007 used a sequentially numbered list. For the Ermis 2001

study we have retained the judgement we made in the original

review, Rojas-Urdaneta 2007 used a list created using a random

number generator (calculator).

Of these 32 studies, the treatment allocation of each participant

was kept concealed in 11 studies, in 4 studies it was not kept con-

cealed, and in the remaining 17 studies it was not clear whether

or not the treatment allocation was kept concealed, and we were

unable to obtain further information regarding allocation conceal-

ment from the study authors. Methods of allocation concealment

included: using centralised telephone randomisation (Sassi 2008);

using opaque, sealed envelopes containing cards with treatment

allocations written on them (Radmanesh 2006; Rodriguez-Martin

2009); generationof the allocation sequence by a third party not as-

sociated with the study (Mehrabi 2006; Middelkamp-Hup 2007;

Rojas-Urdaneta 2007; Sanclemente 2008; Van Geel 2004) or hav-

ing a trial pharmacist responsible for preparing treatments and

who was the only person to know which active agent they had

dispensed, revealing this information only at the end of the study

(Cestari 2001; Yones 2007). Of note, some studies did not ran-

domise participants but instead randomised vitiligo lesions within

the same participant to different interventions (otherwise known

as a within-participant comparison).

In 25 of the included studies, the method of generation of the

randomisation sequence was unclear, even after obtaining further

information regarding generation of the randomisation sequence

from some of the authors. Of these 25 studies, none gave definite

information about concealment of the treatment allocation.

Blinding

Although some studies were described as double-blind or implied

double-blinding, not all of them provided details as to how blind-

ing was maintained. Many studies were within-participant com-

parisons of different interventions so it was not possible for the

participants (and sometimes the clinicians) to be blinded in these

studies. Blinding of participants was also not possible in some

other studies because two differentmodalities of intervention were

being assessed (e.g. oral versus topical interventions).

In seven studies, participants, clinicians and assessors were all

blinded. Eight studies were described as double-blind (or implied

double-blinding) but did not have separate blinded outcome as-

sessors. In 11 studies it was stated that the outcome assessor was

blinded but either the participants and/or the clinicians were not

blinded. Many of these studies were within-participant compar-

isons of different interventions. In fourteen studies blinding of

participants, clinicians or outcome assessors was not stated.

Further details are provided in the Risk of bias tables (see

Characteristics of included studies) where we have summarized

the adequacy of blinding with particular attention to those stud-

ies, twenty-eight in total, that addressed > 75% repigmentation,

a primary outcome of this review. We have grouped the studies

according to intervention types.Where combinations of interven-

tions were used (e.g. topical and light therapy together), we have

included the studies under the heading of what we consider to

be the main intervention. Where studies examined two or more

different modalities of intervention, we have mentioned the study

under each intervention heading.

Incomplete outcome data

The overall number of participants lost to follow-upwas 521/3139

i.e.16.6% of the total number of study participants included in

the review. In 23 of the included studies (40%), data were analysed

on an intention-to-treat basis. This was either because there were

no losses to follow-up (16 studies) or because data from dropouts

were included in an explicit intention-to-treat analysis. For 2 of the

studies (Farah 1967; Rath 2008) it was unclear whether or not an

intention-to-treat analysis had been performed. Only 7 of the 23

studies with intention-to-treat (ITT) analysis (30%) addressed the

primary outcome ofmore than 75% repigmentation of this review.

Twenty of the included studies which assessed repigmentation did

not use ITT or had losses to follow-up and one (Rath 2008) was

unclear. See Characteristics of included studies and the Risk of

bias tables for each study.



Effects of interventions

In this section we will present the results for the effects of interven-

tions only for studies that examined the primary and secondary

outcomes of interest in this review. Of the thirty-eight new studies,

seventeen (45%) assessed monotherapies (although two included

oral placebo and two included topical placebo). The other twenty-

one (55%) studies assessed combinations of interventions. Nearly

all of the studies examined insufficiently similar interventions to

allow data pooling, with the exception of one meta-analysis.

As not all of the trials reported on the outcomes of interest, forest

plots are only presented for 28 of the 57 studies. Many of the tri-

als had control arms in which none of the participants improved.

This resulted in large risk ratios (RRs) with very large associated

95% confidence intervals, exacerbated by the relatively small par-

ticipant numbers in many of the trials. Thus the magnitude of the

relative RRs quoted should be considered with caution, but some

conclusions may be drawn on statistical significance (or lack of it)

over control at the 95% confidence level.

Due to the number and complexity of the interventions in this

review, we have not repeated our results for those studies with

combination interventions under each of the relevant headings.

For example in Section 3.1,Oral PUVA we did not include details

of the primary outcome, more than 75% of repigmentation, for

Ermis 2001 because the main intervention was calcipotriol, not

PUVA. In these instances please refer to the Characteristics of

included studies for the main comparator for that study and the

outcomes that were addressed.

For the explanation of technical or medical terms found mainly

in Adverse Effects, please refer to the Glossary Table 1.

(1) Topical therapies

1.1 Topical corticosteroids

A total of ten studies examined the effect of topical steroids, either

asmonotherapy or in combination with other interventions. Eight

of these studies examined one or more outcomes of interest, and

two did not examine any outcomes of interest other than adverse

effects.

Primary outcomes

a) Quality of life

One study assessing topical corticosteroids assessed patient-rated

quality of life measures (Sassi 2008). Sassi 2008 compared the ef-

fects of topical hydrocortisone17-butyrate plus laser versus laser

alone. One outcome measure in this study was patient-rated qual-

ity of life, asmeasured by Skindex-29 (higher score indicates higher

quality of life); however no statistically significant difference was

seen between the intervention groups (MD 4.75, 95% CI -1.56

to 11.06, Analysis 1.1).

b) Percentage of repigmentation > 75%

Six studies examining the effect of topical corticosteroids addressed

this outcome, although repigmentation was measured in various

different ways. Four of these six studies examined the effect of

topical steroids as monotherapy against other monotherapies or

combinations of therapies, and the other two examined the effect

of topical corticosteroids in combination with other therapies.

In one RCT of children with vitiligo, Khalid 1995 compared top-

ical clobetasol propionate with topical PUVAsol in children with

vitiligo. Participants receiving clobetasol propionate were signifi-

cantly more likely than those receiving PUVAsol to achieve greater

than 75% repigmentation (RR 4.70; 95%CI 1.14 to 19.39, Anal-

ysis 2.1). Lim-Ong 2005 compared topical clobetasol propionate

plus NB-UVB versus placebo plus NB-UVB. There was no sta-

tistically significant difference between the two groups (RR 1.38;

95% CI 0.71 to 2.68, Analysis 3.1). Sassi 2008 compared topi-

cal hydrocortisone 17-butyrate plus laser versus laser. There was

a statistically significant difference in favour of the combination

treatment; these participants were more than twice as likely to

achieve 75% repigmentation than those receiving laser treatment

alone (RR 2.57; 95% CI 1.20 to 5.50, Analysis 4.1).

Kumaran 2006 compared topical betamethasone dipropionate

with either calcipotriol or betamethasone dipropionate plus cal-

cipotriol. None of the participants achieved greater than 75%

repigmentation (Analysis 5.1). Lepe 2003 compared topical clo-

betasol propionate and 0.1% tacrolimus. With respect to repig-

mentation of greater than 75%, there was no statistically signifi-

cant difference between the two interventions (RR 1.00; 95% CI

0.34 to 2.93 Analysis 6.1). Westerhof 1999 was a parallel-group

comparison of topical fluticasone propionate (FP) versus FP plus

UVA or versus UVA alone. No significant difference was seen be-

tween participants in the FP plus UVA group and those receiv-

ing FP alone, in achieving greater than 75% repigmentation (RR

3.38; 95%CI 0.76 to 15.12 Analysis 7.1). Fluticasone propionate,

alone and with UVA, was superior to UVA alone. (RR 3.94; 95%

CI 1.16 to 13.43 Analysis 8.1).

Secondary outcomes

a) Cessation of spread of vitiligo

Only one study assessed the ability of interventions to halt the

spread of vitiligo.

Lim-Ong 2005 assessed the effect of topical clobetasol propionate

plus NB-UVB versus placebo plus NB-UVB on vitiligo disease



activity by comparing pre- and post-treatment vitiligo disease ac-

tivity (VIDA) scores and permanence of repigmentation and de-

velopment of new lesions within 1 year post-treatment, as docu-

mented by photographs. There was no statistically significant dif-

ference between the two groups (RR 1.00; 95% CI 0.70 to 1.43

Analysis 9.1).

b) Long-term repigmentation

None of the studies assessed this outcome.

c) Adverse effects

All ten studies examining the effect of topical corticosteroids re-

ported adverse effects in some of the participants receiving them.

For combination therapies it is not possible to ascertain which of

these adverse effects were attributable to the topical steroids.

Of the studies that examined the effects of topical corticosteroids

as monotherapy, Khalid 1995 reported mild atrophy (four cases),

telangiectasia (two), hypertrichosis (one), or acneiform papules

(two) in participants treated with clobetasol propionate. Lepe

2003 also reported atrophy (three cases) and telangiectasia (two)

in participants after treatment with clobetasol. Kumaran 2006 re-

ported side-effects in seven participants treated with betametha-

sone, including lesional atrophy, soreness, and hypertrichosis. One

participant treated with a combination of betamethasone and cal-

cipotriol had hypertrichosis and another had dry skin in the le-

sions. Westerhof 1999 detected no evidence of dermal or epider-

mal atrophy on skin biopsy in participants treated with fluticasone

propionate.

Of the studies examining effects of combination therapies, Barman

2004 compared the effect of applying a topical corticosteroid

(0.1% fluocinolone acetonide) after punch grafting with the use

of PUVA therapy after punch grafting. Cobblestoning, depigmen-

tation of the grafts, infection, and graft displacement were seen in

some participants in both of the groups. Lim-Ong 2005 reported

adverse effects of vesicles, acneiform eruptions, hypertrichosis, and

striae in some participants treated with betamethasone plus NB-

UVB. Agarwal 2005 (see Section 2: Oral therapies) reported local

atrophy and telangiectasia in participants from both intervention

groups and acneiform lesions in one participant in the mometa-

sone plus levamisole group. (These adverse effectswere attributable

to topical mometasone, not levamisole, so we have listed them

here, rather than in Section 2: Oral therapies.) Sassi 2008 reported

hyperpigmentation in some participants receiving combination

treatment with hydrocortisone 17-butyrate plus laser, although

this was also seen in participants receiving only laser treatment.

Two studies assessing the effect of topical corticosteroids men-

tioned adverse events but did not address any other outcomes

of interest in this review. Kandil 1974 compared betamethasone

valerate in 50% isopropyl alcohol versus the alcohol base alone.

Adverse effects attributed to betamethasone valerate included hy-

pertrichosis in two participants and a localised acneiform eruption

in three participants. Sanclemente 2008 compared twice daily ap-

plication of betamethasone with a topical catalase/dismutase su-

peroxide. No adverse events were reported in participants receiv-

ing betamethasone.

1.2 Intralesional corticosteroids

One study (Vasistha 1979) assessed the effect of intralesional cor-

ticosteroids, but it did not examine any outcomes of interest other

than adverse effects. This study compared intralesional triamci-

nolone acetonide injections with placebo injections. Adverse ef-

fects in the intralesional steroid group included atrophy in eight

participants, telangiectasia in two, infection in one, and intrader-

mal haemorrhage in one.

1.3 Topical vitamin D analogues - (Calcipotriol andTacalcitol)

A total of nine studies assessed the effect of vitamin D analogues

as monotherapy or in combination with other interventions.

Primary outcomes

a) Quality of life



Kumaran 2006 was the only study to examine the effect of

calcipotriol as monotherapy (see Section 1.1, Topical Corticos-

teroids).

Four studies examining the effect of vitamin D analogues in com-

bination with other interventions assessed percentage of repig-

mentation of greater than 75%.

Parsad 1998 compared calcipotriol plus PUVAsol with placebo

plus PUVAsol. There was no statistically significant difference in

the number of participants achieving greater than 75% repigmen-

tation (RR 1.44; 95% CI 0.86 to 2.43, Analysis 10.1). Ermis

2001 compared calcipotriol plus PUVA with placebo plus PUVA.

The side of participants treated with the calcipotriol plus PUVA

had a significant 4 fold increase in the likelihood of achieving

greater than 75% repigmentation sooner than the side treated

with placebo plus PUVA (paired OR OR 4.25 (95% CI 1.43,

12.64), Analysis 11.1). Lu-Yan 2006 compared tacalcitol plus

308nm monochromatic excimer light (MEL) with placebo plus

MEL. A statistically significantly greater proportion of partici-

pants in the tacalcitol plus MEL group achieved greater than 75%

repigmentation (RR 4.50; 95% CI 1.05 to 19.35, Analysis 12.1).



Rodriguez-Martin 2009 compared tacalcitol plus sunlight versus

placebo plus sunlight, but found no difference between the groups

(RR 0.33; 95% CI 0.01 to 7.89, Analysis 13.1).

Secondary outcomes





c) Adverse effects

Parsad 1998 reported mild skin irritation in three participants

treated with calcipotriol. Ermis 2001 reported mild to moder-

ate erythema, xerosis (dryness), and itching in two participants

treated with calcipotriol. In the Kumaran 2006 study, one partic-

ipant using calcipotriol had perilesional hyperpigmentation and

one had an irritant reaction. Lu-Yan 2006 reported that a total

of six participants had mild to moderate erythema xerosis and it-

ching after combination treatment with tacalcitol and MEL. In

the Rodriguez-Martin 2009 study, a number of participants ex-

perienced itching or contact dermatitis at the site of application

of tacalcitol and a larger number noted transient erythema of the

treated skin.

Two studies did not assess any outcomes of interest other than ad-

verse effects. Leone 2006 compared tacalcitol plus NB-UVB with

NB-UVB alone. In the tacalcitol plus NB-UVB group, they re-

ported erythema and itching (all participants), mild irritation (12

cases), and desquamation (12). Arca 2006 compared calcipotriol

plusNB-UVBwithNB-UVB alone. Adverse effects of itching and

erythemawere mentioned but it was not clear in which group they

occurred.

Two studies did not report adverse effects or any other predeter-

mined outcomes of interest (Goldinger 2007; (see Section 3.7,

Lasers), Akhyani 2005).

1.4 Calcineurin inhibitors (Tacrolimus and Pimecrolimus)

A total of six studies assessed the effect of calcineurin inhibitors as

monotherapy or in combination with other interventions.

Primary outcomes

a) Quality of life



Lepe 2003 was the only study to examine the effect of topical

tacrolimus as monotherapy (see Section 1.1, Topical Corticos-

teroids).

All other studies examined the effect of topical calcineurin in-

hibitors in combination with other interventions.

Kawalek 2004 and Passeron 2004 both compared topical 0.1%

tacrolimus plus 308 nm xenon chloride excimer laser with placebo

plus laser, both studies used a within-participant design. We per-

formed a meta-analysis of these two studies, which demonstrated

that patches treated with the combination of topical tacrolimus

plus laser were more likely to achieve 75% repigmentation than

those treated with laser alone (RR 3.15; 95% CI 1.46 to 6.76,

Analysis 14.1); however this result should be treated with caution

since both of studies used a within-participant design and there-

fore the 95% confidence intervals from the meta-analysis may be

too narrow to reflect this.

Esfandiarpour 2009 compared topical pimecrolimus plus NB-

UVB versus placebo plus NB-UVB, but found no statistically sig-

nificant difference in rates of repigmentation between the groups

(RR 3.38; 95% CI 0.93 to 12.29, Analysis 15.1).

Secondary outcomes


Dawid 2006 compared topical pimecrolimus with vehicle. Ces-

sation of spread was assessed using the VIDA (Vitiligo Index of

Disease Activity) score, and found no significant difference be-

tween the two groups (topical pimecrolimus: median difference

90, range -2046 to +509; vehicle: median difference 114, range -

1230 to +615; P value 0.5, (Wilcoxon signed rank test).





c) Adverse effects

Lepe 2003 reported burning sensations in the skin of two partic-

ipants during the first two weeks of treatment with tacrolimus.

Kawalek 2004 reported mild to moderate erythema in all vitiligo

patches treatedwith tacrolimus plus laser, with blistering occurring

at one site. Eighty per cent of participants treated with this combi-

nation experienced a tingling and burning sensation and erythema

at the treatment site, compared to 30% treated with placebo plus

laser. Passeron 2004 observed moderate to severe erythema at least

one time in all participants from both groups; localised bullous

eruptions were observed in two lesions in both groups. However,

stinging was only observed in five participants treated with laser

and topical tacrolimus.

Dawid 2006 reported no adverse effects in response to topical

pimecrolimus. Esfandiarpour 2009 reported only self-limiting ery-

thema and pruritus with pimecrolimus plus NB-UVB.

Mehrabi 2006 compared topical tacrolimus plus NB-UVB with

placebo plusNB-UVB but the only outcome of interest assessed in

this study was adverse effects. These included erythema, pruritus,

blistering, or a burning sensation. It was not clear which treatment

combination led to these effects.

1.5 Khellin

Two studies examined the effect of khellin in combination with

UVA. Only one of them, Procaccini 1995, which compared the

application of khellin in two different vehicles (5% khellin in O/

W (oil in water) cream or 3% khellin in methyl-2-pyrrolidine,

PYR) plus UVA, with the vehicles alone plus UVA is discussed in

this section. The other study, Cestari 2001, compared topical 2%

khellin plus UVA with PUVA (see Section 3.1, Oral PUVA).

Primary outcomes

a) Quality of life

The study did not assess this outcome.


In Procaccini 1995, there was no statistically significant difference

in repigmentation between khellin 3% in PYR plus UVA, and

PYR plus UVA (RR 1.14; 95% CI 0.47 to 2.75 Analysis 16.1),

khellin 5% in O/W cream plus UVA, versus O/W cream plus

UVA (RR 1.00; 95% CI 0.46 to 2.15 Analysis 17.1), and PYR

plus UVA versus O/W cream plus UVA (RR 3.03; 95% CI 0.70

to 13.04 Analysis 18.1).

Secondary outcomes





c) Adverse effects

None were reported for this study.

1.6 Pseudocatalase and catalase / dismutase superoxide

Two studies examined the effect of pseudocatalase or similar com-

pounds. Schallreuter 2002 compared Dead Sea climatotherapy

plus pseudocatalase cream (PC-KUS) with Dead Sea climatother-

apy plus placebo cream and Dead Sea climatotherapy alone. How-

ever, this study did not examine any outcomes of interest and there

was no mention of adverse effects in the paper.

Sanclemente 2008 (see Section 1.1, Topical Corticosteroids) re-

ported a self-limiting erythematous papular rash in one partici-

pant treated with a topical catalase/dismutase superoxide, but did

not examine any other outcomes of interest.

1.7 Melagenina (human placental extract)

Souto 1997 was the only study to examine the effects of melagen-

ina, which it compared with placebo. However, the study exam-

ined no outcomes of interest and reported two cases of burning

with infra-red light but the group was not specified.

(2) Oral therapies

In this section we present the results of studies that used oral thera-

pies, either asmonotherapy or combined with other interventions.

The majority of these studies examined the effect of oral therapies

in combination with other interventions. Studies examining oral

PUVA as monotherapy are discussed in Section 3.1, Oral PUVA,

although some of the studies in this section used oral PUVA in

combination with other oral therapies. In total, eleven studies ex-

amined the effect of oral therapies.



Primary outcomes

a) Quality of life

Two studies assessed patient-rated quality of life. In Middelkamp-

Hup 2007, participants received eitherPolypodium leucotomos cap-sules plus NB-UVB or placebo plus NB-UVB. Patient-rated qual-

ity of life was assessed using Skindex 29 and no significant differ-

ences were reported in the paper for change in quality of life.

Agarwal 2005 compared oral levamisole plus topical mometasone

furoate with oral placebo plus topical mometasone. Quality of life

was assessed using threemeasures:Dermatology LifeQuality Index

(DLQI), Childrens Dermatology Life Quality Index (CDLQI),

and theWHOQuality of Life Brief Questionnaire (WHOQOL-

BREF). No significant differences were seen between the interven-

tion groups at 6 months follow-up (DLQI: oral levamisole: me-

dian 1, range 0 to 7; placebo: median 1, range 0 to 14: CDLQI:

oral levamisole: median 1, range 0 to 6; placebo: median 1, range

0 to 2. Data were not reported for WHO QOL-BREF).


Four studies assessed this outcome. One study (Parsad 2003b), ex-

amined the effect of an oral intervention as monotherapy, namely

oral Ginkgo biloba, compared with placebo. OverallGinkgo bilobashowed a significant improvement over placebo (RR 4.40; 95%

CI 1.08 to 17.95, Analysis 19.1).

Rath 2008 compared the effect of oral minipulses of betametha-

sone (OMP) with three different combination interventions,

namely: OMP plus PUVA; OMP plus NB-UVB, and OMP plus

BB-UVB. There was a statistically significant difference in favour

of OMP plus NB-UVB compared to OMP alone (RR 7.41; 95%

CI 1.03 to 53.26, Analysis 20.1), but not for OMP plus PUVA

versus OMP alone (RR 3.70; 95% CI 0.47 to 29.28, Analysis

21.1) or for OMP plus BB-UVB versus OMP alone (RR 1.67;

95% CI 0.11 to 24.26, Analysis 22.1).

Radmanesh 2006 compared oral azathioprine plus 8-MOP plus

UVA versus 8-MOP plus UVA. Those in the group receiving

azathioprine were statistically significantly more likely to achieve

greater than 75% repigmentation 4 months after treatment (RR

17.77; 95% CI 1.08 to 291.82, Analysis 23.1).

DellAnna 2007 compared an oral antioxidant pool plusNB-UVB

with NB-UVB alone. No statistically significant difference was

found between the two groups (RR 2.59; 95% CI 0.67 to 10.00,

Analysis 24.1).

Secondary outcomes


Three studies assessed this outcome. Parsad 2003b assessed cessa-

tion of spread of vitiligo as defined by the arrest of progression of

vitiligo in participants with unstable disease, presumably at the

end of the study period (six months). There was a statistically sig-

nificant difference in favour of the group receiving oral Ginkgobiloba (RR 2.20; 95% CI 1.22 to 3.95, Analysis 25.1).Siddiqui 1994 examined the effectiveness of oral L-phenylalanine,

either with or without UVA. Participants received either L-pheny-

lalanine or placebo andwere divided into two groups, one of which

received UVA and the other did not. There was no statistically sig-

nificant difference between the L-phenylalanine plus UVA group

and the no active treatment group (RR 1.36; 95%CI 0.86 to 2.13,

Analysis 26.1) or between the L-phenylalanine alone group versus

the no active treatment group (RR 1.31; 95% CI 0.74 to 2.32,

Analysis 27.1).

Agarwal 2005 compared oral levamisole plus topical mometasone

furoate with oral placebo plus topical mometasone, and assessed

cessation of spread by means of counting the number of any new

vitiligo lesions at each monthly visit. There was no statistically

significant difference between the groups (RR 1.38; 95% CI 0.92

to 2.06, Analysis 28.1).



c) Adverse effects

Of the studies mentioned above, Siddiqui 1994 reported only

mild nausea in participants treated with L-phenylalanine. Agarwal

2005 reported no adverse effects that were likely to be due to

oral levamisole. Parsad 2003b reported nausea in two participants

receiving Ginkgo biloba. Radmanesh 2006 reported two cases ofgastrointestinal upset on participants receiving azathioprine plus

PUVA. Middelkamp-Hup 2007 observed mild and transient it-

ching (10), dryness of the skin due to NB-UVB (5), and mild gas-

trointestinal complaints due to capsule intake (4) in the NB-UVB

plus Polypodium leucotomos. In the NB-UVB alone they observedthe same adverse effects in five participants (mild and transient

itching), three (dryness) and five (mild gastrointestinal complaints

due to capsule intake).

Rath 2008 reported nausea and weight gain in eleven participants

receiving OMP plus PUVA and excessive erythema and blistering

of the skin in five. Weight gain was reported in ten participants

receiving OMP plus NB-UVB. Excessive erythema occurred in six

participants receivingMOP plus BB-UVB and weight gain in five.

Ten participants receiving OMP alone experienced weight gain.



Three studies did not assess any of the predetermined outcomes

of interest other than adverse effects.

Tjioe 2002 compared oral vitamin B12 and folic acid plus NB-

UVBwithNB-UVB alone. The only adverse effects were prickling

sensations on depigmented areas; an occasional phototherapy-in-

duced erythema. Rojas-Urdaneta 2007 compared an antioxidant

and mitochondrial stimulating cream plus oral antioxidants and

phenylalanine with a placebo cream plus oral antioxidants and

phenylalanine and also with the active cream alone and placebo

treatment alone. The only adverse effect reported was mild acne

and pruritus in one participant receiving the active topical treat-

ment. Shi 2008 compared Zengse pill (ZSP) with or without oral

cobamamide plus topical psoralea tincture. There were five or six

reports of skin redness and itching in both groups, and one par-

ticipant receiving ZSP experienced constipation.

Three studies did not observe adverse effects (Agarwal 2005;

DellAnna 2007; Reyes 2006). The latter reported no adverse ef-

fects that were likely to be due to oral levamisole.

(3) Light therapies

3.1 Oral PUVA

Of a total of 10 RCTs evaluating oral PUVA, two studies as-

sessed oral PUVA alone (Bhatnagar 2007; Yones 2007). Eight

studies assessed oral PUVA in combination with other therapies

such as calcipotriol (Akhyani 2005; Ermis 2001), azathioprine

(Radmanesh 2006), Polypodium leucotomos (Reyes 2006), topicaldimethoxyamoidina (khellin) (Cestari 2001), and surgical thera-

pies e.g. punch grafting plus PUVA (Barman 2004), transplanta-

tion of cultured autologous melanocytes plus PUVA (CMP), suc-

tion blister transplantation plus PUVA (SBP) or cryotherapy plus

PUVA (CP) (Czajkowski 2004) and melanocyte transplantation

plus PUVA (VanGeel2004). The psoralen employed inmost stud-

ies was 8-MOP (8 methoxypsoralen), although Bhatnagar 2007

used trimethylpsoralen.

Primary outcomes

a) Quality of life

One study (Yones 2007) measured patient-rated quality of life:

Differences between the groups for the Dermatology Life Quality

Index and Visual Analogue Scale at one year post-treatment were

not statistically significant (reduction in DLQI p=0.8; reduction

in VAS p=0.5).


Ermis 2001; (see Section 1.3, Topical vitamin D analogues).

Radmanesh 2006; (see Section 2, Oral therapies).

There was no statistical difference in repigmentation in partici-

pants treated with PUVA compared to NB-UVB (RR 0.52; 95%

CI 0.02 to 11.98, Analysis 29.1) one year after treatment (Yones

2007), nor in participants treated with TMP plus UVA compared

to NB-UVB (

CD003263 Standard

Documents

Transcript of CD003263 Standard