A New Therapy for Vitiligo Using Fire Needles: A Systematic Review of Evidence from 3618 Subjects

Introduction Fire needle therapy has been reported as an effective treatment for vitiligo. However, current clinical evidence has not been systematically evaluated. The aim of this study was to determine whether fire needle therapy is effective and safe for treating vitiligo. Methods Seven databases were searched until October 2019 for randomized controlled trials on fire needle therapy, with and without conventional treatments, versus any type of conventional therapy for treating vitiligo. The RevMan 5.3.5 software was used to perform meta-analysis of the included studies. Results Forty-seven trials comprising 3618 patients were included. Fire needle combined with conventional vitiligo treatments had a higher efficacy (risk ratio (RR): 1.55, 95% confidence interval (CI): 1.46–1.65, P < 0.00001 and RR: 1.41, 95% CI: 1.24–1.61, P < 0.00001, respectively) and a greater effect on restoring the color of the area of the skin lesion (mean difference (MD): 3.40, 95% CI: 2.11–4.69, P < 0.00001), increasing the pigment point of vitiligo (MD: 0.83, 95% CI: 0.54–1.13, P < 0.00001) and improving the cytokine level (MD: 8.10, 95% CI: 6.94–9.27, P < 0.00001) and effectual time (MD: −4.76, 95% CI: −7.33 to −2.19, P=0.0003) than traditional methods. Limb lesions (RR: 1.60, 95% CI: 1.31–1.95, P < 0.00001) were more effectively treated when the treatments included fire needles, whereas the therapeutic effect of fire needles on either the head and neck (RR: 1.13, 95% CI: 0.78–1.64, P=0.52) or torso lesions (RR: 1.22, 95% CI: 0.82–1.81, P=0.33) was not significantly different compared to that without fire needles. No statistically significant differences in adverse effects (RR: 1.15, 95% CI: 0.89–1.49, P=0.28) and recurrence rates (RR: 0.90, 95% CI: 0.17–4.92, P=0.91) during the follow-up period were observed between treatment with and without fire needles. Conclusions Fire needle therapy combined with other conventional treatments is useful in treating vitiligo. Further studies with larger sample sizes should be performed to make a conclusive judgment. This trial is registered with CRD42018094918.


Introduction
Vitiligo is a frequently acquired depigmentation disease of localized or generalized skin and mucosa. A typical pathological feature of the disease is a lack of melanocytes and melanin granules, and it is characterized by whitening of lesions on the skin [1]. Approximately 1% to 2% of people globally have vitiligo [2], including both adults and children. e cause of vitiligo is still unclear, but it is considered to be related to heredity, immunity, mental and neurological causes, melanin self-destruction, and oxidative stress, among other causes [3]. Vitiligo could lead to serious loss of normal appearance, which may result in negative effects on learning, work, social interaction, and the marriage status of patients. Conventional treatments for vitiligo include topical corticosteroids, calcitonin inhibitors, phototherapy, and transplantation [4]. ese treatments have severe side effects including chemical drug toxicity and phototherapy diffusion in varying degrees [5]. Consequently, there is an urgent need for new approaches to treatment.
Vitiligo is called "Baidian" or "Baibofeng" in traditional Chinese medicine (TCM), which has a long history of treatment for the disease. Fire needle therapy, recorded as early as the time of the Yellow Emperor (475-221 BC), has been an integral part of TCM and the acupuncture method. In the procedure, a special nontoxic stainless-steel needle is inserted into the acupoint after heating (Supplementary file 1), thus improving localized blood flow in the skin lesions and promoting the formation of melanin [6], which has been proved to be effective for treating vitiligo [7,8]. However, to our knowledge, no systematic review has evaluated the benefits of fire needle therapy in patients with vitiligo. erefore, we conducted this systematic review to comprehensively assess the clinical efficacy and safety of this classic treatment for vitiligo, in order to develop and supplement the therapeutic approaches available.

Methods
is systematic review was performed following the Cochrane Handbook for Systematic Reviews of Interventions [9], presented under the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines (Supplementary file 2).

Inclusion Criteria.
In this analysis, we included randomized controlled trials (RCTs) that used interventions of either fire needles alone or fire needles combined with conventional therapy, compared with any conventional therapy for vitiligo without fire needles. We included studies in which the pricking area was on the Ashi points (local skin lesions). e patients must have been diagnosed with vitiligo, regardless of sex, age, and ethnicity.

Outcomes.
e primary outcome measured in this study was the total effectiveness rate for the duration of treatment, defined as the rate of restoration of the vitiligo lesions to normal color.
is was divided into four categories: (i) cured: vitiligo had subsided with the skin color restored to normal; (ii) markedly effective: vitiligo had partially subsided or reduced, with more than 50% of the skin area restored to the normal skin color; (iii) effective: part of the vitiligo had subsided or reduced; and (iv) ineffective: vitiligo had been pigmented, regenerated, or expanded with reduction of the healing number.
If the trials used the number of patients as the unit for statistical outcomes, the effectiveness rates were calculated using the following NP formula: effectiveness rate � (number of cured patients + number of markedly effective patients)/ number of total patients × 100%. If the trials used the number of vitiligo skin lesions as the unit for statistical outcomes, the effectiveness rates were calculated by the following NVSL formula: effectiveness rate � (number of cured vitiligo skin lesions + number of markedly effective vitiligo skin lesions)/number of total vitiligo skin lesions × 100%. e secondary outcomes included total restoration of the area's color, the total increased pigment point, changes in the cytokine level, effectual time, rate of the therapy's effectiveness in different locations, and adverse effects, as well as recurrence rates. e pigment point was defined as follows: 0 points, pure white skin lesions without any pigment production; 1 point, pale white skin with a little pigmentation; 2 points, pale brown skin lesions with a lot of pigmentation; and 3 points, skin damage with tan, basic, or near normal skin color.

Selection of Studies and Data Extraction.
We searched the Cochrane Library, Excerpta Medica data BASE (EMBASE), PubMed, the China National Knowledge Infrastructure Database, Chinese Scientific Journals Full Text Database, Wanfang Data Knowledge Service Platform, and the Chinese Biomedical Literature Service System. We included papers dating from the earliest citation in the databases until October 2019. e search terms included "vitiligo" or "leukoderma," combined with "fire needle".
Two investigators (R.Y. and Z.Y.) independently screened the studies according to the inclusion criteria and extracted the information based on self-designed data-extraction templates, which included the first author, study characteristics (i.e., year, duration, setting, and design), participant characteristics (i.e., mean age, sample size, TCM syndrome, and interventions), and measured outcomes.

Risk of Bias Assessments.
e Cochrane Handbook [9] was used to evaluate the methodological quality of each included study in terms of the following characteristics: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of the outcome assessment, incomplete outcome data, selective reporting, and other biases. e terms "low," "unclear," and "high," respectively, referred to low, uncertain, and high risks of bias. e results were cross checked by two investigators (Z.Q. and L.L.), and the disagreements were settled by discussion between them.

Data Analyses.
e RevMan 5.3.5 software (Cochrane Community, London, United Kingdom) was used for the data analyses. Dichotomous data are expressed as risk ratios (RRs) with 95% confidence intervals (CIs). Continuous data are expressed as mean differences (MDs) with 95% CIs. For each meta-analysis, we evaluated the statistical heterogeneity. If the trial had acceptable homogeneity (P > 0.1, I 2 < 50%), a fixedeffects model was applied; if there was statistical heterogeneity (P < 0.1, I 2 > 50%), we used a random-effect model. Clinical heterogeneity was assessed by reviewing the differences in the distribution of participants' characteristics between trials (i.e., age, sex, and disease duration).

Description of Studies.
After performing primary searches of the seven databases, we identified 342 potential articles. We excluded 293 as they did not conform to the inclusion criteria. Eventually, 47 trials with a total of 3618 patients  were included in this review ( Figure 1). Among the included articles, eight [29-32, 46, 47, 50, 55] were unpublished master's theses. e characteristics of the included trials are presented in Table 1. All 47 trials  were performed in China. Only one trial [31] was a multicenter study, and the rest were single-center studies. e study population included children and adults. All included patients had a confirmed diagnosis of vitiligo. In the diagnosis of TCM syndromes, eight trials [12,19,21,36,46,48,52,53] only included patients with "Qi-stagnancy and blood stasis," "liverkidney yin deficiency," "blockage of the vessel," or "liver depression and Qi-stagnancy" TCM patterns, as described in the TCM dialectic. Interventions included the combination of fire needle therapy on the Ashi points (local skin lesions) and conventional treatments (308-nm excimer laser, narrow-band ultraviolet B light, Western medicine ointment, and other TCM methods). e control participants used conventional treatments, without the use of fire needles. Courses of treatment ranged from 2 to 6 months.

Methodological Quality.
e Cochrane risk of bias for all 47 trials is shown in Figure 2. Two trials [31,32] reported using a sample-size calculation method. Eleven trials Studies excluded, with reasons Self-controlled trial (n = 11) Control group included fire needle (n = 2) Unclear sample size (n = 2) Unclear outcomes data (n = 1) Duplicate publication of data (n = 3) Lack of all primary outcomes (n = 1) Evidence-Based Complementary and Alternative Medicine Evidence-Based Complementary and Alternative Medicine Evidence-Based Complementary and Alternative Medicine   [31,32] used blinding to assess the outcomes. Four trials [10,13,19,20] did not report the loss of patients to follow-up, while only one trial [22] used an intention-totreat analysis.

Total Effectiveness
Rate. Forty-four trials [10-12, 14, 16-46, 48-56] used the number of patients as the unit of statistical outcomes with the total effectiveness rate counted using the NP formula. Of the 44 trials, nine [12,14,17,20,23,24,39,45,49] compared the use of fire needles and a 308-nm excimer laser with a 308-nm excimer laser alone; eight trials [10,16,18,21,35,36,41,48] compared the use of fire needles and other TCM methods with other TCM methods alone; three trials [28,31,54] compared the use of fire needles and tacrolimus ointment with tacrolimus ointment alone; 14 trials [11,19,22,27,33,37,38,40,42,44,46,53,55,56] compared the use of fire needles and other treatments with other treatments alone; three [25,26,50] compared the use of fire needles and other TCM methods with both the phototherapy and the same TCM methods; and seven trials [29,30,32,34,43,51,52] Figure 3) and found that publication bias existed but was small. e remaining three trials [13,15,47] used the number of vitiligo skin lesions as the unit of statistical outcomes with the total effectiveness rate counted using the NVSL formula. All three trials involved interventions of fire needle therapy and other treatments vs. other treatments alone (specific types in Table 1 (A), (B)). Using a fixed-effects model ( Table 2), we found that there was an obvious difference in the effect of the therapies (RR: 1.41, 95% CI: 1.24-1.61, P < 0.00001). [29-32, 35, 36, 42, 50, 51, 54-56] reported data on the total area of the skin lesions both before and after treatment. In a comparison of the restored color on skin lesions using random-effects modeling (Supplementary file 3), we found that treatment including fire needles had a significantly greater effect in restoring the color of the skin than traditional methods without fire needles (MD: 3.40, 95% CI: 2.11-4.69, P < 0.00001). [32,37,52,[54][55][56] reported data on the total pigment point of the skin lesions both before and after treatment. In a comparison of the increased pigment point of these skin lesions using random-effects modeling (Supplementary file 4), we found that treatment including fire needles had a significantly greater effect in increasing the pigment point of the skin than traditional methods without fire needles (MD: 0.83, 95% CI: 0.54-1.13, P < 0.00001).

Changes in the Cytokine
Level. Five trials [35,37,38,52,56] reported various changes in the cytokine level, which demonstrated that fire needle therapy combined with traditional treatments could affect vitiligo-related cytokines. Two trials [38,56] reported the serum interleukin-(IL-) 17 level both before and after treatment. In a comparison of changes in the serum IL-17 level using a fixed-effects model (Supplementary file 5), we found that treatment including fire needles had a significantly greater effect on reducing the serum IL-17 level than traditional methods without fire needles (MD: 8.10, 95% CI: 6.94-9.27, P < 0.00001).

Effectual Time.
ree trials [26,36,45] reported the effectual time of treatments in both experimental and control groups. In a comparison of the effectual times using randomeffects modeling (Supplementary file 6), we found that treatment including fire needles had a significantly therapeutic effect in a shorter time than traditional methods without fire needles (MD: −4.76, 95% CI: −7.33 to −2.19, P � 0.0003).

Rate of the erapy's Effectiveness in Different Lesion
Locations.
ree trials [21,23,26] examined the effectiveness of treating lesions with different locations in both the experimental and control groups using the NVSL formula. With random-effects modeling, we discovered that interventions including fire needles for lesions placed at limbs had a better effect than interventions that did not use fire needles (RR: 1.60, 95% CI: 1.31-1.95, P < 0.00001) (Supplementary file 7). At the same time, when the conventional treatments with and without fire needles on either head and neck lesions (RR: 1.13, 95% CI: 0.78-1.64, P � 0.52) or torso lesions (RR: 1.22; 95% CI: 0.82-1.8, P � 0.33) were compared, there was no statistically significant difference.
Only one trial [11] examined the effectiveness of treating lesions placed at different locations in both the experimental and control groups using the NP formula, which concluded that treatments ("Bailing pian" of Chinese patent drug and NB-UVB) with fire needles on the head, neck, torso, and limbs had a better effect than without fire needles.

Discussion
e data in this study indicated that when compared with traditional methods, fire needles combined with conventional treatments for vitiligo had a higher efficiency, as well as a shorter time before taking effect, and a greater ability to reduce serum cytokines related to vitiligo, restore the color, and increase the pigment point of skin lesions. Limb lesions were more effectively treated when the treatments included fire needles, whereas the effect of treatment on lesions placed at the head, neck, and torso was not significantly different between treatments that used and those that did not use fire needles. ere was no statistically significant difference in adverse effects and the recurrence rate during the follow-up period between the methods that included and those that did not include fire needles.
Vitiligo, a disease with a high incidence, is difficult to treat. e main symptoms of skin leukoplakia affect patients' work, daily life, and mental state, thereby leading to serious problems with sleeping and learning. Improvement in the therapeutic outcomes of the condition would improve patients' quality of life [57]. We based this study on the TCM theory, verifying that fire needle therapy, which had some therapeutic effect without serious side effects and could be applied directly to the skin, might be a new way to treat vitiligo.
Chinese medicine proposes that when the human body is greatly stimulated by spirit or material, the "peak potential" of the body's electrical activity is increased, and the electrical activity of the skin damaged by tyrosinase is decreased. e binding of tyrosinase oxidative dopamine with protein is affected, thus destroying the formation of melanosomes. With abnormal neural electrical activity, the function of the cutaneous nerve crest is damaged, along with affected melanocyte formation and metabolism, thereby causing vitiligo [58]. In recent years, fire needle therapy has been proved to  be an effective and almost side-effect-free external treatment in TCM [59,60], while clinical research on the application of fire needle therapy for vitiligo has become a popular treatment method. A previous study confirmed that, after the treatment of vitiligo with fire needles, color on the lesions emerged; in the meantime, using a confocal laser scanning microscope, it was observed that dendritic melanocytes appeared in the skin lesions, the pigment content gradually increased and, then, formed a complete pigment ring, and the melanin ring gradually became bright [57]. Laboratory studies also demonstrated that fire needle therapy could improve the recoloration rate of hair follicles and promote the recoloration in refractory locations of acro-limbs. It has been reported that fire needle therapy could help to improve localized blood flow, expand the capillaries, and accelerate the circulation of blood. Nutrition could be strengthened to stimulate the activity of tyrosinase, thus facilitating the formation of melanin [6]. However, acupuncture needling into the skin produces localized inflammatory stimulation that might activate amelanotic melanocytes (AMMC) in the hair root sheath outside the hair follicle, thereby promoting the differentiation of AMMCs to normal melanocytes [61]. e efficacy of fire needle therapy for vitiligo involves the immunomodulatory mechanism that is associated with regulating the proportion of 17 cells, IL-17, and IL-23 cytokine levels, as well as the function of T lymphocytes [62,63].
is study had some limitations that should be acknowledged. e studies included in this review generally had poor methodologies, which could have caused bias.
Not all studies mentioned the method of random order generation. In addition, only two trials [31,32] mentioned the blinding method and one trial [32] utilized concealment of allocation. Moreover, the difficulties associated with blinding in studies of acupuncture treatment led to a low methodological quality, causing possible selection bias in RCTs. e safety and efficacy of acupuncture treatment is attracting increasing attention from the domestic and international medical communities. Nevertheless, acupuncture, particularly fire needle therapy, differed significantly in characteristics from drug treatment; thus, effective blinding and the selection of an appropriate placebo had long been recognized as extremely challenging [64]. It was quite clear that the development of more appropriate RCT designs and protocols was an urgent problem that acupuncture therapy faced. Since fire needle therapy was a TCM treatment, all of the studies included were performed in China. e therapy's application in other countries and regions must be studied further. Few studies mentioned the recurrence of lesions and the efficacy of the treatment for lesions placed at different locations, which might have brought about a difference in the results. Additionally, most existing studies had small sample sizes, which might have produced a high level of bias.
e results of this review showed that there is still a lack of well-designed studies on fire needle therapy in the treatment of vitiligo. Since severe adverse effects were not reported in the included studies and the curative effect of therapy with fire needles was definite, we believe that it  would be worthwhile to conduct further rigorously designed RCTs with larger sample sizes and high methodological quality on fire needle therapy in the treatment for vitiligo in the future.

Conclusions
Fire needle therapy combined with other conventional treatments had some effect on vitiligo lesions, were not associated with any side-effects, and could be administered in a straightforward manner. More high-quality studies with larger sample sizes should be conducted to make a conclusive judgment of its broader application for treatment.

TCM:
Traditional Chinese medicine RCTs: Randomized controlled trials NP formula: Effectiveness rate � (the number of cured patients + the number of markedly effective patients)/the number of total patients × 100% NVSL formula: Effectiveness rate � (the number of cured vitiligo skin lesions + the number of markedly effective vitiligo skin lesions)/ the number of total vitiligo skin lesions × 100% RRs: Risk ratios CIs: Confidence intervals MDs: Mean differences NB-UVB: Narrow-band ultraviolet B light AMMC: Activate amelanotic melanocyte IL: Interleukin vs: Versus.

Data Availability
All data generated during this study are included in this article and in the Supplementary Information files.

Conflicts of Interest
e authors have no conflicts of interest to declare.