記住我
ORIGINAL ARTICLE
Year : 2022 | Volume
: 40
| Issue : 2 | Page : 85-93
The optimal concentration of intralesional triamcinolone acetonide for patchy alopecia areata: A systematic review and meta-analysis
Hsuan-An Su1, Yu-Tsung Chen2, Yu-Chia Chen1
1 Department of Dermatology, Far Eastern Memorial Hospital, New Taipei City, Taiwan
2 Department of Dermatology, Taipei Medical University Shuang Ho Hospital; School of Public Health, College of Public Health and Nutrition, Taipei Medical University, Taipei, Taiwan
Correspondence Address:
Dr. Yu-Chia Chen
No. 21, Section 2, Nanya South Road, Banciao District, New Taipei City 220
Taiwan
Source of Support: None, Conflict of Interest: None
DOI: 10.4103/ds.ds_15_22
Background: Intralesional steroid treatment for alopecia areata (AA) has been developed for decades, yet the optimal concentration of triamcinolone acetonide (TrA) is not well-established. Objectives: This review aims to determine the optimal concentration of intralesional TrA in treating patchy AA. Methods: We conducted a systematic review and meta-analysis, and searched the Cochrane Library, Embase, and PubMed databases on July 4, 2021, to identify randomized or nonrandomized comparative studies reporting the response rates and/or adverse events among AA patients treated with various concentrations of TrA. The meta-analysis of proportions and odds ratios was analyzed using random-effects modeling. Results: Nineteen studies and a total of 783 participants were included. The estimated response rate of 5 mg/dl (74.82%, 95% confidence interval [CI] 64.99%–83.50%) was shown to be more efficacious than 2.5/3.33 mg/dl (38.64%, 95% CI 16.98%–62.99%) but similar to 10 mg/dl (71.06%, 95% CI 59.72%–81.20%), while pooled estimate of odds ratios revealed higher efficacy with 10 mg/dl than 5 mg/dl (odds ratio = 1.64, 95% CI 1.05–2.58, P = 0.031). The rates of skin atrophy were 18.05% (95% CI, 10.32%–27.38%), 11.49% (95% CI, 2.86%–24.84%), and 3.85% (95% CI, 1.27%–14.01%) in groups 10, 5, and 2.5/3.33 mg/dl, respectively. Higher concentration is associated with more skin atrophy in a dose-dependent fashion (P = 0.012). Heterogeneity among studies in the meta-analyses was high. Conclusion: The optimal intralesional concentration of TrA for patchy AA is probably 10 mg/dl with acceptable adverse events.
Keywords: Alopecia areata, intralesional injection, intralesional steroid, triamcinolone acetonide
Alopecia areata (AA) is an autoimmune disorder characterized by non-scarring hair loss of the scalp and body hair.[1] Clinical presentations range from several discrete well-circumscribed oval patches of hair loss on the scalp, face or body, involvement in the entire scalp (AA totalis), to involvement in the whole body (AA universalis).[1] AA has complicated pathophysiology including genetic, environmental, immunologic and psychological factors, and unfortunately, its prognosis is unpredictable.[2]
Despite numerous conventional and novel treatment options have been developed, intralesional triamcinolone acetonide (TrA) remains the first-line therapy for AA.[2],[3] Although intralesional TrA has been practiced for decades, the optimal concentration of TrA is not yet well-established. Furthermore, common adverse events, such as skin atrophy and telangiectasia, are of great clinical concern.[3]
In 2010, TrA 5 mg/dl to the scalp and 2.5 mg/dl to the face every 4–6 weeks were the suggested regimen by experts' opinion.[3] In 2019, Yee et al. conducted a systematic review and meta-analysis of limited studies, and reported that 5 mg/dl TrA may offer the greatest benefit to patients with focal AA.[4] Since clinical studies investigating the efficacy of TrA for AA have increased in recent years, there is a need to update the current evidence. With a relatively limited amount of high-quality randomized controlled trials (RCTs) and comparative studies, we aimed to perform a systematic review and meta-analysis of all relevant comparative studies to determine the preferred concentration of TrA for AA.
Materials and MethodsReview protocol
Recommendations of the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) statement, explanation and elaboration document, and checklist were followed as closely as possible to guide our methodology and reporting.[5]
Literature search and search strategy
We conducted a systematic review and meta-analysis of all relevant comparative studies on the preferred concentrations of intralesional TrA in treating AA. We searched the Cochrane Library, Embase, and PubMed for relevant studies on July 4, 2021, using the search strategy listed in [Table 1]. No additional filter or limit was applied.
Study selection
Inclusion criteria were all RCTs or nonrandomized comparative studies that reported rates of hair regrowth with different concentrations of intralesional TrA for AA. Exclusion criteria included review articles, noncomparative single-arm studies, case series, and case reports. Studies with alopecia totalis or alopecia universalis, concurrent therapies, sample size < 10 subjects, written in non-English, unpublished studies without peer-review, and injection modalities other than conventional syringes were also excluded. If overlapping data sets were described in different publications, the study with the larger population was included. Our primary outcome was the response rate in the hair regrowth area or other comparable parameters. Our secondary outcome was adverse events including local skin atrophy, telangiectasia, and pain. Management of references and automatic identification of duplicates were conducted using the EndNote software (version 20, Bld 14672, for Windows, Clarivate™). Two authors (HA Su and YT Chen) independently reviewed all of the titles, abstracts, and/or full-texts of searched results to identify potentially eligible studies. A third author (YC Chen) provided arbitration.
Data extraction
By scrutinizing the methodology, results, figures, and tables, and supporting information of the full-texts, one author (HA Su) extracted the following data from the included studies: (1) study design, (2) types of AA investigated, (3) numbers of subjects, (3) concentrations of TrA, (4) treatment frequency and duration, (5) definition of response to treatment, (6) response rate of hair regrowth, (7) adverse events. If data were incomplete in the text, we extrapolated from available data or tried to contact the authors by E-mail for potentially available data. Other authors verified the extracted data.
The quality of included studies was assessed by one author (HA Su) using the Cochrane Collaboration's tool[6] with the software Review Manager Version 5.4 (The Cochrane Collaboration, 2020) for RCTs, and the Methodological index for nonrandomized studies for the nonrandomized interventional comparative studies.[7] Another author (YC Chen) reviewed and approved the judgment. Missing, unclear, or inadequate information were sought throughout the texts and would be ranked as unclear or high risk of bias.
Statistical analysis
The acquired data were organized with Microsoft Excel and all statistical analyses including the meta-analysis were performed using MedCalc, version 18.11 (MedCalc Software, Ostend, Belgium; https://www.medcalc.org). The studies were grouped for synthesis according to the TrA concentration into three groups: 10 mg/dl, 5 mg/dl, and 2.5/3.33 mg/dl. The response rates were presented as proportions in percentages (%) with 95% confidence intervals (CI) in the meta-analysis of proportions and odds ratios (ORs). Meta-analysis was performed with random-effects model and was illustrated as forest plots. The I2 statistic was used to assess statistical heterogeneity. Kruskal–Wallis test was used to examine the response rates and adverse rates among groups, and Jonckheere–Terpstra trend test was used to determine a trend in the data. Funnel plots were illustrated for potential reporting bias. If reasonable, subgroup analysis would be performed to explore possible causes of heterogeneity. Results were considered statistically significant for P < 0.05 to address the certainty of estimates. Sensitivity analysis was tested by excluding studies enrolling subjects with facial AA involvements, by excluding study with a relatively implicit definition of the assessment criteria, and by including only RCTs comparing different TrA concentrations. The test for the fixed-effect model was not considered for sensitivity analysis because of the apparent between-study and within-study variance.
ResultsLiterature search and characteristics of included studies
[Figure 1] shows the process of literature search and selection of potential studies using the template from the PRISMA 2020 study flow diagram.[5] The search identified 1,266 studies, from which 208 duplicates were excluded, and other 1,023 studies were excluded after screening. Except for 4 studies without available full-text, the remaining 31 studies were examined for eligibility, from which 12 studies were excluded with reasons. The reasons for the exclusion of studies that might meet the inclusion criteria are listed in [Supplementary Table 1]. Finally, 19 studies were included in the review, 17 of which were included in the meta-analysis of the primary outcome.
Figure 1: Preferred reporting items for systematic reviews and meta-analysis flow chart of study identification.Characteristics of the included studies are listed in [Supplementary Table 2]. Twelve of the studies were RCTs, five were prospective comparative studies, and two were retrospective comparative studies. Seven studies conducted intra-patient comparisons.[8],[9],[10],[11],[12],[13],[14] Only three studies compared different TrA concentrations for AA,[11],[15],[16] while the others compared intralesional TrA with other treatments. The subjects were mainly localized patchy AA on the scalp (<50%), with a few subjects having facial involvement.[9],[10],[15],[17],[18],[19] One study reported inclusion of Ikeda type I AA.[20] A patch of AA or an individual subject was viewed as a treatment unit. The TrA concentrations included 10, 5, 3.33 and 2.5 mg/dl. The subjects were treated with a 4-week interval, 3-week interval, and 2-week interval in eight, eight, and three studies, respectively; most of the endpoints were evaluated at the 12th week. With various treatment courses and outcome assessments from the included studies, the following methodologies were considered eligible: (1) outcome assessment tools evaluating the hair regrowth percentage, including hair regrowth scale (RGS),[11] the Severity of Alopecia Tool score system,[21] global assessment for hair regrowth,[22] and mean improvement score by the physician (MISP);[10] (2) definitions for a successful treatment response at the endpoints comparable to >50% regrowth, for example, RGS score 3 and 4, global assessment for hair growth A3 and A4,[22] grade of improvement IV and V,[21] MISP score as marked, excellent or complete,[10] and descriptive criteria “adequate response and complete regrowth;”[20] and (3) the endpoints were defined as the final evaluations with analyzable data. RGS score 4,[15] hair regrowth >60%,[18] complete remission,[8] and adequate response[20] were respectively adopted by 4 studies; one study did not define the response to treatment.[23]
Efficacy of different concentrations of triamcinolone acetonide
Nine studies provided data for the response rate in subjects treated with 10 mg/dl TrA,[11],[15],[16],[17],[20],[21],[23],[24],[25] and the pooled estimate of response rate was 71.06% (n = 306/425, 95% CI 59.72%–81.20%). Eight studies provided data for the response rate in subjects treated with 5 mg/dl TrA,[11],[14],[15],[16],[18],[19],[22],[26] and the pooled estimate of response rate was 74.82% (n = 328/453, 95% CI 64.99%–83.50%). Five studies provided data for the response rate in subjects treated with 2.5/3.33 mg/dl TrA,[8],[9],[10],[11],[15] and the pooled estimate of response rate was 38.64% (n = 114/228, 95% CI 16.98%–62.99%) [Figure 2]. Visual inspection of the funnel plots revealed no significant publication bias in all three meta-analyses [Supplementary Figure 1]. The heterogeneity of data was high as indicated by I2 values ranging from 79.84% to 91.57% (all P < 0.001). Using Kruskal–Wallis test (P = 0.037), 10 mg/dl and 5 mg/dl TrA showed similar response rates, but were both significantly more efficacious in treating AA than 2.5/3.33 mg/dl (P < 0.05). Three studies comparing different TrA concentrations provided data for odds ratios in subjects treated with 10 and 5 mg/dl TrA.[11],[15],[16] The efficacy of 10 mg/dl was higher than that of 5 mg/dl (OR = 1.64, 95% CI 1.05–2.58, P = 0.031) [Figure 3].
Figure 2: Pooled estimates of response rates in localized patchy alopecia areata treated with Triamcinolone acetonide (a) 10 mg/dl, (b) 5 mg/dl, and (c) 2.5/3.33 mg/dl.
Figure 3: Pooled odds ratio of treatment efficacy in localized patchy alopecia areata treated with Triamcinolone acetonide 10 versus 5 mg/dl.Adverse events following the intralesional triamcinolone acetonide
Adverse events were reported in 16 studies, 3 of which reported none observed.[8],[12],[24] Skin atrophy was reported in 9 studies,[10],[11],[15],[16],[17],[18],[20],[21],[25] pain in 7 studies,[10],[13],[14],[17],[18],[21],[22] telangiectasia in 3 studies,[10],[11],[15] hypopigmentation in 2 studies,[15],[25] erythema in 2 studies,[25],[26] burning sensation in 2 studies,[17],[26] hyperpigmentation in 1 study,[15] folliculitis in 1 study,[15] and itching in 1 study [Supplementary Table 2].[17] The endpoint of skin atrophy in Ustuner et al.'s study was set at the 3rd month rather than the final visit at 6th month, due to a dropout rate 62.62% at the 6th month.[15]
Eight studies provided data for the adverse events of skin atrophy in subjects treated with 10 mg/dl TrA,[11],[15],[16],[17],[20],[21],[24],[25] and the pooled estimate of event rate was 18.05% (n = 71/314; 95% CI, 10.32%–27.38%). Four studies provided data for the adverse events of skin atrophy in subjects treated with 5 mg/dl TrA,[11],[15],[16],[18] and the pooled estimate of event rate was 11.49% (n = 42/317; 95% CI, 2.86%–24.84%). Five studies provided data for the adverse events of skin atrophy in subjects treated with 2.5/3.33 mg/dl TrA,[8],[10],[11],[12],[15] and the pooled estimate of event rate was 3.85% (n = 22/196; 95% CI, 1.27%–14.01%) [Figure 4]. Visual inspection of the funnel plots revealed no significant publication bias in all three meta-analyses [Supplementary Figure 2]. The heterogeneity of data was high as indicated by I2 values ranging from 72.58% to 88.99% (all P < 0.001). The result of Kruskal–Wallis test failed to reach statistical differences among groups 10, 5, and 2.5/3.33 mg/dl TrA (P = 0.056), yet a dose-dependent trend of more skin atrophy with increased TrA concentration was identified using the Jonckheere-Terpstra trend test (P = 0.012).
Figure 4: Pooled estimates of skin atrophy rates in localized patchy alopecia areata treated with Triamcinolone acetonide (a) 10 mg/dl, (b) 5 mg/dl,s and (c) 2.5/3.33 mg/dl.Three studies comparing different TrA concentrations provided data for skin atrophy rates in subjects treated with 10,[11],[15],[16] 5,[11],[15],[16] and 2.5/3.33[11],[15] mg/dl TrA. The skin atrophy rates are higher with 10 mg/dl in comparison to 5 mg/dl (OR = 1.72, 95% CI 1.07–2.79, P = 0.027) and 2.5/3.33 mg/dl (OR = 2.73, 95% CI 1.00–7.39, P = 0.049), while the skin atrophy rates were not significantly different between 5 and 2.5/3.33 mg/dl (OR = 1.45, 95% CI 0.79–2.67, P = 0.231) [Figure 5].
Figure 5: Pooled odds ratios of skin atrophy rates in localized patchy alopecia areata treated with Triamcinolone acetonide (a) 10 mg/dl, (b) 5 mg/dl, and (c) 2.5/3.33 mg/dl.Adverse events of temporary and mild pain at the injection site were reported in studied populations variously with a range 10%–75%.[10],[14],[17],[18],[22] Kapoor et al. reported a mean score for pain 2.25 ± 3.27 using visual analog scale (VAS),[21] Mahgoub et al. reported pain (VAS = 3) in 50% subjects,[13] and Abdallah et al. only reported tolerable mild burning pain that did not necessitate discontinuation of the treatment.[14] Telangiectasia was reported as 4.8% and 14.7% with 10 mg/dl TrA from 2 studies,[11],[15] 2.4% and 0% with 5 mg/dl TrA from 2 studies,[11],[15] and 1.6% and 0% with 2.5 mg/dl and 3.33 mg/dl TrA from 3 studies.[10],[11],[15]
Quality of included studies
The level of evidence in [Supplementary Table 2] was evaluated according to the Oxford Centre for Evidence-Based Medicine 2011 Levels of Evidence.[27] The risk of bias summary of the 10 RCTs including in the meta-analysis is illustrated in [Supplementary Figure 3]. These RCTs were rated as having unclear risk of bias on selection bias and high risk of bias on performance and detection bias. Quality assessment for the included nonrandomized interventional comparative clinical studies was tabulated in [Supplementary Table 3]. Overall, because of the inclusion of randomized and nonrandomized comparative studies with unclear to high risk of bias and high degrees of heterogeneity, the certainty of the evidence for this systematic review is low.
DiscussionThis meta-analysis suggested that the intralesional TrA 10 mg/dl has the highest efficacy in treating patchy AA, decreasingly followed by 5 mg/dl and 2.5/3.33 mg/dl. Patients treated with 10 mg/dl TrA also have a higher chance to develop local skin atrophy than 5 mg/dl and 2.5/3.33 mg/dl in a dose-dependent fashion. Despite the higher risk of local adverse events such as skin atrophy, the adverse events were mild, transient, reversible, and avoidable.[3],[11] Therefore, we suggest 10 mg/dl as the preferred concentration of intralesional TrA to start off for patients with patchy AA.
Despite intralesional corticosteroid treatment for AA has been developed for decades, a consensus on the concentration has not been well-established, and physicians often select the concentration by experience.[15] Review articles in 2010 suggested 5 mg/dl be preferred concentrations for adult AA patients with <50% scalp involvement,[3],[28] whereas other studies suggested 10 mg/dl,[15] 5 mg/dl,[11],[16] or 2.5 mg/dl[29] as the preferred concentration. A systematic review and meta-analysis of heterogeneous studies reported that 5 mg/dl may offer the optimal benefit to AA patients.[4] After Yee et al.'s review, many randomized or nonrandomized comparative studies with at least one arm treated with intralesional TrA for AA have been published.[10],[11],[12],[13],[14],[16],[21],[22],[24],[25],[26],[30],[31],[32],[33] Unfortunately, high-quality RCTs comparing the various concentrations of intralesional TA in scalp AA are still limited.[11],[15] Because intralesional treatment for AA is highly operator-dependent on the performers' skill and experience, we excluded non-comparative studies such as case series or observational cohorts trying to minimize, though inevitable, the inter-operator variability.
Rajan et al. and Usturner et al. conducted RCTs and demonstrated both higher response rates and higher chances of skin atrophy with higher TrA concentrations.[11],[15] Since the pooled estimated response rates of 10 mg/dl TrA (71.06%) and 5 mg/dl (74.82%) did not show significant difference examined by the Kruskal-Wallis test (P = 0.037), we further conducted a meta-analysis of odds ratio which showed a consistent result with Rajan et al.'s and Usturner et al.'s studies. Since AA is a disease of multifactorial pathogenesis, we believe that the response rate may reach a plateau as the TrA concentration increases, and the concentration of maximal effect could only be designated after more comparative clinical trials have been completed.
The main adverse events associated with intralesional TrA are skin atrophy, pain, and telangiectasia.[3] Among the adverse events reported, only skin atrophy was meta-analyzed in our study and resulted in a positive dose-dependent association with the TrA concentration. Skin atrophy related to intralesional TrA for AA is mild and could be minimized by operational modifications on treatment frequency, injection depth, injection volume, and number of injection per site.[3],[11] The treatment frequency should be every 2 weeks at least, with a recommended range from 4 to 6 weeks.[3] A physician should inject TrA intradermally and avoid intraepidermal and deep subcutaneous injection. Intraepidermal injection forms a small superficial bump on the skin, generates higher resistance during injection, and causes greater pain to the patients. Deep subcutaneous injection, however, causes less pain with minimal resistance during injection but increases the risk of subcutaneous fat atrophy.[34] Injection volume <1 ml per injection at 1-cm intervals is recommended.[3],[35] Meticulous calculation of the alopecic surface area could further optimize the number of injections and prevent overdosage.[11] Most importantly, a physician should hold the therapy if skin atrophy or telangiectasia was observed. Local mild pain during injection was subjective and various among individuals, and could be alleviated by topical anesthetic[3] or assistant devices such as a multi-injection plate.[36]
This study has several limitations. First, the number of RCTs comparing the efficacy of different TrA concentrations[11],[15] may be too limited to reflect the true efficacy of the concentrations. The effects of demographics of the subjects, baseline condition of AA, and treatment protocols on the response rate were uncontrolled due to the inconsistency of the studies. The methodologies including the technique and treatment course were also highly various among institutions. Since most of the studies did not enroll a placebo control group, the chance of spontaneous remission was not taken into consideration. The spontaneous remission rate was approximately 78% in mild disease within 6 months,[37] the importance of which is not negligible. Although the included studies investigated mainly on scalp AA, a minor part of the subjects presented with facial AA, which usually required a lower dosage of TrA.[3] Since TrA is a slow-release steroid,[38] diffusion of TrA into adjacent partitions in intra-patient trials was possible with unknown confounding effects.[8],[9],[10],[11],[12],[13],[14] Last but not least, readers should be reminded that the efficacy and adverse events of the treatment are determined not only by the concentration of TrA but also by the aforementioned operational parameters, the importance of which could not be analyzed in the present review. Despite the high degrees of heterogeneity among studies in our meta-analysis, the strength of the present report was a greater number of included studies of the higher level of evidence than the previous review.[4] To further reinforce our finding, more high-quality RCTs directly comparing various TrA concentrations for AA with a large sample size may be required.
ConclusionConsidering both efficacy and mild adverse events, the optimal intralesional TrA concentration for patchy AA is probably 10 mg/dl in comparison to 5, 3.33, or 2.5 mg/dl. Higher concentration is associated with higher efficacy and skin atrophy. Operational modifications should be conducted for better treatment response with less adverse events.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
Supplementary Material
References
留言 (0)