FormalPara Key Summary Points

Why carry out this study?

Recent advancements in androgenetic alopecia (AGA) treatment have highlighted the efficacy of botulinum toxin (BoNT). However, inconsistencies in injection sites and depths warrant attention. It remains unclear which injection strategy is most beneficial for patients.

What was learned from this study?

The study compared intramuscular versus intradermal BoNT injection strategies, revealing both BoNT injection strategies enhanced hair diameter 12 weeks post-treatment, with intramuscular injections in surrounding muscles significantly improving hair density more effectively.

Pain evaluation revealed that intradermal injections in balding areas were slightly more painful than intramuscular injections in surrounding muscles. Although the overall safety profile of BoNT remained favorable, one instance of alopecia areata was noted.

There was variability in patient responses to BoNT treatment, indicating the need for identifying patients most likely to benefit from BoNT.

Introduction

Androgenetic alopecia (AGA) is the predominant cause of hair loss, affecting both men and women with a genetically determined sensitivity of hair follicles to androgens. As this condition progresses with age, it results in varying degrees of hair loss, significantly affecting an individual’s appearance and thereby impacting their self-esteem and sense of identity [1]. Despite the availability of US Food and Drug Administration (FDA)-approved treatments such as minoxidil, 5-alpha-reductase inhibitors, and low-level laser therapy (LLLT), in addition to various nonprescription products, patients often find these options limited, either because of unsatisfactory results or side effects [2, 3]. Furthermore, most treatment regimens require frequent interventions, which introduce significant inconvenience to patients’ lives [4].

Recent advancements in AGA treatment have highlighted the efficacy of botulinum toxin (BoNT). Since its initial FDA approval in 1989 for treating strabismus, research and application of BoNT have significantly expanded, leading to new formulations and a broader spectrum of uses [5]. This has cemented its role not only in neurology and cosmetic practices but also in emerging areas handled by urologists and pain specialists [6]. In the field of hair disorders, BoNT was first reported in 2010 for the treatment of cephalalgia alopecia and has since been widely adopted for managing androgenetic alopecia [7]. However, inconsistencies in injection sites and depths warrant attention. Some studies have demonstrated a preference for intradermal injections in areas affected by hair loss, leveraging BoNT’s ability to modulate the production of transforming growth factor-β1 (TGFβ1) by dermal papilla cells (DPC)—a critical factor in the pathogenesis of AGA [8,9,10,11]. Conversely, other approaches have utilized intramuscular or subcutaneous injections targeting periscapular muscles such as the frontalis, temporalis, and occipitalis [12,13,14,15,16]. This method capitalizes on BoNT’s capacity to relax these adjacent muscles, thereby enhancing scalp blood flow and oxygenation, ultimately contributing positively to the hair growth cycle [17].

The primary objective of this randomized controlled trial was to assess the efficacy and safety of two different BoNT injection methods in enhancing hair density and diameter in patients with AGA. The study also evaluates patient pain and treatment tolerability to determine the optimal injection strategy for clinical use.

Methods

Registration and Ethical Consent

This study was a single-center, split-scalp randomized controlled trial involving patients with AGA (ClinicalTrials.gov identifier ChiCTR2400080190). Ethical approval was obtained from the West China Hospital Institutional Review Board (#2023-1458) prior to the commencement of the study, and all patients provided both verbal and written consent for participation. The authors obtained the necessary permissions to publish the patient photographs displayed in the article. All research methods involving human subjects adhered to the ethical guidelines of institutional and national committees and conformed to the ethical principles of the 1964 Helsinki Declaration.

Patients and Methods

To ensure consistent BoNT dosing and equal injection points across muscular and balding regions, we exclusively enrolled adult patients with AGA, who experienced diffuse mid-frontal scalp hair thinning. Specifically, this included patients with a V1–V3 rating or an F1–F3 rating according to the basic and specific (BASP) classification system. The primary exclusion criteria included scalp abrasions or abnormalities; a history of significant diseases relevant to the study; the presence of malignant tumors; a history of BoNT allergy; periods of preconception, pregnancy, or lactation; and patients who have undergone other AGA treatments or anti-androgen therapy within the past 3 months.

The patients’ scalps were divided into left and right sides along the midline. Using an internet-based tool (random.org), a random number—either 1 or 2—was generated. If the resulting number was 1, patients received BoNT intramuscular injections in the surrounding muscle areas and saline intradermal injections in the balding areas on the right side. Simultaneously, they received saline intramuscular injections in the surrounding muscle areas and BoNT intradermal injections in the balding areas on the left side. Conversely, if the generated number was 2, the procedure was reversed, with patients receiving BoNT intradermal injections in the balding areas on the right side and saline intramuscular injections in the surrounding muscle areas, while the left side received saline intradermal injections in the balding areas and BoNT intramuscular injections in the surrounding muscle areas.

For the injection sites, we allocated 10 and 6 intramuscular injection points to the temporalis and occipitalis muscles, respectively, and 16 intradermal injection points in the balding regions on each side. We intentionally omitted the frontalis muscle from our injection plan to avoid the risk of inducing facial asymmetry (Fig. 1). The experimental drug utilized was BoNT (Botox®, Allergan, USA). The operator prepared the solution by diluting 100 U of BoNT powder in 2 ml of saline, resulting in a concentration of 0.5 U/μL. Each injection site was administered 3 units of BoNT or an equivalent volume of saline in accordance with the plan.

Fig. 1
figure 1

Schematic illustration of botulinum toxin and saline injection sites and depths. The diagram depicts a patient randomized to receive intramuscular injections on the left side and intradermal injections on the right. Specifically, the left temporalis and occipitalis muscles received botulinum toxin injections at 10 and 6 points, respectively, while the equivalent right-side areas were administered saline injections. In the balding area, 16 points on the right received botulinum toxin intradermally, with corresponding left-side points treated with saline intradermally. BoNT botulinum toxin

Full size image

Efficacy Assessment

Standardized macrophotography and trichoscopic examination are the cornerstone evaluation methods used in this research. To ensure accurate localization for follow-up trichoscopic assessments, two points were symmetrically tattooed within the balding area, each positioned 3 cm from the midline (Fig. 1). Participants underwent both standardized photography and trichoscopic evaluations at the study’s outset and again 12 weeks after initiating treatment. Macrophotographs were taken with a camera (Canon D200, Japan), and trichoscopic images were captured using the FotoFinder system (FotoFinder Systems GmbH, Germany). FotoFinder’s automated analysis provided detailed measurements of hair density, follicular density, average hair diameter, and the proportion of hairs with varying shaft diameters within the target area. The final statistical outcomes were derived following a review and verification of the FotoFinder results by two blinded dermatologists.

For both hair density and average hair diameter, we calculated the post-treatment relative change using the following formula: (follow-up data − baseline data)/baseline data. On the basis of the relative change, patients were categorized into four groups: the no improvement group (for a relative change ≤ 0), the mild improvement group (for 0 < relative change ≤ 15%), the moderate improvement group (for 15% < relative change ≤ 30%), and the significant improvement group (for a relative change > 30%).

Pain and Safety Assessment

Throughout the treatment process, the experimenters meticulously documented the pain reported by the participants using a visual analog scale (VAS), which spans from 0 (indicating no pain) to 10 (indicating very severe pain), immediately following the needle stick at various injection sites and depths. Treatment-emergent adverse events were thoroughly recorded. Investigators assessed their severity and determined the causal relationship with the study treatment.

Statistical Analysis

With the study designed for a two-sided alpha level of 0.05 and aiming for a statistical power over 80%, we initially required 25 patients. Considering an anticipated dropout rate of 20%, the total recruitment goal was set at 32 patients.

Differences within and between treatment groups were analyzed using paired t tests or Wilcoxon tests, based on whether the data conformed to a normal distribution, as assessed by the Shapiro–Wilk test. This approach was applied to compare baseline and follow-up data for within-group differences and pre- and post-treatment changes for between-group differences. The Mann–Whitney U test was used to detect differences in the proportions of various treatment response groups between the two experimental groups. A p value of less than 0.05 was considered to indicate statistical significance.

Results

From August to November 2023, 42 patients were screened for participation in this study. Concerns about potential pain and reluctance to commit to follow-up visits led to the exclusion of seven individuals. Of the 35 patients who were enrolled, six did not complete the follow-up as scheduled, leaving 29 patients who successfully completed the study (Fig. 2). The average age of these 29 participants was 33.2 years, and the participants included 3 women and 26 men. Detailed information on the age, gender, ethnicity, BASP classification, and treatment history prior to the 3-month washout period for each participant is displayed in the Table 1.

Fig. 2
figure 2

Flow diagram of patient enrollment and follow-up

Full size image
Table 1 Baseline characteristics of the 29 patients who completed the trial
Full size table

Paired t tests at 12 weeks showed that the intramuscular injection side showed a significant increase in hair density, from a baseline mean of 118.3 (standard deviation, SD 32.1) cm2 to 136.1 (SD 35.9) cm2 at 12 weeks (p < 0.001), compared to the intradermal injection side which also improved but not significantly (from 120.5 [SD 37.3] cm2 to 126.2 [SD 34.0] cm2, p = 0.217). This pattern was mirrored in the hair follicle unit density, with significant improvements observed on the intramuscular side (p < 0.001), suggesting that intramuscular BoNT injections might provide a more conducive environment for enhancing hair density in patients with AGA (Table 2, Fig. 3). However, both intramuscular and intradermal injections led to a notable increase in hair diameter, confirming the capacity of BoNT to positively influence hair thickness. However, the lack of a significant difference between the groups (p = 0.433) indicates that both injection routes are comparably effective in this regard (Table 2, Fig. 3).

Table 2 Comparison of hair parameters before and 12 weeks after BoNT injection, and the difference between the two injection strategies
Full size table
Fig. 3
figure 3

Global photographs and trichoscopic images were taken at baseline and 12 weeks after treatment in a patient with androgenetic alopecia (AGA), classified as M2F2 according to the basic and specific (BASP) classification system. BoNT botulinum toxin

Full size image

Although the two treatment groups exhibited differences in the distribution across the “no improvement,” “mild improvement,” “moderate improvement,” and “significant improvement” categories for both hair density and average hair diameter, the Mann–Whitney test confirmed that these variations were not statistically significant, with p values of 0.216 and 0.600, respectively (Fig. 4).

Fig. 4
figure 4

Distribution of treatment response categories 12 weeks post-treatment: no improvement group, mild improvement group, moderate improvement group, and significant improvement group. BoNT botulinum toxin

Full size image

In terms of injection pain, we analyzed data from all 35 patients enrolled at the start of the study. Although 7 patients reported that intradermal injections in the balding areas were more painful, on average, the VAS scores for intramuscular injections in the surrounding muscle areas (mean 4.24, SD 1.47) were significantly lower than those for intradermal injections in the balding areas (mean 4.76, SD 1.51), with a p value of 0.036. Throughout the clinical study, no participants reported adverse reactions such as hematomas, infections, or headaches. However, a 24-year-old male patient with no prior history of alopecia areata developed the condition in the right occipital region 3 weeks after receiving BoNT injections, which was successfully treated with topical corticosteroids.

Discussion

The role of BoNT in promoting hair growth is complex and multifaceted, uncovering a range of mechanisms that underscore its therapeutic potential [18]. The exploration of BoNT as a treatment for hair loss initially centered around its well-documented ability to block acetylcholine release at neuromuscular junctions, a mechanism crucial for inducing muscle relaxation. Kato et al. revealed that local ischemia can lead to a reduction in hair shaft diameter, delayed hair shaft elongation, decreased hair density, and premature termination of the anagen phase [17]. BoNT offers a promising solution due to its capacity to relax specific scalp muscles. This relaxation targets muscles such as the frontalis, occipitalis, temporalis, and periauricular regions, alleviating pressure on blood vessels and thereby enhancing blood flow and oxygenation to the scalp. By improving scalp blood flow and oxygenation, BoNT helps create a healthier scalp environment. This improvement is likely to reduce tissue levels of dihydrotestosterone (DHT) through a mechanism known as the washout effect, subsequently reducing follicular miniaturization, which is considered the primary pathophysiological basis of this condition [19].

Besides its known muscle relaxation effects, recent research has further illuminated the ability of BoNT to impact hair follicles. It regulates the secretion of proinflammatory cytokines, such as tumor necrosis factor alpha (TNFα) and interleukins (ILs), reducing scalp inflammation and thereby creating a nurturing environment for hair follicles [7, 20]. Additionally, a critical aspect of BoNT’s therapeutic action is its ability to target the TGFβ1 pathway. This pathway, activated by DHT within DPC, typically inhibits the proliferation of follicular epithelial cells [11, 21]. By suppressing TGFβ1 secretion, BoNT effectively counters one of the principal pathophysiological processes contributing to AGA, alleviating the inhibition of follicular cell growth [8]. Furthermore, contemporary research has shed light on the role of BoNT in mitigating DHT-induced DPC dysfunction [22]. By downregulating the expression of circ_0135062, BoNT modulates miR-506-3p-targeted Bax levels, reducing apoptosis in DHT-treated DPCs [22].

BoNT has emerged as a promising candidate for AGA treatment. However, the application of BoNT in AGA treatment varies widely, and a consensus on injection protocols including sites, depths, dosages, and frequency is lacking. The most significant debates center on the optimal locations and depths for injections. To address these uncertainties, our study adopted a split-scalp randomized controlled trial design to directly compare the efficacy of BoNT injections in the surrounding muscle areas (intramuscular) to that of direct injections in the balding zones (intradermal) for AGA management. Our findings significantly contribute to this debate, showing that both approaches to BoNT injection significantly increase hair diameter 12 weeks post-treatment. Notably, intramuscular injections were more effective at improving hair density, consistent with BoNT’s recognized effects on neuromuscular blockade. Importantly, our findings should not diminish the observed efficacy of intradermal injections, especially considering that Shon et al.’s study demonstrated significant improvements in hair density at a 24-week follow-up [8]. The safety profile of BoNT, as echoed in our findings, remains favorable, supporting its potential as a low-risk option for individuals seeking alternative AGA treatments.

Our investigation of BoNT for AGA treatment highlights two critical areas for further exploration. Initially, the idea of integrating both intramuscular and intradermal injections raises possibilities for synergistic effects on hair regrowth. However, replicating the BoNT dosage used in our study, where the integration of intramuscular and intradermal injections necessitates a cumulative BoNT dose of 192 units, presents possible synergistic advantages for hair regeneration but concurrently introduces significant concerns related to safety and risk management [23]. Interestingly, research indicates that low-dose BoNT injections significantly benefit AGA in the Chinese population, highlighting the necessity for thorough future research to confirm the effectiveness and safety of this combined strategy [13]. Moreover, our findings revealed notable variability in patient responses to BoNT treatment for AGA. While BoNT generally shows effectiveness, especially with intramuscular injections that significantly enhance hair density, a specific group of patients—six in our study—did not exhibit any improvement in hair density on the BoNT intramuscular injection side. Additionally, despite significant improvements in hair diameter with both injection approaches, a portion of our study participants (seven and nine sides which received intramuscular and intradermal injections, respectively) did not experience benefits. This variability in AGA treatment response highlights the challenge of identifying patients most likely to benefit from BoNT. Understanding the factors influencing treatment outcomes is crucial, and could lead to more targeted and effective strategies, for maximizing the benefits of BoNT.

Compared to earlier studies that have primarily confirmed the efficacy and safety of BoNT in treating AGA, our research not only compares the two predominant methods of BoNT injections but also reaffirms that both methods can benefit patients with AGA. Importantly, our findings indicate that intramuscular injections significantly enhance hair density more effectively than intradermal methods. However, this study has several limitations that merit consideration. The primary limitation is the relatively short follow-up period of 3 months, which may not fully capture the long-term efficacy and safety of BoNT injections for patients with AGA. Additionally, the study’s sample size, while adequate for initial comparisons, limits the generalizability of the findings across diverse populations and different degrees of hair loss severity. Future studies should consider incorporating larger sample sizes and multicenter trials to enhance the robustness and applicability of the results.

Conclusion

Our research revealed that both BoNT injection strategies enhanced hair diameter 12 weeks post-treatment, with intramuscular injections significantly improving hair density more effectively. Despite the promising outcomes, the variability in patient responses underscores the necessity for personalized approaches and further research to refine injection protocols for optimized efficacy and safety.