Is Botox Really Safe? The Answer from Meta-Analysis Data on 9,669 People

1. What is Botox?

Botulinum toxin type A is a medical and cosmetic intervention widely used for the treatment of facial wrinkles (Cochrane 2021). It is mainly used to improve glabellar lines, forehead lines, and crow's feet lines (M. Gostimir 2023).

The types of products whose efficacy and safety have been evaluated include OnabotulinumtoxinA, AbobotulinumtoxinA, IncobotulinumtoxinA, DaxibotulinumtoxinA, PrabotulinumtoxinA, HBTX-A, CBFC26, Neuronox, and Liquid AbobotulinumtoxinA. These are each administered in different units (U), and their efficacy and safety profiles compared with placebo (a dummy drug) have been studied in detail (Cochrane 2021).

2. The safety data of the meta-analysis

In a comprehensive meta-analysis on the safety of Botox injection, it was confirmed that the risk from systemic spread is extremely low, or does not exist (M. Gostimir 2023). As a result of the influence analysis, the overall pooled risk ratio was 1.51 (M. Gostimir 2023), with a 95% confidence interval of 1.32 to 1.72 (M. Gostimir 2023). The I-squared statistic at this time was 27.2%, indicating that the heterogeneity between studies is relatively low (M. Gostimir 2023).

Major past meta-analyses also back up this safety. For example, a study published in 2009 analyzed 11 papers and 1,063 participants (age range 31 to 59 years), and the overall quality of evidence was assessed as good (Ghadia 2009). Similarly, a meta-analysis of a global clinical registration study using individual data of 1,678 participants (Brin 2009), and other systematic reviews on facial rejuvenation (Guo 2015) (Jia 2016) (Naumann 2004), also reported the high tolerability and safety of botulinum toxin type A. Additionally, the latest Cochrane review set the target age broadly at 18 to 65 years and assessed the certainty of evidence in the range of "very low" to "moderate" (Cochrane 2021).

3. The types and incidence of side effects

The frequency of side effects is, on the whole, similar between the botulinum toxin type A group and the placebo group, but only the incidence ofeyelid ptosis(blepharoptosis) tends to be significantly higher than in the placebo group (Cochrane 2021). In most cases, side effects are mild to moderate and transient, consistent with the expected pharmacodynamics (M. Gostimir 2023).

• Major adverse events: eyelid ptosis, sensory disturbance of the eyelid, strabismus, and so on have been evaluated, but the occurrence of serious events is extremely rare. No major serious side effects were observed with the 20U administration of IncobotulinumtoxinA (incidence 0) (Cochrane 2021).
• With the 20U administration of IncobotulinumtoxinA, the general side-effect risk is estimated at 37 per 100 people (95% confidence interval: 7 to 100 people). By contrast, the risk in the placebo group was 1 per 100 people (Cochrane 2021).
• In the 45U administration group of AbobotulinumtoxinA, the frequency of side effects is reported as 5 out of 40 people (13%) (Cochrane 2021).
• In the analysis using OnabotulinumtoxinA 20U (Carruthers 2014) (Harii 2017) (Moers-Carpi 2015) (Wu 2019), the RR of adverse events after 4 weeks was 12.38 (95% confidence interval 8.93 to 17.16). This risk ratio decreased over time, and after 16 weeks the RR dropped to 5.46 (95% confidence interval 3.19 to 9.32) (Cochrane 2021).
• In the analysis using PrabotulinumtoxinA 20U, among 948 participants (737 in the treatment group, 211 in the placebo group), the overall side-effect RR was 1.14 (95% confidence interval 0.91 to 1.43) (Beer 2019a) (Beer 2019b) (Rzany 2019).
• In the analysis of Liquid AbobotulinumtoxinA 50U, the overall side-effect RR among 255 participants was 1.11 (95% confidence interval 0.72 to 1.71) (Ascher 2018) (Ascher 2020).

📊 A comparison of side-effect risk by botulinum toxin type A product (Cochrane 2021 / 9,669-person meta-analysis)

Source: Cochrane Database of Systematic Reviews 2021 (9,669-person analysis) / RR = risk ratio (vs. placebo). The lower the value, the lower the side-effect risk.

4. The evidence of the Cochrane review

The 2021 Cochrane database systematic review rigorously evaluated the efficacy and safety of botulinum toxin type A for facial wrinkles (Cochrane 2021).

• The duration of therapeutic effect:The mean duration of therapeutic effect was 99.7 days, and it was shown that the mean difference (MD) compared with placebo was 17.3 days longer (95% confidence interval: 15.82 days longer to 18.78 days longer) (Cochrane 2021).
• Assessment of risk of bias:The quality of the studies varies widely. For example, a study that clearly stated that the vials were identical, that they were identified only by patient number and study number, and that the dilution and injection procedures were identical, was judged to have a low risk of selection bias (Carruthers 2004). A study that used the SAS random-number generation function was also assessed as low risk (Harii 2008). On the other hand, a study that reported only the results after 4 weeks was judged to have a high selective reporting bias (Lee 2013), and a study whose randomization method was unknown was listed as risk unclear (Park 2014). Also, a study that reported all prespecified outcomes was considered low risk (Satler 2010).

5. The satisfaction data for the forehead and glabella

The treatment success for wrinkles of the upper face (glabella, forehead, crow's feet) has been highly evaluated through analysis of physician and participant scores and scales (Cochrane 2021). The following data show the risk ratio (RR) of success compared with placebo and the 95% confidence interval.

Data for OnabotulinumtoxinA

The participant-assessed success rate with 44U administration of OnabotulinumtoxinA (Carruthers 2015) (Moers-Carpi 2015):

• After 4 weeks: the RR based on 2 studies (808 participants) is 11.09 (95% confidence interval 4.12 to 29.83).
• After 8 weeks: RR 9.94 (95% confidence interval 1.78 to 55.44).
• After 12 weeks: RR 5.96 (95% confidence interval 0.60 to 58.98).

The participant-assessed success rate with 64U administration of OnabotulinumtoxinA (De Boulle 2018):

• After 4 weeks: RR 8.32 (95% confidence interval 4.53 to 15.30) (469 participants).
• After 8 weeks: RR 106.50 (95% confidence interval 6.66 to 1702.48).
• After 12 weeks: RR 44.50 (95% confidence interval 2.76 to 717.83).
• After 16 weeks: RR 26.50 (95% confidence interval 1.63 to 431.99).
• After 20 weeks: RR 5.50 (95% confidence interval 0.31 to 98.84).
• After 24 weeks: RR 1.50 (95% confidence interval 0.06 to 36.61).

The success rate with 12U administration of OnabotulinumtoxinA (Harii 2017):

• After 4 weeks: RR 6.86 (95% confidence interval 3.45 to 13.62).
• After 8 weeks: RR 4.29 (95% confidence interval 2.10 to 8.76).
• After 12 weeks: RR 5.55 (95% confidence interval 1.68 to 18.34).

Data for AbobotulinumtoxinA

The physician-assessed success rate with 50U administration of AbobotulinumtoxinA:

• After 4 weeks: the RR based on 5 studies (915 participants) is 21.22 (95% confidence interval 7.43 to 60.56) (Ascher 2018) (Brandt 2009) (Kane 2009) (Monheit 2019) (NCT02450526).
• After 8 weeks: RR 39.89 (95% confidence interval 14.10 to 112.88) (725 participants) (Ascher 2018) (Monheit 2019) (NCT02450526).
• After 12 weeks: RR 28.83 (95% confidence interval 10.16 to 81.81) (725 participants) (Ascher 2018) (Monheit 2019) (NCT02450526).
• After 16 weeks: RR 11.78 (95% confidence interval 4.12 to 33.66) (371 participants) (Ascher 2018) (Monheit 2019).
• After 20 weeks: RR 5.33 (95% confidence interval 1.67 to 16.99) (300 participants) (Monheit 2019).

The success rate with 50u administration of Liquid AbobotulinumtoxinA (Ascher 2018) (Ascher 2020):

• After 4 weeks: RR 16.73 (95% confidence interval 2.84 to 98.58) (255 participants).
• After 8 weeks: RR 48.98 (95% confidence interval 9.99 to 240.21).
• After 12 weeks: RR 35.93 (95% confidence interval 7.30 to 176.90).
• After 16 weeks: RR 21.25 (95% confidence interval 2.95 to 152.88).
• After 20 weeks: RR 25.86 (95% confidence interval 1.60 to 417.34).

The success rate with 75u administration of Liquid AbobotulinumtoxinA (Ascher 2018):

• After 4 weeks: RR 64.75 (95% confidence interval 4.12 to 1018.73).
• After 8 weeks: RR 60.64 (95% confidence interval 3.85 to 955.48).
• After 12 weeks: RR 56.53 (95% confidence interval 3.58 to 892.24).
• After 16 weeks: RR 42.14 (95% confidence interval 2.65 to 670.89).

Data for DaxibotulinumtoxinA

The physician-assessed success rate with 40U administration of DaxibotulinumtoxinA:

• After 4 weeks: RR 21.10 (95% confidence interval 11.31 to 39.34) (683 participants) (Bertucci 2020) (Carruthers 2017).
• After 8 weeks: RR 15.75 (95% confidence interval 8.85 to 28.03) (609 participants) (Bertucci 2020).
• After 12 weeks: RR 24.51 (95% confidence interval 11.12 to 54.05) (609 participants) (Bertucci 2020).
• After 16 weeks: RR 12.74 (95% confidence interval 6.80 to 23.89) (683 participants) (Bertucci 2020) (Carruthers 2017).
• After 20 weeks: RR 13.60 (95% confidence interval 6.13 to 30.18) (609 participants) (Bertucci 2020).
• After 24 weeks: RR 10.25 (95% confidence interval 4.01 to 26.20) (683 participants) (Bertucci 2020) (Carruthers 2017).
• Regarding the duration of therapeutic effect, the mean difference reached 22.80 weeks (95% confidence interval 20.74 to 24.86) (Carruthers 2017).

Data for other products

• With 54U to 64U administration of IncobotulinumtoxinA, the success RR for the glabella after 4 weeks was 87.81 (95% confidence interval 5.56 to 1386.93), the success RR for the forehead was 35.94 (95% confidence interval 5.14 to 251.27), and the success RR for the crow's feet was 32.54 (95% confidence interval 4.65 to 227.82) (Kerscher 2015).
• With 50u administration of HBTX-A, the RR after 4 weeks was 42.68 (95% confidence interval 6.08 to 299.41) (NCT02493946).

6. Points for minimizing risk

To keep the risk in Botox treatment to a minimum, there are several important points.

• Appropriate management of dose:The effect on safety strongly depends on the individual injection volume and the total volume injected per treatment session (M. Gostimir 2023).
• Use of validated assessment tools:It is strongly recommended to incorporate assessment based on objective indices, such as a validated assessment scale for neck volume (Sattler 2012).
• Adherence to international consensus:It is important to follow the international consensus recommendations on the use of botulinum toxin type A for cosmetic purposes, published in 2010 (Ascher 2010), and the evidence-based review and recommendations on complications, updated in 2016 (Sundaram 2016).

If you find yourself thinking, "What about my case?"

Book a free consultation

There is no pushy sales. Please feel free to get in touch.

7. Summary

Botulinum toxin type A is a treatment that shows extremely high efficacy for facial wrinkles such as the glabella, forehead, and crow's feet. As multiple large-scale meta-analyses show, there is almost no concern of serious side effects from systemic spread, the majority of side effects that occur are mild and transient, and they are due to the pharmacodynamic properties of the product (M. Gostimir 2023). Although local risks such as eyelid ptosis are significantly higher compared with the placebo group, that risk is considered sufficiently manageable through appropriate dose management and procedures based on international guidelines (Ascher 2010) (Sundaram 2016). The duration of effect is also stable over the long term at an average of about 100 days (about 3 to 4 months), and very high satisfaction and success rates from both patients and physicians have been confirmed in many clinical trials (Cochrane 2021).

References

1. Gostimir M, Liou V, Yoon M. Safety of Botulinum Toxin A Injections for Facial Rejuvenation: A Meta-Analysis of 9,669 Patients. Ophthalmic Plastic & Reconstructive Surgery. 2023 DOI
2. Camargo C, Xia J, Costa C, et al. Botulinum toxin type A for facial wrinkles. Cochrane Database of Systematic Reviews. 2021 DOI
3. Ogilvie P, Rivkin A, Dayan S, et al. OnabotulinumtoxinA for Treatment of Forehead and Glabellar Lines: Subject-Reported Satisfaction and Impact From a Phase 3 Double-Blind Study. Dermatologic Surgery. 2019 DOI
4. Ozdemir Cetinkaya P, Karaosmanoglu N, Özkesici Kurt B, et al. Functional and esthetic effects of botulinum toxin injection into the masseter muscles: evaluation of 80 patients from a dermatological perspective. International Journal of Dermatology. 2025 DOI

Author of this article

Hiromitsu Nakamura Physician

Zetith Beauty Clinic Ginza, Osaka, Fukuoka

He has a track record of research presentations at domestic and international academic conferences, and is also engaged in technical guidance and education for Zetith Beauty Clinic as a whole. He specializes in precise aesthetic medicine based on anatomical grounds, pursuing natural results tailored to each person's bone structure and tissue.