Jie Su1,2  Xiaoyan Ren2*  Xiaoying Lyu1*

Institute of Clinical Basic Medicine of Traditional Chinese Medicine,

China Academy of Chinese Medical Sciences, Beijing 100700, China

Swiss University of Traditional Chinese Medicine, Bad Zurzach 5330, Switzerland

Abstract

Objective: To systematically evaluate the efficacy and safety of acupuncture for facial skin aging-related manifestations and to summarize the characteristics of currently available randomized controlled trials (RCTs). Methods: Eight Chinese and English databases, including CNKI, VIP, Wanfang, SinoMed, PubMed, Cochrane Library, Web of Science, and Embase, were searched from inception to June 2025. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the Cochrane risk-of-bias tool. Meta-analysis was conducted using R 4.4.1, and effect sizes were expressed as relative risk (RR) with 95% confidence intervals (CI). Results: Twelve RCTs involving 1109 participants were included. Meta-analysis suggested that acupuncture may improve overall response rate compared with injection filling techniques (RR=1.27, 95%CI: 1.15-1.40), facial care (RR=1.08, 95%CI: 1.00-1.17), and sham acupuncture (RR=2.32, 95%CI: 1.37-3.91). Acupuncture combined with laser/photon therapy may improve patient satisfaction. Reported adverse events were generally mild and transient, mainly including pruritus, ecchymosis, and bruising. Conclusion: Current evidence suggests that acupuncture may offer certain clinical benefits for facial skin aging-related manifestations. However, the overall certainty of the evidence remains limited because of small sample sizes, considerable heterogeneity in interventions and outcome assessments, and generally suboptimal methodological quality of the included studies. More rigorously designed high-quality RCTs are needed.

Keywords: Acupuncture; Facial skin aging; Wrinkles; Systematic review; Meta-analysis; Randomized controlled trial

Skin aging is a gradual deterioration of skin morphology, structure, and function driven by both intrinsic aging and extrinsic factors. It often presents as dry and rough skin, reduced elasticity, dull complexion, laxity and thinning, as well as wrinkles, hyperpigmentation, and telangiectasia [1]. Among these, facial wrinkles are one of the most intuitive and concerning clinical manifestations of skin aging. With increasing public demand for facial rejuvenation, interventions such as laser and light-based therapies, injectable fillers, surgical procedures, and daily skincare have been widely used. However, some methods still have limitations including high cost, prolonged recovery time, limited duration of effect, or adverse reactions.

In traditional Chinese medicine (TCM) literature, manifestations related to facial skin aging are often categorized under terms such as “面始焦”(facial withering), “皮槁” (dry and withered skin), and “面色黄”(yellowish complexion)[2]. Acupuncture, as an important component of TCM external therapies, has received sustained attention for improving localized wrinkles, regulating overall body status, and promoting recovery [3–6]. Although previous reviews have summarized the anti-aging effects of acupuncture [7], there remains a relative lack of systematic evaluations based on randomized controlled trial (RCT) evidence specifically targeting facial skin aging. Therefore, this study conducts a systematic review and meta-analysis of existing clinical RCTs, aiming to provide a reference for the clinical application and future research of acupuncture in facial skin aging.

1. Materials and Methods

1.1 Data Sources and Search Methods

(1) Databases searched

This study searched the China National Knowledge Infrastructure (CNKI), VIP, Wanfang, SinoMed, PubMed, Cochrane Library, Web of Science, and Embase databases for studies on acupuncture treatment for skin aging and facial wrinkles.

(2) Search strategy

The Chinese database search formula was: Topic=(“acupuncture” OR “needling”) AND [Topic=(“aging” OR “senescence” OR “beauty” OR “wrinkle removal” OR “wrinkles”) AND All=(“skin” OR “face”) AND All=(“randomized”)]. The English database search formula was: (acupuncture [Title / Abstract]) AND ((skin aging[Title/Abstract]) OR (photoaging[Title/Abstract]) OR (wrinkle[Title / Abstract]) OR (facial rejuvenation[Title/Abstract])) AND (randomized controlled trial[Title/Abstract] OR random[Title/Abstract]). Journal articles, dissertations, and conference papers were also searched, supplemented by manual searching of reference lists.

(3) Search period

From database inception to June 2025.

1.2 Inclusion and Exclusion Criteria

(1) Inclusion criteria

① Study subjects were individuals seeking aesthetic improvement due to skin aging.

② Study type was a randomized controlled trial (RCT) published in Chinese or English.

③ The intervention in the experimental group was any form of acupuncture or acupuncture combined with other non-acupuncture measures; the control group intervention could be blank, non-acupuncture measures, or acupuncture treatment.

④ The study reported the relative risk (RR) and its 95% confidence interval (CI) for different treatment groups, or the data allowed calculation of these values.

⑤ For studies using the same dataset, the analysis with the largest sample size, longest follow-up, and most adjustments for confounding factors was included.

(2) Exclusion criteria

① Reviews, systematic reviews, or meta-analyses.

② Studies for which the full text could not be obtained.

③ Studies with missing data that could not be obtained from the authors.

④ Study subjects were individuals seeking aesthetic improvement due to trauma or disease.

⑤ Studies where both the experimental and control groups received acupuncture, but the research objective was to explore the efficacy of non-acupuncture measures.

1.3 Study Selection and Data Extraction

(1) Study selection

Records retrieved from each database were imported into NoteExpress reference management software. Duplicates were removed. Following the above inclusion and exclusion criteria, an initial screening was performed by reading the titles and abstracts. Subsequently, full texts of potentially eligible studies were downloaded and reviewed to confirm inclusion. Finally, the included studies underwent quality assessment and data extraction. Study selection was conducted independently by two researchers, and disagreements were resolved through discussion with a third researcher.

(2) Data extraction

For studies finally included in the analysis, a pre-designed data extraction form was used to collect the following information: first author, publication year, study design type, study subjects, sex, age, sample size, number of responders, and response rate. Data extraction was performed independently by two researchers, and the results were cross-checked. Disagreements were resolved through discussion with a third researcher.

(3) Outcome measures

Relevant studies lacked unified efficacy evaluation criteria, and substantial differences existed in wrinkle scoring, grading tools, and reporting methods; thus, continuous outcomes were not suitable for direct pooling. Therefore, this study used the most frequently reported clinical response rate in primary studies as the primary outcome measure, satisfaction as the secondary outcome measure, and also recorded adverse events. The response rate was defined as (number of responders / total number of cases) × 100%. “Cured, markedly effective, effective” or “excellent, good, moderate / acceptable” as reported in the literature were classified as “responded”, while “ineffective” or “poor” were classified as “non-responded”. Satisfaction was defined as (number of satisfied cases / total number of cases) × 100%; all levels other than “dissatisfied” in the original studies were classified as “satisfied”. If a study reported multiple follow-up time points, data from the first post-treatment visit were included for analysis.

1.4 Quality Assessment

This study used the Cochrane Collaboration’s risk of bias tool [8] to assess the methodological quality of the included studies. The assessment included random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting, and other biases. The above assessments were conducted independently by two researchers, and disagreements were resolved through discussion or by consulting a third researcher. Each item was judged as “low risk”, “high risk”, or “unclear risk”.

1.5 Statistical Analysis

RR was used as the effect measure for response rate and satisfaction. Statistical heterogeneity between studies was assessed using I², and the effect model was selected in combination with clinical heterogeneity: a fixed-effect model was used when heterogeneity was low and clinical characteristics were similar; a random-effects model was used when significant heterogeneity was present; when heterogeneity was too high and could not be reasonably explained, effect sizes were not pooled, and only qualitative description was provided.

2. Results

2.1 Literature Search Results

A total of 537 articles were retrieved from the Chinese and English databases, and 3 articles were identified from reference lists. After removing duplicates, 460 articles remained. After reading titles and abstracts, 437 articles were excluded, leaving 23 articles that potentially met the inclusion criteria. Full texts of these 23 articles were downloaded and reviewed. We excluded 6 articles where the control group received acupuncture, 1 article involving patients without skin aging, 1 article where the intervention was not acupuncture, 1 article where the study objective was to evaluate non-acupuncture efficacy, 1 article without response rate/satisfaction outcomes, and 1 article without a control group. Ultimately, 12 RCTs were included in the meta-analysis. The literature screening flow diagram is shown in Figure 1.

Figure 1 Literature Screening Flowchart

2.2 Basic Characteristics of Included Studies

The 12 studies were published between 2011 and 2025, with a relatively even distribution across years. The total sample size was 1,109 cases, including 65 males and 1,044 females, with an age range of 24–59 years. The basic characteristics of the included studies are shown in Tables 1 and 2.

Table 1 Basic characteristics of the included studies 

Table 2 Basic characteristics of the included studies

Note: ① – Effective rate: if multiple follow up data are available, the first follow up data after treatment are included; ② – Satisfaction rate: if multiple follow up data are available, the first follow up data after treatment are included.

Acupoint representation: pinyin of the acupoint (code). The codes are based on the WHO Standard International Acupuncture Nomenclature officially issued by the World Health Organization (WHO) in 1991

2.3 Methodological Quality Assessment of Included Studies

A total of 12 randomized controlled trials were included. As shown in Figure 2, two studies[11,17] used SPSS to generate the random sequence, two studies[10,12] used a random number table method, and one study[20] used an online randomization system; these five studies were rated as having “low risk” for random sequence generation. One study[19] used odd-even patient numbers for randomization and was rated as “high risk”. The remaining six studies did not specify the method of random sequence generation and were rated as having “unclear risk of bias”. For allocation concealment, two studies [11,20] used sealed envelopes and were rated as “low risk”; one study[17] used envelopes but did not specify whether they were opaque; the other nine studies did not report allocation concealment, and these ten studies were rated as having “unclear risk of bias”. Regarding blinding, one study[17] explicitly blinded both participants and outcome assessors, and thus measurement bias and performance bias were rated as “low risk”. One study[20] stated that blinding of participants and personnel was not feasible but that outcome assessors were blinded; measurement bias was rated as “low risk” and performance bias as “high risk”. The other ten studies did not mention blinding, and blinding appeared unlikely to have been implemented; these were rated as “high risk”. Five studies[9,11,17,19,20] reported dropout data and reasons for dropout, with dropout rates below 20%; these were rated as having “low risk” of incomplete outcome bias. The remaining studies had the same number of participants randomized and included in the final analysis and did not describe whether any dropouts occurred; these were rated as having “unclear risk of bias” for incomplete outcome data. All studies reported clinically important outcomes, and selective reporting bias was assessed based on whether such outcomes were reported; all 12 studies were rated as having “low risk” for selective reporting bias. Two studies[13,14] did not report baseline comparability and lacked clear inclusion/exclusion criteria; these were rated as having “high risk” for other bias. Two studies[18,19] reported baseline comparability but lacked clear inclusion / exclusion criteria; these were rated as having “unclear risk of bias” for other bias. The remaining eight studies had comparable baseline characteristics between groups and clearly defined inclusion/exclusion criteria, and were rated as having “low risk” for other bias.

Figure 2 Risk of bias assessment of randomized controlled trials

2.4.2 Satisfaction Analysis

A total of six studies reported satisfaction. The forest plot for satisfaction is shown in Figure 4.

(1) Acupuncture vs. injectable fillers

One study [9] reported this outcome, with an RR (95% CI) of 1.13 (0.98, 1.31), which does not allow a clear conclusion that acupuncture leads to higher satisfaction than injectable fillers.

(2) Acupuncture vs. facial skincare

Two studies [10,14] reported this outcome. The heterogeneity test showed I² = 10.1%, and a fixed-effect model was used. The pooled RR (95% CI) was 1.25 (1.12, 1.40), suggesting that acupuncture may result in higher satisfaction than facial skincare.

(3) Acupuncture vs. laser and intense pulsed light (IPL) techniques

Two studies [12,16] reported this outcome. The heterogeneity test showed I² = 17.7%, and a fixed-effect model was used. The pooled RR (95% CI) was 1.14 (1.05, 1.24), suggesting that acupuncture may result in higher satisfaction than laser and IPL techniques.

(4) Acupuncture vs. blank control

One study [20] reported this outcome, with an RR (95% CI) of 53.00 (3.35, 837.36), which does not allow a clear conclusion that acupuncture leads to higher satisfaction than blank control.

2.4.3 Safety Analysis

Among the included studies, seven studies [9-11,16-18,20] reported this outcome. The adverse events were primarily mild itching, bruising, and the like, which gradually resolved without treatment or after interventions such as warm compresses.

Figure 4 Forest plot of satisfaction rate

3. Discussion

Skin aging, particularly the formation of facial wrinkles, involves multiple factors including the degradation of collagen and elastic fibers, oxidative stress, chronic low-grade inflammation, and alterations in skin support structures [1]. This study systematically reviewed the evidence from randomized controlled trials on acupuncture for facial skin aging. The overall results suggest that, under certain comparator conditions, acupuncture may improve clinical response rates or patient satisfaction, and the reported adverse events were mostly mild and transient. This indicates that acupuncture may have potential application value in facial rejuvenation interventions. However, the interpretation of the existing evidence should remain within strict evidentiary boundaries.

3.1 Advantageous Scenarios, Synergistic Potential, and Uncertainties of Anti-aging Acupuncture

This study showed that acupuncture may be associated with higher clinical response rates when compared with injectable fillers, facial skincare, and sham acupuncture, and may improve satisfaction when combined with laser/intense pulsed light (IPL) techniques. It should be noted that different comparator types address different clinical questions, and the results of the subgroups should not be simply compared horizontally to conclude that “acupuncture is superior to other treatments overall.” In particular, the number of studies using blank or sham acupuncture controls is small, and some effect estimates have wide confidence intervals, suggesting that the impact of small sample sizes and bias cannot be ignored. When directly compared with laser/IPL techniques, the current evidence is insufficient to demonstrate a clear advantage of acupuncture. Therefore, a more reasonable interpretation is that acupuncture may have complementary value in specific populations and specific scenarios, especially in combination therapies and recovery support.

3.2.1 Theoretical Basis in Traditional Chinese Medicine for Acupoint Selection Principles

The acupoint selection protocols used in the included studies show certain commonalities, suggesting that the most common approach in current clinical practice is a combination of local acupoints and distal regulatory points. Locally, points are often selected from areas around the wrinkles, such as Ashi points, Yintang (EX-HN3), Cuanzhu (BL2), Sizhukong (TE23), Taiyang (EX-HN5), and Sibai (ST2). Distally, points such as Zusanli (ST36), Sanyinjiao (SP6), Hegu (LI4), Taixi (KI3), and Guanyuan (CV4) are frequently used, aiming to achieve both local improvement and systemic regulation.

From the perspective of TCM theory, this acupoint selection pattern is often associated with concepts such as “spleen and kidney deficiency,” “qi and blood disharmony,” and “stagnation in the collateral vessels.” That is, while improving local facial manifestations, attention is also given to nourishing qi and blood and regulating visceral functions. However, it should be emphasized that this study only summarizes the frequency and characteristics of acupoint use based on the included RCTs, which is insufficient to establish a standardized acupoint selection model or to directly interpret these patterns as a treatment standard supported by high-level evidence.

Regarding compatibility characteristics, the combination of local facial acupoints with distal tonic points is a relatively common pattern, which may reflect a clinical approach that emphasizes both “local dredging” and “systemic regulation.” Combinations such as “Zusanli + Sanyinjiao,” “Hegu + Taichong,” and “Taixi + Shenshu” appear frequently in the relevant literature; however, their synergistic effects still need to be further verified in more rigorously designed clinical studies.

Therefore, regarding the acupoint selection rules for anti-facial-aging acupuncture, it is currently more appropriate to describe that “certain commonalities and empirical trends exist” rather than stating that a unified, mature, and directly generalizable standard protocol has been established.

3.2.2 Associations with Potential Mechanisms of Action from a Modern Medicine Perspective

Existing basic research and reviews suggest that acupuncture-related interventions may exert effects by improving local microcirculation, regulating inflammatory responses, influencing tissue repair, and participating in the neuro-endocrine-immune network regulation [23-26]. However, the studies included in this review are primarily clinical efficacy studies and did not directly evaluate molecular mechanisms or changes in objective biomarkers. Therefore, explanations involving promotion of collagen synthesis, modulation of fascial tension, or improvement of the systemic internal environment are currently more suitable as mechanistic hypotheses rather than as conclusions directly confirmed by this study.

4. Conclusion

In summary, the existing randomized controlled trial evidence suggests that acupuncture may offer certain clinical benefits in improving manifestations associated with facial skin aging, particularly by enhancing patient satisfaction in some combined treatment regimens. At the same time, the included studies commonly have limitations such as small sample sizes, inadequate reporting of randomization and blinding, lack of standardized intervention protocols and outcome measures, and limited follow-up durations. Therefore, the current evidence is insufficient to form a clear and robust clinical recommendation. Future high-quality randomized controlled trials with adequate sample sizes, appropriate comparator settings, objective and standardized outcome measures, and medium-to-long-term follow-up are needed to further verify the efficacy and safety of acupuncture for facial skin aging.
References

[1] HE L. Cosmetic dermatology[M]. 2nd ed. Beijing: People’s Medical Publishing House, 2011: 12. [In Chinese]

[2] OU Denggeerli, ZHOU G. Discussion on pathogenesis and syndrome characteristics of female skin aging[J]. Xinjiang Journal of Traditional Chinese Medicine, 2009, 27(2): 30-31. [In Chinese]

[3] XUE Kaiyang, CUI Jin. Current status analysis of acupuncture for wrinkle removal[J]. Journal of Guizhou University of Traditional Chinese Medicine, 2021, 43(6): 74-78. [In Chinese]

[4] LI Hongrong, MA Jing, YIN Yujie, et al. Research progress on anti-aging effects of characteristic therapies of traditional Chinese medicine[J]. World Chinese Medicine, 2023, 18(24): 3608-3615. [In Chinese]

[5] ZHONG Li, AI Bingwei, XIA Yi. Clinical study on anti-free radical and anti-aging effects of acupuncture at Zusanli and Guanyuan points[J]. Jiangsu Journal of Traditional Chinese Medicine, 2015, 47(6): 56-58. [In Chinese]

[6] LI Yanmei. Effects of warm acupuncture at different time on superoxide dismutase and T lymphocyte subsets in the elderly[J]. Journal of Traditional Chinese Medicine, 2011, 52(21): 1840-1842. [In Chinese]

[7] WU Xia, ZHAO Caijiao, LIN Juan, et al. Meta-analysis of the efficacy of acupuncture in delaying aging[J]. Hunan Journal of Traditional Chinese Medicine, 2016, 32(8): 176-180. [In Chinese]

[8] HIGGINS JP, ALTMAN DG, GØTZSCHE PC, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials[J]. BMJ, 2011, 343: d5928.

[9] ZHANG Tianqi. Clinical study of SVF gel filling combined with acupuncture in improving nasolabial fold aging[D]. Nanjing: Nanjing University of Chinese Medicine, 2019. [In Chinese]

[10] GAO Lan. Observation on the efficacy of magnetic needle horizontal needling plus body acupuncture combined with traditional Chinese medicine mask for wrinkle removal[J]. Journal of Heze Medical College, 2022, 34(1): 63-66. [In Chinese]

[11] XU Jing. Clinical study of rolling needle combined with bFGF in the treatment of crow’s feet in young people[D]. Chengdu: Chengdu University of Traditional Chinese Medicine, 2011. [In Chinese]

[12] ZHU Xuejie, ZHAO Xia. Application of ActiveFXTM fractional CO2 laser combined with traditional Chinese medicine acupuncture in removing periorbital wrinkles[J]. Chinese Journal of Aesthetic Medicine, 2019, 28(7): 34-36. [In Chinese]

[13] LIU Songchun. Application of acupuncture points in skin beautification and wrinkle removal[J]. China Medical Cosmetology, 2014, 4(2): 131, 119. [In Chinese]

[14] HE Li, WU Ruiyao, ZHANG Yao. Analysis of the application effect of acupuncture wrinkle removal in facial rejuvenation[J]. Chinese Scientific Journal Database (Full Text Edition) Medical and Health, 2023(10). [In Chinese]

[15] SHEN Ling. Effect of mesoderm acupuncture therapy on improving lower eyelid wrinkles[J]. Medical Aesthetics and Cosmetology, 2022, 31(13): 88-90. [In Chinese]

[16] DAI Liying. Observation on the effect of traditional Chinese medicine acupuncture combined with laser in removing periorbital wrinkles[J]. Healthy Women, 2022(25): 184-186. [In Chinese]

[17] ZHOU Shuanglin. Clinical and experimental study on catgut embedding at Foot Yangming Meridian in the treatment of facial skin aging[D]. Beijing: Beijing University of Chinese Medicine, 2014. [In Chinese]

[18] HUANG Rong, GAO Fenghua. Observation on the efficacy of ultramicro acupotomy combined with microneedle therapy on static wrinkles of the lower face[J]. China Medical Cosmetology, 2016, 6(10): 59-61. [In Chinese]

[19] LI Shuqing. Acupoint catgut embedding therapy for dynamic wrinkles of the upper face[J]. World Latest Medicine Information, 2020, 20(15): 87-89. [In Chinese]

[20] HAGHIR H, YAZDANPANAH MJ, FARAHMAND SK, et al. Is acupuncture effective in diminishing frown lines? Evidence from a randomized controlled trial[J]. Journal of Cosmetic Dermatology, 2025, 24: e70144.

[21] SUN Jing. Role of syndrome differentiation and treatment of traditional Chinese medicine in cosmetology[J]. Chinese Journal of Aesthetic Medicine, 2011, 20(3): 463-464. [In Chinese]

[22] BAI Yan, MA Xintian, LIU Hao, et al. Observation on 31 cases of nasolabial fold wrinkles improved by facial acupuncture combined with edge needle[J]. Chinese Journal of Traditional Chinese Medicine Science and Technology, 2024, 31(2): 281-284. [In Chinese]

[23] YU Maoji. Meridian therapy and cosmetology[M]. Shanghai: Shanghai University of Traditional Chinese Medicine Press, 2001. [In Chinese]

[24] SU Wenwu, ZHENG Zhaoshun. Acupoint catgut embedding for dynamic wrinkles of the upper face[J]. China Health Care & Nutrition, 2013, 2(2): 134-135. [In Chinese]

[25] CHEN Bo, LYU Mingzhuang, HE Zhiguang, et al. Experimental study on anti-free radical oxidative damage to DNA by acupoint catgut embedding plus moxibustion in aging model rats[J]. Chinese Acupuncture & Moxibustion, 2003, 23(4): 223-227. [In Chinese]

[26] WANG Keyuan, ZHI Yaolong, GU Pingwei, et al. Research progress on mechanism of acupuncture delaying skin aging[J]. Journal of Shaanxi University of Chinese Medicine, 2025, 48(2): 123-127. [In Chinese]

Appendix: Additional Outcome Measures of Included Studies