Parkinson’s disease (PD) ranks among the most prevalent degenerative brain disorders, second only to dementia. It manifests as a chronic and progressive degeneration of the nervous system, characterized by the gradual deterioration of the nigrostriatal pathway due to a deficiency in dopamine resulting from the loss of neurons in the substantia nigra of the midbrain. This degeneration leads to symptoms such as bradykinesia, resting tremors, and muscle rigidity [1].

The prevalence of PD is estimated to range from 5 to 35 cases per 100,000 individuals, with its incidence increasing with age, affecting approximately 2–3% of those aged 60 or older [2]. Notably, data from the National Health Insurance Service of Korea indicates a steady rise in the number of individuals receiving treatment for PD, with figures climbing from 96,499 in 2016 to 111,311 in 2020, reflecting an average annual growth rate of 3.6% [3].

Standard management of PD typically involves the administration of levodopa, a precursor to dopamine. While initially effective in alleviating symptoms, dopamine replacement therapy is associated with side effects such as wearing-off phenomenon and dyskinesia in later stages [4].

In Korean medicine, approaches like herbal medicine aim to suppress hyperkinesia, enhance blood circulation, and reduce tremors, while acupuncture is utilized to alleviate back muscle tension and enhance lower limb muscle function. Combining bee venom and pharmacopuncture with acupuncture has shown greater efficacy in symptom improvement compared to treatment with Western medicine alone. Additionally, surgical interventions such as deep brain stimulation, alongside physical and exercise therapies, are employed to manage the condition [5].

Acupuncture therapy for PD is undergoing exploration and examination through diverse modalities such as cranial acupuncture, conventional acupuncture, bee venom therapy, and auricular acupuncture [6]. Research indicates that the acupuncture treatment group for PD exhibited significant efficacy in managing both exercise-related and non-exercise-related symptoms compared to control groups [7]. Furthermore, analysis spanning 5 years from May 2016 to April 2021 has revealed a growing trend in utilizing various acupuncture techniques to address specific symptoms associated with PD [4].

Despite these findings, research concerning acupuncture therapy for PD has predominantly remained at a conceptual level, lacking in-depth individual study analyses. Therefore, the authors aim to elucidate the latest international research trends regarding acupuncture therapy for PD and conduct a meta-analysis to furnish a foundation for understanding the methodology and efficacy of acupuncture treatment in PD. This endeavor seeks to further increase its clinical application and provide a groundwork for future clinical investigations.

1. Data sources and search methods

The search encompassed eight databases: PubMed (British), EMBASE (British), CNKI (Chinese), as well as Korean literature databases including OASIS, KISS, KMBASE, RISS, and ScienceOn. This search was conducted on September 8, 2023, with a focus on studies published after January 2014 to analyze research conducted over the past decade (Appendix 1).

2. Inclusion and exclusion criteria for literature

This study exclusively considered clinical randomized controlled trials (RCTs) involving patients with PD. Case reports, case series, various crossover study designs, previous studies, validation studies, simple reviews, mechanism-oriented studies, and experimental investigations were excluded. Patients diagnosed with PD were eligible for inclusion, without restrictions based on age, gender, race, or duration of illness. Interventions in the treatment group were only included if they involved acupuncture administration. Various acupuncture modalities such as electroacupuncture, scalp acupuncture, pharmacopuncture, cranial acupuncture, and auricular acupuncture were considered. Additionally, interventions combining acupuncture with conventional medication drugs like Madopar were deemed suitable for assessing acupuncture treatment effects. However, interventions combining acupuncture with other integrative therapies that could potentially influence the evaluation of acupuncture efficacy were excluded from the analysis.

3. Data selection and extraction

The data from the articles underwent validation and extraction based on pre-established criteria. Initially, titles and abstracts were screened to eliminate duplicates and unsuitable studies, following which full-text articles were reviewed to extract relevant information. Extracted data included details on participants, study methodologies, interventions (such as acupuncture points and treatment frequency), outcomes (both primary and secondary), study results, and any reported adverse events (AEs).

4. Risk of bias assessment

Cochrane’s risk of bias tool 5.4.1 was employed to assess the quality of the included studies. This assessment covered various aspects, including random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, completeness of outcome data, selective outcome reporting, and other potential sources of bias. Each category was categorized as exhibiting either a “low risk of bias,” “unclear risk of bias,” or “high risk of bias.” Both authors (JO, YBC) independently assessed the risk of bias, with any discrepancies resolved through consultation with a third researcher (YK) if needed.

5. Data collection, analysis, and management

Statistical analysis involved the examination of outcome indicators from studies amenable to quantitative synthesis. The Review Manager software application (version 5.3) was utilized for data analysis. Mean differences (MDs) and odds ratios, accompanied by 95% confidence intervals (CIs), were calculated using the generic inverse variance estimation method. Distinctly, fixed-effect and random-effect models were applied based on the degree of heterogeneity observed. The I² test was employed to assess the degree of heterogeneity. Notably, the evaluation of publication bias through the funnel plot, typically effective with a minimum of ten studies, was omitted due to insufficient study inclusion for meta-analysis.

1. Literature selection

Initially, a comprehensive search yielded 1,199 documents. After removing 206 duplicate cases, the titles and abstracts of 993 documents were scrutinized. Subsequently, 73 documents were initially chosen, excluding 888 cases that did not align with the predefined criteria regarding subject matter, intervention, and study design, as well as 32 duplicate cases. Upon thorough examination of the full text of the initially selected papers, 38 documents were ultimately selected, following the exclusion of 34 cases (nonrandomized studies [n = 2], literature reviews [n = 1], conference abstracts [n = 2], articles [n = 1], protocols [n = 2], other studies not meeting the criteria [n = 26], and duplicate case [n = 1]). Among these, 12 documents were used for the meta-analysis (Fig. 1).

Figure 1. Flow diagram of the study selection process for this review.

2. Qualitative analysis of the included literature

1) Characteristics of study subjects

All 38 studies ultimately included in the analysis were RCTs, with the number of subjects ranging from 21 to 128, totaling 2,786 participants. The average age of the study subjects was 41.22 years (Table 1) [8-45].

Table 1 . Summary of key data in included randomized controlled trials.

Study (y)	Participants (E/C) (Age, Dx duration, H&Y)	Experimental group treatment (points)	Intervention (n)	Control (n)	Outcome measures	Results	AEs (G:N)	Sub-symptom
Yao (2014) [8]	40–68 (58)/42–70 (60) 0.5–10/0.42–9 NA	GV21, GV19, BL7, CV23, TE10, SI4, SI3, LI10, LI11, GB34, LR3, KI7, EX-HN12, EX-HN13, Waitianzu, Shegen, and Shejuan	A. Medication (same as B) + AT Frequency: 1 time a day Period: 1 month (n = 30)	B. Medication Madopar Period: 1 month (n = 27)	Webster scale	A > B p < 0.05	NA
Li (2015) [9]	59.7 ± 6.7/59.8 ± 6.9 NA NA	PC7, TE4, SI6, LI5, LI11, TE14, HT3, PC3, LU5, LI15, LR4, GB30, ST41, SI9, KI3, SP9, GB34, BL36, ST36, CV12, PC6, CV10, CV6, ST37, ST39, BL39, Zuhegu, etc.	A. Medication (same as B) + AT Frequency: 1 time a day Period: 3 months (n = 60)	B. Medication Madopar Period: 3 months (n = 60)	TER	A > B p < 0.05	NA
Lin (2015) [10]	61.2 ± 7.1/60.7 ± 6.7 15.8 ± 4.9/14.2 ± 4.1 NA	GV20, BL10, GB12, GV15, HT2, LI11, ST40, KI3, SP6	A. Medication (same as B) + AT) Frequency: 1 time a day Period: 1 month (n = 31)	B. Medication Madopar Period: 1 month (n = 31)	1) Webster scale 2) TER	1) A > B p < 0.05 2) A > B p < 0.05	I:19/C:23 Needle fainting (I:1/C:0), dizziness (I:2/C:3), nausea and vomiting (I:5/C:6), loss of appetite (I:7/C:9), palpitation (I:4/C:5)
Shi (2015) [11]	67 ± 10/67 ± 8 8.92 ± 2.29/7.79 ± 2.15 2.42 ± 0.72/2.17 ± 0.79	GV12	A. Medication (same as B) + AT Frequency: 3 times a week Period: 3 months (n = 31)	B. Medication Madopar Period: 3 months (n = 30)	UPDRS	A > B p < 0.01	NA
Jiang (2016) [12]	56 ± 6/56 ± 6/57 ± 6 1.09 ± 0.30/1.24 ± 0.34/1.07 ± 0.45 NA	LI4, LR3, GB20, BL18, BL23, SP6, ST40, CV12, SP9, CV6, SP10, ST36, and EX-HN1	A. Medication (same as C) + AT + rehabilitation training Frequency: 1 time every other day Period: 2 months (n = 32)	B. Medication (same as C) + rehabilitation training (same as A) Frequency: 1 time every other day Period: 2 months (n = 30) C. Medication Madopar Frequency: 1 time every other day Period: 2 months (n = 31)	1) UPDRS 2) BBS	1) A > B, C p < 0.05 2) A, B > C p < 0.05	NA
Li (2016) [13]	60.4 ± 3.3/60.5 ± 3.2 NA NA	TE4, PC7, LI5, LU5, SI6, HT3, LI11, PC3, TE14, LI15, SI9, KI3, LR4, ST41, GB34, SP9, GB30, BL40, BL36, PC6, ST36, CV12, LI4, CV6, CV10, ST37, ST39, BL39, etc.	A. Medication (same as B) + AT Frequency: 1 time a day Period: 3 months (n = 30)	B. Medication Madopar Period: 3 months (n = 30)	TER	A > B p < 0.05	Dry mouth (I:36/C:49), constipation (I:30/C:43), insomnia (I:15/C:30), nausea (I:10/C:23), palpitations (I:13/C:20), abnormal movements (I:10/C:14), wearing-off phenomenon (I:3/C:13), switch phenomenon (I:3/C:12), irritability (I:5/C:10), and mental disorders (I:1/C:5)
Liu (2016) [14]	65 ± 6/65 ± 7 4.1 ± 1.3/4.3 ± 1.4 NA	Shang-sanhuang, Xia-san huang, Linggu, and Dabai + (Tung’s acupuncture points)	A. Medication (same as B) + AT Frequency: 1 time every other day Period: 20 days (n = 46)	B. Medication Levodopa Period: 20 days (n = 46)	1) Walk across 2) Step/quick turn 3) TUGT	1) A > Bp < 0.05 2) A > Bp < 0.05 3) A > Bp < 0.05	NA
Xie (2016) [15]	71.51 ± 6.06/71.51 ± 6.06 5.7 ± 4.1/5.7 ± 4.1 NA	GV20, PC6, LR3, CV17, LR14, LR2, ST36, BL15, BL20, EX-HN1, and EX-HN3	A. Medication (same as B) + AT Frequency: 1 time a day Period: 2 months (n = 53)	B. Medication (bifid triple viable capsules) Period: 2 months (n = 53)	1) HAM-D 2) UPDRS	1) A > Bp < 0.05 2) A > Bp < 0.05	NA	Depression
Kluger (2016) [16]	64.4 ± 10.3/63.0 ± 13.0 NA 1–3/1–4	GV20, GV24, CV6, LI10, HT7, ST36, and SP6	A. AT Frequency: 2 times a week Period: 6 weeks of biweekly (n = 47)	B. Sham acupuncture Frequency: 2 times a week Period: 6 weeks of biweekly (n = 47)	1) MFIS 2) The UPDRS motor 3) HADS 4) PDSS 5) ESS 6) AES	All p > 0.05	Constipation (I:1/C:0)	Fatigue
Li (2017) [17]	59.18 ± 3.15/58.51 ± 2.82 NA 2–4	HT5, PC3, SP6, ST36, LR3, KI3, LI4, GV4, CV4, GB39, GV20, GB20, SP9, and GB34	A. Medication (same as B) + AT (n = 64)	B. Medication Madopar Period: 3 months (n = 64)	UPDRS	A > B (p < 0.05)	Constipation (I:30/C:46), insomnia (I:18/C:31), nausea (I:11/C:24), switch phenomenon (I:4/C:15), wearing-off phenomenon (I:4/C:14) (A < B) (p < 0.05)
Zhao (2017) [18]	53.3 ± 0.2/53.4 ± 0.6 4.15 ± 0.22/4.08 ± 0.17 1–3	LR3, GV20, GB20, LI4, ST36, SP10, CV6, SP6, BL23, and BL18	A. Medication (levodopa) + AT Frequency: 1 session: once a day for 5 days, next session after 2 days Period: 3 months (n = 54)	B. Medication Levodopa Period: 3 months (n = 54)	1) UPDRS 2) Webster scale	A > B (p < 0.01)	NA
Aroxa (2017) [19]	65 ± 10/56 ± 12 NA 1–3	LR3, SP6, LI4, TE5, HT7, PC6, LI11, and GB20	A. Medication (same as B) + AT Period: 8 weekly sessions of 30 min (n = 11)	B. Medication (antiparkinsonian medication) (n = 11)	PDSS	A > B (p < 0.02)	NA	Sleep disorders
Lin (2018) [20]	63.4 ± 4.3/61.6 ± 5.8 5.3 ± 2.6/4.7 ± 2.3 NA	LI4, LR3	A. Medication (same as B) + AT Frequency: 4 weekly sessions of 30 min (n = 30)	B. Basic anti-Parkinson treatment (n = 30)	UPDRS	A > B (p < 0.05)	NA
Sheng (2018) [21]	60.28 ± 3.15/61.07 ± 3.28 NA NA	LI5, SI6, TE4, PC7, HT3, LU5, LI11, LI15, PC2, SI9, ST41, KI3, LR4, BL36, GB30, BL40, GB34, SP9	A. Medication (same as B) + AT Frequency: 1 session: once a day for 15 days, rest 5–7 days. Total 3 sessions (n = 38)	B. Medication Madopar (n = 38)	Walking distance and motion completion time	A > B (p < 0.05)	NA
Cao (2018) [22]	56.31 ± 9.62/55.93 ± 10.08 3.73 ± 2.09/3.62 ± 1.53 NA	GB39, GV20, and EX-HN1	A. Cognitive function training (same as B) + AT Frequency: 1 session: once a day for a week Period: 6 weeks (n = 48)	B. Cognitive function training (n = 48)	1) UPDRS III 2) The 6-minute walk test 3) MMSE 4) MoCA	1) A > B (p < 0.05) 2) A > B (p < 0.05) 3) A > B (p < 0.05) 4) A > B (p < 0.05)	NA
Shen (2018) [23]	74 ± 11/72 ± 11 8.53 ± 1.55/7.13 ± 1.47 2.5–5	ST25, CV6, ST29, TE6, ST36, and ST37	A. Medication (same as B) + AT Frequency: Once a day, every other day Period: 1 month (n = 30)	B. Medication (2 Ma Ren capsules) Frequency: Three times a day (n = 30)	1) BBS 2) PAC-QOL 3) UPDRS III	1) A > B (p < 0.05) 2) A > B (p < 0.05) 3) A > B (p < 0.05)	NA	Constipation
Kong (2017) [24]	66.4 ± 6.5/62.9 ± 9.7 7.27 ± 4.43/4.18 ± 2.2 NA	PC6, LI4, ST36, SP6, KI3, and CV6	A. Real needles AT Frequency: twice-weekly sessions, a total of 10 sessions of acupuncture Period: 5 weeks (n = 20)	B. Sham needles AT Frequency: twice weekly sessions, 10 sessions of acupuncture Period: 5 weeks (n = 20)	1) MFIS 2) The UPDRS motor 3) PDQ-39 4) GDS 5) ESS	1) NA 2) NA	Skull fracture after fall (I:1) Pelvic fracture after fall (I:1) Exacerbation of anxiety (I:1) All adverse events were deemed unrelated to acupuncture treatment	Fatigue
Zhan (2019) [25]	61.87 ± 6.25/63.14 ± 7.28 5.89 ± 4.62/5.75 ± 3.17 NA	ST36, ST37, ST25, BL25, KI6, and TE6	A. AT Frequency: six times a week Period: 2 months (n = 20)	B. Medication (Maren pill 10 g) Frequency: twice a day Period: 2 months (n = 20)	1) TER 2) BSFS and PAC-QOL	1) A > B (p < 0.05) 2) A > B (p < 0.05)	NA	Constipation
Shen (2019) [26]	73.9 ± 11.5/71.9 ± 11.4 8.5 ± 1.6/7.1 ± 1.5 2.5–5	ST25, CV6, ST29, TE6, ST36, and ST37	A. Medication (same as B)+ AT Frequency: every other day Period: 1 month (n = 36)	B. Medication (3 Maren soft capsules) Frequency: every other day Period: 1 month (n = 36)	1) TER 2) Constipation symptom score 3) UPDRS II 4) Medication compliance	1) A > B (p < 0.05) 2) A > B (p < 0.05) 3) A > B (p < 0.05) 4) A > B (p < 0.05)	NA	Constipation
Wang (2020) [27]	59 ± 10/59 ± 10 5.26 ± 1.02/5.31 ± 1.08 NA	GV24, GV20, EX-HN3, BL10, GB20, GB12, EX-HN13, EX-HN12, KI16, LU7, Pricking bloodletting at Yanhoubi	A. Medication (same as B) + oral sensorimotor training (same as B) +AT Frequency: once a day for 5 days, rest 2 days Period: 1 month (n = 45)	B. Medication Madopar + oral sensorimotor training (n = 45)	1) Kubota’s water swallowing test score 2) FOIS score 3) VFSS score 4) NRS 2002 5) Cerebral hemodynamics of the bilateral cerebral arteries (1) Vm (2) Vs (3) RI 6) MBI score 7) SWAL-QOL score	1) A > B p < 0.05 2) A > B p < 0.001 3) A > B p < 0.001 4) A > B p < 0.001 5) (1), (2) A > B p < 0.001 (3) A > B p < 0.05 6), 7) A > B p < 0.001	NA	Dysphagia
You (2020) [28]	58.35 ± 3.34/58.41 ± 3.39 11.23 ± 3.25/11.18 ± 3.23 NA	TE4, LU5, LI11, HT3, SI6, PC7, LI5, PC3, LI15, SI9, TE14, LR4, KI3, ST41, GB34, SP9, GB30, PC6, ST36, BL40, BL36, ST37, ST39, BL39, CV12, LI4, and CV6	A. Medication (same as B)+ AT Frequency: once a day Period: 3 months (n = 42)	B. Medication (Madopar) Period: 3 months (n = 42)	1) TER 2) Levels of CD+3, CD+4, CD+8, and CD+4/CD+8 3) Levels of GP and PUT (SWI) 4) Incidence of dry mouth, constipation, nausea, palpitations, and insomnia	1) A > B p < 0. 05 2) A > B p < 0. 05 3) A > B p < 0. 05 4) A > B p < 0.05	NA
Zhang (2020) [29]	52.16 ± 3.56/55.32 ± 3.02 4.46 ± 1.32/4.12 ± 1.21 NA	GB20, GV20, and EX-HN3	A. Medication (same as B) + AT Frequency: once a day for 6 days, rest 1 day. One session lasts for 2 weeks Period: 2 months (n = 48)	B. Medication Pramipexole (n = 48)	1) YKL-40 level 2) BDNF levels 3) UPDRS 4) Effective rate	1) A > B p < 0.05 2) A > B p < 0. 05 3) A > B p < 0. 05 4) A > B p < 0. 05	NA
Luo (2020) [30]	58.93 ± 2.44/58.95 ± 2.45 4.09 ± 0.51/4.11 ± 0.52 < III	LR3, LI4	A. Medication + deep brain stimulation (same as B) + AT Frequency: once a day Period: 2 months (n = 30)	B. Medication Madopar + deep brain stimulation (n = 30)	1) UPDRS 2) WHO-QOL-2 3) Inflammatory response and oxidative stress (IL-1β, IL-6, SOD, MDA, DA) 4) Electromyographic tremor (1) Intensity (2) Frequency 5) TER	1) A > B p < 0.05 2) A > B p < 0.05 3) A > B p < 0.05 4) (1) A < B p < 0.05 (2) A > B p < 0.05 5) A > B p < 0.05	NA
Kim (2020) [31]	NA NA NA	GB20, LI14, DU14, and DU16	A. Qigong meditation (same as B) + AT (n = 10)	B. Qigong meditation (n = 11)	1) UPDRS I 2) TSI	p > 0.05	NA
Jang (2020) [32]	65.38 ± 7.81/61.46 ± 8.33 6.92 ± 4.83/8.38 ± 3.88 1.92 ± 0.64/1.85 ± 0.69	STRICTA 2010 guidelines	A. Conventional therapy + AT Frequency: twice a week Period: 1 month (n = 13)	B. Conventional therapy (n = 13)	1) UPDRSM 2) GAITRite parameters (1) Cadence (2) Stride time (3) Swing time (4) Single support times	1) A > B p = 0.02 2) A > B (1) p = 0.004 (2) p = 0.006 (3) p = 0.001 (4) p = 0.001	NA
Wu (2021) [33]	60 ± 6/59 ± 5 4.27 ± 0.75/3.92 ± 0.81 2–3	LR3, LI4, GV20, GB34, GB20, KI3, SP6, BL18, BL23, and CV6	A. Medication (same as B) + AT Frequency: once a day for 6 days, rest 1 day Period: 3 months (n = 48)	B. Medication Madopar (n = 48)	1) IL-2, IL-6, and homocysteine 2) Catalas, NT-3, IGF-1, and DA 3) UPDRS 4) MoCA 5) TER	1) A < B p < 0.05 2) A > B p < 0.05 3) A > B p < 0.05 4) A > B p < 0.05 5) A > B p < 0.05	I:7/C:9 Abnormal liver function (I:1/C:1) Gastrointestinal reactions (I:3/C:4) Arrhythmia (I:2/C:3) Kidney function damage (I:1/C:1)
Fan (2022) [34]	40–75/ 40–75 NA 1–3	Du three needles (DU14, DU8 and DU4) Hand Luan three needles (HT 1, LU5 and PC6) Foot Luan three needles (SP9 and SP6)	A. Medication (same as B) + AT Frequency: four times a week Period: 2 months (n = 58)	B. Medication (anti-PD drug) + Sham acupuncture (n = 54)	1) UPDRS III 2) 10-m walk test (time and speed) 3) Ashworth scale score 4) BDNF levels 5) α-synuclein 6) TER	1–3) p < 0.01 4) p < 0.05 5) p > 0.05 6) A > B p < 0.05	NS	Anxiety
Zhang (2022) [35]	67.80 ± 2.73/67.53 ± 2.78 3.04 ± 1.25/3.18 ± 1.23 NA	EX-HN1, GV16, GV26, GB20, PC6, LI4, LR3, GB34, SP6, KI3, and ST36	A. Medication (same as B) + conventional rehabilitation treatment (same as B) + AT Frequency: once a day for 6 days, rest 1 day Period: 2 months (n = 32)	B. Medication (Madopar) + conventional rehabilitation treatment Period: 2 months (n = 32)	1) UPDRS I 2) UPDRS II 3) UPDRS III 4) UPDRS IV	1–4) A > B p < 0.05	NA
Brandín-de la Cruz (2022) [36]	69.9 ± 7.2/69.9 ± 7.2 NA NA	Dry needle: semitendinosus, medial gastrocnemius, soleus, and rectus femoris muscles	A. Dry needle performed only once (n = 18)	B. Sham Dry needle (n = 15) - performed only once	1) TUGT 2) 10-m walk test, 6-minute walk test 3) MyotonometryBefore, immediately after, and 7 days after the intervention	1–3) No significant differences	NS
Fan (2022) [37]	61.03 ± 9.80/62.66 ± 6.94 4–9/2–8 1–4	GV24, GV29, HT7, SP6, and EX-HN1	A. Medication (same as B) + AT Frequency: once per day, 3 times per week Period: 2 months (n = 32)	B. Medication (anti-PD drug) + Sham acupuncture (a noninsertion procedure) (n = 32)	1) HAM-A 2) UPDRS and UPDRS I 3) PDQ-39 (ADL, EW) 4) Levels of the Adrenocorticotropic hormone 5) Levels of Cortisol (1) Post-treatment: 1–5) (2) Follow-up (2 months after): 1–3)	1. Post-treatment 1–3) p > 0.05 4) p < 0.001 5) p = 0.82 2. Follow-up (2 months after treatment) A > B 1–3) p < 0.001	Bleeding (I:1) Subcutaneous hematoma (I:1) Mental tension (I:1) Muscle spasm (I:1)
Han (2022) [38]	61 ± 3/61 ± 3 4.1 ± 2.8/4.4 ± 2.6 1–3	EX-B2 from C2 to L5	A. Medication (same as B) + AT Frequency: once a day for 5 days, rest 2 days Period: 4 months (n = 48)	B. Medication (madopar 125–250 mg) Frequency: three times a day Period: 4 months (n = 48)	1) UPDRS III 2) UPDRS IV 3) TCM symptomps score 4) PDQ-39Post-treatment and follow-up (1 month after)	1–4) A > B p < 0.05	I:5/C:14 (p < 0.05) Dizziness (I:0/C:4), vomit (I:1/C:3), hypotonia (I:2/C:3), fatigue (I:1/C:0), abnormal liver and kidney function (I:0/C:2), lethargy (I:1/C:2)
Li (2022) [39]	63 ± 6.73/59 ± 9.28 5.94 ± 1.64/5.92 ± 1.68 2.68 ± 0.39/2.67 ± 039	GV24, GV20, KI6, and GB20	A. Medication (same as B) + AT Frequency: once a day (except day 5, 6 and 28, 29 after admission) Period: 1 month (n = 30)	B. Medication Levodopa + sham AT Not percutaneous (n = 27)	1) UPDRS III 2) PDSS-2 3) ESS 4) Sleep latency, sleep efficiency, and total sleep time using ActiGraph 5) Sleep efficacy using ActiGraph	1) A > B but, p > 0.05 2) A > B but, p > 0.05 3) A > B p < 0.01 4) A > B p < 0.01 5) A > B; days 5–7, days 26–28 p < 0.01	NA	Sleep disorders
Feng (2023) [40]	71 ± 5/72 ± 6 3.8 ± 2.1/4.2 ± 2.6 NA	GV16, GV20, GV24, EX-HN3, GB20, EX-HN12, EX-HN1, Yansanzhen, Yanhoubi, Yanhoubi, hejian, and Shesanzhen	A. Medication (same as B) + rehabilitation training (same as B) + AT Frequency: once a day, 6 times a week Period: 1 month (n = 28)	B. Medication Madopar + rehabilitation training (n = 27)	1) Kubota water swallowing test 2) Standardized swallowing assessment 3) SWAL-QOL	1), 2) A > B, p < 0.05 3) A > B p < 0.001	NA	Dysphagia
Wu (2023) [41]	66 ± 1/66 ± 1 4.29 ± 0.35/4.34 ± 0.32 NA	LI4, LR3	A. Medication (same as B) + Yi Shen Xi Feng decoction (same as B) + AT Frequency: once a day for 5 days per week Period: 2 months (n = 51)	B. Medication madopar + Yi Shen Xi Feng decoction Period: 2 months (n = 52)	1) SOD, MDA, and glutathione levels 2) UPDRS II 3) TER	1) A > B p < 0.05 2) A > B p < 0.05 3) A > B p < 0.05	NA
Luo (2023) [42]	23–68 (42.81)/24–67 (41.22) 1–13 (12.34)/1–12 (11.68) 2–4	GV24, GV20, EX-B2(C4), BL14, Tung’s acupuncture points, LI10, LU5, PC7, GB29, GB33, BL58, BL40, BL23, BL24	A. Medication (same as B) + AT Frequency: 3 days per week Period: 1 month (n = 30)	B. Medication Madopar (n = 30)	1) UPDRS III 2) HAM-D and HAM-A 3) Pittsburgh sleep quality index 4) TCM syndrome 5) TER	1–4) A > B p < 0.05 5) A > B p < 0.05	NA
Zhao (2023) [43]	69.18 ± 12.33/68.29 ± 13.67 1.31 ± 0.55/1.26 ± 0.49 NA	TE4, LI5, PC7, LU5, LI1, PC3, and HT3	A. Medication (same as B) + AT Frequency: once a day Period: 3 months (n = 39)	B. Medication Madopar (n = 39)	1) GUSS score 2) Albumin, Hb, and TG	1), 2) A > B p < 0.05	I:3/C:7 Switch phenomenon (I:2/C:3), abnormal mental state (I:1/C:2), taste disorder (I:0/C:2)	Dysphagia
Gu (2023) [44]	51.37 ± 7.26/52.07 ± 7.56 4.67 ± 0.82/4.21 ± 0.79 NA	GV20, EX-HN3, EX-HN1, PC6, HT7, GB40, and PC7	A. Medication (same as B) + AT Frequency: once a day for 5 days, rest 2 days Period: 3 months (n = 52)	B. Medication Madopar and sertraline hydrochloride Period: 3 months (n = 51)	1) UPDRS 2) Yale–Brown obsessive compulsive scale 3) HAM-A and HAM-D 4) Improvement rate of obsessive state and effective rate of PD	1–4) A > B p < 0.05	I:6/C:5 Loss of appetite (I:2/C:1), nausea (I:2/C:2), vomiting (I:1/C:1), insomnia (I:1/C:1) (p > 0.05)	Obsessive state
Li (2023) [45]	63.90 ± 7.34/63.74 ± 9.24 5.74 ± 3.95/6.05 ± 4.37 ≤ 3	EX-HN1, GV24, GV29, ST25, CV4, and ST37	A. AT Frequency: 3 times a week Period: 1 month (n = 36)	B: Sham AT Frequency: 3 times a week Period: 1 month (n = 35)	1) Complete spontaneous bowel movements (weekly) 2) Constipation symptom and efficacy assessment scale 3) PAC-QOL 4) UPDRS and UPDRS I Post-treatment (week 4), follow-up (week 8)	1. Post-treatment (week 4) A > B 1) p < 0.001 2) p < 0.001 3) p = 0.005 4) p = 0.147 2. Follow-up (week 8) 1) p < 0.001 2) p < 0.001 3) p = 0.004 4) p = 0.134	I:6/C:0 bleeding (I:3/C:0), subcutaneous hematoma (I:2/C:0), sharp pain (I:1/C:0) p = 0.276	Constipation

E/C, experimental/comparison; Dx, diagnosis; H&Y, Hoehn and Yahr stage; AT, acupuncture treatment; NA, not applicable; NS, not significant; TER, total effective rate; UPDRS, unified Parkinson’s disease rating scale; BBS, Berg balance scale; TUGT, timed up and go test; HAM-D, Hamilton depression scale; MFIS, modified fatigue impact scale; HADS, hospital anxiety and depression scale; PDSS, Parkinson’s disease sleep scale; ESS, Epworth sleepiness scale; AES, apathy evaluation scale; MMSE, mini-mental state examination; MoCA, Montreal cognitive assessment; PAC-QOL, patient assessment of constipation-quality of life; PDQ-39, Parkinson’s disease questionnaire-39; GDS, geriatrics depression scale; BSFS, Bristol stool form scale; FOIS, functional oral intake scale score; VFSS, video fluoroscopic swallowing study; NRS 2002, nutrition risk screening 2002; Vm, mean blood flow velocity; Vs, peak systolic blood flow velocity; RI, resistance index; MBI, modified Barthel index; SWAL-QOL, swallowing-related quality of life; GP, globus pallidus; PUT, putamen; SWI, susceptibility weighted imaging; YKL-40, tyrosine (Y), lysine (K) and leucine (L), and its molecular mass of 40 kDa 14; BDNF, brain-derived neurotrophic factor; WHO-QOL, World Health Organization quality of life; IL, interleukin; SOD, superoxide dismutase; MDA, malondialdehyde; DA, dopamine; TSI, test of smell identification; STRICTA, Standards for Reporting Interventions in Clinical Trials of Acupuncture; NT-3, neurotrophin-3; IGF-1, insulin-like growth factor 1; UPDRSM, motor section of the unified Parkinson’s disease rating scale; PD, Parkinson’s disease; TCM, traditional Chinese medicine; HAM-A, Hamilton anxiety scale; GUSS, Gugging Swallowing Screen Score; TG, triglyceride; Hb, hemoglobin..

2) Intervention

Acupuncture was consistently employed as a treatment intervention across all study groups. Among the 38 studies, 25 incorporated a combination of acupuncture and Western medicine, while acupuncture alone was utilized in 5 studies. Additionally, one study employed a combination of acupuncture, herbal medicine, and Western medicine, while another combined acupuncture and qigong medication. Six studies implemented a multifaceted approach involving acupuncture, Western medicine, rehabilitation, and cognitive treatment.

In contrast, control groups predominantly received Western medicine treatments such as Levodopa in 21 studies, with 4 studies combining Western medicine and rehabilitation treatments. Sham acupuncture treatment alone was administered in four studies, while in three studies, a combination of Western medicine and sham acupuncture was utilized. Furthermore, three studies solely administered Maren capsules, while one study employed both herbal and Western medicines. Lastly, qigong medication or rehabilitation therapy was the intervention in one study each.

3) Acupuncture points

In PD studies, the most frequently employed acupuncture points were GV20 and ST36, utilized in 13 studies, followed by and LI4 in 12 studies. Additionally, GB20, LR3, and SP6 were utilized in 11 studies each, while EX-HN1, CV6, KI3, and PC6 were employed in 9 studies. LI11 was utilized in eight studies, and GV24, GB34, LU5, SP9, and PC7 were used in seven studies.

4) Evaluation tool

The primary assessment tools utilized across most studies included the unified Parkinson’s disease rating scale (UPDRS) and the Webster scale, in conjunction with the total effective rate (TER).

Furthermore, assessments of balance and walking capabilities were conducted using the Berg balance scale (BBS), various walk tests (such as walk across, step/quick turn, timed up & go test, walking distance and motion completion time, 6-minute walk test, and 10-m walk test), as well as GAITRite parameters.

The Hamilton depression scale (HAM-D), Hamilton anxiety scale (HAM-A), hospital anxiety and depression scale, geriatrics depression scale, and apathy evaluation scale were used to evaluate symptoms of depression and anxiety. Additionally, the modified fatigue impact scale (MFIS) was used to evaluate fatigue symptoms.

To evaluate cognitive improvement, the mini-mental state examination and Montreal cognitive assessment were administered. Sleep quality was evaluated using the Parkinson’s disease sleep scale (PDSS), PDSS-2, Pittsburgh sleep quality index, Epworth sleepiness scale (ESS), and ActiGraph. Evaluation of dysphagia involved assessments such as Kubota’s water swallowing test score, functional oral intake scale score, video fluoroscopic swallowing study (VFSS) score, swallowing-related quality of life score, Gugging swallowing screen (GUSS) score, and Standardized Swallowing Assessment. For the assessment of constipation, tools including the Bristol stool form scale, patient assessment of constipation-quality of life, complete spontaneous bowel movements, constipation symptom and efficacy assessment scale, and constipation symptom score were used.

Furthermore, the modified Barthel index score, World Health Organization quality of life, test of smell identification, Ashworth scale score, Yale–Brown obsessive compulsive scale, and medication compliance were used once in the respective studies.

Moreover, individual studies assessed cerebral hemodynamics of bilateral cerebral arteries, as well as the globus pallidus and putamen, utilizing susceptibility weighted imaging. Additionally, some investigations included blood tests to determine levels of brain-derived neurotrophic factor, interleukin (IL)-1β, IL-2, IL-6, superoxide dismutase, malondialdehyde, dopamine, catalase, neurotrophin-3, insulin-like growth factor-1, tyrosine (Y), lysine (K) and leucine (L), and its molecular mass of 40 kDa 14, homocysteine, glutathione, albumin, hemoglobin, and triglyceride, as well as CD+3, CD+4, and CD+8 values.

3. Quantitative synthesis and results analysis (meta-analysis)

This study examined 12 out of 26 studies utilizing Total UPDRS, UPDRS I, UPDRS II, UPDRS III, and UPDRS IV, which are representative evaluation scales for PD, and performed a meta-analysis. Among them, seven studies compared acupuncture treatment with Western medicine alone, and five studies compared acupuncture treatment with both Western medicine and sham acupuncture treatment alongside Western medicine.

Studies conducted by Shi and Zhang [11] and Lin et al. [20] were excluded due to methodological heterogeneity regarding general acupuncture treatment. Shi and Zhang [11] utilized acupuncture treatment involving the transverse insertion of long acupuncture needles at GV12, while Lin et al. [20] focused solely on acupuncture treatment at Siguan points (LI4, LR3), with the definition of basic treatment not clearly specified. Additionally, Brandín-de la Cruz et al. [36] was excluded as it differed from other studies in that acupuncture treatment was administered only once.

In the studies by Shen et al. [23] and Shen et al. [26], the herbal medicine “Maren pill” served as the basic treatment, while Wu et al. [41] incorporated the herbal medicine “Yi Shen Xi Feng decoction” into the basic treatment. Xie et al. [15] was excluded from the analysis due to intervention heterogeneity, as it used lactobacilli as the basic treatment, unlike the other studies.

Zhang et al. [29] solely utilized pramipexole as a control, differing from the predominant use of levodopa as a control group in most studies. Due to this heterogeneity, Zhang et al.’s [29] study was excluded from the analysis. Furthermore, Luo et al. [30], Cao et al. [22], and Kim et al. [31] were excluded due to inclusion of interventions such as deep brain stimulation, cognitive function training, and qigong medication, which were inconsistent with other studies.

Li [17] used UPDRS but reported the rate of change rather than the original score, while Jang et al. [32] reported only the sum of UPDRS partial scores. Li et al. [39] was excluded from the analysis as quantitative synthesis was not feasible due to the absence of reported research results.

1) Effects of acupuncture with Western medicine using the UPDRS total score

Three studies [18,29,33] comparing acupuncture with Western medicine versus Western medicine alone, and two studies [37,45] comparing acupuncture with Western medicine versus sham acupuncture with Western medicine, were analyzed.

An analysis of 3 RCTs involving 307 patients revealed a significant improvement in the total UPDRS score with acupuncture combined with Western medicine compared to Western medicine alone (MD, 4.92; 95% CI, 3.79–6.05; p < 0.01; I² = 3%) (Fig. 2-1.1).

Figure 2. Total unified Parkinson’s disease rating scale score. WM, Western medicine; SD, standard deviation; IV, inverse variance; CI, confidence interval; Acu, acupuncture.

Similarly, an analysis of 2 RCTs comprising 142 patients showed. Significant improvement in the total UPDRS score with acupuncture combined with Western medicine compared to sham acupuncture with Western medicine (MD, 3.39; 95% CI, 2.49–4.29; p < 0.01; I² = 0%) (Fig. 2-1.2).

Furthermore, an analysis of 5 RCTs involving 449 patients demonstrated a significant improvement in the total UPDRS score with acupuncture combined with Western medicine compared to both Western medicine alone and sham acupuncture with Western medicine (MD, 3.99; 95% CI, 3.28–4.69; p < 0.01; I² = 37%).

2) Improvement of non-exercise symptoms in acupuncture with Western medicine using UPDRS Part 1

A meta-analysis examined one study [35] comparing acupuncture with Western medicine versus Western medicine alone, and two studies [37,45] comparing acupuncture with Western medicine versus sham acupuncture with Western medicine.

Analysis of 1 RCT involving 64 patients comparing acupuncture with Western medicine versus Western medicine alone revealed no statistically significant difference between the two groups (MD, 0.50; 95% CI, 0.04–0.96; p = 0.03) (Fig. 3-1.4.1).

Figure 3. Unified Parkinson’s disease rating scale I. WM, Western medicine; SD, standard deviation; IV, inverse variance; CI, confidence interval; Acu, acupuncture.

Similarly, an analysis of 2 RCTs comprising 142 patients comparing acupuncture with Western medicine versus sham acupuncture with Western medicine showed no statistically significant difference between the two groups (MD, 1.33; 95% CI, −1.08 to 3.74; p = 0.28; I² = 95%) (Fig. 3-1.4.2).

Furthermore, analysis of 3 RCTs involving 206 patients revealed no statistically significant difference between acupuncture with Western medicine and control groups (MD, 1.05; 95% CI, −0.46 to 2.57; p = 0.17; I² = 95%).

3) Improvement of exercise experience in daily life in acupuncture with Western medicine using UPDRS Part 2

A meta-analysis examined one study [35] comparing acupuncture with Western medicine versus Western medicine alone.

Analysis of 1 RCT involving 64 patients revealed no statistically significant difference between the two groups (MD, 2.10; 95% CI, −0.15 to 4.35; p = 0.07) (Fig. 4).

Figure 4. Unified Parkinson’s disease rating scale II. WM, Western medicine; SD, standard deviation; IV, inverse variance; CI, confidence interval; Acu, acupuncture.

4) Improvement of exercise-related symptoms in acupuncture with Western medicine using UPDRS Part 3

A meta-analysis investigated four studies [12,35,38,42] comparing acupuncture with Western medicine versus Western medicine alone, and three studies [16,24,34] comparing acupuncture with Western medicine versus sham acupuncture with Western medicine.

An analysis of 4 RCTs involving 281 patients comparing acupuncture with Western medicine versus Western medicine alone revealed a significant improvement in UPDRS III scores with acupuncture combined with Western medicine compared to herbal medicine (MD, 3.75; 95% CI, 2.56–4.94; p < 0.01; I² = 3%) (Fig. 5-1.2.1).

Figure 5. Unified Parkinson’s disease rating scale III. WM, Western medicine; SD, standard deviation; IV, inverse variance; CI, confidence interval; Acu, acupuncture.

Similarly, an analysis of 3 RCTs comprising 240 patients comparing acupuncture with Western medicine versus sham acupuncture with Western medicine showed no statistically significant difference between the two groups (MD, 2.34; 95% CI, −1.77 to 6.45; p = 0.26; I² = 69%) (Fig. 5-1.2.2).

Furthermore, analysis of 7 RCTs involving 521 patients demonstrated a significant improvement in UPDRS III scores with acupuncture combined with Western medicine compared to Western medicine alone and sham acupuncture with Western medicine groups (MD, 3.19; 95% CI, 1.71–4.67; p < 0.01; I² = 44%).

5) Improvement of exercise-related complications in acupuncture with Western medicine using UPDRS Part 4

A meta-analysis investigated two studies [35,38] comparing acupuncture with Western medicine versus Western medicine alone.

Analysis of 2 RCTs involving 160 patients comparing acupuncture with Western medicine versus Western medicine alone revealed a significant imoprovement in UPDRS IV scores with acupuncture combined with Western medicine compared to herbal medicine alone (MD, 0.86; 95% CI, 0.54–1.17; p < 0.01; I² = 93%) (Fig. 6).

Figure 6. Unified Parkinson’s disease rating scale IV. WM, Western medicine; SD, standard deviation; IV, inverse variance; CI, confidence interval; Acu, acupuncture.

6) Assessment of risk of bias

Regarding random sequence generation, 11 studies were categorized as “Low,” while 1 study [44] was deemed “Unclear.”

For Allocation concealment, nine studies were labeled “Unclear,” with two studies [24,45] rated as “Low” and one study [44] marked as “High.”

In terms of blinding of participants and personnel, nine studies were assessed as “High,” while three studies [24,37,45] were considered “Low.” As for blinding of outcome assessment, nine studies received a rating of “High,” whereas three studies [16,37,45] were rated as “Low.”

In assessing incomplete outcome data, all 12 studies were classified as “Low” since no missing values were observed. Regarding selective reporting, 11 studies were categorized as “Unclear,” with 1 study [44] being rated as “Low.” Moreover, all 12 studies were categorized as “Low” for other sources of bias (Fig. 7).

Figure 7. Risk of bias summary and risk of bias graph. The assessment by review authors of each risk of bias item for all included studies is depicted as percentages in both the risk of bias summary and graph.

4. Adverse event reports

Out of the 38 studies examined, 2 studies [34,36] reported no side effects, while AEs were reported in 11 studies. The remaining 25 studies did not document any adverse effects. Gastrointestinal symptoms, encompassing nausea, vomiting, loss of appetite, constipation, and taste disorder, were documented in six studies [10,13,17, 38,43,44]. Cardiovascular symptoms, such as hypotension, dizziness, needle fainting, palpitations, and arrhythmia, were reported in four studies [10,13,33,38], while mental symptoms and sleep disorders, including insomnia, fatigue, irritability, anxiety, mental tension, hypotonia, and lethargy, were documented in seven studies [13,17,19,24,37,43,44]. Additionally, the wearing-off phenomenon was noted in two studies [13,17], and the switch phenomenon was observed in three studies [13,17,43]. Subcutaneous hematoma and bleeding were reported in two studies [37,45], and decreased liver and kidney functions were documented in two studies [33,38]. Furthermore, dry mouth [13], muscle spasm [37], and abnormal movements [13] were reported in one study. Two cases of fractures resulting from falls were reported in a single study, though they were not directly attributed to acupuncture [24]. Notably, no serious AEs associated with acupuncture were recorded across all 38 studies.

Recent systematic literature reviews have delved into acupuncture treatments for PD. Pereira et al. [46] conducted a study in 2022 revealing that combining acupuncture with conventional treatment yielded favorable outcomes in UPDRS scores compared to conventional treatment alone. Additionally, Kim et al. [4] noted in their study that recent research has employed various acupuncture methods for PD treatment and evaluated their long-term effects. However, acupuncture treatment research predominantly remains a topic of discussion, lacking in-depth individual study analyses.

Considering this, the authors aimed to discern the latest trends in foreign research on acupuncture treatment for PD and conducted a meta-analysis. This endeavor seeks to establish a foundation for understanding the methodology and efficacy of acupuncture treatment for PD, potentially fostering its increased clinical use in the future and providing a cornerstone for further clinical investigations.

Due to the challenges associated with early diagnosis of PD, it is highly probable that the majority of patients seeking medical assistance at primary healthcare facilities are undergoing treatment with Western medicine. Therefore, in this study, we focused on assessing the efficacy of combining Western medicine with acupuncture. The primary Western medicine utilized in this study is levodopa, which serves as the main approach for addressing dopamine deficiency in PD [47]. Levodopa’s adverse effects may include nausea, vomiting, hallucinations, and sleep disorders. Prolonged usage of the medication can also lead to motor fluctuations, characterized by shortened or irregular durations of drug effectiveness, and dyskinesia, resulting in excessive movements [48].

The UPDRS and Webster scale were the primary assessment tools utilized across most studies, complemented by the TER. In this study, UPDRS, a prominent measure specific to PD outcomes, was selected as the primary evaluation metric. The UPDRS comprises a total score of 265 points, with higher scores indicating greater disease severity. It encompasses four parts: Part 1 assesses non-exercise symptoms in daily life, Part 2 evaluates exercise-related symptoms in daily life, Part 3 examines exercise-related symptoms in clinical settings, and Part 4 evaluates exercise-related complications. The total UPDRS score and Parts 2 and 3 are mainly used to evaluate exercise-related symptoms associated with PD, while Part 1 evaluates symptoms linked to cognitive impairment, depression and anxiety, sleep disorders, constipation, and fatigue. Often, parallel evaluation scales for each symptom are utilized concurrently [49].

The meta-analysis focused solely on acupuncture treatment, excluding electroacupuncture, pharmacopuncture, and cranial acupuncture. For quantitative synthesis, 122 studies were analyzed, reviewing both the total score and detailed scores of UPDRS (UPDRS I, II, III, IV), which serve as representative indicators of PD.

The primary endpoint of this study was the total UPDRS score. The test group, receiving both acupuncture and Western medicine, exhibited a significant improvement in the total UPDRS score compared to the control group treated solely with Western medicine or with a combination of sham acupuncture and Western medicine (MD, 4.92; 95% CI, 3.79–6.05; p < 0.01; I² = 3%). Subgroup analysis based on the use of sham acupuncture revealed that studies comparing acupuncture combined with Western medicine to Western medicine alone entailed 3-month treatment periods with 5 or 6 weekly sessions, demonstrating a significantly greater effect in the test group compared to the control group (MD, 3.39; 95% CI, 2.49–4.29; p < 0.01; I² = 0%). The acupuncture points included in the studies were GV20, EX-HN3, EX-HN1, PC6, HT7, GB40, PC7, LR3, LI4, GB34, GB20, KI3, SP6, BL18, BL23, CV6, ST36, and SP10. In the control group, sham acupuncture combined with Western medicine treatment was administered for 1 or 2 months, three times a week. The treatment effect in the test group was significantly higher than that in the control group (MD, 3.99; 95% CI, 3.28–4.69; p < 0.01; I² = 37%).

The acupuncture points examined in the studies encompassed GV24, GV29, HT7, SP6, EX-HN1, GV20, EX-HN3, PC6, GB40, and PC7. Despite efforts to minimize heterogeneity among interventions, subgroup heterogeneity between the two studies was found to be relatively high at 76%. Consequently, further analysis of this heterogeneity was conducted. The two studies utilizing sham acupuncture primarily focused on anxiety in Fan et al. [37] and subsymptoms in Li et al. [45], employing the same type of nonpenetrating sham acupuncture device. Given that nonspecific effects can manifest even in sham acupuncture [17] and symptoms like anxiety are subjectively influenced [8], it is probable that the placebo effect within the sham acupuncture group contributed to variations in effect size between the two subgroups. According to this background, researches reached a consensus that sufficient explanation of the heterogeneity observed during the analysis was feasible, enabling the synthesis and estimation of results from both subgroups to determine the actual effect size of acupuncture treatment.

UPDRS Part 1 evaluates non-exercise functions such as cognition, behavior, and mood, with a total of 16 points. Zhang et al. [35], focusing on a control group treated solely with Western medicine, observed clinical symptoms in PD associated with liver and kidney yin deficiency. Meanwhile, Fan et al. [37], exploring the effects of acupuncture on anxiety within a combined treatment control group involving sham acupuncture and Western medicine, and Li et al. [45], examining acupuncture’s impact on constipation, employed similar interventions, controls, and treatment regimens. However, the study subjects differed between Fan et al. [37] and Li et al. [45]. Consequently, UPDRS Part 1 served as a secondary endpoint. Given that the baseline scores for Part 1 in both studies were 5.50 and 3.41 out of 16, respectively, it was deemed that the mean difference attributable to the treatment effect may not have been adequately reflected. Therefore, the presence of heterogeneity should be considered, as evidenced by I² = 95% among sham acupuncture studies in UPDRS Part 1 and I² = 95% among all three studies. A more definitive conclusion may be reached with continued research in the future.

UPDRS Part 2 comprises 52 points and assesses exercise experiences in daily life. Due to the limited number of final selected studies, quantitative synthesis was not feasible. Based on the findings of Zhang et al. [35], the group receiving acupuncture alongside Western medicine demonstrated greater treatment effects compared to the group receiving Western medicine alone. However, this difference did not reach statistical significance (p > 0.07).

UPDRS Part 3 evaluates exercise function and comprises a total of 108 points. Among the four studies analyzed for the control group receiving only Western medicine, involving 281 subjects, a statistically significant difference in treatment effect was observed between the two groups (MD, 3.75; 95% CI, 2.56–4.94; p < 0.01; I² = 3%). However, for the three studies comparing the control group receiving sham acupuncture and western medicine combined treatment, there was high heterogeneity in the study results with an I² = 69%. Kluger et al. [16] and Kong et al. [24] had treatment durations of twice a week for 6 weeks and 5 weeks, respectively, while Fan et al. [37] was treated four times a week for 8 weeks, suggesting that variations in treatment duration may have contributed to differences in study outcomes. Kong et al. [24] mentioned the lack of deli confirmation during acupuncture as a potential factor affecting the specific effects of acupuncture, contributing to the observed heterogeneity among studies. Therefore, the authors agreed that acupuncture treatment significantly improved UPDRS Part 3 exercise function compared to no acupuncture treatment in PD patients undergoing Western treatment (MD, 3.19; 95% CI, 1.71–4.67; p < 0.01; I² = 44%) but noted a moderate heterogeneity of I² = 44%, which may affect the actual effect.

UPDRS Part 4 assesses exercise-related complications and has a total score of 32. Both studies in this category were control groups receiving only western medicine treatment. Upon synthesizing the data from these two studies involving 160 patients, the experimental group demonstrated a statistically significant therapeutic effect compared to the control group (MD, 0.86; 95% CI, 0.54–1.17; p < 0.01; I² = 93%). However, there was high heterogeneity (I² = 93%) between the two studies, rendering the heterogeneity difficult to interpret.

Based on these findings, acupuncture treatment appears to be effective in ameliorating Parkinson’s disease symptoms as assessed by UPDRS, with a low likelihood of significant side effects. Considering that UPDRS encompasses various clinical manifestations, it is plausible that more detailed insights into symptoms could be gleaned from future studies.

Out of the 38 studies examined, two reported no side effects, while 11 documented side effects. The remaining 25 studies did not mention any side effects. Among these, eight studies noted gastrointestinal issues and mental health or sleep-related symptoms associated with Western medicine. Interestingly, in one clinical trial, acupuncture was observed to mitigate the gastrointestinal side effects of levodopa [50]. This suggests that acupuncture might potentially diminish the adverse effects of Western medicine treatments in PD patients, warranting further investigation through observational studies.

The limitations of this study are outlined as follows: To begin with, the pool of studies available for synthesis was limited. PD encompasses diverse clinical manifestations beyond exercise-related symptoms [51]. Consequently, the number of studies eligible for quantitative analysis was restricted due to the considerable heterogeneity in evaluation methodologies during study design. The authors opted for UPDRS as the primary outcome measure in this study, considering its widespread adoption as a pivotal assessment scale for PD in both domestic and international research.

A more definitive conclusion could potentially be drawn if a broader array of research topics were explored in RCTs, such as overall efficacy, the Webster scale, the BBS for assessing balance and walking capabilities, various walking tests, the Hamilton scale for evaluating depression and anxiety symptoms, and the MFIS for assessing fatigue symptoms.

The second limitation pertains to the restriction of acupuncture treatment to pure acupuncture. Many acupuncture studies typically involve electroacupuncture. Despite being labeled as acupuncture studies, some employ electroacupuncture. Notably, acupuncture and electroacupuncture treatments are frequently combined in clinical practice in Korea, highlighting the need for further research encompassing electroacupuncture. Additionally, as various acupuncture modalities like pharmacopuncture, cranial acupuncture, and auriculotherapy are being explored for PD management, future studies are anticipated to investigate diverse acupuncture techniques.

Lastly, regarding the assessment of bias risk, it is imperative to utilize sham acupuncture that can be compared with genuine acupuncture treatment while enhancing the quality of blinding for participants and personnel. Reliable sham acupuncture methods such as the “Park Sham needle” [24] or those referenced in studies utilizing the self-developed “nonpenetrating sham acupuncture” [37] could be employed. Moreover, a three-arm study featuring a waiting-list control group can offer a clearer understanding of treatment effects.

Through this study, we conducted a systematic review of research examining the impact of acupuncture on symptoms experienced by individuals with PD, followed by a meta-analysis to consolidate the findings. The meta-analysis affirmed that acupuncture treatment, when used in conjunction with conventional western medicine approaches, yielded notably beneficial effects in alleviating symptoms among PD patients. Consequently, acupuncture emerges as a viable primary or adjunctive intervention for individuals with PD who are undergoing conventional Western medical treatments.

Despite the aforementioned limitations, our study serves as foundational groundwork for future investigations into the efficacy of acupuncture as a therapeutic approach for managing PD symptoms in real-world clinical settings. Given the scarcity of systematic reviews assessing the efficacy of acupuncture treatment for PD, our findings hold promise for informing future research endeavors in this domain.

A systematic review and meta-analysis were conducted to evaluate the efficacy of combining acupuncture with Western medicine in treating patients with PD, focusing on studies published over a decade from January 2014 to August 2023. The analysis revealed notable enhancements in the overall UPDRS score among patients receiving acupuncture alongside Western medicine, with no significant AEs attributed to acupuncture reported in the study findings. However, due to the limited number of studies incorporated, definitive conclusions regarding acupuncture’s efficacy in real-world clinical settings remain elusive. To comprehensively assess the efficacy of acupuncture treatment for PD, further high-quality RCTs are imperative.

Conceptualization: YK, JUK. Data curation: YL, SA. Formal analysis: Yoomin Choi, JO. Funding acquisition: JUK, Yoomin Choi. Investigation: JUK. Methodology: Yoomin Choi, YL. Project administration: YK. Resources: JO, Yoonbyeong Chae. Software: SA, YL. Supervision: YK, Yoomin Choi. Validation: JUK. Visualization: Yoomin Choi. Writing – original draft: YL, JO. Writing – review & editing: YK, Yoonbyeong Chae.

The authors have no conflicts of interest to declare.

This study received support from Woosuk University.

This research did not involve any human or animal experiment.