This meta-analysis follows the guidelines provided in the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) and proceeds based on the PICO principles for study design. P-Participants/population: Participants were eligible for clinical diagnosis of spastic CP without restriction of age, gender, or ethnicity. I - intervention(s): The study group was treated with acupuncture on meridian tendons based on the treatment given to the control group; C - Comparator (s)/control: The control group, which did not receive treatment of acupuncture on meridian tendons, was not limited to rehabilitation treatment modalities; O - Outcome (s): The main outcome indicators included the overall effectiveness, Modified Ashworth Spasticity (MAS) Scale, Gross Motor Function Measure (GMFM), Fine Motor Function Measure (FMFM), and Comprehensive Spasticity Scale (CSS).

An electronic search of databases, including MEDLINE/PubMed, Web of Science, the Cochrane Library, EMBASE, CBM, CNKI, WanFang, and VIP, was performed to collect randomized controlled clinical studies on the treatment of acupuncture on meridian tendons for CP from the beginning of the database construction until 31 October 2023. Medical Subject Headings (MeSH) and non-MeSH phrases were used, including acupuncture, meridian tendons, needle, needles and knives, prick, bladed needle, CP, spasticity, and their corresponding English keywords, and the search language was limited to Chinese and English. Taking the Pubmed database as an example, the detailed search strategy is shown in Table 1.

(1) Study type: randomized controlled trial (RCT) either in Chinese or English language, without limitation on the publication time, and including single-blind, double-blind, or non-blind methods. (2) Subjects: Cases with a diagnosis that meets the criteria set by the China Cerebral Palsy Rehabilitation Guidelines (19) and can be diagnosed as CP. (3) Interventions: The treatment group used Jingjin acupuncture therapy in combination with conventional treatment or pure Jingjin acupuncture therapy, and the control group utilized basic treatment. (4) Outcome evaluation indices: total effective rate, MAS Scale, GMFM, FMFM, and the CSS.

(1) Study type: Non-RCT studies. (2) Study population: Non-CP cases and animal experiment-type literature. (3) Interventions: the literature that did not meet the requirements. (4) Efficacy evaluation: the literature whose indicators did not meet the inclusion requirements. (5) Literature with duplicate publications, very low-quality literature, and unclear reporting of original literature were excluded.

The study was independently searched and screened by two evaluators, and then the results were compared, with third-party adjudication in case of disagreement. The results were analyzed by first author, year of publication, number of cases (treatment and control groups), intervention (treatment and control groups), test parameters, duration of treatment, and consistency at baseline. The quality of the literature was assessed using the “Risk of Bias Assessment” tool recommended by the Cochrane Handbook of Systematic Reviews, which assesses the risk of bias for seven items, including generation of randomized sequences, concealment of allocation, blinding of investigators and subjects, blinding of study endpoints, completeness of endpoint data, selective reporting of results, and other biases. Each entry was assessed for risk of bias and categorized as high, low, or uncertain.

The results were analyzed using RevMan 5.4 software. Measurement data were evaluated using the advantages ratio (AR), while continuous variable data were evaluated using mean difference (MD) as the effect size, both expressed with a 95% confidence interval (CI). First, a heterogeneity test was conducted. If I² ≤ 50%, the fixed-effects model was used; otherwise, sensitivity analysis and subgroup analysis were conducted first. Following this, either the fixed-effects model or the random-effects model was used directly, depending on the results of the analyses. Publication bias was assessed using a funnel plot, where a symmetrical plot indicated an absence of publication bias, and an asymmetrical plot suggested the presence of publication bias.

A total of 8,239 documents were searched across 9 databases. After checking and reading the titles and abstracts, 356 documents were initially screened. After a full-text screening, documents that did not meet the screening conditions were excluded, resulting in a final inclusion of 20 documents. The process of literature screening is shown in Figure 1.

All 20 trials were single-center clinical RCTs, with the number of patients in each study ranging from 36 to 137, for a total of 1,453 cases. The included literature met the predefined inclusion and exclusion criteria, and the basic characteristics of all the original studies are shown in Table 2.

Twelve of the 20 papers specifically described random sequence generation methods, 8 (24, 30–36) used random number table methods, 2 (23, 27) used systematic randomization, 1 (29) applied random assignment card methods, and 1 (18) used paired design random table methods. In terms of concealment of protocol allocation, only 1 (29) mentioned concealment of protocol allocation by using numbered sealed envelopes, and none of the blinding methods were explicitly applied. A total of 20 papers had complete data without missing, and selective reporting was complete and unbiased. In terms of other biases, five (21, 23, 27, 34, 37) documents did not report whether the baseline data were comparable. A literature risk of bias assessment map was generated by RevMan 5.4 software, as shown in Figures 2, 3.

Publication bias was assessed by plotting a funnel plot of the total effectiveness rate of the outcome metrics for the most included studies. The 10 studies were distributed roughly symmetrically around the center line, indicating that publication bias was insignificant and had a negligible effect on the amount of combined effect (see Figure 11).

Spastic CP is divided into spastic quadriplegia, spastic hemiplegia, and spastic diplegia. The site of damage in spastic CP is mainly concentrated in the pyramidal system. When the lesion damage in the brain involves the motor neurons in the cerebral cortex, the motor neurons in the spinal cord lose the constraints of the higher motor neurons (43). This results in the weakening of the downward inhibitory effect and a relative enhancement of the excitatory effect of the peripheral afferent nerves. Consequently, there is a decrease in the excitatory threshold of the detrusor receptors, leading to the stimulation of the muscle shuttles to activate the α-motor neurons and the contraction of the extra-shuttle muscles. This phenomenon is manifested as abnormal spasticity of the limbs. Therefore, reducing muscle tone and relieving limb spasticity are essential for treating spastic CP (44).

According to TCM, the pathogenesis of CP is mostly due to the congenital insufficiency of the endowment of the child. This results in a lack of nourishment of the liver, spleen, kidneys, and heart, resulting in a deficiency of qi and blood in these organs. This deficiency further leads to the loss of nourishment of the tendons and veins of the affected child, resulting in bone softness and weakness. Additionally, there is a loss of fullness in the medulla oblongata and lack of mental clarity, which manifests as developmental delays, impairment of limb movement, language disadvantage, and mental retardation. At the same time, dysfunction of the four organs (liver, spleen, kidney, and heart) also leads to long-term insufficiency of qi and blood in the children’s meridians and skin, which is not moistened. This condition, combined with phlegm and stasis blocking the collaterals, prevents qi and blood from running smoothly through the meridians. This creates a vicious circle, making CP an intractable disease (45).

Jingjin acupuncture therapy is based on the principle of “treating with burnt needles and robbing and stabbing, counting with knowledge and losing with pain.” It combines this principle with the theory of Jingjin, using custom-made needles to superficially stab the subcutaneous fascial layer. This process aims to promote the body’s surface Wei qi, soothe and regulate meridians and tendons, and apply treatment to meridian tendons. This approach is a special kind of acupuncture method (46), characterized by its ease of operation and rapid onset of action. It is known for being easy to operate, fast-acting, minimally painful, and highly accepted by patients. In clinical use, compared to ordinary millimeter needling, it is safer and more effective, with better comfort and less resistance to needling in the treatment process (47). Jingjin needling involves a transverse, shallow puncture of the subcutaneous fascia. This technique aims to promote Wei qi, and to dredge and regulate the meridians and tendons (48). Wei qi is the key factor for the effectiveness of Jingjin acupuncture, and the technique of superficial subcutaneous stabbing with Jingjin needles is a means to regulate the operation of Wei qi. Jingjin acupuncture is gentle, and does not cause an increase in muscle tension in children, as it does not elicit a strong feeling of acidity and distension. In the practical application of rehabilitation therapy, we can look for positive reaction points (muscle tension points, muscle fiber nodules, and subcutaneous strips) along the direction of muscle fibers in the lesion area of the child and then insert the needle along the muscle direction, with the needle pointing in the direction of the lesion area.

The results of this study showed that the experimental group exhibited a significantly higher total effective rate of spasticity improvement, GMFM score, and FMFM score compared to the control group. Additionally, the MAS score and CSS score were significantly lower in the experimental group than in the control group, these findings suggest that Jingjin acupuncture therapy can significantly reduce the degree of spasticity in patients with CP, improve their motor function, and enhance their ability to perform daily life activities. These results provide a basis for the argument in favor of using Jingjin acupuncture therapy to treat spasticity in CP patients.

The large heterogeneity observed between studies on the MAS and GMFM scales may be due to differences in study quality. Subgroup analyses were performed according to the high and low quality of the literature. While the heterogeneity of the high-quality literature disappeared, it persisted in the lower-quality literature. This suggests an increased likelihood of false-negative and false-positive results regarding significance. This meta-analysis differed in the setting of the control group, which in the 20 studies consisted of routine rehabilitation (18, 23, 28, 29, 31, 35, 36, 38), acupuncture therapy (16, 20), and traditional rehabilitation therapies such as external immobilization in plaster. Conventional rehabilitation was the most frequently used, and although all rehabilitation therapies were utilized, the specific implementation of the course of treatment, location, and means of rehabilitation were not the same. In addition, there are some differences in the treatment group settings in this study. Due to the diversity of the means of Jingjin acupuncture therapy, the treatment measures of the treatment groups in the 20 studies, which included therapies such as needle knife (21, 26, 31), millimeter needle needling (34), and Zhuang medicine picking and stabbing (32), were not completely standardized. There are also fewer studies included in the FMFM and CSS scales, which may increase the probability of bias.

First, the 20 studies included in the meta-analysis were all Chinese studies, with no foreign studies, so differences in implementation at home and abroad could not be more fully evaluated. Second, there is a lack of use of the Cochrane-recommended GRADE methodology to assess the quality of evidence. Third, some studies had small overall sample sizes, there were no large samples or multicenter studies, random allocation methods were not described in detail in some studies, and few studies involved allocation concealment or blinded designs. Measurement bias and implementation bias are present, which may exaggerate clinical treatment effects. Fourth, the presence or absence of dropout cases was not mentioned in each study. In addition, most of the studies were not followed up to observe long-term efficacy. Fifth, some studies had unequal sample sizes in the control and trial groups, which may pose a risk of bias. Sixth, some studies were of low quality, and there may be inaccuracies in the interventions or testing indicators during the study. The above shortcomings may have a certain impact on the evaluation results and limit the strength of the argument of this study to a certain extent.

Steps to make the treatment more scientific and standardized are as follows: (1) standardizing clinical diagnostic criteria, inclusion criteria, intervention criteria, and efficacy criteria. (2) designing reasonable and scientific, conducting high-quality multi-center, large-sample randomized controlled clinical trials, and paying attention to describing the method of random allocation, allocation concealment, blinding, adverse reactions, and follow-up. (3) studying in-depth the timing, selection, depth, intensity, manipulation, and course of acupuncture points.