Acupuncture for Chronic Pain: 2025–2026 Evidence on Dose, Mechanisms, and Functional Outcomes

article-type: evidence-review
target-keyword: acupuncture for chronic pain
review-window: 2024-01-01/2026-10-02
clinical-scope: dose + mechanisms + functional outcomes
status: educational; not a diagnosis or individualized treatment order

evidence snapshot

A 2025 evidence map in Frontiers in Medicine reviewed 111 systematic reviews covering 35 musculoskeletal pain conditions. Short-term outcomes generally favored acupuncture, but all included reviews were rated low or critically low in methodological quality. Long-term evidence remained insufficient for consistent pain improvement.[1]

The current evidence does not establish one universal acupuncture dose for chronic pain. Treatment response depends on condition, baseline impairment, protocol, outcome measure, and comparator. The most defensible clinical model is a monitored multi-session course with reassessment based on pain, function, sleep, activity, and quality-of-life outcomes.

dose-and-delivery evidence

A 2025 Journal of Pain Research systematic review and meta-regression included 12 randomized controlled trials and 1,155 participants with chronic musculoskeletal pain. Acupuncture produced a pooled Hedges’ g of 0.57, with substantial heterogeneity. No statistically significant dose-response relationship was identified for:

• total sessions: typically 6–20
• treatment frequency
• needle-retention time

For total sessions, the meta-regression coefficient was 0.002, with p = 0.909. The result indicates that increasing the dose within the ranges commonly tested in trials did not reliably increase the pooled effect size.[2]

The 2018 Vickers individual-patient-data update remains the most recent Vickers IPD meta-analysis through October 2026. It included 39 trials and 20,827 patients. A patient-level dose effect was observed in trials using no-acupuncture controls: the effect size increased by approximately 0.10 standard deviations per five additional treatments. No clear association was found for treatment frequency or session duration.[3]

A 2025 Norwegian prospective observational study reported a median of six treatments among patients with treatment-resistant chronic pain. Participants receiving six or more treatments showed stronger 12-month improvements than those receiving one to five treatments. The study lacked randomization, had substantial follow-up attrition, and could not exclude regression to the mean or treatment-context effects. The finding supports six sessions as a reasonable review point, not a universal minimum dose.[4]

A practical evidence-based range is therefore:

• initial course: 6–10 sessions
• frequency: commonly 1–3 sessions/week
• course length: approximately 4–8 weeks
• reassessment: after early functional change, not pain score alone
• continuation: based on measurable benefit, tolerance, and treatment goals

Daily treatment or courses exceeding 16–20 sessions have not shown proportionally superior outcomes across chronic musculoskeletal pain trials.

proposed mechanisms

Mechanistic evidence is derived primarily from laboratory studies, neuroimaging, physiological experiments, and translational reviews. These mechanisms remain proposed rather than established explanations for every clinical response.

• adenosine/a1 receptor signalling: needle stimulation may increase local adenosine, with activity at adenosine A1 receptors on sensory and glial cells. This pathway is associated with reduced nociceptive signalling in preclinical models.
• vagal-adrenal axis: low-intensity electroacupuncture at selected limb points has been proposed to activate sensory-autonomic pathways involving the vagus nerve, adrenal medulla, catecholamines, and inflammatory cytokine regulation.
• endogenous opioids: acupuncture may engage μ-, δ-, and κ-opioid systems, with frequency-dependent effects described mainly in experimental research.
• descending pain modulation: brainstem and spinal pathways involving endogenous opioids, serotonin, norepinephrine, the periaqueductal gray, and the rostroventromedial medulla may alter nociceptive transmission.
• neuroinflammation and central sensitization: modulation of microglia, astrocytes, cytokines, NMDA-related signalling, and synaptic plasticity has been proposed in neuropathic and inflammatory pain models.

A 2024 review of De-Qi and brain modulation described these pathways as biologically plausible but incompletely connected to patient-level outcomes.[5] Mechanistic plausibility does not establish that a specific point prescription, sensation, or electrical frequency will produce a predictable result in an individual.

condition-specific findings

chronic low back pain

A 2024 prospective study evaluated 60 patients with chronic non-specific low back pain. A subgroup received 10 treatments over four weeks, with a maximum of three sessions weekly. Temporal summation of pain did not predict response. After the treatment series, pain and Hannover Functional Ability Questionnaire scores improved, but the study was small and uncontrolled.[6]

The 2025 evidence map reported favorable short-term acupuncture outcomes for chronic low back pain compared with sham, usual care, medication-based controls, and no treatment. Long-term benefit remained less certain. Functional improvement was reported in multiple trials, including disability and activity measures.[1]

knee osteoarthritis

Evidence for knee osteoarthritis supports short-term improvements in pain and physical function, particularly compared with no-acupuncture controls. Results against sham acupuncture are less consistent. A 2025 randomized trial evaluated different acupuncture techniques for knee osteoarthritis, reinforcing that treatment type and dosing parameters remain active research questions.[7]

Relevant outcomes include:

• Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC)
• walking and stair-climbing capacity
• joint stiffness
• physical-function subscales
• patient global assessment

chronic neck pain

The 2025 evidence map found generally positive results for acupuncture compared with no-treatment or usual-care controls. Results were mixed for electroacupuncture versus sham or no treatment. Evidence for non-specific musculoskeletal neck pain is stronger than evidence for cervical radiculopathy, where guideline consensus remains limited.[1,8]

fibromyalgia

A 2026 meta-analysis of 17 randomized trials and 773 participants reported short-term pain improvement with an SMD of −0.72 and longer-term pain improvement with an SMD of −0.53. Stiffness also improved. Overall physical-function outcomes were not statistically significant, although regimens delivered at least twice weekly showed better short-term physical-function results in subgroup analysis.[9]

The findings remain limited by heterogeneity, small trials, publication bias for short-term pain, and mixed comparator designs. Acupuncture should be evaluated as one component of fibromyalgia care rather than as a stand-alone intervention.

chronic musculoskeletal pain

Across conditions, improvements in disability, physical function, health-related quality of life, work ability, and symptom interference are clinically relevant outcomes. Pain intensity alone does not capture whether a treatment changes daily activity or participation.

sham-controlled findings and response predictors

Sham acupuncture is not a uniform control. Penetrating needles, non-penetrating devices, superficial needling, and non-acupuncture locations may produce different sensory and physiological effects. The Vickers update found smaller acupuncture–sham differences when sham procedures included needle penetration or higher-intensity stimulation.[3]

There is no 2025 successor IPD meta-analysis from the Acupuncture Trialists’ Collaboration that replaces the 2018 Vickers update. Recent 2024–2026 evidence consists primarily of aggregate meta-analyses, evidence maps, predictor studies, and guideline reviews.

The most consistent response variables are:

• higher baseline pain severity
• greater baseline functional limitation
• early improvement during the initial treatment phase
• patient and practitioner expectations
• coping capacity and treatment engagement

The 2024 low back pain study found that temporal summation did not predict response. A 2025 fibromyalgia predictor study reported that pressure-pain sensitivity and perceived pain impact may predict short- and long-term response differently.[10] No validated clinical test currently identifies acupuncture responders with sufficient accuracy for routine treatment allocation.

functional and patient-reported outcomes

Trials increasingly report outcomes beyond pain intensity:

• disability: Roland-Morris Disability Questionnaire, Oswestry Disability Index
• joint function: WOMAC, Oxford Knee Score
• general health: SF-36, SF-12
• fibromyalgia impact: Fibromyalgia Impact Questionnaire
• neck function: Neck Disability Index, Northwick Park Neck Pain Questionnaire
• activity: walking, work ability, daily living tasks
• symptom burden: sleep, fatigue, stiffness, mood, and pain interference

A clinically meaningful response may involve improved walking tolerance, reduced activity limitation, improved sleep continuity, or increased work participation without complete pain elimination.

guideline position

A 2025 systematic review examined 17 clinical practice guidelines and 35 recommendations for chronic musculoskeletal pain. Sixty percent supported or suggested acupuncture; 54.3% were weak or conditional recommendations and 5.7% were strong recommendations. Another 22.9% offered no explicit guidance, and 17.1% advised against acupuncture in specified contexts.[8]

Current guideline positioning is therefore:

• chronic low back pain: usually an adjunctive option
• knee osteoarthritis: commonly considered for pain and function
• non-specific neck pain: possible adjunct; radicular pain less certain
• fibromyalgia: potentially beneficial; recommendation strength generally limited
• multimodal care: preferred over acupuncture as isolated treatment

safety-and-screening record

Acupuncture is generally associated with transient soreness, bruising, minor bleeding, or dizziness when delivered by a trained practitioner using sterile, single-use needles. Screening should document:

• anticoagulant or antiplatelet medication
• thrombocytopenia or known bleeding disorder
• pregnancy and pregnancy-related treatment restrictions
• infection, impaired immunity, or poor wound healing
• implanted electrical devices when electroacupuncture is being considered
• history of fainting or needle intolerance
• unexplained neurological symptoms

Medical evaluation is required before acupuncture for red-flag symptoms, including new progressive weakness, loss of bowel or bladder control, saddle anesthesia, fever with spinal pain, unexplained weight loss, history of cancer with new pain, major trauma, suspected infection, or acute chest pain or shortness of breath.

bottom-line/evidence-certainty block

dose-certainty: low-to-moderate
best-supported-course: approximately 6–10 sessions with reassessment
frequency-certainty: no universal optimum; 1–3 sessions/week commonly studied
condition-certainty: strongest for short-term musculoskeletal outcomes
functional-outcomes: shown in several conditions but heterogeneous
mechanisms: biologically plausible; predominantly proposed or preclinical
long-term-pain-certainty: limited
clinical-position: individualized adjunct within multimodal care

EnerQi root-cause assessment

EnerQi Wellness evaluates chronic pain through a broader history that includes symptom patterns, lifestyle factors, prior treatment response, sleep, stress physiology, movement, and inflammatory features. The Root Cause Method is designed for individualized acupuncture, Chinese herbal medicine, and personalized treatment protocols rather than a fixed point prescription.

The initial pathway is a 15-minute Root Cause Discovery Call. The purpose is preliminary assessment, identification of appropriate next steps, and determination of whether a comprehensive evaluation is indicated.

references/resources

1. Ang L, et al. Frontiers in Medicine. 2025;12:1575226. doi:10.3389/fmed.2025.1575226
2. Dose-response relationship between acupuncture parameters and pain outcomes. Journal of Pain Research. 2025. Full text
3. Vickers AJ, et al. Journal of Pain. 2018;19:455–474. doi:10.1016/j.jpain.2017.11.005
4. Lindberg V, Baak J. Frontiers in Pain Research. 2025;6:1521466. doi:10.3389/fpain.2025.1521466
5. Zhong Z, et al. Frontiers in Neuroscience. 2024;18:1386108. doi:10.3389/fnins.2024.1386108
6. Baeumler P, et al. Frontiers in Neurology. 2024;15:1335356. doi:10.3389/fneur.2024.1335356
7. Sun J, et al. Pain and Therapy. 2025;14:737–751. doi:10.1007/s40122-025-00713-x
8. Ho L, et al. BMC Complementary Medicine and Therapies. 2025;25:322. doi:10.1186/s12906-025-05070-y
9. Jin L, et al. Journal of Pain Research. 2026;19:568235. doi:10.2147/JPR.S568235
10. Murphy AE, et al. Journal of Integrative Medicine. 2025;23:152–158. doi:10.1016/j.joim.2025.01.005
11. Chen W, et al. CNS Neuroscience & Therapeutics. 2025;31:e70327. doi:10.1111/cns.70327
12. International Association for the Study of Pain. Acupuncture for pain relief.