Skip to content

Incidence of adverse effects during acupuncture therapy—a multicentre survey

Ernst et al. · Complementary Therapies in Medicine · 2003

📊Multicenter Observational Study👥n=409 patients⚠️Safety - High Impact

Evidence Level

MODERATE
75/ 100
Quality
4/5
Sample
4/5
Replication
3/5
🎯

OBJECTIVE

Assess the incidence of adverse events during acupuncture treatments in routine clinical practice

👥

WHO

409 patients treated by physicians and other professionals in Berlin, Germany

⏱️

DURATION

12 months (December 1995 to December 1996)

📍

POINTS

Routine acupuncture (10-15 needles per session)

🔬 Study Design

409participants
randomization

Acupuncture patients

n=409

Standard acupuncture treatment with 10-15 needles

⏱️ Duration: 12 months of follow-up

📊 Results in numbers

0%

Treatments with adverse events

0%

Patients who experienced adverse events

0%

Mild hemorrhage

0%

Hematoma

0%

Dizziness

Percentage highlights

11.4%
Treatments with adverse events
37.4%
Patients who experienced adverse events
2.9%
Mild hemorrhage
2.2%
Hematoma
1.0%
Dizziness

📊 Outcome Comparison

Main adverse events per treatment

Mild hemorrhage
2.9
Hematoma
2.2
Dizziness
1
💬 What does this mean for you?

This German study followed more than 400 patients during acupuncture treatments to assess safety. It found that acupuncture is generally very safe, with the majority of adverse effects being mild (such as small bleeding or hematomas). Serious events are extremely rare when acupuncture is performed correctly.

📝

Article summary

Plain-language narrative summary

This German multicenter study represents one of the most comprehensive investigations of acupuncture safety in routine clinical practice. Conducted between December 1995 and December 1996 in Berlin, the study followed 409 patients (279 women, 125 men) who received a total of 3,535 acupuncture treatment sessions. The researchers collaborated with 13 general medicine clinics and outpatient services, plus 16 other professionals, to systematically document all adverse events occurring during treatments. The study methodology was rigorous, using standardized questionnaires completed by therapists after each session.

Practitioners questioned patients in a structured manner during treatment ('How do you feel now?') and before each subsequent session ('How did you feel after the last acupuncture therapy?'). A checklist was provided that included hemorrhage, hematoma, infections, neurological abnormalities, fainting, vestibular symptoms, nausea, prolonged DeQi effect, and increased pain, with free space to document other observed effects. The results revealed that adverse events were observed in 402 treatments (11.4%) involving 153 different patients (37.4% of the total). Women had an average of 9.01 treatments, while men had 7.85.

The main adverse events were mild hemorrhage (2.9% of treatments), hematoma (2.2%), dizziness (1.0%), and other systemic symptoms (2.7%). Less frequent events included fainting (0.1%), nausea (0.2%), prolonged DeQi effect (0.5%), and increased pain (0.4%). A notable case was the report of aphasia lasting one hour after acupuncture. The most common systemic symptoms included fatigue (51 treatments), generalized sweating (11), feeling cold (3), isolated hand sweating (14), increased peristalsis (4), and feeling hot (3).

The study found significant correlations between minor hemorrhages and hematomas and the number of needles used. Hemorrhages were less frequent in female patients, while hematomas were less observed in young patients, who in turn expressed more pain. The use of acetylsalicylic acid led to a higher incidence of hemorrhage. Notably, no local infections were reported, despite the practice of several therapists of using sterile needles without disinfecting the skin before insertion.

No contact allergic reactions were observed either, despite the high prevalence of nickel allergy in the general population (12.9%). The study found no differences in adverse event incidence rates between physicians and other professionals, suggesting careful documentation and treatment by the participating professionals. The clinical implications are significant for patients and professionals. The study confirms that acupuncture is a safe therapeutic method when practiced according to established safety rules and in appropriate anatomical regions.

However, patients should be informed about possible minor adverse events, especially those using anticoagulant medications such as acetylsalicylic acid. The medical history before treatment should include questions about coagulation disorders, immunodeficiencies, possible cardiac diseases, pregnancy, medications in use, and existing anatomical abnormalities.

Strengths

  • 1Substantial sample of more than 3,500 documented treatments
  • 2Systematic methodology with standardized questionnaires
  • 3Multicenter study reducing individual biases
  • 4Prospective documentation of adverse events
  • 5Analysis by both treatments and patients
⚠️

Limitations

  • 1Possible underreporting bias by the therapists themselves
  • 2Absence of a control group for comparison
  • 3Possible selection bias among participating centers
  • 4Insufficient sample size to detect very rare adverse events
  • 5Lack of information on practitioner training
Dr. Marcus Yu Bin Pai

Expert Commentary

Dr. Marcus Yu Bin Pai

MD, PhD · Pain Medicine · Physical Medicine and Rehabilitation · Medical Acupuncture

Clinical Relevance

The safety profile of acupuncture is one of the most frequent questions I receive from colleagues before they refer patients, and this multicenter survey offers concrete data for that conversation. With more than 3,500 sessions prospectively documented, the findings allow us to state, on an empirical basis, that the adverse events observed are predominantly local and self-limited — mild hemorrhage in 2.9% of treatments, hematoma in 2.2%, dizziness in 1.0%. For daily practice in musculoskeletal rehabilitation, this directly guides pretreatment screening: patients on acetylsalicylic acid or anticoagulants deserve heightened attention to the number of needles and puncture sites. The absence of local infections, even without skin antisepsis in part of the cases, reinforces the safety of sterile disposable needles as a nonnegotiable standard.

Notable Findings

The dissociation between incidence per treatment (11.4%) and incidence per patient (37.4%) deserves attention: it means that a significant portion of patients experienced at least one adverse event over the course of treatment, although most of these events occurred in isolated sessions and with spontaneous resolution. The positive correlation between the number of needles and the incidence of hemorrhage and hematoma is actionable data — reducing the number of needles in patients with higher bleeding risk is a rational decision grounded in these data. The isolated report of transient aphasia of one hour should be remembered not as alarmism, but as sufficient reason for post-session monitoring in the first minutes. The absence of nickel allergic reactions, despite a 12.9% prevalence in the general population, suggests that the superficial and brief contact with the needle rarely triggers a clinically relevant response.

From My Experience

In my practice at the musculoskeletal pain clinic, I usually use this type of safety data in informed consent — not to frighten the patient, but to be transparent and professional. Anticoagulated patients receive a protocol with a reduced number of needles, points with lower vascular risk, and local compression for 30 seconds after withdrawal. I have observed that post-session dizziness, cited in 1% of treatments here, is more frequent in debilitated, anxious, or prolonged-fasting patients — I routinely advise them to eat lightly before the session. The finding about fatigue as the most common systemic event (reported in 51 sessions) corresponds to what I see: a portion of patients feel drowsiness and deep relaxation in the first sessions, which I usually frame as an expected autonomic response and not as a concerning adverse event. For acupuncture-naive patients, I start with 8 to 10 needles to assess tolerance before expanding the protocol.

PhD in Health Sciences, University of São Paulo. Board-certified in Pain Medicine, Physical Medicine and Rehabilitation, and Medical Acupuncture.

Full original article

Read the full scientific study

Complementary Therapies in Medicine · 2003

DOI: 10.1016/S0965-2299(03)00004-9

Access original article

Scientific Review

Marcus Yu Bin Pai, MD, PhD

Marcus Yu Bin Pai, MD, PhD

CRM-SP: 158074 | RQE: 65523 · 65524 · 655241

PhD in Health Sciences, University of São Paulo. Board-certified in Pain Medicine, Physical Medicine and Rehabilitation, and Medical Acupuncture. Scientific review and curation of every entry in this library.

Learn more about the author →
⚕️

Medical disclaimer: This content is for educational purposes only and does not replace consultation, diagnosis, or treatment by a qualified professional. Some information may be assisted by artificial intelligence and is subject to inaccuracies. Always consult a physician.

Content reviewed by the medical team at CEIMEC — Integrated Centre for Chinese Medicine Studies, a reference in Medical Acupuncture for over 30 years.