Atrial fibrillation (AF) is an irregular, often fast heart rhythm in which the upper chambers of the heart quiver instead of contracting in time. It comes and goes in paroxysmal AF and lasts more than a week in persistent AF. Trials of acupuncture for AF measure one of three things: whether the rhythm returns to normal, called sinus rhythm, when acupuncture is added to drugs; whether AF comes back after cardioversion, an electric shock that restores sinus rhythm, or after catheter ablation, a procedure that scars the tissue where the irregular rhythm starts; and whether AF first appears in the days after surgery, called postoperative AF. The control differs from trial to trial and decides what a result can show: a sham procedure made to resemble acupuncture, no added treatment, or antiarrhythmic drugs. Sham acupuncture describes the kinds of sham in use. No trial on this page tested acupuncture in place of anticoagulation, the drugs that lower the risk of stroke in AF.
Is there a Cochrane review?
Cochrane has no review of acupuncture for atrial fibrillation. Its review of interventions to prevent AF after heart surgery, by Arsenault and colleagues and updated in January 2013, searched to July 2011 and included 118 trials with 17,364 participants. The interventions it found trials for were amiodarone, beta-blockers, sotalol, magnesium, atrial pacing and posterior pericardiotomy, a cut in the sac around the heart that lets fluid drain; acupuncture was not among them.
Systematic reviews
Four reviews have pooled randomised trials of acupuncture for AF, and an overview has graded them. The reviews share many of their trials, so they are not four independent confirmations.
| Review | Searched to | Trials (people) | Comparison | Pooled result | Certainty (GRADE) |
|---|---|---|---|---|---|
| Liu 2018 | February 2016 | 9 (638), all arrhythmias | Acupuncture alone against amiodarone, in AF and atrial flutter | No significant difference in response rate | Not graded |
| Fei 2019 | January 2018 | 8 (633) | Acupuncture against or added to another active treatment | More conversions to sinus rhythm with acupuncture | Not graded |
| Li 2022 | May 2021 | 11 (1,234) | Acupuncture added to drugs against drugs alone | More conversions overall and in paroxysmal AF; no significant difference in persistent AF | Not graded |
| Liu 2024 | November 2023 | 15 (1,960) | Acupuncture added to drugs, cardioversion or ablation against those alone or with sham | More conversions to sinus rhythm; slower ventricular rate | Not graded |
Liu and colleagues reviewed acupuncture for all heart rhythm disorders and rated only one of their nine trials at low risk of bias; they wrote that they could not reach a robust conclusion because the overall evidence was of low quality. Fei and colleagues rated all but one of their trials in their lowest quality grade, found only one that was double-blind, and noted that most of their trials came from China. Li and colleagues asked whether adding acupuncture to drugs helped AF convert to sinus rhythm, and reported the largest effect in trials whose sessions lasted 20 minutes or less. Liu and colleagues, in the largest of the four, found that 4 of their 15 trials reported blinding and 2 reported concealing the allocation; they noted that some antiarrhythmic drugs the trials used are not sold outside China, which makes the comparisons hard to interpret. Each review's main pooled outcome was conversion to sinus rhythm or a response rate, and none graded the certainty of its own results.
An overview by Ci and colleagues, published in December 2025, searched to December 2023 and appraised 10 reviews of acupuncture for heart rhythm disorders, 3 of them on AF alone. It applied GRADE, the scale reviews use to say how far a result can be trusted, to the 34 outcomes those reviews reported: none was rated high, 4 moderate, 13 low and 17 very low. The overview rated the methodological quality of the reviews themselves very low.
Trials after cardioversion or ablation
Three randomised trials have asked whether acupuncture changes how often AF returns once sinus rhythm has been restored.
| Trial | People | Control | Course | Result |
|---|---|---|---|---|
| Lomuscio 2011, Italy | 80 with persistent AF after cardioversion; 54 randomised | Sham, and no treatment; a non-randomised group already on amiodarone for reference | 10 weekly sessions | AF recurred within 12 months in 35% with acupuncture, 69% with sham, 54% with no treatment and 27% on amiodarone |
| Lee 2021, Korea | 44 with persistent AF after cardioversion; 31 completed | Sham | 12 weekly sessions of electroacupuncture | No significant difference in recurrence at 3 months; AF returned in more than half of each group |
| Yin 2019, China | 85 with persistent AF after pulmonary vein isolation, a form of ablation | Amiodarone alone; no sham | Daily for 7 days, with amiodarone | Early recurrence within 3 months in 12.5% with acupuncture and 33.3% without |
The Milan trial by Lomuscio and colleagues, published in 2011, needled PC06 (Neiguan), HT07 (Shenmen) and BL15 (Xinshu); the sham group attended the same sessions without those points being needled. After adjusting for ejection fraction, hypertension and the size of the left atrium, recurrence with acupuncture did not differ significantly from the amiodarone group, while recurrence with sham and with no treatment was significantly higher. The randomised groups held 17, 13 and 24 people, and the amiodarone group was not randomised. The Korean pilot trial by Lee and colleagues gave electroacupuncture at PC05 (Jianshi), PC06, ST36 (Zusanli) and ST37 (Shangjuxu); it found a reduction in the size of the left atrium with acupuncture and none with sham, and no serious adverse events. Its authors recommended that later trials enrol only people with symptoms and treat several times a week. The Wuhan trial by Yin and colleagues needled PC06 alone and measured recurrence only in the three-month blanking period after ablation, when early recurrences are common; inflammatory markers were also lower in the acupuncture group.
AF after surgery
The authors of the ACU-Heart trial wrote in 2023 that acupuncture had not been examined after cardiac surgery or for postoperative AF. Three trials have tested it around heart or lung operations.
| Trial | Surgery and people | Control | Result |
|---|---|---|---|
| ACU-Heart, Feingold 2023, United States | Heart valve surgery; 100 without previous AF | Standard care; no sham | Postoperative AF in 13.7% with acupuncture and 32.7% with standard care |
| Zhou 2025, China | Lung cancer surgery; 90, of whom 81 were analysed | Non-penetrating sham electroacupuncture | New AF within 72 hours in 2.5% with electroacupuncture, 14.6 percentage points fewer than with sham |
| Liu 2025, China | Thoracoscopic lung cancer surgery; 77 analysed | Needling at non-acupoints near the same sites | Supraventricular arrhythmia, AF included, within 24 hours in 10.5% and 33.3% |
ACU-Heart, a pilot trial at Northwestern University in Chicago, gave a mean of 3.8 inpatient sessions at nine body and ear points, PC06 and HT07 among them, starting the day after surgery. Its primary aim was feasibility. Patients and acupuncturists knew who received acupuncture, but a committee blinded to the groups decided which patients had postoperative AF from their records, and patients were monitored continuously until discharge. The groups differed in age, about 56 years against 61. The Shanghai trial by Zhou and colleagues gave three sessions at PC06, HT07, LI04 (Hegu) and TB05 (Waiguan) on the side of the operation, with outcome assessors blinded. The Guangzhou trial by Liu and colleagues stimulated PC06 and SP04 (Gongsun) from before anaesthesia to the end of surgery and again at 24 hours, and counted AF together with atrial flutter and other supraventricular tachycardias, fast rhythms that start above the ventricles, so it does not report AF on its own.
Safety and anticoagulants
Anticoagulants lower the risk of stroke in AF, and bleeding is their main complication. A chart review at a Korean stroke centre by Kwon and colleagues covered 316 patients and 10,177 acupuncture sessions in 2017; bleeding that stopped within 30 seconds occurred at similar rates in people taking newer oral anticoagulants, antiplatelet drugs or neither, and no major bleeding was recorded. The trials above that reported adverse events recorded none that were serious. Bleeding and anticoagulants covers needling in people on these drugs.
What guidelines say
The 2023 ACC/AHA/ACCP/HRS guideline on the diagnosis and management of AF, by Joglar and colleagues for the American College of Cardiology, the American Heart Association, the American College of Clinical Pharmacy and the Heart Rhythm Society, was published on 30 November 2023 and updates the 2014 guideline and its 2019 focused update. It adds recommendations on risk factor modification and AF prevention. Its full text does not mention acupuncture. The 2024 ESC guideline for the management of AF, by Van Gelder and colleagues for the European Society of Cardiology with the European Association for Cardio-Thoracic Surgery, was published in the European Heart Journal in 2024; its full text does not mention acupuncture either. No newer version of either guideline had been published by October 2026. Clinical guidelines compares other bodies’ positions.
What the reviews could not settle
The pooled trials mostly add acupuncture to drugs and compare the result with the drugs alone, without a sham and usually without blinding, so the reviews cannot separate the needling from the attention and expectation that come with it. Their main outcome, conversion to sinus rhythm during a course of treatment, is not the question for people whose AF has already been converted or ablated. On that question, the two sham-controlled trials disagree: Lomuscio 2011 found less recurrence than sham, Lee 2021 found no difference, and both were small. In Lomuscio 2011, AF returned more often after sham than after no treatment at all. The trials after surgery are single-centre, small, run around different operations and stop counting within days of the operation. No trial on this page measured stroke, hospital admission or death. Where the trials come from discusses what the country of a trial tells a reader.
A multicentre trial registered in China, whose protocol was published in 2024, plans to randomise 146 people after ablation to 18 sessions of acupuncture or sham over 12 weeks and follow them for 9 months, with quality of life as its primary outcome and AF burden and recurrence among its secondary ones.
AcuiQ’s atrial fibrillation, postoperative atrial fibrillation, arrhythmia and palpitations pages list the protocols individual studies prescribed, each with its citation.