Article Text


Summaries of recent papers
  1. Adrian White1,
  2. Prem Bajaj2,
  3. Michael Meinen3,
  4. Amer Sheikh4
  1. 1Primary Care Group, Plymouth University Peninsula Schools of Medicine and Dentistry, Plymouth, UK
  2. 2Modern Health, Uxbridge, UK
  3. 3Denton Park Medical Group, Denton Park Health Centre, Newcastle Upon Tyne, UK
  4. 4Ashford Health Centre, Ashford Hospital, Ashford, Middlesex, UK
  1. Correspondence to Dr Adrian White, N21 ITTC Building, Tamar Science Park, Plymouth PL6 8BX, UK; Adrian.white{at}

Statistics from

Migraine prophylaxis

▸ Facco E, Liguori A, Petti F, et al. Acupuncture versus valproic acid in the prophylaxis of migraine without aura: a prospective controlled study. Minerva Anestesiol 2013;79:634–42.

An open-label randomised controlled trial (RCT) (n=100).


A total of 100 consecutive patients with migraine without aura were recruited from outpatient departments in Italy. They were randomised into two equal groups to receive acupuncture (group A), or valproic acid (group V) which is an antiepileptic drug shown to be effective in migraine prophylaxis.

Acupuncture was carried out by experienced practitioners using manual stimulation to achieve de qi and needles retained for 30 min. Acupuncture points were chosen according to Traditional Chinese Medicine (TCM) syndrome theory. Patients in group A received 2 courses of 10 acupuncture treatments, twice a week, with 1 week between the 2 courses. Patients in group V were treated with 600 mg/day of valproic acid for 3 months in order to correlate with the time span taken for treatment in group A. Both groups were allowed to take risatriptan for relief in acute attacks of migraine.

The primary outcome measure was the Midas Index, which measures negative patient experiences including disability, and has been shown to correlate with quality of life. Secondary outcomes included the Pain Relief Score (PRS), risatriptan intake and pain intensity (PI) on a 0–10 scale. Measurements were taken at baseline (T0), and at 3 months (T1) and 6 months (T2) from the start of treatment.


Both groups showed similar and significant improvements in Midas Score at T1 (p=0.1) and T2 (p=0.1) compared to baseline, as well as improvement in the number of days with pain (p=0.63 at T1 and p=0.1 at T2), shown in figure 1. PI was better at T1 in group V (p<0.0001), whereas group A showed better improvement at T2 for PI (p=0.02) and PRS (p=0.02). Risatriptan use and PI both decreased from T1 to T2 in group A, whereas they increased in group V. Interestingly, there were no reported adverse events in group A, compared to 20 patient issues in group V.

Figure 1

Changes in migraine with acupuncture (A) and valproate (V). Based on Minerva Anestesiol 2013;79:634-42.


This study corroborates the findings of previous studies showing the efficacy of acupuncture and valproic acid in migraine prophylaxis. The efficacy of valproic acid in terms of reduction in PI and acute medication use was shown to be greatest during the 3-month period during which it was taken. In contrast, acupuncture efficacy continued beyond the treatment period to at least 3 months afterwards. The authors note that one of the limitations of this study was a lack of a placebo group, but this would have been against the Declaration of Helsinki.


▸ Qu SS, Huang Y, Zhang ZJ, et al. A 6-week randomized controlled trial with 4-week follow-up of acupuncture combined with paroxetine (PRX) in patients with major depressive disorder. J Psychiatr Res 2013;47:726–32.

Three-arm RCT (n=160), open-label.


In this 6-week, randomised, controlled trial with 4-week follow-up, 160 patients with major depressive disorder were randomly assigned to paroxetine (PRX) alone (n=48) or combined with 18 sessions of manual acupuncture (MA, n=54) or electroacupuncture (EA, n=58). MA used 10 points: GV20, EX-HN2 (Yintang), GV16, GV14, GB20, PC6, SP6, with stimulation to elicit de qi. In the EA group, after manual stimulation to the same points, EA was added: 2/100 Hz EA between GV20 and EX-HN2, and between bilateral GB20. In all, 18 sessions were given over 6 weeks.

Treatment outcomes were measured mainly using the 17-item Hamilton Depression Rating Scale (HAMD-17), Self-rating Depression Scale (SDS), clinical response and remission rates. Average PRX dose taken and proportion of patients who required an increased PRX dose (up to maximum 40 mg/day) due to symptom aggravation were also obtained.

Figure 2

Hamilton Depression Scale scores. EA, electroacupuncture; MA, manual acupuncture; PRX, paroxetine. Based on J Psychiatr Res 2013;47:726–32.


Additional MA and EA both produced a significantly greater reduction from baseline in score on HAMD-17 and SDS (figure 2). The clinical response was markedly greater in MA and EA (both 70%) groups than the group treated with PRX alone (41.7%, p=0.004). The proportion of patients who required an increase dose of PRX due to symptom aggravation was significantly lower with MA (6%) and EA (9%) than PRX alone (23%, p=0.019). At 4 weeks follow-up after completion of acupuncture treatment, patients with EA, but not MA, continued to show significantly greater clinical improvement. Incidence of adverse events was not different in the three groups.


A clear effect of acupuncture overall was seen in this well conducted study.

Benign prostatic hyperplasia

▸Wang Y, Liu B, Yu J, et al. Electroacupuncture for moderate and severe benign prostatic hyperplasia (BPH): a randomized controlled trial. PLoS ONE 2013;8:e59449.

Sham-controlled RCT (n=97).


Men with benign prostatic hyperplasia (BPH) and an International Prostate Symptom Score (IPSS) ≥8 were enrolled to receive either 16 sessions of electroacupuncture (EA) at BL33 bilaterally, or the same number of treatments with needles inserted 2 cun laterally to BL33 as control over a period of 4 weeks. Outcome measures were improvement in IPSS as well as post-micturition residual volume (PVR) and maximum urinary flow (Qmax).


At 6 weeks, IPSS had improved from 20.16±6.52 to 12.84±5.87 in the acupuncture group and from 18.76±6.06 to 16.42 in the control group (p<0.001), see figure 3. The authors stated that there was further significant improvement after 18 weeks, although they did not report the data. No significant improvements were seen in PVR and Qmax. There were only two adverse events in the study (minor haematoma). The authors concluded that EA was effective in improving the symptoms of BPH.

Figure 3

Prostate symptoms (International Prostate Symptom Score (IPSS)). Based on PLoS ONE 2013;8:e59449.


This was a small study, marred by a high dropout rate, without the benefit of a published power calculation. I was surprised that needling and stimulating just 2 cun laterally from the traditional point could have such a dramatic difference in response. Perhaps an inert control might have shown an even greater difference. I would also have welcomed a longer follow-up period. BPH is a common and chronic condition, and 16 EA sessions will be costly; it would therefore be beneficial to know how long the benefit may last. Most puzzling of all was how can a treatment that does not improve residual volume or flow bring about an improvement in symptoms?

Postcolectomy ileus

▸Deng G, Wong WD, Guillem J, et al. A Phase II, randomized, controlled trial of acupuncture for reduction of postcolectomy ileus. Ann Surg Oncol 2013;20:1164–9.

Sham-controlled RCT (n=90) in patients with bowel cancer.


Patients undergoing elective resection of bowel cancer were randomised to either true acupuncture (n=46; 30 min of treatment at ST36, PC6, LI4, SP6, SP9, ST25, with electroacupuncture (EA) at ST36 and PC6) or non-penetrating sham at the same points (n=44). Primary endpoint was GI-3 (the later of either time at which the patient first tolerated solid food, or when first passed flatus). Secondary endpoint was GI-2 (the later of either time when patient first tolerated solid food or first passed a bowel movement). Pain, nausea or need for insertion of a nasogastric tube were also assessed.


Seven patients withdrew in the acupuncture group and two in the sham group. Time to GI-3 was 149 h (SD 71 h) in the acupuncture group and 146 h (SD 62 h) in the sham group (p=0.9), figure 4. Time to GI-2 was 149 h (SD 71 h) in the acupuncture group and 152 h (SD 60 h) in the sham group (p=0.8). Results for other endpoints were also not significant. The authors concluded that acupuncture does not make a significant difference in the prevention of a postoperative ileus. However, they conceded that their study may have been underpowered as the standard deviations in their data were much larger than anticipated.

Figure 4

Time to flatus and bowel movement post colectomy. Based on Ann Surg Oncol 2013;20:1164–9.


This was indeed a relatively small study, although well planned and carried out, and a larger study population may well have given better results. However, a larger study would of course be more expensive. It will be interesting to see whether such a study is performed, and if so what results it will show.

Preoperative or postoperative electroacupuncture for prevention of nausea?

▸Lee S, Lee MS, Choi DH, et al. Electroacupuncture on PC6 prevents opioid-induced nausea and vomiting after laparoscopic surgery. Chin J Integr Med 2013;19:277–81.

RCT (n=178)


A total of 178 patients who had received intravenous patient-controlled analgesia with fentanyl were assigned randomly to 3 groups: a preoperative electroacupuncture (EA) group (PrEA), a postoperative EA group (PoEA) and a non-acupuncture control group. The main outcomes were severity and frequency of postoperative nausea and vomiting, which were measured with a self-reported questionnaire and confirmation from a blinded anaesthetist.


The incidence of nausea and vomiting was significantly lower in the preoperative EA group than both the other groups during 48 h after surgery.


Preoperative EA seems most successful in preventing nausea, but the full text of this paper was not available for review.

Postoperative knee swelling

▸Mikashima Y, Takagi T, Tomatsu T, et al. Efficacy of acupuncture during post-acute phase of rehabilitation after total knee arthroplasty. J Tradit Chin Med 2012;32:545–8.

Open-label RCT (n=80).


Following total knee arthroplasty, 80 patients were randomly assigned to either an acupuncture or a no-acupuncture control. Acupuncture points were ST31, ST32, ST38, SP6, BL23, BL25, BL37 and BL57, with manual stimulation only, treated three times/week from postoperative day 7 until day 21. Outcome measures were pain visual analogue scale (VAS), swelling (ratio of circumferences of both knees) and range of movement (ROM).


There was no difference in pain scores on the 6th postoperative day, but on days 14 and 21 the acupuncture group scored significantly less pain. The acupuncture group showed significantly greater reduction in swelling (see figure 5) as well as faster return to full ROM.

Figure 5

Postoperative knee swelling: mean ratio of knee circumferences. Based on J Tradit Chin Med 2012;32:545–8.


There is no mention of blinding of the observer.

Hot flushes in patients with breast cancer

▸Bokmand S, Flyger H. Acupuncture relieves menopausal discomfort in breast cancer patients: A prospective, double blinded, randomized study. Breast 2013;22:320–3.

Three-arm RCT: acupuncture, sham and no acupuncture, (n=94).


We randomised 94 women (mean age 61 years) into the study with breast cancer and troublesome flushes whether or not they were on antioestrogen treatment (57 were taking some form). In all, 31 had acupuncture, 29 had sham acupuncture and 34 had no treatment. True acupuncture points were PC6, KI3, SP6 and LR3, with manual stimulation; sham acupuncture was superficial insertion into nearby locations that were non-points. Treatments were for 15–20 min weekly for 5 weeks.

Participants used 3-day logbooks to rate the extent of symptoms on a visual analogue scale (VAS) after each treatment and at 6 and 12 weeks after the end of treatment. Night disturbance was scored as yes or no.

Plasma oestradiol was measured in acupuncture groups before and after the first acupuncture treatment, and after the last; and in the control group at baseline and 5 weeks. Side effects of the treatment were registered.


In the acupuncture group, 16 patients (52%) experienced a significant effect on hot flushes compared with 7 patients (24%) in the sham group (p<0.05). The effect was seen after the second acupuncture session and lasted for at least 12 weeks after the last treatment, (figure 6). In addition, a statistically significant positive effect was seen on sleep in the acupuncture group compared with the sham-acupuncture and no-treatment groups.

Figure 6

Hot flush disturbance, visual analogue scale (VAS). Based on Breast 2013;22:320–3.

There was no increase in plasma oestradiol, which is reassuring.


Previous sham-controlled studies in this population have shown no difference, so this positive finding of a quick-acting and lasting effect of straightforward standardised acupuncture is interesting. Unfortunately, details of manual stimulation (especially de qi) were not given; and hot flushes were not measured using a validated method. However, the inclusion of a no treatment group was masterly design.

It is reassuring that acupuncture did not increase oestrogen levels in patients with breast cancer.

Stroke (subacute)

▸Zhu Y, Zhang L, Ouyang G, et al. Acupuncture in subacute stroke: no benefits detected. Phys Ther. Published Online First: 30 May 2013. doi:10.2522/ptj.20110138

Multicentre, single-blinded, RCT (n=188): acupuncture as an adjunct to rehabilitation.


This study from Nanjing included 188 patients with subacute stroke (<30 days) admitted to hospital. They were randomised into an acupuncture group and a conventional rehabilitation only group. All patients underwent conventional stroke rehabilitation (about 4 h per day, 5 days a week) from physiotherapists, occupational and speech therapists. A combination of body and scalp acupuncture was used 5 days a week for 3 months in the acupuncture group: HT1, PC6, LI4, Baxie (EX-UE9), ST31, ST32, SP9, SP6, ST36 LR3; scalp points MS6 and MS7; plus symptom-related points, for example, for facial paralysis (ST4, ST6, ST7) or dysarthria (GV15, GB20, EX-HN12, EX-HN13, CV23, HT5). De qi was achieved by manual stimulation.

The Fugl–Meyer Assessment (FMA) and Barthel Index (BI) were performed at baseline, 1 month, 3 months and 6 months after inclusion in the study by physiotherapists blinded to group allocation.


Significant improvements were found in each group following treatment. But no statistically significant differences were found between the groups using the FMA motor scores and the BI scores at baseline, 1 month, 3 months and 6 months (see figure 7).

Figure 7

Upper limb Fugl–Meyer Assessment (FMA) scores Based on Phys Ther 2013;doi:10.2522/ptj.20110138


This is a well conducted study that reached its estimated sample size (with only 3% dropout!), the largest study in stroke to date. Literature reviews have reached different conclusions in the past about the effectiveness of acupuncture in acute or subacute stroke. The intensive conventional rehabilitation in this study may have achieved the maximum possible central nervous system stimulation in these patients, that is, a ‘ceiling’ effect.

Systematic reviews

▸Urroz P, Colagiuri B, Smith CA, et al. Effect of acute acupuncture treatment on exercise performance and postexercise recovery: a systematic review. J Altern Complement Med 2013;19:9–16.

Review of four RCTs: three on exercise performance and one on recovery.


In a crossover study, 17 healthy volunteers carried out 4 bicycle exercise tests with, in random order: no electroacupuncture (EA); 30 min EA (2 Hz with a current of 1–2 mA) at PC5–PC6; the same at LU7–LI4; and the same at GB37–GB39 (intended as control). EA at PC5–PC6 and LU7–LI4 significantly reduced exercise induced increases in blood pressure, rate–pressure product and peak power output versus control. An effect was seen in approximately 70% of subjects. In the second study, (n=20) 20 min MA at ST36, GB34, LI11, LR3 and GV20 compared with sham and no acupuncture had an effect on rating of perceived exhaustion, but not exercise performance on cycle ergometer or blood lactate concentrations. In the third study (n=10) acupuncture at GV20, LI15, LI13, PC6, ST36, SP6 compared to nearby sham points had no effect on physiological parameters.

Another study (n=30) on post-exercise recovery used treadmill exercise and compared acupuncture with sham and control. The acupuncture group (PC6 and ST36) had significantly lower maximum heart rate, VO2max and blood lactic acid than the sham and the normal groups at the 30th min post exercise. Blood lactic acid of the acupuncture group was also significantly lower than that of the other two groups in the 60th min post exercise


The sample size may have been too small to produce significant results. There are inherent difficulties in standardisation of the type of individuals in terms of their training and therefore comparing different studies are difficult. Effects originating from acupuncture control points, and not calculating ‘a priori’ sample sizes require more rigorous study designs and methods to demonstrate the usefulness of Acupuncture treatments

Safety: systematic review

▸ Xu S, Wang L, Cooper E, et al. Adverse events of acupuncture: a systematic review of case reports. Evid Based Complement Alternat Med 2013;doi:10.1155/2013/581203

This review updates a previous report, and covers 2000–2011.

Infections were the main complication of acupuncture, including outbreaks due to reuse of needles, and nine cases of methicillin-resistant Staphylococcus aureus (MRSA) infected by their acupuncturist with poor technique (table 1). There were no reported cases of hepatitis.

Table 1

Acupuncture adverse events worldwide, 2000–2011

One patient died of infection: the hospital had diagnosed neutropoenia but not informed his general practitioner, who gave the acupuncture. One pneumothorax case was fatal, as was a case in which an abdominal needle created a direct communication between the aorta and the gastrointestinal (GI) tract.

In one case, pancreatitis was caused by three 13 cm needles inserted into the abdomen. Injuries are still being caused by migration of indwelling needles.

Safety: new case reports: epidural abscesses, cardiac tamponade, myositis ossificans, bilateral pneumothoraces

▸ Yu HJ, Lee KE, Kang HS, et al. Teaching NeuroImages: Multiple epidural abscesses after acupuncture. Neurology 2013;80:e169.

An 80-year-old woman presented with 2 days of progressive quadriparesis, difficulty voiding and fever. She had received acupuncture in the cervical and lumbar regions for 5 days beginning a week earlier. Whole-spine MRI showed multiple epidural abscesses at levels C3–C7, L3–L5 and L5–S1. Broad-spectrum antibiotics were administered, and the abscesses were surgically drained. Strains of Staphylococcus aureus were isolated.

▸ Her AY, Kim YH, Ryu SM, et al. Cardiac tamponade complicated by acupuncture: hemopericardium due to shredded coronary artery injury. Yonsei Med J 2013;54:788–90.

A 62-year-old man had cardiac tamponade due to coronary artery injury after acupuncture into the substernum. After resuscitation of cardiac arrest, we performed emergency pericardiocentesis. Nevertheless, cardiac arrest recurred and an emergency cardiopulmonary bypass was performed. We identified haemopericardium due to shredded acute marginal branch of right coronary artery, which was successfully ligated. The patient was discharged without any other complications.

▸ Lee DG, Lee SH, Hwang SW, Kim ES, Eoh W. Myositis ossificans in the paraspinal muscles of the neck after acupuncture: a case report. Spine J 2013;13:e9–e12.

A 26-year-old woman presented with posterior neck pain that had begun 2 months earlier with neck swelling after acupuncture. No abnormal finding was reported after x-ray or MRI. Computed tomography-guided biopsy was performed and a diagnosis of myositis ossificans was made. The patient was conservatively treated through rest and analgesics. Neck pain and swelling improved over several months.

▸ Harriott A, Mehta N, Secko M, et al. Sonographic diagnosis of bilateral pneumothorax following an acupuncture session. J Clin Ultrasound. Published Online First: 9 May 2013. doi: 10.1002/jcu.22046.

A 57-year-old woman presented with acute onset of chest pain and dyspnoea, which started while undergoing acupuncture for neck pain. A bedside ultrasound revealed bilateral pneumothoraces, which were confirmed radiographically.

Basic research

▸ Choi YJ, Lee JE, Moon WK, et al. Does the effect of acupuncture depend on needling sensation and manipulation? Complement Ther Med 2013;21:207–14.


A total of 53 healthy volunteers received 3 different forms of acupuncture in a single-blinded crossover design: superficial needling (0.3 cm), deep needling (2 cm) and needling with bidirectional rotation. The effects of acupuncture were evaluated by using the pressure pain threshold. Acupuncture sensation was measured.


Total acupuncture sensation and increase of the pressure pain threshold were maximum in needling with rotation, followed by deep needling and superficial needling. Further, there was a significant correlation (p=0.002) between the acupuncture sensation in the three groups and changes in pressure pain threshold (see figure 8). However, between individuals, the ‘responders’ who experienced the greatest sensation for any particular level of stimulation did not also show the greatest analgesia.

Figure 8

Correlation between acupuncture sensation and pain threshold. PPT, pressure pain threshold. Based on Complement Ther Med 2013;21:207–14.


This is novel work, but the finding is only surprising in that the analgesia response for any individual cannot be predicted from the sensation they experience.

View Abstract


  • Competing interests None.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.