Acupuncture for CTTH

…plus a protocol for CAT in migraine?

Inspired by Zheng et al 2022,[1] and the protocol by Li et al 2022.[2]

Photo by Andrea Piacquadio on

CTTH – chronic tension-type headache
TTH – tension-type headache (episodic or chronic)
ART – Acupuncture Randomised Trial (part of the Modellvorhaben Akupunktur)
GERAC – GERman ACupuncture trial (part of the Modellvorhaben Akupunktur)
APs – acupuncture points
CAT – contralateral acupuncture (with reference to migraine protocol)
IAT – ipsilateral acupuncture (with reference to migraine protocol)

key to acronyms

This week we have another paper from Chengdu with 3 of the same authors as last week. That was entirely a chance occurrence I can assure you. I will also briefly mention a protocol for a migraine trial based a little farther south in Kunming.

The main paper I am highlighting is the biggest controlled trial in TTH (n=218; 2 arms) since the two large 3-armed studies of the Modellvorhaben Akupunktur:[3] the ART TTH study (n=270; 3 arms);[4] and the GERAC TTH study (n=409; 2 arms – 3rd arm was stopped).[5]

Chronic TTH… 15 or more headache days in 4 weeks
Episodic TTH… less than 15 headache days in 4 weeks

This study from Chengdu included only CTTH, whereas the German trials included both episodic and chronic TTH. The difference is that CTTH requires 15 or more days with headache over 4 weeks and episodic TTH is less than 15 days over the same period. This is a relatively arbitrary distinction, and the actual baseline means of days with headache in the respective trials were as follows: Chengdu 21; ART 17.5; GERAC 14.

The treatment approaches were rather similar in that all three studies used manual acupuncture versus a superficial needling control. The interventions were either standardised (set protocol in Chengdu) or semi-standardised (set points plus optional others in the German trials). 20 sessions over 8 weeks were applied in Chengdu as opposed to 12 sessions over 8 weeks (ART) or 10 sessions over 6 weeks with the option of 5 more if there was a moderate (20% to 50%) response to the first 10 (GERAC). Approximately 10 to 15 needles were used in each session in all the trials.

In Chengdu the superficial control was applied to the same acupuncture points, but in the ART and GERAC trials, so-called non points were used away from classical APs and meridians. Personally, I doubt this makes any difference to the physiological stimulus, but it was the principal reason for performing this trial. The team from Chengdu wanted to measure the difference between deqi and no deqi at the same points rather than deqi at real points compared with no deqi and non-points, as in both ART and GERAC studies.

The results in the three trials are all relatively similar in terms of headache days, although they each used different primary outcomes and therefore their designation as positive or negative trials differed. In terms of headache days, the ART trial (the smallest and least statistically powerful) did not show a difference between standard and superficial needling. In the real acupuncture group (standard needling) headache days reduced from 17.5 to 9.9 and from the same baseline to 10.8 in the superficial needling group. In the GERAC trial (with over 200 in each arm, and therefore greater statistical power) headache days reduced from 15.6 to 6.2 in the real acupuncture group, and this was statistically superior to the superficial off-point needling group in which the reduction was from 16.4 to 8.5. In the Chengdu trial (with just over 100 in each arm) headache days reduced from 20.4 to 7.5 in the real acupuncture group and from 22.6 to 11.9 in the superficial acupuncture group.

RR50 = % of patients with a 50% or more reduction in days with headache over 4 weeks

So, good manual acupuncture seems to have a better effect in TTH and the responder rate can reach almost 70% (in Chengdu at least). A responder is defined as a patient who experiences a reduction in days with headache of more than 50% – sometimes this is referred to as the RR50. In the superficial needling group, the RR50 reached 50% by the end of the Chengdu trial. In the German trials the RR50 varied considerably based on different methods of analysis and reporting. In the ART trial all patients with missing data were categorised as non-responders and the RR50 was 46% and 35% in the standard and superficial needling groups respectively. In the GERAC trial these figures were either 66% and 55% or 33% and 27% depending on whether or not patients with changes to their medication were counted as non-responders or not. In Chengdu, missing values were imputed using a bootstrapping algorithm.

In TTH: the RR50 for standard needling ~70%
and the RR50 for superficial needling ~50%

Over my career of needling, my impression has been that migraine responds slightly better than TTH, excluding the miracle cures we see with myofascial headache. These results seem to suggest the results can be just as good if not better with TTH. I do have to admit that I have never used 20 sessions in 8 weeks of course.

That brings us to the migraine protocol just out in BMJ Open.[2] This aims to randomise 243 women with unilateral migraine without aura to one of 3 parallel arms: CAT; IAT; or sham using the Park sham device. The CAT and IAT groups will also use the Park sham device, but with real needles instead of a retractable non-penetrating needle.

The points to be used are all GB and TE meridian points: GB20; GB8; TE5; TE3; GB34; and GB41. With less than 100 in each needling arm, I would not expect any difference to be detected between CAT and IAT, but I would hope that there is a measurable difference between the real needling and the non-penetrating sham. I will report back when the results are out in a year or two.


1          Zheng H, Gao T, Zheng Q-H, et al. Acupuncture for Patients With Chronic Tension-Type Headache: A Randomized Controlled Trial. Neurology Published Online First: 22 June 2022. doi:10.1212/WNL.0000000000200670

2          Li Q, Feng J, Zhang X, et al. Efficacy of contralateral acupuncture in women with migraine without aura: protocol for a randomised controlled trial. BMJ Open 2022;12:e061287. doi:10.1136/bmjopen-2022-061287

3          Cummings M. Modellvorhaben Akupunktur–a summary of the ART, ARC and GERAC trials. Acupunct Med 2009;27:26–30. doi:10.1136/aim.2008.000281

4          Melchart D, Streng A, Hoppe A, et al. Acupuncture in patients with tension-type headache: randomised controlled trial. BMJ 2005;331:376–82. doi:10.1136/bmj.38512.405440.8F

5          Endres HG, Böwing G, Diener H-C, et al. Acupuncture for tension-type headache: a multicentre, sham-controlled, patient-and observer-blinded, randomised trial. J Headache Pain 2007;8:306–14. doi:10.1007/s10194-007-0416-5

Declaration of interests MC