Acupuncture in patients with tension-type headache: randomised controlled trial

The authors speculate that participants have not been unblinded by
the publication of the detailed study protocol before the trial was
completed. They maintain that "unblinding of patients recruited or
observed after the protocol publication by reading details .... in a
specialized medical journal or in the abstract in the internet is
theoretically possible but highly unlikely."

What exactly does "highly unlikely" mean statistically and how can
this assumption be evidence based? Quite the opposite seems to be true,
because the study plan was freely available in German libraries to all
patients and the internet sources could be found easily by a simple google
or pubmed search

http://www.angelfire.com/sc/naturheilverfahren/cgi-bin/s1.htm

http://www.angelfire.com/sc/naturheilverfahren/cgi-bin/s4.htm

http://content.karger.com/ProdukteDB/produkte.asp?Aktion=ShowPDF&Produkt... 229609&ArtikelNr=73473&filename=73473.pdf

The authors now mention an additional analysis restricted to the
patients in whom data collection for the main outcome measure was
completed before the protocol publication. They say that "this analysis
yielded results which were almost identical to those of the analysis
including all patients."

Since they do not give any figures for this additional analysis its
results must be disputed. The term "almost identical" is rather
unscientific.

The study under discussion did not observe good clinical practice
rules.

Competing interests:
None declared

This brave work confirms a long line of studies that Godfrey et al
1978 started in Toronto when an experienced physician, a graduate of
Shanghai 2nd Medical College, was flown in to administer ‘appropriate’
acupuncture for relief of chronic dull non-inflammatory pain: the benefits
he provided were compared with those achieved by Canadian practitioners
who not only were ignorant of acupuncture but had been misinformed as to
where the patient’s site of pain actually was. A screen separated the
patient’s body from the head and neck so he or she could not spot which
practitioner was which. A communicator explained in a bland voice why
acupuncture required needles to be placed where they were even if they
were at the other end and on the opposite side of the body. There was no
significant difference between the two groups: in excess of 56% of both
groups obtained relief.

The fact that people in pain obtain so much relief when needled suggests
this stimulus is of value wherever it is applied in relation to the site
of pain.

A lot goes on in the brains of those who volunteer to have a needle
inserted, especially those who volunteer for a trial. Who would normally
wish to lie still when someone advances upon them with a needle,
especially when they might receive the ‘wrong’ treatment? A considerable
degree of inhibition must be in place in the minds of those to allow this
treatment to occur. Attention in the patient's mind is paid to the
effects of needling, which anyone will remember if they have been
subjected to a splinter or thorn. The desire for this new event not only
to be painless but also relieve pain would be expected to promote changes
within the CNS that are associated with pain inhibition, as molecular
imaging studies by positron emission tomography have shown (Zubieta et al
2005).

The existence of pathways where ascending activity arising from Aδ
input excited by acupuncture can stimulate descending pain inhibitory
pathways that affect all segments have been discovered (Bowsher 1998) and
similar descending pain inhibitory pathways become active during the
administration of morphine and induction of placebo.
Unfortunately we still do not have access to the gold-standard double-
blind needling system - where neither the practitioner nor the patient
knows whether they have or have not been needled. Until we do, one’s
desire to perform an analysis of needling locations and techniques versus
placebo can not be satisfied. Park et al 1999 and Streitberger et al 1998
have both designed different types of sham needles: however neither sham
needle fools the experienced practitioner.

The fact we do not have access to a double-blind placebo controlled trial
of acupuncture does not necessarily condemn a technique that is bound to
excite complex peripheral and central changes (Sjölund 2005). If the
Chinese have discovered the ‘perfect placebo’, then I say well done.
However there are clues that suggest otherwise.
What is so interesting is that if acupuncture did not provide a specific
beneficial effect one would expect its benefit to be reduced as the
treatment is repeated – but generally the reverse is true. One would not
expect the treatment to be dose-related: too little needling, no effect;
just right, relief is produced; but too much noxious stimulation can be
followed by an increase in the patient’s complaint, a worsening that is
often speedily resolved by further stimulation. It is not easy to explain
why the onset of such changes occur usually after a delay of 12 to 48
hours. If these surprising effects could be documented and if we could
agree these are ‘specific effects’ as compared with the so called non-
specific effects usually associated with placebo then at least some
advance in the credibility of this subject could be maintained.

I wonder if in their more detailed unpublished records the authors
documented effects that in their opinion were unlikely to be those of
placebo and whether these differed in any significant way between the two
groups?

REFERENCES

Bowsher D (1998) Mechanism of acupuncture. In: Medical Acupuncture, a
Western Scientific Approach. (Eds) Filshie J, White A. Churchill
Livingstone. pp69-82

Godfrey CM, Morgan P (1978) A controlled trial of the theory of
acupuncture in musculoskeletal pain. Journal of Rheumatology 5 121-124

Park J, White A, Lee H, Ernst E (1999) Development of a new sham needle.
Acupuncture in Medicine 17 110-112

Streitberger K, Kleinhenz J (1998) Introducing a placebo needle into
acupuncture research. The Lancet 352 364-365

Sjölund B (2005) Acupuncture or acupuncture? Pain 114:311-312

Zubieta J-K, Bueller JA, Jackson LR, Scott DJ, Xu Y, Koeppe RA, Nichols
TE, Stohler CS (2005) Placebo effects mediated by endogenous opioid
activity on µ-opioid receptors. The Journal of Neuroscience 25(2005)7754-
7762

Competing interests:
None declared

In addition to my Rapid Response

Was it a good idea to publish the study-design before the end of the
study ?

I want to illustrate the extend of the unblinding information given
in the article

Melchart D, Linde K, Streng A, Reitmayr S, Hoppe A, Brinkhaus B,
Becker-Witt C, Wagenpfeil S, Pfaffenrath V, Hammes M, Willich SN,
Weidenhammer W: Acupuncture Randomized Trials (ART) in Patients with
Migraine or Tension-Type Headache - Design and Protocols. Forsch
Komplementärmed Klass Naturheilkd 2003;10:179- 184 (DOI:
10.1159/000073473)

This article was published online (download) in August 2003 and was
also made available through German libraries free of charge also to
patients. Recruitment for the study under discussion ended in January
2004, that means this study run fully unblinded!

Find Details under

http://www.angelfire.com/sc/naturheilverfahren/cgi-bin/points.JPG

Competing interests:
None declared

In a recent article Melchart et al demonstrated that acupuncture was
not better than ‘sham’ acupuncture at non acupuncture sites.1
Nevertheless, in both groups the amount of responders (>50% reduction
in days with headache) was very high (46 and 35% respectively). This was
in contrast with the controle group who were placed on a waiting list.
Other studies with acupuncture also failed to demonstrate a significant
better outcome compared to placebo groups (‘sham‘ acupuncture). Only laser
acupuncture seemed to be better than placebo (same procedure but output
power was set to zero).2 The therapy was given for 3 weeks, but the
beneficial effect lasted for three months.
Also other therapies failed to demonstrate any improvement compared with
placebo: e.g. injection of botulinum toxine.3-5
These studies look for a difference between ‘real’ and ‘sham’ procedures.
In both groups patients are hoping to get much benefit from the procedure
so a placebo effect can be expected. Just taking care of patients, listen
to their complaints can induce a reduction of tension. Moreover, feeling
at ease because other diagnoses (e.g. cerebral tumour) were ruled out,
could reduce headache complaints.
On the other hand, placing patients on a waiting list could induce an
adverse effect: this could negatively influence the complaints because
patients experience they are not been taken care of.
As far as I know just one other remedy proved to be better than placebo:
tiger balm reduced headache significantly better than topical balm with
menth essence and revealed to be as beneficial as acetaminophen.6

What should a clinician advise to his patients? A basic rule of
medicine is not to harm patients. So I would not advise the very expensive
botulinum toxine. Patients reported muscle weakness of the neck or eye
lids…
Also acupuncture causes some extra complaints. Tiger balm on the contrary
is easy to use, on daily basis, is very cheap and side effects are not
reported.

References:
1. Melchart D, Streng A, Hoppe A, Brinkhaus B, Witt C, Wagenpfeil S,
Pfaffenrath V, Hammes M, Hummelsberger J, Irnich D, Weidenhammer W,
Willich SN, Linde K. Acupuncture in patients with tension-type headache:
randomised controlled trial. BMJ. 2005; 331(7513):376-82.
2. Ebneshahidi NS, Heshmatipour M, Moghaddami A, Eghtesadi-Araghi P. The
effects of laser acupuncture on chronic tension headache--a randomised
controlled trial. Acupunct Med. 2005 ;23(1):13-8.
3. Padberg M, de Bruijn SF, de Haan RJ, Tavy DL.Treatment of chronic
tension-type headache with botulinum toxin: a double-blind, placebo-
controlled clinical trial.
Cephalalgia. 2004; 24(8):675-80.
4. Schulte-Mattler WJ, Krack P; BoNTTH Study Group.Pain. Treatment of
chronic tension-type headache with botulinum toxin A: a randomized, double
-blind, placebo-controlled multicenter study. 2004;109(1-2):110-4.
5. Rollnik JD, Tanneberger O, Schubert M, Schneider U, Dengler R.
Treatment of tension-type headache with botulinum toxin type A: a double-
blind, placebo-controlled study. Headache. 2000;40(4):300-5.
6. Schattner P, Randerson D. Tiger Balm as a treatment of tension
headache. A clinical trial in general practice. Aust Fam Physician. 1996;
25(2):216, 218, 220

Competing interests:
None declared