Content uploaded by Hanns Ackermann
Author content
All content in this area was uploaded by Hanns Ackermann on Oct 04, 2014
Content may be subject to copyright.
THE JOURNAL OF ALTERNATIVE AND COMPLEMENTARY MEDICINE
Volume 14, Number 1, 2008, pp. 17–25
© Mary Ann Liebert, Inc.
DOI: 10.1089/acm.2007.0560
ORIGINAL PAPERS
Efficacy of Arnica montana D4 for Healing of Wounds After
Hallux Valgus Surgery Compared to Diclofenac
JENS-HAGEN KAROW, M.D.,
1
HANS-PETER ABT, M.D.,
1
MARKUS FRÖHLING, M.D.,
1
and HANNS ACKERMANN, Ph.D.
2
ABSTRACT
Objective: This study was undertaken to answer the question: “Is Arnica D4 as efficacious as diclofenac in
relation to symptoms and wound healing after foot surgery?”
Methods: In this randomized double-blinded, parallel-group study (GCP-standard), the efficacy of Arnica D4
10 pillules (taken orally, 3 times per day) and diclofenac sodium, 50 mg (taken orally, 3 times per day) were in-
vestigated for equivalence in 88 patients 4 days after hallux valgus surgery. Outcome parameters were (1) postop-
erative irritation, (2) patient mobility, (3) rated pain, and (4) use of analgesics. The hierarchic equivalence test based
on one-sided Wilcoxon-Mann-Whitney-U confidence intervals (CIs) was used. Equivalence was perceived, when
the lower margin of the 95% CI was 0.36 corresponding to a range of equivalence of 1/2 standard deviation.
Results: Arnica D4 and diclofenac were equivalent for wound irritation (lower margin of the 95% CI on day
4: 0.4729 for rubor; 0.3674 for swelling; 0.4106 for calor) and patient mobility (0.4726). A descriptive analy-
sis showed the superiority of Arnica D4 with respect to patient mobility (p 0.045).With respect to pain, Ar-
nica D4 was inferior to diclofenac (lower margin of the 95% CI 0.026). No significant differences were found
regarding the use of additional analgesics during the 4 postoperative days (Dipidolor,
®
Janssen-Cilag, Neuss,
Germany; p 0.54; Tramal,
®
Grünenthal, Aachen, Germany; p 0.1; and Novalgin,
®
AVENTIS-Pharma, Bad
Soden, Germany; p 0.1). Arnica D4 was significantly better tolerated than diclofenac (p 0.049). Nine (9)
patients (20.45%) of the diclofenac group and 2 (4.5%) of the Arnica D4 group reported intolerance. There was
no disturbance in wound healing in any of the patients. Arnica D4 is 60% cheaper than diclofenac.
Conclusions: After foot operations, Arnica D4 can be used instead of diclofenac to reduce wound irritation.
17
INTRODUCTION
F
or patients with Hallux valgus and Hallux rigidus, or-
thopedic surgeons, after exhausting conservative mea-
sures, use osteotomies of the first ray to correct the defor-
mity. The post-traumatic edemata caused by surgery can,
inter alia, due to abnormal blood circulation caused by com-
pression and dilating, lead to problems in the healing of the
wounds and pain. For this reason, nonsteroidal an-
tirheumatica (NSAR) of the diclofenac type are prescribed
to lessen swelling and as analgesics. The efficacy of di-
clofenac to lessen pain and swelling following surgery is
verified in placebo-controlled clinical studies.
1–3
The positive desirable effects of diclofenac are in con-
trast to possible side-effects such as gastrointestinal bleed-
ing and allergies.
4–6
The development of a new generation
of NSAR with selective cyclo-oxygenase-2 inhibitors has,
to date, also not been able to resolve the difficulty of unde-
sired effects.
7,8
In the field of homeopathy, Arnica montana preparations
have been used, with very good levels of tolerance, for acute
traumata such as strains, contusions, and bruises.
9,10
The ex-
1
Orthopaedic University Hospital Friedrichsheim, Frankfurt Medical School, Johann Wolfgang Goethe-Universität, Frankfurt, Germany.
2
Centre for Biomathematics, Frankfurt Medical School, Johann Wolfgang Goethe-Universität, Frankfurt, Germany.
ternal use of Arnica essence (not diluted) may cause skin ir-
ritation, and internal use can cause lethal allergic reactions.
These effects are not known for homeopathic pillules of Ar-
nica D4. Some studies to date relating to the postoperative use
of Arnica have produced positive results but, due to method-
ical weaknesses, are not valid.
11
Newer studies from the sur-
gical field showed that Arnica preparations had no advantage
over placebo.
12,13
Due to the inconsistent nature of the data
and a presumed potential, further research is necessary. We
wanted to clarify whether Arnica D4 has an influence on symp-
toms and healing of wounds after standardized foot surgery.
PATIENTS AND METHODS
Eighty-eight (88) patients took part in this randomized,
double-blinded, parallel-group study, which was carried
out, in accordance with good clinical practice (GCP) stan-
dards, in an orthopedic hospital from November 2000 un-
til May 2002. Men and women between the ages of 20 and
65 years with the surgical indication “Hallux valgus” or
“Hallux rigidus” on the left and/or right metatarsal I were
included.
Criteria for exclusion were clotting disorders (also due to
low-dose acetylsalicylic acid), rheumatic diseases, serious
metabolic disorders (e.g., diabetes mellitus), arterial occlu-
sive disease, varicosis, lymphoedema, peptic ulcer, and el-
evated transaminase levels.
The division of the patients in to the Arnica or the di-
clofenac group was carried out in 11 randomized blocks,
each with 8 patients. The surgeries were divided into cate-
gories 1–3 according to the difficulty of the operation as
well as soft-tissue and bone trauma and assigned to blocks
as follows:
14–16
KAROW ET AL.
18
FIG. 1. First level of the hierarchical analysis, postoperative irritation with the single criterion rubor, swelling, or calor (N 88). Me-
dian, percentage, maxima, and minima on the fourth day after hallux valgus surgery. *Significant p-value.
T
ABLE
1. F
IRST
L
EVEL OF THE
H
IERARCHICAL
A
NALYSIS
, P
OSTOPERATIVE
I
RRITATION WITH THE
S
INGLE
C
RITERION
R
UBOR
, S
WELLING
,
OR
C
ALOR
: M
EDIAN
, P
ERCENTILE
, M
AXIMA
,
AND
M
INIMA ON THE
F
OURTH
D
AY AFTER
H
ALLUX
V
ALGUS
S
URGERY
(N 88)
Confidence
Mann- interval at Mann- Lower limit of the
Whitney- Whitney-value one-sided 95%
Median Minimum Maximum p-value U-value for p 0.90 confidence range
Rubor
Diclofenac 10 0 75 0.049 0.62 0.5271
–0.7167 0.4729
Arnica D4 25 0 75
Swelling
Diclofenac 27,5 0 80 0.58 0.53 0.4361–0.6326
0.3674
Arnica D4 32,5 0 100
Calor
Diclofenac 20 0 75 0.89 0.49 0.3925–0.5894
0.4106
Arnica D4 20 0 50
Bolding signifies significant p-value. Underlining signifies the side of the limit (used for one-sided 95% confidence range).
Category 1—Operations with subcapital corrective os-
teotomy on the first ray and if necessary on the proximal
phalange. (Chevron-osteotomy and modifications ac-
cording to Akin) 5 blocks
Category 2—Diaphysial osteotomies on the first ray (Scarf-
osteotomy) or athrodesis of the first tarsometatarsal joint
(modified Lapidus arthrodesis) 4 blocks
Category 3—Complex corrections of the forefoot with op-
erations on the first ray according to Scarf Ludloff as well
as corrective osteotomies on the small toes (OP in
accordance with Helal, Weil, Hohmann cited in Refs.
14–16) 2 blocks.
To exclude iatrogenic effects on the degree of trauma, all
operations were conducted by the same surgeon.
For 4 days postoperatively, the patients received either di-
clofenac sodium, 50 mg, 3 times daily or 10 pillules of Ar-
nica D4 (WALA) 3 times per day. The pillules were produced
by WALA, Bad Boll, Germany, in accordance with the HAB
1, rule 39a (the rule is in the Homeopathic Pharmaceutical
Manual [in German], Stuttgart: German Pharmacist Associ-
ation, 2001). The drugs were prepared in a gelatin capsule,
which, as the content was identical in appearance, were vi-
sually indistinguishable from the Arnica pillules. In addition,
patients from both groups received ranitidine, 150 mg, twice
daily, which was needed in combination with diclofenac as
gastroprotective therapy.
Postoperative irritation was recorded with the criteria “ru-
bor,” “swelling,” and “calor.” On the fourth day after
surgery, the criteria were determined for each patient by the
surgeon. After inspection and palpation, the results were en-
tered on a visual analogue scale (VAS) from 0 to 100, for
which 0 none, 25 weak, 50 medium strong, 75
strong, and 100 very strong. The patients’ convalescence
was recorded in terms of patient mobility. On the fourth
postoperative day, in the evening, each patient indicated on
the questionnaire how long she or he had been out of bed.
Pain on the fourth postoperative day was calculated as an
area under the curve (AUC). In the morning and in the evening,
patients indicated pain by using the same VAS as noted above.
The analgesics Dipidolor,
®
(Janssen-Cilag, Neuss, Ger-
many), Tramal,
®
(Grünenthal, Aachen, Germany), and No-
valgin
®
(AVENTIS-Pharma, Bad Soden, Germany) were
administered, according to hospital standards, in individual
or combined doses as requested by the patients and docu-
mented to record the additional need for analgesics.
EFFICACY OF ARNICA MONTANA D4 FOR HEALING OF WOUNDS
19
T
ABLE
2. S
ECOND
L
EVEL OF THE
H
IERARCHICAL
A
NALYSIS
, C
ONVALECENCE
, R
ECORDED FROM
P
ATIENT
M
OBILITY
(M
INUTES
O
UT OF
B
ED
)
FROM THE
F
IRST TO THE
F
OURTH
D
AY AFTER
H
ALLUX
V
ALGUS
S
URGERY
(N 88)
Mann- Confidence Lower limit of
Whitney interval at Mann-- the one-sided
p-value U-value- Whiney-value for 95% confidence
Median Minimum Maximum (4th day) (4th day) (4th day) p 0.90 range
1st day
Diclofenac 2.5 0 180
Arnica D4 4.5 0 90
2nd day
Diclofenac 25 0 240
Arnica D4 30 0 360
3rd day
Diclofenac 37.5 0 330
Arnica D4 50 0 450
4th day
Diclofenac 60 0 300 0.045 0.62 0.5274
–0.7205 0.4726
Arnica D4 120 0 360
Bolding signifies significant p-value of the fourth day taken for the endpoint to be evaluated. Underlining signifies the side of the
limit (used for the one-sided 95% confidence range).
T
ABLE
3. T
HIRD
L
EVEL OF THE
H
IERARCHICAL
A
NALYSIS
, P
AIN
Lower limit of the
Mann- confidence interval Lower limit of the
Whitney- at Mann-Whitney- one-sided 95%
Median Minimum Maximum p-value U-value value (p 0.90) confidence range
AUC
Diclofenac 2325 0 7800 0.027 0.6366 0.7337 0.2662
Arnica D4 3450 480 6240
Note: The average total pain recorded by the addition of the 8 visual analogue scale pain-related details over 4 days, calculated as
AUC [area under the curve] is shown.
Bolding signifies significant p-value.
The study was conceived as proof of therapeutic equiva-
lence (i.e., noninferiority). A hierarchical test for therapeu-
tic equivalence was set up and outcome parameters were
based on 4 levels:
17
1. Equivalence test for postoperative irritation (rubor,
swelling, and calor)
2. Equivalence test convalescence
3. Equivalence test rated pain
4. Equivalence test for use of analgesics.
The collective range from n1 n2 44 was recorded on
the bases of the Wilcoxon-Mann-Whitney U test where
0.05 and 0.1 and the Mann-Whitney-U estimator
0.70. The global significance level was determined with
0.05. The therapeutic equivalence test was carried out on the
basis of the Wilcoxon-Mann-Whitney-U test, for which the
Mann-Whitney-U value was calculated as the lower 95%
confidence limit and compared with the equivalence limit
of 0.36 (corresponding to 1/2 standard deviation). Equiva-
lence was considered proven when the lower margin of the
confidence interval (CI) was 0.36.
18
The four levels were
ordered in terms of their significance for the study and were
tested one after another in a priori ordered sequence. The
process was terminated when zero hypothesis Ho ( non-
equivalence) in relation to a group could not be rejected and
therefore therapeutic inferiority was to be assumed. After
the confirmatory test procedure had been terminated, further
test results had to be handled on a descriptive basis. For this,
the Moses CIs, which relate to the difference of the medi-
ans, were used to interpret the results.
Working with the data was done on the basis of biomet-
rical analysis of samples (BiAS; EPSILON-Verlag,
Hochheim-Darmstadt, Germany, 2002).
19
RESULTS
All 88 patients included in the study were evaluated per
protocol. No missing values were replaced.
20
First level, postoperative irritation
The lower margins of the 95% CI for the individual cri-
teria swelling ( 0.3674), calor ( 0.4106), and rubor (
0.4729) were over the critical threshold for therapeutic
equivalence of 0.36. Arnica D4 and diclofenac were thus
therapeutically equivalent (Fig. 1; Table 1).
Second level, convalescence
To determine convalescence, patient mobility was eval-
uated, as this is dependent on the stage of wound healing.
On the fourth day, the lower margin of the 95% CI for
convalescence was 0.4726 and thus over the confidence limit
of 0.36. Arnica is not therapeutically inferior to diclofenac.
The descriptive test for group differences indicated the
superiority of Arnica D4: On the 4th postoperative day pa-
tients in the Arnic a group were out of bed, on average, twice
as long as the patients in the diclofenac group (p 0.045;
Table 2). The comparison on the first 3 postoperative days
showed no significant differences in the groups (p for day
1 0.78, p for day 2 0.67, p for day 3 0.41).
Third level, the patients’ level of pain
The lower limit of the 95% confidence range for pain was
0.2662 on the VAS, below the CI for therapeutic equiva-
lence of 0.36. Arnica D4 was therefore therapeutically in-
ferior to diclofenac in relation to pain (Table 3).
KAROW ET AL.
20
T
ABLE
4. C
OURSE OF THE
V
ISUAL
A
NALOGUE
S
CALE
D
ETAILS
R
ELATING TO
P
AIN
O
VER
D
AYS
1–4
AFTER
H
ALLUX
V
ALGUS
S
URGERY
(N 88)
Lower limit of Moses
confidence interval
Median Minimum Maximum (p 0.95) p-value
1st day Morning Diclofenac 75 0 100 5 0.81
Arnica 67.5 0 100
Evening Diclofenac 50 0 100 25 0.012
Arnica 65 20 100
2nd day Morning Diclofenac 27.5 0 100 25 0.027
Arnica 50 0 100
Evening Diclofenac 25 0 100 25 0.031
Arnica 50 0 100
3rd day Morning Diclofenac 25 0 100 15 0.099
Arnica 30 0 60
Evening Diclofenac 25 0 85 15 0.089
Arnica 25 0 75
4th day Morning Diclofenac 10 0 80 10 0.20
Arnica 25 0 50
Evening Diclofenac 15 0 80 10 0.18
Arnica 25 0 75
Bolding signifies significant p-values.
In accordance with the specifications for a hierarchical
group closed testing procedure for therapeutic equivalence,
the confirmatory part of the testing had to be terminated at
this point. Further statistical comparisons relating to the cri-
teria for pain and use of medication were dealt with de-
scriptively.
With respect to intensity of pain there were significant
differences in the groups on the evening of the first post-
operative day (p 0.012) as well as in the morning and
evening of the second postoperative day (both p 0.03;
Table 4, Fig. 2). The difference of the medians at these three
points in time was 25 (i.e., at any one time the median for
the patients in the Arnica group was 25 VAS points higher
pain score). Furthermore, in relation to the total of all VAS
pain, the patients in the Arnica group had more pain than
the patients in the Diclofenac group (p 0.027; Table 3)
Use of analgesics
Dipidolor, Novalgin, and Tramal were used as required
for the treatment of pain, individually or combined, until
each patient was satisfied with the pain alleviation.
Each dose of analgesic was documented exactly with
amount and time of administration. The total use of the med-
EFFICACY OF ARNICA MONTANA D4 FOR HEALING OF WOUNDS
21
FIG. 2. Course of the visual analogue scale details relating to pain over days 1–4 after hallux valgus surgery (N 88), median, per-
centile, maxima, and, minima.
ications Dipidolor (p 0.54), Tramal (p 0.1), and No-
valgin (p 0.1) over 4 days revealed no significant differ-
ences between the groups. When each day was analyzed,
Tramal (p 0.05) was used in the Arnica group signifi-
cantly more often on the first day. The Moses CI for the use
of Tramal was 20 on the first postoperative day. With a
certainty of 95%, the patients in the Arnica group needed
therefore an average 20 drops more Tramal than the patients
in the diclofenac group. The CIs on the other days were 0
(i.e., there was no additional consumption of medication).
The Moses interval for total use was 0 for Dipidolor and
20 respectively for Tramal and Novalgin. The patients in
the Arnica group used 20 drops more Tramal and Novalgin
over all 4 postoperative days combined (Table 5; Fig. 3).
Intolerance to medications and termination
of studies
Nine (9) of the 44 patients in the diclofenac group
(20.45%) reported intolerance, in the Arnica group 2 of the
44 patients did (4.5%) so. In individual cases, intolerance
led to termination of the study. The difference at p 0.049
was significant (Table 6; Fig. 4).
DISCUSSION
Diclofenac is standard for the therapy of postoperative
pain. A Cochrane review revealed a clear superiority to
placebo (relative benefit 3.2 CI 2.6–4.0).
3
With respect to
postoperative swelling, which is especially relevant to chi-
ropody because of the possible complications of wound
dehiscence and aberrations in the healing of wounds, di-
clofenac, 3 times 50 mg per day, orally, is generally recom-
mended,
15
and was significantly superior to placebo in
placebo-controlled studies for both postoperative use
21
and
for post-traumatic swelling, which is subject to a similar
mechanism as postoperative swelling.
2,22
A control to placebo
was therefore refused by the Ethics Commission for our study.
KAROW ET AL.
22
T
ABLE
5. R
ESULTS OF
U
SE OF
N
OVALGIN
, T
RAMAL
,
®
AND
D
IPIDOLOR
®
IN THE
D
ICLOFENAC
AND
A
RNICA
G
ROUPS
O
VER
D
AYS
1–4
AFTER
H
ALLUX
V
ALGUS
S
URGERY
(N 88)
Lower limit of the Moses
Median Minimum Maximum confidence interval (p 0.95) p-value
1st day Dipidolor
®
Diclofenac 0 0 2 0.0 0.58
Arnica 00 4
Tramal
®
Diclofenac 0 0 125 20 0.049
Arnica 0 0 160
Novalgin
®
Diclofenac 0 0 80 0.0 0.50
Arnica 0 0 100
2nd day Dipidolor Diclofenac 0 0 1.5 0.0 0.86
Arnica 0 0 1.00
Tramal Diclofenac 0 0 100 0.0 0.11
Arnica 0 0 160
Novalgin Diclofenac 0 0 60 0.0 0.12
Arnica 0060
3rd day Dipidolor Diclofenac 0 0 0 0.0 0.85
Arnica 0 0 0.5
Tramal Diclofenac 0 0 120 0.0 0.54
Arnica 0080
Novalgin Diclofenac 0 0 40 0.0 0.11
Arnica 0060
4th day Dipidolor Diclofenac 0 0 0 0.0 1.0
Arnica 00 0
Tramal Diclofenac 0 0 120 0.0 0.58
Arnica 0 0 100
Novalgin Diclofenac 0 0 20 0.0 0.45
Arnica 0040
Total Dipidolor Diclofenac 0 0 3.5 0.0 0.54
Arnica 00 5
Tramal Diclofenac 20 0 300 20 0.11
Arnica 20 0 320
Novalgin Diclofenac 0 0 180 20 0.097
Arnica 22.5 0 220
Dipidolor,
®
Janssen-Cilag, Neuss, Germany.
Tramal,
®
Grünenthal, Aachen, Germany.
Novalgin,
®
AVENTIS-Pharma, Bad Soden, Germany.
Bolding signifies significant p-value.
Due to its particular relevance to chiropody, postoperative ir-
ritation was chosen as the primary outcome parameter for the
hierarchical analysis. The equivalence limit of 1/2 standard
deviation lay clearly below the distance between diclofenac
and placebo established for postoperative pain.
3
Arnica D4, 3 10 pillules per day, was equivalent to di-
clofenac, 3 50 mg per day, for relieving postoperative
swelling and calor. The rubor of the feet were more strongly
pronounced when Arnica D4 was used than when diclofenac
was used (p 0.049), which was assessed as a positive ef-
EFFICACY OF ARNICA MONTANA D4 FOR HEALING OF WOUNDS
23
FIG. 3. Graph of the use of Novalgin,
®
Tramal,
®
and Dipidolor
®
in the diclofenac and Arnica groups over days 1–4 after hallux val-
gus surgery (N 88). Novalgin, AVENTIS-Pharma, Bad Soden, Germany. Tramal, Grünenthal, Aachen, Germany. Dipidolor, Janssen-
Cilag, Neuss, Germany.
fect. As is known, diclofenac interferes in an inhibitory man-
ner in prostaglandin synthesis and, in this way, acts as an
antiexudative and lessens the vasodilation mediated by
prostaglandin.
23
It appears that Arnica D4 does not suppress
the hemoperfusion as dicofenac does, which contributes to
a more rapid healing of the wound.
On the fourth postoperative day, mobility, as an addi-
tional clinically important parameter, was even better with
Arnica D4 than with diclofenac, possibly due to this factor.
According to recent research, helenalin in A. montana in-
hibits NF-kappaB in the micromolar region, a transcription
factor that plays a substantial role in the inflammatory
process.
24
This finding is apt, given the fact that Arnica D4
has a positive effect, particularly, on inflammatory irritation,
although the effective concentration of helenalin in Arnica
D4 is probably lower than the concentration in micromolar
units generally recognized for a drug effect. Whether and
which additional factors are responsible for the antiswelling
effects cannot be answered in this study.
According to this study Arnica D4 can be used instead
of diclofenac for the healing of wounds after foot opera-
tions. The fact that the tolerance for Arnica D4 was clearly
better than that for diclofenac speaks even more so in Ar-
nica D4’s favor. (Table 6, Fig. 4). There is no gastropro-
tective therapy necessary. The costs of using Arnica D4 as
a medication are 60% less than than the costs of using di-
clofenac. However, patients in the Arnica group were clearly
in more pain on the first 2 postoperative days than the pa-
tients in the diclofenac group (Table 4; Fig. 2). Interestingly,
despite the clearly higher level of pain, the use of analgesics
in the Arnica group with a maximum of 20 drops of Tramal
and Novalgin over 4 days, was only slightly higher. There-
fore, with the exception of day 1, the pain appeared to have
been relatively tolerable, as otherwise the use of analgesics
should have been considerably higher.
ACKNOWLEDGMENTS
The authors wish to thank Roman Huber, M.D. (Univer-
sitätsklinik Freiburg, Freiburg, Germany) for his critical
reading of the text as well as his valuable suggestions for
making alterations.
REFERENCES
1. Albrecht J. Study comparing the decongestant effect of di-
clofenac-natrium and escin drug in traumatology [in German].
Z Allg Med 1981;29:2012–2016.
2. Diebschlag W. Volumetric monitoring of swollen ankle joints
treated with diclofenac [in German]. Fortschritte der Medizin
1986;21:51–54.
3. Barden J, Edwards J, Moore RA, McQuay HJ. Single dose oral
diclofenac for postoperative pain. Cochrane Database of Syst
Rev 2006;1:CD00509.
4. Rösch W: Undesired effects of NSAR in therapy [in German].
Med Trib 1999;34:50.
5. Incorvaia C. Allergy to Medication: Ospedale Maggiore Mi-
lan: Ärztliche Praxis 1999;11:12.
6. Bolten W. Pain reduction of arthritis through physiotherapy
[in German]. Ärzte Zeitung 1999;9:30.
KAROW ET AL.
24
T
ABLE
6. I
NTOLERANCE TO
M
EDICATION
AND
T
ERMINATION OF
T
REATMENT
Number of patients Termination
Diclofenac group
11 nervous disorders
14 abdominal complaints 2nd day
27 abdominal complaints 2nd day
36 abdominal complaints 2nd day
44 abdominal complaints 2nd day
58 perioral Efflorescence
65 abdominal complaints 4th day
71 abdominal complaints
82 Exanthem (rash) 2nd day
Arnica group
45 abdominal complaints 2nd day
56 racing heart 1st day
FIG. 4. Complications caused by the test medication.
7. Solomon DH, Schneeweiss S, Glynn RJ, Kiyota Y. Relation-
ship between selective cyclooxygenase-2 inhibitors and acute
myocardial infarction in older adults. Circulation 2004;109;
2068–2073.
8. Edwards JE. Efficacy and safety of valdecoxib for treatment
of osteoarthrosis and rheumatoid arthritis. Pain 2004;20:1–11.
9. Ausmeier E. Medicine plant therapy in complementary prac-
tice: Arnica montana: A great remedy for trauma in phy-
totherapy. [in German]. Das Seminar 1988;1;22-30.
10. Madaus G. Textbook of Medical Plant Therapy [in German].
Ravensburg, Germany: Mediamed-Verlag, 1987.
11. Ernst E, Pittler M. Efficacy of homeopathic arnica: A sys-
tematic review of placebo-controlled clinical trials. Arch Surg
1998;133:1187–1190.
12. Ramelet A-A, Buchheim G, Lorenz P, Imfeld M. Homeopathic
arnica in postoperative haematomas: A double-blind study.
Dermatology 2000;201:347–348.
13. Stevinson G, Devaraj VS. Homeopathic arnica for prevention
of pain and bruising: Randomized placebo-controlled trial in
hand surgery. J R Soc Med 2003;96:60–65
14. Coughlin MJ. Hallux valgus. J Bone Joint Surg 1996;78:932–
966.
15. Engelhardt P: Orthopedic Foot Surgery [in German]. Darm-
stadt: Steinkopff-Verlag, 2001.
16. Valtin B, Jakob R, Fulford P, Horan F. Changing concepts in
the surgery of hallux valgus: London European instructional
course lecture. Soc Bone Joint Surg 1999;4:119–127.
17. Maurer W, Hothorn LA, Lehmacher W. Multiple comparisons
in drug clinical trials and preclinical assays: A priori ordered
hypotheses. In: Vollmar J, ed. Biometrical Analysis in the
Chemical and Pharmaceutical Industry [in German]. Stuttgart-
Jena-New York: Gustav Fischer-Verlag, 1995:3–21.
18. Colditz GA, Miller JN. Measuring gain in the evaluation of
medical technology: The probability of a better outcome. Int
J Technol Health Care 1988;4:637–642
19. Ackermann H. BIAS: Biometrical Analysis of Samples [in
German]. Hochheim-Darmstadt, Germany: Epsilon Verlag,
2002.
20. The European Agency for the Evaluation of Medicinal Prod-
ucts, Human Medicines Evaluation Unit, ed. ICH Topic E6:
Guideline for Statistical Principles in Clinical Trial. London:
Document III/3630/92-EN 1998.
21. Henrikson PA, Thilander H, Wahlander LA. Voltaren as an
analgesic after surgical removal of a lower wisdom tooth. Int
J Oral Surg 1985;14:333–338.
22. Giani E, Rocchi L, Tavoni A, et al. Telethermographic eval-
uation of NSAIDs in the treatment of sport injuries. Med Sci
Sports Exerc 1989;21:1–6.
23. Wellhöner HH. General and Systematic Pharmacology and
Toxicology [in German]. Berlin-Heidelberg: Springer 1988.
24. Lyss G, Schmidt TJ, Merfort I, Pahl HL. Helenalin, an anti-
inflammatory sesquiterpene lactone from Arnica, selectively
inhibits transcription factor NF-kappa. Biol Chem 1997;378:
951–961.
Address reprint requests to:
Jens-Hagen Karow, M.D.
Oberbornstraße 25
61381 Friedrichsdorf/Ts
Germany
E-mail: karow@bias-online.de
EFFICACY OF ARNICA MONTANA D4 FOR HEALING OF WOUNDS
25
