SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1 11

CORRESPONDENCE TO:
Prof. C Mucha
Medizinische Rehabilitation
und Prävention
Deutsche Sporthochschule
Carl-Diem-Weg 6
D-50933
Köln
E-mail: mucha@dshs-koeln.de

inflammatory response and fibrovascular
hyperplasia (Reveille 1997).

Besides processes of aging and other
degeneration processes, local hypoxia in
this tendon tissue plays a crucial role
in the pathogenesis (Almekinders and
Temple 1998, Fassbender 1984). This is
intensified further by the pain-induced
hypertension with contraction ischaemia
in the extensor muscle.

Accordingly, therapy would have to
be suitable to enhance local blood flow
and metabolism besides alleviating pain
and lowering muscle tone. Several forms
of physiotherapy are appropriate, but
hydrogalvanic partial-immersion bath
and rising-temperature arm bath are able
to combine the single effects specified
(Mucha 1987, Sadil and Sadil 1994).

The combination of these forms
of treatment in “therapy-resistant”
epicondylitis conditions was investigated
in an observational study on patients,
and its effectiveness was confirmed
(Mucha 1987).

In the current study, the efficacy of
this concept of treatment was therefore

ABSTRACT: The efficacy of two different regimens of physiotherapy for epicon-
dylitis was compared. A combination treatment with hydrogalvanic four-cell bath
and arm bath with rising temperature, which had showed good effects in treatment
of tennis elbow in an earlier observational study (Mucha 1987), was compared with
the analgesic interference current treatment often recommended in the literature
(Sadil and Sadil 1994, Noteboom et al 1994, Becker and Reuter 1982).

For this study, 60 patients with epicondylitis that was resistant to conservative
treatment were randomized into two groups for comparison. In group 1, interference currents were administered twice
a day for six weeks and group 2 received combination treatment with the hydrogalvanic four-cell bath and rising-
temperature arm bath once a day for six weeks.

Criteria for inclusion, control and appraisal were laid down prospectively. Several parameters were used, recorded
and statistically evaluated as outcome measures. These were active joint range of movement of the elbow, grip strength,
pain provocation with muscle contraction, palpation pain and pain with functional activities.

The results showed a significant superiority of combination treatment over therapy with interference current. It is
therefore recommended that hydrogalvanic four-cell bath and arm bath with rising temperature should be carried out
before considering surgical treatment for chronic epicondylitis.

KEY WORDS: EPICONDYLITIS, HYDROGALVANIC THERAPY, INTERFERENCE THERAPY, COMPARATIVE STUDY.

HYDRO-GALVANIC AND RISING-TEMPERATURE
BATH THERAPY FOR CHRONIC ELBOW

EPICONDYLITIS: A COMPARATIVE STUDY

R E S E A R C H
A R T I C L E

INTRODUCTION AND
STATEMENT OF THE PROBLEM
Epicondylitis is frequently encountered
by medical practitioners. Of these lateral
epicondylitis is most commonly seen
(Piligian et al 2000).

Etiologically, this condition in the
upper limb can be classified as a primary
degenerative disease (sometimes with
secondary inflammatory components) of
the tendon units (Piligian et al 2000,
Reveille 1997). However, the aetiology
and pathology is still largely unclear.
The main characteristics of epicondylitis
are likely to be its local origin and
development in the muscle-tendon units
without connection to intra-articular
structures. Its causation is evidently
multifactorial. According to present-day
knowledge, psychological overstimula-
tion and physical overuse of the a
diverse kind frequently lead to triggering
of this soft tissue disease (Becker and
Reuter 1982, Hotchkiss 2000, Noteboom
et al 1994, Piligian et al 2000). The
development of small tears in the
extensor aponeurosis is followed by an

compared with interference current
treatment, which is often recommended
for treatment of epicondylitis (Becker
and Reuter 1982, Noteboom et al 1994,
Sadil and Sadil 1994). These interference
currents comprise three superimposing
circuits of the same frequency (5000 Hz)
and variable phasing. The star-shaped
electrodes enabling just one bundled
pair of electrodes to be attached sepa-
rately to the patient's elbow.

The effect is determined by the choice
of beat frequency: analgesic sedative
(100-200 Hz), blood flow-enhancing
(100 Hz) or muscle-relaxant (25 Hz)
effects.

MUCHA C, MD1
1

Medizinische Rehabilitation
und Prävention,
Deutsche Sporthochschule Köln.

12 SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1

PATIENTS AND METHODS
Sixty patients were allocated randomly
to two groups: interference currents were
administered in group 1 and combina-
tion treatment with four-cell baths and
rising-temperature arm bath was applied
in group 2.
The following inclusion criteria were
applied:
- Patients of both sexes were included

in the study.
- No age limits.
- Their lateral epicondylitis was regarded

as resistant to treatment by at least
two mutually independent treatment
centres, so that the question of surgi-
cal treatment had been raised.

- The treatment period just completed
had to have exceeded 12 weeks.

- Epicondylitis had to have been mani-
fested for more than six months.

TREATMENT
The interference current in group 1 was
applied for 15 minutes in the elbow
region with the Stereodynator® as Sedat.
and Vegetat. Stimulation III Program
(200 Hz and 0.1 - 1 Hz) via plate
electrodes. In order to obtain quanti-
tative equivalence with the combination
therapy in group 2, the application was
repeated after a 60-minute break. In
group 2, the four-cell bath with cathode
switching on the arms and sensory
threshold current strength was applied
for 20 minutes and the rising-tempera-
ture arm bath was given to the affected
arm after a break of one hour. The
temperature was raised in steps of 1˚C
from an initial 39˚C to the maximum
tolerable temperature of 44˚ - 45˚C. The
temperature elevation took place within
20 minutes during a single application.
Both therapies were applied daily for
6 weeks.

On admission of the patients to our
hospital, the course of the disease history
was documented in addition to the
patients’ demographic data.
1. Specific clinical functional status was

established.
2. Joint status including measurements

of the circumference of the upper
limb, and active elbow movement
(see results).

3. Descriptive muscle status including
the shoulder girdle, tenderness to

palpation, localization of tenderness,
swelling, number and localization of
myogelosis.

4. Grip strength of the hand was tested
with the Vigorimeter®. Each side was
compared and classified as follows:
1 = equal strength on each side
(without consideration of lateral
dominance), 2 = up to 25% strength
reduction in the comparison between
sides, 3 = up to 50% strength reduc-
tion in the comparison between sides,
4 = up to 75% strength reduction in
the comparison between sides.

5. Medial or lateral elbow pain testing
included: provocation pain was tested
with resistance in dorsal and/or volar
flexion of the wrist (lateral or medial
or bilateral epicondylitis) and the
following classifications were taken
as the basis: 1 = no pain, 2 = slight
pain with good strength, 3 = still
tolerable pain with reduced strength,
4 = intensive pain without resistance.

6. To check the loading pain, the patient
had to lift 2 kg, 1 kg and 500 g weights
with the elbow extended: 1 = 2 kg
weight without pain, 2 = 2 kg weight
with pain, 3 = 1 kg weight with
pronounced pain, 4 = 500 g weight
with intensive pain.

7. Further pain qualities were scaled as
follows:

a: Resting pain generalized to elbow
area: 1 = no pain, 2 = slight pain under
maintained arm position, 3 = distinct
to excruciating pain in maintained
arm position, 4 = almost constant and
excruciating pain, even at night.

b: Movement pain of the forearm and
hand: 1 = no pain in routine everyday
movement without loading, 2 = slight
and transient pain in some load-relief
movements, 3 = severe pain in many
load-relief movements, 4 = distinct
pain in any movement and in some
cases with drastic increase of intensity.

c: Pressure pain over the epicondyle:
1 = no pain, 2 = moderate pain, 3 =
pronounced pain, 4 = severe pain.

All these measurements were repeated
after two weeks, four weeks, six weeks,
12 weeks and again after two years.

Treatment was always administered
for six weeks. With a positive course

and the good regeneration prognosis,
the same therapy was continued until
complete freedom from symptoms and
restoration of complete everyday routine
function. When no or only inadequate
effects of treatment were shown after six
weeks, administration of the alternative
treatment of the comparison group was
envisaged. When there was no success
under the new therapy, the treatment
was to be concluded if appropriate after
a further six weeks of treatment. These
cases had to be considered as “failed
treatments”.

Two years after the end of treatment,
the patients underwent follow-up
examinations. In addition they were
asked about the progress of their illness
during the preceding period and in
particular about any fresh treatment
required. They were also asked about
their functional capacity at work and in
everyday life.

The data obtained were broken down
into initial, progress and final criteria
and their absolute and relative fre-
quencies were compared by means of
descriptive statistics. For the six-week
treatment phase, the data in the course
were calculated and plotted and the group
comparison at the end of this phase of
treatment was analyzed with the �2 test
with Yates correction. The probabilities
of error were 5% (* p < 0.05) and
1% (** p < 0.01).

RESULTS
The mean age of the overall patient
population was 41.9 (19-68) years.
Women predominated (53%). Sex and
age distributions in the comparison
group did not show any statistically
significant differences. The same also
applied to the distribution of occu-
pational groups. Only 13.1% of patients
engaged in heavy physical work.

Exactly defined triggering causes
could only be rarely established, so
that the symptoms developed “spon-
taneously” in more than 61% of all
patients. One-sided strain in recreational
activities was the cause in 21.3%. The
remainder was caused by sport and
occupational strain.

The right elbow was involved in
66%, the left elbow in 34%, the lateral
epicondyles in 81%, the medial epi-

SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1 13

condyles in 12% and both epicondyles
in 7%. Only three patients in the overall
population were left-handed.

All patients had complained of recur-
rent symptoms for more than six months
(= criteria). The longest disease phase
was two and a half years (three patients).
Four patients had already undergone
surgical treatment. The conservative
therapy measures used are listed in
Table 1. The second most frequent
treatment was local cryotherapy, followed
by immobilization in plaster casts. The
majority of these patients reported an
alleviation of pain during immobiliza-
tion, but without a decisive effect on the
subsequent mobilization phase. This
benefit was not sustained. The numerous
forms of physiotherapy, which were
mainly employed in combination could
not always be adequately specified
(e.g. electrotherapies) and the parameters
of implementation (e.g. iontophoresis)
could not always be precisely established.
As reported, the forms of treatment were
mostly changed, when patients did not
report an improvement within seven to
14 days.

Functional pain, often associated with
transient resting pain was always clini-
cally prominent. Not uncommonly, there
were distinctly painful myogelosis at
the homolateral shoulder girdle and the
neck musculature owing to unilateral
guarding, so that patients often found

the shoulder and neck pain to be just as
intensive as the elbow pain. The pain
always radiated into the pertinent exten-
sor or flexor muscles of the forearm.
Neurological signs could only be found
in patients who had been previously
operated on, in some of whom there was
hypoesthesias in the vicinity of the scar.
Locally, a distinct swelling could be
observed in about 20% of the overall
population, and rarely a slight difference
in temperature comparing the two
sides. Unequivocal erythema was never
found. If present, differences in circum-
ference were not statistically significant.
However, there were distinct, even if
only terminal, movement restrictions in
mobility of the elbow. On average, there
was an active flexor deficit of 10˚ and an
extensor deficit of 15˚ ± 5˚. These deficits
were the same in lateral and medial
epicondylitis. The difference between
the sides was significant (p < 0.01) at
the beginning of treatment. This was
also the case for active pronation and
supination of the forearm, which were
terminally restricted by an average of 10˚.
These restrictions of movement which
were evidently due to pain regressed
spontaneously on reduction of the pain,
so that there were no longer any relevant
differences between the two sides after
six weeks of treatment, e.g. in group 2.
However, the differences from group 1
were then statistically significant.

There were significant (p < 0.01)
differences between the groups with
regard to all other test parameters after
six weeks of treatment. The distribution
results are summarized in Figures 1 - 6.
Development of the pain parameters
for the six-week course of treatment in
the two comparison groups is shown in
Figures 7 - 12. There were no group dif-
ferences in any of these test parameters
at the beginning of treatment. The
manifestation of differences between
the groups occurred with increasing
duration of therapy and became ever-
more distinct especially after the second
week of treatment. They were statistically
significant group differences in the grip
strength and all provocation pain as
early as the fourth week of treatment.
There was as yet no significant group
difference at this time of treatment only
in the case of rest pain. However, move-
ment pain attained threshold values,
although the trend of the overall course
is equivalent to that of the other test
parameters. There were statistically sig-
nificant (p < 0.01) group differences for
this parameter after six weeks of treat-
ment. It is also striking that in contrast to
provocation pain movement and rest pain
were very much less pronounced at the
beginning of therapy. The latter sub-
sided more slowly in the overall course.
Apart from the small number of affected
patients, these courses are likely to be
due in particular to severe epicondylitis.

Even though the effects were much
stronger in group 1, almost all para-
meters showed relatively slight effects
of treatment in the first two weeks of
treatment. They become increasingly
more pronounced only after this latency
period. However, it must be assumed
that the effects of treatment subside
again beyond the sixth week of treatment,
since the average duration of treatment
in group 2 was 9.6 ± 2.3 weeks. Owing
to the very much less pronounced effect
of treatment in group 1, these patients
afterwards received continued treatment
with the combination therapy of group 2
in accordance with our prospective
implementation criteria. The average
duration of treatment in these cases was
18.2 ± 8.3 weeks.

Four patients of group 1 and two
patients of group 2 did not attain any

Types of therapy Number of patients

pharmacotherapy oral 59

local 57

cortisone injections (>2x) 27 (21)

surgery 4

splint (10-36 days) 54

cryotherapy 57

massages 44

heat packs 13

ultrasound 43

interference currents 31

iontophoresis 19

other electrotherapies 22

exercise therapy 33

Table 1: Types of prior treatments in the total patient population (n=60)
(multi-referencing)

14 SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1

symptom-free routine everyday function,
so that treatment in these patients (10%
of the overall population) is to be
considered as unsuccessful according to
our efficiency criteria.

In the subsequent two-year observa-
tion period, 44 patients from the total
patient population (73.3%) remained
asymptomatic. Six patients (10%) had
one recurrence and four patients (6.7%)
had two or more recurrences that could
be treated conservatively. The remaining
six patients (10%) were our therapy
failures and had to be treated surgically.
Of these, two patients were not free

of symptoms even after the operation
and one patient was retired early for this
reason.

DISCUSSION
Clinically, pain and pain-dependent
functional impediments are doubtless
most prominent in epicondylitis. Since
this is primarily a degenerative local
soft-tissue disease, it does not have any
systemic clinical test parameter. For this
reason, exclusively subjective parame-
ters are available for objective evaluation
of therapeutic effects. However, taken
together in particular with regard to their

progress characteristics they are repre-
sent (Hotchkiss 2000). They also allow
quantitative appraisal of this pain syn-
drome (Benjamin et al 1999, Deveraux
and Hazleman 1984, Hotchkiss 2000,
Kurvers and Verhaar 1995, Pienimäki
et al 2002, Piligian et al 2000). All
specific diagnostic parameters were there-
fore used for multivariate comparison in
this study in order to detect possible
therapeutic differences in the compari-
son group. Active elbow movement,
differences of circumference of the
upper limb, grip strength, pain under
provocation tests, pain at rest, pain in

Score

100

Grip strength

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 1: Number of patients with reduced grip
strength at the end of six weeks of treatment in the
comparison groups.

Score

100

Provocation pain

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 2: Distribution differences in provocation pain
at the end of six weeks of treatment in the therapy
groups.

Score

100

Pain on weight bearing

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 3: Distribution differences in pain on weight
bearing at the end of six weeks of treatment in the
therapy groups.

Score

100

Pain on palpation

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 4: Distribution differences in pain on palpa-
tion at the end of six weeks of treatment in the
therapy groups.

SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1 15

movements and daily activity were
established in the course of treatment
and their differences in normalization
were compared between the treatment
groups. The salient demographic data of
the overall patient population in this
investigation hardly differs from those
described in the literature (Dijs et al 1990,
Kurvers and Verhaar 1995, Noteboom et
al 1994, Pienimäki et al 2002, Sevier
and Wilson 1999). There are sometimes
differences in the occupational groups
with those engaged in heavy work being
represented very much less than was
reported in earlier investigations, which
indicated that epicondylitis appears to
be more frequent in this occupational
category. On the other hand, more recent
publications (Piligian 2000) already
indicate that occupational features as
well as sport strain have recently become
less prominent than other characteristics,
in particular recreational activities. It
is probable that qualitative overstrain
plays a greater role under the techno-
logically more sophisticated working
conditions of the present day. However,
such differences are hardly likely to
affect the acute course of the disease and
its treatment.

Different distributions of the other
initial baseline criteria in the comparison
groups would be of crucial importance
for our treatment study, since they influ-
ence the course of the disease. However,
they are homogeneously distributed in

the two comparison groups. For this
reason, differences in the development
in the comparison groups during treat-
ment must be attributed to their effect.
Distinct divergences between the therapy
groups are already shown after two
weeks of treatment for all test para-
meters. A very much more pronounced
and more rapid regression of all clinical
symptoms took place in group 2. The
greater intensities manifested in the
provocation tests at the beginning of
treatment are entirely consistent with
clinical experience. The movement pain
in load removal and rest pain are corre-
spondingly less. These were only rarely
permanent and in most instances do not
affect all patients. However, their mani-
festation is likely to involve especially
pronounced disease courses, which are
the more difficult to influence. The
slower regression of these disease
characteristics (Fig. 11 and 12) in our
treatment course is doubtless attri-
butable to this.

All test parameters show a typical
time course. Initially, they show no or
only slight effects of therapy in the
first two weeks. The time course of the
clinical effect is thus the opposite of
what is usually encountered in pharma-
cotherapy. This observation is evidently
a special feature of many forms of
physiotherapy which as reaction therapy
are accompanied by delayed onset of
action (Lange 2003). Not only for this

reason, they therefore have to be applied
in series and planned as long-term
treatment.

This is important in the present con-
text because the kinds of physiotherapy
that were used in the prior treatments in
this patient population had mostly been
switched after one or two weeks owing
to reported lack of effect. Consequently,
there were deficits in their implementa-
tion. The effectiveness of physical forms
of treatment is crucially determined by
the quality of their implementation
(Lange 2003).

Furthermore, the question as to the
effectiveness of the treatments employed
remains unresolved. There are only a
few controlled studies on this topic
(Hotchkiss 2000). Strikingly, for most
data reported in the literature (Becker
and Reuter 1982, Dijs et al 1990,
Noteboom et al 1994) the indication
criteria for the individual forms of
physiotherapy were differentiated hardly
at all or only to a small extent. In most
cases, analgesia was the prime objective
of treatment. Its efficacy in treatment
of epicondylitis is mostly appraised
positively. This also applies to pharma-
cotherapy and suggestions for surgical
treatment (Hotchkiss 2000, Kurvers and
Verhaar 1995, Noteboom et al 1994,
Reveille 1997). Such results are likely to
be based on insufficient differentiation
of inclusion criteria in patients with
epicondylitis (Hotchkiss 2000). It is

Score

100

Movement pain

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 5: Distribution differences in movement pain
at the end of six weeks of treatment in the therapy
groups.

Score

100

Resting pain

Group 1 (n=30)

Group 2 (n=30)

N
u
m

b
e
r

o
f

P
a
ti
e
n
ts

i
n
%

1 2 3 4

Figure 6: Distribution differences in resting pain at
the end of six weeks of treatment in the therapy
groups.

16 SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1

Weeks
0 2 4 6

Grip strength

Group 1

Group 2

S
co

Figure 7: Development of grip strength for the six
week-course of treatment in the comparison groups.

Weeks
0 2 4 6

Provocation pain

Group 1

Group 2

S
co

Figure 8: Development of provocation pain for the
six week-course of treatment in the therapy groups.

Weeks
0 2 4 6

Pain on weight bearing

Group 1

Group 2

S
co

Figure 9: Development of pain on weight bearing for
the six week-course of treatment in the therapy groups.

Weeks
0 2 4 6

Pain on palpation

Group 1

Group 2

S
co

Figure 10: Development of pain on palpation for the
six week-course of treatment in the therapy groups.

Weeks
0 2 4 6

Movement pain

Group 1

Group 2

S
co

Figure 11: Development of movement pain for the
six week-course of treatment in the therapy groups.

Weeks
0 2 4 6

Resting pain

Group 1

Group 2

S
co

Figure 12: Development of resting pain for the six
week-course of treatment in the therapy groups.

SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1 17

known that they tend to show a varying
course and partial avoidance of strain
in consequence of pain already leads
to spontaneous improvement in a high
percentage of cases. It is therefore
problematical to determine the specific
effect of treatment.

In an attempt to rule out such sponta-
neous improvements in the course, only
patients whose disease course was rated
as resistant to therapy by at least two
independent treating physicians and who
showed a chronic course according to
the definition were included in this study.
The effect of a positive spontaneous
course can therefore be ruled out at least
for our appraisal phase of six weeks. In
addition, such effects are not likely to be
significant for the controlled group
comparison, because they would affect
the course of the disease to an equal
extent in both groups given the demon-
strated homogeneous distribution of the
random sample.

On the other hand, the regression of
all clinical functional symptoms was
very much more rapid in group 2 than
in comparison group 1. After six weeks
of treatment, all parameters show sig-
nificant (p < 0.001) group differences.
The unequivocal therapeutic superiority
of group 2 in which four-cell baths and
rising-temperature arm baths were used
as combination therapy was already
manifested in this early phase compared
to group 1 in which interference currents
were applied.

Moreover, the result is underscored
by the fact that despite the six-week
delay in treatment, the patients of group 1
could afterwards likewise be treated
successfully with the combination the-
rapy. However, an additional delay effect
had to be accepted, since the average
treatment time of these patients was 18.2
weeks, and was thus substantially
prolonged beyond the first six weeks of
treatment. Early use and consistent
implementation of combination therapy
can therefore additionally improve the
overall success of treatment.

Therapeutic effects of interference
current therapy in group 1 are doubtless
discernible, even though these occurred
with a delay in the course of treatment
and are demonstrably less than in
quantitative terms those in therapy

group 2. It is already known from
comparative studies (Mucha and Zysno
1984) on the analgesic effect in ischaemic
muscle pain that interference current
therapy is less effective than hydro-
galvanic baths. This is also confirmed in
the present condition-specific treatment.
Moreover, it must be assumed from
investigations published so far (Mucha
1993) that with interference current in
contrast to hydrogalvanic baths the
increase in blood flow in muscle opti-
mally and tendon tissue is inadequate.
The pronounced effect of hydrogalvanic
baths on the increase in blood flow is
also known and documented in deep
tissue layers (Maslov and Smirnov 1993,
Sadil and Sadil 1994). It can therefore
be assumed that this combination of
effects of the four-cell bath in decreasing
pain and increasing blood flow are
able to influence the pathogenetically
inadequate blood flow and supply in
the affected tendons and muscles in
epicondylitis. There may also be an
increase of blood flow in the bone, so
that the major sources of blood supply to
the tendon are likely to be activated.
Consequently, there is probably an
improvement in the overall supply of
nutrients.

Corresponding effects can be attained
with the rising-temperature arm bath.
A reaction interval of at least one hour
is indispensable since new reaction
sequences and thus additive effects can
only be expected again after the vascular
tonus has normalized. With our appli-
cation form, an additional local effect
of heat (maximum tolerable temperature
application) was to be exercised in addi-
tion to the vascular stimulus in order to
increase the local rate of metabolism in
the bradytrophic tendon tissue besides
the effects on the contraction ischaemia
in the musculature. Moreover, a heat-
induced inhibition of proliferation of
the connective tissue cells can also be
assumed (Schmidt and Hatzfeld 1979).
An increased proliferation of the connec-
tive tissue cells in the affected tendon
areas is also known from morphological
investigations (Fassbender 1984). It is
probably a result of the relative cell
hypoxia.

Since the effects both of the four-cell
bath and the rising-temperature arm

bath should be able to influence impor-
tant pathogenetic factors in lateral
epicondylitis, we implemented this con-
cept of combination therapy in a first
observational study (Mucha 1987). The
success attained therefore justifies using
this concept of therapy as a reference
therapy in the present comparison study.
The overall results confirm once more
the convincing effectiveness of this
concept of therapy with its mainly
pathogenetic approach.

The present results show that combi-
nation treatment with four-cell baths and
rising-temperature arm baths is clearly
superior to interference current therapy.
The mainly analgesic effect of the inter-
ference current (Mucha and Zysno 1984,
Sadil and Sadil 1994) is evidently not
sufficient for effective treatment of
epicondylitis. An attenuation of the
effect by habituation can be ruled out in
the application of interference current
(twice a day) since the progress results
of all test parameters show a tendency to
regression with increasing length of
application, and there are thus no indica-
tions for delayed effects which would
otherwise be expected.

With some reservations, the distinct
differences in effectiveness of the therapy
forms used also allow conclusions
with regard to possible effects of other
forms of physiotherapy. The forms of
pretreatment used during the unsuccess-
ful treatment in this test population also
did not always fulfill the theoretical con-
ditions for an absolute indication. This is
to be assumed above all for cryotherapy
and the immobilization in plaster casts.
Although cryotherapy has good analgesic
effects (Mucha and Zysno 1984), the
simultaneous suppression of blood flow
and metabolism (Liman et al 1982) are
likely to exacerbate the pathogenetic
conditions in epicondylitis. Secondary
inflammatory reactions in epicondylitis
mainly involve proliferative forms of
inflammation. However, an effect selec-
tion can only be achieved by a patho-
genetic and disease-specific adaptation
of the forms of therapy. Relief might be
attained by plaster cast immobilization,
but the question as to concomitant nega-
tive immobilization effects on the course
of the disease remains unanswered.
Vascular dysregulations for example and

18 SA JOURNAL OF PHYSIOTHERAPY 2004 VOL 60 NO 1

insufficient blood supply to bradytrophic
tissue associated with this already occur
in brief immobilization (Mucha et al
1987). Compared to radical immobili-
zation, however, functional relief is
indicated and evidently sufficient by
specific monitoring of movement activity
in the course of treatment. The results of
controlled investigations reported by
Little (1984) also indicate that better
results cannot be expected from plaster
cast immobilization.

The following conclusions can be
drawn from our results: with the appro-
priate qualitative and quantitative imple-
mentation criteria (e.g. reaction pause of
at least one hour between the individual
applications), combination treatment with
four-cell bath and rising-temperature
arm bath are efficient even in severe
chronic epicondylitis and the results that
can be attained are entirely comparable
to those of surgery (Kurvers and Verhaar
1995, Piligian 2000). This regime should
therefore be recommended before decid-
ing to carry out surgery. Ultrasound and
iontophoresis have been promoted and
are useful treatments. But there is no
evidence in the literature to support the
use of any of one modality over another
(Hotchkiss 2000). More investigation is
needed to determine the efficacy.

As a rule there are no major functional
impediments in these patients and they
are spontaneously reactivated with the
reduction of pain, exercise therapy is not
necessary during acute treatment. The
extent to which it should be used as a
secondary prevention measure in the
asymptomatic stages must be established
individually and depends on the motor
deficits actually present (Barry and
McGuire 1996). Since the objectives of
treatment differ greatly from patient to
patient, the establishment of the indi-
cation must of course be considered
critically since the commencement of
unusual activities is the most frequent
cause for the manifestation of epicon-
dylitis in this test population.

REFERENCES

Almekinders LC, Temple JD 1998 Etiology,
diagnosis, and treatment of tendonitis: an
analysis of the literature. Medicine and
Science in Sports Science 30:1183-1190

Barry NN, McGuire JL 1996 Overuse syn-
dromes in adult athletes. Rheumatic Disease
Clinics of North America 22:515-530

Becker R, Reuter H 1982 Epicondylitis
humeri: Diagnose und Therapie. Zeitschrift
für physikalische Medizin, Balneologie,
medizinische Klimatologie 11:496-501

Benjamin SJ, Williams DA, Kalbfleisch JH,
Gorman PW, Panus PC 1999 Normalized
Forces and Active Range of Motion in
Unilateral Radial Epicondylalgia (Tennis
Elbow). Journal of Orthopaedic and Sports
Physical Therapy 29:668-676

Devereaux M, Hazleman B 1984 Grip
strength as a measure of response to treatment
for lateral epicondylitis. British Journal of
Rheumatology 23:154-155

Dijs H, Mortier G, Driessens M, De Ridder A,
Willems J, De Vroey T 1990 A retrospective
study of the conservative treatment of tennis-
elbow. Medica Physica 13:73-77

Fassbender HG 1984 Strukturelle Ansatzpunkte
für eine Behandlung mit nichtsteroidalen
Antirheumatika. In: Miehlke K (ed) Fortschritte
in der perkutanen Therapie rheumatischer
Erkrankungen, pp15-31. pmi, Frankfurt/Main

Hotchkiss RN 2000 Epicondylitis - Lateral
And Medial. A Problem-Oriented Approach.
Hand Clinics 16:505-508

Kurvers H, Verhaar J 1995 The Results of
Operative Treatment of Medial Epicondylitis.
The Journal of Bone and Joint Surgery
77:1374-1379

Lange A 2003 Physikalische Medizin, pp l4-
14. Springer Verlag, Berlin Heidelberg

Liman W, Fricke R, Taghawinejad M,
Bernstein H 1982 Arterielle Durchblutung
unter Kryotherapie bei chronischer Poly-
arthritis. Zeitschrift für physikalische Medizin,
Balneologie, medizinische Klimatologie
11:196-201

Little TS 1984 Tennis elbow - to rest or not
to rest? Practitioner 228:457

Maslov AG, Smirnov VP 1993 Vliianie
transkardial'noi galvanizatsii na sostoianie
mikrotsirkuliatornogo rusla periinfarktnoi
zony pri eksperimental'nom infarkte miokarda.
(The effect of transcardiac galvanization on
the function of the microcirculatory bed of the
peri-infarct area in experimental myocardial
infarct). Bjulleten Ekspimentalnoj Biologii
Mediciny 116:85-86

Mucha C 1987 Sind Kombinationsbehand-
lungen sinnvoll? Therapiewoche 37:3460-3474

Mucha C 1993 Einflufl motorisch erregender
Stromformen auf die Unterarmdurchblutung.
Physical Therapy 14:9-14

Mucha C, Wieland B, Zysno EA 1987
Veränderungen der peripheren Durchblutung
unter Immobilisation und funktioneller
Frührehabilitation. Zeitschrift für physika-
lische Medizin, Balneologie, medizinische
Klimatologie 16:183-189

Mucha C, Zysno EA 1984 Vergleichende
Untersuchungen zur analgetischen Wirkung
medizinisch-physikalischer Therapieverfahren
auf experimentelle ischämische Muskel-
schmerzen - Teil 1 und 2. Zeitschrift für
physikalische Medizin, Balneologie, medi-
zinische Klimatologie 11, 544-557, 1982 und
13, 253-259, 1984

Noteboom T, Cruver R, Keller J, Kellogg B,
Nitz AJ 1994 Tennis Elbow: A Review.
Journal of Orthopaedic and Sports Physical
Therapy 19:357-366

Pienimäki TT, Siira PT, Vanharanta H 2002
Chronic Medial and Lateral Epicondylitis:
A Comparison of Pain, Disability, and
Function. Archives of Physical Medicine and
Rehabilitation 83:317-321

Piligian G, Herbert R, Hearns M, Dropkin J,
Landsbergis P, Cherniack M 2000 Evaluation
and Management of Chronic Work-Related
Musculoskeletal Disorders of the Distal Upper
Extremity. American Journal of Industrial
Medicine 37:75-93

Reveille JD 1997 Soft-Tissue Rheumatism:
Diagnosis and Treatment. The American
Journal of Medicine 102:23S-29Sm

Sadil V, Sadil S 1994 Elektrotherapie. Wiener
Medizinische Wochenschrift 144:509-520

Schmidt KL, Hatzfeld Th 1979 Zur Wirkung
kurzdauernder heifler B‰der auf expe-
rimentell ausgelöste Entz¸ndungen. Zeitschrift
für Physikalische Medizin 8:145-148

Sevier TL, Wilson JK 1999 Treating Lateral
Epicondylitis. Sports Medicine 28:375-380