522

RESEARCH REPORT

Progressive Strengthening and

Stretching Exercises and Ultrasound for
Chronic Lateral Epicondylitis
Tuomo T Pienimaki
Tuula K Tarvainen
Pertti T Siira
Heikki Vanharanta
Key Words
Epicondylitis, physical exercise, ultrasound, pain, muscle
strength.

Summary
Thirty-nine patients suffering from chronic lateral epicondylitis
were randomised into two treatment groups. The first group
(n = 20) was treated with progressive slow, repetitive wrist and
forearm stretching, muscle conditioning and occupational exercises, which were intensified in four steps. The second group
(n = 19) was treated with pulsed ultrasound. The effect of six to
eight weeks treatment was measured by a pain questionnaire
(visual analogue scale), isokinetic muscle performance testing of
wrist and forearm, and isometric grip strength measurements.
In the follow-up visit after eight weeks treatment, pain at rest and
under strain had decreased and subjective ability to work
increased in the exercise group significantly more than in the
ultrasound group (p = 0.004,0.04 and 0.004). Correspondingly,
sleep disturbance was alleviated significantly more in the exercise group (p = O.Ol).The isokinetic torque of wrist flexion
increased by 45% in the exercise group and declined by 4% in the
ultrasound group (p = 0.0002). Maximum isometric grip strength
increased 12 % in the exercise group and remained unchanged
in the ultrasound group (p = 0.05).
During treatment six of eight patients in the exercise group and
three of nine patients in the ultrasound group became able to
work. All clinical manual provocation tests for tennis elbow
improved within the exercise group. The results indicate that
progressive exercise therapy is more effective than ultrasound in
treating chronic lateral epicondylitis, reducing pain and improving
patients ability to work.

Introduction
The tennis elbow syndrome, o r lateral
epicondylitis, is generally a work-related or sportrelated pain disorder of the common extensor
origin of the arm, usually caused by excessive
quick, repetitive movements of the wrist and
forearm. These quick movements may rupture the
proximal attachment of the long extensor muscles
and cause local inflammation and pain. Cervical
root irritation, shoulder problems, local bursitis or
radiohumeral synovitis and posterior interosseus
nerve entrapment are implicated in epicondylalgia (Goldie, 1964) and adverse neural tension is
also said to contribute t o the epicondylar pain
(Yaxley and Gwendolen, 1993).

Physiotherapy, September 1996, vol82, no 9

Acute cases will usually improve when the cause

is avoided as well as with anti-inflammatory
drugs or local corticosteroid injections. According
to several studies of both conservative and operative methods, chronic cases are often difficult to
treat. They cause prolonged disability and
inability t o work (Binder and Hazleman, 1983).
The tennis elbow syndrome is an important, separate, strain injury-related cause of premature
retirement from working life.
Outcome from conservative treatment methods is
variable. Local steroid injections are widely used
in chronic cases (Labelle et al, 1992) and have
been shown to be more effective than ultrasound
and placebo ultrasound therapy in a clinical
trial (Binder et al, 1985). Manipulative and electrotherapy methods have been used with limited
effect (Wadsworth, 1987; La Freniere, 1979; Mills,
1928). Continuous and pulsed ultrasound treatment has produced conflicting results (Binder and
Hazleman, 1983; Binder et al, 1985; Haker and
Lundeberg, 1991; Kivi, 1983; Lundeberg et a l ,
1988; Reginssen, 1990) and pulsed ultrasound has
been reported to be no better than placebo ultrasound (Haker and Lundeberg, 1991).
In practice, treatment is usually a combination
of therapies which makes the effects of each separate method difficult to assess. In an attempted
meta-analysis, Labelle et al (1992) showed the
contradictory results of different therapies. Therefore, randomised therapy studies in the treatment
of lateral tennis elbow syndrome are needed t o
provide more scientific data.
There are few publications on the effect of therapeutic exercises as the sole treatment method
(La Freniere, 1979; Fillion, 1991). Strengthening
the damaged attachment of the wrist extensors
so that it can better tolerate repetitive movements might be beneficial. Continuous passive
movement improves tissue healing in many
connective tissue problems (Akeson et al, 1991)
and active rehabilitation programmes have been
successful in treating chronic back patients.
Therefore, we decided t o test if an active programme would be effective in the treatment of
chronic epicondylitis. As pulsed ultrasound is
widely used for epicondylitis (Reginssen, 1990;
Haker and Lundeberg, 1991), it was selected

523

for comparison. Increased protein synthesis in

tissues has been demonstrated at low intensities
below 0.5 W/cm2 (Dyson and Suckling, 1978).
Haker and Lundeberg (1991) used 1 MHz frequency and 1 W/cm2intensity in treating lateral
epicondylalgia with pulsed ultrasound.
Evaluation of different treatment methods is
usually based on pain relief or subjective wellbeing assessed by pain questionnaires. The effects
of therapies on arm function are rarely assessed.
Isometric grip strength appears to be a suitable
method for evaluating treatment outcome in
lateral epicondylitis (Thurtle et al, 1984). Isokinetic testing of muscle performance is a newer
method that is used more commonly in sports and
rehabilitation clinics to test function and outcome
from rehabilitation. Although it has several limitations in practical use, it is safe even in painful
conditions (Friedlander et al, 1991). Isokinetic
wrist flexion, extension, supination and pronation
torques may describe the functional limitation in
the disorder better than isometric grip strength.
Additionally, produced work, which describes arm
function better, can be evaluated with isokinetic
devices. Therefore we decided t o measure arm
function with both the isometric and isokinetic
testing methods.
The aim of the study was to explore the effects of
a progressive stretching and strengthening exercise programme in chronic lateral epicondylitis
and to compare the effects of this treatment with
the results of local pulsed ultrasound treatment.

Methods
Subjects
Thirty-nine patients (14 men, 25 women, mean
age 42.3 f 5.4 years, range 31-53 years) with clinically diagnosed chronic unilateral epicondylitis
took part in this study. They were selected from
patients referred to the Oulu University Hospital
for clinical evaluation and treatment by their
general practitioners because they had not
responded t o primary treatment. Thirty of them
had had symptoms for over three months, and
nine of them for over one year. All had been
treated in different ways before entering the
study. General data and previous treatments are
listed in tables 1and 2.
Clinical Examination and
Exclusion and Inclusion Criteria
All referred patients were examined by a physician who also examined selected patients after
the treatment period. The principal exclusion
criteria were cubital osteo-arthritis, carpal or
radial tunnel syndrome, rheumatoid arthritis,
severe cervical spondylosis or cervical radicular

Table 1: Patient data (N = 39)

Exercise group

UItrasound group

No of patients
20
Menlwomen
8112
Mean age (years)
43 (33-53)
Mean height (cm)
166 (152-182)
Mean weight (kg)
73 (50-98)
Smokerslnon-smokers 10/10
Strenuous hobby using
arms
11
Regular sport activity
6
Duration of symptoms
Under 6 months
9
Over 6 months
11
Duration of sick leave
(mean, weeks)
6.3
Non-participants*
1

19
6113
41 (31-53)
166 (154-179)
75 (52-108)
10/9
9
5
11
8
7.1
2

*Excluded from results above and from statistical analysis in

the study. See text.

Table 2: Previous treatments

~

~ _ _ _ _ _ _ _ _

Number of patients (N = 39)

Exercise gp Ultrasound gp Total
lmmobilisation
Local injections (n = 82)
Oral medication
Percutaneous medication
support
Physical therapy

4
20
18
11
11
12

4
15
18

8
7
7

8
35
36
19
18
19

syndrome, painful shoulder or rotator cuff

tendinitis, previous fractures of arm causing
limitations in arm function, and no clinical
signs of tennis elbow syndrome.
To be included in the study, all patients had
a positive Mills test (Wadsworth, 1987) and
resisted wrist and/or middle finger extension
produced typical pain at the origin on the lateral
epicondyle. A third inclusion criterion was local
tenderness on palpation over the lateral
epicondyle.
The results of the three manual provocative
tests were recorded for evaluation of treatment
outcome. Sick leave was recorded and ability
to work assessed.

Test and Measurements

Patients completed a pain and disability questionnaire including a visual analogue scale WAS).
The questions covered pain at rest, pain under
strain, ability to work, ability to lift objects,
restrictions on hobbies and sleep disturbance.
Patients indicated the intensity of the pain on a
10 cm long non-segmented line. The questions
were to determine the extent of pain in the upper
limb. Pain under strain was also assessed during
testing of muscle strength.
Muscle function was measured independently by
a third physiotherapist who was blind to the
patients therapy group and who did not treat the
patients at all.

Physiotherapy, September 1996, vol82, no 9

524

Isometric grip strength was measured with a

strain-gauge dynamometer (Newtest Oy, Oulu,
Finland) whose repeatability and usefulness
has been demonstrated (Malkia, 1983). Three
measurements were made of three different grips
(1 cm, 2 cm and 3 cm). The mean value of nine
repetitions was calculated and the maximum
value of repetitions was recorded.
The isokinetic muscle performance of the wrist
and forearm was measured with a n isokinetic
multi-joint active device (Loredan, Davis, California), the Lido Active multi-joint dynamometer
which has previously been demonstrated to be
both valid and reliable (Patterson and Spivey,
1992). It was used at a velocity of 90"/second in
wrist flexion-extension movements and in
forearm supination-pronation movements. The
muscle work used was concentric. Ten repetitions
were made at each position. From the versatile
collected data, peak torque and total work results
were analysed.

Procedure
The physician selected patients for the trial on the
basis of his clinical assessment and randomly allocated them to exercise and ultrasound groups by
drawing lots. There were three immediate dropouts, one in the exercise group and two in the
ultrasound group, for personal reasons which
prevented participation.
The 20 patients in the exercise group were trained
in a four-step home exercise programme. They
visited the physiotherapist once every other week
for follow-up examination and t o receive a new,
more intensive programme. The exercises started
with slow fist-clenching, resisted wrist movements, and wrist rotations with a stick (step 11,
followed by movements against a band (step 2)
and two-way resisted wrist rotations and pressing
hands against a wall (step 3 ) . The patients
performed the exercise programmes four t o six
times daily a t home. Each programme included
ten repetitions in two or three series for each exercise. The fourth step was a versatile occupational

Fig 1: Clenching fist strongly

Physiotherapy, September 1996, vol 82, no 9

training programme. Every exercise period ended

with stretching for at least 30 seconds in both
flexion and extension and each individual exercise
movement was done slowly while the patient
counted to eight. The steps of the programme are
presented in table 3 and figures 1to 10.
Table 3: Progressive exercise programme, steps 1 to 4

Exercises

Figures

Step 1
Clenching fist strrongly
Resisted wrist extension
Resisted wrist flexion
Wrist rotation with a stick
Towards the little finger
Towards the thumb
End: stretching at least 30 seconds
to flexion and extension
Step 2
Exercises against an elastic band for:
Wrist extension
Wrist flexion
Wrist radial deviation
Wrist ulnar deviation
End: stretching 30 seconds (as in step 1)
10 x 3 series, several repetitions daily
Step 3
Combined wrist rotary movements
using eg tabie top as a support
Upwards, resisted from below
Towards the little finger
Towards the thumb
Donwards, resisted from above
Towards the little finger
Towards the thumb
Pressing hands against a wall
End: stretching 30 seconds
10 x 3 series, several repetitions daily
Step 4
An occupational training programme, including:
Soft ball compressing exercises
Transferring buttons from cup into another
Twisting a towel into a roll
Rotating hand on a table, in both directions, etc

7
8

10

End: stretching 30 seconds

This program can be performed together with one of steps 1-3
Each separate movement and exercise in each step must be
done while slowly counting to eight.

Fig 2: Resisted wrist extension exercises

525

Fig 3: Resisted wrist flexion exercises

Fig 7: Wrist flexion exercises against an elastic band

Fig 4: Wrist rotation with a stick towards the little finger

Fig 8: Exercises for radial deviation

Fig 5: Stretching to flexion

Fig 9: Pressing hands against a wall

Fig 6: Wrist extension exercises against an elastic band

Fig 10: Twisting a towel into a roll

Physiotherapy, September 1996, vol 82, no 9

526

The 19 patients in the ultrasound group received

local pulsed ultrasound from 0.3 t o 0.7 W/cm2
two t o three times a week from a Diter S 3000
ultrasonic machine (Diter-Elektroniikka Oy,
Turku, Finland). The pulse ratio was 1:5, the
duration of pulse 2 ms, and frequency 1MHz. The
radiated area was 5 cm2 over the common
extensor origin. Treatment time was 10 t o 15
minutes.
The treatment for each group was continued for
six t o eight weeks during which time they had
no other treatment. The same physiotherapist
carried out the exercise programmes and treated
four of the 19 patients in the ultrasound group.
The other 15 patients were treated by another
physiotherapist with exactly the same treatment
protocol.
The questionnaire was completed and clinical
examination, isometric and isokinetic muscle
testing were done with exactly the same protocols
before and after the treatment period.

Analysis of Data
Changes in sick leave, clinical manual tests, pain
scores, grip strength and isokinetic muscle performance of the upper limbs were calculated from
the data and compared between the groups.
Statistically the results within and between
the treatment groups were compared using the
Students t-test, ANOVA with repeated measures,
the Wilcoxon test for matched pairs or the MannWhitney test.

Results
The Sample
The mean age of the sample (42 years) is typical
for tennis elbow patients. Data are summarised in
table 1.There were minor differences between the
study groups. Patients in the ultrasound group
were two years younger than patients is the exercise group (average 41 and 43 years respectively).
There were a few more men in the exercise group
(40%) than in the ultrasound group (32%)(and the
mean initial values of isometric grip strength and
isokinetic torques were consequently lower in
the ultrasound group). The occupations of the
patients were similar i n both groups although
more patients in the ultrasound group who
had heavy work were unable to work. Patients in
the ultrasound group had also a little more pain
under strain and were more restricted is pursuing
their hobbies at the beginning of treatment. They
had had a variety of treatments before being
referred by their general practitioners (table 2).
Analysed data were from three sources: the pain
questionnaire filled in by patients providing

Physiotherapy, September 1996, vol 82, no 9

subjective data, objective tests of muscle function

carried out by the physiotherapist and the
manual provocative tests and assessment of
ability to work carried out by the physician.

Pain (table 4)
Pain at rest increased in the ultrasound group
0.2 cm on the VAS and decreased 1.9 cm on the
VAS in the exercise group during treatment. The
difference between the groups was statistically
significant (p = 0.004). Pain under strain diminished 1.4 cm in the ultrasound group and 3.5 cm
in the exercise group (p = 0.04). From the pain
questionnaire scales, the mean subjective
inability to work declined 3.3/10 cm WAS) in the
exercise group and 0.5/10 cm in the ultrasound
group. The mean final values were 3.7 and 6.7 cm
in the groups. The difference between the groups
was statistically significant (p = 0.004). Additionally, mean sleeping disturbance decreased 1.9 cm
on VAS in the exercise group and increased 0.2 cm
in the ultrasound group. The difference between
the groups was statistically significant (p = 0.01).
Changes in restrictions on hobbies were almost
the same in both groups.
Isokinetic Tests (table 5)
In the exercise group, the mean isokinetic torque
of wrist flexion increased by 45% from 10.1 to 14.7
Nm. The torque declined from 9.6 to 9.3 Nm (4%)
in the ultrasound group. Both the difference
between the groups (p = 0.0002) and the change
within the exercise group (p = 0.0001) were statistically significant . The results were quite similar
in flexor work production (p = 0.0002 between
group difference),which correlated well with peak
torque values. The mean values of work produced
by wrist flexion increased by 62% from 10.0 to
16.1 joules per repetition. In the ultrasound
group, the mean values were 9.7 and 9.2 joules
and the work produced decreased 5%. The extent
of change was statistically significant between
the groups (p = 0.0002). Total work done per
repetition in wrist extension improved in the
exercise group and the difference between the
groups was statistically significant (p = 0.05).
No significant differences between the groups
were found for wrist extension peak torques,
forearm pronation or supination peak torques or
produced work.
Grip Strength (table 6)
Although the maximal isometric grip strength
increased 12.2% in the exercise group and
remained unchanged in the ultrasound group, the
difference between the groups was not statistically significant (p = 0.05). The change in
isometric grip strength was similar to the isokinetic muscle performance in wrist flexion.

527

Table 4: Mean changes (decreases/increases in cm) on visual analogue scale of different pain qualities in
both groups
f a i n quality

Exercise group
(n = 20)
Initial
Change
value
cm (SO)

Pain at rest
Pain under strain
Working inability
Lifting inability
Hobby limitations
Sleep disturbance

3.7
7.3
7.0
4.2
4.6
3.8

-1.9
-3.5
-3.3
-2.1
-1.5
-1.9

Ultrasound group
(n = 19)
Initial
Change
value
cm (SO)

(1.8)
(3.5)
(2.7)
(4.8)
(2.0)
(1.9)

3.7
7.8
7.0
4.9
6.1
4.4

0.2
-1.4
-0.5
-1.3
-1.0
0.2

Difference
of changes

-0.7,
-9.1,
-0.9,
1.7,
1.5,
-0.6,

-2.1
-2.1
-2.8
-0.8
-0.5
-2.1

(2.6)
(2.9)
(2.9)
(2.7)
(3.8)
(3.0)

95% CI
of difference
between
groups
-3.6
-4.3
-4.6
-3.3
-2.4
-3.8

P between
groups

0.004
0.04
0.004
0.52*
0.06*
0.01

The table contains initial mean values and changes (cm) and standard deviations (SD) of changes. Negative change (-) shows
decrease in reported pain. Statistical significance between the groups is tested with Student's unpaired t-test. *Not significant

Table 5: Changes (mean and SD) between initial and final isokinetic peak torque (Nm) and isokinetic work per
repetition (joules) values of wrist and forearm muscle performance in both treatment groups at an angular
velocity of 90"/second
Movement

Exercise group
(n = 20)
Initial
Change
value
cm (SO)

Ultrasound group
( n = 19)
Initial
Change
value
cm (SO)

Difference
of changes

95% CI
of difference
between
groups

P between
groups

Wrist
Flexion
Torque (Nm)
Work (J)

10.1
10.0

+4.6 (3.9)
+6.1 (5.0)

9.6
9.7

-0.4
-0.4

(2.9)
(3.5)

5.0
6.5

2.7, 7.4
3.5, 9.6

0.0002
0.0001

Torque (Nm)
Work (J)

5.7
6.0

+0.6 (1.7)
+0.9 (2.0)

5.4
5.4

-0.1
-0.3

(1.6)
(1.3)

0.7
1.2

-0.4, 1.9
0.0, 2.4

NS
0.05

Forearm
Supination
Torque (Nm)
Work (J)

6.4
7.3

+0.9 (2.1)
+0.9 (2.6)

6.0
6.7

0.0 (1.5)
-0.8 (2.3)

0.9
1.7

-0.5,
-0.2,

2.4
3.6

NS
NS

6.1
6.7

+0.9 (1.3)
+0.6 (1.2)

5.7
6.1

-0.1
-0.4

(1.8)
(2.4)

1.o
1.o

-0.2,
-0.3,

2.1
2.4

NS
NS

Extension

Pronation
Torque (Nm)
Work (J)

Table 6: Mean and maximal isometric grip strength (newtons) of involved arms before and after treatment in
both groups
~~

Before treatment
Mean (SD)

After treatment
Mean (SO)

Change

P within
group

Differences
between groups

P between
groups

Mean isometric grip strength 9 repebtions

EG

305

(152)

357

(149)

+52

0 003

UG

244

(100)

255

(103)

+11

0 45*

0 02

41

0 6*

41

0 05

Maximal isometric grip strength

EG

361

(159)

404

(164)

+43

UG

301

(114)

303

(122)

+2

0 92*

Values are means (SD) Change within each treatment group is analysed with Student's t-test and statistical difference between the
treatment groups using analysis of variance with repeated measures EG = exercise group (n = 20) UG = ultrasound group (n = 19).
*Not significant

Physiotherapy,September 1996, vol82, no 9

528

Manual Provocative Test (table 7)

To meet the inclusion criteria, the three tests
were positive for all patients before treatment.
Nine patients in the exercise group and 16 in the
ultrasound group had positive Mills test after
treatment. The change was statistically significant within the exercise group. Changes in
palpation and in resisted wrist or middle finger
extension were significant during the treatment
period for the exercise group also. The changes
in the provocative tests were not statistically
significant for the ultrasound group.
Table 7: Positive clinical manual tests in both treatment
groups before and after treatment
~

Before
treatment

After
treatment

Change Zvalue
within
group

Pvalue
within
group

Palpa tion, lateral epicondyle

EG
UG

20
19

13
17

-7
-2

-2.4
-1.3

0.02
0.18*

-8
-1

-2.5
-0.53

0.01
0.59*

Resisted wrist extension

EG
UG

20
18

12
17

Resisted middle finger extension

EG
UG

19
17

11
15

-8
-2

-2.2
-1.34

0.02
0.18*

20
19

9
16

-11
-3

-2.9
-1.6

0.003
0.11*

Mills test
EG
UG

Values are number of patients = number of positive tests. Statistical significance within each group is tested with non-parametric
Wilcoxon test for matched pairs.
EG = exercise group (n = 20). UG = ultrasound group (n = 19)
*Not significant

Ability to Work and Sick Leave (table 8)

Before treatment, 40% of the exercise group and
47.4% of the ultrasound group were unable t o
work. After the therapy, only 10% of the exercise
group and 31.6% of the ultrasound group were
absent from work. The difference between the
groups was not statistically significant (p = 0.07).
Table 8: Patients unable to work before and after the treatment in both groups
Before treatment
No
(A)
Exercise group (n = 20)
Ultrasound group (n = 19)

8
9

(40.0)
(47.4)

After treatment
No
(A)
2
6

(10.0)
(31.6)

Discussion
The study fulfilled the aims of exploring the
effects of progressive exercise therapy on chronic
lateral epicondylitis and comparing these effects
with outcomes from pulsed ultrasound. The
progressive strengthening and stretching exercise

Physiotherapy, September 1996, vol 82, no 9

therapy resulted in significantly better subjective

and objective short-term outcomes than pulsed
ultrasound therapy. It reduced pain at rest and
under strain, improved arm function and
muscular performance of the wrist and forearm.
Results of clinical tests improved in the exercise
group showing a good correlation with the healing
process. The exercise treatment also had good
effects on subjective and objective working ability,
which is important in the prevention of prolonged
disability.

Changes in Level of Pain, Function and

Ability to Work
The main purpose of statistical analysis was
to identify significant changes in pain, muscle
function of the involved upper limbs, working
ability and clinical manual testing within and
between the treatment groups and t o compare
differences between the groups. Therefore the
extent of change, not the level of initial values,
is the important factor.
The wrist flexion torque is reduced in the tennis
elbow syndrome. Not only was it improved significantly by the exercise programme but the
improvement correlated well with functional
improvement in the upper limb and changes in
maximum isometric grip strength. Little change
was noted in the extension torques of the wrist
but improved work production values show that
ability to use the extensor muscles improved in
the exercise group. Exact measurement analysis
was more difficult on the extensor side, because
the peak torque values were small (mostly only 3
to 5 Nm), than on the flexor side where torques
were much greater. The relatively small effect on
extension forces can be explained by the fact that
the damage on the extensor side of the wrist is
linked to eccentric rather than concentric muscle
work, and eccentric muscle work was not tested
in this study.
The isometric grip strength did not increase as
much as wrist isokinetic flexion torque, perhaps
because of greater painful strain on the affected
muscle tendon region in isometric muscle contraction. Maximal isometric grip strength increased
significantly more in the exercise group, reflecting
short-term ability to use the hand. The treatment
period was not long enough t o provide results
about endurance.
Pain at rest and under strain also declined significantly more in the exercise group. This shows
that active exercises may have an effect on pain
experience in patients. Reported pain under
strain represents both painful isometric and
isokinetic muscle work in patients responses
in the pain questionnaire. Pain has an important role underlying decreased muscle function

529

and cannot be ignored in analysis of results from

muscle function tests.
Additionally, characteristics of a patients occupation affect the ability to work. Different types of
work produce different strains on the upper limb
in different patients and, therefore, the change
in ability to work noted within each treatment is
important. In the exercise group, more patients
became able t o do their usual work than in the
ultrasound group.
Finally, while the results are overwhelmingly in
favour of exercise therapy, tennis elbow syndrome
is reported to be a self-limiting disorder in some
cases (Cyriax, 1936). It is impossible t o know
whether or not some of the patients in this study
had such a self-limiting disorder and, therefore,
whether all measured improvements in pain
ratings, muscle function o r ability t o work are
attributable to treatment.

Therapeutic Considerations
The programme of progressive exercise therapy
used in this study appears to offer considerable
health gain. I t is inexpensive, because the
patients take a n active role and require few
treatments and only four follow-up visits t o a
physiotherapist. It aims to strengthen damaged
tissues. Using slow, repetitive exercise movements for strengthening the soft tissues of the
upper limb appears to have beneficial effects in
treating chronic strain injuries. As the aetiology of
strain injuries such as the tennis elbow syndrome
may be repeated rapid movements, slow progressive strengthening exercises may allow tissue
healing.
Early mobilisation is reported t o have good effects
on the tensile strength of connective tissue scars
in muscle injury in acute cases (Celberman et al,
1988; Kannus et al, 1992). Eccentric exercises
seem t o be stressful to the myotendinous unit
(Appell, 1990) and a trend from avulsion-type
failures in immobilisation, t o insertion and
midsubstance-type failures in remobilisation,
has been noted in animal knee ligament studies
(Larsen et a l , 1987). Therefore, a progressive,
stepwise exercise programme can promote
healing without traumatisation. The damaged
epicondylar attachment area is an osteotendinal
region with the properties of inflamed and
atrophied tendon and, in prolonged cases,
bony atrophy too. Tipton et a1 (1987) say that
prescribed exercises which increase the forces
being transmitted t o ligaments, tendons and
bones will maintain and generally increase the
strength and functional capacity of these structures. The same principle seems to be valid in the
treatment of chronic tennis elbow syndrome. The
progressive exercise treatment used in this study

started with slow soft tissue-stretching exercises.

The whole programme exercised muscles, tendons
and ligaments and also the osteotendineal insertion region, and the fourth step was a more
intensive occupational programme t o promote
patients daily living and ability to work.
Pulsed ultrasound was not found to be effective as
a sole treatment in treating chronic epicondylar
pain. Although pulsed ultrasound 1MHd0.5 W cm2
has been found t o have beneficial effects on
protein synthesis (Harvey et al, 1975; Dyson and
Suckling, 1978),this kind of ultrasound treatment
procedure cannot be recommended. Although the
intensity applied in this study was lower, the
treatment procedure was similar to the procedure
used in the study by Haker and Lundeberg (1991)
in which they reported no beneficial effect for
pulsed ultrasound over placebo ultrasound.

Conclusions and Recommendations

The results indicate that progressive strengthening and stretching exercise treatment is more
effective than pulsed ultrasound in treating
chronic lateral epicondylitis: it reduced chronic
pain and improved upper limb function and the
ability t o work of patients in the study. I t may
correct the ill-effects of prolonged immobilisation,
counter patients fear of using the forearm and
hands, and help them to return to work.
Authors
T Tuomo Pienimaki MD was head researcher and principal
author. He carried out the clinical evaluation and data analysis.
K Tuula Tarvainen PT is a physiotherapist. T Siira Pertti P T
carried out the muscle function testing. Heikki Vanharanta MD
is a professor of physical medicine and rehabilitation in Oulu
University and was the study supervisor.
This article was received on October 20, 1995, and accepted on
May 16, 1996.

Address for Correspondence

Dr Tuomo Pienimaki, Department of Physical Medicine and Rehabilitation, Oulu University Hospital, Kajaanintie 50, FIN-90220
Oulu, Finland.

References
Akeson, W H, Amiel, D, Woo, S L-Y, Abitbol, J-J and Garfin, S R
(1991). Concepts of soft tissue homoeostasis and healing in:
Mayer, T G, Mooney, V and Gatchel, R J (eds). Contemporary
Conservative Care for Painful Spinal Disorders, Lea and Febiger,
Pennsylvania, pages 84 -101.
Appell, H-J (1990). Muscular atrophy following immobilisation.
A review, Sports Medicine, 10, 48-58.
Binder, A I and Hazelman, B L (1983). Lateral humeral
epicondylitis - Natural history and the effect of conservative
therapy, British Journal of Rheumatology, 22, 73-76.
Binder, A, Hodge, G, Greenwood, A M, Hazelman, B Land Page
Thomas, D P (1985). Is therapeutic ultrasound effective in
treating soft tissue lesions? British Medical Journal, 290,
512-514.
Celberman, R, Coldberg, V, An, K-N and Banes, A (1988).
Tendon in: Woo, S L-Y and Buckwalter, J A (eds) Injury and
Repair of the Musculoskeletal Soft Tissues, American Academy
of Orthopaedic Surgeons, Park Ridge, II, pages 5-40.

Physiotherapy, September 1996, vol 82, no 9

530

Cyriax, J (1936). 'The pathology and treatment of tennis elbow',

Journal of Bone and Joint Surgery, 18, 921-940.

LaFreniere, J G (1979). 'Tennis elbow: Evaluation, treatment and

prevention', Physical Therapy, 59, 742-746.

Dyson, M and Suckling, J (1978). 'Stimulation of tissue repair

by ultrasound: A survey of mechanisms involved', Physiotherapy,
64,4, 105-108.

Larsen, N P, Forwood, M R and Parker, A W (1987). 'Immobilisation and retraining of cruciate ligaments in the rat', Acta
Orthopedica Scandinavica, 58, 260-264.

Fillion, P L (1991). 'Treatment of lateral epicondylitis', American

Journal of Occupational Therapy, 45, 340-343.

Lundeberg, T, Abrahamsson, P and Haker, E (1988). 'A comparative study of continuous ultrasound, placebo ultrasound and
rest in epicondylalgia', Scandinavian Journal of Rehabilitation
Medicine, 20, 99-101.

Friedlander, A L, Block, J E, Byl, N N, Stubbs, H A, Sadowsky,

H S and Genant, H K (1991). 'Isokinetic limb and trunk muscle
performance testing: A short-term reliability', Journal of Orthopedics and Sports Physical Therapy, 14, 220-224.
Haker, E and Lundeberg, T (1991). 'Pulsed ultrasound treatment
in lateral epicondylitis', Scandinavian Journal of Rehabilitation Medicine, 23, 115-118.
Harvey, W, Dyson, M, Pond, J 6 and Grahame, R (1975). 'The
"in vitro" stimulation of protein synthesis in human fibroblasts by
therapeutic levels of ultrasound', Exerpta Medica, lnternational Congress Series No 363, pages 10-21.
Kannus, P, Jozsa, L, Renstrom, P, Jarvinen, M, Kvist, M, Lehto,
M, Oja, P and Vuori, I (1992). 'The effects of training, immobilisation and remobilisation on musculoskeletal tissue', review article,
Scandinavian Journal of Medicine and Science in Sports, 2,100118.
Kivi, P (1983). 'The etiology and conservative treatment of
humeral epicondylitis', Scandinavian Journal of Rehabilitation
Medicine, 15, 37-41.
Labelle, H, Guibert, R, Joncas, J, Newman, N, Fallaha, M, and
Rivard, C-H (1992). 'Lack of scientific evidence for the treatment
of lateral epicondylitis of the elbow', Journal of Bone and Joint
Surgery (Br), 74-8,646-651.

Malkia, E (1983). Muscular Performance as a Determinant of

Physical Ability in Finnish Adulf Population. Publications of the
Social Insurance Institution, Finland, AL:23, page 145.
Mills, G P (1928). 'Treatment of tennis elbow', British Medical
Journal, 12-13.
Patterson, L A and Spivey, W E (1992). 'Validity and reliability of
the LIDO active lisokinetic system', Journal of Orthopaedic and
Sports Physical Therapy, 15, 32-36.
Reginssen, T (1990). 'Effekt av ultralydbehandling - En klinisk
kontrollert studie', Fysioterapeuten, 57, 5-8.
Tipton, C M, Vailas, A C and Matthes, R D (1987). 'Experimental
studies on the influences of physical activity on ligaments,
tendons and joints: A brief review', Acfa Medica Scandinavica,
711 , 157-68.
Wadsworth, T G (1987). 'Tennis elbow: Conservative, surgical
and manipulative treatment', British Medical Journal, 294, 621624.
Yaxley, G A ,and Gwendolen, A J (1993). 'Adverse tension in
the neural system. A preliminary study of tennis elbow', Australian
Physiotherapy Journal, 39,1, 15-22.

periodical review
This feature outlines the principal contents of journals and newsletters of
Specific Interest Groups. To obtain the publications please contact the
honorary secretary of the appropriate group, published in the Annual Report
(dispatched to members in April).

Association of Chartered Physiotherapists in

Sports Medicine

A personal view of Parkinson's disease

Physiotherapy in Sport

1996, vol XIX, no 2, June/July

Tennis elbow - And eccentric rehabilitation

The effects of foam, athletic shoes and aircasts upon

L Curry, W Harpur
the postural sway of healthy females
National Vocational Qualifications

U McC Persson

H J Challis

Sporting injuries to the elbow

Tennis elbow and raquet design
Acupuncture and the treatment of tennis elbow

P Ludgate
R J D'Souza

Steroid phonophoresis - An alternative

to injections?

R W Watson

AGILE: Association of Chartered

A Kinloch

C Rogerson, M Williamson

Association of Chartered Physiotherapists

in Animal Therapy
Journal

,gg6, June
The quack physio
Rehabilitation for dogs - Whyever not?

P Windle-Baker

0 Hanney

Computerise your practice

P Verey

Setting the standards?

T Crook

Physiotherapists with Elderly People

Agility

1966, January
Integrated medicine: A physiotherapy view
Social service contributions to the
rehabilitation of older people
Falls in the elderly: What can we learn
from the FlCSlT trials?
Functional reach in clinical practice
Solving footwear problems

Physiotherapy, September 1996, vol 82, no 9

A Culot

Organisation of Chartered Physiotherapists

in Private Practice
In Touch
1996, no 80, Summer

H Mandelstam

Countdown to self-assessment

J Bruce
E Wilson

R Smith

The Physiolympics
Accreditation
Muscle energy technique

J Kelly, C Wigley
E Wilson

J Dow

Sleep and its side effects

B Ancell

J Simpson