ORIGINAL ARTICLE

Treatment of Breast-Cancer-related Lymphedema

With or Without Manual Lymphatic Drainage
A Randomized Study
Lene Andersen, Inger Højris, Mogens Erlandsen and Jørn Andersen

From the Department of Oncology (L. Andersen, I. Højris, J. Andersen), Biostatistics (M. Erlandsen) and the
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

Danish Cancer Society, Department of Experimental Clinical Oncology (I. Højris), Aarhus University Hospital,
Denmark

Correspondence to: Lene Andersen, Department of Oncology, Aarhus University Hospital, Nørrebrogade 44,
DK-8000 Aarhus C, Denmark Tel: +45 89 49 25 45 Fax:+ 45 89 49 25 40

Acta Oncologica Vol. 39, No. 3, pp. 399–405, 2000

A prospective randomized study was carried out to investigate whether the addition of manual lymphatic drainage (MLD) to the standard
therapy could improve treatment outcome in women with lymphedema of the ipsilateral arm after breast cancer treatment. Forty-two
patients were randomly assigned to receive standard therapy or standard therapy plus MLD 8 times in 2 weeks and training in
self-massage. The standard therapy consisted of use of a compression garment, exercises and information about lymphedema and skin
For personal use only.

care. The efficacy of treatment was evaluated by reduction in lymphedema volume during treatment and by improvement in symptoms
potentially related to lymphedema. The patients were followed-up for a total of 12 months. The study showed that both groups obtained
a significant reduction in edema and that MLD did not contribute significantly to reduce edema volume.
Recei6ed 12 January 2000
Accepted 24 February 2000

One of the complications of breast cancer treatment is (8). Intensive treatment is mainly used for severely swollen
lymphedema of the ipsilateral arm (1, 2). Lymphedema is or misshapen limbs, where an elastic garment cannot be
defined as a swelling of the arm caused by insufficient fitted (3, 11). MLD is a gentle massage technique, which
lymph drainage (3). It may result in cosmetic deformity, stimulates the lymphangiomotoric activity, as demon-
loss of functional ability, physical discomfort, recurrent strated by Mislin (12). This directs the lymphatic flow
episodes of erysipelas and psychological distress (4, 5). The away from the edematous part of the trunk and arm and
incidence of lymphedema after treatment for breast cancer thereby decreases the edema and fibrous changes in the
arm. MLD is a part of DLT and should be adjusted to the
ranges between 6% and 38%, depending on the extent of
individual patient, but MLD alone has been found inade-
axillary surgery and the use of radiotherapy (5 – 7).
quate (13) (14). The main constituent of DLT is compres-
Lymphedema can be divided into 3 stages. During the first
sion by elastic sleeves. The patient’s own contribution
‘reversible’ stage a protein-rich edema is present. Stage 2,
includes skin care, exercises and if necessary in combina-
designated as ‘spontaneously irreversible’, presents
tion with MLD. The intensity of application of the indi-
fibrosclerotic alterations and an increase in the number of
vidual components of DLT depends on the stage of
ceratinocytes and connective tissue cells. Stage 3, ‘elephan-
lymphedema at the time of treatment starts (13, 14).
tiasis’, is characterized by massive hyperceratosis and by a Uncomplicated cases of lymphedema can be treated in an
tremendous increase in the volume of the limb (8). outpatient setting (8) (13). In a study using DLT, Oliver
Lymphedema may arise immediately after treatment or Leduc et al. found that the most important reduction of
show up after several years. the edema was obtained in the first week. During the
Decongestive lymphatic therapy (DLT) can be effective second week, the results obtained were stabilized (14). In
in reducing lymphedema (9, 10). DLT is a combination of our experience, women treated for breast cancer with an
intensive treatment using compression by bandages, man- uncomplicated edema in stage 1 or 2 have a notable
ual lymphatic drainage (MLD), exercises enhancing the reduction effect with a standard therapy consisting of daily
lymphatic flow, and skin care. This is usually followed by use of compression sleeves, exercises, skin care and
daily use of compression garments, exercises, and skin care precautions.

© Taylor & Francis 2000. ISSN 0284-186X Acta Oncologica

 400 L. Andersen et al. Acta Oncologica 39 (2000)

The aim of the present study was to investigate whether mobility in two plans, i.e. extension – flexion and adduc-
addition of MLD to our standard therapy improved the tion – abduction. The patients completed a questionnaire
outcome in women with modest lymphedema stage 1 or 2 on symptoms possibly related to lymphedema, which they
after treatment for breast cancer. As the most important graded from 1 to 7. In addition, the patients were asked to
reduction of the edema is obtained in the first two weeks what extent they complied with the treatment instructions
(14), we chose to add MLD, given 8 times in 2 weeks. given.
The women also completed the EORTC QLQ-C30 ques-
MATERIAL AND METHODS tionnaire for breast cancer, but these data are not reported
in the present study.
Patients
This study included 42 women who after treatment for Study design
early breast cancer had developed unilateral lymphedema The patients were randomly assigned to receive standard
of the arm. The patients were seen in the outpatient therapy or standard therapy plus MLD and training in
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

lymphedema clinic at the department of oncology, if self-massage. The standard therapy consisted of a custom-
lymphedema was found at a routine follow-up visit or if made sleeve-and-glove garment providing 32 – 40 mmHg
the patients referred themselves to the department because compression (Jobst-Elvarex, compression class 2, Beiers-
of a swollen arm. Prior to enrolment in the study, each dorf, Sweden), educational information and recommenda-
patient was examined to exclude lymphedema caused by tions about lymphedema, instruction in physical exercises
recurrence of breast cancer. The women underwent a to enhance the lymph flow, education in skin care and
physical examination, ultrasound of the axillary and peri- safety precautions. In the experimental arm, MLD was
clavicular regions and x-ray of the shoulder. All the given 8 times in 2 weeks, and the patients were further-
women were outpatients. more educated in daily self-massage using a simple form of
The criteria for entry in the study were: one or more MLD. One hour was reserved for each visit.
symptoms of lymphedema (numbness, tightness, stiffness, Patients randomized to standard therapy alone were
pain, aching, heaviness or other kinds of discomfort), a
For personal use only.

allowed to crossover after three months, if they found that

difference in volume between the two arms of at least 200 the treatment response up until then was unsatisfactory.
ml (measured to a level 15 cm above the elbow), and/or a Assessments including objective measurements and
difference between the circumference of the two arms of at questionnaires about symptoms related to lymphedema
least 2 cm (measured 15 cm above or 10 cm below the and compliance with the use of compression sleeves, exer-
elbow). Eligible patients had to be at least four months cises, and self-massage were performed after 1, 3, 6, 9 and
after surgery, because some cases of lymphedema accom- 12 months. A further follow-up visit after 4 months was
panying operation and radiotherapy can resolve sponta- arranged for women randomized to standard therapy
neously. Women were not eligible for enrolment to the alone who in addition chose to have the MLD and educa-
study if they showed evidence of recurrence, had bilateral tion in self-massage. The social circumstances were ap-
breast cancer, or if they had received treatment for praised and patients could be referred to a social welfare
lymphedema during the preceding three months. Patients officer. The study participants were encouraged to contact
with severe lymphedema, defined as a difference in arm the lymphedema clinic whenever any unexpected problems
volume exceeding 30%, were not included in the study but arose, in order to tackle these without delay.
were offered DLT, including compression bandaging. The endpoints of the study were the change in volume of
However, if they declined to receive this more extensive the ipsilateral arm compared to the contralateral arm, and
treatment, they were allowed to participate in the study. patient-reported symptoms potentially related to
lymphedema.
Assessments
At the time of enrolment, a complete history was obtained Statistical methods
from each woman on type and side of operation, the The study was designed to detect a percentage reduction in
number of excised axillary lymph nodes, the number of absolute edema volume of 20% after 3 months compared
tumor-positive lymph nodes, radiotherapy technique, adju- to the baseline absolute edema volume with a significance
vant systemic treatment, duration of lymphedema, previ- level of 5% and a type II error of 10% (power of 90%).
ous episodes of infections, and injuries to the arm. The (For example, a reduction in absolute edema volume from
circumference of both arms was measured starting at the 30% to 50%.) The absolute edema volume is defined as the
wrist and repeated for every 5 cm proximally for a total of difference in volume of the lymphedema arm compared to
40 cm. The volume of each arm was calculated from these the contralateral arm. This required inclusion of 42 pa-
measurements using numerical integration by piecewise tients. The effect of treatment was analyzed by intention to
quadratic approximation, known as Simpson’s rule of treat. The level of statistical significance was set to 5%. All
integration. Shoulder function was measured as the active the estimated p-values are two-tailed.
 Acta Oncologica 39 (2000) Treatment of lymphedema 401

For assessment of differences in baseline characteristics arm. T is the time of treatment assessment and B is
between groups, Fisher’s exact test was used if the data baseline (pretreatment) assessment. The changes in the
were categorical and the Wilcoxon test was used for con- natural logarithm to the ratio of the volume of the
tinuous variables. lymphedema arm compared with the volume of the con-
tralateral arm were calculated as follows: change in ln
ratio =ln (Vo /Vc)B-ln (Vo/Vc)T.
Analysis of 6olume changes.
Analysis of such longitudinal studies has to take into
Comparison of the treatment effects of the groups was account that each individual is measured several times
performed as a two-way analysis of variance with repeated during the study (repeated measurements), causing correla-
measurements over time. The natural logarithm to the tion between observations on the same individual. For
ratio of the volume of the lymphedema arm compared to statistical reasons it is not a good idea to base the analysis
the contralateral arm was chosen as the response variable on the changes in absolute difference, since for a given
in order to fulfil the statistical assumptions for such an individual the same set of baseline values are used in the
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

analysis. This implies that the estimated levels at each time computation on each of the changes. In this study the
point correspond to relative changes in volume of the diagnostic checking of the statistical model (inspection of
lymphedema arm compared to the contralateral arm. The residuals) led us to conclude that the analysis should be
model included three variance components: (a) inter-indi- based on the differences in natural logarithm to the ratio
vidual, (b) intra-individual and (c) residual variation (mea- instead of the absolute differences. The differences between
surement error). The intra-individual variation is modeled the two methods are illustrated in Table 1, in three selected
as a random change in level 6 months after the first patients. Subjective inspection of the raw data (baseline
measurement. This model gave a substantially better fit and 3 months) suggests that the reduction in lymphedema
than the more simple compound symmetry model, which volume is most pronounced for patient no. 15, followed by
includes only the variance components (a) and (c) above patient 39 and least for patient 34. In patient 15, both
and is often used to analyze repeated measurements. The methods agree and give a large reduction. In patient 34,
model also included an extra variable to indicate whether
For personal use only.

the reduction in absolute difference compared with the

the lymphedema arm was identical to the dominant arm or baseline gives the impression of a marked reduction,
not. And finally, the duration of lymphedema was exam- whereas the reduction in the natural logarithm to the ratio
ined by adding this variable as a linear regressor. The of the volumes is only modest. In patient 39, the changes
analyses were performed by means of Proc Mixed in SAS are somewhat reversed compared with those for patient 34.
(version 6.12). Thus evaluation of treatment results by calculating the
In accordance with the original study design, the reduc- change in the natural logarithm to the ratio of the volumes
tion in absolute edema volume after 3 months compared is in closer agreement with the subjective impression than
with the baseline absolute edema volume was assessed as by calculating the change in absolute edema volume com-
well. Student’s unpaired t-test was used to compare the pared with the baseline absolute edema volume.
treatment groups.
The change in absolute edema volume compared with
the baseline absolute edema volume was calculated as Analysis of subjecti6e measures.
follows: change in absolute edema =((Vo – Vc)B-(Vo – Although the measures are scores on an ordinal and not a
Vc)T))/(Vo – Vc)B. Where Vo is the volume of the continuous scale, we have analyzed this data in the same
lymphedema arm and Vc is the volume of the contralateral way as the volume changes described above. Since the

Table 1
Comparison of the estimated treatement outcome in 3 selected patients by calculation of the change in absolute edema and by calculation of
the change in the natural logarithm to the ratio

Patient Baseline (B) 3 months (3 mths) Absolute difference3 Change Ln (ratio) difference4 Change
number %
V10 (ml) V2c (ml) V0 (ml) Vc (ml) Baseline 3 months Baseline 3 months
ml ml

15 2 871 1 687 1 972 1 628 1 184 344 70.0 0.532 0.192 0.340
34 1 866 1 647 1 684 1 590 219 94 57.1 0.125 0.057 0.067
39 2 749 1 451 2 263 1 519 1 298 744 42.7 0.639 0.399 0.240
1
V0 is the volume of the lymphedema arm.
2
Vc is the volume of the contralateral arm.
3
Change in absolute edema =(V0−Vc)B−(V0−Vc)3 months/(V0−Vc)B
4
Change in ln ratio=ln (V0/Vc)B−ln (V0/Vc)3 months.
 402 L. Andersen et al. Acta Oncologica 39 (2000)

and one was randomized less than 4 months after surgery).

Of the 42 women included in the analysis, 22 were ran-
domized to standard therapy alone, and 20 were random-
ized to standard therapy plus MLD. Ten patients
randomized to standard therapy crossed over to standard
treatment plus MLD after 3 months. One woman random-
ized to standard therapy plus MLD died of a heart attack
before the 12-month evaluation. Two patients randomized
to standard therapy withdrew from the study after the 1-
and 4-month evaluations, respectively, because of breast
cancer recurrence. One woman withdrew from the study
after the 3-month evaluation because of her husband’s
Fig. 1. Trial profile.
terminal disease. One woman did not return for her 12-
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

month evaluation because of depression. This allowed data

scale includes 7 points and the data do not seem to be
to be obtained on 42 patients at 1 month, on 41 patients at
concentrated on a smaller subset, we do not believe that
3 months, on 39 patients at 6 and 9 months and 38
this will have major impact on the results. This avoids the
patients at 12 months.
use of more specialized software to analyze ordinal data
The median age of the patients was 53 years (range:
with repeated measurements. The data have also been
25 – 77 years) and the median edema volume at baseline
examined for a possible time trend by means of a linear
was 346 ml (range: 78 ml – 1297 ml). Significantly more
regression, which, apart from the variance components
patients randomized to standard therapy had received
mentioned above, includes three terms: (a) a baseline
endocrine treatment compared with patients in the stan-
value, (b) a level change to describe the new baseline value
dard therapy plus MLD group, but apart from this, the
after one month of treatment and (c) a possible linear
characteristics of the patients in the two groups were
trend over time.
For personal use only.

similar (Table 2), as were their answers to the baseline

Ethics questionnaire on arm discomfort, tightness, heaviness,
aching, pain, function and loss of shoulder mobility (data
The study was carried out in accordance with the Helsinki-
not shown).
II declaration and was approved by the regional Scientific
Analysis on an intention-to-treat basis showed no sig-
Committee of Aarhus. All patients received verbal and
nificant differences in the estimated reduction in
written information, and gave informed consent before
lymphedema over time between the two randomization
inclusion.
groups (p =0.66) (Fig. 2). The analysis was repeated,
dividing the patients into 3 groups: one group randomized
RESULTS to standard therapy plus MLD right from the start, one
The trial profile is presented in Fig. 1. Forty-four patients group randomized to standard therapy and continuing on
were randomized, but 2 patients, one in each treatment standard therapy and the last group consisting of the
group, were subsequently found not to be eligible (one was patients randomized to standard therapy but with the
found to have lymphedema caused by a local recurrence addition of MLD after 3 months (Fig. 3). There was no

Table 2
Characteristics of patients by treatment

Standard treatment Standard treatment plus MLD p-value

(n =22) (n =20)
Median (range) Median (range)

Age (years) 56 (29–77) 53 (25–73) 0.41

Edema volume (ml) 361 (78–1184) 340 (161–1297) 0.71
Duration of lymphedema (months) 12 (4–126) 15 (5–183) 0.26
Lymph nodes removed 11 (5–21) 15 (3–30) 0.10
Number Number
Endocrine therapy 10 2 0.02
Chemotherapy 6 9 0.19
Radiotherapy (RT) 0.93
No RT 11 8 –
RT without including the axillae 7 10 –
RT including the axillae 4 2 –
 Acta Oncologica 39 (2000) Treatment of lymphedema 403

and 66%, respectively. The lymphedema reduction after

one month was statistically significant compared to the
baseline (p B 0.001). The mean residual lymphedema after
12 months was 166 ml (median residual volume =119 ml;
range: −99 ml – 938 ml).
The percentages for patients reporting improvement or
no change in symptom score after 3 months compared to
baseline according to treatment group are given in Fig. 4.
Only patients reporting the symptom in question at base-
line are included in the description. Analysis of the symp-
tom scores did not suggest any differences between the
groups, and the results for each treatment were conse-
quently pooled. The analysis revealed that, after one
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

month, the patients experienced a significant reduction in

all the symptoms. Over time, there was a further slight
Fig. 2. Treatment outcome according to randomization group. improvement, but this was not significant (data not
The estimated increase in the ratio of the arm volume between the shown).
lymphedema arm and the contralateral arm according to the time There was no difference between groups in the compli-
since treatment start. MDL =manual lymphatic drainage.
ance of the patients concerning use of compression sleeves
evidence of an improved effect of adding MLD to the or performance of arm exercise. The analysis showed that
treatment (p =0.86). the effect of treatment on lymphedema was significantly
The mean percentage reduction in absolute edema vol- related to the use of compression sleeves in both groups
ume after 3 months compared with the baseline absolute (p B0.001). This effect was constant over time.
edema volume was 60% (95% CI: 43%–78%) among pa-
For personal use only.

tients assigned to standard treatment alone versus 48%

DISCUSSION
(95% CI: 32%–65%) among patients assigned to standard
treatment plus MLD. Lymphedema can be a serious and disabling complication
As no evidence of a treatment effect from MLD could of breast cancer treatment. There is no cure for this
be demonstrated, the results were pooled to estimate the condition, and the aim of the treatment is to reduce the
overall extent of treatment effect. To allow direct compari- swelling, increase joint mobility and to decrease discom-
son with treatment results from other studies, these results fort. Management with DLT is currently a popular and
are given on the untransformed data and as the reduction widespread treatment approach, and was recently recom-
in absolute edema volume. The mean percentage reduc- mended by a workgroup of the American Cancer Society
tions in absolute edema volume after 1 and 12 months in Lymphedema Workshop (15). DLT is a combined method
relation to the baseline absolute edema volume were 43% of treatment, and the relative efficacy of each of the
components of this comprehensive treatment program has
not previously been investigated in randomized studies.
One person (LA), an experienced and certified
lymphotherapist according to the Vodder School of prac-
tice, carried out all treatments in the present study. Our
intention was to study only the majority of patients with
minor lymphedema, defined as a difference in arm volume
of less than 30% (absolute difference), and for this reason
we found it justifiable to omit the bandages usually used as
an integrated part of the DLT. This is in accordance with
the approach used by others (3).
MLD is generally believed to be an important part of
DLT, though inadequate as the sole treatment (13, 14).
Our study showed a lack of effect of MLD as a supple-
ment to standard therapy for management of minor
lymphedema. Our findings suggest, that the standard ther-
apy approach without MLD is a sufficient and adequate
Fig. 3. Treatment outcome according to the treatment received.
The estimated increase in the ratio of the arm volume between the
treatment for this patient category. Both groups obtained
lymphedema arm and the contralateral arm according to the time a significant reduction in limb volume, a decrease in
since treatment start. MDL = manual lymphatic drainage. discomfort and an increased joint mobility during treat-
 404 L. Andersen et al. Acta Oncologica 39 (2000)
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

Fig. 4. Change in symptom scores after 3 months. Percentage of patients reporting improvement or no change in symptom score after
3 months compared with baseline according to treatment group. Black columns: improvement in symptoms. White columns: unchanged
symptoms.The number of patients reporting the symptom in question at baseline is given in parentheses and only these patients are
included in the description. ST =standard treatment; MDL =manual lymphatic drainage.
For personal use only.

ment. Less than half of the patients randomized to the 1 680 ml and the mean duration of lymphedema was 7.2
standard therapy chose the addition of MLD, indicating years compared with a mean volume of 350 ml and a mean
that they did not have additional treatment requirement, duration of 1.2 years in our study. Comparison of the
and furthermore the addition of MLD at crossover did not respective outcomes of the treatment in the two studies
further improve edema reduction. After 3 months, 27 out suggests that, to obtain the best results, it is important to
of 41 patients had an edema volume of less than the 200 start compression treatment in the early stage of
ml, usually regarded as a threshold for clinically significant lymphedema.
edema. The outcome of the randomized study by Brorson et al.
In the present study the mean reduction in absolute indicates that standard therapy/controlled compression
lymphedema in the whole treatment group was 43% after therapy might not be sufficient in larger lymphedema. A
one month. This is in accordance with the absolute better treatment for larger lymphedema (greater than 1 000
lymphedema reduction of 47% found in a Swedish study ml) could be liposuction combined with controlled com-
by Brorson et al. using ‘controlled compression therapy’ pression therapy, which in the above study was shown to
(16). Controlled compression therapy is similar to the reduce the lymphedema significantly more effectively than
standard therapy used in the present study, with the compression therapy alone, and resulted in a reduction in
exception that our patients wore compression garments absolute edema volume of more than 100% (16).
during daytime only, while patients in the Swedish study The effect of DLT has to our knowledge been assessed
used compression garments both day and night. Even in non-randomized studies only. In a study by Boris et al.
though the two studies yielded approximately the same comprising 16 patients with a mean lymphedema before
results in the percentage of reduction of lymphedema, treatment of 690 ml, the edema reduction was 73% (9) and
there are considerable differences in the treatment results in a study from the Adelaide Lymphedema Clinic an
when comparing the mean volumes of edema remaining average reduction of 64% after one month was described
after treatment. In our patient group the mean for the first 78 consecutive arm lymphedemas treated in
lymphedema left after 12 months was 166 ml (range: − 99 the clinic (17). Long-term results of DLT have been pub-
ml –938 ml) compared with 873 ml (range: 340 ml– 2 275 lished by Boris et al. (18), who reported an initial arm
ml) in the Swedish study. This difference in treatment lymphedema reduction after DLT averaging 62.6% in 56
outcome is most likely due to differences in patient charac- consecutive patients. After 36 months’ follow-up, this re-
teristics rather than in the treatment given. The patients in duction was maintained.
the Swedish study had a much larger and more severe It is possible that DLT including MLD has a place in
lymphedema. The mean volume of edema at baseline was the treatment of larger lymphedema, but randomized stud-
 Acta Oncologica 39 (2000) Treatment of lymphedema 405

ies evaluating the efficacy of the different components of 2. Willich N, Zeschick A, Zeschick L, Wendt TH. Ergebnisse
this combined treatment should be performed in this pa- der bestrahlung nach mastectomie. In: Sauer H. Mam-
makarzinom Hrsg. München: Zuckschwerdt. 1989: 56 – 63.
tient category. Further studies should include objective
3. Mortimer P, Badger C, Hall JG. Lymphoedema. In: Doyle
measurements of both arms as well as some kind of D, Hanks G, MacDonald N, eds. Oxford textbook of pal-
subjective assessment. The patients should be followed-up liative medicine. Oxford: Oxford University Press, 1993:
for at least one year, to allow evaluation of long-term 407 – 15.
treatment outcome. 4. Tobin MB, Lacey HJ, Meyer L, Mortimer PS. The psycho-
logical morbidity of breast cancer related arm swelling. Psy-
As shown in our study, the results of the treatment
chological morbidity of lymphoedema. Cancer 1993; 72:
depend on the compliance of the patients, assessed by their 3248 – 52.
use of the compression garment. This is in accordance with 5. Petrek JA, Heelan MC. Incidence of breast carcinoma-re-
the experience of others (10, 13, 16). To achieve the lated lymphedema. Cancer 1998; 83 (Suppl 12 American):
maximal compliance of the patient, two things are very 1776 – 81.
important: the compression sleeve has to fit and the pa- 6. Swedborg I, Wallgree A. The effect of pre- and postmastec-
Acta Oncol Downloaded from informahealthcare.com by University of California San Francisco on 05/01/15

tomy radiotherapy on the degree of edema, shoulder-joint

tient must understand why and how to use it. For the first
mobility, and gripping force. Cancer 1981; 47: 877 – 81.
couple of treatment weeks our patients used decreasing 7. Kissin MW, Querci Della Rovere G, Easton D, Westbury
sizes of Jobst compression garments to reduce the edema. G. Risk of lymphoedema following the treatment of breast
Then measurements were taken for a custom-made com- cancer. Br J Surg 1986; 73: 580 – 4.
pression garment. In general, the garments were replaced 8. Földi E. Treatment of lymphedema and patient rehabilita-
tion. Anticancer Res 1998; 18: 2211 – 2.
every 2–6 months to maintain the proper amount of
9. Boris M, Weindorf S, Lasinski B, Boris G. Lymphedema
compression. reduction by noninvasive complex lymphedema therapy. On-
Two to three hours were reserved for each patient at the cology (Huntingt) 1994; 8: 95 – 106.
first appointment and one hour was reserved for each 10. Casley-Smidt JR, Boris M, Weindorf S, Lasinski B. Treat-
follow-up visit. The total time spent on each patient ran- ment for lymphedema of the arm — the Casley-Smidt
domized to the standard therapy was thus eight hours. For method. A noninvasive method produces continued reduc-
tion. Cancer 1998; 83 (Suppl 12 American): 2843 – 60.
For personal use only.

patients randomized to MLD, a further one hour was

11. Dawn C, Rose K. Treatment leads to significant improve-
reserved 8 times for MLD and training in self-massage. ment. Effect of conservative treatment on pain in
The time taken up by the standard therapy was thus only lymphoedema. Prof Nurse 1992; 10: 32 – 5.
half the time used with the standard treatment plus MLD. 12. Mislin H. In: Altmann HW. Handbuch der allgemeinen
As a consequence of the less comprehensive treatment, it Pathologie. Berlin, Heidelberg, NewYork: Springer Verlag,
might be easier for patients to continue their normal 1972: 3. band 6. teil: 222.
13. Földi E. The treatment of lymphedema. Cancer 1998; 83
working and social lives while being treated.
(Suppl 12 American): 2833 – 4.
Our conclusion from the present study is that standard 14. Leduc O, Leduc A, Bourgeois P, Belgrado JP. The physical
therapy is an effective and simple way of treating minor treatment of upper limb edema. Cancer 1998; 83 (Suppl 12
lymphedema. MLD does not—at least in the early American): 2835 – 9.
stage—improve the treatment outcome. 15. Rockson SG, Miller LT, Senie R, et al. American Cancer
Society Lymphedema Workshop. Workgroup III: Diagnosis
and management of lymphedema. Cancer 1998; 83 (Suppl
ACKNOWLEDGEMENTS 12 American): 2882 – 5.
Financial support for this study was provided by the Research 16. Brorson H, Svensson H. Liposuction combined with con-
Initiative, the Research Foundation for the Radium Station, the trolled compression therapy reduces arm lymphedema more
Clinical Research Unit at the Oncology Center, all from Aarhus effectively than controlled compression therapy alone. Plast
University Hospital, and by grants in memory of nurse Ingrid Reconstr Surg 1998; 102: 1058 – 67.
Lorentsen and in memory of Holger and Inez Petersen. 17. Morgan RG, Casley-Smith JR, Mason MR, Casley-Smith
JR. Complex physical therapy for the lymphoedematous
arm. [Br] J Hand Surg 1992; 17: 437 – 41.
REFERENCES
18. Boris M, Weindorf S, Lasinkski S. Persistence of
1. Brennan MJ. Lymphedema following the surgical treatment lymphedema reduction after noninvasive complex
of breast cancer: a review of pathophysiology and treat- lymphedema therapy. Oncology (Huntingt) 1997; 11: 99–
ment. J Pain Symptom Manage 1992; 7: 110–6. 109.