Support Care Cancer
DOI 10.1007/s00520-010-0906-x
ORIGINAL ARTICLE
Factors influencing breast cancer-related lymphedema
volume after intensive decongestive physiotherapy
Stéphane Vignes & Raphaël Porcher & Maria Arrault &
Alain Dupuy
Received: 18 December 2009 / Accepted: 5 May 2010
# Springer-Verlag 2010
Abstract
Background Lymphedema treatment is based on an inten-
sive decongestive physiotherapy phase of volume reduction
followed by a long-term maintenance phase. Factors
influencing the morbid lymphedema volume increase
during maintenance were analyzed.
Materials and methods Among 867 consecutive women
recruited and followed in a single lymphology unit, 682
were analyzed. The other 185 were not analyzed because of
an initial lymphedema volume <250 ml, <20% lymphe-
dema volume decrease during the intensive phase, or they
were lost to follow-up. Lymphedema volume was recorded
prior to and at the end of intensive phase, and at each
follow-up visit. During follow-up, treatment failure was
defined as a lymphedema volume increase of ≥50% of the
total reduction obtained during the intensive phase.
Results Median lymphedema volume was 936 ml before
and 335 ml after intensive decongestive physiotherapy (P<
0.0001). Median follow-up was 28 months. During the
maintenance phase, the risk of treatment failure at 1, 2, and
S. Vignes (*) : M. Arrault
Department of Lymphology, Hôpital Cognacq-Jay,
15, rue Eugène Millon,
75015 Paris, France
e-mail: stephane.vignes@hopital-cognacq-jay.fr
R. Porcher
Department of Biostatistics and Medical Informatics,
Hôpital Saint-Louis AP–HP, Université Paris 7,
INSERM U717,
Paris, France
A. Dupuy
Department of Dermatology 2, Hôpital Saint-Louis,
AP–HP, 1, Avenue Claude Vellefaux,
75475 Paris, Cedex 10, France
4 years was estimated to be 38.1%, 53.1%, and 64.8%,
respectively. Wearing an elastic sleeve during the day and
an overnight multilayer low-stretch bandage (median, four
nights per week; interquartile range, 2–6) significantly
decreased the risk of treatment failure [hazard ratio, 0.53,
(0.34–0.82), P=0.004], whereas manual lymph drainage
adjunction to those therapeutic components did not. The
risk of treatment failure was also associated with weight
and body mass index at inclusion.
Conclusion Risk of maintenance-therapy failure after in-
tensive decongestive physiotherapy was associated with
patients characteristics (younger age, higher weight, and
body mass index), while elastic sleeve and bandage were
associated with better maintenance results. Paradoxical
effect of manual lymph drainage is likely to be related to
indication bias.
Keyword Breast cancer . Compliance . Elastic garment .
Low-stretch bandage . Lymphedema . Physiotherapy
Introduction
Lymphedema is the accumulation of lymph that results
from a defective lymphatic transport capacity. It remains a
major problem for women treated for breast cancer,
affecting 12% to 28% after axillary dissection and 3.5%
to 7.5% after sentinel node biopsy [1–3]. This chronic and
debilitating dysfunction can be responsible for severe
physical, psychological, and social morbidity. Limb swell-
ing causes the disproportionate increase in the size of a part
of the body and can also affect patients’ self-perception and
alter the quality of life [4]. The major aim of lymphedema
treatment is to reduce the volume at long-term and
hopefully, by doing so, avoid or decrease the risk of

infection. Current management of secondary arm lymphe-
dema is based on the widely advocated regimen called
complete (or complex) decongestive physiotherapy, includ-
ing multilayer low-stretch bandaging, manual lymph drain-
age, exercises, and skin care [5, 6]. Lymphedema therapy is
usually divided into two successive phases [7–9]. The first
consists of intensive treatment and may be implemented in
the hospital to achieve substantial lymphedema volume
reduction, whereas the second phase consists of long-term
maintenance therapy at home to stabilize lymphedema
volume [7]. The modalities of the latter phase are much
less precise than those of the intensive phase. Our objective
was to evaluate lymphedema volume during the mainte-
nance phase and to determine the factors that influence it.
Materials and methods
Patients
Patients eligible for inclusion in this prospective cohort
study were all consecutive women referred, between
January 2001 and June 2008, to our center dedicated to
lymphedema management, for treatment of unilateral arm
lymphedema after breast cancer treatment. Lymphedema
therapy was managed as follows: first, a 2 weeks hospital-
ization period to implement the intensive phase; second, at
discharge, the maintenance phase was pursued by the
patient and family at home and rarely by a physiotherapist.
The first day of hospitalization for intensive therapy was
the day of inclusion in the cohort. Patients with no follow-
up were secondarily excluded from the analysis.
Exclusion criteria were as follows: patients with initial
lymphedema volume <250 ml (n=10) were not eligible
because variations around a small volume are of uncertain
meaning; and because we aimed at maintaining a gain
obtained after the intensive phase, those with insufficient
volume reduction at the end of the intensive phase
compared to baseline (<20% volume reduction) were
excluded (n=107). The population not included in the
analysis was not different from the population analyzed
regarding age, BMI, breast cancer characteristics, and
lymphedema characteristics.
Complete decongestive physiotherapy
All patients underwent complete decongestive physiother-
apy, consisting of a combination of manual lymph drainage
according to Földi's techniques, multilayer low-stretch
bandages, specific exercises, and skin care, as previously
recommended by the international consensus guidelines [5,
6]. This first intensive phase aims to obtain a 40–60%
lymphedema volume reduction [8, 9]. Manuel lymph
Support Care Cancer
drainage was done by a physiotherapist specialized in
lymphatic techniques, who also taught the patient the
multilayer bandaging technique and specific exercises.
Manual lymph drainage lasts 30 min. It begins by
manual stimulation of the lymph nodes in adjacent drainage
regions (neck, subclavicular, contralateral axilla, and back)
and is followed by decongestion of the involved trunk,
shoulder, arm, forearm, and, if necessary, hand. Then, a
low-stretch compressive bandage (Somos®, BSN Medical,
Vibraye, France) was wrapped in multiple (two to four)
layers after covering the affected limb with foam (N/N®,
Thuasne, Levallois-Perret, France) or cotton batting
(Cellona®, Lohmann, Rengsdorf, Germany). Bandages
were worn 24 h/day throughout the intensive phase and
reapplied every day, 5 days/week (the last bandage of the
week was retained until Monday). After compressive
bandages had been applied, soft and moderate remedial
exercises (shoulder rotations and stretches, neck stretches,
elbow and wrist circles, opening/closing fingers and finger
dexterity, posture corrections, and breathing exercises) were
performed with the bandage in place to enhance lymphatic
flow from peripheral to central compartments.
Each patient, and sometimes family members, was
taught the self-bandaging technique during the intensive
phase under the supervision of the physiotherapist. Several
learning sessions were proposed for each patient to improve
and verify good understanding and practice. At least three
overnight bandages per week were recommended during
the long-term maintenance phase. During this phase,
follow-up visits were scheduled at 6 and 12 months, and
then annually.
Complete decongestive physiotherapy also included
meticulous skin care. Dry skin was systematically treated
with moisturizer (cold cream) before applying the bandage.
Patients were instructed to avoid cutaneous abrasions (e.g.,
cuts, burns, insect bites, cat scratch, and cracks in dry skins)
and to protect their skin during daily activities (e.g., using
gloves for gardening, thimble when sewing). Overweight
and obese patients had a specific consultation with a
dietician or nutritionist to incite them to lose weight
because weight control is a major component of lymphe-
dema management [10].
Data collected
The following information was recorded: characteristics of
breast cancer stage and treatment [age at cancer diagnosis,
type of surgery (mastectomy or tumorectomy), radiother-
apy, chemotherapy, antiestrogen drugs, complications
(radiation-induced brachial plexopathy), past cellulitis
(erysipelas), patient characteristics, body mass index
[BMI; calculated as weight (kg)/height (m2)], date of
lymphedema onset and duration, and lymphedema volume
Support Care Cancer
at inclusion]. Lymphedema volume was calculated for
each 5-cm segment using the formula for a truncated cone:
H
ðC 2 þ Cc þ c2 Þ=12p, H = height, C = circumference
of the top of the cone, c = circumference of the base of the
cone [11]. This method demonstrated excellent inter- and
intra-observer reproducibility when compared to water
displacement, which remains the gold standard [12].
Lymphedema volume was defined as the difference
between the lymphedematous limb (VL) and the healthy
limb volume (VH). Arm volumes were measured at
inclusion, at the end of intensive decongestive physiother-
apy and at each of the follow-up visits. At discharge, a
custom-made elastic sleeve (Sirlex Radiante®, Châtelle-
rault, France) class 2 (15–19 mmHg) was fitted for 9% of
the patients, class 3 (20–36 mmHg) for 90%, and class 4
(>36 mmHg) for 1%. The elastic sleeve was to be worn
everyday from morning to night and it was recommended
that it be replaced every 3 or 4 months.
Follow-up
At each follow-up visit, the woman was questioned about
the actual treatment she applied during the time elapsed
since the previous visit and her responses were recorded,
i.e., whether she actually used manual lymph drainage, self-
applied bandages (number of times per week), and wore the
elastic sleeve.
Treatment failure during follow-up was defined as the
lymphedema volume increase ≥50% of the total reduction
obtained during the first intensive decongestive physiother-
apy. Follow-up duration was stopped when the woman was
again hospitalized for a second intensive phase but the
patient was included in this analysis.
Statistical analysis
Data are presented as numbers (%) for categorical variables
and median, interquartile range (first quartile, third quartile
(Q1, Q3)) or mean (standard deviation, SD) for quantitative
variables, unless otherwise stated.
The probability of failure over time was estimated in
a competing-risks framework by the usual estimator of
the cumulative incidence function, treating a new
hospitalization without failure as a competing event.
The association of baseline characteristics with the risk
of failure was analyzed using proportional cause-specific
hazard models. Also, the different components of
treatment during follow-up were analyzed as time-
dependent covariates in the proportional cause-specific
hazard models. All tests were two-sided, and P values<
0.05 were considered significant. Analyses were per-
formed using R 2.6.2 statistical software (R Development
Core Team, Vienna, Austria, 2008).
Results
Among 867 women screened for our study, 682 were analyzed
(Fig. 1). The main clinical characteristics of the patients, their
breast cancer treatment, and arm lymphedema are reported in
Table 1. Before intensive decongestive physiotherapy, median
lymphedema volume was 936 ml (Q1, Q3=660, 1,284) and it
declined to 335 ml (Q1, Q3 = 216.8, 507; P<0.0001) after it.
Follow-up
During follow-up of the maintenance phase, the median
number of visits was 4 (range, 1–17) and 32 (4.7%) patients
died. Median duration of follow-up was 28 (range, 4–101)
months. During follow-up, the risk of treatment failure at 1,
2, and 4 years was estimated to be 38.1%, 53.1%, and
64.8%, respectively (Fig. 2). As shown in Table 2, wearing
the elastic sleeve during the day and multilayer low-stretch
bandage overnight (median, four nights per week; Q1, Q3 =
2, 6) significantly decreased the risk of treatment failure,
while the adjunction of manual lymph drainage (one to
three times per week) to the regimen did not; wearing the
elastic sleeve alone during the day only tended to decrease
the risk of treatment failure.
Clinical risk factors positively associated with treatment
failure were the patient's weight and BMI ≥30 at inclusion,
while age at inclusion was negatively associated (Table 3).
The interval from the start of cancer treatment to lymphe-
dema onset, the duration of lymphedema, type of surgery,
and previous cancer treatment were not associated with the
risk of treatment failure.
Discussion
This is the first study to analyze the long-term evolution of
lymphedema volume after breast cancer treatment. In this
867 patients
screened
117 patients not eligible,
Intensive phase
10 with lymphedema volume <250 ml
107 with volume reduction <20% at the
end of intensive phase
750 patients
included
Maintenance phase
68 patients lost-to-
follow-up
682 patients
analyzed
Fig. 1 Flow chart of the patients included in the cohort
 Support Care Cancer

Table 1 Patient characteristics at inclusion, before intensive decon- Table 2 Lymphedema maintenance-therapy components: their asso-
gestive physiotherapy ciation with outcome (risk of maintenance-therapy failure after
intensive decongestive physiotherapy)
Characteristic Value n=682
Treatment Hazard ratio P value
Age, median (range), years 62 (55–71) (95% CI)
Body mass index, median (Q1; Q3), kg/m2 27.3 (24.5; 31.2)
None 1 Reference
Breast cancer characteristics
category
Age at breast cancer, median (Q1; Q3), years 52 (45; 60)
MLD alone 1.91 (1.07–3.42) 0.03
Type of surgery
Elastic sleeve alone 0.65 (0.40–1.08) 0.1
Mastectomy, n (%) 326 (47.8)
MLD + elastic sleeve 1.09 (0.68–1.73) 0.73
Tumorectomy, n (%) 341 (50)
Bandage + elastic sleeve 0.53 (0.34–0.82) 0.004
None, n (%) 15 (2.2)
MLD + bandage + elastic sleeve 0.73 (0.47–1.11) 0.14
Radiotherapy, n (%) 664 (97.4)
Chemotherapy, n (%) 450 (66) MLD manual lymph drainage
Antiestrogen drugs, n (%) 354 (51.9)
Patients with metastatic cancer, n (%) 117 (17.2)
Mammilary reconstruction, n (%) 76 (11.1) ≥50% of the total reduction obtained with the first intensive
Lymphedema decongestive physiotherapy, was selected to estimate a
Time to its onset, median (Q1; Q3), months 22 (6; 69) prolonged benefit. With this end-point criterion, the risk of
Its duration, median (Q1; Q3), months 32 (13; 97) failure was estimated to 38.1%, 53.1%, and 64.8% after 1,
Past erysipelas, n (%) 229 (33.6) 2, and 4 years, respectively.
Associated with radiation-induced brachial 44 (6.5) In light of the high percentage of breast cancer patients
plexopathy, n (%) affected, a wide variety of potential risk factors of treatment
failure were investigated during follow-up, including the
patients' characteristics and the different treatment compo-
large, prospective cohort, we observed a gradual lymphe- nents: low-stretch multilayer bandages, manual lymph
dema volume increase during follow-up after the intensive drainage, and elastic garment. In agreement with previous
physiotherapy phase. The volume increased mainly findings, lymphedema volume during the maintenance
throughout the first year. The stringent definition of phase in our patients mainly reflected treatment compliance
treatment failure, i.e., a lymphedema volume increase of [13]. We previously showed that 1 year after the intensive
phase, lymphedema volume was stable in 20% of patients,
increased in 52%, and continued to decrease in 28% by
≥10% as compared with lymphedema volume at the end of
intensive decongestive physiotherapy [14]. Compression
therapy, combining the wearing of an elastic sleeve during
the day and a bandage overnight, represented the main
means to stabilize the lymphedema volume during the
maintenance phase.
A surprising finding in our results is the association
between MLD and failure for maintenance of lymphedema
volume. Patients using MLD alone appear to have an
increased risk of failure. Moreover, adding MLD to any

Table 3 Clinical risk factors associated with lymphedema mainte-

nance therapy (risk of maintenance-therapy failure after intensive
decongestive physiotherapy)

Risk factor Hazard ratio P value

(95% CI)

Fig. 2 Risk of maintenance-therapy failure after intensive deconges- Age at inclusion (per 10 years) 0.85 (0.77–0.94) 0.001
tive physiotherapy for arm lymphedema after breast cancer treatment. Weight at inclusion (per 10 kg) 1.09 (1.01–1.16) 0.018
Month 0 represents the start of the in-hospital intensive decongestive Body mass index at inclusion ≥30 1.29 (1.04–1.60) 0.023
therapy
Support Care Cancer
other treatment (i.e., elastic sleeve and elastic sleeve
associated to bandage) tend to increase the risk of failure.
In a recent randomized controlled trial, an MLD-based
early physiotherapy program was proved effective reducing
the risk of secondary lymphedema after surgery for breast
cancer [15]. The hypothesis of a direct deleterious effect of
MLD in our population is therefore not a reasonable one.
The reasons underlying this puzzling finding are unclear.
They may be related to unknown confounders, among
which confounding by indication would need to be further
assessed in subsequent studies. We could not directly
compare the baseline characteristics and lymphedema
volume in patients with vs without MLD since each patient
accounted for different treatments at different periods
during follow-up. Adding MLD to a previous treatment
regimen could have been a marker of unsatisfying results
associated with this regimen, with an increasing volume,
while not yet reaching the arbitrary threshold of 50%
increase defining failure in our study. On the other hand, in
order to alleviate the burden related to a multicomponent
successful regimen, MLD would have been the first
component withdrawn. Such attitudes would both contrib-
ute to the paradoxical effect of MLD in our study. Only a
randomized controlled design would be free of bias
regarding confounding by indication.
Older age at study inclusion was a clinical factor
associated with decreased risk of treatment failure. Unfor-
tunately, we do not have precise data to verify whether this
effect could be due to better treatment adherence by older
women. The second major intrinsic clinical parameter
influencing the risk of failure treatment was the patient’s
weight and a BMI ≥30 at inclusion. Overweight or high
BMI at cancer onset or weight gain after cancer treatment
heightens the risk of developing lymphedema and its
severity [16–18]. Weight control combined with compres-
sion therapy is required to treat women with arm
lymphedema [10].
The main strengths of our study are that it was
conducted in a single specialized lymphedema center and
it was prospective, including only women with lymphe-
dema after breast cancer treatment. Intensive physiotherapy,
follow-up visits, and lymphedema volume measurements
were ensured by trained physiotherapists and practitioners.
However, our study has some limitations. First, because the
treatments were not randomly allocated, we cannot exclude
that their effects were actually due to unidentified con-
founding factors. Second, assessment of compliance with
the different maintenance-therapy components was based
on patient declarations.
We concluded that wearing an elastic sleeve and
overnight bandage are the major efficacy components of
maintenance therapy for lymphedema after intensive de-
congestive physiotherapy. The apparent negative impact of
MLD is likely due to indication bias. Among patient
characteristics, older age, lower weight, and lower BMI
were associated with a better maintenance of the good
results obtained after intensive decongestive physiotherapy.
Prolonged stabilization of lymphedema volume might
participate in improving the quality of life of these women
and decrease the risk of complications, especially infectious
ones.
Conflict of interest None.
References
1. Lucci A, McCall LM, Beitsch PD et al (2007) Surgical
complications associated with sentinel lymph node dissection
(SLND) plus axillary lymph node dissection compared with
SLND alone in the American College of Surgeons Oncology
Group Trial Z0011. J Clin Oncol 25:3657–3663
2. Langer I, Guller U, Berclaz G et al (2007) Morbidity of sentinel
lymph node biopsy (SLN) alone versus SLN and completion
axillary lymph node dissection after breast cancer surgery: a
prospective Swiss multicenter study on 659 patients. Ann Surg
245:452–461
3. McLaughlin SA, Wright MJ, Morris KT et al (2008) Prevalence of
lymphedema in women with breast cancer 5 years after sentinel
lymph node biopsy or axillary dissection: objective measure-
ments. J Clin Oncol 26:5213–5219
4. Heiney SP, McWayne J, Cunningham JE et al (2007) Quality of
life and lymphedema following breast cancer. Lymphology
40:177–184
5. Harris SR, Hugi MR, Olivotto IA, Levine M, Steering
Committee for Clinical Practice Guidelines for the Care and
Treatment of Breast Cancer (2001) Clinical practice guidelines
for the care and treatment of breast cancer: 11. Lymphedema.
CMAJ 164:191–199
6. Lymphoedema Framework (2006) Best practice for the manage-
ment of lymphoedema. International consensus. MEP Ltd,
London
7. Cheville AL, McGarvey CL, Petrek JA et al (2003) Lymphedema
management. Semin Radiat Oncol 13:290–301
8. Szuba A, Cooke JP, Yousuf S, Rockson SG (2000) Decongestive
lymphatic therapy for patients with cancer-related or primary
lymphedema. Am J Med 109:296–300
9. Ko DS, Lerner R, Klose G, Cosimi AB (1998) Effective treatment
of lymphedema of the extremities. Arch Surg 133:452–458
10. Shaw C, Mortimer P, Judd PA (2007) A randomized controlled
trial of weight reduction as a treatment for breast cancer-related
lymphedema. Cancer 110:1868–1874
11. Sitzia J (1995) Volume measurement in lymphoedema treatment:
examination of formulae. Eur J Cancer Care 4:11–16
12. Galland C, Auvert JF, Flahault A, Vayssairat M (2002) Why and
how post-mastectomy edema should be quantified in patients with
breast cancer. Breast Cancer Res Treat 75:87–89
13. Boris M, Weindorf S, Lasinski B (1997) Persistence of lymphe-
dema reduction after noninvasive complex lymphedema therapy.
Oncology (Williston Park) 11:99–109, discussion 110, 113–114
14. Vignes S, Porcher R, Arrault M, Dupuy A (2007) Long-term
management of breast cancer-related lymphedema after intensive
decongestive physiotherapy. Breast Cancer Res Treat 101:285–29

15. Torres Lacomba M, Yuste Sánchez MJ, Zapico Goñi A et al
(2010) Effectiveness of early physiotherapy to prevent lymphoe-
dema after surgery for breast cancer: randomised, single blinded,
clinical trial. BMJ 340:b5396. doi:10.1136/bmj.b5396
16. Kocak Z, Overgaard J (2000) Risk factors of arm lymphedema in
breast cancer patients. Acta Oncol 39:389–392
Support Care Cancer
17. Petrek JA, Senie RT, Peters M, Rosen PP (2001) Lymphedema in
a cohort of breast carcinoma survivors 20 years after diagnosis.
Cancer 92:1368–1377
18. Vignes S, Arrault M, Dupuy A (2007) Factors associated with
increased breast cancer-related lymphedema volume. Acta Oncol
46:1138–1142