Institute of Cancer Research Repository
Please direct all emails to:
This is an author produced version of an article that appears in:
The internet address for this paper is:
RADIOTHERAPY AND ONCOLOGY
L Gothard, A Stanton, J MacLaren, D Lawrence, E Hall, P
Mortimer, E Parkin, J Pritchard, J Risdall, R Sawyer, M
Woods, J Yarnold (2004) Non-randomised phase II trial of
hyperbaric oxygen therapy in patients with chronic arm
lymphoedema and tissue fibrosis after radiotherapy for early
breast cancer, Radiotherapy and Oncology, Vol. 70(3), 217-
Non-randomised Phase II Trial of Hyperbaric Oxygen Therapy in Patients
with Chronic Arm Lymphoedema and Tissue Fibrosis
after Radiotherapy for Early Breast Cancer
Lone Gothard , Anthony Stantonbc
, Julie MacLaren , David Lawrenced,
Emma Hall , Peter Mortimerc, Eileen Parkinee
, Joyce Pritchard ,
Jane Risdall , Robert Sawyer , Mary Woodsca,*
, John Yarnold
aDepartment of Radiotherapy, Royal Marsden Hospital NHS Trust, Sutton, Surrey, UK
bDivision of Physiological Medicine (Dermatology), St George's Hospital Medical School, London, UK
cLymphoedema Services, Royal Marsden Hospital NHS Trust, Sutton, Surrey, UK
dClinical Trials and Statistics Unit, Institute of Cancer Research, Sutton, Surrey, UK
eRadiotherapy Action Group Exposure, 24 Edgeborough Way, Bromley, Kent, UK
fInstitute of Naval Medicine, Alverstoke, Gosport, Hampshire, UK
* Corresponding author:
Professor John Yarnold, Consultant Clinical Oncologist
Academic Radiotherapy Unit, The Royal Marsden Hospital
Downs Road, Sutton, Surrey SM2 5PT, UK
Tel: 0044 20 8661 3388 / Fax: 0044 20 8661 3107 / Email: email@example.com
Total number of pages: 18, Tables: 2, Figures: 1
Running head: Hyperbaric oxygen therapy for chronic arm lymphoedema and tissue fibrosis
Key words: hyperbaric oxygen therapy, clinical trials, radiotherapy complications, lymphoedema
Radiation-induced arm lymphoedema is a common and distressing complication of curative treatment
for early breast cancer. Hyperbaric oxygen (HBO2) therapy promotes healing in bone rendered
ischemic by radiotherapy, and may help some soft-tissue injuries too, but is untested in arm
Twenty-one eligible research volunteers with a minimum 30% increase in arm volume in the years
after axillary/supraclavicular radiotherapy (axillary surgery in 18/21 cases) were treated with HBO2.
The volunteers breathed 100% oxygen at 2.4 ATA for 100 minutes in a multiplace hyperbaric chamber
on 30 occasions over a period of 6 weeks. The volume of the ipsilateral limb, measured opto-
electronically by a perometer and expressed as a percentage of contralateral limb volume, was selected
as the primary endpoint. A secondary endpoint was local lymph drainage expressed as fractional
removal rate of radioisotopic tracer, measured using lymphoscintigraphy.
Three out of 19 evaluable patients experienced >20% reduction in arm volume at 12 months. Six out
of 13 evaluable patients experienced a >25% improvement in 99Tc-nanocolloid clearance rate from the
ipsilateral forearm measured by quantitative lymphoscintigraphy at 12 months. Overall, there was a
statistically significant, but clinically modest, reduction in ipsilateral arm volume at 12 months follow
up compared with baseline (p = 0.005). The mean percentage reduction in arm volume from baseline
at 12 months was 7.51. Moderate or marked lessening of induration in the irradiated breast, pectoral
fold &/or supraclavicular fossa was recorded clinically in 8/15 evaluable patients. Twelve out of 19
evaluable patients volunteered that their arms felt softer, and six reported improvements in shoulder
mobility at 12 months. No significant improvements were noted in patient self-assessments of quality
Interpretation is limited by the absence of a control group. However, measurement of limb volume by
perometry is reportedly reliable, and lymphoscintigraphy is assumed to be operator-independent.
Taking all data into account, there is sufficient evidence to justify a double-blind randomised
controlled trial of hyperbaric oxygen in this group of patients.
Clinical syndromes associated with radiotherapy complications are variable, but all are progressive and
irreversible . The commonest indication for high dose radiotherapy in the UK is the management
of women with early breast cancer. A proportion of this population develop arm lymphoedema, painful
hardening and shrinkage of the breast, rib fracture, lung damage, cardiac injury and, rarely, nerve
damage . One of the commonest and most distressing of these consequences is lymphoedema of the
upper limb when the axilla has been irradiated, especially following axillary surgery [6,26]. In severe
cases, the volume of the limb is increased by 40% or more, causing disability by virtue of added
weight and impeded joint movement . In a population-based retrospective study in the South of
England, 28% of 1,077 women remaining disease-free after treatment of breast cancer reported some
degree of arm swelling . No reliable data exist on NHS workload related to this group of patients,
but the experience of patient advocate groups suggests that the need for rehabilitation services far
exceed the resources available . Conventional treatment consists of skin-care, exercises to maintain
and promote shoulder movement, massage (either self- or professionally-applied) and compression
garments to improve lymph drainage and soften the tissues. In many cases, the condition can be
managed with appropriate advice and compression sleeves for life . In a minority, compression
bandaging and other measures fail to prevent progression to a grossly swollen limb. In response to the
need for integrated management by a range of health professionals with appropriate expertise, clinical
oncologists, breast care nurses, pain specialists, physiotherapists, occupational therapists and
complementary therapists have been identified in each NHS Region .
Data from our own study testing HBO2 in patients with radiation-induced brachial plexopathy (RIBP)
 showed no reliable evidence to support the hypothesis that HBO2 therapy slows or reverses RIBP
in a substantial proportion of affected individuals, although improvements in warm sensory threshold
offered some suggestion of therapeutic effect. However, 2/6 cases with marked chronic arm
lymphoedema reported major and persistent improvements in arm volume for at least 12 months after
treatment with HBO2. This observation was not anticipated, and forms the basis of the current
hypothesis under test, namely that hyperbaric oxygen therapy improves long-standing arm
lymphoedema in a proportion of patients. The study was performed in collaboration with the patient
lobby group Radiotherapy Action Group Exposure (R.A.G.E.) and was funded by the UK Medical
Research Council. The protocol was approved by the Research Ethics Committees of the Royal
Marsden NHS Trust and the Institute of Naval Medicine.
Patients and Methods
Eligibility and pretreatment assessment
Inclusion criteria included ipsilateral arm lymphoedema following treatment for breast cancer causing
≥30% increase in arm volume, freedom from cancer recurrence, physical and psychological fitness for
HBO2, availability for follow up and written informed consent. Pretreatment baseline assessments
included magnetic resonance imaging (MRI) of the supraclavicular fossa, axilla and brachial plexus to
exclude cancer recurrence, clinical assessment of subcutaneous induration within the radiotherapy
volume, measurement of arm volume using a perometer, quantitative lymphoscintigraphy to assess
local lymph drainage in the forearm, clinical photographs and patient self-assessments using the
EORTC Quality of Life Questionnaires QLQ-C30 and BR23 .
Clinical examination was undertaken by an oncologist to confirm trial eligibility and grade induration
of the breast boost site, pectoral fold and supraclavicular fossa. The latter was done by palpation using
a scale 0-3 (0 = none, 1 = a little, 2 = quite a lot, 3 = very much) and was repeated at follow up
assessments by the same clinician to ensure consistency. Response was defined as an improvement of
at least 2 grades (e.g. 3→1, 3→0, 2→0) at 12 months, since this was regarded as a clinically
worthwhile improvement that could not easily be attributed to measurement error. Clinical
photographs were taken of the upper body (hands on hips and above head) and of both arms, but
response (change in photographic appearance at 12 months) was not predefined.
Measurement of arm volume
Arm volumes were measured in an operator-independent manner using a perometer (Model 400T,
Pero-System GmbH, Wuppertal, Germany [5,24]). The volume from wrist to axilla was determined by
placing the arm vertically inside a square measuring frame containing rows of infrared light-emitting
diodes on two adjacent sides. On moving the frame along the arm, volume was calculated from pairs
of diameter measurements every 3 mm, assuming a circular or elliptical cross-section. The volume of
the ipsilateral limb was expressed as a percentage of the contralateral (control) limb volume. Response
was defined in the protocol as a >20% reduction in ipsilateral arm volume on the grounds that this was
considered a clinically worthwhile improvement unlikely to be due to supportive measures or changes
Lymphoscintigrapy was performed on as many volunteers as possible. Patients were acclimatised to
their surroundings for at least 45 minutes before starting. The patient sat with both arms resting on a
table with palms facing up and the gamma camera (Starcam GE 400ACT, General Electric Medical
Systems, Milwaukee, WI, USA), wide-field-of-view, low energy, high resolution collimator)
positioned above the arms for ventral viewing. A site on the ventral surface of the ipsilateral forearm
particularly affected by swelling was selected for injection (typically one-third of the distance from the
antecubital crease to the wrist). The site was carefully marked and the distance to the outstretched
fingertip and to the olecranon process was measured in cm to enable relocation to the same position
for the repeat scan at 12 months. The corresponding site on the contralateral arm was also marked. 0.2
ml of 99mTc-Nanocoll in saline, containing approximately 35 MBq, was injected subcutaneously in
each forearm using a 1-ml syringe and a 23-gauge needle. The site was not massaged. The camera was
immediately lowered to 2 cm above the injection site and a 60-second static acquisition was
performed. Acquisitions were repeated at 30 min, 60 min, 90 min, 120 min and 180 min. In between
acquisitions the patient sat in the waiting room and read. No exercise was performed. The number of
counts recorded within a rectangular region of interest (area 54 cm2), encompassing the depot, was
obtained from the computer (GE acquisition computer). The removal rate constant for the radiotracer,
k (local lymph flow per unit volume of distribution of tracer, units: % min-1), was determined from the
regression slope of the plot of logn fraction of counts remaining at the depot against time . No
formal definition of response was agreed prior to the study.
Quality of life
The EORTC core questionnaire QLQ-C30 and breast module BR23 were chosen for the study. The
core questionnaire QLQ-C30 incorporates five functional scales (physical, role, cognitive, emotional,
and social), three symptom scales (fatigue, pain, and nausea and vomiting), a global health status/QoL
scale, and a number of single items assessing additional symptoms commonly reported by cancer
patients (dysphnoea, loss of appetite, insomnia, constipation and diarrhoea) and perceived financial
impact of the disease. The breast module BR23 comprises 23 questions assessing disease symptoms,
side effects of treatment (surgery, chemotherapy, radiotherapy and hormonal treatment), body image,
sexual functioning and future perspective . Volunteers were asked to complete set of a baseline
questionnaires before they started HBO therapy.
Hyperbaric oxygen therapy
Research volunteers were registered on the first day of treatment by a telephone call to the Clinical
Trials & Statistics Unit, Institute of Cancer Research. Volunteers were compressed to 2.4 ATA
(243 kba) in a multiplace Category 1 hyperbaric chamber at the Royal Hospital Haslar. Patients
breathed 100% oxygen at pressure via a transparent hood. The total time at 2.4 ATA was 100 minutes,
including two 5-minute ‘air breaks’. Each participant received a total of 30 pressure exposures,
treating five days a week for six weeks.
Clinical assessment and perometer measurement of arm volumes were repeated within 1 week of
completing six weeks of HBO2 therapy and 6 and 12 months after start of treatment. Self-assessments
of quality of life were completed at the same intervals plus3 and 9 months after start of treatment.
Lymphoscintigraphy and clinical photographs were repeated at 12 months only.
The primary endpoint was defined as an absolute change of >20% in the relative volume of the
ipsilateral arm vs. contralateral arm at 12 months. In addition, the relative change in excess arm
volume (arm swelling) between baseline and 12 months was analysed as an unplanned endpoint.
Planned secondary endpoints included i) lymphoscintigraphy, ii) patient self-assessments of arm
swelling, tissue hardness and physical functioning, and iii) physician assessments of photographic
appearance and palpable induration of the breast/chest wall, pectoral fold and/or supraclavicular fossa.
In addition, comments on arm softening and shoulder mobility volunteered by patients at clinical
assessment were incorporated as unplanned endpoints.
The appropriate size of the study group was determined by the feasibility of testing a complex
treatment as much as by statistical considerations. A sample size of 21 patients was judged to be large
enough to provide some indication of whether HBO has any effect on arm lymphoedema.
Change in ipsilateral arm volume was analysed on a continuous scale as the percentage change in
excess volume of the swollen arm as a percentage of the contralateral arm at baseline (bl) and after
therapy. In the protocol definition of response was an absolute change of >20% in the relative volume
of the ipsilateral arm vs. contralateral arm at 12 months (absdif = voldif12-voldifbl) e.g. for case 1 (see
Table 1) voldif12-voldifbl = 131-137 = -6 (minus sign indicates a reduction in volume). Change from
baseline at 12 months was examined using a paired t-test, since an inspection of the histogram and
normal probability plot suggested that a t-test was appropriate. In addition, the relative change in
excess arm volume (arm swelling) between baseline and 12 months was analysed using the formula
reldif = ((voldif12/voldifbl) x 100)-100 (where reldif is the relative difference in excess volume at 12
months e.g. for case 1 ((31/37) x 100)-100 = -16% relative change (reduction)). Health status scales
were derived from the questionnaires using standard methods . Change in lymphoscintigraphy,
breast induration (palpation), breast appearance (clinical photographs), arm softening and shoulder
mobility were recorded for each individual volunteer. No formal statistical analyses were planned of
the secondary endpoints, since the sample size was not large enough to provide reliable results on
multiple endpoints. However, it was felt appropriate to analyse the lymphoscintigraphy data formally.
Paired t-tests were used after confirming the normality of the data.
The median age of the 21 volunteers (20 female and 1 male) at the start of HBO2 was 64 years (range
53-76). The median time from primary therapy for breast cancer and onset of lymphoedema was 2
years (range 1 month-34 years). The median time from radiotherapy treatment to HBO2 was 14 years
(range 7-35). 10/21 volunteers had wide local excision as part of their primary treatment for breast
cancer, and 8 of these patients had some form of axillary surgery as well. Eleven out of 21 volunteers
underwent mastectomy, and 10 of these volunteers had some form of axillary surgery as well. The
level of axillary surgery performed was described in various terms, including “level I” and “axillary
clearance”, but the exact description was rarely available in the patients’ operation notes. All
volunteers had radiotherapy to the breast/chest wall and axilla +/- supraclavicular fossa.
Compliance with the treatment protocol was 100%. All planned 126 patient self-assessment
questionnaires were completed and returned on time, and only two individuals missed follow up
assessments (one moved away and declined her 12-month assessment; and another was hospitalised
with septicaemia and was advised by her medical practitioners not to travel). Hence, 81/84 (96%) of
perometer measurements and clinical assessments were completed as planned. Due to logistical
problems, only 15/21 (71%) patients entered the lymphoscintigraphy protocol. We found ourselves in
a situation where we were ready to start HBO2 therapy, but unable to fit in a sufficient number of
pretreatment lymphoscintigraphy scans. It was therefore decided to enter as many volunteers as
possible within the timeframe given, and the selection of individuals was entirely random. Fourteen
out of 15 (93%) completed the lymphoscintigraphy protocol (one volunteer did not attend for her 12-
month scan). Unfortunately, one follow up scan was accidentally erased from the computer system,
and this volunteer was therefore not evaluable for the study.
Results of arm volume measurements before HBO , one week post-HBO
22, at 6 months and 12 months
from study entry are listed in Table 1. Twenty-one out of 21 volunteers were assessed one week after
the end of therapy (7 weeks after start of HBO ), 20/21 at 6 months after start of HBO
22 and 19/21 at 12
months. For the purpose of the analysis only the 19 patients assessed at 12 months were included. The
median volume of the ipsilateral limb expressed as a percentage of contralateral limb volume at
baseline was 154% (range 131-213%). One week after the end of HBO2 therapy the median volume
was 159% (range 128-205%). 6 months after start of therapy this was reduced to 150% (range 114-
202%), and at 12 months the median volume of the ipsilateral limb expressed as a percentage of
contralateral limb volume was 144% (range 115-199%). The mean difference calculated at the same
time points changed from 157% at baseline to 149% 12 months after start of therapy.
In the trial protocol, a 'response' was defined as >20% reduction in arm volume at 12 months. This
level of improvement was considered to be both clinically significant and unlikely to be attributable to
supportive measures or change in lifestyle. 3 patients (volunteers 3, 14 and 15) were responders
according to this classification (range 21-29%), 16 patients were non-responders and 2 patients
(volunteers 7 and 21) could not be assessed, as they did not complete their 12-month clinical
assessments. Examining the relative reduction in arm swelling (excess arm volume rather than total
arm volume) as an unplanned secondary endpoint, 5/19 patients (volunteers 3, 6, 8, 14 and 15)
experienced >20% (range 25-66%) reduction in arm swelling at 12 months.
There was a statistically significant reduction in arm volume from baseline at 12 months (p = 0.005).
The mean percentage reduction in arm volume from baseline at 12 months was 7.68 (95% CI 2.65-
The lymphoscintigraphy outcome is shown graphically in Fig 1. Lymphoscintigraphy was performed
before HBO2 and 12 months after start of therapy as a planned secondary endpoint. This was not a
condition for participation in the study, but 13/21 volunteers consented and were evaluable. The mean
ratio between the lymphatic clearance rate in the arm affected by lymphoedema and the contralateral
arm (kipsilat/kcontralat) at baseline was 0.76, ie a reduction of 24% in the ipsilateral arm (constant with
published findings ). At 12 months after the start of HBO the mean value for k
2 ipsilat had increased
from 0.0244 to 0.0334 min (59%), whereas the mean value for k
contralat had increased from 0.0366 to
0.0386 min-1 (19%). As a result, the mean ratio kipsilat/kcontralat was increased to 0.92, i.e. an
improvement of 16% at 12 months post start of HBO .
There was a statistically significant difference between kipsilat and kcontralat at baseline (p = 0.05), but not
at 12 months after start of therapy (p = 0.20). Comparing kipsilat at baseline and kipsilat at 12 months
there was a statistically significant improvement (p = 0.03), whereas there was no significant
difference between k
at baseline and k
contralat contralat at 12 months (p = 0.71). Finally, when comparing
the ratio ipsilateral/contralateral baseline versus 12 months there was a significant difference (p =
Clinical assessments including tissue hardness and clinical photographs
The results of the serial clinical assessments of tissue hardness are summarised in Table 2. At baseline
17/19 patients had some degree of tissue hardness in irradiated areas of the breast/chest wall (n = 8),
pectoral fold (n = 16) and/or supraclavicular fossa (n = 12). Softening of a substantial number of these
sites was noted at the first post-therapy assessment one week after end of HBO2, and further softening
was recorded in some cases at 12 months post-therapy. The total improvement at 12 months were 1/8
(13%) sites on the breast/chest wall, 8/16 (50%) sites in the pectoral fold and 4/12 (33%) sites at the
supraclavicular fossa (total 13/36 (36%) sites in 8/17 (47%) patients). Clinical photographs were taken
of the volunteers at baseline and 12 months post start of HBO2, but these were not found to be
informative, recording no changes in appearance that were not recorded by perometry.
QLQ-C30 & BR23 and patient self-reports
The data were analysed but recorded no significant changes for better or worse over the study period
(data not shown). However, a proportion (12/19, 63%) of volunteers reported a definite softening in
the tissues of their affected arms. Six volunteers reported improvement in shoulder mobility and, in
cases of marked volume reduction, improved posture and less pain and feeling of heaviness, see Table
2. In Table 2, it is clear that there are no obvious associations between any of the endpoints reported.
Fibrosis is believed to contribute to the development of lymphoedema. The pathophysiology of
lymphoedema after radiotherapy and/or surgery involves obstruction of lymphatic flow causing an
imbalance between capillary filtration and lymph drainage [16,23]. Although physical removal of
lymphatic vessels at surgery offers a partial explanation, the variable onset, progression rate and
ultimate severity indicate that this is not the only mechanism. Radiotherapy to the axilla is a potent
cause of arm lymphoedema in its own right, more so after any kind of surgical disturbance of the axilla
. The continuous accumulation and contraction of scar tissue over many years is considered to be a
cause of progressive lymphatic obstruction in response to radiotherapy. The most vivid accounts of
fibrosis come from the surgical records of affected patients describing scar tissue infiltrating and
compressing axillary structures, especially the neurovascular sheath.
The traditional view regards radiation fibrosis as a passive phenomenon, representing the residual
extracellular matrix of a tissue depleted of cells. As such, fibrosis is regarded as an end-state beyond
effective therapeutic intervention. An alternative model considers radiation fibrosis in terms of
deregulated collagen metabolism with fibroblasts playing a central role [4,21]. The active model shares
features with other fibrotic states characterised by chronic imbalances between collagen deposition and
resorption . The significance of the active model is that it raises possibilities for reversing
Hyperbaric oxygen (HBO2) therapy currently provides the strongest proof of principle in support of an
active model for normal tissue responses to radiotherapy, with two comparative studies reported in
patients with heavily damaged tissues. The first study tested preoperative hyperbaric oxygen against
conventional penicillin cover in a randomised study of 74 patients requiring dental extraction
following radical mandibular irradiation . Only 5.4% of the patients in the hyperbaric oxygen
group compared to 30% of the patients in the penicillin-treated group experienced failure of wound
healing 6 months after surgery (p<0.05). In 160 patients treated by hyperbaric oxygen or standard
postoperative care following major soft tissue surgery for radiotherapy injury, four-fold reductions in
wound dehiscence, infection and delayed wound healing were seen in the hyperbaric oxygen treated
group (p<0.01) . The pathological correlates of the response to hyperbaric oxygen in irradiated
tissues have been studied in animals, and include neovascularisation, organisation and marked
reductions in fibrous tissue . However, it is not currently clear how these relate to the clinical
effects reported in humans.
There are several reports of HBO2 in the treatment of patients with haemorrhagic cystitis following
pelvic radiation and refractory to conventional measures. Bevers et al. describe 40 patients treated for
20 sessions of HBO2 for severe haematuria . Thirty-seven of the 40 patients showed improvement in
symptoms and no adverse effects were reported. Lee et al. treated 20 similar patients from 1989-1992
and again showed improvement of haematuria in 90% . Schoenrock and Cianci reported a patient
with severe haemorrhagic cystitis and a vesicocutaneous fistula which healed with HBO2 . Nakada
et al. gave 6 patients with radiation cystitis HBO2 and also found improvement in all but one patient
. Similar reports on small patient numbers are given by Kindwall, Weiss and Norkool [8,19,28].
No adverse effects of HBO were reported in any of these studies.
Turning to the current data, the evidence is highly suggestive of a therapeutic effect, but falls short of
providing reliable evidence for clinical benefit. Assuming HBO2 has reduced lymphoedema, it is not
possible to say if this relates to the changes induced by surgery, radiotherapy or both. Where
experimental design is concerned, the important decision to conduct a non-randomised study was
determined by the assumptions i) that perometry offers an operator-independent measure of arm
volume and ii) that a sustained reduction of >20% in arm volume could not occur spontaneously, or in
response to other interventions such as more intensive arm care by lymphoedema clinics. The average
reduction in arm volume reported was just below 10%, with only 3/19 patients recording >20%
reductions in arm volume (5/19 with >20% reductions in arm swelling) at 12 months. Although this
reproduces our previous observations, the modest level of effect makes it difficult to be certain that
responses were HBO2-induced, rather than seasonal fluctuation (this appears to be ruled out by
reviewing the calendar months of measurements), changes in life-style and/or more intense interest
shown by lymphoedema clinics responsible for routine management (we have no evidence for either
of these effects).
Despite these uncertainties, quantitative lymphoscintigraphy offers strong supportive evidence for a
durable effect attributable to HBO2. The robustness of the technique is indicated by the similarity of
the pre-HBO2 difference (24%) with published data . The baseline clearance rate of radioactive
tracer was <0.9 (ratio of kipsilat to kcontralat) in the ipsilateral arm compared to the contralateral limb in
9/13 patients before treatment and in only 4/13 patients 12 months later, a statistically significant, and
highly clinically significant, improvement. It is difficult to ignore the softening of the ipsilateral arm
reported by so many (12/19) evaluable patients, not considered as a planned endpoint, but mentioned
by so many patients without prompting. Assuming the recorded changes in arm volume, 99Tc-Nanocoll
clearance rate and arm hardness to be consequences of HBO2, it is possible to speculate that limb
tissues can be insufficiently elastic to respond to restoration of normal clearance rates by volume
The physician assessments of tissue softening recorded in Table 2 were planned secondary endpoints.
Whilst these are highly subjective, it is considered unlikely that improvements of 2 grades represent
measurement errors or spontaneous change. The pathological changes underlying induration are not
clear, although tissue fibrosis and oedema are likely to be important. The same pathological features
may be relevant to the pathogenesis of arm lymphoedema after lymphatic surgery and radiotherapy to
HBO2. If so, the response at one site might be expected to be associated with a response at another.
Possibly due to the very small sample size, these associations are not apparent.
The improvements noticed in patient self-assessments of quality of life were not statistically
significant. Patients commented that the choice of questionnaires was inappropriate. They felt that the
questions would have been relevant to them in the first few years after their treatment for breast
cancer/onset of lymphoedema, but not several years later when they had adjusted to their situation and
In summary, it is perhaps unsurprising that firmer conclusions cannot be drawn from a sample size of
19 evaluable patients lacking a randomised control group. Nevertheless, even if the volume changes
are difficult to attribute, the lymphoscintigraphy results are highly suggestive of a real effect in a
group of patients receiving primary treatment a median of 14 years ago (range 7-35). We think further
work is certainly justified in the context of a double-blind, randomised controlled clinical trial, and
this is under development.
We are grateful for the enormous support and commitment shown by the volunteers participating in
this study. We acknowledge the contribution of staff at the Nuclear Medicine Department at The
Royal Marsden Hospital, the Hyperbaric Medicine Unit at the Royal Hospital Haslar and the technical
staff at the Defence Evaluation and Research Agency (DERA). The active support of the Royal
College of Radiologists (UK) is also recognised. We acknowledge the Medical Research Council and
the Ministry of Defence that jointly funded the work. Statistical input was undertaken at the ICR
Clinical Trials and Statistics Unit with funding from Cancer Research UK.
 Bates, T. and Evans, R. G. B. Report of the independent review commissioned by the Royal
College of Radiologists into brachial plexus neuropathy following radiotherapy for breast
carcinoma. The Royal College of Radiologists, London 1995.
 Bevers, R. F., Bakker, D. J. and Kurth, K. H. Hyperbaric oxygen treatment for haemorrhagic
radiation cystitis. Lancet 1995; 346: 803-5.
 Fayers, P., Aaronson, N., Bjordal, K., Curran, D. and Groenvold, M., EORTC ALA-C30
Scoring Manual. Brussels. EORTC Quality of Life Study Group 1999.
 Finkelstein, J. N., Johnston, C. J., Baggs, R. and Rubin, P. Early alterations in extracellular
matrix and transforming growth factor beta gene expression in mouse lung indicative of late
radiation fibrosis. Int J Radiat Oncol Biol Phys 1994; 28: 621-31.
 Fischbach, J. U., Mönter, B. and Göltner, E. Messungen von Armödemen durch
optoelektronische Volumetrie. Phlebologie und Proctologie 1986; 15: 184-189.
 Forrest, A. P., Everington, D., McDonald, C. C., Steele, R. J., Chetty, U. and Stewart, H. J. The
Edinburgh randomized trial of axillary sampling or clearance after mastectomy. Br J Surg 1995;
 Grotendorst, G. R., Paglia, L., McIvor, C., Barsky, S., Martinet, Y. and Pencev, D.
Chemoattractants in fibrotic disorders. Ciba Found Symp 1985; 114: 150-63.
 Kindwall, E. P. Hyperbaric oxygen treatment of radiation cystitis. Clin Plast Surg 1993; 20:
 Kissin, M. W., Querci della Rovere, G., Easton, D. and Westbury, G. Risk of lymphoedema
following the treatment of breast cancer. Br J Surg 1986; 73: 580-4.
 Lee, H. C., Liu, C. S., Chiao, C. and Lin, S. N. Hyperbaric oxygen therapy in haemorrhagic
radiation cystitis: a report of 20 cases. Undersea Hyperb Med 1994; 21: 321-7.
 Logan, V. Incidence and prevalence of lymphoedema: a literature review. J Clin Nurs 1995; 4:
 Maher, E. J. Maher Report. Management of adverse effects following breast radiotherapy.
Maher Committee, London: The Royal College of Radiologists. 1995.
 Marx, R. E. Radiation injury to tissue. Hyperbaric Medicine Practice. Kindwall EP (Ed)
Flagstaff, Arizona (Publ) 1994; 447-503.
 Marx, R. E., Ehler, W. J., Tayapongsak, P. and Pierce, L. W. Relationship of oxygen dose to
angiogenesis induction in irradiated tissue. Am J Surg 1990; 160: 519-24.
 Marx, R. E., Johnson, R. P. and Kline, S. N. Prevention of osteoradionecrosis: a randomized
prospective clinical trial of hyperbaric oxygen versus penicillin. J Am Dent Assoc 1985; 111:
 Anonymous. Arm oedema following breast cancer treatment. Drug and Therapeutics Bulletin
2000; 38 (6): 41-43.
 Mortimer, P. S., Bates, D. O., Brassington, H. D., Stanton, A. W. B., Strachan, D. P. and
Levick, J. R. The prevalence of arm oedema following treatment for breast cancer. Q J Med
1996; 89: 377-380.
 Nakada, T., Yamaguchi, T., Sasagawa, I., Kubota, Y., Suzuki, H. and Izumiya, K. Successful
hyperbaric oxygenation for radiation cystitis due to excessive irradiation to uterus cancer. Eur
Urol 1992; 22: 294-7.
 Norkool, D. M., Hampson, N. B., Gibbons, R. P. and Weissman, R. M. Hyperbaric oxygen
therapy for radiation-induced hemorrhagic cystitis. J Urol 1993; 150: 332-4.
 Pritchard, J., Anand, A., Broome, J., Davis, C., Gothard, L., Hall, E., Maher, J., McKinna, F.,
Millington, J., Misra, V. P., Pitkin, A. and Yarnold, J. Double-blind randomised phase II study
of hyperbaric oxygen in patients with radiation-induced brachial plexopathy. Radiother Oncol
2001; 58(3): p. 279-86.
 Rubin, P., Finkelstein, J. and Shapiro, D. Molecular biology mechanisms in the radiation
induction of pulmonary injury syndromes: interrelationship between the alveolar macrophage
and the septal fibroblast. Int J Radiat Oncol Biol Phys 1992; 24: 93-101.
 Schoenrock, G. J. and Cianci, P. Treatment of radiation cystitis with hyperbaric oxygen.
Urology 1986; 27: 271-2.
 Stanton, A. W. B., Levick, J. R. and Mortimer, P. S. Current puzzles presented by
postmastectomy oedema (breast cancer related lymphoedema). Vasc Med 1996; 1: 213-25.
 Stanton, A. W. B., Northfield, J. W., Holroyd, B., Mortimer, P. S. and Levick, J. R. Validation
of an optoelectronic limb volumeter (Perometer). Lymphology 1997; 30: 77-97.
 Stanton, A. W., Svensson, W. E., Mellor, R. H., Peters, A. M., Levick, J. R. and Mortimer, P. S.
Differences in lymph drainage between swollen and non-swollen regions in arms with breast-
cancer-related lymphoedema. Clin Sci (Lond) 2001; 101: 131-40.
 Swedborg, I. and Wallgren, A. The effect of pre- and postmastectomy radiotherapy on the
degree of edema, shoulder-joint mobility, and gripping force. Cancer 1981; 47: 877-81.
 Turesson, I. The progression rate of late radiation effects in normal tissue and its impact on
dose-response relationships. Radiother Oncol 1989; 15: 217-26.
 Weiss, J. P., Mattei, D. M., Neville, E. C. and Hanno, P. M. Primary treatment of radiation-
induced haemorrhagic cystitis with hyperbaric oxygen: 10-year experience. J Urol 1994; 151:
Table 1 Perometer measurements in 21 patients taken before hyperbaric oxygen therapy (HBO2)
and 12 months after therapy.
Table 2 Summary of results of study including change in arm volume and arm swelling, tissue
softening, lymphoscintigraphy and patient self-reports in 19 patients evaluable at 12
months post HBO .
Figure 1 Removal rate constant, k, pre-HBO2 and 12 months after hyperbaric oxygen therapy
(HBO ). p-values represent between-arms comparisons using the paired-t test.
Perometer measurements in 21 patients taken before hyperbaric oxygen therapy (HBO2
) and 12 months after therapy
at 12 months
at 12 months
Ipsilat / contralat
Ipsilateral / contralateral
at 12 months
3108 / 2153
4528 / 2977
3890 / 2474
3282 / 2201
4008 / 2354
3102 / 1886
3178 / 2335
3275 / 2277
2852 / 1849
2936 / 1930
2739 / 1994
4311 / 3247
3808 / 2306
3092 / 2360
3475 / 2023
3973 / 2331
3849 / 2287
5345 / 3061
3115 / 2043
4622 / 1621
2528 / 2173
3979 / 3118
3899 / 2859
3003 / 2227
5963 / 2982
3635 / 2277
3042 / 1970
3127 / 2462
3067 / 2262
2639 / 1824
2946 / 2036
4229 / 3272
3505 / 2133
2938 / 2229
3571 / 2054
4214 / 2418
4193 / 2408
5954 / 3202
* - = decrease; + = increase
n/k = not known
Summary of results of study including change in arm volume and arm swelling, tissue softening,
lymphoscintigraphy and patient self-reports in 19 patients evaluable at 12 months post HBO2
Change in volume Tissue softening
12 months after start of
06 14 -15 nil
04 21 -14 -12 no
05 14 -11 -16 nil
19 7 -10 -16 nil
08 21 -9 no
11 22 -9 -20 no
16 14 -9 -17 no
13 11 -8 -15
01 11 -6 -16 no
09 8 -3 -9 nil
02 16 -1 -2 nil
10 32 0 0 nil
20 10 3 4 nil
12 14 4 6 nil
17 20 6 9 no
18 9 11 15 nil
Lymphoscintigraphy Patient self-reports
12 months after start of HBO2
arm tissue 12
start of HBO
start of HBO
clearance rate 12
start of HBO2
Unevaluable: no change or no measurement at baseline.
Evaluable: Improvement at 12 months post HBO2 from baseline.
Evaluable: No improvement at 12 months post HBO2 from baseline.
Fig 1 Download full-text
p = 0.05 p = 0.20
Pre-HBO2 12 months post-HBO2
Ipsilateral Contralateral Ipsilateral
Removal rate constant, k