BJD. Summary. British Journal of Dermatology THERAPEUTICS

Transcription

1 THERAPEUTICS BJD British Journal of Dermatology A preliminary randomized controlled study to determine the application frequency of a new lymphoedema bandaging system C.J. Moffatt,* P.J. Franks, à D. Hardy, M. Lewis,** V. Parker and J.L. Feldmanàà *Royal Derby Hospital Lymphoedema Service, Derby, U.K. Faculty of Medicine, Division of Nursing and Healthcare, University of Glasgow, Glasgow, U.K. àcentre for Research & Implementation of Clinical Practice, 174 St Luke s Crypt, Sydney Street, London SW3 6NH, U.K. London South Bank University, Faculty of Health & Social Care, London, U.K. Kendal Lymphology Centre, Kendal, U.K. **Swansea Lymphoedema Service Singleton Hospital, Swansea, U.K. University of Missouri, Columbia, MI, U.S.A. àànorthshore University HealthSystem Research Institute, Evanston, IL, U.S.A. Summary Correspondence Peter J. Franks. peter.franks@cricp.org.uk Accepted for publication 1 November 2011 Funding sources Funding was provided by 3M. Conflicts of interest C.J.M. and P.J.F. have both received research income on behalf of the Centre for Research & Implementation of Clinical Practice from 3M prior to and during this study. The study was registered under clinicaltrials.gov no. NCT DOI /j x Background Bandaging plays an important role in the treatment of lymphoedema. Objective To investigate efficacy and safety of the 3MÔ CobanÔ 2 compression system (Coban 2 system) with different application frequencies in comparison to short-stretch bandaging. Methods A multicentre, randomized, prospective study was performed with 82 patients suffering from arm or leg lymphoedema stage II or late stage II. Patients were allocated to traditional short-stretch bandaging five times per week or to the Coban 2 system applied two, three or five times per week for 19 days. Limb volume and adverse events were recorded at each study visit. The primary endpoint was percentage volume reduction. Results The highest lymphoedema volume reduction was achieved with the Coban 2 system applied two times per week. A mean reduction of 18Æ7% (SD 14Æ5) in legs and 10Æ5% (SD 8Æ3) in arms was achieved. More frequent bandage changes of three and five times per week did not demonstrate additional benefits. Shortstretch bandaging five times per week showed a mean volume reduction of 10Æ9% (SD 5Æ2) and 8Æ2% (SD 3Æ1) for legs and arms, respectively. Bandage slippage was low for all treatment groups. A relevant change in overall mobility was achieved during the use of the Coban 2 system. The adverse reactions were in agreement with already known side-effects and did not differ remarkably between the treatment groups. Conclusion The 3MÔ CobanÔ 2 compression system applied twice weekly demonstrated a high rate of volume reduction and a good safety profile. Oedema reduction was still effective with 4 days between bandage change, which allows a constant therapeutic effect in routine practice. This should give the patient a high degree of independence and mobility. Lymphoedema is a chronic overload of protein-rich interstitial volume, secondary to drainage failure of a compromised lymphatic system. A study in the U.K. revealed a prevalence of 1Æ33 per 1000 inhabitants. 1 Lymphoedema often occurs in the extremities, probably because of the low number of alternate lymph pathways for drainage. Further accumulation of protein-rich fluid results in fibrosis. The oedema is initially a pitting oedema, but over time the subcutaneous tissue becomes fibrotic, resulting in nonpitting oedema. Eventually, the involved skin becomes hyperkeratotic, hyperpigmented and papillomatous. Several studies show that compression is the first choice of treatment, as pressure is the main prerequisite for substantial oedema reduction. 2 5 Mechanisms possibly explaining the efficacy of compression therapy include 624

2 Lymphoedema bandaging, C.J. Moffatt et al. 625 reduction of capillary filtration into the interstitial tissue, increase in lymphatic reabsorption and muscular contraction of lymph vessels, enhanced venous return, and breakdown of fibrosclerotic tissue. Badger et al. 2 demonstrated in a randomized clinical trial that bandaging followed by compression stockings results in a better and more sustained outcome than compression stockings alone. Bandaging plays an outstanding role in the treatment of lymphoedema, reflected by the fact that a complete European Wound Management Association 6 focus document was produced on Lymphoedema bandaging in practice. This randomized clinical trial investigated the efficacy and safety of the 3MÔ CobanÔ 2 compression system (Coban 2 system; 3M Deutschland GmbH, Neuss, Germany), with three different application frequencies, in comparison to a standard short-stretch bandage. Methods This randomized, multicentre, open-label, controlled study was conducted in seven specialist lymphoedema centres in the U.K. and four in the U.S.A. The protocol was approved by independent ethics committees or investigational review boards, depending on the legal requirements of each country. Study participants granted written informed consent prior to inclusion into the study. Subjects were included if they were 18 years or older and were suffering from lymphoedema stage II or late stage II according to the International Society of Lymphology (ISL) staging. 7 These stages are defined as: ISL stage II limb elevation alone rarely reduces swelling, and pitting is manifest; ISL late stage II there may or may not be pitting as tissue fibrosis is more evident. In addition, patients required intense bandaging therapy and had completed all primary and adjuvant cancer treatments at least 6 months prior to randomization. In contrast to patients with leg lymphoedema, patients with arm lymphoedema were only included with secondary and unilateral disease. Patients who were pregnant were excluded, as were patients with evidence of active cancer, a period of intensive bandaging during the preceding month, decompensated heart failure or clinically relevant kidney or liver disease, known arterial disease, paralysis, clinical infection or wounds requiring dressing change more than once weekly on the study limb, deep vein thrombosis within the last 3 months, history of allergic reactions to the study material or participation in another prospective clinical trial that could potentially interfere with this study. According to Laplace s equation, the pressure applied is a relation of the bandage tension, width, number of layers and radius of the limb. Taking into consideration different circumferences for arms and legs, the Coban 2 system is provided in two versions for the upper (3MÔ CobanÔ 2 Lite) and lower extremities (3MÔ CobanÔ 2). In contrast to other currently available compression systems for intense lymphoedema treatment, the Coban 2 system is designed to be applied without padding to achieve a thin and comfortable system allowing the patient a high degree of independence and mobility. The Coban 2 system consists of a comfort layer comprising medical foam laminated to a nonwoven backing and a self-adherent compression layer. The comfort layer is applied without tension whereas the compression layer is applied at full stretch. For anatomical areas with small circumferences, including arms, fingers and toes, 3MÔ CobanÔ 2 Lite was used, whereas for legs, 3MÔ CobanÔ 2 was applied. For both systems no additional materials such as padding or foams were used. For the Coban 2 system a new application technique was developed. The short-stretch bandage (Comprilan Ò ; BSN Medical Ltd, Hull, U.K.) was wrapped according to a standardized protocol developed on the basis of the manufacturer s application guidelines, including a tubular retention bandage and Cellona Ò (Activa Healthcare Ltd, Burton-upon-Trent, U.K.) as padding material. The bandaging systems were always applied by study personnel trained before the study start and supported by detailed application instructions in the protocol. All patients were randomly allocated to one of the following treatment groups with equal numbers of patients with leg and arm lymphoedema (Table 1). For each bandage renewal, patients had to come to the site during a total treatment period of 19 days (three working weeks). To support the therapeutic effect, patients were instructed to perform daily physical exercises developed by one of the participating lymph therapists. Table 1 Treatment groups and number of patients per analysis set Stratum Treatment group Total applications Safety analysis set Arm (n = 42) Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week Total Leg (n = 40) Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week Total Full analysis set

3 626 Lymphoedema bandaging, C.J. Moffatt et al. The primary study objective was to assess the percentage reduction in limb volume. Secondary objectives were safety, slippage, health-related quality of life and health-economic parameters. The last two will be the subject of another publication. Further parameters included absolute volume reduction and mobility. Adverse events were recorded as and when they occurred. The limb circumference measurements were done along the study limb in 4-cm intervals. The limb circumference data were used for absolute limb volume calculation using the following formula: V ¼ðhÞðC 2 þ Cc þ c 2 Þ=12ðpÞ where C and c are the circumferences at each end of the truncated cone. Percentage volume reduction was calculated based on the relation of absolute volumes at baseline to the volumes at each subsequent follow-up visit. The slippage for leg bandaging was assessed by measuring the length of the bandage immediately after application (from the top of the bandage to the floor, when the patient s foot was resting flat on the floor, standing) and measured again before the next bandage change, in the same position. For arms, slippage was assessed by measuring the length of the bandage immediately after application (from the top of the bandage to the nail bed of the little finger on the ulnar aspect of the arm, when the arm was placed on a table with the hand palm down) and measured again before the next bandage change. Study limb and overall mobility were assessed by patients at baseline and at the end of each week on a 10-point scale with 0 as worst and 10 as the best rating. The absolute change from study start was calculated at the end of each week of treatment. Adverse events were documented, including their description, date of onset, severity, seriousness, expectedness, relation to the device, action taken and outcome. Adverse events were coded using the Medical Dictionary for Regulatory Activities (MedDRA) version Itching and pain were assessed separately from adverse events. Statistical considerations and data analysis Sample size calculations to provide the basis for a powered statistical analysis were not performed. The sample in this study was designed to indicate trends for decisions on application frequencies and to allow sample size calculations for future appropriately powered clinical studies. The randomization list was generated by ClinResearch GmbH (Cologne, Germany), with a block size of four. Separate randomization lists were generated for the arms and legs with sealed envelopes. ClinResearch was informed immediately after the randomization of each patient and kept central lists of the numbers of patients in each stratum and treatment group. As soon as the first treatment group in each stratum had 10 patients enrolled, randomization by envelope was stopped and replaced by central telephone randomization. This central telephone randomization continued to work along the randomization list, but was free to skip allocation to selected treatment arms if the number of patients in this arm exceeded 10. The central telephone randomization was performed by 3M. Blinding of the treatment procedure was not possible due to the obvious differences of the compression devices. The study data were evaluated by descriptive summary statistics. Results Between February and September 2010, 82 patients were treated in this study. Forty-two suffered from arm lymphoedema and 40 from leg lymphoedema. All 82 patients included in the study qualified for the safety analysis. Figure 1 gives the flow diagram of the patients entered into this trial. At least one completed bandage wear episode was documented for 76 patients and qualified for the full analysis set (FAS). Of the 76 patients, 38 suffered from arm and 38 from leg lymphoedema. Table 1 presents the number of patients randomized to each treatment group. In the arm stratum, all the patients were women while in the leg stratum five of the 38 (13%) patients were men. The mean height, weight and body mass index (BMI) were similar between the treatment groups of the arm stratum. In the leg stratum, patients on the Coban 2 system renewed three times per week had a slightly lower body weight than the patients in the other treatment groups. The patients on short-stretch bandaging were slightly younger (Table 2). In the arm stratum, all patients had lymphoedema due to cancer apart from one patient in the five times per week Coban 2 system group who had axillary lipoma and arm vein thrombosis. In the leg stratum, 21 of 38 patients (55%) had secondary lymphoedema, of which seven were related to cancer. In both strata, all patients received manual lymph drainage (MLD) before the study start. MLD was mandatory at each clinic visit for a duration of up to 45 min. The MLD actually performed prior to the application of each study bandage was approximately 20 min on average for all treatment groups. Limb volume reduction The highest mean percentage volume reduction of 18Æ65% (SD 14Æ5) over 19 days was achieved in the leg stratum with the Coban 2 system applied twice weekly (Table 3). The second highest volume reduction was found in the five times per week Coban 2 system group at 12Æ0% (SD 10Æ48). For the arm stratum, the highest mean volume reduction of 10Æ48% (SD 8Æ33) was achieved by the Coban 2 system applied twice weekly followed by 8Æ47% (SD 6Æ37) for the Coban 2 system renewed five times per week. The mean volume reduction achieved with short-stretch bandaging was 10Æ87% (SD 5Æ20) for legs and 8Æ19% (SD 3Æ13) for arms. The box plot for percentage reduction is presented graphically in Figure 2. As expected, no comparisons between treatments achieved a standard level of statistical significance (all P >0Æ05) due to the small sample size and low power. The

4 Lymphoedema bandaging, C.J. Moffatt et al. 627 Assessed for eligibility (n = 141) Excluded (n = 59) Not meeting inclusion criteria (n = 7) Declined to participate (n = 39) Other reasons (n = 13) Randomized (n = 82) Allocated to 3M 5 times/wk (n = 22) Received allocated intervention (n = 22) Did not receive intervention (n = 0) Allocated to 3M 3 times/wk (n = 20) Received allocated intervention (n = 20 ) Did not receive intervention (n = 0) Allocated to 3M 2 times/wk (n = 22) Received allocated intervention (n = 22) Did not receive intervention (n = 0) Allocated to Comprilan 5 times/wk (n = 18 ) Received allocated intervention (n = 18) Did not receive intervention (n = 0) Lost to follow-up (n = 0) Discontinued intervention (n = 4 ) AE = 2, Personal = 1, Compliance = 1 Lost to follow-up (n = 0) Discontinued intervention (n = 2 ) AE = 1 Personal = 1 Lost to follow-up (n = 0) Discontinued intervention (n = 4), AE = 2, Personal reason and compliance = 2 Lost to follow-up (n = 0) Discontinued intervention (n = 3 ) AE = 1, Personal = 1, other = 1 Analysed (n = 22) Excluded from analysis (n = 0) Analysed (n = 20) Excluded from analysis (n = 0) Analysed (n = 18) Excluded from analysis (n = 4) Did not complete at least 1 wear episode (4) Analysed (n = 16) Excluded from analysis (n = 2) Did not complete at least 1 wear episode (2) Fig 1. Trial flow diagram. AE, adverse event. Table 2 Age, height, weight and body mass index (BMI) per stratum and treatment group Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week Arm, n Age (years) 61Æ18 (9Æ42) 65Æ20 (9Æ53) 62Æ44 (12Æ34) 63Æ75 (10Æ46) Height (cm) 156Æ86 (4Æ78) 159Æ50 (5Æ36) 159Æ59 (8Æ02) 161Æ17 (9Æ39) Weight (kg) 78Æ07 (14Æ77) 74Æ96 (13Æ93) 81Æ68 (25Æ44) 74Æ44 (13Æ72) BMI (kg m )2 ) 31Æ62 (4Æ85) 29Æ47 (5Æ04) 31Æ63 (8Æ01) 28Æ66 (5Æ15) Leg, n Age (years) 63Æ36 (16Æ66) 58Æ50 (18Æ09) 57Æ00 (9Æ92) 48Æ50 (16Æ54) Height (cm) 165Æ31 (8Æ67) 161Æ92 (3Æ85) 161Æ10 (7Æ63) 161Æ49 (6Æ81) Weight (kg) 106Æ25 (33Æ39) 84Æ91 (33Æ80) 102Æ77 (33Æ08) 102Æ19 (41Æ24) BMI (kg m )2 ) 38Æ87 (11Æ78) 32Æ43 (12Æ77) 39Æ35 (11Æ74) 39Æ13 (16Æ14) Values are mean (SD). Table 3 Mean volume reduction after 19 days Volume reduction of study limb: end of treatment compared with baseline Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week Arm, n Per cent 8Æ47 (6Æ37) 6Æ78 (7Æ03) 10Æ48 (8Æ33) 8Æ19 (3Æ13) Millilitres 261 (307) 181 (234) 352 (335) 211 (105) Leg, n Per cent 12Æ00 (10Æ48) 7Æ43 (7Æ55) 18Æ65 (14Æ50) 10Æ87 (5Æ20) Millilitres 1093 (798) 831 (1076) 1552 (1221) 870 (454) Values are mean (SD).

5 628 Lymphoedema bandaging, C.J. Moffatt et al. % Volume reduction Legs Arms Coban 2 5 /week Coban 2 3 /week Coban 2 2 /week Short stretch Coban 2 5 /week Coban 2 3 /week Treatment groups Fig 2. Percentage volume reduction over 19 days. Coban 2 2 /week Short stretch Absolute change over time on a 10-point rating scale Study Start Friday Week 1 Friday Week 2 Friday Week 3 Coban 2 Arm Coban 2 Leg Short-stretch Arm Short-stretch Leg comparison of twice-weekly Coban 2 with Comprilan achieved a significance level of P =0Æ48 in the arm patients and P =0Æ17 in the leg patients. Analysing the percentage volume reduction over time for all groups, the main volume reduction was seen during the first 5 days of treatment (results not shown). Slippage was compared over time between all treatment groups. In general, the averages were similar between the treatment groups and indicated a mean slippage of approximately 2 cm at the arms and 4 7 cm at the legs. However, high SDs were found for the Coban 2 system applied twice weekly in both strata and for the Coban 2 system applied three times per week on the leg (Table 4). To investigate the influence of the bandage on overall and limb mobility, individual ratings were obtained at baseline and the end of each treatment week. For the evaluation of the overall mobility and the mobility of the study limb, the three different treatment groups on the Coban 2 system were pooled into one group and compared against short-stretch bandaging (Fig. 3). Data of the Coban 2 groups were pooled due to comparable results across all treatment groups. For arm and leg patients, the overall patient mobility continuously increased each week for the Coban 2 systems, whereas for short-stretch bandaging no relevant differences were seen between baseline and the end of treatment. However, during the therapy there was a transient reduction in mobility for the leg patients with short-stretch bandaging. The study limb mobility with the Coban 2 systems continuously improved for the leg patients whereas with the Coban 2 Fig 3. Overall mobility over time Absolute change over time on a 10-point rating scale Study Start Friday Week 1 Friday Week 2 Friday Week 3 Coban 2 Leg Coban 2 Arm Fig 4. Mobility of study limb over time. Short-stretch Arm Short-stretch Leg system for arms it stayed on the baseline level. Temporary reduction of limb mobility was observed in the short-stretch bandaging groups, especially in the leg patients (Fig. 4). Due to the same application procedure in all the Coban 2 groups, application time, i.e. the time needed to bandage the study limb, was evaluated after pooling the three Coban 2 system groups and compared against time needed for short-stretch Table 4 Mean slippage of bandage between Monday week 1 and Friday week 3 Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week Arm, n First volume reduction (cm) 2Æ26 (1Æ57) 2Æ08 (1Æ40) 2Æ13 (5Æ12) 1Æ35 (1Æ00) Leg, n First volume reduction (cm) 3Æ74 (3Æ04) 7Æ10 (5Æ97) 6Æ03 (7Æ41) 4Æ84 (3Æ21) Values are mean (SD).

6 Lymphoedema bandaging, C.J. Moffatt et al. 629 Table 5 Premature study termination Stratum Treatment Reason for premature termination Arm (n = 42) Coban 2 5 week Adverse event Yes Coban 2 5 week Adverse event Yes Coban 2 3 week Personal reason Yes Coban 2 3 week Adverse event Yes Coban 2 2 week Adverse event No Coban 2 2 week Lack of compliance and No personal reason and other Short-stretch Other Yes Short-stretch Personal reason No Short-stretch Adverse event No Leg (n = 40) Coban 2 5 week Personal reason Yes Coban 2 5 week Lack of compliance Yes Coban 2 2 week Adverse event No Coban 2 2 week Lack of compliance and personal reason No Included into full analysis set bandage application. The time ranged between 8 and 45 min for all groups. On average, the application time in minutes for the Coban 2 systems were slightly higher at a mean (SD) of 21Æ4 (5Æ4) for arms and 24Æ9 (8Æ2) for legs compared with short-stretch application times of 18Æ8 (6Æ7) for arms and 19Æ1 (5Æ2) for legs. Five of 38 (13%) patients from the FAS stopped the study prematurely in the arm stratum, whereas in the leg stratum two of 38 (5%) ceased study participation (Table 5). The safety evaluation was based on all randomized patients; of these, 42 were in the arm stratum and 40 were in the leg stratum. Adverse reactions to the study device were reported by 23 of 42 (55%) patients in the arm stratum and 22 of 40 (55%) patients in the leg stratum. Fifty-eight and 42 adverse events were reported in the group of the arm patients and leg patients, respectively, and of these 54 and 40 events were related to the study device. Table 6 shows details of the adverse reactions that were assessed as unanticipated by the investigators. The adverse reactions that led to withdrawal from the study are indicated. Although assessed by the investigators as unanticipated in nature or severity, the adverse reactions listed in Table 6 mainly represented symptoms caused by the mechanical effects of the bandages on the skin, subcutaneous tissue and musculoskeletal system and thus were compatible with generally known risks of bandaging. No medically relevant differences between frequency and nature of adverse reactions during the use of the Coban 2 system and short-stretch bandaging were identified. Itching and pain were assessed independently from the adverse events (Table 7). No relevant differences could be observed. Discussion The study was designed to consider which application frequency may be optimal for patients with lymphoedema using a new 3MÔ CobanÔ 2 compression system in comparison to a well-accepted bandage (Comprilan Ò ). For this purpose, a randomized, prospective, multicentre study on 82 patients with lymphoedema was performed. All groups had similar demographic profiles with exceptions of slight variations in age and weight. However, it is expected that these differences do not have a major impact on the recorded outcomes presented here. This study was conducted with patients with lymphoedema stage II and late stage II, this being the largest patient group who would require a period of intensive bandage treatment. Percentage volume reduction was chosen as a primary endpoint to enable comparability between treatment groups and strata (arms and legs). The highest volume reductions were achieved with the Coban 2 system when renewed twice weekly. A mean reduction of 18Æ7% (1552 ml) and 10Æ5% (352 ml) was documented for legs and arms, respectively. More frequent application of the Coban 2 system did not translate into better results. The reason why the three times per week application showed the lowest oedema reduction remains unclear. While we do not have strong evidence to indicate a real effect, the volume reduction profile in the arm patients was similar to that in the leg patients. It is acknowledged that this may be a chance event due to the relatively small sample size in this study. In both strata, short-stretch bandaging showed a lower mean volume reduction with 10Æ9% (870 ml) for legs and 8Æ2% (211 ml) for arms compared with the twice-weekly application of the Coban 2 system. As expected, the large SDs and small sample sizes in this study led to a low power. Because of this, the study was not expected to detect significant differences between treatment groups. Using these results, it is possible to determine a sample size estimate for a fully powered clinical trial. Using the mean (SD) values from Table 3 it is estimated that to demonstrate a difference between Coban 2 twice weekly and Comprilan a sample size of 238 patients in total would be required for the arm group, and a sample size of 62 (31 in each group) would be

7 630 Lymphoedema bandaging, C.J. Moffatt et al. Table 6 Adverse reactions assessed as unanticipated Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week MedDRA 13Æ0, preferred term n % n % n % n % Arm group Unanticipated in nature a 2 18Æ2 3 30Æ0 3 27Æ3 2 20Æ0 Medical device discomfort 0 0Æ0 1 10Æ0 1 9Æ1 0 0Æ0 Application-site discomfort 0 0Æ0 1 10Æ0 0 0Æ0 1 b 10Æ0 Pain in extremity 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Application-site irritation 0 0Æ0 1 10Æ0 0 0Æ0 0 0Æ0 Folliculitis 1 9Æ1 0 0Æ0 1 9Æ1 0 0Æ0 Anxiety 1 c 9Æ1 0 0Æ0 0 0Æ0 0 0Æ0 Dyspnoea 0 0Æ0 0 0Æ0 1 d 9Æ1 0 0Æ0 Joint sprain 0 0Æ0 0 0Æ0 0 0Æ0 1 e 10Æ0 Neuralgia 0 0Æ0 0 0Æ0 1 d 9Æ1 0 0Æ0 Open wound 0 0Æ0 1 10Æ0 0 0Æ0 0 0Æ0 Rash macular 0 0Æ0 0 0Æ0 1 d 9Æ1 0 0Æ0 Unanticipated in severity 0 0Æ0 1 10Æ0 0 0Æ0 0 0Æ0 Fibrosis 0 0Æ0 1 f 10Æ0 0 0Æ0 0 0Æ0 Leg group Unanticipated in nature a 4 36Æ4 1 10Æ0 4 36Æ4 2 25Æ0 Skin irritation 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Medical device discomfort 1 9Æ1 0 0Æ0 0 0Æ0 1 12Æ5 Application-site discomfort 1 9Æ1 0 0Æ0 0 0Æ0 0 0Æ0 Excoriation 1 9Æ1 0 0Æ0 0 0Æ0 0 0Æ0 Application-site irritation 0 0Æ0 1 10Æ0 0 0Æ0 0 0Æ0 Application-site reaction 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Application-site vesicles 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Cellulitis 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Heat rash 0 0Æ0 0 0Æ0 0 0Æ0 1 12Æ5 Joint swelling 1 9Æ1 0 0Æ0 0 0Æ0 0 0Æ0 Limb discomfort 1 9Æ1 0 0Æ0 0 0Æ0 0 0Æ0 Skin chapped 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Unanticipated in severity 0 0Æ0 0 0Æ0 1 9Æ1 0 0Æ0 Back pain 0 0Æ0 0 0Æ0 1 g 9Æ1 0 0Æ0 Bold indicates patient withdrawn. a Multiple responses per patient are possible. b Patient withdrawn from study due to chest discomfort and arm being heavy. c Patient withdrawn from study due to patient being emotionally unable to cope with arm being bandaged. d Patient withdrawn from study because of developing whole body oedema recovering within 6 h after the bandages were removed. The body oedema was assessed as anticipated while dyspnoea, neuralgia and macular rash were assessed as unanticipated. All four adverse reactions started on the same day. e The joint sprain was specified as pain around the thumb resulting from slipped bandages. f The fibrosis was specified as increased fibrosis in the lower arm and wrist. g Patient withdrawn from study because of severe pain radiating from lower back to behind knee; the patient had no history of slipped disk pinched nerve. required (power 80%, a <0Æ05) to detect a difference in the patients with leg lymphoedema. The study indicated high variation in volume reduction during this study. This may be in part due to measurement error of the circumferential measures. In this study all assessors were tested to ensure that their measurement technique was standardized. The reason for higher variation within the Coban 2 treatment groups may be related to the lower degree of experience with the new application technique, compared with a frequently used short-stretch bandage. There is some anecdotal evidence to support this with comments from therapists, who felt more competent using the new bandage system as they gained more experience through the course of the study. Slippage can be related to several circumstances including failure of the bandage, high volume reduction or poor technique. In this study the Coban 2 system applied twice weekly showed a slippage of 6Æ0 cm for legs whereas the five times per week group had the lowest slippage of 3Æ7 cm. However, although the slippage was higher for the twiceweekly application, a large mean leg volume reduction of 10% was achieved within in the first 4 days. This remarkable volume loss, obtained by the initial bandage application, which remained on the leg for 4 days, supported the excellent product properties. For patients with arm bandaging, lower slippage values of between 1Æ4 and 2Æ3 cm were observed for all treatment groups. A volume loss of 4% for the first bandage interval also

8 Lymphoedema bandaging, C.J. Moffatt et al. 631 Table 7 Itching and pain Coban 2 5 week Coban 2 3 week Coban 2 2 week Short-stretch 5 week n % n % n % n % Arm group Itching Improvement 2 18Æ2 1 10Æ0 0 0Æ0 0 0Æ0 Unchanged 5 45Æ5 7 70Æ0 6 54Æ5 6 60Æ0 Deterioration 4 36Æ4 2 20Æ0 5 45Æ5 4 40Æ0 Pain Improvement 3 27Æ3 2 20Æ0 1 9Æ1 3 30Æ0 Unchanged 7 63Æ6 6 60Æ0 6 54Æ5 3 30Æ0 Deterioration 1 9Æ1 2 20Æ0 4 36Æ4 4 40Æ0 Leg group Itching Improvement 1 9Æ1 0 0Æ0 2 18Æ2 0 0Æ0 Unchanged 8 72Æ7 8 80Æ0 8 72Æ Æ0 Deterioration 2 18Æ2 2 20Æ0 1 9Æ1 0 0Æ0 Pain Improvement 2 18Æ2 4 40Æ0 4 36Æ4 4 50Æ0 Unchanged 8 72Æ7 5 50Æ0 6 54Æ5 4 50Æ0 Deterioration 1 9Æ1 1 10Æ0 1 9Æ1 0 0Æ0 demonstrated an excellent outcome for the twice-weekly Coban 2 arm group. In general, movement is an important prerequisite for oedema reduction. With bandaging in place, mobility can be compromised compared with the unbandaged condition. In contrast, overall and limb mobility continuously increased in the Coban 2 system for both arm and leg strata. For limb mobility, this was especially evident for the leg patients. The thinner profile and lower weight of the Coban 2 system resulted in a greater ability of the patients to wear their own clothes and shoes, which should have enabled them to be more physically active. This could again be a condition for more effective oedema reduction under the inelastic bandage sleeve, which further leads to increased mobility. More than 95% treatment compliance was achieved over a treatment period of 19 days (three working weeks) for all patients. Adverse events were equally distributed across all groups and reflected the spectrum of already known sideeffects of compression therapy such as local discomfort and mild skin irritation. Taking into consideration the good safety profile under all treatment groups of the Coban 2 system on one side and effective oedema reduction on the other side it would appear that a twice-weekly application is probably at least as good as a more frequent reapplication. The study results have indicated a good clinical benefit when applying the bandage twice weekly. However, these results need to be substantiated by an appropriately powered randomized controlled trial. At this stage it is uncertain whether the twice-weekly bandage application does give the best volume reduction, which appears to be at odds with clinical knowledge of compression application. Some justification for this is that both the arm and leg subgroups experienced a similar effect. Clearly definitive evidence of this can only be provided in a larger trial of bandage change frequency. However, if correct, this reduced frequency of bandage change is likely to reduce the burden on the healthcare system and may enable the lymphoedema centres to treat more patients. Typically, compression bandaging is used for the intensive phase of lymphoedema management. 2,8 In the maintenance phase, stockings are the principal method of compression application, which allows the patient to be independent from frequent clinic visits. This trial evaluated only the acute phase of the management of patients with lymphoedema, without long-term follow-up. Clearly there is a need to determine the benefits of this reduction in limb volume following this acute phase in the maintenance of oedema reduction throughout a prolonged follow-up period. If the clinical benefits of a twice-weekly application of the Coban 2 system are confirmed, this regimen may be considered for patients in the maintenance phase of their treatment, either in combination with stockings or as a replacement, where stockings are considered problematic. This might be especially considered for patients who are not able to apply the stockings or where there is evidence of skin reactions to the hosiery. Although the sample sizes were low for the individual treatment groups the trial has demonstrated high oedema volume reduction with the Coban 2 system. The new two-layer system does not need additional padding. This thin and flexible compression system has obvious potential advantages for patients in reducing the impact of the bandaging on their lives. The lower number of clinic visits could have positive effects on total treatment costs, which requires further evaluation. 9

9 632 Lymphoedema bandaging, C.J. Moffatt et al. What s already known about this topic? In chronic lymphoedema of the extremities, bandaging together with exercise and manual lymph drainage and skin care is the standard intensive therapy. Once lymphoedema has been reduced significantly, hosiery is used in the maintenance phase. What does this study add? Indicates the optimal treatment regimen of a new bandaging system. Comparison against traditional bandage systems. Data on volume reduction over time. Measurement of bandage slippage. Change in mobility. Safety data. Acknowledgments We thank the patients and staff who undertook this study. The authors acknowledge the collaboration with the other participating centres: St Oswald s Hospice Lymphoedema Clinic, Gosford, Newcastle-upon-Tyne, U.K.; St Giles Hospice Lymphoedema Service, Whittington, Lichfield, U.K.; Loros Hospice, Leicester, U.K.; London Road Clinic, Milborne Port, Dorset, U.K.; Enfield Macmillan Lymphoedema Service, London, U.K.; University of Pennsylvania, Philadelphia, PA, U.S.A.; Memorial Hermann Hyperbaric Center, Houston, TX, U.S.A. References 1 Moffatt CJ, Franks PJ, Doherty DC et al. Lymphoedema: an underestimated health problem. QJM 2003; 96: Badger CMA, Peacock JL, Mortimer PS. A randomized, controlled, parallel-group clinical trial comparing multilayer bandaging followed by hosiery versus hosiery alone in the treatment of patients with lymphoedema of the limb. Cancer 2000; 8: McNeely ML, Magee DJ, Lees AW et al. The addition of manual lymph drainage to compression therapy for breast cancer related lymphedema: a randomized controlled trial. Breast Cancer Res Treat 2004; 86: Badger C, Preston N, Seers K, Mortimer P. Physical therapies for reducing and controlling lymphoedema of the limbs. Cochrane Database Syst Rev 2004; 4:CD Mayrovitz HN. The standard of care for lymphedema: current concepts and physiological considerations. Lymphat Res Biol 2009; 7: European Wound Management Association. Lymphoedema Bandaging in Practice. Focus Document. London: MEP Ltd, International Society of Lymphology. The diagnosis and treatment of peripheral lymphedema. Consensus document of the International Society of Lymphology. Lymphology 2003; 36: Vaillant L, Mueller C, Gousse P. Treatment of limbs lymphedema. Presse Med 2010; 39: Partsch H, Stout N, Forner-Cordero I et al. Clinical trials needed to evaluate compression therapy in breast cancer related lymphedema (BCRL). Proposal from an expert group. Int Angiol 2010; 29: