The patients were consecutively selected by one physician from the inpatient clinic in the Department of Rehabilitation Medicine at Kosin University Hospital, Korea, from January 2010 to December 2010. Cancer survivors, who were diagnosed with malignant lymphedema by a physician, were included. The criteria used for the diagnosis of malignant lymphedema in cancer survivors were 1) a greater than 2 cm circumference difference between the affected upper or lower limb and the normal upper or lower limb; 2) a diagnosis of lymphedema via lymphoscintigraphy (decreased uptake or no uptake by the lymph nodes, dermal backflow, poor to no visualization of the collateral and main lymphatics, and decreased or no clearance of radioisotope from the injection site) [7]; 3) lymphatic channel or lymph node invasion in the axilla, the pelvis, or the lumbar or inguinal areas diagnosed via imaging study; and 4) limb swelling that appeared to be aggressively and rapidly progressing. Exclusion criteria included 1) age older than 80 years, 2) bilateral lymphedema of the upper or lower limbs, 3) patients with benign lymphedema before malignant metastasis, 4) patients with induced lymphedema during radiotherapy or chemotherapy, 5) vascular disorders that might cause lymphedema, and 6) poor communication skills or low treatment compliance.

We recruited 33 patients, who had never received treatment for their lymphedema. The patients provided informed consent, and the study protocol conformed to the ethical guidelines of the Declaration of Helsinki. The patients completed a survey in which the purpose of the study was explained to them. Those who agreed to participate were included. Eleven patients were excluded during this process, leaving a total of 22 patients who were selected to proceed with the study.

All of the 22 patients received the following components of CDPT: non-elastic bandage compression therapy, remedial exercise, and skin care. Only the manual lymphatic drainage (MLD) was not utilized. Therapy was provided five times a week for two weeks by one physical therapist. Drugs for anticancer treatment and pain were allowed, but were not changed during the two weeks of the study.

We measured the volume of the upper or the lower limbs at 3 cm intervals along the axis of ordinates of each extremity. This axis was based on the anatomic reference line that connects the bilateral epicondyles or condyles of the elbow and knee. The proximal part of the upper and lower limbs was measured from the anatomic reference line to levels of the axilla and pubic symphysis. And distal portion was measured from the anatomic reference line to as the hands and feet area distal 6 cm the medial and lateral styloid process of the wrist and malleolus of the ankle. The circumference measurements obtained in this manner were then used to calculate the volume of each segment, which was treated as a truncated cone using the following formula: V=h(C1²+C1C2+C2²)/12π [8]. The sum of the volumes of the proximal and distal portions of the upper or lower limbs was taken as the total volume. The volume measurements were done twice-before and after the two weeks of CPDT.

Patient QOL was also assessed at the measurement visit. The Korean version of the Medical Outcomes 36-item Short Form Health Survey (SF-36) 0-100 score was used. The SF-36 includes eight multi-item scales that contain two to ten items each, plus a single item to assess the state of health. The multi-item scales cover the dimensions of physical functioning, physical role functioning, bodily pain, general health perceptions, mental health, emotional role functioning, social role functioning, and vitality [9]. The multi-item scales are separated into the physical component summary (PCS), which includes physical functioning, physical role functioning, bodily pain and general health perceptions, and the mental component summary (MCS), which includes mental health, emotional role functioning, social role functioning, and vitality. These summaries were used to confirm changes that occurred as a result of treatment.

We used the ten-point visual analog scale (VAS) to evaluate pain intensity. Patients were asked to indicate the intensity of pain in the limbs with lymphedema. The pain intensity was measured before and after two weeks of CDPT.

SPSS ver. 14.0 (SPSS Inc., Chicago, IL, USA) was used for all analyses. A paired t-test was employed to assess the significance of differences in volume of the upper and lower limbs (total, proximal and distal volume), intensity of pain and QOL before and after CDPT. And Student t-test was applied to confirm differences of volume change in the proximal and distal parts of the upper and lower limbs, respectively. In parts of PCS and MCS, we used the paired t-test before and after CDPT. In order to confirm change of PCS and MCS depending on the upper or lower limbs, Student t-test was used to compare change in parts of PCS and MCS before and after CDPT. A p-value less than 0.05 was considered statistically significant.

Twenty-two patients were enrolled in this study, and their average age was 60.0 years. There were 7 (31.8%) and 15 (68.2%) patients with malignant lymphedema of the upper and lower limbs, respectively. There were nine patients, who had received radiation therapy. No patient developed an infection related to malignant lymphedema (Table 1).

The total volume of the upper or the lower limb decreased by 3.7% and 10.9%, respectively after two weeks of CDPT (p=0.001, respectively) (Tables 2, 3). Among 22 patients, 19 patients improved in volume, but the volume of the others were increased. The reduction in the proximal area volume was greater than that of the distal area volume in both the upper limbs (p=0.015) and the lower limbs (p=0.016).

Using the VAS, the patients reported an average decrease of 1.6 points compared with pretreatment VAS points. However, one patient reported no change in pain intensity, while another patient reported an increase in pain.

In the QOL evaluation, five items such as physical functioning (p=0.001), physical role functioning (p=0.001), bodily pain (p=0.001), social role functioning (p=0.008), and emotional role functioning (p<0.001) showed a statistically significant improvement. An improvement was also noted when the PCS and MCS components were analyzed separately (p=0.006, p=0.001, respectively) (Table 4). The PCS value decreased in seven patients, and four patients reported on the MCS that their quality of life had decreased. Patients with malignant lymphedema of the upper limb showed more improvement on the MCS than patients with malignant lymphedema of the lower limb (p=0.04). However, there was a greater improvement on the PCS in patients with malignant lymphedema of the lower limb than in patients with upper limb involvement (p=0.03).