Treatment of Breast Cancer-related Lymphoedema

Patients with unilateral arm lymphoedema were enrolled and all were hospitalized. The participating patients were randomly divided into two groups.

Demographic features of the patients including age, body mass index [BMI = weight (kg) / height2 (m2)], duration of lymphoedema (months) and number of axillary lymph nodes removed were all recorded. The volume (V) of every part of the limb was calculated by the truncated cone formula according to circumferential measurements and the total volume is estimated by the sum of the increments. Circumference measurements were taken at 4-centimetre intervals along the arm from the level of the ulnar styloid to the shoulder with a flexible non-stretch tape measure. The volume between every two circumferential measurement levels was calculated using the following formula:

Volume = h (C2 + Cc + c2) / 12π In this formula, C is the first circumferential measurement (circumference of the top of the cone), c is the second circumferential measurement (circumference of the base of the cone) and h is the height from measurement C to c. Both the affected and non-affected limbs were evaluated by the same examiner. An Excel-based software program was used to convert these values into limb volumes in mL.

The severity of lymphoedema was characterized by the percentage excess volume (PEV) PEV = [(VLE - VH) / VH] x 100 where VLE denotes the volume of the lymphoedema arm and VH is that of the healthy arm. PEV is the preferred quantity for defining the severity of lymphoedema than the absolute difference volume. Also, this method minimizes the effects of BMI on the volume estimates. The efficacy of complex decongestive therapy, the response to the therapeutic intervention, was also quantified by the percentage reduction of excess volume (PREV) PREV = [(pre-treatment VLE - post-treatment VLE) / excess volume)] x 100. Other assessments of clinical origin were also conducted on the patients by physical therapists who were blind to the applied treatment and the timing of the data collection point. Pitting oedema was evaluated at multiple points on the forearm by pressing the thumb on the region to be investigated for a minute as hard as possible, as such test is considered as an indicator of the presence of excess free fluid accumulated in the superficial interstitial tissue spaces. The Stemmer's sign was performed as an inability to pinch the fold of skin at the base of the finger.

Clinical symptoms of pain (during activity, resting and night), heaviness and tightness in the affected arm were evaluated with Visual Analog Scale (VAS) with a 0-10 numerical rating scale. The patients' functional situation was assessed by the Quick Disabilities of the Arm, Shoulder and Hand (QuickDASH) outcome questionnaire which is a self-reported assessment tool for the measurement of physical function and symptoms in individuals with a musculoskeletal disorder of the upper limb. This instrument was also reported as a convenient, reliable, and valid patient-reported outcome measure to assess upper extremity disability in patients with breast cancer. The scores indicated the level of disability and severity, ranging from 0 (no disability) to 100 (most severe disability).

Grip strength was measured by using Jamar® hydraulic hand dynamometer (Sammons Preston, Bolingbrook, IL, USA). Both hands of the patient's grip strength were assessed by using the second level of resistant (3.75 cm) as the elbow is at 90º flexion, the forearm is in a neutral position. All measurements were performed bilaterally and three times. The average of the obtained values was recorded in kg. Beck Depression Inventory, a 21-item, self-administered inventory that measures clinical depression, was also administered to the patients. Scores ranged from 0 to 63, and higher scores were indicative of depression. The validity and reliability of the Beck Depression Inventory for the Turkish population have already been tested previously.

Kolmogorov-Smirnov test was used to test whether the data were normally distributed. All of the measured variables failed the normality test and thus median (25-75 percentiles) were given for nonparametric descriptive statistics. Wilcoxon T test was used for the intragroup comparison of the measurements between before and after the treatments. Mann-Whitney U-test was used for the intergroup comparison of the measurements after the treatments. Categorical variables were shown as number (%) and Chi-square tests were used for the comparison of categorical variables. P<0.05 was accepted as an indicator of the statistically significant difference.