Gothard, L., Haviland, J., Bryson, P., Laden, G., Glover, M., Harrison, S., . . . Yarnold, J. (2010). Randomised phase II trial of hyperbaric oxygen therapy in patients with chronic arm lymphoedema after radiotherapy for cancer. Radiotherapy and Oncology: Journal of the European Society for Therapeutic Radiology and Oncology, 97(1), 101–107.

To assess the effectiveness of hyperbaric oxygen (HBO) therapy on reducing arm lymphedema in patients with early-stage breast cancer

Participants were divided into the control and the experimental group at a ratio of 1 to 2, respectively. The experimental group received HBO therapy and breathed 100% oxygen at 2.4 atmospheres absolute for 100 minutes with two five-minute air breaks. The sessions were conducted 30 times in a six-week period or five times a week. Both the control and experimental groups received patient education on standard care for lymphedema and hosiery when appropriate.

• The study sample was comprised of 58 patients.
• Mean age of the control group was 62.1 years and the experimental group was 63.2 years.
• All patients had breast cancer and most (97%) had diagnosed lymphedema.
• Two patients had Hodgkin lymphoma with radiotherapy.
• Patients had finished adjuvant radiotherapy.
• Patients had a minimum of 15% increase in limb volume.

The study took place in multiple hospitals in the United Kingdom.

Patients were undergoing active lymphedema treatment.

The study used a randomized phase II study design.

• A Perometer measured limb volume.
• An 8% reduction was maintained as a clinically significant reduction in arm volume.
• Lymphoscintigraphy was performed to measure the fractional removal rate of the radioisotopic tracer at baseline and 12 months.
• Acquisitions were performed at 20, 60, 90, 120, and 180 minutes.
• The removal rate was calculated by the computer using a regression slope.
• Extracellular water content was measured at baseline and 12 months using dielectric constant measurements.
• Two measurements were taken: one third the distance from the antecubital crease to the forearm and one half the distance from the antecubital crease to the upper arm.
• Quality of life was assessed using the United Kingdom SF-36, completed before randomization into groups, at baseline, and at 3, 6, 9, and 12 months.

There was not a statistically significant change in limb volume between the control group (p = 0.64) and the experimental group (p = 0.50) at 12 months after baseline. The investigators define a positive response as an 8% reduction in arm volume. Thirty percent of the experimental group versus 18.8% of the control group responded to meet these criteria but was also statistically insignificant (p = 0.50). Lymphatic clearance rates were similar among groups and were not significant findings. Quality of life findings were similar among both groups and were not significant.
 

The study suggests that HBO therapy when added to best standard treatment of lymphedema in patients with breast cancer is not effective. The study does not confirm earlier reports of a therapeutic effect of HBO.

• The sample size was small, with less than 100 participants.
• The sample size for the control and experimental group were different, which could have affected the statistical analysis comparison and leaves room for imbalances.
• The study had a risk of bias because it had no blinding and the sample characteristics.
• The findings are not generalizable.
• The average time from radiotherapy treatment post-operatively to participation in the study was 12 years, which may also effect the results of remodeling of fibrotic tissues from HBO.

The study suggests that HBO therapy is an ineffective therapy for treating lymphedema in patients with breast cancer. The therapy should not be enacted into practice. Nursing researchers should analyze the randomized trial and non-randomized trials of this treatment program to identify confounding variables that may have made the non-randomized trial results significant and the randomized trial results not significant. If the study is repeated a larger sample size should be used.