Effect of yoga on QOL, cortisol rhythm, and HRV for women with breast cancer undergoing radiotherapy

Reporter: Lara Bonner Millar, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 5, 2011

Presenter: L. Cohen
Presenter's Institution: University of Texas MD Anderson Cancer Center


  • Yoga is an ancient practice that originated in India; this "old" practice has a new following, as complementary medicine utilization has increased in recent years within the general population and among cancer patients.
  • Yoga is both a physical and spiritual experience that may help some cancer patients manage stress and place life in perspective. In recent years, studies have been performed to examine the effects of yoga among cancer patients and survivors; however, these studies have had limitations including the heterogeneity of patient population, various outcome measurements used, and non-standardization of yoga teaching
  • Among breast cancer patients, radiotherapy (XRT) can decrease quality of life (QOL), disrupt cortisol rhythm and contribute to cardiovascular disease. An ongoing need to find interventions to help reduce these side effects exists.
    • Cortisol regulation may be particularly important because higher stress hormone levels have been linked to worse outcomes in patients with advanced breast cancer.
  • The purpose of this study was to conduct a randomized, controlled, clinical trial examining the efficacy of yoga for improving cortisol slope, heart rate variability (HRV) and QOL among breast cancer survivors.


  • 163 patients with stage 0-III disease, and a mean age of 52 years were recruited prior to XRT and randomized to one of three groups: Yoga (YG-n=53) or Stretching (STR-n=56) 3 times a week for 6 weeks during XRT or Waitlist Control (WLC-n=54).
    • The control group received no instructions in yoga or stretching.
    • The stretching and yoga groups had 18 planned sessions each (3 per week) for six weeks.
    • There were roughly equal proportions of stage I, II and III patients enrolled.
  • Cortisol, which peaks in the morning and declines throughout the day, was measured five times per day for five consecutive days during the intervention and each follow-up period.
    • In terms of generating a "cortisol slope," a steeper slope indicated a greater decrease in cortisol, and therefore a greater physiologic impact of the intervention.
  • Self-report measures of fatigue (Brief Fatigue Inventory scale), depression (Center for Epidemiological Studies Depression Scale ), QOL (San Franciso-36), benefit finding (BF), and spirituality (Functional Assessment of Chronic Illness Therapy-Spiritual Well-Being Scale) were completed.
  • Saliva and 10-minute ECGs were collected at baseline, end of treatment, and 1, 3, and 6 months later.
  • The authors examined change from baseline for questionnaires and HRV data and slope analyses for cortisol.


  • By the end of XRT, the YG and STR groups had a reduction in fatigue while the WLC had an increase (YG: -0.23, STR: -0.45, WLC: 0.52; p<.05).
  • Class attendance was high overall with 80% of stretching and yoga participants attending at least 14/18 classes.
  • At 1, 3, and 6 months after XRT, the YG group had a greater increase in SF-36 physical functioning compared to STR and WLC (1 month: 5.8, 2.0, 0.8; 3 months: 6.5, 3.4, -0.2; 6 months: 6.1, 3.4, 1.1; p’s<.05), with differences between STR and WLC at 3 months (p<.02) and similar outcomes for SF-36 general health scores.
  • By 3 and 6 months after XRT, there was a significant increase in BF, the ability to derive meaning from illness, for the YG group (3 months: 3.1, -2.5, -2.5; 6 months: 1.1; -3.9; -4.7; p<.05).
  • There were no differences for spirituality and depression.
  • Cortisol slope was steepest for the YG group compared to STR and WLC (end of XRT: -0.12, -0.08, -0.08; 1 month: -0.12, -0.08, -0.07; p < .01 for all comparisons).
    • Women who practiced yoga also had the largest decline in their cortisol levels across the day, indicating that yoga had the ability to improve regulation of stress hormone.
  • Within group changes of the standard deviation of all n-n intervals (SDNN) from baseline to the end of XRT revealed significant increases in HRV for YG (p<.05; 64.3ms) but not for STR or WLC (5.8ms, -8.1ms).
  • At one, three and six months after radiation therapy, women who practiced yoga during the treatment period reported greater benefits to physical functioning and general health. They were also more likely to perceive positive life changes from their cancer experience than either of the other groups.
  • In the yoga group, the participants who attended the highest number of classes had continuous improvement in physical functioning scores. Six months post- yoga, those who attended the most classes continued to have improvement.

Author's Conclusions

  • While stretching improved fatigue and physical functioning, yoga buffered changes associated with XRT in terms of fatigue, QOL and benefit finding, and resulted in steeper cortisol slopes, and increased HRV.
  • This is the first yoga study to include an active control group, suggesting that the benefits of yoga are due to more than simple stretching, social support or other indirect effects.
  • The next step is a phase III study with a goal of recruiting 600 patients to a single blinded study with 3 arms: yoga, stretching, and relaxation controls; doing a single blinded study should diminish the possibility that an effect in favor of yoga is due to the placebo effect, since all participants will know they are going to be assigned to an intervention arm, though they will not know what the other arms are.

Clinical Implications

  • The transition from active therapy back to everyday life can be stressful. Teaching patients a mind-body technique like yoga as a coping skill may make the transition less difficult. However, patients in this study group may be self-selected or interested in yoga or other interventions to help QOL, which may bias them to demonstrate a greater benefit in QOL and fatigue scores.
  • Having said this, the change seen in cortisol slope is unlikely to be influenced by a patient's perceptions of the intervention. The examination of cortisol and other biomarkers in important in supportive care research, since such research has been frequently discounted for its use of "soft" endpoints.
  • The intervention was given for only 6 weeks. It is unclear if this amount of yoga would have continued effects beyond 6 months, or if patients would need to continue the intervention for a longer period of time to sustain benefit.
  • There is a growing body of literature on symptom management for cancer patients. Because patients rarely have one isolated symptom or need, interventions like yoga that can target multiple outcomes may have a great impact on symptom management. To that end, economic outcomes may be important: does yoga intervention decrease long-term health care utilization? Is the cost and availability of yoga an obstacle to its use? Patient risk, patient response, and patient treatment preference should play into the determination of appropriate supportive care interventions as we seek to further personalize medicine and supportive care.


Click on any of these terms for more related articles