Knols, R., Aaronson, N. K., Uebelhart, D., Fransen, J., & Aufdemkampe, G. (2005). Physical exercise in cancer patients during and after medical treatment: a systematic review of randomized and controlled clinical trials. Journal of Clinical Oncology, 23, 3830–3842.

Databases searched were MEDLINE, CINAHL, Cochrane Library, CANCERLIT, and PEDro through June 2004 to identify randomized, controlled trials and controlled trials (those with a comparison group but without explicit use of randomization for purposes of group allocation).

To be included in the review, the trials had to have examined the effects of physical exercise after surgery or during or after chemotherapy, radiotherapy, and/or hormonal therapy. Only exercise interventions designed to improve endurance or muscular strength were included.

Studies of relaxing exercises (e.g., yoga or tai-chi) were excluded.

The methodologic quality (using the Delphi criteria list—a set of nine criteria for quality assessment of clinical trials) and substantive results of 34 randomized, controlled trials and controlled trials was examined. Of the 34 studies examined, 22 examined the effectiveness of physical exercise during medical treatment, whereas 12 focused on the period after medical treatment.

Outcomes were fatigue, health-related quality of life, symptom distress, psychological distress, body composition, physical exercise capacity (maximal oxygen consumption [VO₂] max), self-reported exercise/physical activity level, and other physical performance measures, such as walk time. Various physical exercise modalities were used, differing in type (walking, cycling, swimming, resistive exercises, or combined exercises), intensity (with most programs at 50% to 90% of the estimated VO₂ maximum heart rate), frequency (ranging from two times per week to up to two times daily), and duration (ranging from two weeks up to one year). In some studies, the experimental group was compared with a group that received some form of training of a lesser intensity, frequency, and/or duration (e.g., stretching, self-directed exercises, strength exercises, aerobic exercise of a lesser intensity, swimming, behavioral therapy). In other studies, the comparison group did not receive any exercise program or advice, was on a waiting list, or participated in a cross-over trial.

The studies during medical treatment were divided into three subcategories: (1) exercise during breast cancer treatment, (2) exercise during bone marrow and peripheral blood stem cell transplantation, and (3) exercise during medical treatment for mixed solid tumors. The studies after medical treatment were divided into those involving exercise after breast cancer treatment and exercise after medical treatment for other solid tumors. The authors used this strategy to reflect not only differences in cancer diagnosis and the timing of physical exercise programs, but also possible differences in motivation, safety, feasibility and efficacy of exercise. The sample sizes for the intervention groups ranged from 12 to 188 participants.

A clinically significant or statistically significant positive effect of physical activity specifically on fatigue was noted during breast cancer treatment  (three studies) or after breast cancer treatment (two studies), and during treatment (three studies) or after treatment (one study) in a mixed solid tumor population. The median quality criteria score on the Delphi list (range 1–7) was four for studies of exercise during and after cancer treatment. Twenty-five of the trials satisfied more than three criteria on the Delphi criteria list. The most commonly observed methodologic problems were with concealment of treatment allocation, blinding of the outcome assessor, and failure to use an intention-to-treat data analysis strategy.

Overall, the authors concluded that the included trials were of moderate methodologic quality, with a trend toward more methodologic rigor in more recent studies.