RESUMO
INTRODUCTION: Body composition is one of the main variables of interest in clinical practice in cancer patients. Specific markers from bioelectrical impedance analysis, such as phase angle (PhA), have been assuming increasing relevance in this population. The aim of the present systematic review was to study and systematise the effect of exercise on PhA in cancer survivors, as compared to control conditions, namely usual care, with no exercise. The effect of exercise on PhA in the population of cancer survivors is not yet established. EVIDENCE ACQUISITION: This systematic review was conducted on October 13, 2021, through PubMed, Web of Science, Wiley Online Library, Directory of Open Access Journals, Science Direct and JSTOR, following PRISMA guidelines and PICOS model that include: cancer survivors with ≥18 years; intervention of any exercise program for the target group; comparison between intervention group and control group that followed control conditions, namely usual care, with no exercise; outcome related to PhA; and studies of randomized control trials. EVIDENCE SYNTHESIS: We founded a total of 1244 publications, using selected keywords. Eight studies were included in this systematic review, after inclusion/exclusion criteria considered. Compared with the control conditions, exercise training programs seem associated with a positive effect on PhA, both in solid tumours and haematologic cancer types, but only when using resistance exercise alone. According to the best evidence synthesis criteria, we could not conclude the superiority of any exercise program analysed in the value of the PhA. CONCLUSIONS: There were several exercise details that may have potential to be beneficial for PhA in cancer patients, including an early start of the exercise intervention (during treatment and immediately after discharge from hospital), the use of resistance exercise or/and aerobic exercise, and mainly a long follow-up period (≥ 4 months) to verify the structural effects of exercise on the PhA. However, there were no effects on the PhA value immediately after the intervention, regardless of the type of protocol and the intervention time.