Background Chronic Kidney Disease (CKD) patients frequently develop life-impairing bone mineral disorders. Despite the reported impact of exercise on bone health, systematic reviews of the evidence are lacking. This review examines the association of both physical activity (PA) and the effects of different exercise interventions with bone outcomes in CKD. Methods English-language publications in EBSCO, Web of Science and Scopus were searched up to May 2019, from which observational and experimental studies examining the relation between PA and the effect of regular exercise on bone-imaging or -outcomes in CKD stage 3–5 adults were included. All data were extracted and recorded using a spreadsheet by two review authors. The evidence quality was rated using the Cochrane risk of bias tool and a modified Newcastle-Ottawa scale. Results Six observational (4 cross-sectional, 2 longitudinal) and seven experimental (2 aerobic-, 5 resistance-exercise trials) studies were included, with an overall sample size of 367 and 215 patients, respectively. Judged risk of bias was low and unclear in most observational and experimental studies, respectively. PA was positively associated with bone mineral density at lumbar spine, femoral neck and total body, but not with bone biomarkers. Resistance exercise seems to improve bone mass at femoral neck and proximal femur, with improved bone formation and inhibited bone resorption observed, despite the inconsistency of results amongst different studies. Conclusions There is partial evidence supporting (i) a positive relation of PA and bone outcomes, and (ii) positive effects of resistance exercise on bone health in CKD. Prospective population studies and long-term RCT trials exploring different exercise modalities measuring bone-related parameters as endpoint are currently lacking.