Background: We quantified baseline and observed change in peak VO(2) , quality of life, cardiac function, strength and energy intake following exercise training in haemodialysis patients and optimal exercise delivery for producing greatest adherence, safety and patient improvements.

Methods: A systematic literature search was completed in August 2010 to identify randomized, controlled trials of exercise training studies in haemodialysis patients. A subsequent meta-analysis was conducted and the search repeated in December 2010.

Results: Fifteen studies, yielding 565 patients were included. Baseline, peak VO(2) values were 70% of age-predicted values, exercise intervention patients improved post-training peak VO(2) to 88% predicted. Exercise training produced mean 26 ± 12% improvements in eight studies that reported peak VO(2) , mean difference 5.22 mL O(2) /kg per min (95% confidence interval 3.86, 6.59, P < 0.00001). Equivocal results for change in short-form 36 health questionnaire scores were reported post-training. Heart rate variability was improved after exercise training of normal to normal interval, mean difference 1634 milliseconds (95% confidence interval 8.3, 24.3, P < 0.0001). Significant improvements in lean body mass, quadriceps muscle area, knee extension, hip abduction and flexion strength were also reported (all P < 0.0001). Exercise training appears safe, with no deaths directly associated with exercise in 28 400 patient-hours and no differences in withdrawal rates between exercise and control participants, P = 0.98. Exercise training for 6 months or more conveyed larger improvements in peak VO(2) than shorter programmes. Data indicate about 25% of patients were excluded from exercise training studies for medical reasons.

Conclusion: Exercise training is safe and imparts large improvements in peak VO(2) , and heart rate variability.

© 2011 The Authors. Nephrology © 2011 Asian Pacific Society of Nephrology.