Jennifer M Ryan, Elizabeth E Cassidy, Stephen G Noorduyn, Neil E O'Connell

Cerebral palsy (CP) is a neurodevelopmental disorder resulting from an injury to the developing brain. It is the most common form of childhood disability with prevalence rates of between 1.5 and 3.8 per 1000 births reported worldwide. The primary impairments associated with CP include reduced muscle strength and reduced cardiorespiratory fitness, resulting in difficulties performing activities such as dressing, walking and negotiating stairs.Exercise is defined as a planned, structured and repetitive activity that aims to improve fitness, and it is a commonly used intervention for people with CP. Aerobic and resistance training may improve activity (i.e. the ability to execute a task) and participation (i.e. involvement in a life situation) through their impact on the primary impairments of CP. However, to date, there has been no comprehensive review of exercise interventions for people with CP. To assess the effects of exercise interventions in people with CP, primarily in terms of activity, participation and quality of life. Secondary outcomes assessed body functions and body structures. Comparators of interest were no treatment, usual care or an alternative type of exercise intervention. In June 2016 we searched CENTRAL, MEDLINE, Embase, nine other databases and four trials registers. We included randomised controlled trials (RCTs) and quasi-RCTs of children, adolescents and adults with CP. We included studies of aerobic exercise, resistance training, and 'mixed training' (a combination of at least two of aerobic exercise, resistance training and anaerobic training). Two review authors independently screened titles, abstracts and potentially relevant full-text reports for eligibility; extracted all relevant data and conducted 'Risk of bias' and GRADE assessments. We included 29 trials (926 participants); 27 included children and adolescents up to the age of 19 years, three included adolescents and young adults (10 to 22 years), and one included adults over 20 years. Males constituted 53% of the sample. Five trials were conducted in the USA; four in Australia; two in Egypt, Korea, Saudi Arabia, Taiwan, the Netherlands, and the UK; three in Greece; and one apiece in India, Italy, Norway, and South Africa.Twenty-six trials included people with spastic CP only; three trials included children and adolescents with spastic and other types of CP. Twenty-one trials included people who were able to walk with or without assistive devices, four trials also included people who used wheeled mobility devices in most settings, and one trial included people who used wheeled mobility devices only. Three trials did not report the functional ability of participants. Only two trials reported participants' manual ability. Eight studies compared aerobic exercise to usual care, while 15 compared resistance training and 4 compared mixed training to usual care or no treatment. Two trials compared aerobic exercise to resistance training. We judged all trials to be at high risk of bias overall.We found low-quality evidence that aerobic exercise improves gross motor function in the short term (standardised mean difference (SMD) 0.53, 95% confidence interval (CI) 0.02 to 1.04, N = 65, 3 studies) and intermediate term (mean difference (MD) 12.96%, 95% CI 0.52% to 25.40%, N = 12, 1 study). Aerobic exercise does not improve gait speed in the short term (MD 0.09 m/s, 95% CI -0.11 m/s to 0.28 m/s, N = 82, 4 studies, very low-quality evidence) or intermediate term (MD -0.17 m/s, 95% CI -0.59 m/s to 0.24 m/s, N = 12, 1 study, low-quality evidence). No trial assessed participation or quality of life following aerobic exercise.We found low-quality evidence that resistance training does not improve gross motor function (SMD 0.12, 95% CI -0.19 to 0.43, N = 164, 7 studies), gait speed (MD 0.03 m/s, 95% CI -0.02 m/s to 0.07 m/s, N = 185, 8 studies), participation (SMD 0.34, 95% CI -0.01 to 0.70, N = 127, 2 studies) or parent-reported quality of life (MD 12.70, 95% CI -5.63 to 31.03, n = 12, 1 study) in the short term. There is also low-quality evidence that resistance training does not improve gait speed (MD -0.03 m/s, 95% CI -0.17 m/s to 0.11 m/s, N = 84, 3 studies), gross motor function (SMD 0.13, 95% CI -0.30 to 0.55, N = 85, 3 studies) or participation (MD 0.37, 95% CI -6.61 to 7.35, N = 36, 1 study) in the intermediate term.We found low-quality evidence that mixed training does not improve gross motor function (SMD 0.02, 95% CI -0.29 to 0.33, N = 163, 4 studies) or gait speed (MD 0.10 m/s, -0.07 m/s to 0.27 m/s, N = 58, 1 study) but does improve participation (MD 0.40, 95% CI 0.13 to 0.67, N = 65, 1 study) in the short-term.There is no difference between resistance training and aerobic exercise in terms of the effect on gross motor function in the short term (SMD 0.02, 95% CI -0.50 to 0.55, N = 56, 2 studies, low-quality evidence).Thirteen trials did not report adverse events, seven reported no adverse events, and nine reported non-serious adverse events. The quality of evidence for all conclusions is low to very low. As included trials have small sample sizes, heterogeneity may be underestimated, resulting in considerable uncertainty relating to effect estimates. For children with CP, there is evidence that aerobic exercise may result in a small improvement in gross motor function, though it does not improve gait speed. There is evidence that resistance training does not improve gait speed, gross motor function, participation or quality of life among children with CP.Based on the evidence available, exercise appears to be safe for people with CP; only 55% of trials, however, reported adverse events or stated that they monitored adverse events. There is a need for large, high-quality, well-reported RCTs that assess the effectiveness of exercise in terms of activity and participation, before drawing any firm conclusions on the effectiveness of exercise for people with CP. Research is also required to determine if current exercise guidelines for the general population are effective and feasible for people with CP.