Aerobic fitness in children with cerebral palsy compared to typically developing peers: A systematic review and meta-analysis.

Sixteen of the 36 studies shared participants, so the same children appear in more than one paper. The reviewers removed the duplicates they could identify, but some pooled figures rest on fewer unique children than the study count suggests. The authors could not use a standard risk-of-bias tool or a GRADE rating. Quality was judged instead on whether each study defined how it measured peak oxygen uptake and reported who completed a maximal test, so the usual quality safeguards are missing. The included studies differed a lot from one another: different populations, different exercise protocols, and testing in both laboratories and the field. Some studies included children who did not meet the criteria for a true peak, and reaching a maximum is hard for children, so the pooled values are likely an underestimate of true aerobic capacity. The figure for GMFCS III/IV comes from a single study, and no child at GMFCS V could perform a maximal test at all, so the most disabled children are barely represented. The difference between boys and girls with cerebral palsy had a confidence interval that crossed zero, so it is not a settled finding. Only group averages were available, so child characteristics could not be combined in one model to work out which ones matter most.

The authors declare no competing interests.

Do not conclude from this paper that boys with cerebral palsy are more affected than girls. That claim rests on a subgroup analysis whose difference between boys and girls was 4.79 mL/kg/min with a confidence interval running from −8.10 to 17.68, which includes the possibility of no difference at all. The single pooled value of 32.84 mL/kg/min is a group average from overlapping study samples, not a threshold to judge an individual child against.