The methodology of this review is sound, but the trials included were
small and too heterogeneous to pool for statistical analysis. One
might therefore question the power of the study, but, as the reviewers
surmised, the differences that were not significant were so small
that they were not clinically relevant.