A total of 22 studies were included in the review (n=1,095, range 12 to 185). Functional outcome was assessed in 16 studies (12 retrospective case-series, three prospective case-series and one randomised controlled trial). Radiographic outcome was assessed in 10 studies (eight retrospective case-series, one prospective case-series and one cohort study). Publication bias was present for both outcomes.
Functional outcome: The pooled success rate was 83.9% (15 studies, n=665); significant heterogeneity was present for this analysis (I2=69.6%). Subgroup analysis found no significant difference in clinical outcome between those patients treated with a brace to those treated without.
Radiographic outcome: The pooled success rate was 28.0% (10 studies, n=847); significant heterogeneity was present for this analysis (I2=85.7%). Subgroup analysis showed that: unilateral defects healed significantly more than bilateral defects (71% versus 18.1%, p<0.0001); and acute defects healed at a rate of 68.1%, significantly more than progressive lesions (28.3%, p<0.0001) and terminal lesions (1.8%, p<0.0001)