Poor Outcomes of Children and Adolescents with Femoral Neck Fractures: A Meta-Analysis Based on Clinical Studies

To comprehensively assess the differences in outcome between open reduction and closed reduction for children and adolescents with femoral neck fractures.

Based on the predetermined strategies, eligible studies were obtained by searching Embase, the Cochrane Library, and PubMed databases (retrieval time: June 2018) and through manual retrieval for paper documents. The 95% confidence intervals (CI) and risk ratios (RR) were used as evaluation indexes. Moreover, the results of avascular necrosis, coxa vara, or non-union were compared between open reduction and closed reduction under random or fixed effects models. After sensitivity analysis was carried out, publication bias was evaluated for the eligible studies using Egger's test.

Six studies were included in our meta-analysis. No significant heterogeneity was found among the included studies (P ≥ 0.05) and, thus, the fixed effects model was used for merging the effect sizes of avascular necrosis (RR [95% CI] = 0.50 [0.26, 0.98], P = 0.04), coxa vara (RR [95% CI] = 0.16 [0.04, 0.70], P = 0.01), and non-union (RR [95% CI] = 0.22 [0.05, 0.93], P = 0.04). Sensitivity analysis suggested that the results of avascular necrosis were not stable (RR = 0.50, 95% CI = 0.25 1.17, P = 0.12), while those of coxa vara and non-union were stable. There was no significant publication bias among the eligible studies (t = -0.70, P = 0.522).

Femoral neck fractures treated by open reduction had less adverse outcomes compared with those treated by closed reduction.