Перегляд вихідного коду сторінки Something Nearly everybody Dislikes Concerning ARAF Also As to why

The funnel plot [24] was used to assess publication bias. If there were asymmetrical plots, we used the trim and fill analysis to assess the stability [25]. When allowed, the subgroup analyses were performed in isolation for DTF. We also used the Grading of [http://en.wikipedia.org/wiki/ARAF ARAF] Recommendations Assessment, Development and Evaluation (GRADE) system to evaluate the quality of evidence by each outcome [26]. The literature search initially yielded 6620 relevant studies, from which 2391 redundant publications were excluded. According to our criteria of inclusion and exclusion, 4213 studies were excluded. In the remaining full texts, three studies had data duplication and only the newest one was included [7]. One study carried another study further [8]?and?[9] and we included both. Finally, 14 studies including five prospective trials [8], [9], [10], [11]?and?[12] and 12 retrospective trials [7], [13], [14], [15], [16], [17], [18], [19]?and?[20] [http://www.selleckchem.com/products/GDC-0941.html Pictilisib nmr] with 842 participants were included (Fig. 1). The weighted kappa for the agreement on eligibility between reviewers was 0.84 (95% CI: 0.71�C0.93). No publication bias was found in the funnel plot (Fig. 2). IMN decreased the infection rate by 48% (N?=?695, RR: 0.52 (0.30, 0.89); p?=?0.02), but increased the malunion rate by 147% (N?=?842, RR: 2.47 (1.58, 3.85); p?[http://www.selleckchem.com/products/ldk378.html click here] reduced (p?=?0.04). We also found medium heterogeneity in the secondary surgery rate (I2?=?67%) and the implant removal rate (I2?=?52%). When excluding the data of Seyhan et al. [23], which included two different IMNs, the heterogeneity reduced to 37% (p?=?0.007) and 16% (p?=?0.04), respectively. IMN increased the functional score (N?=?383, SMD: ?0.26 (?0.47 to ?0.06); p?=?0.01) compared with plating. However, no difference was found in the pain score (N?=?245, SMD: 0.23 (?0.23 to ?0.70); p?=?0.33). All the outcomes did not change if the studies with moderate-to-low quality were omitted. We found medium heterogeneity (I2?=?67%) in the pain score. We noted three studies [7], [8]?and?[10] that assessed the ankle pain except for the study by Janssen et al. [14] that assessed knee pain.