Background: 

Hip fractures are a leading cause of disability worldwide, with displaced femoral neck fractures being of particular concern. A recent meta-analysis reported that total hip arthroplasty (THA) was superior to hemiarthroplasty (HA) in terms of reoperations, but inferior in terms of dislocations. However, publication of 4 additional randomized controlled trials that enrolled nearly 1,780 additional patients merits an updated meta-analysis.

Methods: 

We conducted a literature search of 4 databases to identify randomized controlled trials comparing THA and HA in patients with displaced femoral neck fractures. For patient-reported outcomes, the minimally important difference informed calculation of risk differences. We performed a subgroup analysis to address the possible impact of risk of bias and performed meta-regression to assess the possible impact of duration of follow-up.

Results: 

Sixteen studies that enrolled 3,084 patients randomized to undergo THA (n = 1,521) or HA (n = 1,563) proved eligible. There were no significant differences between the 2 groups in terms of the revision rate at up to 5 years of follow-up or functional outcome at up to 3 years. Health-related quality of life was superior in the THA group (mean difference [MD] = 0.05, 95% confidence interval [CI] = 0.02 to 0.07, minimally important difference, 0.145). There was no significant difference between the groups in terms of dislocation or periprosthetic fracture incidence. Operative time was significantly shorter in the HA group (MD = 22 minutes, 95% CI = 9 to 35 minutes). Analyses addressing risk of bias and length of follow-up did not reveal subgroup differences. Certainty of evidence for all outcomes was rated as moderate.

Conclusions: 

The best evidence showed, with moderate certainty, that HA and THA likely result in similar revision rate, function, mortality, periprosthetic fracture, and dislocation at up to 5 years, with a small, possibly unimportant benefit in health-related quality of life with THA. More specifically, the improvements are well below established cutoffs for clinical importance. Almost half of all patients were from a single large randomized controlled trial, although the results were consistent across the studies. In addition, HA likely results in a clinically unimportant reduction in operative time.

Level of Evidence: 

Therapeutic Level I. See Instructions for Authors for a complete description of levels of evidence.

中文翻译：

---

全髋关节置换术与置换术治疗股骨颈骨折：系统评价和荟萃分析的随机对照试验。

背景： 

髋部骨折是全球范围内致残的主要原因，尤其是移位的股骨颈骨折。最近的一项荟萃​​分析报告说，就再手术而言，全髋关节置换术（THA）优于半髋置换术（HA），但在脱位方面则较差。然而，另外4项随机对照试验的发表使另外1780名患者入选，值得进行更新的荟萃分析。

方法： 

我们对4个数据库进行了文献检索，以识别比较股骨颈骨折患者THA和HA的随机对照试验。对于患者报告的结果，最小重要差异可指导风险差异的计算。我们进行了亚组分析以解决偏倚风险的可能影响，并进行了荟萃回归以评估随访时间的可能影响。

结果： 

16项研究纳入3084例随机接受THA（n = 1,521）或HA（n = 1,563）的患者，证明是合格的。两组之间的随访率（随访至5年）或功能预后（随访至3年）无明显差异。THA组与健康相关的生活质量更高（平均差异[MD] = 0.05，95％置信区间[CI] = 0.02至0.07，最小重要差异0.145）。在脱位或假体周围骨折发生率方面，两组之间无显着差异。HA组的手术时间明显缩短（MD = 22分钟，95％CI = 9至35分钟）。关于偏倚风险和随访时间的分析并未揭示亚组差异。所有结果的证据确定性为中等。

结论： 

最好的证据表明，HA和THA可能在长达5年的时间内导致相似的翻修率，功能，死亡率，假体周围骨折和脱位，对健康相关的生活质量影响很小，甚至可能不重要THA。更具体地说，对于临床重要性而言，这些改进远低于既定的标准。尽管所有研究的结果都是一致的，但几乎所有患者中有一半来自一项大型随机对照试验。此外，HA可能会导致临床上不重要的手术时间减少。

证据级别： 

治疗I级。有关证据水平的完整说明，请参见《作者说明》。