Moon Edward S, Mehlman Charles T
Musculoskeletal Outcomes Research Fellow, Cincinnati Children's Hospital Medical Center, University of Cincinnati College of Medicine, Cincinnati, OH 45229, USA.
J Orthop Trauma. 2006 May;20(5):323-9. doi: 10.1097/00005131-200605000-00005.
Avascular necrosis (AVN) is one of the most serious complications of femoral neck fractures in children. Variability in the reported rates of AVN and lack of statistical evidence has minimized the prognostic power of individual studies. The purpose of our study was to review our own cases as well as those from the literature in an effort to identify the risk factors for AVN.
We conducted a computerized search of medical records to identify all children with femoral neck fractures treated at our institution from 1980 to 2003. Records were reviewed to determine fracture type (Delbet type I to IV), displacement, age, treatment, and incidence of AVN. A structured search of PubMed was also performed to identify all reports published in English, on femoral neck fractures in children, from 1965 to 2003. Hand searches of major orthopedic journals and reference lists of publications identified additional cases. Of the 275 citations initially identified through our computerized search, 47 met the criteria for further evaluation. These citations were reviewed by 2 investigators, and data was abstracted from 20 reports that provided patient-level data and met our criteria for inclusion.
Twenty-four patients with 25 femoral neck fractures from our institution were identified including 12 boys and 12 girls with an average age of 8 years (range 1.5 to 16). Over 300 patients who met our study criteria were also identified from the literature, and a total of 360 patients were included in the analysis. Fracture type, displacement, age, and treatment were all statistically significant independent predictors of AVN with P values <or=0.05. With logistic regression analysis, however, fracture type and age were identified as the only significant predictors of AVN. Older children were 1.14 times more likely to develop AVN for each year of increasing age. Type I to III fractures were 15, 6, and 4 times, respectively, more likely to develop AVN than type IV fractures. AVN rate by Delbet class was I=38%, II=28%, III=18%, and IV=5%.
Although several factors may contribute to the development of AVN, our meta-analysis provides statistical evidence that fracture type and age are the most significant predictors.
股骨头缺血性坏死(AVN)是儿童股骨颈骨折最严重的并发症之一。AVN报告发生率的差异以及缺乏统计证据,使得个体研究的预后能力最小化。我们研究的目的是回顾我们自己的病例以及文献中的病例,以确定AVN的危险因素。
我们对病历进行了计算机检索,以确定1980年至2003年在我们机构接受治疗的所有股骨颈骨折儿童。审查记录以确定骨折类型(Delbet I型至IV型)、移位情况、年龄、治疗方法以及AVN的发生率。还对PubMed进行了结构化检索,以确定1965年至2003年发表的所有关于儿童股骨颈骨折的英文报告。对主要骨科期刊进行手工检索以及对出版物的参考文献列表进行检索,以确定更多病例。在最初通过计算机检索确定的275篇文献中,47篇符合进一步评估的标准。这47篇文献由2名研究人员进行审查,并从20份提供患者层面数据且符合我们纳入标准的报告中提取数据。
我们机构确定了24例患有25处股骨颈骨折的患者,其中包括12名男孩和12名女孩,平均年龄为8岁(范围为1.5至16岁)。从文献中还确定了300多名符合我们研究标准的患者,共有360名患者纳入分析。骨折类型、移位情况、年龄和治疗方法均为AVN的统计学显著独立预测因素,P值≤0.05。然而,通过逻辑回归分析,骨折类型和年龄被确定为AVN的唯一显著预测因素。年龄每增加一岁,大龄儿童发生AVN的可能性增加1.14倍。I型至III型骨折发生AVN的可能性分别比IV型骨折高15倍、6倍和4倍。按Delbet分类的AVN发生率为:I型=38%,II型=28%,III型=18%,IV型=5%。
尽管有几个因素可能导致AVN的发生,但我们的荟萃分析提供了统计证据,表明骨折类型和年龄是最显著的预测因素。