Department of Orthopaedics and Trauma, Medical University of Graz, Auenbruggerplatz 5, Graz 8036, Austria.
Injury. 2021 Nov;52(11):3483-3488. doi: 10.1016/j.injury.2021.01.015. Epub 2021 Jan 19.
Periprosthetic fractures (PPF) following total knee (TKA) and hip arthroplasty (THA) have become more common over the years. The aim of the present study was to assess morbidity and mortality following surgery for PPF of hip and knee.
Altogether, 124 patients (mean age: 77 years; 77.4% female) with PPF of the hip (n=97) and knee (n=27), treated between 2005 and 2017 at a level-1 trauma centre, were retrospectively included. In order to assess risk factors for postoperative morbidity, Fine and Gray's model was used to compensate for death as the competing event. Risk factors for mortality were estimated with uni- and multivariate Cox-regression models.
Vancouver B2 fractures were most common (n=39; 42.4%), followed by B1 fractures (n=23; 25.0%). Lewis-Rorabeck Type I fractures (n=14; 51.9%) were most frequent in PPF of the knee. Overall complication rates were 44.0% and 29.9% for PPF of the knee and hip, respectively, with three patients having both early and late complications, 25 patients developing early complications and 19 patients undergoing surgery for implant-related, late complications. In the multivariate Fine and Gray model, advanced patient age (HR: 0.956; 95%CI: 0.922-0.991; p=0.014) and prosthesis exchange (vs. ORIF; HR: 0.242, 95%CI: 0.068-0.859; p=0.028) were associated with lower risk of implant-related complications, irrespective of gender (p=0.450) and a surgical delay > 2 days (p=0.411). One- and 5-year overall survival-rates were 97.9% and 93.1%, respectively. Gender, type of fixation (ORIF vs. prosthesis exchange), surgical delay > 2 days, BMI and age at surgery were neither in the univariate, nor multivariate Cox-regression model associated with an increased mortality rate.
Postoperative morbidity caused by implant-related complications is higher in younger patients and those receiving ORIF. With the statistical approach used, potential underestimation of actual complication rates may have been avoided, taking into account death as the competing event. Despite being based on a retrospective, heterogenous patient collective treated at a level-1 trauma centre, our results indicate that careful planning of the surgical procedure beyond 2 days, taking into consideration both patient's age and activity level, has no negative effect on patient outcome.
全膝关节置换术（TKA）和髋关节置换术（THA）后发生的假体周围骨折（PPF）近年来变得更为常见。本研究旨在评估髋关节和膝关节 PPF 手术后的发病率和死亡率。
共纳入 124 例（平均年龄：77 岁；77.4%为女性）在 2005 年至 2017 年期间于 1 级创伤中心接受治疗的髋关节（n=97）和膝关节（n=27）PPF 患者。为了评估术后发病率的危险因素，使用 Fine 和 Gray 模型来补偿因死亡作为竞争事件而导致的缺失数据。使用单变量和多变量 Cox 回归模型来估计死亡率的危险因素。
温哥华 B2 型骨折最为常见（n=39；42.4%），其次是 B1 型骨折（n=23；25.0%）。膝关节 PPF 中最常见的是 Lewis-Rorabeck Ⅰ型骨折（n=14；51.9%）。膝关节和髋关节的总体并发症发生率分别为 44.0%和 29.9%，其中 3 例患者既有早期并发症也有晚期并发症，25 例患者发生早期并发症，19 例患者因与植入物相关的晚期并发症而接受手术。在多变量 Fine 和 Gray 模型中，患者年龄较大（HR：0.956；95%CI：0.922-0.991；p=0.014）和假体置换（与 ORIF 相比；HR：0.242，95%CI：0.068-0.859；p=0.028）与较低的植入物相关并发症风险相关，与性别（p=0.450）和手术延迟>2 天（p=0.411）无关。1 年和 5 年的总生存率分别为 97.9%和 93.1%。性别、固定方式（ORIF 与假体置换）、手术延迟>2 天、BMI 和手术时年龄在单变量和多变量 Cox 回归模型中均与死亡率的增加无关。
接受 ORIF 治疗的年轻患者和患者术后与植入物相关的并发症导致的发病率更高。使用所采用的统计学方法，考虑到死亡作为竞争事件，可能避免了实际并发症发生率的潜在低估。尽管基于在 1 级创伤中心接受治疗的回顾性、异质性患者群体，但我们的研究结果表明，在 2 天以上的手术中进行仔细的手术计划，同时考虑到患者的年龄和活动水平，对患者的预后没有负面影响。
