Abstract：
BACKGROUND: Avascular necrosis of the humeral head (AVN) is characterized by osteonecrosis secondary to disrupted blood flow to the glenohumeral joint. Following collapse of the humeral head, arthroplasty, namely total shoulder arthroplasty (TSA) or humeral head arthroplasty (hemiarthroplasty) is recommended standard of care. The literature is limited to underpowered and small sample sizes in comparing arthroplasty modalities. Therefore, the aims of this study were (1) to compare the 10-year survivorship of TSA and hemiarthroplasty in the treatment of AVN of the humeral head and (2) to identify differences in their revision etiologies.
METHODS: Patients who underwent primary TSA and hemiarthroplasty for AVN were identified using the PearlDiver database. TSA patients were matched by age, gender, and Charlson Comorbidity Index (CCI) to the hemiarthroplasty cohort in a 4:1 ratio since TSA patients were generally older, sicker, and more often female. The 10-year cumulative incidence rate of all-cause revision was determined using Kaplan-Meier survival analysis. Multivariable analysis was conducted using Cox Proportional Hazard modeling. Chi-squared analysis was conducted to compare the indications for revisions between matched cohorts including periprosthetic joint infection (PJI), dislocation, mechanical loosening, broken implants, periprosthetic fracture, and stiffness.
RESULTS: In total, 4,825 patients undergoing TSA and 1,969 patients undergoing hemiarthroplasty for AVN were included in this study. The unmatched 10-year cumulative incidence of revision for patients who underwent TSA and hemiarthroplasty was 7.0% and 7.7%, respectively. The matched 10-year cumulative incidence of revision for patients who underwent TSA and hemiarthroplasty was 6.7% and 8.0%, respectively. When comparing the unmatched cohorts, TSA patients were at significantly higher risk of 10-year all-cause revision (HR: 1.39; P = 0.017) when compared to hemiarthroplasty patients. After matching, there was no significant difference in risk of 10-year all-cause revision (HR: 1.29; P = 0.148) and no difference in the observed etiologies for revision (P > 0.05 for all).
CONCLUSIONS: After controlling for confounders, only 6.7% of TSA and 8.0% hemiarthroplasties for humeral head AVN were revised within 10-years of index surgery. The demonstrated high and comparable long-term survivorship for both modalities supports the utilization of either for the AVN induced humeral head collapse.
METHODS: Patients who underwent primary TSA and hemiarthroplasty for AVN were identified using the PearlDiver database. TSA patients were matched by age, gender, and Charlson Comorbidity Index (CCI) to the hemiarthroplasty cohort in a 4:1 ratio since TSA patients were generally older, sicker, and more often female. The 10-year cumulative incidence rate of all-cause revision was determined using Kaplan-Meier survival analysis. Multivariable analysis was conducted using Cox Proportional Hazard modeling. Chi-squared analysis was conducted to compare the indications for revisions between matched cohorts including periprosthetic joint infection (PJI), dislocation, mechanical loosening, broken implants, periprosthetic fracture, and stiffness.
RESULTS: In total, 4,825 patients undergoing TSA and 1,969 patients undergoing hemiarthroplasty for AVN were included in this study. The unmatched 10-year cumulative incidence of revision for patients who underwent TSA and hemiarthroplasty was 7.0% and 7.7%, respectively. The matched 10-year cumulative incidence of revision for patients who underwent TSA and hemiarthroplasty was 6.7% and 8.0%, respectively. When comparing the unmatched cohorts, TSA patients were at significantly higher risk of 10-year all-cause revision (HR: 1.39; P = 0.017) when compared to hemiarthroplasty patients. After matching, there was no significant difference in risk of 10-year all-cause revision (HR: 1.29; P = 0.148) and no difference in the observed etiologies for revision (P > 0.05 for all).
CONCLUSIONS: After controlling for confounders, only 6.7% of TSA and 8.0% hemiarthroplasties for humeral head AVN were revised within 10-years of index surgery. The demonstrated high and comparable long-term survivorship for both modalities supports the utilization of either for the AVN induced humeral head collapse.
摘要：
背景：肱骨头（AVN）的缺血性坏死的特征是继发于肱骨关节血流中断的骨坏死。肱骨头塌陷后,关节成形术,即全肩关节置换术（TSA）或肱骨头置换术（半髋关节置换术）是推荐的标准护理。在比较关节成形术方式时，文献仅限于动力不足和样本量小。因此,这项研究的目的是(1)比较TSA和半髋关节置换术治疗肱骨头AVN的10年生存率,以及(2)确定其翻修病因的差异.
方法：使用PearlDiver数据库确定接受原发性TSA和半髋关节置换术治疗AVN的患者。TSA患者按年龄匹配，性别,和Charlson合并症指数（CCI）以4：1的比例加入半髋关节置换术队列，因为TSA患者通常年龄较大，病情加重,更多的是女性。使用Kaplan-Meier生存分析确定全因翻修的10年累积发生率。采用Cox比例危险模型进行多变量分析。进行卡方分析以比较包括假体周围关节感染（PJI）在内的匹配队列之间的修订适应症。位错,机械松动,破裂的植入物,假体周围骨折,和刚度。
结果：总计，本研究包括4,825名接受TSA的患者和1,969名接受半髋关节置换术的患者。接受TSA和半髋关节置换术的患者的10年累积翻修发生率分别为7.0％和7.7％，分别。接受TSA和半髋关节置换术的患者的10年累积翻修发生率分别为6.7％和8.0％，分别。当比较不匹配的队列时，与半髋关节置换术患者相比，TSA患者10年全因翻修的风险明显更高（HR：1.39；P=0.017）。匹配后,10年全因修订的风险无显著差异(HR:1.29;P=0.148),观察到的修订病因无显著差异(均P>0.05).
结论：控制混杂因素后，在初次手术的10年内,仅有6.7%的TSA和8.0%的肱骨头AVN半关节置换术进行了修正.两种方式的高生存率和可比较的长期生存率支持利用AVN引起的肱骨头塌陷。
方法：使用PearlDiver数据库确定接受原发性TSA和半髋关节置换术治疗AVN的患者。TSA患者按年龄匹配，性别,和Charlson合并症指数（CCI）以4：1的比例加入半髋关节置换术队列，因为TSA患者通常年龄较大，病情加重,更多的是女性。使用Kaplan-Meier生存分析确定全因翻修的10年累积发生率。采用Cox比例危险模型进行多变量分析。进行卡方分析以比较包括假体周围关节感染（PJI）在内的匹配队列之间的修订适应症。位错,机械松动,破裂的植入物,假体周围骨折,和刚度。
结果：总计，本研究包括4,825名接受TSA的患者和1,969名接受半髋关节置换术的患者。接受TSA和半髋关节置换术的患者的10年累积翻修发生率分别为7.0％和7.7％，分别。接受TSA和半髋关节置换术的患者的10年累积翻修发生率分别为6.7％和8.0％，分别。当比较不匹配的队列时，与半髋关节置换术患者相比，TSA患者10年全因翻修的风险明显更高（HR：1.39；P=0.017）。匹配后,10年全因修订的风险无显著差异(HR:1.29;P=0.148),观察到的修订病因无显著差异(均P>0.05).
结论：控制混杂因素后，在初次手术的10年内,仅有6.7%的TSA和8.0%的肱骨头AVN半关节置换术进行了修正.两种方式的高生存率和可比较的长期生存率支持利用AVN引起的肱骨头塌陷。
