Abstract：
BACKGROUND: In primary shoulder arthroplasty (SA), intravenous (IV) cefazolin has demonstrated lower rates of infectious complications when compared to IV vancomycin. However, previous analyses included SA cohorts with both complete and incomplete vancomycin administration. Therefore, it is currently unclear whether cefazolin still maintains a prophylactic advantage to vancomycin when it is appropriately indicated and sufficiently administered at the time of surgical incision. This study evaluated the comparative efficacy of cefazolin and complete vancomycin administration for surgical prophylaxis in primary shoulder arthroplasty with respect to infectious complications.
METHODS: A retrospective cohort study was conducted utilizing a single institution total joint registry database, where all primary SA types (hemiarthroplasty, anatomic total shoulder arthroplasty, reverse shoulder arthroplasty) performed between 2000 to 2019 for elective and trauma indications using IV cefazolin or complete vancomycin administration as the primary antibiotic prophylaxis were identified. Vancomycin was primarily indicated for patients with a severe self-reported penicillin or cephalosporin allergy and/or MRSA colonization. Complete administration was defined as at least 30 minutes of antibiotic infusion prior to incision. All included SA had at least 2 years of clinical follow-up. Multivariable Cox proportional hazard regression was used to evaluate all-cause infectious complications including survival free of prosthetic joint infection (PJI).
RESULTS: The final cohort included 7,177 primary SA, 6,879 (95.8%) received IV cefazolin and 298 (4.2%) received complete vancomycin administration. Infectious complications occurred in 120 (1.7%) SA leading to 81 (1.1%) infectious reoperations. Of the infectious complications 41 (0.6%) were superficial infections and 79 were (1.1%) PJIs. When categorized by administered antibiotics, there were no differences in rates of all infectious complications (1.6% vs. 2.3%; P = .352), superficial complications (0.5% vs. 1.3%; P = .071), PJI (1.1% vs. 1.0%; P = .874), or infectious reoperations (1.1% vs. 1.0%; P = .839). On multivariable analyses, complete vancomycin infusion demonstrated no difference in rates of infectious complications compared to cefazolin administration (hazard ratio [HR], 1.50 [95% confidence interval (CI), 0.70 to 3.25]; P = .297), even when other independent predictors of PJI (male sex, prior surgery, and Methicillin-resistant Staphylococcus aureus colonization) were considered.
CONCLUSIONS: In comparison to cefazolin, complete administration of vancomycin (infusion to incision time greater than 30 minutes) as the primary prophylactic agent does not adversely increase the rates of infectious complications and PJI. Prophylaxis protocols should promote appropriate indications for the use of cefazolin or vancomycin, and when necessary, ensure complete administration of vancomycin to mitigate additional infectious risks after primary SA.
METHODS: A retrospective cohort study was conducted utilizing a single institution total joint registry database, where all primary SA types (hemiarthroplasty, anatomic total shoulder arthroplasty, reverse shoulder arthroplasty) performed between 2000 to 2019 for elective and trauma indications using IV cefazolin or complete vancomycin administration as the primary antibiotic prophylaxis were identified. Vancomycin was primarily indicated for patients with a severe self-reported penicillin or cephalosporin allergy and/or MRSA colonization. Complete administration was defined as at least 30 minutes of antibiotic infusion prior to incision. All included SA had at least 2 years of clinical follow-up. Multivariable Cox proportional hazard regression was used to evaluate all-cause infectious complications including survival free of prosthetic joint infection (PJI).
RESULTS: The final cohort included 7,177 primary SA, 6,879 (95.8%) received IV cefazolin and 298 (4.2%) received complete vancomycin administration. Infectious complications occurred in 120 (1.7%) SA leading to 81 (1.1%) infectious reoperations. Of the infectious complications 41 (0.6%) were superficial infections and 79 were (1.1%) PJIs. When categorized by administered antibiotics, there were no differences in rates of all infectious complications (1.6% vs. 2.3%; P = .352), superficial complications (0.5% vs. 1.3%; P = .071), PJI (1.1% vs. 1.0%; P = .874), or infectious reoperations (1.1% vs. 1.0%; P = .839). On multivariable analyses, complete vancomycin infusion demonstrated no difference in rates of infectious complications compared to cefazolin administration (hazard ratio [HR], 1.50 [95% confidence interval (CI), 0.70 to 3.25]; P = .297), even when other independent predictors of PJI (male sex, prior surgery, and Methicillin-resistant Staphylococcus aureus colonization) were considered.
CONCLUSIONS: In comparison to cefazolin, complete administration of vancomycin (infusion to incision time greater than 30 minutes) as the primary prophylactic agent does not adversely increase the rates of infectious complications and PJI. Prophylaxis protocols should promote appropriate indications for the use of cefazolin or vancomycin, and when necessary, ensure complete administration of vancomycin to mitigate additional infectious risks after primary SA.
摘要：
背景：在初次肩关节成形术（SA）中，静脉注射(IV)头孢唑啉与静脉注射万古霉素相比,感染并发症发生率较低.然而,之前的分析包括万古霉素完全和不完全给药的SA队列.因此,目前尚不清楚头孢唑林是否仍保持对万古霉素的预防性优势，如果在手术切口时给予适当的适应症和足够的给药。这项研究评估了头孢唑林和完全万古霉素给药在初次肩关节置换术中手术预防感染性并发症的比较疗效。
方法：使用单一机构的总联合注册数据库进行了一项回顾性队列研究，所有原发性SA类型(半髋关节置换术，解剖全肩关节置换术，反向肩关节成形术）在2000年至2019年间进行，用于选择性和创伤适应症，使用头孢唑林静脉注射或完全万古霉素作为主要抗生素预防。万古霉素主要用于严重自我报告的青霉素或头孢菌素过敏和/或MRSA定植的患者。完全施用被定义为在切口之前至少30分钟的抗生素输注。所有包括的SA都有至少2年的临床随访。多变量Cox比例风险回归用于评估全因感染并发症，包括无人工关节感染（PJI）的生存率。
结果：最终队列包括7,177名主要SA，6,879(95.8%)接受IV头孢唑啉,298(4.2%)接受完全万古霉素给药。120例（1.7％）SA发生感染性并发症，导致81例（1.1％）感染性再次手术。在感染并发症中，41例（0.6％）是浅表感染，79例（1.1％）是PJI。当按使用抗生素分类时，所有感染并发症的发生率没有差异(1.6%vs.2.3%；P=.352），浅表并发症（0.5%vs.1.3%；P=0.071），PJI（1.1%与1.0%；P=.874），或传染性再次手术(1.1%与1.0%；P=.839）。在多变量分析中，与头孢唑林相比，万古霉素完全输注显示感染并发症的发生率没有差异(风险比[HR]，1.50[95%置信区间(CI)，0.70至3.25]；P=.297)，即使当PJI的其他独立预测因子(男性，之前的手术,和耐甲氧西林金黄色葡萄球菌定植）被考虑。
结论：与头孢唑林相比，万古霉素的完全给药（输注至切口时间超过30分钟）作为主要预防剂不会不利地增加感染并发症和PJI的发生率。预防方案应促进使用头孢唑啉或万古霉素的适当适应症，必要时,确保万古霉素的完全给药，以减轻原发性SA后的额外感染风险。
方法：使用单一机构的总联合注册数据库进行了一项回顾性队列研究，所有原发性SA类型(半髋关节置换术，解剖全肩关节置换术，反向肩关节成形术）在2000年至2019年间进行，用于选择性和创伤适应症，使用头孢唑林静脉注射或完全万古霉素作为主要抗生素预防。万古霉素主要用于严重自我报告的青霉素或头孢菌素过敏和/或MRSA定植的患者。完全施用被定义为在切口之前至少30分钟的抗生素输注。所有包括的SA都有至少2年的临床随访。多变量Cox比例风险回归用于评估全因感染并发症，包括无人工关节感染（PJI）的生存率。
结果：最终队列包括7,177名主要SA，6,879(95.8%)接受IV头孢唑啉,298(4.2%)接受完全万古霉素给药。120例（1.7％）SA发生感染性并发症，导致81例（1.1％）感染性再次手术。在感染并发症中，41例（0.6％）是浅表感染，79例（1.1％）是PJI。当按使用抗生素分类时，所有感染并发症的发生率没有差异(1.6%vs.2.3%；P=.352），浅表并发症（0.5%vs.1.3%；P=0.071），PJI（1.1%与1.0%；P=.874），或传染性再次手术(1.1%与1.0%；P=.839）。在多变量分析中，与头孢唑林相比，万古霉素完全输注显示感染并发症的发生率没有差异(风险比[HR]，1.50[95%置信区间(CI)，0.70至3.25]；P=.297)，即使当PJI的其他独立预测因子(男性，之前的手术,和耐甲氧西林金黄色葡萄球菌定植）被考虑。
结论：与头孢唑林相比，万古霉素的完全给药（输注至切口时间超过30分钟）作为主要预防剂不会不利地增加感染并发症和PJI的发生率。预防方案应促进使用头孢唑啉或万古霉素的适当适应症，必要时,确保万古霉素的完全给药，以减轻原发性SA后的额外感染风险。
