术前硬膜外激素注射是否增加腰椎术后手术部位感染风险的Meta分析

Meta analysis on the relationship between preoperative epidural steroid injections and increased risk of postoperative surgical site infection

中文关键词:  硬膜外激素注射  腰椎手术  手术部位感染  Meta分析

英文关键词:Epidural steroid injection  Lumbar surgery  Surgical site infection  Meta analysis

中文摘要:

  【摘要】 目的：通过Meta分析明确术前腰椎硬膜外激素注射（lumbar epidural steroid injection，LESI）是否增加后续腰椎手术术后手术部位感染（surgical site infection，SSI）风险。方法：检索Pubmed、Embase、Cochran Trail数据库，检索时间均是从建库到2018年7月。筛选出比较腰椎术前行LESI及未行LESI患者术后90d内SSI发生率的对照研究，各研究的观察（暴露）组为接受LESI并后续相同节段行首次腰椎手术，对照（非暴露）组为未行LESI而于相同节段行首次腰椎手术。各组病例数均大于100，随访时间至少90d。评价指标为术后90d内SSI发生率。依据LESI注射距后续腰椎手术时间分亚组（1个月内，1~3个月，3~6个月，6~12个月），合并SSI比值比（OR）。使用Newcastle-Ottawa Quality Scale（NOS）评分评价纳入研究质量。分别行亚组分析探讨异质性，敏感性分析测定合并结果稳定性，Egger′s回归检测发表偏倚。结果：共5篇文献纳入研究，其中4篇为回顾性病例对照研究，1篇为回顾性队列研究，研究质量NOS评分均在6分以上。共纳入225801例患者，其中暴露组38452例，对照组187349例。经Meta分析，术前1个月内行LESI，暴露组与对照组术后90d内SSI发生率比较OR=2.15（95%CI，1.30~3.55），P＜0.05；术前1~3个月行LESI，暴露组与对照组术后90d内SSI发生率比较OR=1.54（95%CI，1.36~1.74），P＜0.05；术前3~6个月行LESI，暴露组与对照组术后90d内SSI发生率比较OR=1.09（95%CI，0.94~1.26），P＞0.05；术前6~12个月行LESI，暴露组与对照组术后90d内SSI发生率比较OR=1.30（95%CI，0.88~1.93），P＞0.05。Meta分析亚组分析证实基线资料不匹配为显著异质性来源，敏感性分析证实Meta分析结果稳定，发表偏倚检测提示未见发表偏倚。结论：术前3个月内行LESI显著增加后续腰椎手术术后90d内SSI风险。

英文摘要:

  【Abstract】 Objectives: To investigate if there is relationship between preoperative epidural steroid injections and increased risk of postoperative surgical site infection(SSI). Methods: The search was made in PubMed, Embase, Cochran Trail from inception to July 2018. All control studies comparing the incidence of SSI within postoperative 90 days between the group with preoperative LESI and non-LESI were screened out for analysis. The observation(exposure) groups in each study received LESI and subsequent initial lumbar surgery in same segment while the control(non-exposure) group did not perform LESI while received lumbar surgery. The number of cases in each group was more than 100, and the follow-up time was at least 90 days. The evaluation index was the incidence of SSI within 90 days after lumbar surgery. The analysis was performed based on the time of LESI prior to surgery(＜1 month, 1-3 months, 3-6 months, 6-12 months). Newcastle-Ottawa Quality Scale(NOS) was applied to evaluate the quality of included studies. Subgroup analysis was used to explore heterogeneity and sensitivity analysis was used to determine the robustness of combined results, publication bias assessment was also performed by using Egger′s regression test. Results: Five studies which included 225801 patients(38452 patients in exposure group, 187349 patients in non-exposure group) were enrolled. All articles were from English literature, among them 1 was retrospective cohort study while the rest were retrospective case-control studies. The scores of NOS score system were 6 and above. LESI within 1 month before subsequent surgery significantly increased the risk of SSI in 90 days after surgery, OR=2.15(95%CI, 1.30-3.55), P＜0.05. LESI within 1-3 months before surgery significantly increased the risk of SSI in 90 days after surgery, OR=1.54 (95%CI, 1.36-1.74), P＜0.05. LESI within 3-6 months before surgery did not increase the risk of SSI, OR=1.09(95%CI, 0.94-1.26), P＞0.05. LESI within 6-12 months before surgery did not increase the risk of SSI, OR=1.30(95%CI, 0.88-1.93), P＞0.05. The subgroup analysis confirmed that unmatched baseline condition was a significant source of heterogeneity, and the sensitivity analysis confirmed our results were robust, publication bias were not found in included studies. Conclusions: Our results indicate that LESI within 1 month or 1-3 months before subsequent lumbar surgery significantly increase the risk of SSI.