Emory University, Atlanta, Georgia, USA.
Ann Thorac Surg. 2010 Nov;90(5):1669-73; discussion 1673. doi: 10.1016/j.athoracsur.2010.06.129.
Historically, esophageal perforation has been associated with significant mortality. Improvements in diagnosis, critical care, and surgical and endoscopic techniques may lead to lower mortality rates in the modern era. We reviewed our experience with the management of esophageal perforation to determine whether outcomes have improved.
We retrospectively reviewed all cases of esophageal perforation from 1997 through 2008 at our institution. Univariate and propensity-matching analysis were performed.
We reviewed the charts of 147 patients, and 97 met eligibility criteria. There were 45 women, (46.4%); mean age was 60.7 ± 15.6 years. Etiologies included iatrogenic in 50 (51.6%), spontaneous in 23 (23.7%), and idiopathic in 22 (22.7%). Treatment within 24 hours of presentation occurred in 55.2% of patients; 22.7% of patients were septic on presentation. Treatment included surgery in 72 patients (74.2%) and nonoperative management in 25 (25.8%). Forty-one patients (42.3%) underwent primary repair, 5 (6.9%) underwent esophageal resection, 4 (5.6%) underwent exclusion, and 22 (22.7%) underwent drainage or stent placement. Thirty-day mortality rate for the entire cohort was only 8.3% (8 patients). The mortality rate for the primary repair patients was 7.7%, and none of the resection patients died. There was similar in-hospital mortality rate between operative and nonoperative treatment groups (p = 0.96). Propensity-matching analysis showed equal morbidity (p = 0.74) and 30-day mortality (p = 0.35) between operative and nonoperative treatment groups.
Our study represents a large series of patients treated for esophageal perforation. The results demonstrate that the overall mortality from esophageal perforation can be less than 10%. Primary repair should be considered as first-line treatment when appropriate even in patients who present more than 24 hours after perforation. Nonoperative management, in appropriate patients, can also lead to good success rates and low mortality.
historically （历史上），食管穿孔与较高的死亡率相关。诊断、重症监护、手术和内镜技术的改进可能会导致现代食管穿孔死亡率的降低。我们回顾了我们对食管穿孔的管理经验，以确定结果是否有所改善。
我们对本机构 1997 年至 2008 年间所有食管穿孔病例进行了回顾性研究。进行了单变量和倾向匹配分析。
我们回顾了 147 例患者的病历，其中 97 例符合入选标准。女性 45 例（46.4%）；平均年龄为 60.7 ± 15.6 岁。病因包括医源性 50 例（51.6%）、自发性 23 例（23.7%）和特发性 22 例（22.7%）。在发病后 24 小时内进行治疗的患者占 55.2%；22.7%的患者在发病时出现败血症。治疗方法包括手术 72 例（74.2%）和非手术治疗 25 例（25.8%）。41 例（42.3%）患者行一期修复，5 例（6.9%）患者行食管切除术，4 例（5.6%）患者行隔离术，22 例（22.7%）患者行引流或支架置入术。整个队列的 30 天死亡率仅为 8.3%（8 例）。一期修复患者的死亡率为 7.7%，无切除术患者死亡。手术治疗和非手术治疗组的住院死亡率相似（p = 0.96）。倾向匹配分析显示，手术治疗和非手术治疗组的发病率（p = 0.74）和 30 天死亡率（p = 0.35）相当。
我们的研究代表了一组接受食管穿孔治疗的大量患者。结果表明，食管穿孔的总死亡率可低于 10%。在适当的情况下，即使在穿孔后超过 24 小时的患者中，一期修复也应被视为一线治疗方法。在适当的患者中，非手术治疗也可以获得良好的成功率和低死亡率。
