Reporting of short-term clinical outcomes after esophagectomy: a systematic review

Ann Surg. 2012 Apr;255(4):658-66. doi: 10.1097/SLA.0b013e3182480a6a.


Objective: This review summarizes reporting of complications of esophageal cancer surgery.

Background: Accurate assessment of morbidity and mortality after surgery for cancer is essential to compare centers, allow data synthesis, and inform clinical decision-making. A lack of defined standards may distort clinically relevant treatment effects.

Methods: Systematic literature searches identified articles published between 2005 and 2009 reporting morbidity and mortality after esophagectomy for cancer. Data were analyzed for frequency of complication reporting and to check whether outcomes were defined and classified for severity and whether a validated system for grading complications was used. Information about reporting outcomes adjusting for baseline risk factors was collated, and a descriptive summary of the results of included outcomes was undertaken.

Results: Of 3458 abstracts, 224 full papers were reviewed and 122 were included (17 randomized trials and 105 observational studies), reporting outcomes of 57,299 esophagectomies. No single complication was reported in all papers, and 60 (60.6%) did not define any of the measured complications. Anastomotic leak was the most commonly reported morbidity, assessed in 80 (80.1%) articles, defined in 28 (28.3%), but 22 different descriptions were used. Five papers (5.1%) categorized morbidity with a validated grading system. One hundred fifteen papers reported postoperative mortality rates, 25 defining the term using 10 different definitions. In-hospital mortality was the most commonly used term for postoperative death, with 6 different interpretations of this phrase. Eighteen papers adjusted outcomes for baseline risk factors and 60 presented baseline measures of comorbidity.

Conclusions: Outcome reporting after esophageal cancer surgery is heterogeneous and inconsistent, and it lacks methodological rigor. A consensus approach to reporting clinical outcomes should be considered, and at the minimum it is recommended that a "core outcome set" is defined and used in all studies reporting outcomes of esophageal cancer surgery. This will allow meaningful cross study comparisons and analyses to evaluate surgery.

Publication types

  • Research Support, Non-U.S. Gov't
  • Review
  • Systematic Review

MeSH terms

  • Clinical Trials as Topic
  • Esophageal Neoplasms / mortality
  • Esophageal Neoplasms / surgery*
  • Esophagectomy* / mortality
  • Hospital Mortality
  • Humans
  • Postoperative Complications*
  • Risk Adjustment
  • Treatment Outcome