COLORADO SPRINGS – Minimally invasive esophagectomy was associated with a significantly lower rate of postoperative major morbidity as well as a mean 1-day briefer length of stay than open esophagectomy in a propensity-matched analysis of the real-world American College of Surgeons-National Quality Improvement Program database, Mark F. Berry, MD, reported at the annual meeting of the Western Thoracic Surgical Association.
However, both of the study’s discussants questioned whether the reported modest absolute reduction in major morbidity was really attributable to the minimally invasive approach or could instead have resulted from one of several potential confounders that couldn’t be fully adjusted for, given inherent limitations of the ACS-NSQIP database.
“There was a statistically significant difference in morbidity,” replied Dr. Berry of Stanford (Calif.) University. “It was a 4% absolute difference, which I think is probably clinically meaningful, but certainly it’s not really, really dramatic.”
Dr. Mark F. Berry
“What I think we found is that it’s safe to do a minimally invasive esophagectomy and safe for people to introduce it into their practice. But it’s not necessarily something that’s a game changer, unlike what’s been seen with minimally invasive approaches for some other things,” said Dr. Berry, who added that he didn’t wish to overstate the importance of the observed difference in morbidity.
Studies from high-volume centers show that minimally-invasive esophagectomy (MIE) reduces length of stay, postoperative major morbidity, and features equivalent or even slightly lower mortality than traditional open esophagectomy, the generalizability of these findings beyond such centers is questionable. That’s why Dr. Berry and his coinvestigators turned to the ACS-NSQIP database, which includes all esophagectomies performed for esophageal cancer at roughly 700 U.S. hospitals, not just those done by board-certified thoracic surgeons.
He presented a retrospective cohort study of 3,901 esophagectomy patients during 2005-2013 who met study criteria, 16.4% of whom had MIE. The use of this approach increased steadily from 6.5% of all esophagectomies in 2005 to 22.3% in 2013. A propensity-matched analysis designed to neutralize potentially confounding differences included 638 MIE and 1,914 open esophagectomy patients.
The primary outcome was the 30-day rate of composite major morbidity in the realms of various wound, respiratory, renal, and cardiovascular complications. The rate was 36.1% in the MIE group and 40.5% with open esophagectomy in the propensity-matched analysis, an absolute risk reduction of 4.4% and a relative risk reduction of 17%. Although rates were consistently slightly lower in each of the categories of major morbidity, those individual differences didn’t achieve statistical significance. The difference in major morbidity became significant only when major morbidity was considered as a whole.
Mean length of stay was 9 days with MIE and 10 days with open surgery.
There was no significant difference between the two study groups in 30-day rates of readmission, reoperation, or mortality.
Discussant Donald E. Low said “esophagectomy is being analysed regarding its place in all sorts of presentations, stages, and situations, so the aspect of making sure that we’re delivering the services as efficiently as possible is going to become more important, not less important.”
That being said, he noted that there is no specific CPT code for MIE. That raises the possibility of an uncertain amount of procedural misclassification in the ACS-NSQIP database.
Also, the only significant difference in major morbidity between the two study groups was in the subcategory of intra- or postoperative bleeding requiring transfusion, which occurred in 10.8% of the MIE and 16.7% of the open esophagectomy groups, observed Dr. Low, director of the Esophageal Center of Excellence at Virginia Mason Medical Center, Seattle.
“Some of us believe that blood utilization and transfusion requirement is really a quality measure and not a complication,” the surgeon said. And if that outcome is excluded from consideration, then there is no significant difference in major morbidity.
Discussant Douglas E. Wood, MD, professor and chair of the department of surgery at the University of Washington, Seattle, took the opportunity to share a self-described “pet peeve” about analyses of national surgical databases: these databases typically don’t contain key details necessary to correct for provider and hospital characteristics.
“The small differences that you demonstrate could easily have been completely driven by providers who choose to do minimally invasive esophagectomy and are in higher-volume, more specialized centers,” he said. “I’m not convinced of your conclusion that MIE produces less morbidity based on a 4% difference and no analysis of provider characteristics.”
