WASHINGTON – Annual case volume appears to be falling out as a safety factor in bariatric surgery, Dr. Oliver Varban said at the annual clinical congress of the American College of Surgeons.
In the early days of the procedure, facilities that performed more than 125 operations each year had significantly better safety outcomes than did those performing fewer operations. But that difference is fading, particularly as laparoscopy continues to supplant open surgery, said Dr. Varban of the University of Michigan Health Systems, Ann Arbor.
"Over time, safety is improving in both high- and low-volume centers, with lower rates of complications, morbidity, and mortality. The inverse relationship does still persist, but that effect is attenuating."
Dr. Varban conducted a review of more than 270,000 bariatric surgery procedures performed in 12 hospitals from 2004 to 2009. He separated the hospitals by annual case volume: 125 or more and less than 125 per year.
In each year, about two-thirds of the procedures were completed in high-volume hospitals. The type of surgery also varied over the study period. During 2004-2005, laparoscopic and open procedures were about equally common. By 2006-2007, laparoscopic operations made up 65% of all bariatric procedures, and that held steady through 2008-2009.
The type of procedure evolved as well, Dr. Varban noted. In 2004-2005, gastric banding comprised about 5% of the operations. By 2006-2007, that had risen to 20%, and by 2008-2009, the number was close to 30%.
In the first era, low-volume hospitals had significantly higher rates of any complication than did high-volume centers (9.3% vs. 6%). By 2006-2007, the difference had narrowed but was still statistically significant (7% vs. 4.8%). By 2008-2009, the difference was no longer significant (5.6% vs. 4.5%).
A multivariate analysis that controlled for type of surgery and patient demographics found a similar trend. The risk of any complication remained significantly elevated at low-volume centers during all three periods (odds ratio, 1.33 in 2004-2005; 1.35 in 2006-2007; and 1.21 in 2008-2009).
The pattern of serious complications (problems that extended the length of stay beyond 5 days) was similar, with a rate of 4.6% vs. 2.7% in the early era; 3% vs. 1.8% in the middle era; and 2.4% vs. 1.4% in the final era. The risk of serious complications was significantly higher in the low-volume group in every era (OR 1.35, 1.43, and 1.44, respectively).
The rates of reoperation were higher in low-volume centers in every era, and declined as time went on. However, the difference between low- and high-volume centers in terms of reoperation rates was nonsignificant at every time point (1.5% vs. 1.06%; 1% vs. 0.75%; 0.78% vs. 0.67%). Similarly, the adjusted odds ratios were nonsignificant (OR 1.22, 1.28, and 1.20).
Although overall the mortality rates were low and remained low, they declined significantly in both groups over the study period (from 0.22% vs. 0.1% and 0.09% vs. 0.04%). Only in the first era was the difference statistically significant, with an adjusted OR of 1.71.
Dr. Varban had no financial disclosures.