GOLDEN, Colo. -- Practice makes better bariatric surgery, with fewer complications and quicker hospital stays, according to investigators here.
GOLDEN, Colo., July 30 -- Practice makes better bariatric surgery, with fewer complications and quicker hospital stays, according to investigators here.
A survey of more than 166,000 bariatric surgery procedures performed in 19 states over three years showed that patients treated at the most experienced hospitals had 64% fewer complications and 26% shorter stays than similar patients treated in low-volume centers, found Samantha Collier, M.D., and colleagues, of HealthGrades, a healthcare ratings company.
"Assuming all hospitals in the 19 states studied could perform at the level of the top-performing hospitals 5,902 patients could have potentially avoided one or more major in-hospital complications associated with bariatric surgery," they wrote, in a report titled Bariatric Surgery Trends in American Hospitals.
The authors found that from 2003 through 2005, 166,410 bariatric surgery procedures were performed in the 19 states, and that only about 1,300 more procedures were performed in 2005 than in 2003.
A total of 731 hospitals performed at least one bariatric surgery procedure during the years of the study, but only 481 had sufficient volume to merit a rating, as determined by a minimum 30 cases across three years of data and at least five cases in the most current year.
They also found that 70% of the procedures in 2005 were laparoscopic rather than open gastric bypass, and that the laparoscopic method, as might be expected, was associated with fewer in-hospital complications.
The authors also ranked the hospitals with one, three, or five stars, with five star representing those institutions with significantly better performance than predicted by multivariate logistic regression. Three stars represented average performance, not statistically different from that predicted, and one star was awarded to those hospitals with significantly worse than predicted outcomes.
Of the 481 centers studied, 109 (22.7%) earned five stars, 257 hospitals (53.4%) were given three, and 15 (23.9%) just one.
They found that the five-star rated hospitals performed almost twice the number of procedures as one-star rated hospitals (533 versus 293), and that a typical patient undergoing a bariatric surgery procedure in one of the highest-ranked hospitals would have, on average, a 64% lower risk of one or more major in-hospital complications compared with a low-ranking center.
Compared with all one- and three-star centers, patients treated in a five-star center had a 41% lower risk for complications. The most frequent complications were surgical complications of the respiratory system, occurring in 1.45% of all patients, gastrointestinal system problems in 1.28% post-operative pulmonary insufficiency in 1.19%, hemorrhage complicating the procedure in 1.14%, and accidental operative laceration in 1.08%.
"Two hundred and fifty patients died in hospital from complications of bariatric surgery during the study period," the authors wrote. "The average U.S. mortality rate during the study period was 0.15%, or 1.5 patients per 1,000. While the absolute overall mortality rate was low for most hospitals, the mortality rate in five-star rated hospitals was almost half the rate of all other U.S. hospitals."
They also found that average length of stay among bariatric surgery patients was 26% shorter for those treated in the top-echelon hospitals compared with those treated in one-star centers.
Individual hospital quality results from this study are available at www.healthgrades.com.
"Bariatric surgery has been demonstrated to be highly effective for those with morbid obesity, but the relatively new procedures are not yet regulated or a credentialed surgical subspecialty," said Dr. Collier. "So it is important that patients considering surgery know how hospitals rate."
The authors noted several study limitations, including the possibility that the cases may have been coded incorrectly or incompletely by the hospital, and that there could be missing, incorrect, or outdated data.