ORLANDO – Patients undergoing bariatric surgery had a significantly reduced rate of subsequent myocardial infarctions and strokes and significantly increased survival, compared with similar, morbidly obese patients who had other types of surgery, in a retrospective cohort study of more than 9,000 U.S. patients.
The results “add to the growing evidence that bariatric surgery plays a role in temporizing the risk factors for major cardiovascular events. We believe our analysis builds on prior reports and takes them a step further by evaluating actual events,” Dr. John D. Scott said at the meeting.
In his study of bariatric surgery patients in South Carolina during 1996-2008, the combined rate of MIs, strokes, and deaths was 52% below the rate in patients undergoing gastrointestinal surgery, and 28% below the rate of those who had orthopedic surgery – both statistically significant differences. The results also showed significant drops in each component of the combined end point (MIs, strokes, and deaths).
“Previous literature demonstrated that cardiovascular risk declined after bariatric surgery. [This study looked] at the rate of actual cardiovascular events, which significantly declined after bariatric surgery,” said Dr. Scott, a bariatric surgeon at University Medical Center Greenville (S.C.) Hospital System.
The new study used hospital in-patient records collected during 1996-2008 through the South Carolina Office of Research and Statistics, and death data collected by the South Carolina Department of Health and Environmental Control. The analysis included morbidly obese patients aged 40-79 years who underwent nonemergency surgery (4,747 patients who had any form of bariatric surgery, 3,066 who had joint replacement or spinal surgery, and 1,327 who had a cholecystectomy, hernia repair, or lysis of GI adhesions). Those with a prior MI or stroke were excluded. Patients were followed for an average of 14 months after bariatric surgery, 25 after orthopedic surgery, and 26 months after GI surgery.
In a multivariate analysis that controlled for age, sex, race, hypertension, dyslipidemia, diabetes, coronary artery disease, obstructive sleep apnea, and a history of transient ischemic attack, patients undergoing bariatric surgery had a significant 41% reduced rate of first MI, compared with the orthopedic surgery patients, and a significant 51% lower rate than the GI surgery patients.
Mortality in the bariatric surgery patients dropped by a significant 19% and 55% relative to the orthopedic andGI patients, respectively, and the stroke rate was also significantly lower after bariatric surgery compared with the rates in each control group.
Notably, bariatric surgery “reduced cardiovascular events, as opposed to obesity-drug treatments that may actually increase the risk for cardiovascular events,” he noted. “Bariatric surgery has been rigorously tested and [proved] over the past 20 years, and it has a dramatic effect on all aspects of patient health. [Most] medical treatments for obesity don't have 20 years of data, and some medications actually cause heart problems. We don't know how bariatric surgery reduced MI and stroke, but it's probably several factors: weight loss, and resolution of diabetes, hypertension, and sleep apnea.
Dr. Scott has been a speaker for Gore Medical.
View on the News
Control-Group Issues Complicate Comparisons
Dr. Scott and his associates have attempted to address a quintessential question about bariatric surgery: Does it reduce the long-term mortality associated with obesity? About 10 prior reports in the literature have also attempted to address this.
All of these studies have weaknesses, mostly involving the control group. Because nonsurgical patients who received medical management typically are used as the control group, this often raises the question of whether the control patients were sicker than the surgical patients.
Dr. Scott's study avoided this weakness by comparing bariatric surgery patients with other surgery patients. This eliminated the bias of greater sickness, as all patients in the study were healthy enough to undergo elective surgery. It also eliminated any bias stemming from access to surgical and medical care.
Despite this, the bariatric surgery and control groups differed in demographics and comorbidities. It seems as though the between-group differences were too extensive to allow for adequate adjustment by a multivariate analysis. In addition, the study included no information on body mass index, so no adjustment was possible for this variable.
The impact of bariatric surgery can only be reliably tested in a randomized, controlled trial. The biases embedded in databases cannot be fully eliminated; the only way to address this question objectively is with a randomized trial.
PHILIP SCHAUER, M.D., is director of the bariatric and metabolic institute at the Cleveland Clinic. He made these comments as the designated discussant of Dr. Scott's paper. He has received teaching grants from Allergan and Covidien; consulting fees as a member of the advisory board of and research support from Bard/Davol and Ethicon Endo-Surgery; consulting fees from Baxter Healthcare, Cardinal Health, and Stryker Endoscopy; and support from RemedyMD.
