SAN FRANCISCO – Laparoscopic duodenal switch should be considered a revisional procedure rather than a primary procedure for morbidly obese patients unable to lose enough weight with other bariatric procedures, a retrospective database review indicates.
"I see a role for laparoscopic DS [duodenal switch], for example, following a failed laparoscopic sleeve ... in patients who have a high BMI ... . If they lose all of their weight or at least [reach] a reasonable weight, that’s fine. If they can’t lose the weight, then adding laparoscopic DS in patients who have lost some weight becomes an easier operation," said Dr. Namir Katkhouda, director of the bariatric surgery program at the Keck School of Medicine of the University of Southern California in Los Angeles.
Data from a review of open DS cases for morbid obesity between 1993 and 2011 at the Keck Medical Center showed that the DS procedure is associated with a high ongoing morbidity, Dr. Katkhouda said at the annual meeting of the American Surgical Association. Complications such as strictures and reflux could be linked, and the open approach was responsible for a high rate of hernias. Therefore, DS should not be routinely recommended as a primary procedure for morbid obesity, he noted.
"With the continued scrutiny for the safety of bariatric surgery, I agree with the authors that the DS is an operation that should only be performed in selected high-volume centers and on patients with severe concomitant medical metabolic disease who are willing to return for follow-up" to assess complications, said invited discussant Dr. Ninh T. Nguyen, chief of the gastrointestinal surgery at the University of California in Irvine.
Dr. Katkhouda and his colleagues performed a retrospective study of a prospective bariatric surgery database and electronic medical records. They included all patients who had a DS procedure for morbid obesity between 1993 and 2011. Follow-up data included all hospital readmissions and standardized telephone questionnaires. Mortality was assessed by using the Social Security death index database.
Among the 1,162 patients they identified, the median age was 42 years and most patients (80%) were women. Their median BMI was 51 kg/m2. Median follow-up was 115 months. Common comorbidities included hypertension, pulmonary disease, diabetes, hyperlipidemia, and cardiac disease. Most patients had one to five comorbidities.
The overall readmission rate was 52%. The mean number of readmissions in patients with complications was 2.4. The median length of stay was 6 days for patients with complications vs. 5 days for those without complications, a significant difference. Overall, 40% of patients had reoperations and 14% had two or more reoperations.
The overall complication rate was 58% and the median time to complication was 19 months. Among patients with complications, the reoperation rate was 68%. For patients with fistula, leak, DS failure, dehiscence, small bowel obstruction, or hernia, the reoperation rate was greater than 50%.
Gastroesophageal reflux disease occurred in 64 patients – half de novo and half persistent. Among patients with GERD, the reoperative rate with conversion to Roux-en-Y gastric bypass was 39%. The rate of small bowel obstruction was 7%; the reoperation rate in this group was 74%. Severe malabsorption occurred in 11% of patients. Of these, 29% had hypoalbuminemia, 21% were iron deficient, and 21% had hypocalcemia. An additional 29% had other types of malabsorption. Overall, 42% of patients with malabsorption required reoperation.
On univariate analysis, coronary artery disease, pulmonary disease, dyslipidemia, and degenerative joint disease were identified as risk factors for complications. The presence of more than two comorbidities and age were also risk factors for complications. On multivariate analysis, patients with degenerative joint disease had a fivefold increased risk of complications. Patients with dyslipidemia had a 42% greater risk of complications; African Americans had an 87% greater risk.
The overall mortality was 7%. The median age at death was 51 years. Perioperative mortality was 0.9%; three patients had fatal cardiac events, one had pulmonary embolism, two had respiratory failure, and four had multiorgan failure.
On univariate analysis, gender, race, coronary artery disease, hypertension, age, and BMI were significant predictors of mortality. On multivariate analysis, only age and African American race were significant predictors.
Limitations of the study include the fact that it is a single-center, observational study, and results were not compared with those of another method such as gastric bypass, Dr. Katkhouda said, noting that outcomes could be different in a multicenter study.
Dr. Nguyen said that duodenal switch is not a common bariatric procedure; it accounts for less than 1% of all bariatric procedures performed in the United States. "The perioperative mortality at 0.9% in this series is fourfold higher than that of contemporary data for banded, bypass, or sleeve," he said.