Why we Disagree: A Different Editorial of the Article “Weight Loss, Cardiovascular Risk Factors, and Quality of Life After Gastric Bypass and Duodenal Switch. A Randomized Trial”

Dear Bariatric Times Editor:

Recently, an article appeared on Reuters Health website entitled, “Rare surgery brings more weight loss, more risks.”[1] The article was based on a paper that was published in the September 6, 2011, Annals of Internal Medicine entitled, “Weight loss, cardiovascular risk factors, and quality of life after gastric bypass and duodenal switch: a randomized trial.”[2] The Reuters article was significantly influenced by the accompanying editorial by Dr. Edward Livingston, “Primum non nocere”[3] (a Latin phrase meaning ‘First, do no harm’), which also ran in the September 6, 2011, Annals of Internal Medicine. After reading this article from Reuters and the referenced study, we believe that a balanced perspective must be presented.

In his editorial, Livingston concluded that duodenal switch (DS) should not be offered and that there are better options for bariatric surgery. He stated that extra weight loss is not significant because medical conditions improve with nominal weight loss. This pays no attention to recidivism and the medical impact of weight regain. He did not analyze the techniques used in surgery and paid little attention to the statistically significant aspects of the study, instead highlighting the aspects that were not statistically significant. The summary provided was based on his bias, rather than the data generated in the actual study.

The study was conducted at two European Centers and randomized 60 patients with body mass indices (BMI) of 55kg/m2 or more to gastric bypass or DS. The bypass included a divided lesser curvature pouch and a 150cm alimentary limb with 50cm biliopancreatic limb. The switch included a sleeve gastrectomy over a 30 French bougie with a 200cm alimentary limb and 100cm common channel.

Patients were prescribed nominal vitamin supplementation. The authors of this article report the two-year follow-up data.
The data revealed that weight loss was far greater with DS with a reduction of 25 BMI units compared to 17, with a p value of <0.001. Improvement in cholesterol was also significantly better for DS, also with a p value of <0.0001. At two-year followup, 26 percent of the gastric bypass patients were still morbidly obese compared to none of the DS patients.

In his editorial, Livingston highlighted the higher rate of adverse events. However, when you look at post 30-day morbidity, there was no statistically significant difference between the DS and gastric bypass groups. There were 12 events outside of 30 days in the DS group and nine in the bypass group. This results in a P value of 0.32, which means that there is a better-than-30-percent chance that these results were due to random chance. Similarly, early complications in the DS group were slightly higher but without statistical significance.

In the Annals of Internal Medicine article, the authors commented on a case of protein malnutrition and their belief that there is a tendency for lower micronutrient levels in DS cases. In summary, this study showed better weight loss and cholesterol improvement in DS, offset by the possibility or probability of greater risk of protein and micronutrient deficiency. It also showed that the bypass has questionable efficacy in patients with super morbid obesity. Twenty-six percent of patients still had a BMI over 40kg/m2. With a mean BMI of 38kg/m2 at the nadir period of two years, it is safe to assume a significant number will be morbidly obese by the five-year point. The significance of this remains unknown, but it is doubtful that these are the results that patients desire. Furthermore, if weight regain does occur, medical problems will likely reappear.

More concerning is the construction of the DS that was utilized in this study. The Québec group (comprised of Dr. Simon Biron et al from Laval University, Québec, Canada) advocates a common channel of 75 to 100cm, but performs the sleeve gastrectomy over a 60 bougie. At Lenox Hill Hospital, we use a 38 bougie but lengthen the common channel to 125 to 150cm. With this approach, patients have 1 to 3 bowel movements daily. A DS over a 30 Fr bougie combined with a 100cm common channel is far more aggressive than standard.

At the Third International Concensus Summit for Sleeve Gastrectomy (December 2–4, 2010, New York, New York), the average bougie size for a standalone sleeve was between 34 and 36. If you combine a tight sleeve with a short common channel, it is not surprising that weight loss will be high and there will be a risk of significant malnutrition. What is surprising is that the post 30 day complication rate in the article by Søvik et al[2] was not statistically significant. Also, the supplementation regimen used in this study is standard for gastric bypass. Surgeons who perform DS use a far more robust protocol, emphasizing high supplements of vitamins A, D, E, K, and iron, calcium, and zinc.

The most important thing we can learn from this study is the importance of standardizing operations before we compare them. Of course, weight loss was higher with this extreme procedure. It would also be unfair to conclude that the weight loss would be as good with a larger sleeve or longer common channel. This is what we need to study.

The conclusion that the DS should not be done is not correct. Instead, what has to be determined is whether we can combine an appropriately sized sleeve gastrectomy with a post-pyloric bypass that provides better weight loss and less recidivism than gastric bypass, without increased nutritional side effects. We believe the answer to that question will be “yes.”

After performing thousands of bypass procedures, a substantial emphasis of my facility’s research has been on weight regain after gastric bypass. Initially, we were interested in endoscopic procedures to determine if we could retighten the bypass to restore restriction. Through the process we began to believe that the majority of relapse comes from inter-meal hunger. What may occur in many patients is an increase in compliance of the gastrojejunostomy and resultant loss of restriction. Consequently, food exits the pouch rapidly. This appears to stimulate a rapid rise in insulin, causing reduction of blood glucose. This crescendo/ decrescendo cascade may cause hunger and frequent eating in many patients with weight regain. This is especially true when foods high on the glycemic index are eaten.

We have begun to test this hypothesis in a prospective trial that we presented at the 2011 meeting of the Society of American Gastrointestinal and Endoscopic Surgeons (SAGES). In this study, we showed that six months following surgery, bypass patients had peak insulin levels greater than preoperatively when challenged with glucose. That rise was not seen in DS or sleeve patients.
As a result, we believe that preserving the pyloric valve is important in regulating the release of food from the pouch, and thus prevents this abrupt rise in insulin. We speculate that operations that preserve this important structure may give us better long-term results. Since the fundus is resected in SG and DS, emptying will never be normal. However, the pylorus may give us better and perhaps more physiologic, regulation than a gastrojejunostomy.

While we understand the reservations and technical challenge of performing a laparoscopic DS, we do not know of any physiologic reason that a bypass above the pyloric valve is better than one beneath the pylorus.

After performing close to 500 DS operations, we have reason to believe the lower anastamosis may be better. The marginal ulcer and stricture rates are lower. To date, we have seen better weight loss as well. By lengthening our limb lengths, we rarely get complaints of frequent bowel movements or significant malnutrition.

An emphasis needs to be placed on understanding what the intestinal bypass does and what is important. Is it the length of the total alimentary limb or the length of the common channel? Can we do a loop reconstruction, eliminating the distal attachment and make this procedure simpler? As more data are produced and our experience in this field grows, we must be careful not to use our preconceived bias when analyzing the data.

The article by Søvik et al[2] shows that gastric bypass will not get a significant number of patients to the BMI they desire, and also that a DS with a very small sleeve may be too aggressive. For meaningful studies to be performed, there needs to be standardization of operative technique and better understanding of the intestinal bowel lengths needed to maximize weight loss and minimize unintended side effects.

References
1.    Norton A. Rare surgery brings more weight loss, more risks. Reuters Health. September 6, 2011. http://www.reuters.com/article/2011/09/06/us-surgery-weightloss-idUSTRE7855YR20110906
2.    Søvik TT, Aasheim ET, Taha O, et al. Weight loss, cardiovascular risk factors, and quality of life after gastric bypass and duodenal switch: a randomized trial. Ann Intern Med. 2011;155(5):281–291.
3.    Livingston EH. Primum non nocere. Ann Intern Med. 2011;155(5):329–330.

With regards,

Mitchell S. Roslin, MD, FACS
Lenox Hill Hospital/North Shore-Long Island Jewish Health System (NSLIJ)
Northern Westchester Hospital Center
New York, New York

Paresh C. Shah, MD, FACS
Lenox Hill Hospital/North Shore-Long Island Jewish Health System (NSLIJ)
Northern Westchester Hospital Center
New York, New York

Category: Letters to the Editor, Past Articles