To the Editor I read with great interest the article by Hjorth et al, 1 where nonrandomized patients from the Swedish Obese Subjects (SOS) study were examined for revisional bariatric surgeries (ie, banding, vertical banded gastroplasty [VBG], and gastric bypass [GBP]). Revisions were frequent after banding (40.7%) and VBG (28.3%) but rare after GBP (7.5%). This article, dealing with reoperations, should have included all kinds of interventions. Ingmar Näslund, MD, PhD, one of the authors, 1 reported increased admissions for gastrointestinal tract surgery (24.4%), gallstone problems (7.3%), and internal hernia (4.3%) following GBP. 2 The combined aspects would boost reoperations to more than 30%.There is an assumption that all bariatric operations can be equally revised and easily reoperated, but this is not the case. Following open GBP, there are multiple technical issues and difficulties; adhesions are present that are perhaps severe and extensive, often with hernias or mesh repairs, and are complicated by retrocolic or retrogastric jejunal limbs, and various choices exist for revisions. These permutations and lack of long-term results may explain why bariatric surgeons are hesitant to perform revisions in patients who have received GBP. Valid options include the addition of a band to a pouch or distalization of the Roux (both rarely performed in Sweden) or conversion to duodenal switch (rarely performed as a primary operation in Sweden). Those, I think, explain the lower revision rates of GBP in this study and the fact that revisions for other procedures are simpler and easier.Hjorth et al 1 hypothesized, "The greater weight loss after GBP 25 is the likely main reason for a lower request for conversions in this subgroup compared with the subgroups that undergo banding and VBG." One has to see that body weight after GBP in the SOS study has climbed from a nadir of 32% body weight loss after 2 years to 25% after 10 years. This implies that, with a very conservative estimate, at least 15% to 20% of patients have failed and that patients who received GBP were either denied revisions for lack of valid options or fear of technical difficulties.Further, Sjöström 3 reported in 2004 that of patients who underwent surgery with remission of diabetes at 2 years, 50% had relapsed after 10 years. Näslund has also seen a nearly 40% relapse at 7 years and de novo diabetes in patients without diabetes after GBP. 4 Therefore, patients with diabetes should have been candidates for revisions but were not subjected to reoperations in the present study. 1 Of note, only 7.4% of surgical patients in the SOS study had diabetes at the time of surgery, while most patients in North America have higher rates of diabetes (20% to 30%).In an Invited Commentary, Cohen 5 suggested that we should be doing more Roux-en-Y gastric bypass and less of the other operations (ie, sleeve gastrectomy) based on lower revisions rates. I caution the reader that the reasons explained above make this conclusion unattainable, apart from the fact that sleeve ga...