Correspondence
Bariatric Surgery and Mortality
N Engl J Med 2007; 357:2633-2634December 20, 2007
- Article
To the Editor:
Sjöström et al. (Aug. 23 issue)1 report that bariatric surgery was associated with decreased long-term overall mortality, but because of statistical limitations, they were not able to determine definitively whether this finding could be explained by weight loss or by other beneficial effects of the surgery. In view of the marked reduction in cancer-related deaths in the surgery group as compared with the control group (29 vs. 47 deaths), a further specification of the types of cancer that occurred would be interesting. If the decrease in cancer-related deaths could mainly be attributed to a reduced incidence of cancers for which obesity is known to be a risk factor, such as endometrial carcinoma,2,3 this would provide indirect evidence of the mortality-reducing effect of weight loss rather than surgery-related factors.
3 ReferencesDiane van der Woude, M.D.
Leiden University Medical Center, 2333 ZA Leiden, the Netherlands
dvanderwoude@lumc.nl 1
Sjostrom L ,Narbro K ,Sjostrom D , et al. Effects of bariatric surgery on mortality in Swedish obese subjects. N Engl J Med 2007;357:741-752
Full Text | Web of Science | Medline2
Giovannucci E ,Michaud D . The role of obesity and related metabolic disturbances in cancers of the colon, prostate, and pancreas. Gastroenterology 2007;132:2208-2225
CrossRef | Web of Science | Medline3
Modesitt SC ,van Nagell JR Jr . The impact of obesity on the incidence and treatment of gynecologic cancers: a review. Obstet Gynecol Surv 2005;60:683-692
CrossRef | Web of Science | Medline
To the Editor:
The study reported by Sjöström et al. showed that bariatric surgery in obese patients is associated with long-term weight loss and decreased overall mortality. However, lack of randomization of subjects may have introduced a major selection bias, which in turn may have influenced these results, permitting self-selection of highly motivated patients who could undergo surgery with its attendant complications to improve their health, as compared with a “control” group. Such motivated people would have been more likely to have a healthy lifestyle, including smoking cessation, engagement in physical activity, strict glucose monitoring, and good dietary habits, as compared with subjects in the control group. This might explain the sustained weight loss and reduced mortality in the surgery group and might also explain the gradual mortality benefit observed in the surgery group over a period of years, which does not coincide with the pattern of weight loss observed after surgery. It might be useful to reanalyze the results of this study after obtaining further information about these lifestyle variables at the end of each observation period and stratifying the data on the basis of these variables.
Jagan Beedupalli, M.D., M.P.H.
Louisiana State University Health Sciences Center, Shreveport, LA 71115
jaganbeed@yahoo.com To the Editor:
From a psychiatrist's perspective, I found two points of particular interest in the report by Adams et al. (Aug. 23 issue)1 on mortality after gastric bypass surgery. First, a doubling of the suicide rate among patients who underwent gastric bypass procedures was attributed to underlying psychopathology rather than to the treatment itself, as has been the case with antidepressants. Second, suicide was classified as a “nondisease” cause of death. The notion that completed suicide is unrelated to a disease flies in the face of modern psychiatry.
1 ReferencesMark Beale, M.D.
Charleston Psychiatric Associates, Charleston, SC 29407
bealem@aol.com 1
Adams TD ,Gress RE ,Smith SC , et al. Long-term mortality after gastric bypass surgery. N Engl J Med 2007;357:753-761
Full Text | Web of Science | Medline
Author/Editor Response
As van der Woude correctly points out, there seemed to be a substantial reduction in cancer-related mortality in the surgery group in our study. However, the study did not have sufficient power to prove that cancer-related mortality was, in fact, significantly reduced in the surgery group. Our study had even less power to look at mortality reduction with respect to specific cancers, but the qualitative impression we had was that deaths from obesity-related cancers as well as deaths from cancers unrelated to obesity were less common in the surgery group. Because of long-lasting effects of surgery, a beneficial effect on obesity-related cancers in the surgery group as compared with the control group would not necessarily have supported weight loss over surgery as the cause of the benefit.
We also agree with Beedupalli that the lack of randomization in our study is a drawback. However, a matched design was the only type of trial for which we could get ethical approval, owing to expected postoperative mortality. However, we think the likelihood is low that the favorable effect of surgery on mortality was due to a selection of subjects for surgery who had a healthier lifestyle. At the matching examination, subjects in the surgery group smoked more frequently than those in the control group (proportion of subjects who smoked daily, 27.9% vs. 20.2%), were 2.3 kg heavier, and had higher insulin levels and a lower current health score (see Table 1 of our article). Furthermore, between the matching and baseline examinations, the surgery group had an increase in weight, whereas the control group had a decrease. These divergent weight changes caused several risk factors to be higher in the surgery group at baseline (see Table 1 of our article). If anything, these observations suggest a less healthy lifestyle in the surgery group up to the start of the study intervention. In spite of the survival disadvantages at baseline, the surgery group had a lower mortality in univariate as well as multivariate analyses.
As we have reported previously, lifestyle improved more in the surgery group than in the control group after inclusion in the study.1 Thus, energy intake decreased and physical activity increased more in the surgery group. Thus, we believe that it is reasonable to state that our findings were due to the surgical procedure and its associated weight loss rather than to selection bias.
1 ReferencesLars Sjöström, M.D., Ph.D.
Ted Lystig, Ph.D.
Lena Carlsson, M.D., Ph.D.
Sahlgrenska Academy, 405 30 Göteborg, Sweden
lars.sjostrom@medfak. gu. se 1
Sjostrom L ,Lindroos A-K ,Peltonen M , et al. Lifestyle, diabetes, and cardiovascular risk factors 10 years after bariatric surgery. N Engl J Med 2004;351:2683-2693
Full Text | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
Andreas Oberbach, Martin von Bergen, Susann Blüher, Stefanie Lehmann, Holger Till. (2011) Combined Serum Proteomic and Metabonomic Profiling After Laparoscopic Sleeve Gastrectomy in Children and Adolescents. Journal of Laparoendoscopic & Advanced Surgical Techniques110929122032005
CrossRef
