Metabolic bariatric surgery (MBS) is the most effective long-term treatment of severe obesity1,2 with Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG) being the most common procedures worldwide3. Despite previous RCTs4–8, the comparative efficacy and safety of these procedures remains somewhat unclear. Although RCTs offer the highest level of evidence, there is an inherent challenge concerning external validity, that is the generalizability of outcome in RCTs compared to ‘real-world’ results9. The aim of this study was to perform exclusion- and external validity-analysis of the Swedish patients included in the Bypass Equipoise Sleeve Trial (BEST), a Swedish–Norwegian RCT comparing SG and RYGB, compared to all other patients not included in BEST who underwent MBS during the same time period in Sweden and were registered in the Scandinavian Obesity Surgery Registry (SOReg). Briefly, BEST10 is an RCT comparing SG and RYGB with two co-primary endpoints: non-inferiority for weight loss and superiority for risk of adverse events (AE) both assessed at 5 years. Adult patients in Sweden and Norway with BMI 35–50 kg/m2 were randomized 1:1 to either RYGB or SG. The study design and perioperative outcomes have been previously reported in detail10,11. SOReg, a nationwide quality registry with high validity capturing virtually all MBS in Sweden12, is the case report form for BEST enabling comparison of Swedish BEST outcomes to eligible non-included patients, as well as the general MBS population. The exclusion analysis compares the eligible included patients in BEST to eligible but non-randomized patients declining to participate at the 20 participating surgical centres. The external validity analysis compares BEST patients to all other patients who underwent a primary SG or RYGB. Any statistical differences between groups were additionally assessed regarding clinical relevance. Further methodological details are available in the Supplementary material. During the inclusion period (April 2016–March 2022), 1556 Swedish BEST patients, 6816 eligible non-randomized patients, and 13 101 MBS patients not evaluated for RCT inclusion underwent RYGB or SG in Sweden. Figure S1 shows the distribution and follow-up of all 21 473 patients. Differences in baseline characteristics between randomized and non-randomized eligible patients were minor and not considered clinically relevant according to the prespecified margins, except for dyspeptic symptoms requiring pharmacological treatment (Table 1). The proportion of patients with dyspepsia did not differ between RCT and eligible non-randomized patients undergoing SG, whereas the eligible non-randomized RYGB patients had a higher proportion of dyspeptic symptoms compared to the RCT RYGB patients (Table 2). There were no clinically relevant differences between the BEST and non-randomized eligible patients with regard to AE (Fig. 1a, Table S3). a Exclusion analysis, adverse events in patients eligible for BEST, included versus not included. There were no deaths within 30 days, and 1 death within 90 days (RYGB not included). b External validity analysis, adverse events in BEST versus non-BEST patients. There were 4 deaths within 30 days (RYGB non-BEST) and 8 deaths within 90 days (2 SG non-BEST, 6 RYGB non-BEST) Exclusion analysis—patient characteristics at time of surgery: patients eligible for BEST, included versus not included Presented as number (percent) or mean(sd). *Missing data—smoking: SG BEST 6, Eligible not included 22, RYGB BEST 2, Eligible not included 27. †Missing data—prior deep vein thrombosis (DVT) or pulmonary embolism (PE): RYGB BEST 2. Exclusion analysis—patient characteristics at time of surgery: patients eligible for BEST, included versus not included Presented as number (percent) or mean(sd). *Missing data—smoking: SG BEST 6, Eligible not included 22, RYGB BEST 2, Eligible not included 27. †Missing data—prior deep vein thrombosis (DVT) or pulmonary embolism (PE): RYGB BEST 2. There were minor differences in baseline characteristics between RCT and MBS patients not evaluated for RCT enrolment, which were not considered clinically relevant according to prespecified margins. RCT patients operated with SG had more co-morbidities compared to the patients not eligible for randomization (Table 2). RCT patients operated with RYGB had a lower readmission rate compared to patients undergoing RYGB outside RCT enrolment, but there were no clinically relevant differences in AEs between RCT and patients not eligible for randomization (Fig. 1b, Table S4). External validity analysis—patient characteristics at time of surgery: external validity, BEST versus non-BEST Presented as number (percent) or mean(sd). *Did not reach prespecified margin of clinically significant difference. †Missing data—smoking SG: BEST 6, non-BEST 462, RYGB BEST 2 non-BEST 377. ‡Missing data—prior deep vein thrombosis (DVT) or pulmonary embolism (PE): SG BEST, 0 non-BEST 30, RYGB BEST 2, non-BEST 17. External validity analysis—patient characteristics at time of surgery: external validity, BEST versus non-BEST Presented as number (percent) or mean(sd). *Did not reach prespecified margin of clinically significant difference. †Missing data—smoking SG: BEST 6, non-BEST 462, RYGB BEST 2 non-BEST 377. ‡Missing data—prior deep vein thrombosis (DVT) or pulmonary embolism (PE): SG BEST, 0 non-BEST 30, RYGB BEST 2, non-BEST 17. This study shows that randomized patients in the BEST RCT were generally representative of all patients who underwent MBS with SG and RYGB in Sweden during the study period. The only clinically relevant difference in the baseline characteristics was a lower prevalence of reported dyspeptic symptoms among patients undergoing RYGB in BEST compared to both RYGB comparison groups. This likely represents the exclusion of patients with symptoms indicating gastroesophageal reflux disease (GERD) from BEST. A higher prevalence of GERD was only seen in patients not eligible for RCT enrolment reflecting that RYGB was the preferred procedure in patients with GERD4, that is the results from BEST may not be fully generalizable to patients with GERD. The differences in both adverse events and serious adverse events between SG and RYGB in this entire study appear lower than previously reported8. The lower risk may be an effect of the improvements and ‘maturity’ in surgical techniques, especially for RYGB in Sweden. The 30-day complication rates also compare favourably to the benchmark of 5.5% for Clavien–Dindo 3a or higher at 90 days as suggested by Gero et al.13. As these results represent national MBS practice, this low risk should be within reach by other countries and centres. As there were no clinically relevant differences in the baseline characteristics and perioperative results in BEST RCT were regarding both eligible non-randomized patients and overall Swedish MBS practice in adult patients with BMI 35–50 kg/m2, the future results from the RCT can be considered generalizable in larger MBS cohorts. This work was supported by The Swedish Research Council (Medicine and Health), Region Västra Götaland, Region Östergötland, Region Örebro County, The Erling-Persson Foundation (Grant # 2022 0163 to A.T.). None of the sponsors had any role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. M.N. was funded by the Swedish Research Council (grant no. 2019-00738). Suzanne Hedberg (Data curation, Investigation, Methodology, Writing—original draft, Writing—review & editing), Erik Stenberg (Investigation, Resources, Writing—review & editing), Johanna Österberg (Investigation, Writing—review & editing), Ellen Andersson (Investigation, Writing—review & editing), Markku Peltonen (Formal analysis, Methodology, Software, Supervision, Validation, Writing—review & editing), Erik Näslund (Conceptualization, Investigation, Writing—review & editing), Martin Neovius (Conceptualization, Writing—review & editing), Anders Thorell (Conceptualization, Project administration, Writing—review & editing), Torsten Olbers (Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Writing—review & editing), and Johan Ottosson (Conceptualization, Data curation, Funding acquisition, Investigation, Methodology, Project administration, Supervision, Writing—review & editing) This study was submitted as an original article, and has now been converted to a research letter. An abstract was presented as an oral presentation at the IFSO-EC meeting in Vienna, Austria on 2 May 2024, and at the 7th Nordic–Baltic Bariatric meeting in Turku, Finland on 5 June 2024. Unrelated to this work, T.
No takes yet. Share an insight, caveat, or question.
Hedberg et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: