Roux‐en‐Y versus one‐anastomosis gastric bypass (RYSA study): weight loss, metabolic improvements, and nutrition at 1 year after surgery, a multicenter randomized controlled trial

Although it has been suggested that one‐anastomosis gastric bypass (OAGB) is metabolically superior to the “gold standard,” i.e., Roux‐en‐Y gastric bypass (RYGB), there is little robust evidence to prove it. Because this result may arise from the typically longer length of bypassed intestine in OAGB, here, the authors standardized the bypass length in RYGB and OAGB and compared weight loss and metabolic outcomes in a randomized controlled trial.

marked and similar (p > 0.05) but the use of antihypertensive medications was lower (p = 0.037) and hypertension tended to improve more (p = 0.053) with RYGB versus OAGB at 12 months.Higher rates of vitamin D-25 deficiency (p < 0.05) and lower D-25 levels were observed with OAGB versus RYGB throughout the follow-up (p < 0.001).No differences in adverse effects were observed.
Conclusions: RYGB and OAGB were comparable in weight loss, metabolic improvement, remission of diabetes and hypercholesterolemia, and nutrition at 1-year follow-up.Vitamin D-25 deficiency was more prevalent with OAGB, whereas reduction in antihypertensive medications and hypertension was greater with RYGB.There is no need to change the current practices of RYGB in favor of OAGB.

INTRODUCTION
Roux-en-Y gastric bypass (RYGB) has for long been the "gold standard" for bariatric surgery.One-anastomosis gastric bypass (OAGB) was introduced in 2001 as a simplified but similarly beneficial procedure in terms of weight loss for grade 3 obesity [1], but it has been met with controversial reception in the field.The benefits of both RYGB and OAGB for weight loss and metabolism in the treatment of severe obesity are evident [2][3][4][5][6][7].To date, OAGB has not been shown to be metabolically superior or generally safer than RYGB in the few randomized controlled trials conducted [8,9].
In randomized controlled trials, similar weight loss [8,9], metabolic improvements [8,9], and remission of T2DM [9] or metabolic syndrome [8] but increased rates of nutritional adverse events, diarrhea, and steatorrhea [9] and lower levels of hemoglobin [8] have been reported with OAGB compared with RYGB.Thus far, there is a lack of evidence of better metabolic or insulin-sensitizing effects with OAGB when the length of bypassed intestine is similar to that of RYGB.
In this trial, we compared RYGB and OAGB, standardized with equal lengths of bypassed intestine, at 6 months and 1 year post procedure.
Nine patients were excluded as screening failures (5 withdrew consent, 4 were not eligible).One patient from the RYGB group withdrew consent immediately after randomization, resulting in 60 patients for RYGB and 60 patients for OAGB at study baseline.One operation, originally allocated to RYGB, was converted to sleeve gastrectomy due to intra-abdominal adhesions discovered during surgery, resulting

Study Importance
What is already known?
• It has been suggested that one-anastosmosis gastric bypass (OAGB) more frequently leads to remission of type 2 diabetes mellitus than Roux-en-Y gastric bypass (RYGB).
• OAGB may lead to malnutrition more often than RYGB.
• There are few randomized controlled trials on the subject with standardized lengths of bypassed intestine.

What does this study add?
• RYGB and OAGB are comparable regarding weight loss and metabolism at 1 year.
• RYGB and OAGB result in similar improvements in glucose tolerance.
• Vitamin D-25 deficiency is more prevalent with OAGB.
• RYGB may lead to a better reduction in hypertension.

How might these results change the focus of clinical practice?
• There is no need to change current practices of RYGB in favor of OAGB.
• OAGB may be selected when appropriate because it may be easier to conduct in certain patient groups who are not prone to malnutrition or bile reflux. in 59 patients operated on for RYGB.Three patients were lost to followup.Therefore, the final number of patients was 56 (14 men, 42 women) for RYGB and 60 (19 men, 41 women) for OAGB at 12 months.
The patients for this study were recruited from two academic centers, the Department of Gastrointestinal Surgery, Helsinki University Hospital (Helsinki, Finland) and the Department of Gastrointestinal Surgery, Oulu University Hospital (Oulu, Finland) [21].

Trial design
The primary outcome of the Roux-en-Y Gastric Bypass versus Single-Anastomosis Gastric Bypass (RYSA) trial was the difference in percent excess weight loss (%EWL) at 1 year.In this article, we report the 6-and 12-month results of %EWL; percent total weight loss (%TWL); anthropometric and body composition measurements (bioimpedance); metabolic, i.e., glucose homeostasis by homeostatic model assessment (HOMA) index and oral glucose tolerance test (OGTT; secondary outcome), lipids, and basic metabolic markers; nutritional and inflammatory values; remission of metabolic diseases; and medication use between RYGB and OAGB patients.
The full trial design and eligibility criteria have been previously published and described in detail [21].In brief, patients were required to lose 5% of total body weight before being accepted for surgery during 6 months of conservative treatment supervised by an endocrinologist or an internal medicine specialist.A surgeon thereafter evaluated candidates' suitability for surgery and for the study.The participants were examined at steady state circa 8 weeks before surgery (baseline).Subsequently, participants underwent a 4-to 6-week very low-caloric diet until the operation.The same surgeons, clinicians, and study nurses evaluated the patients, operated on them with standardized procedure, and followed the patients throughout the trial.Continuous training for this was applied.One chief surgeon in Helsinki and one in Oulu supervised all procedures.A summary of the trial protocol is presented in Figure 1B.
Sample size was determined to detect a 10% difference in EWL between the groups at 2 years [21].After trial commencement, the primary endpoint defined in ClinicalTrials.gov(NCT02882685) was decided to be assessed at 1 year instead of 2 years and a full followup protocol of 1, 3, 6, 12, and 24 months and 5 and 10 years to avoid a multiplicity problem in the analyses.
The study was designed as a superiority trial based on a previous cohort series [8].According to power calculations, with the assumptions of mean %EWL = 60 (SD = 17) in the RYGB group at both 1 and 2 years, and with α = 0.05, 50 patients at 1 and 2 years in each group would be sufficient to reach a power of 80%.A dropout rate of 10% at 2 years was considered, i.e., 55 patients per group [8].A dropout rate of 7% at 1 year was considered, i.e., 54 patients per group.Recruitment was set at 60 patients per group.Allocation ratio was 1:1.Power calculations were performed by simulation using the Mann Whitney Wilcoxon test (PASS version 13.0, NCSS Inc., Kaysville, Utah).
The trial was approved centrally by the Helsinki University Hospital Ethics Committee (HUS/1706/2016) and reviewed by the Helsinki University Hospital Research Review Board (HUS269/2017).Oulu University Hospital did not require additional local ethical approval.
Informed consent was obtained from all study participants.The trial is registered at ClinicalTrials.gov(NCT02882685).

Trial treatments
In RYGB, at 5 cm below the gastroesophageal junction, an $20 to 40-mL small pouch was created with linear staplers.The omentum was divided, and an antecolic 80-cm biliopancreatic (BP) limb was anastomosed endto-side with a linear 45-mm stapler.A 130-cm alimentary limb was enteroanastomosed with a 60-mm linear stapler.Both mesenteric defects were closed.In OAGB, a tubular gastric pouch was created along a 38Fr bougie, starting at the crow's foot and using staplers.An antecolic 210-cm BP limb was anastomosed end-to-side with a linear 45-mm stapler.Stiches between end of the biliary limb (BL) and gastric pouch were done to prevent biliary reflux [22].A 210-cm biliopancreatic limb in OAGB and 80-cm (biliopancreatic) and 130-cm (alimentary) limbs in RYGB were chosen to obtain a similar amount of excluded small intestine in both groups.
The lengths of the loops were measured with standard instruments.The procedures have been described in detail previously [21].

Primary outcome: assessment of weight loss
Weight and height were measured after an overnight fast in light clothing.Weight loss was reported as kilograms of weight lost as % TWL and as %EWL at 6 and 12 months.%TWL was calculated as the percentage of weight lost since the baseline visit, before the preoperative very low-caloric diet (Figure 1B).%EWL was calculated as the percentage of excess weight lost since the baseline visit compared with an ideal weight with a body mass index (BMI) of 25 kg/m 2 .The patients were grouped into categories according to the amount of weight lost.For %TWL, the categories were <10%, 10% to 19.9%, 20% to 29.9%, 30% to 39.9%, and ≥40%.For %EWL, the categories were 0% to 24.9%, 25% to 49.9%, 50% to 74.9%, and ≥75%.

Body composition
Whole-body composition was measured using a Bioelectrical Impedance (BIA) Tanita MC-980 (patients in Helsinki) or BIA InBody 720 (patients in Oulu).

Glucose homeostasis
An OGTT of 3 h (75 g glucose) was performed after an overnight (10 h) fast.A fasting blood sample was collected before ingestion of the glucose drink, and post-glucose samples were collected at 0, 30, 60, 120, and 180 min.Glucose, insulin, and C-peptide levels were determined, and HOMA insulin resistance and Matsuda insulin sensitivity indices were calculated, as described elsewhere [23,24].

Metabolic and nutritional status
Routine laboratory tests including blood count, high-sensitivity C-reactive protein (CRP), lipids, and thyroid and liver function tests were performed, as described previously [23,25].Circulating nutritional status was assessed by a panel of vitamins and nutrients using standard methods in the accredited laboratory of the Helsinki University Hospital, HUSLAB (www.huslab.fi/ohjekirja).

Use of medications and nutrients
Use of medications and vitamin/nutrient supplements was recorded before the start of the trial and at each visit through an interview and questionnaires, the medical records of the hospital, and the National Medication List of the Finnish Institute for Health and Welfare.After the operation, patients were instructed to take the following supplements: multivitamins containing vitamins A, E, B, and C; and the minerals folate, selenium, magnesium, and zinc, as well as separate supplementations for vitamin D-25 (30 μg), vitamin B12 (1 mg), iron (50-100 mg), and calcium (1 g).Individual plasma concentrations were recorded at visits, and supplements were adjusted accordingly.Nutrient deficiencies were defined by the standards of American and European recommendations for surgical patients and by the standards of Helsinki University Hospital, HUSLAB (Table S1).

Quality of life
Quality of life was assessed through 15D-instrument of health-related quality of life questionnaire [26] and Bariatric Analysis and Outcome Reporting System (BAROS) [27].

Statistical analyses
Statistical analyses were performed using Stata statistical software (Release 16.0, StataCorp LLC, College Station, Texas).Primary and secondary outcome analyses were conducted in the per-protocol population, including all patients who were randomized and provided data.Independent t test was used for continuous normally distributed variables, with log-10 correction for continuous non-normally distributed data and Fisher exact test for categorical/ordinal variables to test the group differences (RYGB vs. OAGB) at baseline and at 6 and 12 months.Repeatedmeasures ANOVA with Tukey post hoc tests was used for continuous variables to test the differences among the three visits (baseline, 6 months, and 12 months) in RYGB and OAGB.Non-normally distributed variables were log-10-corrected for analysis.Friedman one-way repeatedmeasures ANOVA by ranks with Wilcoxon signed rank tests as post hoc was used for categorical variables to test the differences among the three visits in RYGB and OAGB.A p value < 0.05 was considered significant.
Changes in body composition did not differ between groups at 6 months (Table 1, Table S2).

Metabolic outcomes
Metabolic parameters in the RYGB and OAGB groups improved during the follow-up, with no difference between groups (Table 1, Table S2, Figure 3B).The procedures resulted in similar improvements in fasting plasma glucose and insulin, HbA1c, and HOMA and Matsuda indices between baseline and 6 and 12 months (Figure 3B).Lipid parameters showed a decrease in total and LDL cholesterol and triglycerides and an increase in HDL cholesterol (Figure 3B).High-sensitivity CRP protein and leukocytes decreased similarly, indicating resolution of low-grade inflammation.Hemoglobin and glucose tolerance decreased in both groups between baseline and 6 and 12 months ( p < 0.05).Blood pressure values decreased significantly and similarly, except pulse rate at 6 months, which decreased more in the OAGB group ( p = 0.035; Table 1, Table S2).The respective values at 6 and 12 months in other clinical parameters are shown in Table 1 and Table S2.These values were similar between groups at baseline and at 6 and 12 months ( p > 0.05).

Remission of metabolic diseases and reduction in medication use
Both surgeries had a significant effect on remission of metabolic diseases (

(A) (B)
F I G U R E 3 Body composition and metabolic parameters between RYGB and OAGB at baseline, operation, and at 6 and 12 months.(A) Significant but similar improvements in body composition between RYGB and OAGB (the loss of fat mass and fat kilograms).There were no differences between the groups in the loss of muscle mass.(B) Significant improvements in various metabolic markers by insulin, glucose, lipids, oral glucose tolerance test, and inflammation levels at operation and at 6 and 12 months with RYGB and OAGB.There were no differences between the groups.T2DM in both the RYGB and in OAGB groups.Remission of T2DM was observed in 73% of those in the RYGB group versus 70% of those in the OAGB group at 6 months (p = 0.990) and in 92% of those in the RYGB group versus 82% of those in the OAGB group at 12 months (p = 0.442).Prevalence of hypertension was 71% in the RYGB group and 60% in the OAGB group at baseline, with remission rates of 48% versus 22% at 6 months (p = 0.440) and 55% versus 25% at 12 months (p = 0.053), respectively.Prevalence of dyslipidemia was 76% in the RYGB group and 81% in the OAGB group at baseline, and remission was observed in 41% versus 30% at 6 months (p = 0.505) and 57% versus 58% at 12 months (p = 0.990; Table 3), respectively.
The proportion of persons having medications for T2DM and hypercholesterolemia in the RYGB and OAGB groups were significantly ( p < 0.05) and similarly ( p > 0.05) reduced during the follow-up (Table 4).The number of persons using antihypertensive medications was reduced after RYGB and OAGB at and 12 months compared with baseline ( p < 0.05).However, the reduction was significantly greater in the RYGB (61%) than in the OAGB (28%) group at 12 months ( p = 0.037 between groups).The number of medications used per person for T2DM, hypercholesterolemia, and hypertension were similar between the RYGB and OAGB groups during the weight loss (Figure S1).
F I G U R E 4 Levels of vitamins and nutrients during the weight-loss period at baseline and at 6 and 12 months compared with baseline in both RYGB and OAGB groups.Data are presented as mean ± SE.P values between RYGB and OAGB timepoints were calculated by t test.P values within timepoints in RYGB and OAGB were calculated with repeated-measures ANOVA and Tukey post hoc tests.Non-normally distributed continuous variables have been log10-corrected.***p < 0.001, **p < 0.01, *p < 0.05.OAGB, one-anastomosis gastric bypass; RYGB, Roux-en-Y gastric bypass.
Changes in nutritional parameters and rates of nutritional deficiencies RYGB and OAGB resulted in significant reductions in plasma levels of zinc, copper, vitamin A, and vitamin E at 6 months and in ferritin, zinc, copper, and vitamin E at 12 months (Figure 4, p < 0.05), with increased phosphate levels at 6 and 12 months ( p < 0.05).In the RYGB group, increased folate ( p = 0.027) and vitamin D-25 ( p = 0.008) levels and, in the OAGB group, increased vitamin B1 ( p = 0.023) and magnesium ( p = 0.049) levels at 12 months were observed.OAGB patients had decreased vitamin A ( p = 0.039) levels at 12 months.Vitamin D-25 level was the only parameter that was significantly lower in the OAGB than in RYGB group at 12 months ( p = 0.001; Figure 4).
At baseline, patients from the RYGB and OAGB groups had deficiencies in albumin, iron, phosphate, and vitamins D-25 and B1 and, at 6 months, in albumin, iron, D-25, and B1; however, B1 levels had improved by 12 months (p < 0.001; Table S3).Iron deficiency was more common in both during the follow-up (p < 0.01).Deficiency in D-25 was already greater in the OAGB group at baseline (OAGB 81% vs. RYGB 61%) and remained so at 6 months (72% vs. 53%, p = 0.049) and 12 months (75% vs. 41%, p < 0.001).Incidence of anemia or hypoalbuminemia did not differ between the groups, although a greater increase in anemia was observed with RYGB at 6 and 12 months (p = 0.014; Table S3).The amounts of vitamins used (Table S4) and the number of persons using supplements (Table S5) increased during follow-up as expected but were similar between groups at each timepoint (p > 0.05).

Adverse events and operative aspects
One patient had a laparoscopic cholecystectomy and a hernioplasty due to a primary epigastric hernia 11 months after RYGB.Additional

DISCUSSION
In this randomized controlled trial, RYGB and OAGB with similar standardized lengths of the bypassed intestine had comparable outcomes in weight loss, improvement in glucose homeostasis and dyslipidemia and other markers of metabolism and nutrition, and incidence of adverse events at 6 and 12 months after the operations.RYGB resulted in greater improvement in hypertension, whereas vitamin D-25 deficiency was more common in OAGB patients.
Comparison of RYGB and OAGB has gained increased interest and attention in recent years.We report similar weight loss between the procedures.Previous studies have described similar weight loss [13] or weight loss and improvements in metabolism [2,3,15,16] between the procedures.In contrast, Disse et al. [10] reported superior %EWL with OAGB compared with RYGB at 6 and 12 months (76.3% vs. 60.0%and 89.0% vs. 71.0%,respectively).Bhandari et al. [11] noted better weight loss with OAGB than with RYGB (%EWL 81.6% vs. 66.7%) and better resolution of comorbid conditions with OAGB at 5 years.Almalki et al. [14] reported better %EWL at 1 year, and Magouliotis et al. [3] reported better %EWL at 1, 2, and 5 years.However, these studies had variable lengths of bypassed intestine.Our trial concurs with the similar weight loss between RYGB and OAGB in previous randomized trials [8,9] and in the trial using standardized bypasses [9].
Our trial showed similar impact on T2DM between the procedures based on detailed assessment of glucose tolerance (remission of T2DM; reduction in T2DM medications; fasting plasma glucose and insulin; HbA1c; and glucose and insulin area under the curve from OGTT and HOMA and Matsuda indices).Previous studies have indicated supc in improved insulin sensitivity [3,13,14].Higher rates of T2DM remission with OAGB have been described in a meta-analysis of seven studies by Magouliotis et al. [3] and by Almaki et al. [13].
One study showed a better reduction in HbA1c, but no difference in remission of T2DM [15].In contrast, no differences in the improvement of T2DM between the surgeries were observed at 6 months by Disse et al. [10].Almaki et al. [13] defined remission of T2DM as HbA1c < 6% and no medication use.Our trial used an HbA1c limit of <6.5%, no medication use, and normal OGTT (fasting and 2-h glucose), which may explain the differences in results along with similar length of bypassed small intestine.The bypass length in our trial was standardized between procedures to 210 cm.This is in contrast to the previously mentioned studies, which used more bypassed small intestine for OAGB [13,10].Indeed, bypass length may contribute to T2DM remission [19,20].The randomized controlled trials did not report differences between RYGB and OAGB in improvement of glucose tolerance.Lee et al. described similar remission of T2DM between RYGB and OAGB [8].Robert et al. [9] did not observe differences between the procedures in HbA1c or in remission of T2DM with standardized limb lengths, although the mean decrease in HbA1c was greater in the OAGB group.
Although we observed similar changes in body composition, metabolic profile by blood tests, and remission of dyslipidemia between the groups, there was a slight improvement in hypertension in favor of RYGB.Remission of hypertension and dyslipidemia were similar between RYGB and OAGB in Disse et al. [10] and also in the metaanalysis by Magouliotis [3].Our findings and those of others are thereby consistent with the randomized controlled trials [8,9] and iron.However, some of the deficiencies may require longer follow-up periods to develop.Consistent with our data, a recent study found no differences between procedures in nutritional status (using the Controlling Nutritional Status [CONUT] nutrition score) or malnutrition at 6 months [28].After OAGB, vitamin deficiencies [11] and even malnutrition [3,17] have been reported.Some studies [3,9,19] have suggested that a bypass limb over 200 cm may lead to nutritional problems.We did not observe differences between the procedures with bypassed limbs of 210 cm or in malnutrition in terms of low albumin levels.Our results are consistent with the randomized controlled trial of Robert et al. [9], who described similar vitamin levels and deficiency rates at 2 years after the operations.The trial of Lee et al. described lower hemoglobin levels at 2 years after OAGB but did not assess vitamin or nutrition status [8].Some of the studies reporting malnutrition had longer follow-up periods than ours (1 y), which may explain the differences.Additionally, intensive nutritional counseling and support for systematic use of multivitamins were provided during our trial.
Our trial is one of few to compare weight loss, body composition, metabolism, and nutrition between RYGB and OAGB in detail, with standardized lengths of the bypassed intestine, thereby suggesting that the metabolic and insulin-sensitizing effects of the procedures are independent of the procedures themselves.At 1 year, OAGB is not superior to RYGB in weight loss, glucose homeostasis, metabolism, or nutrition, and procedures have similar rates of adverse events, similar operation times, and similar effect on the improvement of quality of life.Long-term disadvantages of OAGB may include greater incidence of hypoglycemia, malabsorption [3,11,17], and bile reflux [29].
Bariatric surgery has better outcomes in weight loss and remission of T2DM, metabolic syndrome, and hypercholesterolemia than caloric restriction or medication [6,7,[30][31][32][33].However, bariatric surgery is not suitable for all persons with obesity.RYGB and the other mainstream bariatric procedure, sleeve gastrectomy, have comparable outcomes in weight loss and remission of comorbidities in the randomized controlled trials performed [34,35] or greater weight loss is observed with RYGB [36,37].Randomized controlled trials on OAGB versus sleeve gastrectomy reveal similar %EWL and remission of comorbidities at 1 year [38] or 3 years [39] after the procedures.
There is therefore no current evidence to suggest superiority of OAGB over RYGB or sleeve gastrectomy [40].
The strengths of this trial include high attendance throughout the follow-up, as usual in Scandinavia, with few dropouts.A large proportion of the eligible patients were randomized.We evaluated glucose homeostasis, body composition, and metabolism in detail and more broadly than previous studies.We recorded adherence to vitamin supplementation by standardized questionnaires and medication lists,

1
Trial profile.(A) CONSORT diagram of participants and trial design.(B) Study protocol included clinical visits at baseline ($8 weeks before the operation) and at 6 and 12 months.After the baseline visit, the patients started a VLCD of 4 to 6 weeks, ending on the day of surgery.Anthropometric measurements, body composition, and blood tests were performed at each visit.The full study protocol, including visits at operation, 1 month, and 3 months and continuing up to 2, 5, and 10 years, has been described in detail previously [21].CONSORT, Consolidated Standards of Reporting Trials; VLCD, very low-caloric diet; OAGB, one-anastomosis gastric bypass; RYGB, Roux-en-Y gastric bypass.[Color figure can be viewed at wileyonlinelibrary.com] diagnostic endoscopy was performed in one (RYGB) versus four (OAGB) patients ( p = 0.369), nausea was reported by seven versus six patients ( p = 0.769), diarrhea was experienced by four versus ten patients (p = 0.402), abdominal pains were ten versus eight patients (p = 0.450), problems with eating were two versus three patients (p = 0.999), and self-reported dumping was twenty versus fourteen patients (p = 0.148), respectively.Operation time did not differ between RYGB versus OAGB (84.9 min [95% CI: 80.1-89.7] vs. 82.3min [95% CI: 76.6-87.9],p = 0.289).
which have not been reported in previous studies.A possible limitation of this trial is brief follow-up until now.Weight regain, nutritional deficiencies, and metabolic and other complications (such as anastomotic ulcers or internal herniation) may require a longer time to develop.Development of hypoglycemia as an adverse event can be better assessed with longer follow-up.The full trial protocol includes visits at 2, 5, and 10 years after the operation, which will address these questions.Additionally, some variation between measuring the length of the small bowel may have been possible during the procedures performed (10%-20% variation generally), even though a standardized technique was used for measuring.Our trial suggests similar weight loss and improvement in metabolism and comorbidities with RYGB and OAGB at 1 year, without superiority of OAGB over RYGB in glucose tolerance.RYGB patients may have a greater reduction in antihypertensive medications and greater improvement in hypertension, whereas OAGB patients may be more prone to vitamin D-25 deficiency.Although future studies may identify certain subgroups in which OAGB may be preferential, we suggest that RYGB should thus far remain gold standard.O AUTHOR CONTRIBUTIONS Sini Heinonen wrote the manuscript, figures, and tables and participated in data collection and the clinical examinations of the patients.Anne Juuti and Kirsi H. Pietiläinen were responsible for the original study design and patient recruitment.Anne Juuti, Tuure Saarinen, Henna Sammalkorpi, Anne K. Penttilä, Sanna Meriläinen, and Vesa Koivukangas performed the surgical procedures and coordinated the patient follow-up.Anne Juuti, Kirsi H. Pietiläinen, Tuure Saarinen, Sini Heinonen, Sanna Meriläinen, Henna Sammalkorpi, Anne K. Penttilä, Ulla Säiläkivi, Minna Koivikko, and Milla Rosengård-Bärlund examined the patients at the clinical visits.Jaro Karppinen and Pertti Siira provided the body composition data of the patients from Oulu University Hospital.All authors participated in writing of the manuscript and read and approved the final version.and laboratory personnel (Tarja Hallaranta, Elisa Silvennoinen, Emma Paasikivi, Lily-Marleen Trei, and Hanna Kukkonen).

Table 2 ,
, Figure 2).At 12 months, mean weight loss in RYGB versus OAGB was 31.7 kg (95% confidence interval [CI]: 28.8-34.6)versus 33.3 kg (95% CI: 30.4-36.1; Figure Clinical characteristics of the patients Values are mean (95% CI).P values between RYGB and OAGB are calculated by independent t test.P values within timepoints in RYGB and OAGB are calculated with repeated-measures ANOVA and Tukey post hoc tests.Non-normally distributed continuous variables have been log-10-corrected.Per-protocol population analysis.Significant values have been bolded and underlined.n* refers to laboratory or other data available.