Comparison of percentage excess weight loss and body composition after Roux-en-Y gastric bypass versus sleeve gastrectomy: A prospective study

Introduction

Bariatric surgery is the last treatment option for morbid and severely obese patients (body mass index (BMI) ≥ 40 or a BMI ≥ 35 with serious comorbidities) who cannot achieve weight loss with non-surgical treatment. It has been considered a safe and effective treatment for obesity and its comorbidities in the long term.¹ Currently, it is shown to reduce body weight, resulting in complete remission or significant improvement of type 2 diabetes mellitus and other metabolic abnormalities, including hyperlipidemia, obstructive sleep apnea, non-alcoholic fatty liver, and hypertension.^1,2 Furthermore, the benefits of bariatric surgery include long-term weight loss, quality of life improvement, and the reduction of health care expenses.

The purpose of the bariatric procedure is to limit storage volume of the stomach and hinder nutrient absorption. There are many bariatric surgical procedures performed to achieve an ideal weight in patients, however, sleeve gastrectomy (SG) and Roux-en-Y gastric bypass (RYGB) are the two most prevalent bariatric procedures.³ SG is performed by resection along the vertical axis of the stomach, creating a tubular stomach and limiting gastric volume to 60–100 ml with the preservation of the pylorus. Therefore, the continuity of the intestine remains intact. The target weight loss (%TWL) of SG is 25–30%. In contrast, the RYGB procedure divides the stomach into a proximal gastric pouch, which is 15–30 mL in volume, and a lower gastric remnant. The jejunum is divided 15 to 50 cm distally of the ligament of Trietz and the distal alimentary Roux limb is connected to the stomach pouch.⁴ The RYGB procedure reroutes the intestine so that nutrients bypass most of the stomach and the duodenum and flow directly into the jejunum.⁴

With a combination of restrictive and malabsorptive procedures in RYGB, patients tend to lose more weight than those who have undergone SG, at least within the first year follow up.⁵ Weight loss is the primary outcome used for assessing the success of the bariatric procedure. Percentage of excess weight loss (%EWL), the percentage of excess BMI loss (%EBMI loss), the percentage of total weight loss (%TWL), and the percentage of BMI loss (%BMIL) are also considered as indicators for the procedure’s effectiveness. However, the percentage of excess weight loss (%EWL) is used as one of the criteria for success in bariatric surgery.⁶

For the Asian and Pacific regions, the total number of bariatric surgery cases has increased rapidly from 2,770 to 46,110 over 10 years (2003–2013). Both RYGB and SG procedures were frequently used (25% and 49% respectively).⁷ In 2017, a total of 95,125 patients underwent bariatric surgery, including 68% SG, 19.5% RYGB, and 12.5% others. Similarly, in Thailand, patients who underwent bariatric surgery had SG, RYGB and other procedures: 56.4%, 40.5% and 3.1%, respectively.⁸

Studies comparing weight loss and changes in body composition after SG and RYGB have been reported recently. Most of the studies have been conducted on Caucasians.⁹ However, changes in body composition and BMI are related to ethnic-specific factors. It has been found that Asians have a more significant response to metabolic risks than Caucasians with the same BMI, which may be due to the higher body fat in Asians and over-responsiveness to obesity.^9,10 Thus, the efficacy results of body composition changes after bariatric surgery in Caucasians may not be applicable to Asian populations. In addition, studies comparing a RYGB and SG for Asian patients are limited. Therefore, the primary objective of this study was to investigate the impact of two different bariatric surgeries, RYGB and SG, on weight and body composition changes. The secondary objective was to observe the factors related to weight loss mechanisms in patients, including resting energy expenditure, diet intake, changes in hunger and desire for specific food types and physical activity (level and MET-min/week).

Methods

Results

A total of 32 participants (SG (n = 11, two female subjects, nine male subjects) and RYGB (n = 21, 14 female subjects, seven male subjects)) completed the six-month follow-up study (Figure 1).³⁴

Figure 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram.

Sleeve gastrectomy (SG), Roux-en-Y gastric bypass (RYGB).

Baseline characteristics of the SG and RYGB groups are shown in Table 1. There was no significant difference in sex, age, and waist to hip ratio (WHR) between groups. However, mean basal weight before surgery was significantly different according to the surgical procedure (SG (107.39 (15.59)) vs. RYGB (128.87 (19.22)), p = 0.0033). As a result, BMI (SG (40.42 (4.76)) vs. RYGB (46.40 (6.46)), p = 0.0027), fat mass (SG (52.56 (8.20)) vs. RYGB (65.28 (12.92)), p = 0.0033) and fat free mass (SG (54.95 (11.71)) vs. RYGB (64.24 (10.91)), p = 0.0069) were different between surgical procedures at baseline (Table 1). At the baseline, there were no significant differences in comorbidity including type II diabetes mellitus (DM), hypertension (HTN), dyslipidemia (DLP), non-alcoholic fatty live disease (NAFLD), and gastroesophageal reflux disease (GERD) however, there was a significant difference in obstructive Sleep apnea (OSA) between RYGB (95.24%) and SG (63.64%).

In this six-month follow up study, significant weight loss was obtained after three and six months for both procedures. The body weight of subjects in both bariatric procedures were significantly reduced from the baseline, three months and six months: SG group (107.39 (15.39) kg, 90.42 (15.25) kg, and 81.70 (13.52) kg, (p < 0.001)); and RYGB group (128.87 (19.22) kg, 108.30 (17.82) kg and 99.77 (18.09) kg (p < 0.001)), respectively. When comparing the body weight of patients between groups, a significant difference was also found at three months (90.42 (15.25) kg vs. 108.30 (17.82) kg, p = 0.0083) and six months after surgery (81.70 (13.52) kg vs. 99.77 (18.09) kg, p = 0.0068). Similarly, BMI, WHR, and body composition, including fat mass, and fat-free mass were significantly decreased at three and six months following each SG and RYGB procedure, when compared to their baseline (all p ≤ 0.001) (Table 2).

The %EWL and %TWL were reduced significantly after three and six months for both groups (p < 0.001, 0.003 in the SG group, and all p < 0.001 in the RYGB group, respectively). The %EWL and %TWL at three months between the two procedures were not different. However, the %EWL at six months was higher in the SG group than the RYGB group (58.35 (15.30) % vs. 46.54 (12.52) %, p = 0.0257) (Figure 2).

Figure 2. The comparison of body weight (a), body mass index (BMI) (b), fat mass (c), fat-free mass (d) between Roux-en-Y gastric bypass (RYGB) and sleeve gastrectomy (SG) at baseline, three months, and six months post-operation.

Energy and macronutrient intake including carbohydrate, protein, and fat were decreased significantly after three and six months of both procedures (p < 0.05). There was no significant difference in energy and macronutrient intake found when comparing the two groups at each follow-up month.

Fasting blood glucose significantly decreased from 113.55 (30.18) mg/dL at baseline to 93.60 (14.81) mg/dL at three months following SG treatment (p = 0.006). In addition, HbA1C was significantly lower at six months compared to the baseline in both groups (SG from 6.41 (0.96)% to 5.28 (0.57)%, p = 0.022; RYGB from 5.91(0.94)% to 4.96(0.55)%, p ≤ 0.001). Bariatric surgery also affected the lipid profile in both groups. Total cholesterol was significantly lower at three months (161.12 (38.41) mg/dL) compared to the baseline (186.32 (37.85) mg/dL) in RYGB group (p = 0.032). When compared between groups, total cholesterol at six months was significantly lower in the RYGB group (170.71 (30.38) mg/dL) compared to the SG group (200.09 (29.52) mg/dL), p = 0.020. HDL-cholesterol decreased at three months following both surgeries. HDL-cholesterol in the SG group decreased from 41.2 (9.99) mg/dL at baseline to 38.67 (4.85) mg/dL, while in the RYGB group it decreased from 42.67 (9.97) mg/dL at baseline to 36.58 (6.06) mg/dL at three months. However, HDL-cholesterol significantly restored after six months following both surgeries as HDL-cholesterol in SG and RYGB group at six months were 45.36 (6.22) mg/dL and 43.03 (6.66) mg/dL, respectively. There were no significant differences on HDL-cholesterol between groups. Triglyceride level was significantly lower at six months following both surgeries compared to baseline (SG group 156.5 (88.98) mg/dL at baseline to 72.09 (23.91) mg/dL at six months, p = 0.008, RYGB group 143.79 (56.6) mg/dL at baseline to 79.06 (28.63) mg/dL at six months, p ≤ 0.001).

The kidney function was not significantly affected by both surgeries as there were no differences in creatinine and eGFR values at any time points. Meanwhile, the SG group significantly decreased SGOT at month three and month six compared to the RYGB group (at month three, SGOT in SG vs. RYGB were 18.56 (5.53) units/L vs. 25.48 (9.98) units/L, p = 0.036; while at six months, SGOT in SG vs. RYGB were 17.00 (2.83) units/L vs. 21.33 (7.81) units/L, p = 0.039). Albumin level in RYGB was significantly lower in the RYGB group at baseline and six months compared to SG (at baseline were 4.02 (0.28) g/dL vs. 4.36 (0.31) g/dL for RYGB and SG, respectively, p = 0.005; while at six months were 3.89 (0.25) g/dL vs. 4.25 (0.25) g/dL for RYGB and SG, respectively, p = 0.006).

In the SG group, blood biochemical parameters including hemoglobin, hematocrit, mean corpuscular volume (MCV), mean corpuscular hemoglobin (MCH), and mean corpuscular hemoglobin concentration (MCHC) were not changed significantly when compared between follow-up months (p > 0.05). Patients in the RYGB group showed a difference in MCV at three and six months (83.62 (7.03) fl at baseline vs. 83.46 (6.31) fl and 84.67 (6.46) fl, p = 0.012) and MCH was also higher at three and six months (26.92 (2.73) pg at baseline vs. 27.51 (2.41) pg and 27.42 (2.44) pg, p = 0.032). In addition, MCHC at three months (32.91 (0.79) g/dl) was higher than the baseline and six months after surgery (p = 0.003). When comparing blood biochemical parameters, including hemoglobin, hematocrit, MCV, MCH, and MCHC of patients between groups, no significant difference was presented at the three and six month follow-ups (p > 0.05).

The level of vitamin D was increased after six months when compared to the baseline both for the SG (22.93 (9.10) pg/ml vs. 34.47 (11.09) pg/ml, p = 0.008) and the RYGB group (18.84 (7.32) pg/ml vs. 28.81 (10.44) pg/ml, p = 0.002). Also, there was a reduction of folate levels six months post-surgery when compared to the baseline in the SG (11.49 (5.31) ng/ml vs. 5.77 (3.44) ng/ml, p = 0.024) and RYGB group (7.53 (4.81) ng/ml vs. 5.27 (3.15) ng/ml, p = 0.018). Additionally, only vitamin B12 levels were significantly different at six months when comparing procedures (533.55 (136.10) pg/ml vs. 351.89 (121.17) pg/ml, p ≤ 0.001).

The VAS scores of hungers and desires for specific types of patients were shown in Table 3. Patients in the SG group reported a significant increase in hunger (1.17 (0.98) vs. 3.00 (1.41), p = 0.028) and desire for sour food (0.67 (1.21) vs. 2.17 (0.98), p = 0.007) after the six-month follow-up. Meanwhile, patients in the RYGB group reported a higher score of desire for sour and fatty food after the six-month follow-up (1.71 (2.20) vs. 3.21 (2.22), p =0.007 and 0.43 (1.09) vs. 2.07 (2.27), p =0.010, respectively). No significant difference in hunger and desire for specific types between the two procedures was observed (p > 0.05) (Table 3).

Discussion

In the current study, we found that both the SG and RYGB procedures decreased the body weight, BMI, WHR, fat mass, and fat-free mass. Consequently, EWL was reduced at six months by 58.35% and 46.54% for SG and RYGB, respectively. A previous study reported that EWL at six months following SG and RYGB were 64% and 61% (p > 0.05).¹² By contrast, Rondelli et al., reported that EWL in SG was lower compared to RYGB at the one-year follow-up (49% vs 61.5% for SG vs. RYGB, respectively, p = 0.001), however there was no significant difference of EWL between procedures at the two- and three-year follow-ups. A previous study reported several factors affecting the EWL in SG and RYGB, including the characteristics of patients, complications following the surgery, and the technical aspects regarding the surgery (the SG diameter, the volume of RYGB gastric pouch, and the Roux limb length, etc.)⁵

The weight reduction following SG and RYGB procedures is mediated by several mechanisms, including caloric restriction, hormonal, and neural factors in the gastrointestinal tract thus the food intake and appetite were reduced.¹³ Food intake could be the major player of weight loss in the short-term following SG and RYGB procedures. In the current study, the food intake of patients at six months following SG and RYGB procedures significantly decreased from the baseline by 66.4% (-1,216 kcal/day) and 52.7% (-855.4 kcal/day), respectively (p < 0.05). Furthermore, reduction of carbohydrate and fat intake were significantly decreased at the three and six months follow up for both groups when compared to each baseline (p < 0.05). For those patients who underwent SG, the carbohydrate and fat intake decreased by 209.01 g/day and 38.5 g/day at the six-month follow up, while in RYGB, the carbohydrate and fat intake decreased by 132.91 g/day and 30.1 g/day at the six-month follow up when compared to the baseline. A meta-analysis study revealed that RYGB decreased food intake by 1,215.16 kcal/day, while SG decreased 939.38 kcal/day. Janmohammadi et al., also showed that fat intake was decreased following RYGB and SG procedures by -1.91 g/day (95%CI −3.37 to −0.44, p = 0.78) and -2.83 g/day (95%CI −7.07 to 1.42, p = 0.19), respectively. Post-operative adverse effects on gastrointestinal symptoms, such as vomiting, nausea, aversion of food, and dumping syndrome may occur in patients who underwent bariatric surgery which decreased the food intake.¹⁴

In addition, resection of the stomach in SG limits the volume of food intake and increases the release of gut peptides hormones, including glucagon-like-peptide (GLP)-1 and peptide YY (PYY), which promotes satiety and thus reduces the food intake.¹³ In addition, several studies reported that other hormones, such as ghrelin, were changed following bariatric surgery. Ghrelin is a gut hormone that is positively associated with hunger. A previous study reported that the area under the curve of ghrelin was significantly lower by 77% for patients who underwent RYGB when compared to normal-weight controls.¹⁵ Similarly, Frühbeck et al., reported that ghrelin concentration decreased 24 h following RYGB surgery.¹⁶ Decreased ghrelin was accompanied by decreased hunger following SG and RYGB procedures.^17–19

In the current study, we observed the favorable effects of both SG and RYGB on glucose profile. Fasting blood glucose was significantly decreased following SG treatment compared to the baseline, and both treatments significantly lowered the HbA1C compared to the baseline. Similarly, it was reported that SG and RYGB improved the glucose tolerance and decreased the fasting blood glucose report in in vitro and clinical studies.^20–22 The improvement in blood glucose profile is possibly mediated through the increasing of GLP-1 following bariatric surgery treatment.²¹ GLP-1 is known to improve glucose tolerance by inhibiting β-cell apoptosis and stimulating insulin secretion.²¹

Both treatments significantly decreased triglyceride at six months compared to the baseline. Furthermore, RYGB also decreased the total cholesterol in the current study. However, there was no effect of SG on total cholesterol. A similar finding was observed in Liaskos’ study that showed RYGB is more effective in decreasing lipemia in non-diabetic obese patients.²³ This malabsorptive nature of RYGB limits the lipid absorption and allows the greater reduction of total cholesterol level compared to SG procedure.²³

In the current study we observed that liver function as indicated by liver enzymes, including SGOT and SGPT, were significantly lower following SG compared to RYGB. Previous studies also had similar findings that SG treatment decreased SGOT compared to other type of bariatric surgeries at six months²⁴ and two years following surgery.²⁵ The underlying mechanism of the increasing liver enzymes following RYGB remains unclear. However, it was thought that different mechanisms of bariatric surgery possess different effects on endocrine and physiological function in the gut.²⁵ Further studies are needed to confirm the mechanism of liver enzyme reduction in different type of bariatric surgery.

Obesity is associated with vitamin D deficiency because of sequestration of vitamin D and other fat-soluble vitamins in adipose tissue.²⁶ Furthermore, inadequate food intake and reduced absorption area of nutrients following SG and RYGB increases risk of micronutrient deficiency, including vitamin D deficiency.²⁷ In the current study, the vitamin D levels for both groups at the six-month follow-up increased by 50.3% and 52.9% when compared to the baseline. In contrast, folate in both groups decreased by 49.8% and 30%. The increase of vitamin D in this current study could be mediated by the vitamin D supplementation following SG and RYGB procedures. Vitamin D supplementation is recommended for patients who underwent bariatric surgery. The dose may vary depending on the initial serum vitamin D in patients. It is recommended to take 3,000 IU/day of vitamin D, titrate to >30 mg/ml, to prevent the deficiency. In the presence of vitamin D deficiency, it is recommended to consume 50,000 IU of vitamin D once per week for eight weeks, followed by 1,500–2,000 IU/day for maintenance therapy.^28,29

Reduction of serum folate in the current study may be due to inadequate intake of folic acid and vitamin B12. Lower serum vitamin B12 may lower serum folate since vitamin B12 is a coenzyme for folate activation. Additionally, both surgeries may reduce the area of folate absorption.²⁷ The present study also found a reduction of folate among patients for both bariatric procedures. Folate is a water-soluble B-vitamin that functions as a coenzyme for purines and pyrimidines synthesis and amino acid conversion. Deficiency of this vitamin results in cell division and protein synthesis impairment.³⁰ Folates are not synthesized by mammals but can be produced by gut microbes.³¹ It is also present in various types of food including liver, yeast, leafy vegetables, legumes, and fruits. Previous studies indicated that there are factors responsible for folate deficiency in bariatric patients which include eating behavior modification, inadequate dietary intake, malabsorption, gut microbiota diversity reduction, and poor compliance to vitamin supplementation.^32,33 Providing sufficient vitamin supplementation may improve folate levels and prevent related complication from folate deficiency in post-operative patients.

This study enhances knowledge to the limited literature comparing RYGB and LSG regarding body composition changes in the Thai population. In this study, all participants meeting the inclusion criteria were included therefore, the results from this study can be extrapolated into the real-life Thai clinical setting. However, to respect patient right and autonomy, randomization was not feasible. With non-randomized design, selection bias may have occurred. In addition, this study did not examine the exercise patterns, which may confound the results.

Conclusions

Weight loss was comparable between SG and RYGB groups at six months after surgery. Body compositions include body fat percentage and fat mass rapidly decreased during six months similarly for both procedures, and patients in the RYGB group lost more fat-free mass than the SG group at six months post-operation. In addition, no significant difference of hunger and desire for specific types of food between procedures was observed.