Introduction

The worldwide obesity prevalence has expanded to pandemic proportions and is associated with considerable increases in cardiometabolic risks and morbidity and mortality levels [1,2,3]. Obesity is the result of a complex interaction between genetic and environmental factors, but many factors and mechanisms associated with its development are still not fully understood [4]. During the last few years, researchers have demonstrated that the intestinal microbiota is a relevant facto; e.g., individuals with obesity have differences in the diversity and richness of their intestinal microflora compared to individuals without overweight. These differences have been shown to contribute to higher energy extraction from the diet and metabolic pathways dysregulation [5].

Bariatric surgery is the most effective and durable treatment for morbid obesity, and Roux-en-Y gastric bypass (RYGB) is the most frequently performed procedure in Brazil [6,7,8]. The anatomical and physiological changes after RYGB induce a pH and oxygen increase in the intestinal lumen, which can dysregulate the microbiota by impairing the proliferation of microbial species important for maintaining the intestinal barrier and metabolic actions, such Lactobacillus and Bifidobacterium [9, 10].

Oral probiotics administration can be an effective strategy to achieve intestinal eubiosis and improve surgical results. The literature shows different hypotheses about the mechanisms related to supplementation with probiotics and weight loss, the metabolic profile, and the vitamin level improvements, such as increased expression of genes associated with fatty acid metabolism, insulin sensitivity, the expression of adiponectin and AMPK activation, improved micronutrient absorption, and a strengthened gut barrier [11,12,13,14,15].

Even though the scientific evidence has shown that bariatric surgery causes intestinal microbiota changes and alters metabolic and nutritional parameters [16], few studies have evaluated the effects of probiotic supplementation after RYGB or similar surgical techniques, such as one anastomosis gastric bypass-mini gastric bypass (OAGB-MGB). In the few available studies, Lactobacillus and Bifidobacterium strain supplementation has been associated with improvements in gastrointestinal symptoms, inflammatory profile, and vitamin levels. Although supplementation might have beneficial effects on metabolic parameters, increase weight loss, and improve body composition, the evidence is scarce and controversial [16,17,18,19,20].

This study was designed to identify the effects of the supplementation of Lactobacillus acidophilus NCFM and Bifidobacterium lactis Bi-07 on nutritional and metabolic parameters in the early postoperative period after RYGB.

Materials and Methods

This is a randomized, double-blind, placebo-controlled clinical trial conducted with patients undergoing RYGB in a public hospital in Curitiba, Brazil, from April 2018 to January 2019. This study was approved by the Research Ethics Committee of the Pontifical Catholic University of Paraná (PUCPR) (n° 4.252.808) and was registered in the Brazilian Clinical Trials Registry - REBEC (n°RBR-4x3gqp). The inclusion criteria were adult candidates for RYGB, with a body mass index (BMI) of ≥ 35 kg/m2, who signed the informed consent form and did not use antibiotics 4 weeks prior to the beginning of the study. Patients who were submitted to other surgical techniques or reoperation, as well as those participants who had immediate postsurgical complications, ingested antibiotics during the study period, or had an adherence below 90% in the use of the tablets (inadequate use of the probiotics for more than nine consecutive days), were excluded.

Surgical Methods

All surgical procedures were performed by a single team of surgeons based on the standardization of the surgical procedure, which involves midline laparotomy, 100 cm of alimentary limb (Roux limb), jejuno-jejunal anastomosis 200 cm distal from the Treitz ligament, creation of a gastric pouch with a capacity of approximately 30 mL, antecolic gastrojejunal anastomosis, and a negative intraoperative leak test using methylene blue.

Randomization and Treatment

The investigators were responsible for the recruitment and randomization. Patient randomization was performed by a systematic 1:1 allocation process, with randomization codes being drawn to distribute the individuals into the placebo or probiotic group. The supplements were provided in similar packaging by a pharmacist who did not otherwise participate in the study. The participants and investigators were blinded to the product identification code. The probiotic used was FloraVantage® (5 billion Lactobacillus acidophilus NCFM® Strain, 5 billion Bifidobacterium lactis Bi-07®) from Bariatric Advantage (Aliso Viejo, CA, USA) and the placebo was an inert manipulated tablet consisting of starch and 190 mg of lactose. Both were chewable tablets, similar in physical appearance, taste, and color. Patients were instructed to keep the products in places without humidity or sun exposure and to take two of the chewable tablets per day, for 90 days, starting on the seventh postoperative day.

Follow-up Assessments

The first assessment was performed at the first visit, approximately 10 days before surgery. Follow-up assessments were conducted at approximately 12 weeks postoperatively. According to the hospital’s standard protocol, both groups received the same diet orientation and the same multivitamin and protein supplementation prescription. During the intervention period, adherence to the protocol was monitored weekly by phone calls and those who did not follow the protocol or presented with immediate postsurgical complications were excluded. Clinical and anthropometric assessments were performed at both visit times.

Clinical and Anthropometric Assessment

Anthropometric measurements included body weight (kg), height (m), BMI (kg/m2), waist circumference (cm), percentage of excess weight loss (% EWL), body fat (kg and %), and lean mass (kg). BMI was calculated as body weight (kg)/height squared in meter (m2), %EWL was calculated using the method described by Deitel et al. [21], and waist circumference was measured around the largest abdominal perimeter between the last rib and the iliac crest [22]. In addition, body composition assessment was performed using tetrapolar impedance analysis, according to the manufacturer’s instructions (BIA - 450 Bioimpedance Analyzer, Biodynamics, Seattle, EUA).

Fasting glucose, total plasma cholesterol, low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride levels were measured by a colorimetric enzymatic method (Vitros Fusion 5.1 Chemistry System Analyzer, Ortho Clinical Diagnostics, England, UK). The quantification of glycated hemoglobin (HbA1c) was performed according to the specifications of the HbA1c Reagent Kit (Vitros Chemistry Products; Vitros 5.1, Ortho Clinical Diagnostics, England, UK). The serum concentrations of vitamin B12, folate, 25-hydroxy vitamin D, and fasting insulin were determined using an amplified chemiluminescence method (Vitros 3600, Ortho Clinical Diagnostics, England, UK). Insulin sensitivity (QUICKI) and insulin resistance (HOMA-IR) were estimated using the equations proposed by Katz et al. and Matthews et al., respectively [23, 24].

Clinical and sociodemographic data were collected from a nutritional anamnesis form and medical records. Physical activity practice and alcoholic consumption data were collected according to the World Health Organization’s (WHO) criteria [25, 26].

Primary and Secondary Outcomes

The primary outcomes were nutritional status changes, including both the anthropometric data and the serum vitamin levels. Secondary outcomes were metabolic parameter improvements, such as glycemic and lipid markers.

Statistical Analysis

Independent t test and chi-square tests of homogeneity were used to verify differences between groups, considering the numerical and categorical variables, respectively. The probiotic effect on each response variable was analyzed and adjusted by age, sex, body mass index, time since diagnosis, and physical activity. Marginal regression models were fitted, and the interaction effect between the group and patient condition (pre or postoperative) was evaluated. A quasi-likelihood approach was used for fitting the regression models, aiming to lead with non-constant variance, non-normality, and correlated (paired) measures taken from each patient. Robust (sandwich) standard errors were calculated to prevent possible model misspecifications. The link function was chosen among two options: identity or logarithmic. In the first case, the effects are additive, whereas under the logarithmic link function, the effects are multiplicative. The best option for each response variable, among these two link functions, was selected based on the quasi AIC [27,28,29,30]. The results are presented as differences or the ratio of means. All analyses were performed using the IBM SPSS Statistics version 22.0 software (SPSS, Chicago, IL) and R software version 3.6.1 (GLP, Auckland, New Zealand), considering P < 0.05 for statistical significance.

Results

Characteristics of the Participants

From 110 patients initially selected, 9 were not eligible according to the inclusion criteria or refused to participate. Thus, 101 individuals were randomized into the placebo (n = 50) or probiotic (n = 51) group, and 71 (70.3%) completed the protocol (Fig. 1).

Fig. 1
figure 1

The study consort flowchart

Full size image

Participants in the placebo and probiotic groups had 99% ± 2.3 and 99% ± 2.4 of adherence to supplementation, respectively (p = 0.652). None of the participants reported adverse effects during the intervention.

The baseline demographic, clinical, and anthropometric variables analyzed were similar in both groups, except for age and hypertension, which were significantly higher in the placebo group, and total cholesterol levels, which were higher in the probiotic group (Tables 1 and 2; Figs. 2, 3, and 4).

Table 1 Clinical and demographic baseline characteristic
Full size table
Table 2 Anthropometric measurements at the baseline and after 12 weeks of supplementation
Full size table
Fig. 2
figure 2

Vitamin levels of patients in the probiotic and placebo group over the supplementation period. Values are represented by mean (SE)

Full size image
Fig. 3
figure 3

Lipid profile of patients in the probiotic and placebo group over supplementation period. LDL low-density lipoprotein cholesterol, HDL high-density lipoprotein cholesterol. Values are represented by mean (SE)

Full size image
Fig. 4
figure 4

Glycemic profile of patients in the probiotic and placebo group over the supplementation period. FBS fasting blood sugar, HbAlc glycated hemoglobin, HOMA-IR homeostasis model assessment of insulin resistance, QUICKI quantitative insulin check index. Values are represented by mean (SE)

Full size image

Primary Outcomes

After 90 days of intervention, significant reductions were observed in the anthropometric parameters (weight, BMI, WC, body fat) in comparison to the baseline values (p < 0.05) in both groups (Table 2). Serum 25-OH vitamin D increased in both groups compared to baseline; however, this increase was significant only in the probiotic group (p = 0.004). Vitamin B12 levels tended to be significantly higher in the probiotic group than in the placebo group (p = 0.063) (Fig. 2). The remaining anthropometric and nutritional variables were similar in both groups.

Secondary Outcomes

As shown in Figs. 3 and 4, statistically significant improvements in FBS, HbA1c, insulin, HOMA-IR, QUICKI, total cholesterol, HDL, and LDL were observed in both groups regardless of the use of probiotics. However, triglyceride levels showed a significant reduction in the probiotic group only (p < 0.001).

Discussion

This is the first randomized, double-blind, and placebo-controlled study to test the effect of supplementation with Lactobacillus acidophilus NCFM combined with Bifidobacterium lactis Bi-07 on nutritional and metabolic parameters 3 months after RYGB.

The results of the present study showed significant improvements of anthropometric and metabolic parameters in both groups after RYGB, consistent with other clinical trials, confirming that bariatric surgery is a very effective therapy for weight loss management and metabolic improvement in morbidly obese individuals [31,32,33,34]. However, there was a significant improvement in the serum concentrations of 25-OH vitamin D and triglycerides and a trend for increased serum vitamin B12 levels in the probiotic group only.

After RYGB, patients lose an average of 50–80% of their excess body weight, and they generally experience long-term comorbidity remission [35]. Dietary restriction, poor nutrient absorption, reduced ghrelin levels, and microbiota modifications all play important roles in the significant weight loss after surgery [36, 37]. In addition, the literature shows different mechanistic hypotheses related to probiotic supplementation and weight loss, such as increased expression of genes associated with fatty acid metabolism, resensitization to insulin, increased production of adiponectin, and AMPK activation [38, 39]. Karbashian et al. [19] showed a significant reduction in postoperative body weight and %EWL in patients who received Lactobacillus and Bifidobacteria supplementation. However, it is important to emphasize that in that study, probiotic supplementation began 4 weeks before OAGB-MGB and was continued for 12 weeks after the surgery. However, in agreement with our study, Woodard et al. [17] showed no statistically significant difference in %EWL after daily Lactobacillus species supplementation for 6 months after RYGB.

Microbiota-produced vitamins can contribute to serum vitamin levels, particularly vitamin B12. However, after bariatric surgery, changes in the intestinal microbiota can modify the microbial functional capacity, especially during the first 3 months after surgery [40]. Intestinal bacterial overgrowth, preoperative vitamin deficiency, and nonadherence to the use of multivitamin supplements may contribute to worsening serum vitamin B concentrations in the postoperative period [41].

Lactobacillus and Bifidobacteria supplementation may be an appropriate strategy to improve the status of vitamin B12 through intestinal microbiota modulation [14, 42]. Although supplementation did not present significant differences between the groups in this study, there was a tendency toward increased serum B12 concentration in the group supplemented with probiotics. Another study conducted with patients after RYGB who were supplemented with Lactobacillus species for 6 months has reported significantly higher postoperative vitamin B12 levels in the probiotic supplemented group [17].

In this study, the probiotic group showed a significant increase in vitamin D levels, corroborating the findings of Karbashian et al. [19]. This increase in vitamin D levels is expected after rapid weight loss due to the release of vitamin D from adipose tissue [43]. Also, recent studies have demonstrated that vitamin D status is associated with the intestinal microbiota composition and that probiotic treatment could increase 7-dehydrocholesterol synthesis and vitamin D receptor (VDR) expression and activity [44, 45].

Our results showed that probiotic supplementation after RYGB is effective in reducing triglycerides levels. Additionally, in absolute numbers, the average reduction of total cholesterol was almost 18 mg/dl (8.3%) greater in the probiotic group than in the placebo group. Several clinical trials have reported a total cholesterol level reduction without changes in the HDL and triglycerides levels in patients with hypercholesterolemia supplemented with probiotics [46,47,48,49,50,51]. Those studies included both lean and overweight patients, which may have contributed to the different findings with regard to the triglyceride levels. Although the mechanism of action is not well-known, it seems possible that the decrease in triglyceride levels may be due to the upregulation of apolipoprotein A-V (ApoA-V), bile acid receptor (FXR), and PPAR alpha expression in the plasma [46]. Some studies have suggested that probiotic supplementation produces a cholesterol-lowering effect through multiple mechanisms, including increased salt hydrolase expression by lactic acid bacteria, intracellular cholesterol transfer to the cellular surface, cholesterol precipitation by deconjugated bile salt hydrolysate, ferulic acid (FA) synthesis, which can inhibit enzymes involved in endogenous cholesterol production, and higher levels of production of short-chain fatty acids (SCFAs) that can block hepatic cholesterol synthesis [13, 52,53,54,55].

Despite evidence that changes in the intestinal microbiota and the use of probiotics can improve glycemic parameters in animal models and in humans, especially those with type 2 diabetes (T2D) [56], in the present study, the glucose profile and the insulin index did not show any significant improvement after probiotic supplementation. Another study that investigated the effects of probiotics on the glycemic profile after OAGB-MGB also found no significant differences between the control and probiotic groups [19]. These controversial results can be explained by study heterogeneity and many confounding factors, such as the use of drugs, individuality of the intestinal microbiota, BMI, type of surgical procedure, and specific strain effects.

The strengths of our study are the design (randomized, double-blind, placebo-controlled), the use of probiotics and placebo developed for this specific supplementation period (both being chewable and palatable), and the high adherence achieved with the use of probiotics or placebo (over 99% supplementation adherence).

The main limitation of this study is the lack of an intestinal microbiota composition analysis. Randomization, use of the same surgical technique, a standardized diet and multivitamin supplementation, and weekly contact between researchers and participants to monitor their adherence to the research protocol were strategies used to minimize interindividual variability.

In conclusion, anthropometric and metabolic improvements were observed in both groups after RYGB. However, only the probiotic group showed significant improvements in serum 25-OH vitamin D and triglyceride levels. Additional studies, including a longer probiotic supplementation time, are needed to confirm these positive results.