Digestion of adults

The Effects of Probiotic Supplements on Blood Markers of Endotoxin and Lipid Peroxidation in Patients Undergoing Gastric Bypass Surgery; a Randomized, Double-Blind, Placebo-Controlled, Clinical Trial with 13 Months Follow-Up

Abstract

Background The effect of probiotic supplements among subjects undergoing bariatric surgery indicates conflicting results. Moreover, whether these effects remain after ceasing the treatment remained to be elucidated. This study was conducted to assess the effect of probiotic supplements on blood markers of endotoxin (lipopolysaccharides-binding protein: LBP), inflammation and lipid peroxidation (malondialdehyde: MDA) in patients with morbid obesity undergoing the one-anastomosis gastric bypass (OAGB).

Methods This study is a placebo-controlled, double-blind, and randomized clinical trial and 9 months of additional follow-up. Forty-six morbid obese patients undergoing OAGB were randomized to 4 months of probiotic or placebo supplements. Anthropometric indices and blood concentration of LBP, inflammatory markers, MDA, vitamin D3, and B12 were measured at 0, 4, and 13 months of study.

Results Probiotic supplements could improve serum LBP (P = 0.039), TNF-α (P = 0.005), vitamin B12 (P = 0.03), vitamin D3 (P = 0.001), and weight loss (P = 0.01) at month 4 in comparison to placebo; however, only serum MDA concentrations decreased significantly in the probiotic group compared with those in the placebo group (P = 0.013) at the end of follow-up period.

Discussion It was observed that 4 months probiotic supplementation compared with placebo prohibited an elevation in the LBP levels and improved serum TNF-α and 25-OH vitamin D3 concentrations and weight loss in patients undergoing the OAGB surgery. However, these effects did not persist 9 months after the cessation of the treatment. Further investigations are required to find how long supplementation and which dosage of it can benefit body status for the long-term.

1 Department of Clinical Nutrition and Dietetics, Faculty of Nutrition

Sciences and Food Technology, National Nutrition and Food Technology Research Institute, Shahid Beheshti University of Medical Sciences, West Arghavan St., Farahzadi Blvd., P.O. Box: 19395-4741, Tehran 1981619573, Iran

2

Minimally Invasive Surgery Research Center, Iran University of Medical Sciences, Tehran, Iran

3 Center of Excellence for Minimally Invasive Surgery Training, Iran University of Medical Sciences, Tehran, Iran

4 Center of Excellence of European Branch of International Federation for Surgery of Obesity, Tehran, Iran

5

Cellular and Molecular Endocrine Research Center, Research

Institute for Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran

6

Nutrition and Endocrine Research Center, Research Institute for

Endocrine Sciences, Shahid Beheshti University of Medical Sciences, Tehran, Iran

Introduction

Obesity is a worldwide epidemic that increases the risk of chronic diseases, such as cardiovascular disease, type 2 diabetes (T2D), cancers, and osteoarthritis [1]. These risks are associated with low-grade inflammatory state and increased oxidative stress, which are characteristics of obesity [2].

Although several potential sources have been proposed for this inflammation, the alteration in gut microbiome plays a pivotal role in initiation and maintenance of obesity-related inflammation [3]. Innate immunity can be triggered by gut microbial metabolites, especially lipopolysaccharides (LPS), which are glycolipids in the outer membrane of Gramnegative bacteria via interaction with immune receptors and proteins, including the LPS-binding protein (LBP), cluster of differentiation 14 (CD14), myeloid differentiation protein-2 (MD2), and Toll-like receptor 4 (TLR4), ultimately activating nuclear factor-kB (NF-κB) and pro-inflammatory cytokines [3]. LBP, which binds to LPS and presents it to CD14, is a necessary part of the LPS-induced inflammatory response [3]. Taking into account the technical difficulties for quantifying LPS in serum/plasma, their short half-life [4], and increased LBP levels in response to LPS [5], the measurement of LBP is considered to reflect the circulating LPS status. Several studies suggest that obesity is associated with increased LPS and LPB levels. Furthermore, a chronic low-level raise of LPS (endotoxemia) may be involved in obesity-related metabolic disorders [6].

Bariatric surgery (BS) is well approved as an effective treatment for morbid obesity to induce substantial long-term weight loss [7]. Laparoscopic one-anastomosis gastric bypass (OAGB), which is being promoted as a quick and efficient procedure in comparison with the standard Roux-en-Y gastric bypass (RYGB) [8], acts by a combination of restriction and malabsorption of food intake and change in gut hormones involved in appetite control [8]. Despite the advantages of BS, abdominal symptoms and nutritional deficiencies are the side effects of these procedures [9]. Moreover, notwithstanding the noticeable weight loss in the early months after BS, serum LBP levels and some other inflammatory factors remain elevated [10–12], revealing that inflammatory responses still continue. In addition, some evidence reported that oxidative stress markers may not reduce significantly during this time [13, 14]. Bacterial overgrowth, alteration of the intestinal microbial composition, and increased intestinal permeability resulting from the anatomical and physiological modifications of the gastrointestinal tract may contribute to ongoing inflammation [15–17].

Given the growing evidence supporting the involvement of gut microbiota in the inflammation process and the possibility of being affected by BS, it is possible that the use of modulators of gut microbiota, such as probiotics, should be beneficial in patients undergoing this surgery. Extensive evidence supports the potential modulating effects of probiotic supplements, especially containing Bifidobacterium

and Lactobacillus species, on endotoxin [18, 19], inflammatory and oxidative stress status [20, 21], and weight loss [22]. Moreover, the status of certain micronutrients, such as vitamin D3 and vitamin B12, may be affected by the consumption of probiotics [23, 24].

To date, the effect of probiotic supplements among subjects with morbid obesity undergoing BS has been assessed in few studies, indicating conflicting results [16, 25, 26]. Recently, we have shown that probiotic supplements improved inflammation, weight loss, and vitamin D3 status in patients undergoing OAGB [27]. However, the effects of probiotic supplements after ceasing the treatment remained to be elucidated. Patients’ follow-up after the intervention period will provide valuable findings for clinical practice. Thus, the present study was designed to examine the effects of probiotic supplements on serum LBP and malondialdehyde (MDA) levels as a lipid peroxidation marker in patients with morbid obesity undergoing the OAGB surgery and monitor the blood markers of endotoxin (LBP), inflammatory factors, such as tumor necrotizing factor (TNF_α), interlukin-6 (IL-6), and high sensitive C-reactive protein (hs-CRP), lipid peroxidation (MDA), anthropometric indices, and serum levels of 25-hydroxy vitamin D3 and vitamin B12 9 months after the completion of the probiotics supplementation.

Methods and Materials

This was a placebo-controlled, double-blind, randomized clinical trial, and 9 months of additional follow-up. Patients were recruited from Hazrat Rasul Hospital in Tehran, Iran, from May 2015 to March 2017. Subjects who were 18–60 years old, candidates for the laparoscopic OAGB surgery in the next month, morbid obese (BMI ≥ 40 kg/m2 or 40 > BMI > 35 kg/ m2 with comorbidities), and no evidence of chronic gastrointestinal, liver, and kidney disorders, were recruited. Participants who took antibiotics, probiotic supplements, foods fortified with probiotics and/or immunosuppressive treatment, and insulin within 4 weeks before the start of the study and during the study were excluded from the study. Furthermore, subjects were excluded if they were pregnant. The estimated sample size for each group was 23 patients with regarding a power (1 −β) of 90% and α = 0.05 to detect a difference of 10 kg/m2 in BMI measure with a standard deviation of 10 kg/m2, attained from Woodard et al.’s study [16]. Trial was registered at Clinicaltrial.gov (NCT02708589). The ethical committee of Shahid Beheshti University of Medical Sciences approved the study protocol (1394215/787), and written informed consent was obtained from all the participants who were included in the trial.

Surgical Method

Gastric-bypass surgery by OAGB included the creation of a long sleeved gastric tube along the lesser curvature side with a Billroth type II loop gastro-jejunostomy with a 180–200 cm or longer afferent limb [8].

Randomization and Treatment

A total of 46 patients were recruited into the study, and after stratifying (1:1) into two groups based on their type 2 diabetes (T2D) status (with or without T2D), they were randomly allocated to receive either the probiotic supplement (n = 23) or the placebo supplement (n = 23) for 4 months (from a month before surgery to 3 months after the surgery). Randomization sequence was computer-generated by a statistician in blocks of four patients and stratified based on their T2D status (with or without T2D). Patients were allocated to randomization code letters (A or B) in chronological order. Information about the treatment each participant was placed in a sealed envelope, which was not opened until the investigation was completed. Patients were instructed to take one supplement capsule each day and to refrigerate the unused capsules. The patients were requested not to consume the probiotic supplements on the day of surgery until hospital discharge (about 2

days). Each probiotic capsule (ZistTakhmir, Co., Tehran, Iran) contained seven species of probiotic bacteria (Lactobacillus casei (3.5 × 109 CFU/g), Lactobacillus rhamnosus (7.5 × 108 CFU/g), Streptococcus thermophiles (1 × 108 CFU/g), Bifidobacterium breve (1 × 1010 CFU/g), Lactobacillus acidophilus (1 × 109 CFU/g), Bifidobacteriumlongum (3.5 × 109 CFU/g), and Lactobacillus bulgaricus (1 × 108 CFU/g)) and 38.5-mg fructo-oligosaccharide. Placebo capsules contained the same amount of maltodextrin. The surgeon, medical staff related to the care of the patient, the research staff, and patients were all blinded to the treatment assignment.

Follow-up Assessments and Compliance

Follow-up assessments were performed at the months 0 (first visit), 1, 2, 3, 4, and 13 of the study. Adherence to protocol of the study and adverse effects was ascertained at each visit. Compliance with consumption of capsules in the intervention period was determined by supplement count at each visit and weekly telephone call. A loss of more than 10% of the supplements was regarded as incompliance, which resulted in exclusion from the study. Both groups were recommended to adhere dietary and physical activity advices based on clinical guidelines on the identification, evaluation, and treatment of overweight and obesity in adults [28] and clinical practice guidelines for the perioperative nutritional, metabolic, and nonsurgical support of the BS patients [29]. Furthermore, all patients adhered to the protocol of the medical center. Adherence to this protocol was also assessed during followup visits. If participants did not follow the protocol, they were excluded from the study. Assessment of clinical, paraclinical, and dietary intakes was carried out at the baseline and month 4 (3 months after the surgery) and month 13 of the study (12 months after the surgery).

Clinical, Paraclinical, and Dietary Intake Assessment

Anthropometric measurements including weight, height, and waist circumferences; paraclinical parameters including the serum levels of LBP and interleukin-6 (IL-6), tumor necrosis factor-alpha (TNF-α), high sensitivity C-reactive protein (hsCRP), MDA, 25-hydroxy vitamin D3, and vitamin B12; and dietary intakes were measured at the baseline, month 4

(3 months after the surgery), and month13 of the study (12 months after the surgery). Percentage of the excess weight loss (% EWL) was determined by the following formula: (preoperative weight − current weight) / (preoperative weight − ideal weight) × 100, while the ideal weight was considered as the weight of participant with assumed BMI of 25 kg/m2 [30]. All the biochemical parameters were assessed in the same laboratory by standard commercial methodologies. Fasting LBP and inflammatory markers including IL-6, tumor necrosis factor-alpha TNF-α, and hs-CRP concentrations were determined using the enzyme-linked immunosorbent assay (ELISA) (ZellBio, Ulm, Germany, for LBP and hs-CRP; Diaclone, Besancon, France, for TNF-α and IL-6), and the serum levels of MDAwere assayed using a commercial chemical colorimetrical assay kit (MDA assay kit; ZellBio GmbH, Ulm, Germany). Serum concentrations of 25-hydroxy vitamin D3 (Diagnostics Biochem Canada Inc., Ontario, Canada) and vitamin B12 (Zell Bio, Ulm, Germany) were measured by ELISA method according to the manufacturers’ protocols. In addition, the dietary intakes of participants were assessed by a 3-day food recall (two weekdays and one weekend) in months 0, 4, and 13; after that, all the record data were verified by a

dietitian. Dietary intakes were analyzed using Nutritionist V (First Databank, Hearst Corp, San Bruno, CA, USA).

The primary outcome of the study was a significant reduction in serum LBP levels and inflammatory factors measured on months 4 and 13 from the start of treatment. Secondary outcome measures were circulating MDA levels, anthropometric variables, and concentrations of 25-hydroxyvitamin D3 (25OH vitamin D3) and vitamin B12 in serum at the months 4 and 13.

Statistical Analyses

To detect normality of data distribution, the Kolmogorov– Smirnov test was used. To compare variables within each group, the analysis of variance for repeated measures and the Bonferroni post hoc test were used. Student’s t test was performed to find differences between groups. To remove the effects of confounding factors, either in the beginning or during the study, the analysis of covariance test (ANCOVA) was used. The data were analyzed according to the intention-totreat principle. All the statistical analyses were performed using SPSS for Windows (version 19; SPSS Inc., Chicago, IL). P value < 0.05 was considered statistically significant.

Results

From 46 eligible patients, 45 (97.82%) completed the 4-month probiotic intervention and 38 (82.61%) accomplished the 13month follow-up period. Among the patients in the placebo group, four patients were excluded (postponed date of surgery, n = 1; pregnancy, n = 1; withdraw from study, n = 2). Four individuals in the probiotic group were excluded(pregnancy, n = 1; moving to another city, n = 1; withdraw from study, n = 2) (Fig. 1). Dropout rate was similar in both groups, and all enrolled participants were included in the analysis of outcomes. None of the patients reported any serious adverse effects during consumption of probiotic supplements. As shown in Table 1, all baseline characteristicsof theparticipants in both groups were similar except the serum 25-OH vitamin D3 levels, which was significantly greater in the probiotic group compared to those in the placebo group (83.49 ± 25.16 vs. 52.20± 24.30, P < 0.001), and vitamin D3 supplement use at the baseline, which was significantly higher in the placebo group compared to that in the probiotic group (P = 0.026).

Primary Outcome

At the beginning of the study, no significant differences were observed between two groups in LBP and inflammatory cytokines except for hs-CRP levels, which tended to be higher in the placebo group compared to those in the probiotic group (P = 0.050). Within group comparisons of LBP indicated that the probiotic supplements in the intervention group prohibited an augmentation in LBP, whereas in the placebo group a significant rise in LBP was observed at month 4. So, a significant difference between-groups was noted at month 4 (0.06; 95% CI − 3.54 to 3.67 in the probiotic group vs.5.69; 95% CI 1.88 to 9.50 in the placebo group, P = 0.039); although after adjusting for confounder variables, it was disappeared. The

serum LBP levels were similarly unchanged compared to the baseline at month 13 in both groups. Moreover, a significant improvement in serum TNF-α levels was seen in the probiotic group in comparison with the baseline values (− 7.36; 95% CI − 11.86 to2.86, P = 0.01) and the placebo group (P = 0.001)at the end of intervention period. Finally, all of the inflammatory markers significantly decreased atthe end offollow-up period, as compared to the baseline in all patients, without a significant difference between groups .

Atthe study baseline, nosignificant differenceswere observed between the two groups regarding to MDA levels. Withingroup and between-group differences in levels of MDA did not significantly differ in both groups at the end of treatment period. The circulating MDA concentrations decreased significantly in the probiotic group compared with those in the baseline (P = 0.031) and the placebo group (P = 0.013) at the end of follow-up period (Table 2).

As indicated in Fig. 2, within-group differences revealed a significant reduction in the anthropometric measures (weight, BMI, and waist circumference) in both groups at month 4 and month 13 of study in comparison to the baseline. The mean changes in weight, % EWL, and BMI were significantly higher in the probiotic group compared to those in the placebo group (P = 0.026, P = 0.014, and P = 0.027, respectively) just at month 4 of study, and no significant differences in these

Table 2 Serum LBP, inflammatory factors and MDA concentrations of patients in the probiotic group and the placebo group over the study period

Probiotic group Placebo group Pb Pc Pd

LBP (μg/mL)

Baseline (mean± SD) 39.86 ± 10.39 36.04± 11.69 0.253

4 months (mean ± SD) 39.53 ± 10.04 42.01± 9.75 a

13 months (mean ± SD) 39.60 ± 12.07 38.42± 5.29

Change 4 months from baseline (95% CI) 0.06 (− 3.54, 3.67) 5.69 (1.88, 9.50) 0.039 0.493

Change13 months from baseline (95% CI) − 0.68 (− 6.10, 4.74) 2.03 (− 3.09, 7.15) 0.473 0.500

TNF-a (pg/mL) Baseline (mean± SD) 28.18 ± 12.94 24.89± 13.41 0.418

4 months (mean ± SD) 21.33 ± 7.42a 29.09± 20.17

13 months (mean ± SD) 15.05 ± 4.72a 17.93± 6.18a

Change 4 months from baseline (95% CI) − 7.36 (− 11.86, 2.86) 2.30 (− 2.30, 6.91) 0.005 0.001

Change13 months from baseline (95% CI) − 12.91 (− 18.84, − 6.97) − 5.15 (− 11.24, 0.94) 0.079 0.138

lL-6 (pg/mL)

Baseline (mean± SD) 10.89 ± 5.89 11.09 ± 3.51 0.562

4 months (mean ± SD) 8.15± 2.05a 9.90 ± 5.24

13 months (mean ± SD) 7.38± 0.55a 9.34 ± 1.71a

Change 4 months from baseline (95% CI) − 2.78(− 4.86, − 0.70) − 1.14(− 3.32, 1.04) 0.285 0.905

Change13 months from baseline (95% CI) − 2.53(− 3.93, − 1.12) − 1.87(− 3.24, − 0.51) 0.509 0.457

hs-CRP (ng/mL)

Baseline (mean± SD) 8098.411 ± 1364.49 8822.08± 975.388 0.05

4 months (mean ± SD) 5715.96 ± 2715.49 a 6595.17± 2634.45a

13 months (mean ± SD) 1677.75 ± 1880.08 a 2483.08± 3074.33a

Change 4 months from baseline (95% CI) − 2470.66 (− 3547.13, − 1394.19) − 2134.70 (− 3236.14, − 1033.24) 0.667 0.811

Change13 months from baseline (95% CI) − 6418.91 (− 7539.01, − 5298.80) − 6245.11 (− 7365.22, − 5125.01) 0.829 0.863

MDA (μmol/L) Baseline (mean± SD) 5.48± 0.94 5.07 ± 1.47 0.670

4 months (mean ± SD) 5.35± 0.73 5.21 ± 1.50

13 months (mean ± SD) 4.22± 1.53a 4.94 ± 1.67

Change 4 months from baseline (95% CI) − 0.19 (− 0.79, 0.41) 0.10 (− 0.51, 0.71) 0.510 0.269

Change13 months from baseline (95% CI) − 1.46 (− 2.37, − 0.54) 0.06 (− 0.85, 0.98) 0.025 0.013

LBP, lipopolysaccharide binding protein; TNF-a, tumor necrosis factor-alpha; Il-6, interleukin 6; hs-CRP, high-sensitivity C-reactive protein; MDA, malondialdehyde

a

P < 0.05, different from the baseline, based on linear mixed model for repeated measures tests and the Bonferroni post hoc test

b Between-group at the baseline, based on independent t test

c

Based on an ANCOVA and controlling for age

d

Based on an ANCOVA and controlling for age and the mean changes in weight and energy intake from the baseline

parameters were found between the two groups at month 13. According to 3-day dietary records, the dietary components were significantly different within-group at month 4 and month 13 compared to that in the initiation of study, while these changes did not differ significantly between groups at the 4thmonth of intervention and at the end of the study

(Fig. 3).

For the serum 25-OH vitamin D3 levels, an opposite trend was seen during the study period: at the end of 4th month, a significant increase was observed in both groups (P < 0.001), while a reduction was found at the end of follow-up period. Furthermore, the elevation at month 4 and the attenuation at month 13 were significantly greater in the probiotic group (P = 0.001, and P < 0.001, respectively). Within-group differences showed a marginal significant reduction in circulating vitamin B12 concentrations in the placebo group (P = 0.050) at month 4 and no significant difference in the probiotic groups throughout the study; between-group differences revealed a trend toward significant difference (P = 0.078) at month 4 of study

Discussion

Our findings indicate that 4-month consumption of probiotic supplements prohibited an elevation in the LBP levels and improved serum TNF-α and 25-OH vitamin D3 concentrations and weight loss compared to placebo consumption. However, most of the effects of probiotics that were observed at 4th month of supplementation were diminished 9 months after cessation of them. Only MDA, as a marker of oxidative stress status, remained significantly lower in the probiotic group compared to those in the placebo group.

Previous studies indicated that serum LBP or LPS levels do not improve in obese patients after BS; even though they may elevate during first months after RYGB, owing to the metabolic stress induced by surgical stress and significant weight loss [10, 11]. In the present study, an elevation in LBP 3 months after OAGB was suppressed by probiotic supplements. This was accompanied by declining in serum levels of TNF-a, suggesting the probability that the probiotic treatment diminished metabolic endotoxemia and related down-stream inflammatory mediators. However, when the impact on LBP was adjusted for weight loss, this effect disappeared. This might be explained by the effects of probiotics on weight loss improvement.

The evidence suggests that gut microbiota manipulation by probiotics is effective in attenuating increased LPS, LBP, and inflammatory factors induced by high fat diet or dysbiosis in several rodent models and patients with inflammatory conditions [18, 19, 31]. To our knowledge, only limited clinical trials evaluated the impact of probiotics in obese patients following BS. The results of previous studies on inflammatory markers are controversial [16, 25, 26]. This discrepancy may be due to variations in the duration of supplementation, dissimilar dosages and strains of probiotic bacteria, as well as different procedures of surgery. Additionally, one clear and potentially important difference between this study and some previous studies was that they administrated probiotics following surgery; however, in the current study, initiation of the treatment was pre-operation. Decrease in secretion of gastric acid and constraint of anaerobic organisms caused by anatomical changes in gastric bypass may contribute in diminishment of the growth and survival of Bifidobacteria and Lactobacilli [26, 32].

Beneficial influences of probiotics on the blood endotoxin markers and pro-inflammatory cytokines could be resulted from alterations in composition of gut microbiota, leading to a diminution in intestinal endotoxin, modulated intestinal barrier function, attenuation of the LPS-induced NF-κB activation, and other immunomodulatory properties [33, 34].

At the end of follow-up period (1 year after the surgery), in contrast to previous studies [10, 35, 36], mainly conducted on RYGB, no changes were observed in the serum LBP levels. Additionally, the results showed a considerable reduction in all the inflammatory factors among all the patients in comparison with pre-operative findings, which is in agreement with other studies [12, 37, 38]. The exact mechanism of the persistent increase of LBP after the surgery is unclear; however, increased intestinal permeability [15], bacterial overgrowth [16, 39], a decline in Firmicutes and Bacteroidetes (belonging to gram-positive bacteria), and an enhancement of

Proteobacteria (belonging to gram-negative bacteria), which were reported by several studies [17, 32], might contribute to this finding.

In contrast to the previously reported beneficial effects of probiotics on serum MDA [20], we observed the serum levels of MDA were unchanged and not influenced by probiotic administration at the end of intervention period. Given that the impact of probiotics on oxidative stress status is considerably strain-dependent [40], the difference in findings may be due to the different strains of bacteria that were employed. Furthermore, dissimilar examined populations might be attributed to this result. It seems that in early months after surgery, rapid substantial weight loss following BS and the consequent immense load of fatty acids in the liver prevented from decreasing lipid peroxidation [41]. Finally, similar to previous studies in