- Email: [email protected]
Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54
YCLNU3193_proof ■ 26 July 2017 ■ 1/7
Clinical Nutrition xxx (2017) 1e7
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Q8 Q7 Q1
Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity Zhen Li a, b, Rosa-Maria Gueant-Rodriguez a, c, *, 1, Didier Quilliot a, c, d, Marie-Aude Sirveaux c, d, David Meyre a, Jean-Louis Gueant a, Laurent Brunaud a, b, d, **, 1 a
Inserm UMRS 954 N-GERE (Nutrition-Genetics-Environmental Risks), University de Lorraine, Faculty of Medicine, Nancy, France Department of Digestive, Hepato-Biliary and Endocrine Surgery, Regional University Hospital of Nancy (CHRU Nancy), Nancy, France Department of Diabetology and Nutrition, Regional University Hospital of Nancy (CHRU Nancy), Nancy, France d Unit e Multidisciplinaire de la chirurgie de l'ob esit e (UMCO), Regional University Hospital of Nancy (CHRU Nancy), Nancy, France b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 13 March 2017 Accepted 6 July 2017
Background: Low vitamin B12 and high folate during pregnancy are associated with visceral obesity and insulin resistance in offspring. In the general population, high folate exacerbates the increase of methylmalonic acid, a marker of vitamin B12 deficiency. However, the influence of vitamin B12 and folate and their related markers on insulin resistance and metabolic syndrome remains unknown in severe obesity. Aim: To evaluate the influence of vitamin B12 and folate on HOMA-IR and components of metabolic syndrome in severe obesity. Methods: 278 consecutive obese patients were assessed prospectively for HOMA-IR, red blood cell (RBC) folates, homocysteine and methylmalonic acid. We compared the associations with the components of metabolic syndrome during the preoperative multidisciplinary evaluation (period-1) and before bariatric surgery (period-2). Results: The HOMA-IR was higher in patients with highest tertile of RBC and either lowest tertile of plasma B12 or highest tertile of MMA (p < 0.034 and 0.011, respectively). Lg HOMA-IR was negatively correlated with Lg homocysteine (p < 0.0001) and positively correlated with Lg serum folate (p < 0.001). The independent predictors for HOMA-IR at period 2 were either BMI and homocysteine (model 1 without serum folate, p ¼ 0.010 and p ¼ 0.002, respectively) or BMI and MMA (model 2 without homocysteine, p ¼ 0.030 and p ¼ 0.004, respectively). Age and RBC folate remained independently associated with the number of metabolic syndrome components (p ¼ 0.006 and 0.020, respectively). Conclusions: RBC folate, homocysteine, and MMA predict HOMA-IR in severe obesity. Our findings challenge the benefit of folate fortified food in severe obesity, in particular in patients with a deficit of vitamin B12. The cohort study was registered at clinicaltrials.gov as NCT02663388. © 2017 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Obesity Folates HOMA-IR
Abbreviations: APRI, Aspartate aminotransferase-to-platelet ratio; ALAT, Alanine Aminotransferase; ASAT, Aspartate Aminotransferase; CHD, Coronary heart disease; DBP, Diastolic blood pressure; HOMA-IR, Homeostasis model assessment of insulin resistance; IDF, International Diabetes Federation; NASH, Non-alcoholic steatohepatitis; NCEPATP, National Cholesterol Education Program Adult Treatment Panel; MMA, Methyl malonic acid; OSAS, Obstructive sleep apnea syndrome; RBC, Red blood cell; SBP, Systolic blood pressure; SIRT1, Sirtuin-1. * Corresponding author. Inserm UMRS 954 N-GERE (Nutrition-Genetics-Environmental Risks), University de Lorraine, Faculty of Medicine of Nancy, 54500, Vandoeuvre les Nancy, France. ** Corresponding author. Inserm UMRS 954 N-GERE, Department of Digestive, Hepato-Biliary and Endocrine Surgery, CHU Nancy (Brabois), University de Lorraine, 11 Allee du Morvan, 54511, Vandoeuvre les Nancy, France. Fax: þ33 38315312. E-mail addresses: [email protected] (R.-M. Gueant-Rodriguez), [email protected] (L. Brunaud). 1 Equal contribution. http://dx.doi.org/10.1016/j.clnu.2017.07.008 0261-5614/© 2017 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 Q2 113 114 115 116 117 118 119
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 2/7
2
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
1. Introduction
2.2. Definition of metabolic syndrome components
Obesity, a global health problem, is associated with components of metabolic syndrome and other comorbidities such as insulin resistance, type 2 diabetes, hypertension, dyslipidemia, nonalcoholic fatty liver disease, renal disease, and cardiovascular disease [1,2]. It is an important burden for society, and represents one of the primary causes of preventable deaths with an increasing prevalence in developed and developing countries during the last decades [3,4]. Consequently, it has become a worldwide priority to understand mechanisms underlying obesity and related comorbidities because those complications are considered to significantly reduce survival in these patients [5]. It is hypothesized that the metabolic syndrome results from gene-environment interactions [6]. The role of one carbon metabolism on these interactions has been poorly investigated. The studies focused on the clinical impact of homocysteine, a marker of the one-carbon metabolism in obese patients [7e10]. The onecarbon metabolism corresponds to a complex set of biochemical reactions in which methyl groups are used in the methylation cycle that produces methionine and S-adenosylmethionine, the universal methyl donor of epigenomic mechanisms [11]. Methionine synthase catalyses the remethylation of homocysteine in methionine, in presence of 5-methyltetrahydrofolate, as co-substrate and vitamin B12 (cobalamin) as co-factor [11]. Vitamin B12 and folate are critical micronutrients associated with foetal programming of obesity. Studies from India and UK have shown that vitamin B12 and folate status during pregnancy is associated with increased obesity and insulin resistance of the mother and birth weight and risk of insulin resistance and obesity in the offspring [12e15]. Furthermore, some data from the general US population have also suggested that high folate intake is associated with aggravation of the clinical manifestations of vitamin B12 deficiency [16]. Taken together, those data emphasize that vitamin B12 and folate should be considered together to evaluate their respective and combined association with metabolic syndrome components, including insulin resistance in obese patients. However, these associations have never been investigated in patients with severe obesity treated by bariatric surgery. In this study, we aim to evaluate the role of vitamin B12, folate and their related markers, homocysteine and methylmalonic acid on the components of the metabolic syndrome in severely obese patients assessed in two periods, the preoperative multidisciplinary evaluation for bariatric surgery (period 1) and in the hours preceding surgery (period 2).
Metabolic syndrome was defined by using the criteria proposed by the National Cholesterol Education Program Adult Treatment Panel III (NCEP_ATP III) and the international diabetes federation (IDF 2005 and IDF 2009) [2,3]. Waist circumference was evaluated during preoperative multidisciplinary evaluation (period 1) and was considered as a metabolic syndrome component when 94 cm in male and 80 cm in female. Hypertension was defined as SBP 130 mm Hg or DBP 85 mm Hg or the need for antihypertensive medication(s). Dyslipidemia was defined as plasma triglycerides >1.5 g/L or hypo HDL-cholesterol <0.4 (male) or 0.5 (female) g/L. Type 2 diabetes was defined as fasting glycemia 1.0 g/L or the need for treatment. Overall, metabolic syndrome severity was evaluated by counting the presence or absence of those 5 metabolic syndrome components (waist size, hypertension, hypertriglyceridemia, low HDL-cholesterol, diabetes) in all patients during period 1. We also evaluated other criteria corresponding to obesity-related comorbidities during preoperative multidisciplinary evaluation as coronary heart disease, obstructive sleep apnea syndrome, and non-alcoholic steatohepatitis. Coronary heart disease (CHD) was defined as a group of diseases including stable angina, unstable angina, and myocardial infarction. Obstructive sleep apnea syndrome (OSAS) was defined as a sleep disorder involving cessation or significant decrease in airflow in the presence of breathing effort. Non-alcoholic steatohepatitis (NASH) was defined as a type of fatty liver which occurs when fat is deposited (steatosis) in the liver due to causes other than excessive alcohol use.
2. Materials/subjects and methods 2.1. Study population All patients had a BMI 35 kg/m2 and were included in this study during preoperative multidisciplinary evaluation for bariatric surgery. All patients signed specific informed consents. Investigations conformed to the principles outlined in the Declaration of Helsinki and received approval from the local ethical committee. This study was registered at French ministry of health as NCT02663388. Clinical data and biological data were collected during preoperative multidisciplinary evaluation for bariatric surgery (period 1). Then, all patients had a fasting venous blood sample collection in the morning of their bariatric surgery (period 2). Those tubes were centrifuged the same morning and corresponding plasma was collected and stored at 20 C until analysis. Preoperative vitamin B12 deficiency was defined as B12 plasma concentration <148 pmol/L [16]. High preoperative plasma folate concentration was defined as folate plasma concentration >20 nmol/mL [16].
2.3. Biochemical analyses Fasting plasma concentrations of glucose, cholesterol, triglycerides, HDL-cholesterol, ASAT and ALAT were determined by the enzymatic or dry chemistry methods used on automata U 2700 Olympus Corporation, USA. LDL-cholesterol values were calculated using the Friedewald formula: [LDL-chol] ¼ [Total chol] [HDLchol] ([TG]/2.2) using the unit in mmol/L. Plasma insulin were determined by the kit MP Biomedicale (Solon, Ohio 44139, USA) for radioimmunoassay of insulin in serum or EDTA. The HOMA-IR was calculated using the model proposed by Levy et al. and corresponded to HOMA-IR ¼ Glucose Insulin/22.5 using the unit of glucose in mmol/L and the unit of insulin in mUI/ml [17]. Vitamin B12 and folate serum concentrations were determined by the SNB SimulTRAC-box for Radio Immunoassay Vitamin B12 [57Co]/Folate [125]. Red blood cell folate (RBC folate) was determined using the same radio immunoassay method after total blood extract hemolysis in presence of ascorbic acid. Homocysteine and methylmalonic acid (MMA) were determined by UPLC-MS/MS, by using a ACQUITY UPLC BEH C18 column (1.7 mm, 2.1 mm 50 mm, Waters Corporation) [18]. 2.4. Statistical analysis Statistical analyze were performed by SPSS version 23 (IBM corporation, Armonk, NY, USA). The minimal size of our sample was estimated at 250 patients, with a study power 1-b ¼ 0.8 and a ¼ 0.05, assuming a 2.0-fold difference in the folate or B12 concentrations between patients with and without outcomes of metabolic syndrome. Continuous data were reported as mean ± standard deviation. Incidence of nominal variables was reported as percentage. Non-parametric tests were used when data distribution was not normal. KruskaleWallis test was used between groups when there were 3; p-values < 0.05 were considered as significant. Continuous variables that had asymmetrical
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 3/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
distribution were classified in tertiles to study their combined association with HOMA-IR. Multivariate analysis was performed to evaluated independent risk factors for insulin resistance (HOMAIR), type 2 diabetes (fasting glucose > 1.0 g/L or treatment), and metabolic syndrome components (dependent variables) using multiple regression and logistic regression when needed. Corrections for multiple testing were made using the Bonferroni test. 3. Results 3.1. Main clinical and biological characteristics of patients Two hundreds and seventy eight patients were included in this study. The patients were recruited at the Nancy University Hospital Centre during a multidisciplinary evaluation of pathological severe obesity (period 1). They were subsequently evaluated 14.8 months later (range: 6e53 months) in the hours preceding bariatric surgery (period 2). The main clinical and biological characteristics of both periods are reported in Tables 1 and 2. Mean age was 45 ± 11.2 years and 221 patients were female (79.4%). The patients had a mean BMI at 48 þ 7.5 kg/m2 in period 1. We observed that 15 patients had a deficit in vitamin B12 defined by a serum concentration <148 pmol/ L. The concentration of vitamin B12 was not influenced by the treatment with metformin (B12 concentration of patients treated with and without metformin: 252 þ 112 pmol/L vs. 248 þ 98, p ¼ 0.821), in contrast to other studies for which type 2 diabetes was the main recruitment criteria [19,20]. None of the patients had a deficit in folate according to RBC folate. At period 2, we observed that the mean BMI decreased significantly from 48 ± 7.5 to 46 ± 6.6 kg/m2 (p < 0.001) corresponding to a mean loss of weight of 6.1 ± 10 kg. This weight loss resulted from the improved dietary management of the patients between the two evaluations. We reported 37 patients (13.3%) with low serum concentration of vitamin B12 (<148 pmol/L) and 27 patients (9.7%) with folate serum concentration above 20 nmol/mL. The latter were considered to have high folic acid intake. As expected, folate serum concentration was negatively correlated with plasma homocysteine (0.272, 0.377 to 0.159; p < 0.0001). The patients with low serum vitamin B12 had a higher mean plasma concentration of homocysteine (15.6 vs. 13.4 mmol/L; p ¼ 0.039) and MMA (0.178 vs. 0.154 mmol/L; p ¼ 0.026), in comparison with the other patients. As previously reported in other population studies, homocysteine was closely correlated with serum folate (p < 0.001) (Fig. 1A) and the highest concentrations of MMA in patients with low serum vitamin B12
3
Table 2 Metabolic syndrome components and other characteristics in 278 patients accordQ5 ing to IDF definition (2). Comorbidity Metabolic syndrome components Waist size, 94 cm in male and 80 cm in female Hypertension, SBP 130 mm Hg or DBP 85 mm Hg or antihypertensive medication Dyslipidemia - Triglycerides > 1,5 g/L - Hypo HDL-cholesterol < 0.4 (M) or 0.5 (F) g/L - Both - Fasting glycemia 1,0 g/L without treatment - Treated diabetes, fasting glycemia 1,0 g/L or treatment Other characteristics NASH OSAS CHD Male
Number of patients (n)
Percentage of patients (%)
278
100
118
42.4
235 28 108 91 53 84
84.5 10.1 38.9 32.7 19.1 30.2
26 139 18 58
9.4 50.0 6.5 20.8
CHD: coronary heart disease; NASH: non-alcoholic steatohepatitis; OSAS: obstructive sleep apnea syndrome.
were reported in cases with either low or high folate serum concentration, compared to those with intermediate folate concentration (Fig. 1B). 3.2. Predictors of insulin resistance Only the 194 patients who had no known type 2 diabetes and no oral anti-diabetic treatment were assessed for HOMA-IR in period 1 while all patients were assessed in period 2. We observed a significant association of HOMA-IR with RBC folate concentration (p ¼ 0.008) and no significant association with serum vitamin B12, in period 1. In multivariate analysis, HDL-cholesterol and RBC folate remained independent predictors for HOMA-IR (p ¼ 0.034 and p ¼ 0.008, respectively, Table 3). Patients in the lowest tertile of plasma B12 (<238 pmol/L) and highest tertile of RBC (>948 nmol/L) had the highest HOMA-IR (p < 0.05) (Fig. 2A). We confirmed the association of HOMA-IR with low B12/high folate in period 2. Patients with the lowest serum concentration of B12 (tertile < 200 pmol/L) associated with the highest values for plasma folate (tertile > 10.2 nmol/L) were those with the highest HOMA-IR values (Fig. 2B). In addition, patients with the highest concentration
Table 1 Clinical and biological characteristics in the two periods of the multidisciplinary evaluation of patients. Parameters
During multidisciplinary evaluation e period 1 n
Age, years BMI, kg/m2 Fasting insulin, mUI/mL Fasting glucose, g/L HOMA-IR HDL-cholesterol, g/L LDL-cholesterol, g/L Total cholesterol, g/L Triglycerides, g/L RBC folate, nmol/L Plasma folate, nmol/L Plasma B12, pmol/L Homocysteine, mmol/L MMA, mmol/L
278 278 186 192 183 248 248 250 250 220 201 226 ND ND
At bariatric surgery e period 2
Mean ± Standard Deviation
Median
Lower tertile
Upper tertile
n
Mean ± Standard Deviation
Median
Lower tertile
Upper tertile
p-value
46 ± 7.5 24.7 ± 19 0.99 ± 0.18 6.05 ± 4.3 0.45 ± 0.1 1.23 ± 0.33 2.02 ± 0.36 1.74 ± 0.89 936.5 ± 464 13.9 ± 8.3 298 ± 145 ND ND
44.5 20.6 0.96 5.04 0.43 1.21 2.00 1.56 797.5 11.4 274 ND ND
42.2 17.5 0.91 4.15 0.40 1.08 1.86 1.20 681.8 9.6 238 ND ND
47.3 26.5 1.04 6.48 0.48 1.33 2.11 1.90 940.7 14.6 317 ND ND
278 209 278 277 277 277 277 277 277 ND 278 278 278 278
45 ± 11 46.6 ± 6.6 54.7 ± 50 1.06 ± 0.39 16.4 ± 27 0.41 ± 0.1 1.15 ± 0.32 1.87 ± 0.35 1.56 ± 0.69 ND 11.9 ± 7.9 265 ± 294 13.7 ± 4.8 0.16 ± 0.07
45 45.4 43.5 0.94 10.4 0.41 1.13 1.86 1.42 ND 9.6 239 13.0 0.14
40 42.3 33.4 0.86 7.5 0.37 1.01 1.72 1.17 ND 8.1 200 11.4 0.12
51 48.4 54.2 1.05 13.9 0.44 1.28 2.01 1.65 ND 12.7 281 14.4 0.17
NA 0.2537 <0.0001 0.3479 <0.0001 0.0001 0.0061 <0.0001 0.0326 NA 0.0027 <0.0001 NA NA
MMA: methylmalonic acid; NA: no applicable; ND: no data; RBC: red blood cell.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 4/7
4
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
3.3. Predictors of metabolic syndrome severity We evaluated further the predictors of metabolic syndrome severity by considering their association with the number of components, including hypertension, hypertriglyceridemia (>1.5 g:L), low HDL-cholesterol (<0.5 in females and <0.4 in male) and type 2 diabetes defined according to the International Diabetes Federation [2]. The presence of 4, 3 and 2 components was reported in 69, 95 and 95 cases, respectively. Hypertension was the less frequent outcome in patients with 2 components (8 cases) and hypertension the less frequent in patients with 3 components (41 cases). Age, RBC folate, ALAT, ASAT and MMA were the 5 significant predictors of the number of metabolic syndrome components in univariate analysis (Table 4). In multivariate analysis, age and RBC folate remained independently associated with the number of metabolic syndrome components (Table 4, p ¼ 0.010 and 0.020, respectively).
4. Discussion
Fig. 1. A: Correlation (Pearson analysis) of logarithmically transformed homocysteine (Log-HCY) with logarithmically transformed folate. Dotted lines: 95% confidence intervals of the slope. B: Mean MMA concentration was the highest in patients with low vitamin B12 plasma concentration with either the lowest or highest folate plasma concentrations, compared to those with intermediate folate concentration at period 2 (n ¼ 278). Low vitamin B12 plasma concentration (vitamin B12 deficiency) was defined as inferior to 150 pmol/L. MMA values were transformed into logarithms. P for trend analysis was used within plasma vitamin B12 categories.
of MMA had the highest HOMA-IR values (Fig. 2C). Age, BMI, triglycerides, HDL-cholesterol, serum folate, homocysteine and MMA were significantly associated with HOMA-IR in univariate analyses at period 2 (Table 3). Log-transformed HOMA-IR score was negatively correlated with Log-transformed homocysteine (p < 0.0001) (Fig. 3A) and positively correlated with Log-transformed serum folate (p < 0.001) (Fig. 3B) and Log-transformed MMA (Fig. 3C). We performed therefore the multivariate analysis in two models where folate and Hcy were alternatively excluded. BMI and homocysteine (model 1 without serum folate, p ¼ 0.010 and p ¼ 0.002, respectively) or BMI and methylmalonic acid (model 2 without homocysteine, p ¼ 0.030 and p ¼ 0.004, respectively) remained independent predictors for HOMA-IR values (Table 3).
Metabolic syndrome is frequently observed in patients with morbid obesity and is associated with a 5-fold risk of type 2 diabetes and a 2-fold risk of cardiovascular outcomes [21]. The incidence and distribution of the components of metabolic syndrome observed in our study are consistent with these previous data. We identified HDL-cholesterol and RBC folate as independent predictors for HOMA-IR at period 1 while independent predictors for HOMA-IR values at period 2 were either BMI and homocysteine (model 1 without serum folate) or BMI and methylmalonic acid (model 2 without homocysteine). We observed also an influence of markers of B12 deficit, either lowest B12 or highest MMA concentrations in the association between high folate status and HOMA-IR. Age and RBC folate remained independent predictors for the number of metabolic syndrome components. The deficit in vitamin B12 and/or folate leads to the disruption of the one-carbon cycle and its downstream metabolic processes [22]. Our study showed that high folate status was associated with insulin resistance and the number of components of metabolic syndrome in severely obese patients. The potential harmful effects of abnormal folic acid concentration have been suggested in adults and pediatrics population [23,24]. Lowering homocysteine concentration by daily supplementation of vitamins B6, B9 and B12 is not associated with a significant reduction of developing type 2 diabetes, in women with high risk for cardiovascular diseases [25]. High blood folate concentration has been linked to aggravation of anemia and cognitive impairment and increased MMA and
Table 3 Determinants of HOMA-IR in univariate and multivariate analysis at two periods. Parameters
Period 1 Univariate
Age BMI Triglycerides HDL-cholesterol LDL-cholesterol Creatinine RBC folate Serum folate B12 Homocysteine MMA
Q6 Period 2
n
Rho
p-value
0.1054 0.1660 0.2338 0.2213 0.1555 0.0244 0.2135 ND 0.0695 ND ND
0.1874 0.0305 0.0023 0.0041 0.0448 0.7555 0.0085 ND 0.3917 ND ND
Multivariate Coefficient
158 170 168 167 167 165 151 ND 154 ND ND
Univariate p-value
5.6310
0.034
0.0014
0.008
n
Rho
p-value
0.1368 0.1366 0.2439 0.3128 0.2349 0.0697 ND 0.2167 0.0520 0.2286 0.1524
0.0228 0.0491 <0.0001 <0.0001 0.0001 0.2490 ND 0.0003 0.3886 0.0001 0.0111
277 208 277 277 277 275 ND 277 277 277 277
Multivariate Model 1
Multivariate Model 2
Coefficient
p-value
Coefficient
p-value
0.7661
0.0050
0.6816
0.030
1.3852
0.002 83.7915
0.004
MMA: methylmalonic acid; ND: no data; RBC: red blood cell. Corrections for multiple testing were made using the Bonferroni test.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 5/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
5
Fig. 2. A: Patients with the lowest values for plasma B12 (tertile < 241 pmol/L) associated with the highest values for RBC folate (tertile > 953 nmol/L) had the highest HOMA-IR values at period 1 (n ¼ 183). B: Patients at period 2 (n ¼ 277) with the lowest values for plasma B12 (tertile < 200 pmol/L) associated with the highest values for plasma folate (tertile > 10.2 nmol/L) were those with the highest HOMA-IR values. HOMA-IR values in figures A and B were transformed into logarithms. P for trend analysis was used within each plasma vitamin B12 categories (*: p < 0.05). C: Patients with the highest folate and MMA values (tertiles) had the highest mean HOMA-IR values at period 2 (n ¼ 277). HOMA-IR values were transformed into logarithms. P for trend analysis was used within MMA categories.
Fig. 3. A, B, C: Correlation (Pearson analysis) of logarithmically transformed HOMA-IR with logarithmically transformed homocysteine (A, Lg Hcy), folate (B, Lg Folate) and methylmalonic acid (C, Lg MMA).
homocysteine, in patients with low vitamin B12 plasma concentration [16,26]. Consistently, we confirmed the highest MMA concentration in patients with the lowest vitamin B12/highest folate concentrations [26] and a negative correlation between homocysteine and HOMA-IR. Furthermore, transcobalamin 776C/G polymorphism has been shown to be associated with peripheral neuropathy in elderly individuals with high folate intake [27]. In contrast, hyperhomocysteine has been associated with obesity, hypertension and type 2 diabetes mellitus [28]. It correlates also with insulin resistance and low-grade systemic inflammation in obese prepubertal children [29]. Reduced folate concentration is associated with epigenomic changes in liver tissue from subjects with type 2 diabetes [30]. The discrepancy between these and our data may be explained by differences in the severity of obesity and age of patients among studied populations.
We found that the patients with either high folate/low B12 or high folate/high MMA were those with the highest risk of insulin resistance. This was consistent with the negative correlation of HOMA-IR with homocysteine and its positive correlation with folate and MMA. In addition, as much as 10% of our patients had a folate serum concentration above 20 nmol/mL that could reflect a high folate intake. The synergic influence of the high intake of folate and B12 deficit on the risk of insulin resistance is consistent with previous data. An indian study showed that normal to high maternal folate status associated with a low maternal vitamin B12 status led to babies having a higher adiposity and insulin resistance [31]. In the same study, over 60% of pregnant women were B12 deficient. Obese offsprings from mothers with lowest values for plasma B12 and highest values of folate were those with the more severe insulin resistance. Low plasma vitamin B12 during
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 6/7
6
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
Table 4 Predictors of the number of metabolic syndrome components. Criteria
Age, years Male gender, % BMI, kg/m2 OSAS comorbidity, % NASH comorbidity, % CHD comorbidity, % RBC folate, nmol/L Plasma folate, nmol/L Vitamin B12, pmol/L Homocysteine, mmol/L MMA, mmol/L
Number of metabolic syndrome components 1 (n ¼ 21)
2 (n ¼ 93)
3 (n ¼ 95)
4 (n ¼ 69)
40.6 ± 11 5 (24) 47.8 ± 6 7 (33) 0 (0) 0 (0) 835 ± 347 9.9 ± 5 214 ± 72 14.6 ± 4 0.157 ± 0.04
40.5 ± 11 11 (12) 45.1 ± 5 34 (36) 5 (5.4) 1 (1.07) 821 ± 368 10.7 ± 6 239 ± 91 13.0 ± 3 0.147 ± 0.06
46.6 ± 10 20 (21) 45.5 ± 6 56 (59) 12 (12.6) 8 (8.4) 950 ± 412 13.1 ± 11 320 ± 483 13.6 ± 6 0.153 ± 0.08
50.3 ± 10 22 (32) 47.1 ± 8 42 (61) 9 (13) 9 (13) 1116 ± 620 12.6 ± 7 243 ± 101 14.4 ± 5 0.175 ± 0.09
p-for trend
OR
95% C.I.
p-value
<0.001 0.020 0.339 0.001 0.101 0.009 0.016 0.082 0.073 0.270 0.036
1.04
1.01e1.07
0.010
1.02
1.01e3.03
0.020
CHD: coronary heart disease; NASH: non-alcoholic steatohepatitis; OSAS: obstructive sleep apnea syndrome. Corrections for multiple testing were made using the Bonferroni test.
pregnancy is also considered to be associated with an increased risk of gestational and later diabetes leading to contribute to the increased risk in offsprings [32,33]. Folic acid fortification has been introduced in north America two decades ago and its balancing between risks and benefits is under discussion in Europe [31]. Folic acid fortification is clearly associated with increased folate status in general population [16,32]. For example, the introduction of folate fortification in Canada has dramatically increased the maternal RBC folate concentration and cord RBC folate. This highlights the need for investigating long-term health outcomes of high RBC folate in offsprings [33]. There are potential reasons explaining why vitamin B12 deficiency and MMA could influence the risk of insulin resistance in the presence of normal or high folate. B12 deficiency produces a cellular folate trap that lead to impaired methionine synthesis. It produces also an impaired fatty acid oxidation, increased ER stress and impaired expression of SIRT1, a key actor of molecular mechanisms of pathological obesity, in rats [34]. Impaired conversion of MMA to succinyl coA, for which vitamin B12 acts as a co-enzyme, is associated with the accumulation of MMA leading to increased lipogenesis and insulin resistance. Despite folate is not considered to be directly involve in these pathways, higher folate is associated with increased concentration of MMA in the context of vitamin B12 deficiency, as reported in our and previous studies [16,35]. Lastly, vitamin B12 deficiency can increase plasma and RBC folate concentrations [36]. We found an association of age, RBC folate and APRI with the number of components of metabolic syndrome (hypertension, hypertriglyceridemia, low HDL-cholesterol and type 2 diabetes), in multivariate analysis. Previous studies have found an association of nutritional, metabolic and genetic determinants of the one carbon metabolism with some of the components of metabolic syndrome, but none considered the association of folate, B12, homocysteine and MMA with all components together, in severely obese patients. We showed in the OASI population of elderly ambulatory volunteers that plasma total homocysteine correlated negatively with HDL-Cholesterol [37]. In contrast to our results on the association between high RBC values of folate and increased HOMA and dyslipidemia, a double-blind placebo-controlled clinical trial concluded that 5 mg folate supplementation resulted in reduced HOMA-IR and HDL-cholesterol in women with polycystic ovary syndrome [38]. These discrepancies may be explained by the difference between the two populations, since polycystic ovary syndrome is a very specific cause of insulin resistance. Other studies have found a benefit of supplementation with other B vitamins and methyl donors on outcomes of metabolic syndrome. Horigan et al. demonstrated a significant blood pressure-lowering response to intervention with riboflavin for 16 weeks in hypertensive patients [39]. In the HUSK cohort, choline and betaine were associated in
opposite directions with serum triglycerides, glucose, BMI, percent body fat, waist circumference and systolic and diastolic blood pressure [40]. In conclusion, we found that homocysteine, high RBC folate and methylmalonic acid are significant predictors of the insulin resistance and number of metabolic syndrome components in severely obese patients. The highest risk of insulin resistance was observed in patients with highest folate status and either lowest vitamin B12 or highest MMA. These findings should deserve further interest for replicating our study and studying the underlying molecular mechanisms. They may have important implications for the management of severely obese patients and for evaluating the risk/ benefit balance of folate food fortification in this population. Sources of support te This project is supported by Inserm, SOFFCO-MM (SOcie site et des Maladies Française et Francophone de Chirurgie de l'Obe taboliques e 2014 scholarship) and the AGIR grant from the Me University regional hospital of Nancy and The region of Lorraine. ZL was recipient of a graduate student fellowship (#201308070048) from the China Scholarship Council (CSC). Author contributions ZL conducted research, analyzed data, performed statistical analysis, wrote paper and had primary responsibility for final content. LPB, RMGR and JLG designed research, conducted research, analyzed data, performed statistical analysis, wrote paper and had primary responsibility for final content; DQ and MAS conducted research and provided essential materials; DM analyzed databases and revised manuscript. Conflict of interest The authors declare no conflict of interest. References [1] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365: 1415e28. [2] Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120:1640e5. [3] Genser L, Casella Mariolo JR, Castagneto-Gissey L, Panagiotopoulos S, Rubino F. Obesity, type 2 diabetes, and the metabolic syndrome: pathophysiologic relationships and guidelines for surgical intervention. Surg Clin N Am 2016;96: 681e701.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
Q3
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
YCLNU3193_proof ■ 26 July 2017 ■ 7/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
Q4
[4] Malik VS, Willett WC, Hu FB. Global obesity: trends, risk factors and policy implications. Nat Rev Endocrinol 2013;9:13e27. [5] Stern MP, Williams K, Gonzalez-Villalpando C, Hunt KJ, Haffner SM. Does the metabolic syndrome improve identification of individuals at risk of type 2 diabetes and/or cardiovascular disease? Diabetes Care 2004;27:2676e81. [6] Reddon H, Gueant JL, Meyre D. The importance of gene-environment interactions in human obesity. Clin Sci Lond 2016;130:1571e97. [7] Gallistl S, Sudi K, Mangge H, Erwa W, Borkenstein M. Insulin is an independent correlate of plasma homocysteine levels in obese children and adolescents. Diabetes Care 2000;23:1348e52. [8] Kahleova R, Palyzova D, Zvara K, Zvarova J, Hrach K, Novakova I, et al. Essential hypertension in adolescents: association with insulin resistance and with metabolism of homocysteine and vitamins. Am J Hypertens 2002;15:857e64. [9] Sreckovic B, Sreckovic VD, Soldatovic I, Colak E, Sumarac-Dumanovic M, Janeski H, et al. Homocysteine is a marker for metabolic syndrome and atherosclerosis. Diabetes Metab Syndr 2016. S1871-4021:30200e30204. [10] Catena C, Colussi G, Sechi LA. Response to “plasma Homocysteine levels and endothelial dysfunction in cerebro- and cardiovascular diseases in the metabolic syndrome”. Am J Hypertens 2015;28:1490. [11] Gueant JL, Elakoum R, Ziegler O, Coelho D, Feigerlova E, Daval JL, et al. Nutritional models of foetal programming and nutrigenomic and epigenomic dysregulations of fatty acid metabolism in the liver and heart. Pflugers Arch 2014;466:833e50. [12] Knight BA, Shields BM, Brook A, Hill A, Bhat DS, Hattersley AT, et al. Lower circulating B12 is associated with Higher obesity and insulin resistance during pregnancy in a non-diabetic white British population. PLoS One 2015;10, e0135268. [13] Yajnik CS, Chandak GR, Joglekar C, Katre P, Bhat DS, Singh SN, et al. Maternal homocysteine in pregnancy and offspring birthweight: epidemiological associations and Mendelian randomization analysis. Int J Epidemiol 2014;43: 1487e97. [14] Yajnik CS, Deshmukh US. Maternal nutrition, intrauterine programming and consequential risks in the offspring. Rev Endocr Metab Disord 2008;9: 203e11. [15] Stewart CP, Christian P, LeClerq SC, West Jr KP, Khatry SK. Antenatal supplementation with folic acid þ iron þ zinc improves linear growth and reduces peripheral adiposity in school-age children in rural Nepal. Am J Clin Nutr 2009;90:132e40. [16] Selhub J, Rosenberg IH. Excessive folic acid intake and relation to adverse health outcome. Biochimie 2016;126:71e8. [17] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21: 2191e2. [18] Amouzou EK, Chabi NW, Adjalla CE, Rodriguez-Gueant RM, Feillet F, Villaume C, et al. High prevalence of hyperhomocysteinemia related to folate deficiency and the 677C–>T mutation of the gene encoding methylenetetrahydrofolate reductase in coastal West Africa. Am J Clin Nutr 2004;79: 619e24. [19] de Jager J, Kooy A, Lehert P, Wulffele MG, van der Kolk J, Bets D, et al. Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial. BMJ 2010;340, c2181. [20] Aroda VR, Edelstein SL, Goldberg RB, Knowler WC, Marcovina SM, Orchard TJ, et al. Long-term metformin use and vitamin B12 deficiency in the diabetes prevention program outcomes study. J Clin Endocrinol Metabol 2016;101: 1754e61. [21] Grundy SM. Metabolic syndrome: a multiplex cardiovascular risk factor. J Clin Endocrinol Metabol 2007;92:399e404. [22] Finer S, Saravanan P, Hitman G, Yajnik C. The role of the one-carbon cycle in the developmental origins of Type 2 diabetes and obesity. Diabet Med 2014;31:263e72.
7
[23] Sudchada P, Saokaew S, Sridetch S, Incampa S, Jaiyen S, Khaithong W. Effect of folic acid supplementation on plasma total homocysteine levels and glycemic control in patients with type 2 diabetes: a systematic review and metaanalysis. Diabetes Res Clin Pract 2012;98:151e8. [24] Dehkordi EH, Sedehi M, Shahraki ZG, Najafi R. Effect of folic acid on homocysteine and insulin resistance of overweight and obese children and adolescents. Adv Biomed Res 2016;5:88. [25] Song Y, Cook NR, Albert CM, Van Denburgh M, Manson JE. Effect of homocysteine-lowering treatment with folic Acid and B vitamins on risk of type 2 diabetes in women: a randomized, controlled trial. Diabetes 2009;58: 1921e8. [26] Selhub J, Morris MS, Jacques PF. In vitamin B12 deficiency, higher serum folate is associated with increased total homocysteine and methylmalonic acid concentrations. Proc Natl Acad Sci U S A 2007;104:19995e20000. [27] Sawaengsri H, Bergethon PR, Qiu WQ, Scott TM, Jacques PF, Selhub J, et al. Transcobalamin 776C–>G polymorphism is associated with peripheral neuropathy in elderly individuals with high folate intake. Am J Clin Nutr 2016;104:1665e70. [28] Fonseca VA, Fink LM, Kern PA. Insulin sensitivity and plasma homocysteine concentrations in non-diabetic obese and normal weight subjects. Atherosclerosis 2003;167:105e9. [29] Martos R, Valle M, Morales R, Canete R, Gavilan MI, Sanchez-Margalet V. Hyperhomocysteinemia correlates with insulin resistance and low-grade systemic inflammation in obese prepubertal children. Metabolism 2006;55: 72e7. [30] Nilsson E, Matte A, Perfilyev A, de Mello VD, Kakela P, Pihlajamaki J, et al. Epigenetic alterations in Human liver from subjects with type 2 diabetes in parallel with reduced folate levels. J Clin Endocrinol Metabol 2015;100: E1491e501. [31] Smith AD, Refsum H, Selhub J, Rosenberg IH. Decision on folic acid fortification in Europe must consider both risks and benefits. BMJ 2016;352, i734. [32] Smith AD, Kim YI, Refsum H. Is folic acid good for everyone? Am J Clin Nutr 2008;87:517e33. [33] Plumptre L, Masih SP, Ly A, Aufreiter S, Sohn KJ, Croxford R, et al. High concentrations of folate and unmetabolized folic acid in a cohort of pregnant Canadian women and umbilical cord blood. Am J Clin Nutr 2015;102:848e57. [34] Gueant JL, Namour F, Gueant-Rodriguez RM, Daval JL. Folate and fetal programming: a play in epigenomics? Trends Endocrinol Metabol: TEM 2013;24: 279e89. [35] Krishnaveni GV, Hill JC, Veena SR, Bhat DS, Wills AK, Karat CL, et al. Low plasma vitamin B12 in pregnancy is associated with gestational 'diabesity' and later diabetes. Diabetologia 2009;52:2350e8. [36] Krishnaveni GV, Hill JC, Leary SD, Veena SR, Saperia J, Saroja A, et al. Anthropometry, glucose tolerance, and insulin concentrations in Indian children: relationships to maternal glucose and insulin concentrations during pregnancy. Diabetes Care 2005;28:2919e25. [37] Gueant-Rodriguez RM, Spada R, Moreno-Garcia M, Anello G, Bosco P, Lagrost L, et al. Homocysteine is a determinant of ApoA-I and both are associated with ankle brachial index, in an ambulatory elderly population. Atherosclerosis 2011;214:480e5. [38] Asemi Z, Karamali M, Esmaillzadeh A. Metabolic response to folate supplementation in overweight women with polycystic ovary syndrome: a randomized double-blind placebo-controlled clinical trial. Mol Nutr Food Res 2014;58:1465e73. [39] Horigan G, McNulty H, Ward M, Strain JJ, Purvis J, Scott JM. Riboflavin lowers blood pressure in cardiovascular disease patients homozygous for the 677C/T polymorphism in MTHFR. J Hypertens 2010;28:478e86. [40] Konstantinova SV, Tell GS, Vollset SE, Nygard O, Bleie O, Ueland PM. Divergent associations of plasma choline and betaine with components of metabolic syndrome in middle age and elderly men and women. J Nutr 2008;138: 914e20.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100
YCLNU3193_proof ■ 26 July 2017 ■ 1/7
Clinical Nutrition xxx (2017) 1e7
Contents lists available at ScienceDirect
Clinical Nutrition journal homepage: http://www.elsevier.com/locate/clnu
Original article
Q8 Q7 Q1
Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity Zhen Li a, b, Rosa-Maria Gueant-Rodriguez a, c, *, 1, Didier Quilliot a, c, d, Marie-Aude Sirveaux c, d, David Meyre a, Jean-Louis Gueant a, Laurent Brunaud a, b, d, **, 1 a
Inserm UMRS 954 N-GERE (Nutrition-Genetics-Environmental Risks), University de Lorraine, Faculty of Medicine, Nancy, France Department of Digestive, Hepato-Biliary and Endocrine Surgery, Regional University Hospital of Nancy (CHRU Nancy), Nancy, France Department of Diabetology and Nutrition, Regional University Hospital of Nancy (CHRU Nancy), Nancy, France d Unit e Multidisciplinaire de la chirurgie de l'ob esit e (UMCO), Regional University Hospital of Nancy (CHRU Nancy), Nancy, France b c
a r t i c l e i n f o
s u m m a r y
Article history: Received 13 March 2017 Accepted 6 July 2017
Background: Low vitamin B12 and high folate during pregnancy are associated with visceral obesity and insulin resistance in offspring. In the general population, high folate exacerbates the increase of methylmalonic acid, a marker of vitamin B12 deficiency. However, the influence of vitamin B12 and folate and their related markers on insulin resistance and metabolic syndrome remains unknown in severe obesity. Aim: To evaluate the influence of vitamin B12 and folate on HOMA-IR and components of metabolic syndrome in severe obesity. Methods: 278 consecutive obese patients were assessed prospectively for HOMA-IR, red blood cell (RBC) folates, homocysteine and methylmalonic acid. We compared the associations with the components of metabolic syndrome during the preoperative multidisciplinary evaluation (period-1) and before bariatric surgery (period-2). Results: The HOMA-IR was higher in patients with highest tertile of RBC and either lowest tertile of plasma B12 or highest tertile of MMA (p < 0.034 and 0.011, respectively). Lg HOMA-IR was negatively correlated with Lg homocysteine (p < 0.0001) and positively correlated with Lg serum folate (p < 0.001). The independent predictors for HOMA-IR at period 2 were either BMI and homocysteine (model 1 without serum folate, p ¼ 0.010 and p ¼ 0.002, respectively) or BMI and MMA (model 2 without homocysteine, p ¼ 0.030 and p ¼ 0.004, respectively). Age and RBC folate remained independently associated with the number of metabolic syndrome components (p ¼ 0.006 and 0.020, respectively). Conclusions: RBC folate, homocysteine, and MMA predict HOMA-IR in severe obesity. Our findings challenge the benefit of folate fortified food in severe obesity, in particular in patients with a deficit of vitamin B12. The cohort study was registered at clinicaltrials.gov as NCT02663388. © 2017 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Keywords: Obesity Folates HOMA-IR
Abbreviations: APRI, Aspartate aminotransferase-to-platelet ratio; ALAT, Alanine Aminotransferase; ASAT, Aspartate Aminotransferase; CHD, Coronary heart disease; DBP, Diastolic blood pressure; HOMA-IR, Homeostasis model assessment of insulin resistance; IDF, International Diabetes Federation; NASH, Non-alcoholic steatohepatitis; NCEPATP, National Cholesterol Education Program Adult Treatment Panel; MMA, Methyl malonic acid; OSAS, Obstructive sleep apnea syndrome; RBC, Red blood cell; SBP, Systolic blood pressure; SIRT1, Sirtuin-1. * Corresponding author. Inserm UMRS 954 N-GERE (Nutrition-Genetics-Environmental Risks), University de Lorraine, Faculty of Medicine of Nancy, 54500, Vandoeuvre les Nancy, France. ** Corresponding author. Inserm UMRS 954 N-GERE, Department of Digestive, Hepato-Biliary and Endocrine Surgery, CHU Nancy (Brabois), University de Lorraine, 11 Allee du Morvan, 54511, Vandoeuvre les Nancy, France. Fax: þ33 38315312. E-mail addresses: [email protected] (R.-M. Gueant-Rodriguez), [email protected] (L. Brunaud). 1 Equal contribution. http://dx.doi.org/10.1016/j.clnu.2017.07.008 0261-5614/© 2017 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 Q2 113 114 115 116 117 118 119
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 2/7
2
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
1. Introduction
2.2. Definition of metabolic syndrome components
Obesity, a global health problem, is associated with components of metabolic syndrome and other comorbidities such as insulin resistance, type 2 diabetes, hypertension, dyslipidemia, nonalcoholic fatty liver disease, renal disease, and cardiovascular disease [1,2]. It is an important burden for society, and represents one of the primary causes of preventable deaths with an increasing prevalence in developed and developing countries during the last decades [3,4]. Consequently, it has become a worldwide priority to understand mechanisms underlying obesity and related comorbidities because those complications are considered to significantly reduce survival in these patients [5]. It is hypothesized that the metabolic syndrome results from gene-environment interactions [6]. The role of one carbon metabolism on these interactions has been poorly investigated. The studies focused on the clinical impact of homocysteine, a marker of the one-carbon metabolism in obese patients [7e10]. The onecarbon metabolism corresponds to a complex set of biochemical reactions in which methyl groups are used in the methylation cycle that produces methionine and S-adenosylmethionine, the universal methyl donor of epigenomic mechanisms [11]. Methionine synthase catalyses the remethylation of homocysteine in methionine, in presence of 5-methyltetrahydrofolate, as co-substrate and vitamin B12 (cobalamin) as co-factor [11]. Vitamin B12 and folate are critical micronutrients associated with foetal programming of obesity. Studies from India and UK have shown that vitamin B12 and folate status during pregnancy is associated with increased obesity and insulin resistance of the mother and birth weight and risk of insulin resistance and obesity in the offspring [12e15]. Furthermore, some data from the general US population have also suggested that high folate intake is associated with aggravation of the clinical manifestations of vitamin B12 deficiency [16]. Taken together, those data emphasize that vitamin B12 and folate should be considered together to evaluate their respective and combined association with metabolic syndrome components, including insulin resistance in obese patients. However, these associations have never been investigated in patients with severe obesity treated by bariatric surgery. In this study, we aim to evaluate the role of vitamin B12, folate and their related markers, homocysteine and methylmalonic acid on the components of the metabolic syndrome in severely obese patients assessed in two periods, the preoperative multidisciplinary evaluation for bariatric surgery (period 1) and in the hours preceding surgery (period 2).
Metabolic syndrome was defined by using the criteria proposed by the National Cholesterol Education Program Adult Treatment Panel III (NCEP_ATP III) and the international diabetes federation (IDF 2005 and IDF 2009) [2,3]. Waist circumference was evaluated during preoperative multidisciplinary evaluation (period 1) and was considered as a metabolic syndrome component when 94 cm in male and 80 cm in female. Hypertension was defined as SBP 130 mm Hg or DBP 85 mm Hg or the need for antihypertensive medication(s). Dyslipidemia was defined as plasma triglycerides >1.5 g/L or hypo HDL-cholesterol <0.4 (male) or 0.5 (female) g/L. Type 2 diabetes was defined as fasting glycemia 1.0 g/L or the need for treatment. Overall, metabolic syndrome severity was evaluated by counting the presence or absence of those 5 metabolic syndrome components (waist size, hypertension, hypertriglyceridemia, low HDL-cholesterol, diabetes) in all patients during period 1. We also evaluated other criteria corresponding to obesity-related comorbidities during preoperative multidisciplinary evaluation as coronary heart disease, obstructive sleep apnea syndrome, and non-alcoholic steatohepatitis. Coronary heart disease (CHD) was defined as a group of diseases including stable angina, unstable angina, and myocardial infarction. Obstructive sleep apnea syndrome (OSAS) was defined as a sleep disorder involving cessation or significant decrease in airflow in the presence of breathing effort. Non-alcoholic steatohepatitis (NASH) was defined as a type of fatty liver which occurs when fat is deposited (steatosis) in the liver due to causes other than excessive alcohol use.
2. Materials/subjects and methods 2.1. Study population All patients had a BMI 35 kg/m2 and were included in this study during preoperative multidisciplinary evaluation for bariatric surgery. All patients signed specific informed consents. Investigations conformed to the principles outlined in the Declaration of Helsinki and received approval from the local ethical committee. This study was registered at French ministry of health as NCT02663388. Clinical data and biological data were collected during preoperative multidisciplinary evaluation for bariatric surgery (period 1). Then, all patients had a fasting venous blood sample collection in the morning of their bariatric surgery (period 2). Those tubes were centrifuged the same morning and corresponding plasma was collected and stored at 20 C until analysis. Preoperative vitamin B12 deficiency was defined as B12 plasma concentration <148 pmol/L [16]. High preoperative plasma folate concentration was defined as folate plasma concentration >20 nmol/mL [16].
2.3. Biochemical analyses Fasting plasma concentrations of glucose, cholesterol, triglycerides, HDL-cholesterol, ASAT and ALAT were determined by the enzymatic or dry chemistry methods used on automata U 2700 Olympus Corporation, USA. LDL-cholesterol values were calculated using the Friedewald formula: [LDL-chol] ¼ [Total chol] [HDLchol] ([TG]/2.2) using the unit in mmol/L. Plasma insulin were determined by the kit MP Biomedicale (Solon, Ohio 44139, USA) for radioimmunoassay of insulin in serum or EDTA. The HOMA-IR was calculated using the model proposed by Levy et al. and corresponded to HOMA-IR ¼ Glucose Insulin/22.5 using the unit of glucose in mmol/L and the unit of insulin in mUI/ml [17]. Vitamin B12 and folate serum concentrations were determined by the SNB SimulTRAC-box for Radio Immunoassay Vitamin B12 [57Co]/Folate [125]. Red blood cell folate (RBC folate) was determined using the same radio immunoassay method after total blood extract hemolysis in presence of ascorbic acid. Homocysteine and methylmalonic acid (MMA) were determined by UPLC-MS/MS, by using a ACQUITY UPLC BEH C18 column (1.7 mm, 2.1 mm 50 mm, Waters Corporation) [18]. 2.4. Statistical analysis Statistical analyze were performed by SPSS version 23 (IBM corporation, Armonk, NY, USA). The minimal size of our sample was estimated at 250 patients, with a study power 1-b ¼ 0.8 and a ¼ 0.05, assuming a 2.0-fold difference in the folate or B12 concentrations between patients with and without outcomes of metabolic syndrome. Continuous data were reported as mean ± standard deviation. Incidence of nominal variables was reported as percentage. Non-parametric tests were used when data distribution was not normal. KruskaleWallis test was used between groups when there were 3; p-values < 0.05 were considered as significant. Continuous variables that had asymmetrical
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 3/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
distribution were classified in tertiles to study their combined association with HOMA-IR. Multivariate analysis was performed to evaluated independent risk factors for insulin resistance (HOMAIR), type 2 diabetes (fasting glucose > 1.0 g/L or treatment), and metabolic syndrome components (dependent variables) using multiple regression and logistic regression when needed. Corrections for multiple testing were made using the Bonferroni test. 3. Results 3.1. Main clinical and biological characteristics of patients Two hundreds and seventy eight patients were included in this study. The patients were recruited at the Nancy University Hospital Centre during a multidisciplinary evaluation of pathological severe obesity (period 1). They were subsequently evaluated 14.8 months later (range: 6e53 months) in the hours preceding bariatric surgery (period 2). The main clinical and biological characteristics of both periods are reported in Tables 1 and 2. Mean age was 45 ± 11.2 years and 221 patients were female (79.4%). The patients had a mean BMI at 48 þ 7.5 kg/m2 in period 1. We observed that 15 patients had a deficit in vitamin B12 defined by a serum concentration <148 pmol/ L. The concentration of vitamin B12 was not influenced by the treatment with metformin (B12 concentration of patients treated with and without metformin: 252 þ 112 pmol/L vs. 248 þ 98, p ¼ 0.821), in contrast to other studies for which type 2 diabetes was the main recruitment criteria [19,20]. None of the patients had a deficit in folate according to RBC folate. At period 2, we observed that the mean BMI decreased significantly from 48 ± 7.5 to 46 ± 6.6 kg/m2 (p < 0.001) corresponding to a mean loss of weight of 6.1 ± 10 kg. This weight loss resulted from the improved dietary management of the patients between the two evaluations. We reported 37 patients (13.3%) with low serum concentration of vitamin B12 (<148 pmol/L) and 27 patients (9.7%) with folate serum concentration above 20 nmol/mL. The latter were considered to have high folic acid intake. As expected, folate serum concentration was negatively correlated with plasma homocysteine (0.272, 0.377 to 0.159; p < 0.0001). The patients with low serum vitamin B12 had a higher mean plasma concentration of homocysteine (15.6 vs. 13.4 mmol/L; p ¼ 0.039) and MMA (0.178 vs. 0.154 mmol/L; p ¼ 0.026), in comparison with the other patients. As previously reported in other population studies, homocysteine was closely correlated with serum folate (p < 0.001) (Fig. 1A) and the highest concentrations of MMA in patients with low serum vitamin B12
3
Table 2 Metabolic syndrome components and other characteristics in 278 patients accordQ5 ing to IDF definition (2). Comorbidity Metabolic syndrome components Waist size, 94 cm in male and 80 cm in female Hypertension, SBP 130 mm Hg or DBP 85 mm Hg or antihypertensive medication Dyslipidemia - Triglycerides > 1,5 g/L - Hypo HDL-cholesterol < 0.4 (M) or 0.5 (F) g/L - Both - Fasting glycemia 1,0 g/L without treatment - Treated diabetes, fasting glycemia 1,0 g/L or treatment Other characteristics NASH OSAS CHD Male
Number of patients (n)
Percentage of patients (%)
278
100
118
42.4
235 28 108 91 53 84
84.5 10.1 38.9 32.7 19.1 30.2
26 139 18 58
9.4 50.0 6.5 20.8
CHD: coronary heart disease; NASH: non-alcoholic steatohepatitis; OSAS: obstructive sleep apnea syndrome.
were reported in cases with either low or high folate serum concentration, compared to those with intermediate folate concentration (Fig. 1B). 3.2. Predictors of insulin resistance Only the 194 patients who had no known type 2 diabetes and no oral anti-diabetic treatment were assessed for HOMA-IR in period 1 while all patients were assessed in period 2. We observed a significant association of HOMA-IR with RBC folate concentration (p ¼ 0.008) and no significant association with serum vitamin B12, in period 1. In multivariate analysis, HDL-cholesterol and RBC folate remained independent predictors for HOMA-IR (p ¼ 0.034 and p ¼ 0.008, respectively, Table 3). Patients in the lowest tertile of plasma B12 (<238 pmol/L) and highest tertile of RBC (>948 nmol/L) had the highest HOMA-IR (p < 0.05) (Fig. 2A). We confirmed the association of HOMA-IR with low B12/high folate in period 2. Patients with the lowest serum concentration of B12 (tertile < 200 pmol/L) associated with the highest values for plasma folate (tertile > 10.2 nmol/L) were those with the highest HOMA-IR values (Fig. 2B). In addition, patients with the highest concentration
Table 1 Clinical and biological characteristics in the two periods of the multidisciplinary evaluation of patients. Parameters
During multidisciplinary evaluation e period 1 n
Age, years BMI, kg/m2 Fasting insulin, mUI/mL Fasting glucose, g/L HOMA-IR HDL-cholesterol, g/L LDL-cholesterol, g/L Total cholesterol, g/L Triglycerides, g/L RBC folate, nmol/L Plasma folate, nmol/L Plasma B12, pmol/L Homocysteine, mmol/L MMA, mmol/L
278 278 186 192 183 248 248 250 250 220 201 226 ND ND
At bariatric surgery e period 2
Mean ± Standard Deviation
Median
Lower tertile
Upper tertile
n
Mean ± Standard Deviation
Median
Lower tertile
Upper tertile
p-value
46 ± 7.5 24.7 ± 19 0.99 ± 0.18 6.05 ± 4.3 0.45 ± 0.1 1.23 ± 0.33 2.02 ± 0.36 1.74 ± 0.89 936.5 ± 464 13.9 ± 8.3 298 ± 145 ND ND
44.5 20.6 0.96 5.04 0.43 1.21 2.00 1.56 797.5 11.4 274 ND ND
42.2 17.5 0.91 4.15 0.40 1.08 1.86 1.20 681.8 9.6 238 ND ND
47.3 26.5 1.04 6.48 0.48 1.33 2.11 1.90 940.7 14.6 317 ND ND
278 209 278 277 277 277 277 277 277 ND 278 278 278 278
45 ± 11 46.6 ± 6.6 54.7 ± 50 1.06 ± 0.39 16.4 ± 27 0.41 ± 0.1 1.15 ± 0.32 1.87 ± 0.35 1.56 ± 0.69 ND 11.9 ± 7.9 265 ± 294 13.7 ± 4.8 0.16 ± 0.07
45 45.4 43.5 0.94 10.4 0.41 1.13 1.86 1.42 ND 9.6 239 13.0 0.14
40 42.3 33.4 0.86 7.5 0.37 1.01 1.72 1.17 ND 8.1 200 11.4 0.12
51 48.4 54.2 1.05 13.9 0.44 1.28 2.01 1.65 ND 12.7 281 14.4 0.17
NA 0.2537 <0.0001 0.3479 <0.0001 0.0001 0.0061 <0.0001 0.0326 NA 0.0027 <0.0001 NA NA
MMA: methylmalonic acid; NA: no applicable; ND: no data; RBC: red blood cell.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 4/7
4
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
3.3. Predictors of metabolic syndrome severity We evaluated further the predictors of metabolic syndrome severity by considering their association with the number of components, including hypertension, hypertriglyceridemia (>1.5 g:L), low HDL-cholesterol (<0.5 in females and <0.4 in male) and type 2 diabetes defined according to the International Diabetes Federation [2]. The presence of 4, 3 and 2 components was reported in 69, 95 and 95 cases, respectively. Hypertension was the less frequent outcome in patients with 2 components (8 cases) and hypertension the less frequent in patients with 3 components (41 cases). Age, RBC folate, ALAT, ASAT and MMA were the 5 significant predictors of the number of metabolic syndrome components in univariate analysis (Table 4). In multivariate analysis, age and RBC folate remained independently associated with the number of metabolic syndrome components (Table 4, p ¼ 0.010 and 0.020, respectively).
4. Discussion
Fig. 1. A: Correlation (Pearson analysis) of logarithmically transformed homocysteine (Log-HCY) with logarithmically transformed folate. Dotted lines: 95% confidence intervals of the slope. B: Mean MMA concentration was the highest in patients with low vitamin B12 plasma concentration with either the lowest or highest folate plasma concentrations, compared to those with intermediate folate concentration at period 2 (n ¼ 278). Low vitamin B12 plasma concentration (vitamin B12 deficiency) was defined as inferior to 150 pmol/L. MMA values were transformed into logarithms. P for trend analysis was used within plasma vitamin B12 categories.
of MMA had the highest HOMA-IR values (Fig. 2C). Age, BMI, triglycerides, HDL-cholesterol, serum folate, homocysteine and MMA were significantly associated with HOMA-IR in univariate analyses at period 2 (Table 3). Log-transformed HOMA-IR score was negatively correlated with Log-transformed homocysteine (p < 0.0001) (Fig. 3A) and positively correlated with Log-transformed serum folate (p < 0.001) (Fig. 3B) and Log-transformed MMA (Fig. 3C). We performed therefore the multivariate analysis in two models where folate and Hcy were alternatively excluded. BMI and homocysteine (model 1 without serum folate, p ¼ 0.010 and p ¼ 0.002, respectively) or BMI and methylmalonic acid (model 2 without homocysteine, p ¼ 0.030 and p ¼ 0.004, respectively) remained independent predictors for HOMA-IR values (Table 3).
Metabolic syndrome is frequently observed in patients with morbid obesity and is associated with a 5-fold risk of type 2 diabetes and a 2-fold risk of cardiovascular outcomes [21]. The incidence and distribution of the components of metabolic syndrome observed in our study are consistent with these previous data. We identified HDL-cholesterol and RBC folate as independent predictors for HOMA-IR at period 1 while independent predictors for HOMA-IR values at period 2 were either BMI and homocysteine (model 1 without serum folate) or BMI and methylmalonic acid (model 2 without homocysteine). We observed also an influence of markers of B12 deficit, either lowest B12 or highest MMA concentrations in the association between high folate status and HOMA-IR. Age and RBC folate remained independent predictors for the number of metabolic syndrome components. The deficit in vitamin B12 and/or folate leads to the disruption of the one-carbon cycle and its downstream metabolic processes [22]. Our study showed that high folate status was associated with insulin resistance and the number of components of metabolic syndrome in severely obese patients. The potential harmful effects of abnormal folic acid concentration have been suggested in adults and pediatrics population [23,24]. Lowering homocysteine concentration by daily supplementation of vitamins B6, B9 and B12 is not associated with a significant reduction of developing type 2 diabetes, in women with high risk for cardiovascular diseases [25]. High blood folate concentration has been linked to aggravation of anemia and cognitive impairment and increased MMA and
Table 3 Determinants of HOMA-IR in univariate and multivariate analysis at two periods. Parameters
Period 1 Univariate
Age BMI Triglycerides HDL-cholesterol LDL-cholesterol Creatinine RBC folate Serum folate B12 Homocysteine MMA
Q6 Period 2
n
Rho
p-value
0.1054 0.1660 0.2338 0.2213 0.1555 0.0244 0.2135 ND 0.0695 ND ND
0.1874 0.0305 0.0023 0.0041 0.0448 0.7555 0.0085 ND 0.3917 ND ND
Multivariate Coefficient
158 170 168 167 167 165 151 ND 154 ND ND
Univariate p-value
5.6310
0.034
0.0014
0.008
n
Rho
p-value
0.1368 0.1366 0.2439 0.3128 0.2349 0.0697 ND 0.2167 0.0520 0.2286 0.1524
0.0228 0.0491 <0.0001 <0.0001 0.0001 0.2490 ND 0.0003 0.3886 0.0001 0.0111
277 208 277 277 277 275 ND 277 277 277 277
Multivariate Model 1
Multivariate Model 2
Coefficient
p-value
Coefficient
p-value
0.7661
0.0050
0.6816
0.030
1.3852
0.002 83.7915
0.004
MMA: methylmalonic acid; ND: no data; RBC: red blood cell. Corrections for multiple testing were made using the Bonferroni test.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 5/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
5
Fig. 2. A: Patients with the lowest values for plasma B12 (tertile < 241 pmol/L) associated with the highest values for RBC folate (tertile > 953 nmol/L) had the highest HOMA-IR values at period 1 (n ¼ 183). B: Patients at period 2 (n ¼ 277) with the lowest values for plasma B12 (tertile < 200 pmol/L) associated with the highest values for plasma folate (tertile > 10.2 nmol/L) were those with the highest HOMA-IR values. HOMA-IR values in figures A and B were transformed into logarithms. P for trend analysis was used within each plasma vitamin B12 categories (*: p < 0.05). C: Patients with the highest folate and MMA values (tertiles) had the highest mean HOMA-IR values at period 2 (n ¼ 277). HOMA-IR values were transformed into logarithms. P for trend analysis was used within MMA categories.
Fig. 3. A, B, C: Correlation (Pearson analysis) of logarithmically transformed HOMA-IR with logarithmically transformed homocysteine (A, Lg Hcy), folate (B, Lg Folate) and methylmalonic acid (C, Lg MMA).
homocysteine, in patients with low vitamin B12 plasma concentration [16,26]. Consistently, we confirmed the highest MMA concentration in patients with the lowest vitamin B12/highest folate concentrations [26] and a negative correlation between homocysteine and HOMA-IR. Furthermore, transcobalamin 776C/G polymorphism has been shown to be associated with peripheral neuropathy in elderly individuals with high folate intake [27]. In contrast, hyperhomocysteine has been associated with obesity, hypertension and type 2 diabetes mellitus [28]. It correlates also with insulin resistance and low-grade systemic inflammation in obese prepubertal children [29]. Reduced folate concentration is associated with epigenomic changes in liver tissue from subjects with type 2 diabetes [30]. The discrepancy between these and our data may be explained by differences in the severity of obesity and age of patients among studied populations.
We found that the patients with either high folate/low B12 or high folate/high MMA were those with the highest risk of insulin resistance. This was consistent with the negative correlation of HOMA-IR with homocysteine and its positive correlation with folate and MMA. In addition, as much as 10% of our patients had a folate serum concentration above 20 nmol/mL that could reflect a high folate intake. The synergic influence of the high intake of folate and B12 deficit on the risk of insulin resistance is consistent with previous data. An indian study showed that normal to high maternal folate status associated with a low maternal vitamin B12 status led to babies having a higher adiposity and insulin resistance [31]. In the same study, over 60% of pregnant women were B12 deficient. Obese offsprings from mothers with lowest values for plasma B12 and highest values of folate were those with the more severe insulin resistance. Low plasma vitamin B12 during
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65
YCLNU3193_proof ■ 26 July 2017 ■ 6/7
6
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
Table 4 Predictors of the number of metabolic syndrome components. Criteria
Age, years Male gender, % BMI, kg/m2 OSAS comorbidity, % NASH comorbidity, % CHD comorbidity, % RBC folate, nmol/L Plasma folate, nmol/L Vitamin B12, pmol/L Homocysteine, mmol/L MMA, mmol/L
Number of metabolic syndrome components 1 (n ¼ 21)
2 (n ¼ 93)
3 (n ¼ 95)
4 (n ¼ 69)
40.6 ± 11 5 (24) 47.8 ± 6 7 (33) 0 (0) 0 (0) 835 ± 347 9.9 ± 5 214 ± 72 14.6 ± 4 0.157 ± 0.04
40.5 ± 11 11 (12) 45.1 ± 5 34 (36) 5 (5.4) 1 (1.07) 821 ± 368 10.7 ± 6 239 ± 91 13.0 ± 3 0.147 ± 0.06
46.6 ± 10 20 (21) 45.5 ± 6 56 (59) 12 (12.6) 8 (8.4) 950 ± 412 13.1 ± 11 320 ± 483 13.6 ± 6 0.153 ± 0.08
50.3 ± 10 22 (32) 47.1 ± 8 42 (61) 9 (13) 9 (13) 1116 ± 620 12.6 ± 7 243 ± 101 14.4 ± 5 0.175 ± 0.09
p-for trend
OR
95% C.I.
p-value
<0.001 0.020 0.339 0.001 0.101 0.009 0.016 0.082 0.073 0.270 0.036
1.04
1.01e1.07
0.010
1.02
1.01e3.03
0.020
CHD: coronary heart disease; NASH: non-alcoholic steatohepatitis; OSAS: obstructive sleep apnea syndrome. Corrections for multiple testing were made using the Bonferroni test.
pregnancy is also considered to be associated with an increased risk of gestational and later diabetes leading to contribute to the increased risk in offsprings [32,33]. Folic acid fortification has been introduced in north America two decades ago and its balancing between risks and benefits is under discussion in Europe [31]. Folic acid fortification is clearly associated with increased folate status in general population [16,32]. For example, the introduction of folate fortification in Canada has dramatically increased the maternal RBC folate concentration and cord RBC folate. This highlights the need for investigating long-term health outcomes of high RBC folate in offsprings [33]. There are potential reasons explaining why vitamin B12 deficiency and MMA could influence the risk of insulin resistance in the presence of normal or high folate. B12 deficiency produces a cellular folate trap that lead to impaired methionine synthesis. It produces also an impaired fatty acid oxidation, increased ER stress and impaired expression of SIRT1, a key actor of molecular mechanisms of pathological obesity, in rats [34]. Impaired conversion of MMA to succinyl coA, for which vitamin B12 acts as a co-enzyme, is associated with the accumulation of MMA leading to increased lipogenesis and insulin resistance. Despite folate is not considered to be directly involve in these pathways, higher folate is associated with increased concentration of MMA in the context of vitamin B12 deficiency, as reported in our and previous studies [16,35]. Lastly, vitamin B12 deficiency can increase plasma and RBC folate concentrations [36]. We found an association of age, RBC folate and APRI with the number of components of metabolic syndrome (hypertension, hypertriglyceridemia, low HDL-cholesterol and type 2 diabetes), in multivariate analysis. Previous studies have found an association of nutritional, metabolic and genetic determinants of the one carbon metabolism with some of the components of metabolic syndrome, but none considered the association of folate, B12, homocysteine and MMA with all components together, in severely obese patients. We showed in the OASI population of elderly ambulatory volunteers that plasma total homocysteine correlated negatively with HDL-Cholesterol [37]. In contrast to our results on the association between high RBC values of folate and increased HOMA and dyslipidemia, a double-blind placebo-controlled clinical trial concluded that 5 mg folate supplementation resulted in reduced HOMA-IR and HDL-cholesterol in women with polycystic ovary syndrome [38]. These discrepancies may be explained by the difference between the two populations, since polycystic ovary syndrome is a very specific cause of insulin resistance. Other studies have found a benefit of supplementation with other B vitamins and methyl donors on outcomes of metabolic syndrome. Horigan et al. demonstrated a significant blood pressure-lowering response to intervention with riboflavin for 16 weeks in hypertensive patients [39]. In the HUSK cohort, choline and betaine were associated in
opposite directions with serum triglycerides, glucose, BMI, percent body fat, waist circumference and systolic and diastolic blood pressure [40]. In conclusion, we found that homocysteine, high RBC folate and methylmalonic acid are significant predictors of the insulin resistance and number of metabolic syndrome components in severely obese patients. The highest risk of insulin resistance was observed in patients with highest folate status and either lowest vitamin B12 or highest MMA. These findings should deserve further interest for replicating our study and studying the underlying molecular mechanisms. They may have important implications for the management of severely obese patients and for evaluating the risk/ benefit balance of folate food fortification in this population. Sources of support te This project is supported by Inserm, SOFFCO-MM (SOcie site et des Maladies Française et Francophone de Chirurgie de l'Obe taboliques e 2014 scholarship) and the AGIR grant from the Me University regional hospital of Nancy and The region of Lorraine. ZL was recipient of a graduate student fellowship (#201308070048) from the China Scholarship Council (CSC). Author contributions ZL conducted research, analyzed data, performed statistical analysis, wrote paper and had primary responsibility for final content. LPB, RMGR and JLG designed research, conducted research, analyzed data, performed statistical analysis, wrote paper and had primary responsibility for final content; DQ and MAS conducted research and provided essential materials; DM analyzed databases and revised manuscript. Conflict of interest The authors declare no conflict of interest. References [1] Eckel RH, Grundy SM, Zimmet PZ. The metabolic syndrome. Lancet 2005;365: 1415e28. [2] Alberti KG, Eckel RH, Grundy SM, Zimmet PZ, Cleeman JI, Donato KA, et al. Harmonizing the metabolic syndrome: a joint interim statement of the International Diabetes Federation Task Force on Epidemiology and Prevention; National Heart, Lung, and Blood Institute; American Heart Association; World Heart Federation; International Atherosclerosis Society; and International Association for the Study of Obesity. Circulation 2009;120:1640e5. [3] Genser L, Casella Mariolo JR, Castagneto-Gissey L, Panagiotopoulos S, Rubino F. Obesity, type 2 diabetes, and the metabolic syndrome: pathophysiologic relationships and guidelines for surgical intervention. Surg Clin N Am 2016;96: 681e701.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
Q3
66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
YCLNU3193_proof ■ 26 July 2017 ■ 7/7
Z. Li et al. / Clinical Nutrition xxx (2017) 1e7
Q4
[4] Malik VS, Willett WC, Hu FB. Global obesity: trends, risk factors and policy implications. Nat Rev Endocrinol 2013;9:13e27. [5] Stern MP, Williams K, Gonzalez-Villalpando C, Hunt KJ, Haffner SM. Does the metabolic syndrome improve identification of individuals at risk of type 2 diabetes and/or cardiovascular disease? Diabetes Care 2004;27:2676e81. [6] Reddon H, Gueant JL, Meyre D. The importance of gene-environment interactions in human obesity. Clin Sci Lond 2016;130:1571e97. [7] Gallistl S, Sudi K, Mangge H, Erwa W, Borkenstein M. Insulin is an independent correlate of plasma homocysteine levels in obese children and adolescents. Diabetes Care 2000;23:1348e52. [8] Kahleova R, Palyzova D, Zvara K, Zvarova J, Hrach K, Novakova I, et al. Essential hypertension in adolescents: association with insulin resistance and with metabolism of homocysteine and vitamins. Am J Hypertens 2002;15:857e64. [9] Sreckovic B, Sreckovic VD, Soldatovic I, Colak E, Sumarac-Dumanovic M, Janeski H, et al. Homocysteine is a marker for metabolic syndrome and atherosclerosis. Diabetes Metab Syndr 2016. S1871-4021:30200e30204. [10] Catena C, Colussi G, Sechi LA. Response to “plasma Homocysteine levels and endothelial dysfunction in cerebro- and cardiovascular diseases in the metabolic syndrome”. Am J Hypertens 2015;28:1490. [11] Gueant JL, Elakoum R, Ziegler O, Coelho D, Feigerlova E, Daval JL, et al. Nutritional models of foetal programming and nutrigenomic and epigenomic dysregulations of fatty acid metabolism in the liver and heart. Pflugers Arch 2014;466:833e50. [12] Knight BA, Shields BM, Brook A, Hill A, Bhat DS, Hattersley AT, et al. Lower circulating B12 is associated with Higher obesity and insulin resistance during pregnancy in a non-diabetic white British population. PLoS One 2015;10, e0135268. [13] Yajnik CS, Chandak GR, Joglekar C, Katre P, Bhat DS, Singh SN, et al. Maternal homocysteine in pregnancy and offspring birthweight: epidemiological associations and Mendelian randomization analysis. Int J Epidemiol 2014;43: 1487e97. [14] Yajnik CS, Deshmukh US. Maternal nutrition, intrauterine programming and consequential risks in the offspring. Rev Endocr Metab Disord 2008;9: 203e11. [15] Stewart CP, Christian P, LeClerq SC, West Jr KP, Khatry SK. Antenatal supplementation with folic acid þ iron þ zinc improves linear growth and reduces peripheral adiposity in school-age children in rural Nepal. Am J Clin Nutr 2009;90:132e40. [16] Selhub J, Rosenberg IH. Excessive folic acid intake and relation to adverse health outcome. Biochimie 2016;126:71e8. [17] Levy JC, Matthews DR, Hermans MP. Correct homeostasis model assessment (HOMA) evaluation uses the computer program. Diabetes Care 1998;21: 2191e2. [18] Amouzou EK, Chabi NW, Adjalla CE, Rodriguez-Gueant RM, Feillet F, Villaume C, et al. High prevalence of hyperhomocysteinemia related to folate deficiency and the 677C–>T mutation of the gene encoding methylenetetrahydrofolate reductase in coastal West Africa. Am J Clin Nutr 2004;79: 619e24. [19] de Jager J, Kooy A, Lehert P, Wulffele MG, van der Kolk J, Bets D, et al. Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial. BMJ 2010;340, c2181. [20] Aroda VR, Edelstein SL, Goldberg RB, Knowler WC, Marcovina SM, Orchard TJ, et al. Long-term metformin use and vitamin B12 deficiency in the diabetes prevention program outcomes study. J Clin Endocrinol Metabol 2016;101: 1754e61. [21] Grundy SM. Metabolic syndrome: a multiplex cardiovascular risk factor. J Clin Endocrinol Metabol 2007;92:399e404. [22] Finer S, Saravanan P, Hitman G, Yajnik C. The role of the one-carbon cycle in the developmental origins of Type 2 diabetes and obesity. Diabet Med 2014;31:263e72.
7
[23] Sudchada P, Saokaew S, Sridetch S, Incampa S, Jaiyen S, Khaithong W. Effect of folic acid supplementation on plasma total homocysteine levels and glycemic control in patients with type 2 diabetes: a systematic review and metaanalysis. Diabetes Res Clin Pract 2012;98:151e8. [24] Dehkordi EH, Sedehi M, Shahraki ZG, Najafi R. Effect of folic acid on homocysteine and insulin resistance of overweight and obese children and adolescents. Adv Biomed Res 2016;5:88. [25] Song Y, Cook NR, Albert CM, Van Denburgh M, Manson JE. Effect of homocysteine-lowering treatment with folic Acid and B vitamins on risk of type 2 diabetes in women: a randomized, controlled trial. Diabetes 2009;58: 1921e8. [26] Selhub J, Morris MS, Jacques PF. In vitamin B12 deficiency, higher serum folate is associated with increased total homocysteine and methylmalonic acid concentrations. Proc Natl Acad Sci U S A 2007;104:19995e20000. [27] Sawaengsri H, Bergethon PR, Qiu WQ, Scott TM, Jacques PF, Selhub J, et al. Transcobalamin 776C–>G polymorphism is associated with peripheral neuropathy in elderly individuals with high folate intake. Am J Clin Nutr 2016;104:1665e70. [28] Fonseca VA, Fink LM, Kern PA. Insulin sensitivity and plasma homocysteine concentrations in non-diabetic obese and normal weight subjects. Atherosclerosis 2003;167:105e9. [29] Martos R, Valle M, Morales R, Canete R, Gavilan MI, Sanchez-Margalet V. Hyperhomocysteinemia correlates with insulin resistance and low-grade systemic inflammation in obese prepubertal children. Metabolism 2006;55: 72e7. [30] Nilsson E, Matte A, Perfilyev A, de Mello VD, Kakela P, Pihlajamaki J, et al. Epigenetic alterations in Human liver from subjects with type 2 diabetes in parallel with reduced folate levels. J Clin Endocrinol Metabol 2015;100: E1491e501. [31] Smith AD, Refsum H, Selhub J, Rosenberg IH. Decision on folic acid fortification in Europe must consider both risks and benefits. BMJ 2016;352, i734. [32] Smith AD, Kim YI, Refsum H. Is folic acid good for everyone? Am J Clin Nutr 2008;87:517e33. [33] Plumptre L, Masih SP, Ly A, Aufreiter S, Sohn KJ, Croxford R, et al. High concentrations of folate and unmetabolized folic acid in a cohort of pregnant Canadian women and umbilical cord blood. Am J Clin Nutr 2015;102:848e57. [34] Gueant JL, Namour F, Gueant-Rodriguez RM, Daval JL. Folate and fetal programming: a play in epigenomics? Trends Endocrinol Metabol: TEM 2013;24: 279e89. [35] Krishnaveni GV, Hill JC, Veena SR, Bhat DS, Wills AK, Karat CL, et al. Low plasma vitamin B12 in pregnancy is associated with gestational 'diabesity' and later diabetes. Diabetologia 2009;52:2350e8. [36] Krishnaveni GV, Hill JC, Leary SD, Veena SR, Saperia J, Saroja A, et al. Anthropometry, glucose tolerance, and insulin concentrations in Indian children: relationships to maternal glucose and insulin concentrations during pregnancy. Diabetes Care 2005;28:2919e25. [37] Gueant-Rodriguez RM, Spada R, Moreno-Garcia M, Anello G, Bosco P, Lagrost L, et al. Homocysteine is a determinant of ApoA-I and both are associated with ankle brachial index, in an ambulatory elderly population. Atherosclerosis 2011;214:480e5. [38] Asemi Z, Karamali M, Esmaillzadeh A. Metabolic response to folate supplementation in overweight women with polycystic ovary syndrome: a randomized double-blind placebo-controlled clinical trial. Mol Nutr Food Res 2014;58:1465e73. [39] Horigan G, McNulty H, Ward M, Strain JJ, Purvis J, Scott JM. Riboflavin lowers blood pressure in cardiovascular disease patients homozygous for the 677C/T polymorphism in MTHFR. J Hypertens 2010;28:478e86. [40] Konstantinova SV, Tell GS, Vollset SE, Nygard O, Bleie O, Ueland PM. Divergent associations of plasma choline and betaine with components of metabolic syndrome in middle age and elderly men and women. J Nutr 2008;138: 914e20.
Please cite this article in press as: Li Z, et al., Folate and vitamin B12 status is associated with insulin resistance and metabolic syndrome in morbid obesity, Clinical Nutrition (2017), http://dx.doi.org/10.1016/j.clnu.2017.07.008
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100