Abstract
Research suggests that high intake of supplemental vitamin B12 may be associated with increased risk of cancer, with some evidence that this association may vary by gender and smoking status. This investigation evaluates if similar patterns in association are observed for data for 11,757 adults from the National Health and Nutrition Examination Survey (1999–2006). Survey-weighted multivariable-adjusted linear regression was used to evaluate the association between regular B12 supplement use and log-transformed serum B12 levels. Persons taking vitamin B12 through a multivitamin/multimineral (MVMM) had a median supplemental intake of 12 mcg/day (Q1: 6, Q3: 25), compared to 100 mcg/day (Q1: 22, Q3: 500) for persons reporting supplemental B12 intake through a MVMM-exclusive source. MVMM users had a geometric mean serum B12 26% (95% CI: 23%−30%) higher than non-users, whereas MVMM-exclusive users’ geometric mean was 61% (95% CI: 53%−70%) higher than non-users (p-trend<0.001). Although a positive trend (p-trend<0.001) was observed for both men and women, the association was stronger among women (p-interaction<0.001). No interaction was observed for smoking status (p-interaction=0.45). B12 supplementation is associated with higher levels of serum B12, with significant interaction by gender but not smoking. Further work is needed to better understand the interplay of B12 and gender.
Keywords: vitamin B12, supplements, smoking, gender, effect modification
Introduction
Vitamin B12, also known as cobalamin, is a water-soluble vitamin that plays a key role in various bodily functions,1 and whose deficiency has been associated with conditions ranging from anemia to memory loss.2 Vitamin B12 is consumed through both dietary sources (poultry, meat, fish, dairy, and fortified cereals) and use of dietary supplements.1 In the United States, supplemental vitamin B12 use declined from 1999 to 2012, driven by a decrease in use of multi-vitamin/multi-minerals (MVMM), which contain several vitamins/minerals including B12 – but often at lower dosages than individual supplements.3 However, use of vitamin B12 supplements exclusive of MVMM, henceforth referred to as MVMM-exclusive supplements, has increased, with 8.1% of adults reporting use in 2011–2012.3 These trends raise further questions about factors that may affect B12 absorption/metabolism and subsequent serum concentrations. To this end, smokers consistently present with lower serum B12 levels than never-smokers 4–7, and previous cross-sectional studies have found lower serum B12 concentrations and higher prevalence of B12 deficiency among men 8–11.
Additional research suggests that higher intake of B12—usually only possible through supplements— may increase cancer risk.12–16 B vitamins (including vitamin B2, B6, B9 (henceforth referred to as folate), and B12) affect the one-carbon metabolism pathway, which affects DNA methylation.17 Methionine synthase utilizes B12 to transform homocysteine into the essential amino acid methionine, which is later used to create S-adenosylmethionine (SAM). SAM is the universal methyl donor and can deposit methyl groups onto DNA,18 which affects gene expression. Methylation could allow for the expression of protooncogenes or the silencing of tumor suppressor genes, allowing for tumorigenesis.19 As B12 is instrumental in the production of these intermediates, more research is necessary to understand how supplemental intake might impact bioavailable B12.
There has been some suggestion of a positive association between high vitamin B12 levels and cancer risk,15,16,20 and that associations between cancer and supplemental B12 use may vary by gender and smoking status.13,14,21 However, the extent to which these findings reflect differences in B12 intake versus differences in B12 absorption/metabolism by gender and smoking status remains unclear. This is further complicated by a lack of understanding on the extent that such findings apply to intakes of B12 from dietary supplements,22 especially as individuals can ingest substantially higher levels of B12 from supplements than from the diet. Typically, supplement doses range from 5–25 mcg/day23 compared the recommended dietary intake for healthy adults of 2.4 mcg/day.24
Our objective in this study was to address unanswered questions about the association between B12 supplement use (defined as non-use, MVMM use, MVMM-exclusive use) and serum B12 levels using nationally-representative data from the National Health and Nutrition Examination Survey (NHANES). We hypothesize that these associations may be modified by gender and smoking status.
Materials & Methods
Study population
This project was conducted using data from the National Health and Nutrition Examination Survey (NHANES), a nationally-representative study of the non-institutionalized, civilian US population that collects data through in-person interviews, medical examinations, and questionnaires.25
Data for this study include persons surveyed over four continuous two-year cycles (1999–2006) for which serum B12 levels were measured, with each cycle being conducted in an independent sample. Over the 1999–2006 cycles, 17,951 participants aged 20 or older participated in the dietary recall. We excluded participants from our sample if they had a prior history of cancer (exclusive of non-melanoma skin cancer; n = 1,193), and further excluded if they were pregnant at the time of the NHANES interview (n = 1,008), given concern that associations may be markedly different in this group. Exclusions were additionally made for participants missing exposure (n=26), outcome (n=808), or covariate data (n=3,159; among which 1,303 were missing data on tobacco history), leaving 11,757 participants available for study. As de-identified data are publicly available for download, the Memorial Sloan Kettering Institutional Review Board determined that this did not constitute human subjects research and thus did not require local oversight.
Exposures
While vitamins B2 (riboflavin), B6 (pyridoxine), B9 (folate), and B12 (cobalamin) all operate within the one-carbon metabolism, vitamin B12 was selected for this study due to a priori associations with lung cancer, and the suggestion that this association may vary by gender and/or smoking status.13,14 Vitamin B12 supplement use was defined as follows: no regular use, regular vitamin B12 use as part of a multivitamin/multimineral (MVMM), and regular use exclusive of a MVMM supplement. Here, regular use was defined by use at least 15 days per month in order to capture consistent health behaviors. MVMM was defined as a supplement that contains 10 or more vitamins or minerals3; thus, MVMM-exclusive forms of B12 supplementation include B-vitamin complexes, supplements containing solely B12, as well as sources of B12 typically consumed at higher doses than when consumed through an MVMM.
Vitamin B12 supplement use was the primary exposure of interest, but the research team wanted to ensure that results were not influenced greatly by the amount of supplemental or dietary intake. Thus, in sensitivity analyses, we alternatively examined associations pertaining to i) dose of vitamin B12 supplement use (mcg/day), ii) estimated dietary vitamin B12 intake (mcg/day), and iii) total intake from combined supplement and dietary sources (mcg/day). To obtain information on supplemental dose for sensitivity analyses, interviewers asked participants if they had used any supplements in the past 30 days, and if yes, they requested to see the supplement label for ingredient and dosage information. If the interviewer was unable to see the bottle, the participant was asked to provide information to the best of their ability, which trained nutritionists matched to the NCHS dietary supplement database that contains detailed information on supplement products and blends. NHANES documentation provides more details of this matching process and database usage.26 Supplement information was combined with self-reported information on use (e.g. frequency of use, pills/day) to determine total dose/day of supplemental B12 (mcg/day), which was then divided into four groups: non-use and tertiles of supplement dose. Dose of supplemental vitamin B12 ranged from 0.065 mcg/day to 18,000 mcg/day, and the cut-offs for supplement dose tertiles are as follows: 6.14 and 25 mcg/day.
For estimated dietary intake, we utilized NHANES dietary recall data, the methods of which have been detailed elsewhere.27 The NHANES survey collects dietary information from each participant through two 24-hour dietary recalls—one conducted on the day of the interview/medical examination, and for a subset of participants, another conducted over the phone 3 to 10 days later. NHANES estimates the amount of B12 in micrograms (mcg) in each food and calculates a total dietary intake for the recall period. To preserve the temporal sequence, only the first dietary recall was used to create a dietary B12 intake variable because the food would have been ingested before the blood sample was taken. The continuous dietary intake variable (mcg/day) was divided into quartiles, with cutoffs as follows: 2.04, 3.69, and 6.13 mcg/day.
A total supplemental & dietary intake variable was created using the continuous supplemental dosage data with continuous dietary intake data. After combining the dietary intake and supplement dosage, the continuous total intake variable (mcg/day) was divided into quartiles, with total supplement+diet quartile cutoffs as follows: 2.82, 6.16, 12.95 mcg/day.
Outcome
The outcome of interest was blood serum levels of vitamin B12 (pmol/L) measured using Bio-Rad Laboratories Quataphase II Folate/Vitamin B12 radioassay kits.28 The distribution of serum B12 was comparable across NHANES cycles. However, as the B12 blood serum levels were right skewed, a natural log transformation was used to normalize the distribution.
Statistical Analysis
Linear regression was used to model the association between vitamin B12 intake and serum vitamin B12. As the outcome was log-transformed, coefficients are presented in exponentiated form, representing the ratio of geometric mean of serum vitamin B12 for MVMM or MVMM-exclusive users as compared to the non-user reference group. Minimally-adjusted and fully-adjusted linear regression models were conducted for each exposure, including the primary exposure (vitamin B12 supplement use). Minimally-adjusted models adjust for both age and gender. Fully-adjusted models additionally include the following variables: race/ethnicity, education level, poverty-to-income ratio, body mass index (BMI), smoking status, number of pack-years, alcohol use, current health status, diabetes status, heart disease history, history of anemia treatment, and history of memory loss. These covariates were selected a priori based on their expected association with exposure and outcome. Details on covariate modeling are provided in tables and table footnotes. Sensitivity analyses were conducted, with exposure alternatively defined by dietary intake (mcg/day), supplemental dose (mcg/day), and total dietary+supplemental intake (mcg/day). Our research team felt these additional exposures were inherently linked to our primary exposure, but were less descriptive of long-term supplement behaviors as NHANES only collects dietary data over two days.
The multivariate model relating 3-level vitamin B12 supplement use and serum vitamin B12 was first stratified separately by gender and smoking status (current, former, or never smoker). For each, we examined linear trends within strata.
Given prior literature to suggest potential for joint interaction by gender and smoking status, we conducted a secondary analysis of interaction alternatively using six jointly-defined strata: male never smokers, male former smokers, male current smokers, female never smokers, female former smokers, and female current smokers. However, given the small number of MVMM-exclusive vitamin B12 users within each of the 6 strata, this analysis had to be alternatively conducted using the total diet+supplement B12 intake variable.
As NHANES oversamples people 60 and older, African Americans, Asians, and Hispanics, all analyses are weighted to account for oversampling and non-response. All analyses were conducted using StataSE 15 software.29 All P values were two-sided, with P<0.05 considered statistically significant.
Results
In this analysis of US adults, aged ≥20 years, B12 supplement use ranged from 0.065 to 2,800 mcg/day in the MVMM group (median: 12; Q1: 6, Q3: 25) and 2 to 18,000 mcg/day in the MVMM-exclusive group (median: 100; Q1: 22, Q3:500). Distributions of serum B12 were comparable across survey years. Regular use of any supplement containing vitamin B12– defined as MVMM or MVMM-exclusive supplementation at least 15 days of the month -- was most common among those 80 years or older (46.3% reporting use), and least common among 20–29-year-olds (18.6%). This pattern generally held with MVMM-exclusive vitamin B12 supplement use: use was most common amongst those 70–79 years, of whom 9.0% regularly used MVMM-exclusive B12 supplements, as compared to 1.4% of persons 20–29 years of age. Within Non-Hispanic White participants, 5.4% regularly used MVMM-exclusive B12 supplements as compared to 2.2% of non-Hispanic Black participants, and 2.1% of Mexican Americans. MVMM-exclusive B12 use increased with poverty-to-income ratio level, with higher income individuals reporting more use than lower income individuals. Women were more likely be MVMM-exclusive B12 users (5.5%) than men (3.8%). Current smokers were less likely be MVMM-exclusive B12 supplement users (3.1%) compared to never (4.8%) and former (5.9%) smokers (Table 1).
Table 1.
Population Characteristics, by Supplement Use
| Total | Supplement Use | |||
|---|---|---|---|---|
|
|
||||
| N = 11,757 N (Wgt %) | No B12 N = 8304 (67.9%) | MVMM B12 N = 2936 (27.4%) | MVMM-Exclusive B12 N= 517 (4.7%) | |
| Gender | ||||
| Male | 5945 (49.7) | 4372 (71.3) | 1344 (24.9) | 229 (3.8) |
| Female | 5812 (50.3) | 3932 (64.5) | 1592 (30.0) | 288 (5.5) |
| Age | ||||
| 20–29 | 1928 (17.6) | 1598 (81.4) | 308 (17.2) | 22 (1.4) |
| 30–39 | 1982 (20.7) | 1531 (73.8) | 406 (23.6) | 45 (2.6) |
| 40–49 | 2240 (21.9) | 1641 (68.2) | 496 (26.5) | 103 (5.3) |
| 50–59 | 1661 (17.6) | 1109 (62.5) | 460 (31.2) | 92 (6.3) |
| 60–69 | 1903 (11.8) | 1202 (55.0) | 588 (38.5) | 113 (6.5) |
| 70–79 | 1275 (7.2) | 807 (57.1) | 376 (33.8) | 92 (9.0) |
| 80+ | 768 (3.1) | 416 (53.7) | 302 (39.7) | 50 (6.6) |
| Race/ethnicity | ||||
| NH White | 6023 (73.7) | 3749 (63.6) | 1920 (31.0) | 354 (5.4) |
| NH African American | 2343 (10.5) | 1875 (82.4) | 403 (15.4) | 65 (2.2) |
| Mexican American | 2518 (6.9) | 2019 (81.3) | 438 (16.6) | 61 (2.1) |
| Other Hispanic | 476 (4.5) | 385 (79.7) | 76 (17.5) | 15 (2.8) |
| Other Race/Mixed Race | 397 (4.5) | 276 (71.1) | 99 (23.7) | 22 (5.2) |
| Poverty to income ratio | ||||
| <1 (lower income) | 1979 (11.9) | 1649 (81.3) | 287 (15.7) | 43 (3.0) |
| 1–1.99 | 3015 (20.1) | 2277 (75.2) | 614 (20.3) | 124 (4.6) |
| 2–3.99 | 3436 (30.5) | 2362 (68.4) | 906 (26.9) | 168 (4.7) |
| >= 4 (higher income) | 3327 (37.5) | 2016 (59.3) | 1129 (35.5) | 182 (5.3) |
| Education | ||||
| Less than high school | 3470 (18.0) | 2812 (79.6) | 560 (17.4) | 98 (3.0) |
| High school/GED or equivalent | 2789 (25.5) | 1972 (71.1) | 693 (24.3) | 124 (4.6) |
| Some college or AA degree | 3181 (30.6) | 2183 (66.9) | 871 (27.7) | 172 (5.4) |
| College grad or above | 2317 (25.9) | 1382 (57.8) | 812 (37.2) | 123 (5.0) |
| Body Mass Index (BMI) | ||||
| Underweight (<18.5) | 165 (1.6) | 123 (74.2) | 36 (22.9) | 6 (2.8) |
| Normal weight (18.5 – 24.9) | 3466 (32.1) | 2355 (65.9) | 937 (29.4) | 174 (4.7) |
| Overweight (25 – 29.9) | 4176 (34.1) | 2912 (67.1) | 1083 (28.2) | 179 (4.7) |
| Obese (≥ 30) | 3950 (32.2) | 2912 (70.4) | 880 (24.9) | 158 (4.7) |
| Smoking Status | ||||
| Never | 6,592 (55.3) | 4,594 (66.3) | 1,715 (28.8) | 283 (4.8) |
| Formera | 2786 (22.1) | 1814 (61.2) | 815 (32.9) | 157 (5.9) |
| 1st tertile of pack-years | 938 (34.9) | 644 (62.8) | 254 (32.6) | 40 (4.6) |
| 2nd tertile of pack-years | 925 (34.3) | 590 (60.2) | 280 (33.5) | 55 (6.3) |
| 3rd tertile of pack-years | 923 (30.8) | 580 (60.3) | 281 (32.7) | 62 (7.0) |
| Currentb | 2379 (22.6) | 1896 (78.2) | 406 (18.7) | 77 (3.1) |
| 1st tertile of pack-years | 805 (31.5) | 679 (82.7) | 112 (15.8) | 14 (1.5) |
| 2nd tertile of pack-years | 787 (34.5) | 626 (79.1) | 136 (18.1) | 25 (2.9) |
| 3rd tertile of pack-years | 787 (34.0) | 591 (73.2) | 158 (22.1) | 38 (4.7) |
| Alcohol Use | ||||
| < 1 / month | 5991 (45.3) | 4266 (69.1) | 1463 (26.1) | 262 (4.8) |
| ≥ 1 / month to < 4 / week | 3260 (30.1) | 2269 (65.7) | 857 (29.8) | 134 (4.5) |
| ≥ 4 / month to < 2 / day | 1898 (18.5) | 1323 (67.4) | 479 (27.8) | 96 (4.8) |
| ≥ 2 / day | 608 (6.1) | 446 (70.7) | 137 (25.1) | 25 (4.2) |
| Current health status | ||||
| Poor | 482 (3.0) | 336 (74.7) | 97 (21.7) | 19 (3.6) |
| Fair | 2077 (12.8) | 1592 (72.9) | 398 (22.2) | 87 (5.0) |
| Good | 3832 (30.4) | 2793 (70.8) | 898 (25.3) | 141 (3.9) |
| Very Good | 3186 (32.2) | 2116 (65.2) | 892 (29.2) | 178 (5.6) |
| Excellent | 2173 (21.6) | 1432 (67.9) | 649 (31.7) | 92 (4.4) |
| Diabetes | ||||
| No | 10562 (92.9) | 7442 (67.9) | 2669 (27.5) | 451 (4.5) |
| Yes | 1195 (7.1) | 862 (67.1) | 267 (26.4) | 66 (6.4) |
| Heart disease | ||||
| No | 10485 (91.4) | 7438 (68.1) | 2583 (27.4) | 437 (4.5) |
| Yes | 1299 (8.6) | 866 (65.2) | 353 (28.3) | 80 (6.5) |
| Anemia | ||||
| No | 11430 (97.8) | 8101 (68.0) | 2849 (27.5) | 480 (4.5) |
| Yes | 327 (2.2) | 203 (61.2) | 87 (26.8) | 37 (12.0) |
| Memory Loss | ||||
| No | 10855 (93.9) | 7667 (67.7) | 2717 (27.6) | 471 (4.7) |
| Yes | 902 (6.1) | 637 (70.4) | 219 (24.9) | 46 (4.7) |
ABBREVIATIONS: MVMM (Multi-vitamin/mineral supplement), NH (Non-Hispanic)
Tertile 1: (< 5.55 pack-years), Tertile 2: (5.55 – 24.0 pack-years), Tertile 3: (> 24.0 pack-years)
Tertile 1: (< 9.10 pack-years), Tertile 2: (9.10 – 26.0 pack-years), Tertile 3: (> 26.0 pack-years)
In fully-adjusted models, persons taking vitamin B12 through an MVMM source had a 26% (95% CI: 23% - 30%) higher geometric mean than non-users; among those using B12 individual supplements (exclusive of MVMM), the geometric mean of serum B12 was 61% (95% CI: 53%−70%) higher than the non-use group. Secondary analyses for supplemental dose, dietary B12 intake, and total dietary+supplemental intake revealed a similar pattern of association. Specifically, when examining associations for supplemental B12 intake (mcg/day), we observed that that the highest dose users (tertile 3) had a 46% (95% CI: 42% - 51%) higher geometric mean of serum vitamin B12 than B12 non-users (p-for-trend < 0.001).When examining the association between dietary B12 intake and serum B12 levels, the highest dietary intake quartile group’s geometric mean was 18% (95% CI: 14% - 22%) higher than the lowest quartile (p-for-trend < 0.001). Finally, the total intake model (corresponding to dietary+supplemental intake) showed the highest dose quartile group had a geometric mean 43% (95% CI: 39% - 48%) higher than the lowest group (p-for-trend < 0.001) (Table 2)
Table 2.
Association between Vitamin B12 Consumption and Serum Vitamin B12
| Cohort N (Wgt %) | Unadjusted Geometric Mean of B12 serum (pmol/L) | Minimally-Adjusteda Ratio (95% CI) | Multivariable-Adjustedb Ratio (95% CI) | ||||
|---|---|---|---|---|---|---|---|
| N = 11,757 | Mean | 95% CI | Ratio | 95% CI | Ratio | 95% CI | |
|
| |||||||
| Regular Supplement B12 Use | |||||||
| No B12 | 8304 (67.9) | 313.78 | (308.51, 319.05) | 1.00 | Ref | 1.00 | Ref |
| MVMM B12 | 2936 (27.4) | 389.59 | (381.00, 398.18) | 1.24 | (1.21, 1.28) | 1.26 | (1.23, 1.30) |
| MVMM-exclusive B12 | 517 (4.7) | 493.88 | (469.71, 518.04) | 1.58 | (1.50, 1.66) | 1.61 | (1.53, 1.70) |
| p-trend < 0.001 | p-trend < 0.001 | ||||||
| Supplemental B12 Intakec (mcg) | |||||||
| Non-users | 8304 (67.9) | 313.78 | (308.51, 319.05) | 1.00 | Ref | 1.00 | Ref |
| 1st Tertile | 1313 (12.4) | 379.77 | (367.76, 391.77) | 1.21 | (1.17, 1.25) | 1.23 | (1.19, 1.27) |
| 2nd Tertile | 1306 (11.8) | 394.21 | (382.80, 405.62) | 1.26 | (1.21, 1.31) | 1.28 | (1.23, 1.33) |
| 3rd Tertile | 834 (8.0) | 458.05 | (441.43, 474.67) | 1.47 | (1.42, 1.52) | 1.49 | (1.44, 1.54) |
| p-trend < 0.001 | p-trend < 0.001 | ||||||
| Dietary B12 Intaked (mcg) | |||||||
| 1st Quartile | 2945 (22.9) | 319.04 | (309.88, 328.21) | 1.00 | Ref | 1.00 | Ref |
| 2nd Quartile | 2936 (23.6) | 325.94 | (318.04, 333.85) | 1.03 | (0.99, 1.06) | 1.03 | (1.00, 1.07) |
| 3rd Quartile | 2937 (26.5) | 347.13 | (338.63, 355.62) | 1.10 | (1.07, 1.13) | 1.11 | (1.08, 1.15) |
| 4th Quartile | 2939 (27.0) | 365.34 | (358.00, 372.67) | 1.17 | (1.13, 1.20) | 1.18 | (1.14, 1.22) |
| p-trend < 0.001 | p-trend < 0.001 | ||||||
| Dietary+Supplement B12 Intakee (mcg) | |||||||
| 1st Quartile | 2943 (21.7) | 291.83 | (284.75, 298.91) | 1.00 | Ref | 1.00 | Ref |
| 2nd Quartile | 2941 (24.5) | 310.14 | (302.23, 318.05) | 1.06 | (1.04, 1.10) | 1.08 | (1.05, 1.11) |
| 3rd Quartile | 2935 (26.8) | 349.96 | (341.51, 358.42) | 1.21 | (1.17, 1.25) | 1.23 | (1.19, 1.27) |
| 4th Quartile | 2938 (27.0) | 406.52 | (399.48, 413.56) | 1.40 | (1.36, 1.44) | 1.43 | (1.39, 1.48) |
| p-trend < 0.001 | p-trend < 0.001 | ||||||
ABREVIATIONS: MVMM (Multi-vitamin/mineral supplement), mcg (micrograms)
Adjusted for age and gender
Additionally adjusted for race, poverty to income ratio, education, body mass index, smoking status, alcohol use, current health status, diabetes status, heart disease status, anemia treatment, memory loss
Non-users (0 mcg/day), Tertile 1 (0.065– 6.0 mcg/day), Tertile 2 (6.14 – 25 mcg/day), Tertile 3 (> 25.00 mcg/day)
Quartile 1 (< 2.04 mcg/day), Quartile 2 (2.04 – 3.68 mcg/day), Quartile 3 (3.69 – 6.13 mcg/day), Quartile 4 (> 6.13 mcg/day)
Quartile 1 (< 2.82 mcg/day), Quartile 2 (2.82 – 6.15 mcg/day), Quartile 3 (6.16 – 12.95 mcg/day), Quartile 4 (> 12.95 mcg/day)
We examined whether the association between B12 supplement use and serum B12 was modified by participants’ gender and smoking status. While a significant linear trend was observed among both males and females (p < 0.001), the association was stronger among females (p-interaction=0.01). Among males, the MVMM-exclusive group had a geometric mean 51% (95% CI, 44%–58%) higher than the non-use group. Among females, the MVMM-exclusive group’s geometric mean was 69% (95% CI, 55%−85%) higher than non-users. There was no significant interaction between smoking and use of B12 supplements (p=0.45), with significant linear trend observed for all groups (Table 3).
Table 3.
Stratified Association between Vitamin B12 Supplement Use and Serum Vitamin B12
| B12 Supplement Category | p-trend | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
|
|
||||||||||
| None | MVMM | MVMM exclusive | ||||||||
| N (Wgt %) | Ratioa | 95% CI | N (Wgt %) | Ratioa | 95% CI | N (Wgt %) | Ratioa | 95% CI | ||
|
| ||||||||||
| Gender | ||||||||||
| Male | 4372 (71.3) | 1.00 | Ref | 1344 (24.9) | 1.22 | (1.19, 1.26) | 229 (3.8) | 1.51 | (1.44, 1.58) | < 0.001 |
| Female | 3932 (64.5) | 1.00 | Ref | 1592 (30.0) | 1.30 | (1.25, 1.35) | 288 (5.5) | 1.69 | (1.55, 1.85) | < 0.001 |
| p-interaction < 0.001 | ||||||||||
| Smoking Status | ||||||||||
| Never | 4594 (66.3) | 1.00 | Ref | 1715 (28.8) | 1.25 | (1.21, 1.29) | 283 (4.8) | 1.66 | (1.54, 1.78) | < 0.001 |
| Former | 1814 (61.2) | 1.00 | Ref | 815 (32.9) | 1.30 | (1.23, 1.36) | 157 (5.9) | 1.55 | (1.42, 1.69) | < 0.001 |
| Current | 1896 (78.2) | 1.00 | Ref | 406 (18.7) | 1.25 | (1.17, 1.34) | 77 (3.1) | 1.53 | (1.37, 1.72) | < 0.001 |
| p-interaction = 0.45 | ||||||||||
Adjusted for sex, age, race, poverty to income ratio, education, body mass index, smoking status, alcohol use, current health status, diabetes status, heart disease status, anemia treatment, memory loss
Secondary analyses of the association between total vitamin B12 intake and serum B12 levels in gender-smoking jointly-defined strata revealed a statistically non-significant interaction by gender/smoking status (p-interaction: 0.07). Significant trends were observed in all groups (p-trend< 0.05), with the strongest association observed for female former smokers (Q4 vs Q1 Ratio: 1.51; 95% CI: 1.34, 1.70; p-trend < 0.001) and the weakest association observed for male never-smokers (Q4 vs Q1 Ratio: 1.32; 95% CI: 1.23, 1.42; p-trend< 0.001) (Table 4)
Table 4.
Stratified Association between Vitamin B12 Total Intake and Serum Vitamin B12 by Combined Gender and Smoking Status
| Smoking Status | B12 Supplemental & Dietary Combined Intakeb (mcg) | Gender | |||||
|---|---|---|---|---|---|---|---|
|
| |||||||
| Male | Female | ||||||
| N (wgt %) | Ratioa | 95% CI | N (wgt %) | Ratioa | 95% CI | ||
|
| |||||||
| Never Smoker | Quartile 1 | 568 (15.8) | 1.00 | Ref | 1094 (26.2) | 1.00 | Ref |
| Quartile 2 | 727 (26.2) | 1.03 | (0.96, 1.10) | 878 (21.1) | 1.12 | (1.07, 1.17) | |
| Quartile 3 | 780 (30.3) | 1.09 | (1.03, 1.17) | 901 (25.8) | 1.30 | (1.22, 1.38) | |
| Quartile 4 | 710 (27.8) | 1.32 | (1.23, 1.42) | 934 (26.9) | 1.47 | (1.39, 1.55) | |
| p-trend <000.1 | p-trend <000.1 | ||||||
| Former Smoker | Quartile 1 | 356 (15.1) | 1.00 | Ref | 272 (21.0) | 1.00 | Ref |
| Quartile 2 | 459 (25.8) | 1.06 | (0.98, 1.15) | 205 (19.6) | 1.04 | (0.92, 1.17) | |
| Quartile 3 | 429 (27.3) | 1.19 | (1.10, 1.28) | 231 (25.3) | 1.39 | (1.23, 1.56) | |
| Quartile 4 | 506 (31.8) | 1.44 | (1.35, 1.54) | 328 (34.1) | 1.51 | (1.34, 1.70) | |
| p-trend <000.1 | p-trend <000.1 | ||||||
| Current Smoker | Quartile 1 | 306 (19.6) | 1.00 | Ref | 347 (34.5) | 1.00 | Ref |
| Quartile 2 | 412 (290) | 1.11 | (1.04, 1.18) | 260 (28.0) | 1.02 | (0.93, 1.13) | |
| Quartile 3 | 401 (28.2) | 1.20 | (1.10, 1.27) | 193 (19.9) | 1.29 | (1.12, 1.48) | |
| Quartile 4 | 291 (23.3) | 1.34 | (1.25, 1.43) | 169 (17.6) | 1.50 | (1.33, 1.68) | |
| p-trend <000.1 | p-trend <000.1 | ||||||
| p-interaction = 0.07 | |||||||
Adjusted for sex, age, race, poverty to income ratio, education, body mass index, smoking status, pack-years, alcohol use, current health status, diabetes status, heart disease status, anemia treatment, memory loss
Quartile 1: (< 2.82 mcg/day), Quartile 2: (2.82–6.15 mcg/day), Quartile 3: (6.16– 12.95 mcg/day), Quartile 4: (≥ 12.95 mcg/day)
We have also conducted a series of sensitivity analyses to evaluate the robustness of study findings. First, to address concern that supplement use and dietary exposure may not be independent, we conducted a sensitivity analysis using a mutually adjusted model, including both vitamin B12 supplement use and dietary intake. No evident change in coefficients was observed, and therefore these results are not presented. Second, recognizing that the supplemental dose ranged up to 18,000 mcg/day (mean: 82.5, median: 15) for regular users, we were concerned that the threshold of the most highly exposed group may be too low to offer a meaningful comparison. This is particularly true of the dietary+supplemental variable, where the distribution is driven down by non-users (Q4 ranges includes values from 12.95 mcg/day upward). Thus, all total diet+supplemental models were run again alternatively using deciles of total dietary+supplement intake as the exposure variable (top decile cut-point = 100 mcg/day). Results were similar to quartile models, so they are not presented here.Use of metformin, aspirin and proton pump inhibitors (PPIs) have also been associated with vitamin B12 deficiency and absorption issues. Data on metformin was available for all four cycles of NHANES included in this study. However, NHANES only collected data on aspirin use from 1999 to 2004, and during the 2003–2004 cycles, PPIs became available over-the-counter (with over-the-counter use not captured in NHANES, thus making exposure hard to accurately capture from 2003–2004 onward). For each of the medications described, a sensitivity analysis using years with available data was run comparing a model of B12 supplement use level with B12 serum levels adjusted for all covariates including medication of interest (aspirin, metformin, or PPIs) to a similar model not including the medication. Results were unchanged and therefore these three medications were not included in the final models.
Finally, the age range of the dataset (20+) was large, and given concern associations may vary by age, we assessed whether the diet+supplement associations (overall and stratified by gender and smoking status) held when restricted to those age 50+; again, these analyses were conducted using the diet+supplement total exposure variable (as opposed to the 3-level supplement variable), given increased power for this exposure. In a sensitivity analysis of participants 50 years or older (n = 5,607), the multivariate-adjusted model comparing B12 supplementation level to B12 serum had a significant linear trend (see supplemental Table 1). MVMM users had a geometric mean 33% higher than the non-user group, and MVMM exclusive users had one 76% higher. Significant linear trends also were observed for stratified models by gender and smoking status (p < 0.001), with associations stronger than in the original model. Women continued to present with stronger relationships than men between B12 supplement intake group and serum B12 level. MVMM exclusive female and male users had an 85% and 66% higher geometric mean than their designated non-user groups, respectively. However, no significant interaction by gender was observed (see supplemental Table 2). Among the smoking groups, MVMM exclusive never, former, and current smokers showed 85%, 62%, and 74% higher geometric means respectively than non-user groups, again, with the interaction not significant. To provide results comparable to cohorts using 44+ as an age cut-point, so we have also provided supplemental tables for this subset of the study population (n=6,911). With the larger sample size, effect modification by gender was statistically significant (see supplemental Tables 3 & 4).
Discussion
In this large, nationally-representative study, vitamin B12 supplement use was significantly associated with higher blood serum levels than non-use, as expected, with comparable associations observed in sensitivity analyses for supplemental dose, dietary intake, and total supplemental + dietary intake. While significant associations between vitamin B12 supplement use and serum vitamin B12 were observed among both men and women, there was significant heterogeneity by gender, with stronger associations observed among women than men. No interaction was observed by smoking status, with the relationship between vitamin B12 supplement use and vitamin B12 blood levels significant for all groups (never, former, and current smokers).
Regular MVMM-exclusive B12 users presented with 61% higher geometric mean B12 serum levels than non-users (95% CI: 53% - 70%), an association stronger than that observed for lower dose as consumed through a MVMM (Ratio: 1.26; 95% CI: 1.23–1.30). Secondary analyses by dose (supplemental intake, dietary intake, or combined supplemental + dietary intake in mcg/day) revealed a significant linear trend, regardless of source, in agreement with previous studies30–33—including a cross-sectional analysis of B12 intake and plasma B12 from male and female Framingham Heart Study participants which showed linear trends for combined dietary/supplemental intake of B12 and B12 blood concentrations among both supplement and non-supplement users.34 It should be noted that while the range of MVMM and MVMM-exclusive intakes was wide (min-max: 0.065–2,800 and 2–18,000 mcg/day respectively), the thresholds in the highest dose group in our diet+supplemental intake variable (Q4 defined by intake more than 12.95 mcg/day) did not come close to that commonly found in MVMM-exclusive B vitamins. As a water-soluble vitamin, intakes in excess of what the body needs are excreted in the urine, thus no tolerable upper limit has been set for vitamin B12.
Prior analysis of NHANES III with data from 1991–1994 also showed that supplemental B12 use was associated with increased B12 serum levels. As with our findings, the association was attenuated slightly among older individuals ≥50 years, owing to reduced absorption often seen in older adults.35 Similarly, a randomized controlled trial comparing 2000 mcg/day of MVMM-exclusive oral supplementation to intermuscular injections revealed an increase in serum B12 among the oral supplement group.36 In another six-month randomized controlled trial of New Zealand women, significant increases in serum B12 were associated with supplementation even after women were stratified by dietary intake characteristics.30
Previous cross-sectional studies have found lower serum B12 concentrations and higher prevalence of B12 deficiency among men.8–11 However, these articles did not evaluate this association in the context of long-term supplementation, so we further examined the relationship between regular supplement use and serum B12, stratified by gender, and observed that the association between supplementation category and serum B12 was significantly stronger among women than men. Females had a greater difference in geometric mean serum B12 between MVMM-exclusive users and non-users (69%; 95% CI: 58% - 98%) compared to male MVMM-exclusive users and non-users (51%; 95% CI: 44% – 58%). One might expect this result if B12 intake was higher among women to begin with, but male and female intakes among both regular user groups were comparable. Our current finding of a significant interaction between gender and serum B12 provides more evidence that females and males may metabolize vitamin B12 differently.
It is hypothesized that these gender differences could be partially explained by genetic variation. One cross-sectional study found that levels of transcobalamin, the necessary glycoprotein for vitamin B12’s active form, were also higher among females than males.37 Estrogen has been shown to increase levels of important enzymes for the one-carbon metabolism such as phosphatidylethanolamine N-methyltransferase (PEMT).38 Female sex hormones are also associated with fluctuations in homocysteine and folate serum levels during menstrual cycles,39–41 and based on NHANES II data from 1991–1994, pregnant women presented with significantly lower serum homocysteine levels than non-pregnant, non-oral-contraceptive-using women.42 These studies can be extrapolated to the interaction between vitamin B12 and gender as B12 is frequently associated with homocysteine and folate levels in the blood.8,43 These findings might partially explain the weakened interaction by gender in the subset of the study population 50 years or older as hormone levels decline rapidly in postmenopausal women.
A recent cross-sectional study found that the protective association between female sex on serum B12 cannot be explained by estrogen alone.44 Mathematical modeling has described a difference in the amount of one-carbon metabolism enzymes in women and men, with females presenting higher levels of PEMT, choline, betaine, and S-andenosyl-homocysteine, while males present with higher levels of homocysteine and S-andenosylmethionine.45 Differences in the enzymes that drive the one-carbon metabolism could explain gender divides in serum B12 metabolism observed in this study and others, but more research is necessary to fully understand the biological mechanisms affecting this relationship.
As with gender, significant linear trends were observed between supplement category and serum vitamin B12 in all groups, regardless of smoking status (non-, former, and current smokers), with no significant interaction by smoking status. Decreased levels of serum B12 levels among smokers compared to non-smokers have been well documented.4–7 In one randomized controlled trial of hospitalized elderly individuals were given 100% of the Reference Nutrient Intake for B12 and other micronutrients in a micronutrient drink or a placebo drink for six weeks; smokers and ex-smokers presented with lower B12 serum levels than never smokers overall, but no significant difference between groups was observed for change in B12 status.22 Similarly, a case-control study of Iranian men revealed no significant difference between total serum B12 in smokers and non-smokers, but further investigation revealed that the ratio of active forms of B12 to total serum B12 was lower among smokers.46 These results would seem to suggest that future studies should evaluate the relationship between active forms of B12 and smoking status.
Understanding factors that affect B12 absorption and metabolism is important as previous studies have shown an increased risk of some cancers with higher vitamin B12 intakes. B12 is a substrate of the methionine synthase enzyme within the one-carbon metabolism. Higher levels of B12 provide more substrate for S-adenosylmethionine (SAM) production. SAM acts as the universal methyl donor and can alter expression of protooncogenes and tumor suppressor genes, which are associated with cancer.18,19 High plasma B12 was associated with high 1-year risk of cancer (any site) among a Danish cohort within a primary care health improvement setting. Researchers hypothesized that individuals with undiagnosed cancers might metabolize B12 differently,15 as previously observed for folic acid intake and colorectal cancer progression.47–49 Another cohort study of hospital patients found that individuals with persistently high plasma B12 were at higher risk of solid tumor incidence within 5 years.50 A randomized controlled trial of individuals taking 500 mcg/day of B12 as well as 400 mcg/day of folic acid for 2–3 years exhibited increased colorectal cancer risk in the experimental group.16
However, cohort studies appear to disagree on whether women have higher risk of cancer than men after B12 supplementation. A pooled analysis from the Lung Cancer Cohort Consortium revealed a positive association between serum B12 and lung cancer adenocarcinoma risk, but a significant dose-response association was only observed for women, current and former smokers.13 Long-term high-dose MVMM-exclusive B12 supplementation within a prospective cohort study was associated with increased lung cancer risk among men, but not women.14 Further research of the Women’s Health Initiative similarly revealed no association between B12 supplementation and lung cancer among women.21 These results are at odds with findings from our study, which imply that those taking the highest supplements have higher levels of B12 in the blood. Additionally, in our study, women have higher B12 serum levels across supplementation levels than men. Further research on B12 supplementation by sex and smoking status is necessary to understand how differences in metabolism may affect cancer risk.
One strength of this study is that NHANES collects detailed information on supplements from an in-home interview, documenting detailed information, including ingredients and dosages. Furthermore, as dietary recall data were available, we were able to consider both supplemental and dietary sources in our analyses. This large, well-powered study is nationally-representative of non-incarcerated U.S. citizens, facilitating the generalizability of these findings. There are however several important limitations to consider. Firstly, several B vitamins affect the one carbon metabolism,17 but only B12 was analyzed here. One or more of these vitamins may have confounded results. It was also unclear if individuals had ingested their B12 supplements the day before their B12 blood serum levels were taken (even so, by defining use by regular use, we hope to have largely mitigated this issue). Additionally, dose values recorded may not match exactly what was consumed by survey participants because in some cases, the NHANES interviewer may not have seen the supplement bottle during the interview, meaning it would be recorded as a generic B12 supplement. Similarly, there may be measurement error in dietary recall as underreporting of energy intake has been seen in NHANES data collection51 which has been linked to underreporting of micronutrients as well.52 While we would expect such measurement error to attenuate results toward the null overall, it is possible that such measurement error could vary by gender or smoking status, which could bias effect estimates of interaction. Lastly, while extensive effort was made to control for confounding, we were not able to include all factors that might affect B12 blood serum levels, which may have resulted in the low R2 values for the multivariate linear regression models (11%). Even so, efforts were made to adjust for all known factors, so we have some confidence in results.
In summary, this study revealed that increased regular B12 supplement use was associated with higher B12 blood serum levels, with significant effect modification by gender but not smoking status. Further research on this topic would be of merit to establish the biological mechanisms that affect absorption of B12 from supplemental sources.
Supplementary Material
Acknowledgements
We thank the Memorial Sloan Kettering Quantitative Science Undergraduate Research Experience, an NCI-funded training grant (R25 CA214255).
Sources of support:
This work was supported by the National Cancer Institute of the National Institutes of Health under Grant R25 CA214255 and P30 CA008748.
Footnotes
Declaration of interest statement
Authors have no conflict of interest to report.
References:
- 1.Shipton MJ, Thachil J: Vitamin B12 deficiency - A 21st century perspective. Clinical Medicine (London, England) 15, 145–150, 2015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Oh R, Brown DL: Vitamin B12 deficiency. American Family Physician 67, 979–986, 2003. [PubMed] [Google Scholar]
- 3.Kantor ED, Rehm CD, Du M, White E, Giovannucci EL: Trends in Dietary Supplement Use Among US Adults From 1999–2012. JAMA 316, 1464–1474, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Gabriel HE, Crott JW, Ghandour H, Dallal GE, Choi S-W, et al. : Chronic cigarette smoking is associated with diminished folate status, altered folate form distribution, and increased genetic damage in the buccal mucosa of healthy adults. The American Journal of Clinical Nutrition 83, 835–841, 2006. [DOI] [PubMed] [Google Scholar]
- 5.Haj Mouhamed D, Ezzaher A, Neffati F, Douki W, Najjar MF: Effect of cigarette smoking on plasma homocysteine concentrations. Clinical Chemistry and Laboratory Medicine 49, 479–483, 2011. [DOI] [PubMed] [Google Scholar]
- 6.Piyathilake CJ, Macaluso M, Hine RJ, Richards EW, Krumdieck CL: Local and systemic effects of cigarette smoking on folate and vitamin B-12. The American Journal of Clinical Nutrition 60, 559–566, 1994. [DOI] [PubMed] [Google Scholar]
- 7.Northrop-Clewes CA, Thurnham DI: Monitoring micronutrients in cigarette smokers. Clinica Chimica Acta; International Journal of Clinical Chemistry 377, 14–38, 2007. [DOI] [PubMed] [Google Scholar]
- 8.Cohen E, Margalit I, Shochat T, Goldberg E, Krause I: Gender differences in homocysteine concentrations, a population-based cross-sectional study. Nutrition, metabolism, and cardiovascular diseases: NMCD 29, 9–14, 2019. [DOI] [PubMed] [Google Scholar]
- 9.Loikas S, Koskinen P, Irjala K, Lopponen M, Isoaho R, et al. : Vitamin B12 deficiency in the aged: a population-based study. Age and Ageing 36, 177–183, 2007. [DOI] [PubMed] [Google Scholar]
- 10.Alvarez M, Sierra OR, Saavedra G, Moreno S: Vitamin B12 deficiency and diabetic neuropathy in patients taking metformin: a cross-sectional study. Endocrine Connections 8, 1324–1329, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Lim H-S, Heo Y-R: Plasma total homocysteine, folate, and vitamin B12 status in Korean adults. Journal of Nutritional Science and Vitaminology 48, 290–297, 2002. [DOI] [PubMed] [Google Scholar]
- 12.Price AJ, Travis RC, Appleby PN, Albanes D, Barricarte Gurrea A, et al. : Circulating Folate and Vitamin B12 and Risk of Prostate Cancer: A Collaborative Analysis of Individual Participant Data from Six Cohorts Including 6875 Cases and 8104 Controls. European Urology 70, 941–951, 2016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Fanidi A, Carreras-Torres R, Larose TL, Yuan J-M, Stevens VL, et al. : Is high vitamin B12 status a cause of lung cancer? International Journal of Cancer 145, 1499–1503, 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Brasky TM, White E, Chen C-L: Long-Term, Supplemental, One-Carbon Metabolism-Related Vitamin B Use in Relation to Lung Cancer Risk in the Vitamins and Lifestyle (VITAL) Cohort. Journal of Clinical Oncology: Official Journal of the American Society of Clinical Oncology 35, 3440–3448, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Arendt JFH, Sørensen HT, Horsfall LJ, Petersen I: Elevated Vitamin B12 Levels and Cancer Risk in UK Primary Care: A THIN Database Cohort Study. Cancer Epidemiology and Prevention Biomarkers 28, 814–821, 2019. [DOI] [PubMed] [Google Scholar]
- 16.Araghi SO, Jong JCK, Dijk SC van, Swart KMA, Laarhoven HW van, et al. : Folic Acid and Vitamin B12 Supplementation and the Risk of Cancer: Long-term Follow-up of the B Vitamins for the Prevention of Osteoporotic Fractures (B-PROOF) Trial. Cancer Epidemiology and Prevention Biomarkers 28, 275–282, 2019. [DOI] [PubMed] [Google Scholar]
- 17.Friso S, Udali S, De Santis D, Choi S-W: One-carbon metabolism and epigenetics. Molecular Aspects of Medicine 54, 28–36, 2017. [DOI] [PubMed] [Google Scholar]
- 18.Newman AC, Maddocks ODK: One-carbon metabolism in cancer. British Journal of Cancer 116, 1499–1504, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Wajed SA, Laird PW, DeMeester TR: DNA Methylation: An Alternative Pathway to Cancer. Annals of Surgery 234, 10–20, 2001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Hultdin J, Van Guelpen B, Bergh A, Hallmans G, Stattin P: Plasma folate, vitamin B12, and homocysteine and prostate cancer risk: a prospective study. International Journal of Cancer 113, 819–824, 2005. [DOI] [PubMed] [Google Scholar]
- 21.Brasky TM, Ray RM, Navarro SL, Schenk JM, Newton AM, et al. : Supplemental one-carbon metabolism related B vitamins and lung cancer risk in the Women’s Health Initiative. International Journal of Cancer 147, 1374–1384, 2020. [DOI] [PubMed] [Google Scholar]
- 22.Gariballa S, Forster S: Effects of smoking on nutrition status and response to dietary supplements during acute illness. Nutrition in Clinical Practice: Official Publication of the American Society for Parenteral and Enteral Nutrition 24, 84–90, 2009. [DOI] [PubMed] [Google Scholar]
- 23.Office of Dietary Supplements, National Institutes of Health [Internet]: Dietary Supplement Label Database. Bethesda (MD): NIH; c2009 [cited 2021]. Available from: https://ods.od.nih.gov/Research/Dietary_Supplement_Label_Database.aspx.
- 24.Institute of Medicine: Vitamin B12. In: Dietary Reference Intakes: The Essential Guide to Nutrient Requirements. Washington, DC: National Academies Press, 2006, p. 189–195. [Google Scholar]
- 25.National Center for Health Statistics [Internet]: NHANES 2003–2004 public data general release file documentation. [cited 2021 Jun 20] Available from https://www.cdc.gov/nchs/data/nhanes/nhanes_03_04/general_data_release_doc_03-04.pdf.
- 26.National Center for Health Statistics [Internet]: NHANES 2005–2006 Data Documentation, Codebook, and Frequencies: Dietary Supplement Use 30-Day - File 2, Participant’s Use of Supplements [cited 2021 Jun 20]. Available from https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/DSQ2_D.htm.
- 27.National Center for Health Statistics [Internet]: NHANES 2005–2006 Data Documentation, Codebook, and Frequencies: Dietary Interview - Total Nutrient Intakes, First Day. [cited 2021 Jun 20]. Available from https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/DR1TOT_D.htm.
- 28.National Center for Health Statistics [Internet]: NHANES 2005–2006 Data Documentation, Codebook, and Frequencies: Vitamin B12. [cited 2021 Jun 20] Available from https://wwwn.cdc.gov/Nchs/Nhanes/2005-2006/B12_D.htm.
- 29.StataCorp: Stata Statistical Software: Release 15 [software]. 2017. [cited 2021]. Available from: https://www.stata.com/.
- 30.Mearns GJ, Koziol-McLain J, Obolonkin V, Rush EC: Preventing vitamin B12 deficiency in South Asian women of childbearing age: a randomised controlled trial comparing an oral vitamin B12 supplement with B12 dietary advice. European Journal of Clinical Nutrition 68, 870–875, 2014. [DOI] [PubMed] [Google Scholar]
- 31.Bor MV, Lydeking-Olsen E, Møller J, Nexø E: A daily intake of approximately 6 microg vitamin B-12 appears to saturate all the vitamin B-12-related variables in Danish postmenopausal women. The American Journal of Clinical Nutrition 83, 52–58, 2006. [DOI] [PubMed] [Google Scholar]
- 32.Narayanan MN, Dawson DW, Lewis MJ: Dietary deficiency of vitamin B12 is associated with low serum cobalamin levels in non-vegetarians. European Journal of Haematology 47, 115–118, 1991. [DOI] [PubMed] [Google Scholar]
- 33.Ford TC, Downey LA, Simpson T, McPhee G, Oliver C, et al. : The Effect of a High-Dose Vitamin B Multivitamin Supplement on the Relationship between Brain Metabolism and Blood Biomarkers of Oxidative Stress: A Randomized Control Trial. Nutrients 10, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Tucker KL, Rich S, Rosenberg I, Jacques P, Dallal G, et al. : Plasma vitamin B-12 concentrations relate to intake source in the Framingham Offspring study. The American Journal of Clinical Nutrition 71, 514–522, 2000. [DOI] [PubMed] [Google Scholar]
- 35.Evatt ML, Terry PD, Ziegler TR, Oakley GP: Association between vitamin B12-containing supplement consumption and prevalence of biochemically defined B12 deficiency in adults in NHANES III (third national health and nutrition examination survey). Public Health Nutrition 13, 25–31, 2010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Kuzminski AM, Del Giacco EJ, Allen RH, Stabler SP, Lindenbaum J: Effective treatment of cobalamin deficiency with oral cobalamin. Blood 92, 1191–1198, 1998. [PubMed] [Google Scholar]
- 37.Fernandes-Costa F, van Tonder S, Metz J: A sex difference in serum cobalamin and transcobalamin levels. The American Journal of Clinical Nutrition 41, 784–786, 1985. [DOI] [PubMed] [Google Scholar]
- 38.Resseguie M, Song J, Niculescu MD, da Costa K-A, Randall TA, et al. : Phosphatidylethanolamine N-methyltransferase (PEMT) gene expression is induced by estrogen in human and mouse primary hepatocytes. FASEB journal: official publication of the Federation of American Societies for Experimental Biology 21, 2622–2632, 2007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Tallova J, Tomandl J, Bicikova M, Hill M: Changes of plasma total homocysteine levels during the menstrual cycle. European Journal of Clinical Investigation 29, 1041–1044, 1999. [DOI] [PubMed] [Google Scholar]
- 40.Elhadd TA, Neary R, Abdu T a M, Kennedy G, Hill A, et al. : Influence of the hormonal changes during the normal menstrual cycle in healthy young women on soluble adhesion molecules, plasma homocysteine, free radical markers and lipoprotein fractions. International Angiology: A Journal of the International Union of Angiology 22, 222–228, 2003. [PubMed] [Google Scholar]
- 41.Michels KA, Wactawski-Wende J, Mills JL, Schliep KC, Gaskins AJ, et al. : Folate, homocysteine and the ovarian cycle among healthy regularly menstruating women. Human Reproduction (Oxford, England) 32, 1743–1750, 2017. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Morris MS, Jacques PF, Selhub J, Rosenberg IH: Total homocysteine and estrogen status indicators in the Third National Health and Nutrition Examination Survey. American Journal of Epidemiology 152, 140–148, 2000. [DOI] [PubMed] [Google Scholar]
- 43.Hunt A, Harrington D, Robinson S: Vitamin B12 deficiency. BMJ: British Medical Journal 349, 2014. [DOI] [PubMed] [Google Scholar]
- 44.Margalit I, Cohen E, Goldberg E, Krause I: Vitamin B12 Deficiency and the Role of Gender: A Cross-Sectional Study of a Large Cohort. Annals of Nutrition & Metabolism 72, 265–271, 2018. [DOI] [PubMed] [Google Scholar]
- 45.Sadre-Marandi F, Dahdoul T, Reed MC, Nijhout HF: Sex differences in hepatic one-carbon metabolism. BMC systems biology 12, 89, 2018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Shekoohi N, Javanbakht MH, Sohrabi M, Zarei M, Mohammadi H, et al. : Smoking Discriminately Changes the Serum Active and Non-Active Forms of Vitamin B12. Acta Medica Iranica 55, 389–394, 2017. [PubMed] [Google Scholar]
- 47.Mason JB, Dickstein A, Jacques PF, Haggarty P, Selhub J, et al. : A temporal association between folic acid fortification and an increase in colorectal cancer rates may be illuminating important biological principles: a hypothesis. Cancer Epidemiology, Biomarkers & Prevention: A Publication of the American Association for Cancer Research, Cosponsored by the American Society of Preventive Oncology 16, 1325–1329, 2007. [DOI] [PubMed] [Google Scholar]
- 48.Kim Y-I: Role of Folate in Colon Cancer Development and Progression. The Journal of Nutrition 133, 3731S–3739S, 2003. [DOI] [PubMed] [Google Scholar]
- 49.Ulrich CM: Folate and Cancer Prevention—Where to Next? Counterpoint. Cancer Epidemiology and Prevention Biomarkers 17, 2226–2230, 2008. [DOI] [PubMed] [Google Scholar]
- 50.Lacombe V, Chabrun F, Lacout C, Ghali A, Capitain O, et al. : Persistent elevation of plasma vitamin B12 is strongly associated with solid cancer. Scientific Reports 11, 13361, 2021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Archer E, Hand GA, Blair SN: Validity of U.S. Nutritional Surveillance: National Health and Nutrition Examination Survey Caloric Energy Intake Data, 1971–2010 Johannsen D, editor. PLoS ONE 8, e76632, 2013. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 52.Briefel RR, Sempos CT, McDowell MA, Chien S, Alaimo K: Dietary methods research in the third National Health and Nutrition Examination Survey: underreporting of energy intake. The American Journal of Clinical Nutrition 65, 1203S–1209S, 1997. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.