YCLNU3389_proof ■ 19 February 2018 ■ 1/9

Clinical Nutrition xxx (2018) 1e9

55
Contents lists available at ScienceDirect 56
57
58
Clinical Nutrition 59
60
journal homepage: http://www.elsevier.com/locate/clnu 61
62
63
Randomized Control Trials 64
65
1 Effect of two different sublingual dosages of vitamin B12 on cobalamin 66
2 67
3 nutritional status in vegans and vegetarians with a marginal 68
4
5 Q9 deficiency: A randomized controlled trial 69
70
6 71
7 Q8 Cristian Del Bo'*, Patrizia Riso, Claudio Gardana, Antonella Brusamolino, 72
8 Alberto Battezzati, Salvatore Ciappellano 73
9 74

degli Studi di Milano, Milan, Italy
Department of Food, Environmental and Nutritional Sciences, Division of Human Nutrition, Universita
10 75
11 76
12 77
13
a r t i c l e i n f o s u m m a r y
78
14 79
Article history: Background & aims: Vegetarians and vegans are more vulnerable to vitamin B12 deficiency with severe
15 Received 6 November 2017 80
risks of megaloblastic anemia, cognitive decline, neuropathy, and depression. An easy and simple method
16 Accepted 8 February 2018 of supplementation consists of taking one weekly dosage of 2000 mg. However, single large oral doses of 81
17 vitamin B12 are poorly absorbed. The present research evaluates the ability of two different sublingual 82
18 Keywords: dosages of vitamin B12 (350 mg/week vs. 2000 mg/week) in improving cyanocobalamin (vitamin B12) 83
19 Vitamin B12
nutritional status in vegans and vegetarians with a marginal deficiency. 84
Metabolites
20 Sublingual supplements
Methods: A 12-week randomized, double-blind, controlled, parallel intervention trial was performed. 85
21 Forty subjects with marginal vitamin B12 deficiency were enrolled and randomly divided into two
Vegans 86
22 Vegetarians groups: test group Ld (low dose, 350 mg/week) and control group Hd (high dose, 2000 mg/week) vitamin
87
23 B12 supplementation. Blood samples were collected at baseline and after 15, 30, 60, and 90 days from the
88
intervention for the determination of vitamin B12, related metabolic markers, and blood cell counts.
24 89
Results: Two-way analysis of variance showed a significant effect of time (P < 0.0001) and of
25 90
time
treatment interaction (P ¼ 0.012) on serum concentration of vitamin B12 that increased after 90-
26 day supplementation (Ld and Hd) compared to baseline. Both the supplements increased (P < 0.0001, 91
27 time effect) the levels of holotranscobalamin, succinic acid, methionine and wellness parameter, while 92
28 decreased (P < 0.0001, time effect) the levels of methylmalonic acid, homocysteine and folate compared 93
29 to baseline. No difference was observed between groups (Ld vs. Hd). No effect was detected for vitamin B6 94
30 and blood cell count. 95
31 Conclusions: In our experimental conditions, both supplements were able to restore adequate serum 96
32 concentrations of vitamin B12 and to improve the levels of related metabolic blood markers in subjects
97
33 with a marginal deficiency. The results support the use of a sublingual dosage of 50 mg/day (350 mg/week)
98
34 of cobalamin, instead of 2000 mg/week (provided as a single dose), to reach a state of nutritional ade-
quacy of vitamin B12 in this target population.
99
35 100
This study was registered at www.isrctn.org as ISRCTN75099618.
36 101
© 2018 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
37 102
38 103
39 104
40 105
41 106
42 107
Q1 1. Introduction erythrocytes, in the maintenance of the central nervous system,
43 108
and in cognitive performance [1]. Cyanocobalamin is present in
44 109
Vitamin B12 (cyanocobalamin) represents an important and large amounts in animal products such as meat, organ meats,
45 110
essential water-soluble nutrient involved in the formation of shellfish, eggs, milk, and other dairy foods. Plant foods do not
46 111
contain vitamin B12 unless they are fortified (e.g., some breakfast
47 112
cereals); however, the body absorbs animal sources of vitamin B12
48 113
much better than plant sources [1,2]. The physiological absorption
49 * Corresponding author. Department of Food, Environmental and Nutritional 114
of vitamin B12 is mediated by the glycoprotein intrinsic factor (IF).
50 Sciences, Division of Human Nutrition, Division of Human Nutrition, Universit
a 115
degli Studi di Milano, Via G. Celoria 2, 20133, Milano, Italy. Fax: þ390250316721.
For its absorption, the formation of the IF-B12 complex and the
51 116
E-mail address: [email protected] (C. Del Bo'). transport of vitamin B12 across the ileum is required [1,2]. Once
52 117
53 118
https://doi.org/10.1016/j.clnu.2018.02.008
54 0261-5614/© 2018 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved. 119

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 2/9

2 C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9

1 absorbed, vitamin B12 is mainly accumulated in the liver and stored 2. Materials and methods 66
2 for years before using [1,2]. 67
3 The recommendations for B12 intakes vary significantly from 2.1. Subject recruitment 68
4 country to country and individual to individual [3]. Normally, in 69
5 healthy individuals with an ordinary omnivorous diet, a daily The screening of the participants was performed between March 70
6 consumption of a few micrograms of vitamin B12 is enough to 2015 and July 2016 through advertisements on bulletin boards, 71
7 preserve adequate levels of the vitamin [3,4]. In Italy, the National telephone, or e-mail. Subjects were visited for a routine medical 72
8 Reference of Energy and Nutrient Intake Levels (LARN) identified an examination by a physician to assess their eligibility to participate in 73
9 average requirement of 2.4 mg a day for adults and up to 2.6 mg and the trial. The eligibility was assessed by a physician through an ac- 74
10 2.8 mg in pregnancy and lactation, respectively [4]. A deficiency of curate examination and by means of a health/medical questionnaire 75
11 vitamin B12 could be the result of gastrointestinal disorders, celiac to exclude subjects with diseases such as diabetes, renal insuffi- 76
12 disease, Crohn's disease, and genetic polymorphisms leading to ciency, allergies, chronic constipation, diarrhea, or any other 77
13 malabsorption of the nutrient [1,2]. However, this condition is less gastrointestinal disorder. Moreover, a small aliquot of blood was 78
14 frequent; elderly and vegetarians are more susceptible to the con- collected to ascertain vitamin B12 nutritional status. Subjects were 79
15 dition of vitamin B12 deficiency due to their limited intake of meat selected according to the following inclusion criteria: vegan and 80
16 products [5,6]. On the contrary, vegans that exclude animal prod- vegetarian subjects in a condition of marginal vitamin B12 deficiency 81
17 ucts from their diet frequently become deficient in vitamin B12. In (<220 pmol/L) or full-blown (<150 pmol/L), non-smokers or light 82
18 this regard, a recent systematic review evaluated the prevalence of smokers (maximum 5e6 cigarettes/day), and moderate alcohol 83
19 vitamin B12 deficiency in individuals adhering to vegetarian and consumption (up to 14 glasses of wine/beer per week). Subjects 84
20 vegan diets [7]. The authors documented that adherence to a vegan with cardiovascular, coronary, diabetes, hepatic, renal, or gastroin- 85
21 diet was associated with an increased risk of vitamin B12 deficiency testinal diseases were excluded. Subjects were not included if using 86
22 compared to a vegetarian diet [7]. These findings were in line with drugs, medications, and/or supplements at least one month before 87
23 the observations reported by other authors [8e11]. the beginning of the experiment. Moreover, subjects were excluded 88
24 Vitamin B12 deficiency has been associated with several meta- if taking vitamin B12 supplements at least one year before the 89
25 bolic disorders such as macrocytic anemia, hyperhomocysteinemia, experiment. The study was performed in accordance with the 90
26 cardiovascular, cerebrovascular, and neurological disorders ethical standards established in the 2013 Declaration of Helsinki and 91
27 [6,12e14]. However, despite the high risk of developing vitamin B12 approved by the Ethics Committee of the University of Milan (March 92
28 deficiency and related complications, numerous vegans consider 4, 2015, ref. 11/15). The study was registered at www.isrctn.org as 93
29 supplementation unnecessary. The deficiency appears after a long ISRCTN75099618. All participants signed an informed consent form. 94
30 period of depletion (can take years in some), due to the stocks of 95
31 vitamin present in the liver [15]. Individuals with serum levels of 2.2. Experimental design 96
32 B12 < 150 pmol/L are considered deficient [15,16], while subjects 97
33 who have values between 150 and 221 pmol/L are considered A researcher who was not involved in the study and in sample 98
34 marginally deficient [17,18]. In this specific situation, the integra- analysis was appointed to allocate patients to the different treat- 99
35 tion of vitamin B12 by the parenteral route is required. However, ments according to a randomization list obtained through the 100
36 this approach is poorly accepted because the results painful and center's database. The number of participants who were randomly 101
37 expensive [19] as well as substituted by oral formulations. How- assigned to different study groups, the rate of patients completing 102
38 ever, this is not effective in subjects suffering from vomiting or the study, and patients analyzed for the primary outcome are 103
39 diarrhea or are not able to tolerate oral therapies [20]. Moreover, depicted in Fig. 1. Forty subjects were enrolled and randomly 104
40 when high doses of vitamin B12 are given orally, only a small per- divided into two groups of 20 subjects each for a 12-week double- 105
41 centage seems to be absorbed. Recently, the administration of blind (participants and outcome assessors), randomized, 106
42 vitamin B12 in sublingual form has been developed [20]. Although controlled, parallel dietary intervention study. The study was per- 107
43 sublingual vitamin B12 is often promoted for better absorption, formed between May 2015 and October 2016. One group received 108
44 inconsistent results have been obtained as to the effects of the supplement at a low dose (Ld; equivalent to 50 mg/day, 350 mg/ 109
45 administration of low and high doses of vitamin B12. week), while the other group (control) received the supplement at 110
46 The aim of the present study was to evaluate the ability of two a high dose (Hd; equivalent to 2000 mg/week in a single dose). 111
47 different doses (350 mg/week vs. 2000 mg/week) of sublingual Vitamin B12 was provided to the volunteers in one stock at the 112
48 supplements in improving the nutritional status of cyanocobal- beginning of the study. Each subject received 13 boxes containing 113
49 amin in a group of vegans and vegetarians with a marginal defi- the doses for a week in a blind condition. All tablets were packaged 114
50 ciency. The low dose (Ld) consisted of 7 sublingual tablets each and numbered (from 1 to 7) in single-dose blisters. Subjects were 115
51 providing 50 mg/day (350 mg/week) of vitamin B12, while the high instructed to follow the sequence of numbers and to swallow one 116
52 dose (Hd) consisted of 1 sublingual tablet (2000 mg) for the entire tablet per day in the morning before breakfast. The Ld group 117
53 week. The latter represents the most common method of sup- ingested 7 sublingual tablets/week of cyanocobalamin (50 mg each, 118
54 plementation, even if it is administered by the oral or parenteral equivalent to 350 mg), while the Hd group took only 1 sublingual 119
55 route. In this regard, several studies have shown low absorption tablet of vitamin B12 (2000 mg) and 6 sublingual tablets of placebo. 120
56 following the intake of high doses [1,21]. In addition, this practice For both groups (Ld and Hd), the tablets of vitamin B12 consisted of 121
57 could be less tolerated in some subjects; for example, some au- mannitol, maize starch, vegetable stearate magnesium, beet juice, 122
58 thors found adverse effects (e.g., hyperhidrosis and blurred vision) and sucralose. The placebo tablets matched the shape, size, color, 123
59 following supplementation with 1 mg/day of vitamin B12 in in- flavor, and the composition of the vitamin B12 supplements. The 124
60 dividuals with mild and moderate Alzheimer disease [22]. Our sublingual vitamin B12 tablets were obtained from bacteria with a 125
61 hypothesis is that the sublingual administrations of low (350 mg/ manufacturing process compatible with the strictly vegan dietary 126
62 week) and high (2000 mg/week) doses of cyanocobalamin are both requirements. The crystalline form of cyanocobalamin was used for 127
63 able to restore the nutritional adequacy of vitamin B12 within 90 the preparation of the tablets. 128
64 days [23e25] in vegans and vegetarians affected by a marginal Subjects were instructed to maintain their dietary and lifestyle 129
65 deficiency. habits as declared before enrollment. Moreover, they were 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 3/9

C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9 3

1 CONSORT 2010 Flow Diagram 66

2 67
3 Enrollment 68
Assessed for eligibility (n=194)
4 69
5 70
6 71
Excluded (n=154)
7 72
- Not meeting inclusion criteria (n=100)
8 73
- Declined to participate (n=40)
9 74
- Other reasons (n= 14)
10 75
11 76
12 77
13 Randomized (n=40) 78
14 79
15 80
16 81
17 Allocation 82
18 83
19 84
Allocated to intervention (n=20) Allocated to intervention (n=20)
20 85
Received allocated intervention (n=20) Received allocated intervention (n=20)
21 86
22 87
23 88
24 89
25 90
26 91
27 92
28 Lost to follow-up (for personal reasons) (n=2) Follow-Up Lost to follow-up (for personal reasons) (n=2) 93
29 94
30 95
31 96
32 97
33 Analysis 98
34 Analysed (n= 18) 99
Analysed (n= 18)
35 Excluded from analysis (n=0) 100
Excluded from analysis (n=0)
36 101
37 102
38 103
39 Fig. 1. Study flow-chart showing the process of patient selection and enrollment, allocation to the two study groups, and rate of patients completing the study. Ld: group treated 104
40 with low dosage of vitamin B12 (350 mg/week); Hd: group treated with high dosage of vitamin B12 (2000 mg/week). 105
41 106
42 107
43 encouraged to abstain from consuming sources of vitamin B12 (e.g., 2.4. Sampling and analysis of biochemical parameters 108
44 spirulin, yeast, fortified foods). A 24-h record of food consumption 109
45 was kept by each volunteer the day before blood collection to check Blood was collected in the morning by a phlebotomist. Samples 110
46 compliance with the dietary instructions. Every 2 weeks, subjects were drawn into evacuated tubes with or without K2EDTA. Serum 111
47 returned the empty blisters (as evidence of the consumption of the was separated within 1 h, while plasma was separated within 112
48 tablets) and received the new supplements. A 3-day food record 30 min (min) after collection by centrifugation (15 min at 2300
g 113
49 and a weekly direct interview were also scheduled randomly dur- at 4  C). Plasma and serum were aliquoted and stored at 80  C 114
50 ing the experimental period to check compliance with the dietary until analysis. All the samples were analyzed blind. Blood cell count 115
51 instructions and to assure the consumption of the tablets. The day was evaluated by routine laboratories assessment. 116
52 of the experiment, after an overnight fast, subjects reported to the Vitamins B12 levels were measured by a competitive test prin- 117
53 laboratories of the University of Milan. Blood samples were ciple using IF specific for this vitamin. Vitamin B12 was analyzed by 118
54 collected at baseline (time 0) and after 15, 30, 60, and 90 days of electrochemiluminescence immunoassay (ECLIA) using Cobas 119
55 intervention. immunoassay analyzers (Roche Diagnostics, North America). Also, 120
56 the assessment of serum folate was performed with electro- 121
57 2.3. Study variables chemiluminescence immunoassay (ECLIA) using Cobas immuno- 122
58 assay analyzers (Roche Diagnostics, North America). 123
59 The improvement of serum levels of vitamin B12 was considered Holotranscobalamin concentration were determined in serum 124
60 the primary endpoint. The other variables under study were as by immunoenzymatic assay kit (BIOHIT HealthCare, Helsinki, 125
61 follows: holotranscobalamin, methylmalonic acid, succinic acid, Finland). Briefly, the microtiter plate wells were coated with a 126
62 methionine, homocysteine, vitamin B6, folic acid, and complete highly specific monoclonal antibody for BIOHIT Active B12 (holoTC). 127
63 blood count. Since the amount of cobalt provided through the During the first incubation, holoTC specifically bound to the surface 128
64 supplement was negligible with respect to the circulating blood coated with the antibody. Successively, the conjugate was added for 129
65 levels, this variable was not evaluated. the binding of holoTC; the wells were then washed to remove 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 4/9

4 C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9

1 unbound components and holoTC was detected following the in- mean ± standard deviation (SD) baseline vitamin B12 concentration 66
2 cubation with the substrate. Before the analysis, a stop solution was was 140 ± 40 mmol/L and that the treatment would increase the 67
3 added and the absorbance was read at 405 nm (mod. F200 Infinite, levels of cyanocobalamin up to 240 mmol/L. This value represents 68
4 TECAN Milan, Italy). the mean found in an Italian blood donor population [4]. 69
5 Serum vitamin B6 concentrations were evaluated by high per- All analyses were performed using STATISTICA software (Stat- 70
6 formance liquid chromatography method using the relevant com- Soft Inc., Tulsa, OK, USA). Results are expressed as mean ± SD or 71
7 mercial kit (Chromsystems Instruments & Chemicals, Munich, standard error of the mean (SEM). Data were tested for normality of 72
8 Germany) [26]. Homocysteine (HCy), methionine (Met), methyl- distribution by the ShapiroeWilk test. Variables normally distrib- 73
9 malonic acid (MMA), succinic acid (SA), tris(2-carboxyethyl)phos- uted were analyzed by two-way analysis of variance (ANOVA) 74
10 phine hydrochloride (TCEP-HCl), methanol, and formic acid were considering the treatment (350 mg/week vs. 2000 mg/week) and the 75
11 obtained from SigmaeAldrich (St. Louis, MO, USA). Water was ob- time (0, 15, 30, 60, and 90 days) as dependent variables. Data that 76
12 tained from the Milli-Q apparatus (Millipore, Milford, MA). The were not normally distributed were logarithmically transformed. 77
13 determination of HCy, Met, MMA, and SA was performed according Log-transformed data were subjected to analysis by the non- 78
14 to Fu et al. [27], with slight modifications. Briefly, 200 mL of hepa- parametric Friedman test. Differences were considered significant 79
15 rinized plasma was added to 100 mL of water and 100 mL of TCEP- for p < 0.05; the least significant difference test was applied, as well 80
16 HCl (0.1 M). The mixture was vortexed for 10 s (s), incubated for as post hoc analysis, to show differences between treatments. The 81
17 15 min at room temperature, and transferred to an Amicon 10 K Da level of statistical significance was fixed at p < 0.05. 82
18 filter. The filter was centrifuged at 9000 g for 30 min, the filtrate 83
19 was transferred to a microvial, and 5 mL injected into the Ultra 3. Results 84
20 Performance Liquid Chromatography (UPLC)-high resolution (HR)- 85
21 mass spectrometers (MS). The analysis was carried out on an 3.1. Baseline characteristic of the study population 86
22 UHPLC model Acquity (Waters) coupled with a High-Resolution 87
23 Fourier Transform mass spectrometer (Orbitrap) model Exactive Baseline characteristics of the subjects enrolled in each group 88
24 (Thermo Scientific) equipped with an HESI-II probe for electrospray are reported in Table 1. Four subjects (2 for each group) were lost 89
25 ionization and a collision cell (HCD). The column was a 1.8 mm HSS during the follow-up period due to personal reasons. All subjects 90
26 T3 C18 (150
2.1 mm, Waters), flow rate was 0.45 mL/min, and the (n ¼ 36) showed a marginal deficiency of vitamin B12 (<220 pmol/ 91
27 eluents were 0.1% formic acid in water (A) and acetonitrile (B). The L) [3]. Regarding the other biomarkers of cobalamin status: 27 out 92
28 column and sample were kept at 60  C and 15  C, respectively. The of 36 subjects had serum levels of MMA above 750 nmol/L (cut-off 93
29 UHPLC separation was performed by the following linear elution above which cobalamin deficiency is diagnosed), while14 out of 36 94
30 gradient: 100% of A for 5 min, 0e100% B in 1 s, 100% B for 2 min, subjects documented moderate hyperhomocysteinemia (range 95
31 from 100% to 0% B in 1 min, and then isocratic for 2 min. 17.6e33.8 mmol/L) with plasma total homocysteine (HCy-pt) 96
32 For HCy and Met (0e3.2 min), the operative conditions were value  15 mmol/L [3]. Moreover, six subjects had folate levels 97
33 spray voltage þ3.0 kV, sheath gas flow rate 55, auxiliary gas flow (range 7e9 nmol/L) below 10 nmol/L, suggesting a folate deficiency 98
34 rate 20, capillary temperature 320  C, capillary þ47.5 V, tube [29]. Two subjects showed low vitamin B6 levels (<21.3 nmol/L) and 99
35 lens þ110 V, skimmer þ20 V, and heather temperature 120  C. The one also had low holotranscobalamin levels (<21 pmol/L) [3]. No 100
36 acquisition was performed in full-scan mode in the range (m/z)þ abnormalities in blood cell count were observed. The age, sex, he- 101
37 60e180 u. moglobin level, platelet and white blood cell counts, mean 102
38 For MMA and SA (3.2e5 min) the operative conditions were corpuscular volume, and serum cobalamin levels were not signifi- 103
39 spray voltage 3.0 kV, sheath gas flow rate 55, auxiliary gas flow rate cantly different between groups (Table 1). 104
40 20, capillary temperature 320  C, capillary 35 V, tube lens 70 V, 105
41 skimmer 16 V, and heather temperature 120  C. The acquisition 3.2. Compliance 106
42 was performed in full-scan mode in the range (m/z)- 60e130 u and 107
43 the ions with m/z 91.0038, corresponding to the formic acid dimer Subjects were highly motivated to participate in the interven- 108
44 [2M-H]- that was used as the lock mass. The isolation window, tion and confirmed the consumption of the tablets. The compliance 109
45 automatic gain control target, injection time, mass resolution, en- was verified during a weekly direct interview, as previously re- 110
46 ergy, and gas in the collision cell were ±2 ppm, 1
106, 100 ms, 50 K, ported, and confirmed by returning the empty blisters (100% 111
47 20 V, and N2, respectively. The MS data were processed using Xca- 112
48 libur software (Thermo Scientific). The peak identity was ascer- 113
49 tained, evaluating the accurate mass and the fragments obtained in Table 1 114
50 Subjects characteristics at the beginning of the study.a 115
the collision cell. Calibration curves were in the range 0.15e14.8,
51 0.13e33.5, 0.17e42.5, and 0.25e44 mMolar for HCy, Met, MMA, and Ld group Hd group P valueb 116
52 SA, respectively. Finally, the wellness parameter was calculated ac- Number of volunteers 18 18 e
117
53 cording to the Fedosov formula [28]: “wellness parameter”: Male/Female 9/9 9/9 e 118
54 w ¼ log10(holoTCn) þ log10(B12n)  log10(MMAn)  log10(HCyn), Age (years) 43 ± 12 42 ± 13 0.98 119
55 where concentrations are normalized (e.g., MMAn ¼ MMA/MMAn Weight (kg) 63.9 ± 11.5 68.7 ± 17.9 0.36 120
Body mass index (kg/m2) 21.6 ± 2.6 23.2 ± 5.4 0.29
56 normal). 121
Total Vitamin B12 (pmol/L) 146 ± 36 131 ± 56 0.29
57 Erythrocytes (10^6/mL) 4.6 ± 0.4 4.4 ± 0.3 0.06 122
58 2.5. Statistical analysis Mean corpuscular volume (fL) 88.5 ± 3.6 89.5 ± 4.6 0.32 123
59 White blood cells (10^3/mL) 5.3 ± 1.8 4.7 ± 0.8 0.17 124
Hemoglobin (g/dL) 13.7 ± 1.1 13.1 ± 1.0 0.07
60 Sample size was estimated, based on previous studies, in order to 125
Hematocrit (%) 40.9 ± 3.4 39.1 ± 2.7 0.06
61 detect significant differences in the serum vitamin B12 levels Platelets (10^3/mL) 219.4 ± 41.0 249.5 ± 56.7 0.09
126
62 [23e25]. Sixteen subjects per group were considered sufficient to a
127
Data are expressed as mean ± SD. Subjects were randomly assigned to 1 of the 2
63 demonstrate at least a 70% improvement in the levels of vitamin B12 128
groups (Ld vs Hd) and supplemented for 90 days. Ld ¼ low dosage; Hd ¼ high
64 after supplementation with a p value of 0.05 and a power of 80%. dosage. 129
65 The calculation was based on the assumptions that the b
P value derived by one way ANOVA. 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 5/9

C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9 5

1 66
2 67
3 68
4 69
5 70
6 71
7 72
8 73
9 74
10 75
11 76
12 77
13 78
14 79
15 80
16 81
17 82
18 83
19 84
20 Fig. 2. Effect of supplementation on serum circulating levels of total vitamin B12 in the two intervention groups (Ld vs Hd). The concentrations were measured at baseline (T0) and 85
21 after 15, 30, 60 and 90 days. N¼18 for each group. Data are expressed as mean ± SEM. a,b,c,d,eData with different letters are significantly different within the same treatment (time 86
effect; P < 0.05). *,x,#Data with different symbols are significantly different between treatment (treatment effect; P < 0.05).
22 87
23 88
24 compliance). Not one participant declared adverse effects following P < 0.0001). The values increased over time and appeared signifi- 89
25 the supplementation. cantly different between groups after 30 days until the end of the 90
26 experimental period (P < 0.01). Fig. 2A and B shows the levels of 91
27 3.3. Effect of supplementation on serum levels of total, active, and active (holotranscobalamin, HoloTC) (2A) and inactive forms (2B) of 92
28 inactive form of vitamin B12 vitamin B12 measured at baseline and after 15 and 90 days from the 93
29 start of supplementation. The analysis at 15 and 90 days was per- 94
30 The serum levels of total vitamin B12, measured at baseline (time formed based on the prominent absorption observed in vitamin 95
31 0 day) and after 15, 30, 60, and 90 days from the start of supple- B12. On the whole, ANOVA did not show a significant effect of 96
32 mentation, are reported in Fig. 1. Subjects increased the serum treatment and of time
treatment interaction, but revealed an effect 97
33 concentrations of total vitamin B12 to above 240 pmol/L according of time (P < 0.0001) for serum circulating levels of active and 98
34 to our hypothesis. On the whole, repeated measures of ANOVA did inactive vitamin B12 that increased during the treatments. 99
35 not show a significant effect of treatment, but revealed a significant 100
36 effect of time (P ¼ 0.008) and of time
treatment interaction 3.4. Effect of supplementation on serum levels of methylmalonic 101
37 (P ¼ 0.012) for circulating levels of total vitamin B12 that increased acid and homocysteine 102
38 following the treatments. In particular, post-hoc analysis showed a 103
39 significant enhancement after 15 days from the start of the intake of The serum levels of MMA and HCy were measured at baseline 104
40 the supplements (þ51.7% in Ld group vs. þ74.2% in Hd group; (time 0 day) and after 15, 30, 60, and 90 days from the start of 105
41 106
42 107
43 A) B) 108
44 109
160 250
45 110
c
46 111
47 140 c 112
48 200 113
pmol/L inactive vitamin B12

49 120 114
b
50 b 115
pmol/L holoTC

51 100 150 116

52 117
53 80 118
a a
54 100 119
55 60 120
56 121
57 40 Ld group Ld group 122
50
58 Hd group Hd group 123
59 20 124
60 125
0 0
61 126
T=0 T=15 T=90 T=0 T=15 T=90
62 127
Time Time
63 128
64 Fig. 3. Effect of supplementation on serum circulating levels of active (A) and inactive (B) form of vitamin B12 in the two intervention groups (LdvsHd). The concentrations were 129
65 measured at baseline (T0) and after 15 and 90 days from the supplementation. Data are expressed as mean ± SEM. N¼18 for each group. 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 6/9

6 C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9

1 A) B) 66
2 67
25
3 2.0 68
Ld group Ld group
4 69
5 a 70
Hd group Hd group
6 a 20 71
7 1.5 72
8 73
9 74

mmol/L tHcy
15
mmol/L MMA

b
10 75
11 c 76
1.0 b d
12 e 77
13 c 10 78
14 d 79
e
15 80
0.5
16 5 81
17 82
18 83
19 84
0.0 0
20 85
T=0 T=15 T=30 T=60 T=90 T=0 T=15 T=30 T=60 T=90
21 86
22 Time Time 87
23 88
Fig. 4. Effect of supplementation on serum circulating levels of MMA (A) and tHcy (B) in the two intervention groups (Ld vs Hd). The concentrations were measured at baseline (T0)
24 89
and after 15 and 90 days from the supplementation. N¼18 for each group. Data are expressed as mean ± SEM. MMA, methylmalonic acid; tHcy, total homocysteine.
25 90
26 91
27 supplementation, are reported in Fig. 3A and B. ANOVA revealed ANOVA did not show a significant effect of treatment, but revealed a 92
28 only a significant effect of time (P < 0.0001) for serum circulating significant effect of time (P < 0.0001) and time
treatment inter- 93
29 levels of MMA and HCy that decreased over time following both action (P ¼ 0.046). In particular, post-hoc analysis documented a 94
30 Q2 treatments (see Fig. 4). significant improvement over time following the intake of both the 95
31 supplements, with a difference between groups only at specific and 96
32 3.5. Effect of supplementation on serum concentrations of independent time points. 97
33 methionine, succinic acid, vitamin B6 and folate, blood cell count, No effect was documented for serum circulating levels of 98
34 and wellness parameter vitamin B6 and blood cell count (data not shown). 99
35 100
36 The serum levels of Met, SA, vitamin B6, and folate, measured at 4. Discussion 101
37 baseline (time 0 day) and after 15, 30, 60, and 90 days from the start 102
38 of supplementation, are reported in Table 2. ANOVA revealed only a In the present study, we documented that as a little as 350 mg 103
39 significant effect of time for serum circulating levels of folate per week of vitamin B12 supplementation was enough to correct a 104
40 (P < 0.0001), Met (P < 0.0001) and SA (P < 0.0001). In particular, marginal deficiency of cobalamin and to improve holoTC, MMA, 105
41 folate showed a significant decrease over time, while Met and SA and HCy (biomarkers of cobalamin status) in a group of vegans and 106
42 has significant increases. vegetarians. The results obtained support the use of a sublingual 107
43 In Table 2 are reported the values of the wellness parameter supplement at low doses as an effective and non-invasive method 108
44 measured at baseline (time 0 day) and after 15 and 90 days from the to improve the cobalamin status in this target population. 109
45 start of supplementation are reported in Table 2. Since the index It has been reported that the absorption of vitamin B12 from 110
46 derives from a formula that also takes into consideration the levels supplements does not depend only on the dose and frequency of 111
47 of holoTC, this parameter was measured only at times for which the the intake but also on the health status of the subjects. In particular, 112
48 levels of holoTC were detected. On the whole, repeated measures it is widely recognized that subjects suffering from gastric or small 113
49 114
50 115
Table 2
51 Effect of Vitamin B12 supplementation (low dosage versus high dosage) on serum levels of vitamin B6, folates, methionine, succinic acid and Wellness parameter (n ¼ 18)1. Q10,6
116
52 117
53 Variables Treatments T¼0 T ¼ 15 T ¼ 30 T ¼ 60 T ¼ 90 P treatment P time P interaction 118
54 Vitamin B6 nmol/ Ld 55.2 ± 19.2 67.4 ± 37.8 72.3 ± 42.3 69.6 ± 41.2 73.3 ± 42.1 0.76 0.07 0.65 119
55 Hd 61.6 ± 53.4 66.7 ± 38.2 62.6 ± 36.1 70.3 ± 45.4 60.2 ± 47.4 120
Folates nmol/L Ld 22.5 ± 8.8 20.8 ± 9.5 19.3 ± 10.0 18.3 ± 8.3 17.4 ± 10.1 0.23 <0.0001 0.43
56 121
Hd 19.6 ± 9.2 17.2 ± 6.7 18.5 ± 8.0 17.5 ± 7.9 16.1 ± 7.5
57 Methionine mmol/L Ld 17.7 ± 6.5 18.3 ± 6.9 18.0 ± 6.5 17.2 ± 5.2 17.8 ± 5.6 0.31 <0.0001 0.55 122
58 Hd 13.6 ± 3.9 15.6 ± 4.9 15.6 ± 4.8 15.5 ± 4.7 15.8 ± 4.6 123
59 Succinic acid mmol/L Ld 5.8 ± 3.6 6.2 ± 3.8 6.4 ± 3.9 6.7 ± 3.9 6.3 ± 3.6 0.43 <0.0001 0.11 124
Hd 4.3 ± 3.5 5.3 ± 3.4 5.6 ± 3.6 5.7 ± 3.5 5.9 ± 3.4
60 125
Wellness parameter Ld 1.0 ± 0.4a 0.3 ± 0.6b e e 0.2 ± 0.6c 0.88 <0.0001 0.046
61 Hd 1.3 ± 0.7a 0.2 ± 0.7b e e 0.3 ± 0.7c
126
62 127
The variables were measured at baseline (time 0) and after 15, 30, 60, 90 days from the supplementation. Ld, low dosage; Hd, high dosage.
63 128
P values correspond to the treatment, the time and the interaction between treatment and time in the overall two way ANOVA.
64 a,b,c
Data with different letters are significantly different between and within treatments. 129
65 1
Data are expressed as mean ± SD. 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 7/9

C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9 7

1 intestine resections, inflammatory bowel disease, and other com- cobalamin for assessment of B12status. MMA is considered a 66
2 plications related to intestinal absorption may become deficient biomarker of cobalamin function with regard to its role in the 67
3 [30]. Moreover, the capacity of absorption is strictly dependent on functioning of methylmalonyl-CoA mutase. Serum MMA concen- 68
4 saturable active transport and on the efficiency of the aspecific tration increases following an insufficient supply of cobalamin. As 69
5 route. In this regard, different studies have shown that the previously reported values above 750 nmol/L are used to discrim- 70
6 absorptive capacity of vitamin B12 is high when the amount inate a cobalamin deficiency [3]. 71
7 introduced is low. For example, the oral administration of different Plasma HCy is not a specific marker of cobalamin status since it 72
8 doses (1 mg, 10 mg, 50 mg, 500 mg, and 1000 mg) of vitamin B12 are is affected also by dietary factors, such as folate, choline and 73
9 absorbed with an efficiency of 56%, 16%, 3%, 2%, and 1.3%, respec- betaine, as well as renal insufficiency, lifestyle factors (e.g. alcohol 74
10 tively [31]. A plethora of studies investigated the effect of a sup- consumption) and age [41,42]. However, elevated plasma HCy 75
11 plementation on the levels of vitamin B12 and related concentration is commonly observed in subjects with a cobalamin 76
12 cardiovascular markers; however, most of them where performed deficiency. In our experimental conditions, most of the subjects 77
13 in the elderly [6], those with hyperhomocysteinemia [32,33], and showed baseline levels of MMA and HCy above the cut-off values, 78
14 undernourished children [34,35], while very few are involving while only few subjects showed low levels of folate. For these 79
15 vegetarians and/or vegans. A recent 12-week randomized, placebo- reasons, those biomarkers, together with the levels of folate, 80
16 controlled trial performed in vegans documented that the use of a vitamin B6, Met and SA, can be considered a valid support for the 81
17 vitamin B12-fortified toothpaste (about 100 mg/g depending on the assessment of the nutritional status of cobalamin in vegans and 82
18 number of brush sessions) improved serum and plasma concen- vegetarians. In fact, we were able to document a statistically sig- 83
19 trations of cobalamin and related associated markers [36]. Yajnik nificant decrease in the levels of MMA and HCy, and a significant 84
20 et al. [25] found that supplementation of vitamin B12 (500 mg/day), increase in the levels of Met and SA. These results were in line with 85
21 over a 6-week period, significantly increased plasma vitamin B12 those obtained by other authors showing a general improvement 86
22 concentration (from 125 to 215 pmol/L) in a group of healthy, lacto- after cobalamin supplementation [25,35,39,41]. An improvement in 87
23 vegetarian women. The improvement was observed within the first cobalamin nutritional status and a reduction of HCy and MMA may 88
24 2 weeks of intervention, and the levels maintained stability up to 4 be also effective in the prevention of cardiovascular risk and 89
25 weeks. Sharabi and coworkers documented similar findings neurological disorders. However, some studies failed to observe a 90
26 following sublingual and oral administration of 500 mg of cobal- significant modulation in HCy levels. For example, Sharabi and 91
27 amin in subjects with a deficiency [37]. colleagues [37] did not document a decrease in HCy and MMA 92
28 In our experimental conditions, supplementation with low and following 8 weeks of intervention with 500 mg/day of sublingual 93
29 high doses (350 mg/week vs. 2000 mg/week) of cobalamin signifi- and oral B12 administration in subjects with a cobalamin deficiency. 94
30 cantly improved circulating serum levels of vitamin B12, suggesting As previously reported, there is an interrelationship between 95
31 the efficiency and efficacy of both supplements in restoring the vitamin B12 and folate; in particular, vitamin B12 deficiency can lead 96
32 levels of the vitamin (>240 pmol/L) [3]. However, serum levels of to lowered levels of methionine synthetase, which results in folate 97
33 vitamin B12 above the cut-off point does not necessarily indicate an deficiency and an increased proportion of the 5-methyl derivative. 98
34 adequate nutritional status. In fact, there is inconsistency among In our experimental conditions, we did not quantify the levels of 99
35 the scientific community regarding the identification of reference the 5-methyl derivative, but only folate that significantly reduced 100
36 values for cyanocobalamin. Future studies should be performed in following cobalamin supplementation. These results are complex 101
37 order to identify the cut-offs according to individual variability (i.e., to explain; we may hypothesize that the improvement in B12 status, 102
38 age, sex, etc.) and lifestyle habits (i.e., vegans, vegetarians). Holo- also in terms of MMA and HCy, did not require high amounts of 103
39 transcobalamin represents the metabolically active form of vitamin folate to compensate for a cobalamin deficiency. However, we 104
40 B12 that delivers cobalamin to the target cells. Recently, it has been cannot exclude that these fluctuations were attributed mainly to 105
41 recognized as an early and reliable marker to discriminate an physiological changes, since the overall vitamin status was main- 106
42 impaired cobalamin status [38]. However, discrepancies remain tained within the range of normality. 107
43 about mode of application and assignment of these cut-off values to A recent and robust biochemical indicator of cyanocobalamin 108
44 diagnose a deficiency. Based on different populations and criteria, status is the wellness parameter conceived by Fedosov that takes 109
45 cut-off values from 21 to 45 pmol/L have been proposed as “sub- into consideration the levels of total and active B12 forms and those 110
46 optimal” [3]. In our study, subjects have shown levels of holoTC of MMA and HCy [28]. The cut-off to discriminate the wellness 111
47 within the range of normality. This is in line with the characteristics parameter are as follows: deficiency w ¼ 1.49; transition 112
48 of our population that included only individuals with a marginal w ¼ 0.516; normal w ¼ 0.0, and excellent w ¼ þ0.445. In our 113
49 cobalamin deficiency. The supplementation with both dosages experimental conditions, subjects showed a low wellness param- 114
50 significantly increased the levels of holoTC. The improvement was eter at baseline (1.0 for Ld group and 1.3 for Hd group), doc- 115
51 comparable between groups, since only an effect of time, but not of umenting a state of marginal deficiency. The supplementation of 116
52 treatments, was observed. The impact of vitamin B12 supplemen- vitamin B12 significantly improved the wellness parameter in both 117
53 tation on levels of holoTC has been evaluated in different studies the intervention groups. 118
54 [39,40]. In a double-blind, placebo-controlled trial, 12 and 24 weeks Finally, we observed no significant effect on blood cell count 119
55 of supplementation with 1000 mg vitamin B12 or 1000 mg vitamin both at the beginning of the study (see Table 1) and after the 120
56 B12 þ 400 mg folic acid significantly increased the levels of cobal- intervention (data not shown). These results are not surprising, 121
57 amin as well as those of holoTC in elderly subjects [39]. Brito et al. since our subjects were in stage 2e3 of vitamin B12 deficiency and 122
58 [40], reported that a single intramuscular injection of 10 mg this condition does not affect the levels of mean corpuscular vol- 123
59 vitamin B12 (providing 100 mg pyridoxine and 100 mg thiamine) ume and hemoglobin [42]. 124
60 significantly increased, after 4 months, serum vitamin B12 and 125
61 holotranscobalamin levels in a group of 27 community-dwelling 5. Study limitations 126
62 elderly Chileans. 127
63 Other biomarkers of cobalamin status include hematological A possible limitation of the study is the lack of a real control 128
64 changes and the metabolites MMA and HCy. These variables can group (vegans/vegetarians with a marginal deficiency who did not 129
65 add valuable information in conjunction with serum holoTC and/or take supplements). However, by considering that our subjects were 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 8/9

8 C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9

1 affected by a marginal vitamin B12 deficiency, the inclusion of a real References 66

2 placebo group (vegans/vegetarians without supplements) would 67
3 not have been possible for ethical reasons. A second limitation of [1] O'Leary F, Samman S. Vitamin B12 in health and disease. Nutrients 2010;2: 68
299e316.
4 the study is the lack of a follow-up period post-supplementation in [2] Pawlak R, Parrott SJ, Raj S, Cullum-Dugan D, Lucus D. How prevalent is vitamin 69
5 order to verify the changes in the levels of vitamin B12 and related B12deficiency among vegetarians? Nutr Rev 2013;71:110e7. 70
6 metabolic markers along the time. [3] Scientific opinion on dietary reference values for cobalamin (vitamin B12). 71
EFSA panel on dietetic products, nutrition, and allergies (NDA). EFSA Journal
7 2015;13:4150.
72
8 6. Conclusions [4] Societ
a Italiana di Nutrizione Umana, SINU. Livelli di Assunzione di Rifer- 73
9 imento di Nutrienti ed energia per la popolazione italiana, IV revisione. 74
Milano: SICS Editore; 2014.
10 In conclusion, the results obtained have shown that both sup- 75
[5] Stabler SP. Clinical practice. Vitamin B12 deficiency. N Engl J Med
11 plements were able to bring the levels of vitamin B12 from a mar- 2013;10(368):149e60. 76
12 [6] Andre
s E, Affenberger S, Vinzio S, Kurtz JE, Noel E, Kaltenbach G, et al. Food- 77
ginal deficiency to an adequate nutritional status. In particular, we cobalamin malabsorption in elderly patients: clinical manifestations and
13 have documented an increase of serum concentrations of vitamin 78
treatment. Am J Med 2005;118:1154e9. Q4
14 B12 and holoTC, and a reduction of MMA and HCy as markers of [7] Pawlak R, Lester SE, Babatunde T. The prevalence of cobalamin deficiency 79
15 vitamin B12 metabolism. These results are in line with the elevation among vegetarians assessed by serum vitamin B12: a review of literature. Eur J 80
16 Clin Nutr 2014;68:541e8. 81
of the wellness parameter that provides further support for the [8] Gilsing AMJ, Crowe FL, Lloyd-Wright Z, Sanders TAB, Appleby PN, Allen NE,
17 improvement of the nutritional vitamin B12 status. et al. Serum concentrations of vitamin B12 and folate in British male omni- 82
18 Our observations emphasize the importance of supplementation vores, vegetarians, and vegans: results from a cross-sectional analysis of the 83
19 EPIC-Oxford cohort study. Eur J Clin Nutr 2010;64:933e9. 84
in vegetarians and vegans with a marginal deficiency, but it should [9] Majchrzak D, Singer I, Ma €nner M, Rust P, Genser D, Wagner KH, et al. Vitamin
20 be emphasized that the use of pharmacological doses is unnec- status and concentrations of homocysteine in Austrian omnivores, vegetarians 85
21 essary in this target group. Moreover, the absence of a consensus on and vegans. Ann Nutr Metab 2006;50:485e91. 86
22 [10] Woo KS, Kwok TCY, Celermajer DS. Vegan diet, subnormal vitamin B12 status 87
vitamin B12 cut-off values and the high individual variability make it and cardiovascular health. Nutrients 2014;6:3259e73.
23 difficult to identify the real needs for vegans and vegetarians. [11] Obersby D, Chappell DC, Dunnett A, Tsiami AA. Plasma total homocysteine
88
24 Further studies are necessary in order to confirm our findings and status of vegetarians compared with omnivores: a systematic review and 89
25 meta-analysis. Br J Nutr 2013;109:785e94. 90
verify the effects of sublingual supplementation in vegans and [12] Pawlak R. Is vitamin B12 deficiency a risk factor for cardiovascular disease in
26 vegetarians with a severe deficiency and in those affected by 91
vegetarians? Am J Prev Med 2015;48:e11e26.
27 malabsorption and/or impaired metabolism of vitamin B12. [13] Ma Y, Peng D, Liu C, Huang C, Luo J. Serum high concentrations of homo- 92
28 cysteine and low levels of folic acid and vitamin B12are significantly correlated 93
with the categories of coronary artery diseases. BMC Cardiovasc Disord
29 94
Funding 2017;17(1):37.
30 [14] Mahalle N, Kulkarni MV, Garg MK, Naik SS. Vitamin B12 deficiency and 95
31 hyperhomocysteinemia as correlates of cardiovascular risk factors in Indian 96
32 The study was supported by the Phoenix Srl and by intramural subjects with coronary artery disease. J Cardiol 2013;61:289e94. 97
funding. [15] Dali-Youcef N, Andre
s E. An update on cobalamin deficiency in adults. QJM
33 2009;102:17e28. 98
34 [16] Kotilea K, Quennery S, Decroe €s V, Hermans DA. Successful sublingual cobal- 99
35 Author disclosure amin treatment in a child with short-bowel syndrome. J Pediatr Pharmacol 100
Ther 2014;19:60e3.
36 [17] De Benoist B. Conclusion of WHO Technical Consultation on folate and 101
37 Author disclosures: CDB, PR, CG, ABr, Aba and SC declared no vitamin B12 deficiencies. Food Nutr Bull 2008;29:S238e44. 102
38 conflicts of interest. SC is responsible of the grant funding obtained [18] Allen LH. How common is vitamin B12 deficiency? Am J Clin Nutr 2009;89: 103
693Se6S.
39 by Phoenix Srl. The funding source had no role in the study design, 104
[19] Masucci L, Goeree R. Vitamin B12intramuscolar injections versus oral sup-
40 conduct, or interpretation and reporting. plements: a budget impact analysis. Ont Health Technol Assess Ser 2013;13: 105
41 1e24. 106
[20] Delpre G, Stark P, Niv Y. Sublingual therapy for cobalamin deficiency as an
42 Conflict of interest 107
alternative to oral and parenteral cobalamin supplementation. Lancet
43 1999;354:740e1. 108
44 Q3 None. [21] Carmel R. How I treat cobalamin (vitamin B12) deficiency. Blood 2008;112: 109
2214e21.
45 110
[22] Aisen PS, Schneider LS, Sano M, Diaz-Arrastia R, van Dyck CH, Weiner MF,
46 et al. Alzheimer Disease Cooperative Study. High-dose B vitamin supple- 111
Uncited reference
47 mentation and cognitive decline in Alzheimer disease: a randomized 112
48 controlled trial. JAMA 2008;300:1774e83. 113
Q7 [43]. [23] Dhonukshe-Rutten RA, van Zutphen M, de Groot LC, Eussen SJ, Blom HJ, van
49 Staveren WA. Effect of supplementation with cobalamin carried either by a 114
50 milk product or a capsule in mildly cobalamin-deficient elderly Dutch per- 115
51 Acknowledgments sons. Am J Clin Nutr 2005;82:568e74. 116
[24] Wouters-Wesseling W, Wouters AE, Kleijer CN, Bindels JG, de Groot CP, van
52 Staveren WA. Study of the effect of a liquid nutrition supplement on the 117
53 We are grateful to the staff of the International Center for the nutritional status of psycho-geriatric nursing home patients. Eur J Clin Nutr 118
54 Assessment of Nutritional Status (ICANS) of the University of Milan 2002;56:245e51. 119
for the support in the medical examination of the volunteers and [25] Yajnik CS, Lubree HG, Thuse NV, Ramdas LV, Deshpande SS, Deshpande VU,
55 et al. Oral vitamin B12 supplementation reduces plasma total homocysteine
120
56 for the management of the blood sampling. We warm thank Dr. concentration in women in India. Asia Pac J Clin Nutr 2007;16:103e9. 121
57 Michela De Petris and Dr. Luciana Baroni for their precious collab- [26] De Giuseppe R, Venturelli G, Guez S, Salera S, De Vita C, Consonni D, et al. 122
oration in the enrollment of the volunteers. We are grateful to Dr. Homocysteinemetabolism in children and adolescents with epi-
58 123
dermolysisbullosa. BMC Pediatr 2016;16:173.
59 Licia Colombo for the support in providing supplements. We also [27] Fu X, Xu YK, Chan P, Pattengale PK. Simple, fast, and simultaneous 124
60 thank Dr. Arianna Levi and Dr. Camilla Vergnaghi for their support detection of plasma total homocysteine, methylmalonic acid, methionine, 125
61 during the experimental period. Finally, we are grateful to all the and 2-methylcitric acid using liquid chromatography and mass spectrom-
126
etry (LC/MS/MS). JIMD Rep 2013;10:69e78.
62 volunteers for their time and commitment. 127
[28] Fedosov SN. Metabolic signs of vitamin B12 deficiency in humans: computa-
63 All authors provided input into and read and approved the final tional model and its implications for diagnostics. Metabolism 2010;59: 128
64 version of the manuscript. 1124e38. 129
65 130

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008
 YCLNU3389_proof ■ 19 February 2018 ■ 9/9

C. Del Bo' et al. / Clinical Nutrition xxx (2018) 1e9 9

1 [29] Dhonukshe-Rutten RA, de Vries JH, de Bree A, van der Put N, van Staveren WA, [36] Siebert AK, Obeid R, Weder S, Awwad HM, Sputtek A, Geisel J, et al. Vitamin B- 21
de Groot LC. Dietary intake and status of folate and vitamin B12 and their 12-fortified toothpaste improves vitamin status in vegans: a 12-wk ran-
2 22
association with homocysteine and cardiovascular disease in European pop- domized placebo-controlled study. Am J Clin Nutr 2017;105:618e25.
3 ulations. Eur J Clin Nutr 2009;63:18e30. [37] Sharabi A, Cohen E, Sulkes J, Garty M. Replacement therapy for vitamin 23
4 [30] Langan RC, Goodbred AJ. Vitamin B12 deficiency: recognition and manage- B12deficiency: comparison between the sublingual and oral route. Br J Clin 24
5 ment. Am Fam Physician 2017;96:384e9. Pharmacol 2003;56:635e8. 25
[31] Rizzo G, Lagan
a AS, Rapisarda AM, La Ferrera GM, Buscema M, Rossetti P, et al. [38] Nexo E, Hoffmann-Lücke E. Holotranscobalamin, a marker of vitamin B-12
6 Vitamin B12 among vegetarians: status, assessment and supplementation. status: analytical aspects and clinical utility. Am J Clin Nutr 2011;94: 26
7 Nutrients 2016;8. E767. 359Se65S. 27
8 [32] van Dijk SC, Enneman AW, Swart KM, van Wijngaarden JP, Ham AC, de [39] Eussen SJ, de Groot LC, Joosten LW, Bloo RJ, Clarke R, Ueland PM, et al. Effect of 28
Jonge R, et al. Effect of vitamin B12 and folic acid supplementation on bio- oral vitamin B-12 with or without folic acid on cognitive function in older
9 markers of endothelial function and inflammation among elderly individuals people with mild vitamin B-12 deficiency: a randomized, placebo-controlled
29
10 with hyperhomocysteinemia. Vasc Med 2016;21:91e8. trial. Am J Clin Nutr 2006;84:361e70. 30
11 [33] van Dijk SC, Enneman AW, Swart KM, van Wijngaarden JP, Ham AC, Brouwer- [40] Brito A, Grapov D, Fahrmann J, Harvey D, Green R, Miller JW, et al. The human 31
Brolsma EM, et al. Effects of 2-year vitamin B12 and folic acid supplementation serum metabolome of vitamin B-12 deficiency and repletion, and associations
12 in hyperhomocysteinemic elderly on arterial stiffness and cardiovascular with neurological function in elderly adults. J Nutr 2017. pii: jn248278.
32
Q5
13 outcomes within the B-PROOF trial. J Hypertens 2015;33:1897e906. [41] Ganguly Paul. Sreyoshi Fatima Alam. Role of homocysteine in the develop- 33
14 [34] Bahadir A, Reis PG, Erduran E. Oral vitamin B12 treatment is effective for ment of cardiovascular disease. Nutr J 2015;14:6. 34
children with nutritional vitamin B12 deficiency. J Paediatr Child Health [42] Devi S, Mukhopadhyay A, Dwarkanath P, Thomas T, Crasta J, Thomas A, et al.
15 35
2014;50:721e5. Combined vitamin B-12 and balanced protein-energy supplementation affect
16 [35] Siega-Riz AM, Estrada Del Campo Y, Kinlaw A, Reinhart GA, Allen LH, Shahab- homocysteine remethylation in the methionine cycle in pregnant south In- 36
17 Ferdows S, et al. Effect of supplementation with a lipid-based nutrient sup- dian women of low vitamin B-12 status. J Nutr 2017;147:1094e103. 37
plement on the micronutrient status of children aged 6-18 months living in the [43] Herrmann W, Geisel J. Vegetarian lifestyle and monitoring of vitamin B-12
18 38
. Honduras. Paediatr Perinat Epidemiol 2014;28:245e54.
rural region of Intibuca status. Clin Chim Acta 2002;326:47e59.
19 39
20 40

Please cite this article in press as: Del Bo' C, et al., Effect of two different sublingual dosages of vitamin B12 on cobalamin nutritional status in
vegans and vegetarians with a marginal deficiency: A randomized controlled trial, Clinical Nutrition (2018), https://doi.org/10.1016/
j.clnu.2018.02.008