Vitamin B₁₂ Status in Children with Cystic Fibrosis and Pancreatic Insufficiency

Abstract

Objective

Unexpectedly high serum B₁₂ concentrations were noted in most study subjects with cystic fibrosis (CF) and pancreatic insufficiency (PI) participating in a nutrition intervention at the baseline evaluation. The objectives of this study were to determine dietary, supplement-based and enzyme-based B₁₂ intake, serum B₁₂ concentrations, and predictors of vitamin B₁₂ status in children with CF and PI.

Study Design

Serum B₁₂ status was assessed in subjects (5-18 yrs) and categorized as elevated (Hi-B₁₂) or within reference range (RR-B₁₂) for age and sex. Serum homocysteine, plasma B_6, red blood cell folate, height, weight, and body mass index Z scores, pulmonary function, energy, dietary and supplement-based vitamin intake were assessed.

Results

106 subjects, mean age 10.4 ± 3.0 years participated. Median serum B₁₂ was 1083 pg/ml, with 56% in the Hi-B₁₂ group. Dietary and supplement-based B₁₂ intake were both high representing 376% and 667% Recommended Dietary Allowance (RDA). The Hi-B₁₂ group had significantly greater supplement-based B₁₂ intake than the RR-B₁₂ group (1000 vs. 583% RDA, p<0.001). By multiple logistic regression analysis, high supplement-based B₁₂ intake and age >12 years increased risk for Hi-B₁₂, while higher FEV₁ decreased risk (Pseudo-R²=0.18, P<0.001).

Conclusions

Serum B₁₂ was elevated in the majority of children with CF and PI. Supplement-based B₁₂ intake was 6 to 10 times the RDA, and strongly predicted elevated serum B₁₂ status. The health consequences of lifelong high supplement-based B₁₂ intake and high serum B₁₂ are unknown and require further study, as does the inversed correlation between serum B₁₂ and FEV₁.

INTRODUCTION

While fat soluble vitamin status is frequently described in patients with cystic fibrosis (CF) and pancreatic insufficiency (PI), less is known regarding B vitamins. Vitamin B₁₂ (cobalamin) metabolism is a complex process that begins in the oropharynx and is completed in the terminal ileum¹. This process requires an intact gastrointestinal tract, acidification of gastric contents, pancreatic function for alkalinization of duodenal fluid and the provision of trypsin to degrade the R-binder protein and allow B₁₂ to bind to intrinsic factor (IF). The IF-B₁₂ complex is actively absorbed at the terminal ileum¹. There are possible CF and PI- related perturbations in B₁₂ metabolism, specifically impaired trypsin and bicarbonate production^{2, 3}. Approximately 85% of the CF population has PI⁴ and are at risk for malabsorption. Pancreatic enzyme replacement therapy (PERT) partially corrects B₁₂ malabsorption and improves B₁₂ concentrations^5-7 .

Unexpectedly high serum B₁₂ concentrations were noted in most of the study subjects participating in a nutrition intervention trial in children with CF and PI. The objectives of this current report were to determine dietary, vitamin supplement-based and pancreatic enzyme-based B₁₂ intake, serum B₁₂ concentrations, and predictors of vitamin B₁₂ status in children with CF and PI, to try to identify etiology of elevated serum B₁₂.

METHODS

Children with CF and PI ages 5.0 to 17.9 were recruited from ten CF Centers from March 2007 to May 2010 to participate in a randomized placebo-controlled double-blind longitudinal trial to evaluate the impact of a nutritional supplement, LYM-X-SORB™, on nutritional and growth status. LYM-X-SORB™ is an lipid matrix is composed of lysophosphatidyl choline, triglycerides and essential fatty acids that has been previously been shown to be absorbable without pancreatic enzyme therapy (PERT) and to improve nutritional status and clinical outcomes for children with CF and PI⁸. Vitamin B₁₂, B₆ (pyridoxine) and folate status were assessed as part of this larger study due to their role as co-factors in the phospholipid and 1-methyl donor metabolism. Here we report on these B vitamins at the baseline pre-intervention visit. The diagnoses of CF and PI were made by the home CF Center based on clinical symptoms, and duplicate quantitative pilocarpine iontophoresis sweat test and/or by genotype. In all study subjects PI was confirmed by fecal elastase testing (Genova Diagnostics, Asheville, NC), with < 15μg/g stool required for enrollment. Children were excluded if they had an FEV₁ <40% predicted, significant liver disease, history of meconium ileus with significant ileal resection or known current ileal disease, insulin dependent diabetes, or other conditions known to affect growth. The study protocol was approved by the Committee for the Protection of Human Subjects of the Institutional Review Board at the Children’s Hospital of Philadelphia (CHOP), and at the subject’s home institution. Written informed consent and age-appropriate assent were obtained from the parent/legal guardian and each subject, respectively. Subjects with CF were seen at the CHOP Clinical Translational Research Center (CTRC) for overnight admissions while in their usual state of good health. Evaluations included measures of growth and nutritional status, serum vitamin status and other biomarkers, clinical status, spirometry, medication use (including pancreatic enzyme replacement therapy, acid reduction agents as well as antibiotics), dietary and supplemental vitamin intake.

Energy and Nutritional Intake

Dietary intake was assessed with 3-day home-based weighed food records. Subjects and parents/guardians were provided with measuring cups, spoons, and digital food scales after they were given detailed verbal instructions to weigh and record each food/beverage consumed. Diet records were reviewed and analyzed (Nutrition Data System for Research, National Coordinating Center, University of Minnesota, Minneapolis, MN)⁹. Energy intake was assessed using the Dietary Reference Intakes (DRI)¹⁰, and expressed as percent Estimated Energy Requirement (%EER) for active children, as previously determined for children with CF and PI¹¹. Dietary intake of B₁₂, B₆, and folate are expressed as percent Recommended Dietary Allowance (%RDA)^12,13.

Vitamin Intake

Details of the specific brands of vitamin, mineral and caloric supplemental intake, frequency and dose were recorded from study participant report. The supplement-based vitamin product content was verified from the manufacturers’ information. Supplemental intakes for vitamins B₁₂, B₆ and folic acid were expressed as both intake in μg/d or mg/d and as %RDA^{12, 13}. PERT use, including brand, number of capsules and lipase units taken per day were determined by study participant report. The vitamin B₁₂ content of selected PERT products, including Pancreaze™ (11,500 lipase units/capsule; Janssen Pharmaceuticals), Creon™ (11,500 lipase units/capsule; Abbott Laboratories), and Zenpep™ (10,000 lipase units/capsule; Aptalis Pharma) were determined as μg/100g product by the turbidimetric method (Eurofins, Inc, Des Moines, IA).

Vitamin B Status and Biochemistry

Fasting blood samples were collected at baseline and serum vitamin B₁₂ was analyzed by quantitative chemoluminescent immunoassay (CHOP Clinical Laboratory, Philadelphia, PA). Plasma B₆ (pyridoxine-5-phosphate) was determined by high performance liquid chromatography (ARUP Laboratory, Salt Lake City, UT). Whole blood was collected for red blood cell folate concentrations by chemoluminescence (ARUP Laboratory, Salt Lake City, UT) and reported correcting for the hematocrit. For serum B₁₂, subjects were designated as having elevated B₁₂ (Hi- B₁₂) if their B₁₂ concentrations were above versus within the established age- and sex-specific clinical laboratory reference ranges (RR-B₁₂)^{14, 15}. Supplement-based B₁₂ in CF specific and in over the counter vitamin products was in the form of cyanocobalamin (i.e., complexed to cyanide). When it became evident that specifically supplement- based B₁₂ intake was very high, whole blood cyanide concentrations were determined by quantitative colorimetry on a limited subset of study subjects (ARUP Laboratory, Salt Lake City, UT) at the 12 month study visit. Potential cyanide toxicity is associated with blood concentrations > 100 μg/dL¹⁶. CBC with differential, hepatic function panels, and comprehensive metabolic panels were assessed by standard methods; homocysteine, methionine and cysteine were assessed by high performance liquid chromatography (CHOP Clinical Laboratory). High sensitivity C-reactive protein (hsCRP) was assessed by quantitative immunoturbidimetry (ARUP Laboratories, Salt Lake City, UT).

Body Composition, Growth and Clinical Status

Height and weight were measured using standard techniques¹⁷ with a stadiometer accurate to 0.1 cm (Holtain, Crymych, UK), and a digital electronic scale accurate to 0.1 kg (Scaletronix, White Plains, NY). Height was adjusted for genetic potential¹⁸ from measured or reported biological parent heights. Z scores for height (HAZ), mid-parent adjusted height (adjHAZ), weight (WAZ), and body mass index (BMI; kg/m², BMIZ) were computed¹⁹. Pulmonary function was evaluated by standard methods²⁰ by spirometry (Medical Graphics Corporation, Minneapolis, MN). FEV₁ percent predicted (FEV₁%) was calculated using Wang²¹ and Hankinson²² equations recommended for use in children and adolescents with CF and used as the measure of pulmonary function.

Statistical Methods

Measures of central tendency and variability were calculated for each outcome. Descriptive statistics were calculated for continuous variables using means and standard deviations or medians and ranges as appropriate for normally distributed or skewed data. Frequency distributions were used for categorical variables. Significant differences between Hi-B₁₂ versus RR-B₁₂ groups were assessed for continuous variables using unpaired student’s t tests for normally distributed and Mann-Whitney tests for skewed data, and for categorical variables using chi-squared tests. Four categories for B₁₂ supplemental intake were constructed representing low to high B₁₂ supplemental intake as %RDA (Category 1=0-499, 2=500-999, 3=100-1499, and 4= ≥1500 %RDA). Children were also divided into younger (≤12 yrs) and older (>12 yrs) groups. Potential predictors of elevated B₁₂ status were identified as age group, sex, genotype, BMIZ, FEV₁ % predicted, serum B₆ and B₁₂ supplemental intake categories (%RDA). These potential predictors were then tested as both individual predictors in logistic models and also using multiple logistic regression analyses, with Hi-B₁₂ vs. RR-B₁₂ serum status as the dichotomous outcome. Best predictor multiple logistic models for B₁₂ status included only those variables that entered the model at p<0.10 level of significance. All statistical analyses were performed using STATA 12.0 (College Station, TX), and results were considered significant at p<0.05.

RESULTS

Serum B₁₂ was assessed for 106 of the 110 subjects (96%) enrolled in the study; median B₁₂ was 1083 pg/ml, and 56% and 44% of the subjects had Hi-B₁₂ and RR-B₁₂, respectively. There were no children with serum B₁₂ below the age- and sex-specific reference ranges. Baseline demographic, growth and clinical status are presented in Table 1.Serum vitamin status, dietary and supplemental intake for energy, B₁₂, B₆ and folic acid are presented in Table 2 for the entire sample and for subjects in the Hi- B₁₂ and RR-B₁₂ groups. Overall, children were 10.4±3.0 years of age, 72% were <12 yrs and 57% were male. CF genotype was available for 104 subjects, with 59% classified as ΔF508 homozygous, 34% heterozygous (ΔF508/other) and 7% subjects with other CF genotypes. Growth status was suboptimal in these children with mild lung disease. In contrast to serum B₁₂ status, serum B₆ and RBC folate concentrations were within age-and sex-specific reference ranges for the majority of the sample (85% and 94%, respectively). No children were below reference ranges for B₆ and 4% were below for RBC folate concentrations. With respect to hepatic enzyme assessments, AST was mildly above reference range in 28%, ALT in 43%, and GGT in 34% of subjects, and were not associated with B₁₂ status. Dietary and supplemental intake of energy and the B vitamins are also presented in Table 2 for 95 children who completed the 3-day weighed food. Overall, energy intake was higher in subjects with CF (115% EER) than the estimated requirement for healthy active children. Dietary and vitamin supplement-based B₁₂ intake were both high representing 376% and 667% RDA, respectively. Total intake of B₁₂ was 18.4 μg/d, corresponding to 1175 %RDA, or nearly 12 times the recommended intake, primarily due to the high CF specific vitamin supplemental B₁₂ intake. B₆ intake was high, 5 to 6 times the RDA, with majority from vitamin supplement-based intake. Folic acid intake was 2 to 3 times the RDA, and was evenly distributed between dietary and vitamin supplement-based intake.

Table 1. Characteristics of Children with CF and PI by B₁₂ Status Groups.

	Serum B12 Groups
	All	Hi-B12	RR-B12	P-value1
Number	106	59	47
Age, yr	10.4 ± 3.0	11.0 ± 3.3	9.6 ± 2.6	0.02
Age > 12y, %	28	39	15	0.01
Sex, % male	57	51	64	0.18
WAZ	−0.38 ± 0.78	−0.48 ± 0.68	−0.24 ± 0.89	0.11
HAZ	−0.39 ± 0.92	−0.44 ±0.97	−0.32 ± 0.86	0.51
Adj HAZ 2	−0.68 ± 0.89	−0.73 ± 0.92	−0.61 ± 0.85	0.49
BMIZ	−0.19 ± 0.77	−0.32 ± 0.68	−0.03 ± 0.85	0.05
FEV1, % predicted 2	96 ± 23	90 ± 23	103 ± 20	0.005

HAZ, height-for-age Z score; adjHAZ, height-for age Z score adjusted for mid-parental height; WAZ, weight-for-age Z score; BMIZ, body mass index-for-age Z score;

¹Means ± SD and t-tests for normally distributed variables.

²n=103 (n=58 for Hi-B₁₂, n=45 for RR - B₁₂).

Table 2. Blood Biomarkers and Dietary Intake by B₁₂ Status Group.

	Serum B12 Groups
	All	Hi-B12	RR-B12	P-value1
Blood Biomarker
B12, pg/mLA	1083 (371, 3455)	1414 (823, 3455)	853 (371, 1284)	<0.001
B6, ng/mL2,A	16.9 (5.0, 91.7)	17.6 (5.0, 91.7)	15.4 (5.4, 39.4)	0.05
RBC folate, ng/mL3,A	496 (231, 1045)	509 (231, 857)	488 (278, 1045)	0.95
Homocysteine, μmol/LA	4.8 (2.2, 9.0)	4.7 (2.2, 8.9)	5.1 (2.8, 9.0)	0.01
Methionine, nmol/mLA	24.9 (11.1, 65.7)	23.5 (16.1, 65.7)	25.9 (11.1, 56.3)	0.46
Cysteine, nmol/mL	43.7 (12.4, 79.0)	46.5 (12.4, 66.3)	41.4 (28, 79)	0.01
hsCRP, mg/L 4	0.4 (0.1, 48.4)	0.4 (0.1, 48.4)	0.4 (0.1, 5.5)	0.61
WBC, thou/μL 4	7.1 (4.1, 17.8)	7.7 (4.5, 17.8)	6.9 (4.1, 16.7)	0.24
Dietary Intake
Number	95	54	41
Energy, kcal	2325 (865, 4909)	2359 (865, 4909)	2320 (1413, 4014)	0.94
%EER	115 (41, 228)	114 (41, 223)	117 (64, 228)	0.49
Total B12, μg	18.4 (5.5,158.9)	21.9 (8.0, 158.9)	15.1 (5.3, 33.3)	<0.001
Total B12, %RDA	1175 (310, 8826)	1391 (479, 8826)	1000 (310, 1872)	<0.001
Diet , μg	6.0 (1.5, 18.1)	6.4 (2.6, 18.1)	5.6 (1.5, 14.0)	0.17
Diet, %RDA	376 (79, 1085)	379 (79, 1085)	373 (83, 780)	0.80
Supplement, %RDA	667 ( 0, 8333)	1000 (250, 8333)	583 (0,1375)	<0.001
0 - 499 %RDA	15	11	20
500 - 999 %RDA	47	35	63
1000 - 1499 RDA	29	39	17
1500+ %RDA	8	15	0	0.002
Total B6, mg	4.6 (1.9, 78.2)	5.5 (1.9, 78.3)	3.9 (2.4, 8.3)	<0.001
Total B6, %RDA	545 (243, 7827)	591 (243, 7827)	502 (247, 1131)	0.01
Diet, mg	1.9 (0.5, 8.8)	1.9 (0.7, 8.8)	1.8 (0.5, 5.6)	0.13
Diet, %RDA	217 (51, 931)	219 (59, 881)	214 (51, 931)	0.50
Supplement, mg/d	2.3 (0.0, 75.0)	3.0 (0.6, 75.0)	1.9 (0.0, 4.0)	0.01
Supplement, %RDA	317 (0, 7500)	317 (105, 7500)	300 (0, 425)	0.09
Total Folate, μg	755 (335, 1872)	933 (376, 1872)	616 (335, 1518)	<0.001
Total Folate, %RDA	281 (113, 840)	332 (128, 840)	227 (113, 506)	0.003
Diet, μg	413 (140, 1472)	451 (156, 1473)	385 (140, 1251)	0.03
Diet, %RDA	159 (39, 491)	167 (39, 491)	143 (47, 417)	0.14
Supplement, μg	400 (0, 1200)	400 (66, 1200)	200 (0, 800)	0.004
Supplement, %RDA	100 (0, 400)	133 (33, 400)	100 (0, 267)	0.02

hsCRP = high sensitivity C-reactive protein (<3 mg/L normal); WBC, white blood cell count; %EER, percent Estimated Energy Requirement; RDA = Recommended Dietary Allowance

¹Means ± SD and t-tests for normally distributed variables; median (range) and Mann-Whitney for non-normally distributed variables.

²n=103 (n=58 for Hi-B₁₂, n=45 for RR - B₁₂).

³n=97(n=54 for Hi-B₁₂, n=43 for RR - B₁₂).

⁴n=102 (n=57 for Hi-B₁₂, n=45 for RR - B₁₂).

^ABiomarker Reference Ranges: B₁₂, pg/mL: Males: 4-6.9 years: 245-1078; 7-9.9 years 271-1170; 10-12.9 years 183-1088; 13-18 years 214-865; >18 year old 199-732. Females: 4-6.9 years 313-1407; 7-9.9 years 247-1174; 10-12.9 years 197-1019; 13-18 years 182-820; >18 year old 199-732. (CHOP Clinical Laboratory, Philadelphia, PA). B₆, ng/mL: 5-30 (ARUP Laboratories, Salt Lake City, UT). Folate, ng/mL: 280-903 ng/mL (ARUP Laboratories, Salt Lake City, UT). Homocysteine , μmol/L: 6-10.9 years: 0.8-6.5. 11-16.9 years 5.7-11.7. >17 years 10.5-16.7 (ARUP Laboratories, Salt Lake City, UT). Methionine, nmol/mL: 8-49 nmol/mL (ARUP Laboratories, Salt Lake City, UT).

PERT (porcine pancreatic extract) also contributed to B₁₂ intake. Subjects reported a median intake of 18 PERT capsules per day (range 9 to 42), representing a median of 270,000 lipase units per day. There are between 10 and 15 μg B₁₂ in every 100 g of enzymes. One PERT capsule contains approximately 0.2 g of enzyme, depending upon the specific product. Therefore 18 capsules per day contain ~3.6 g enzymes, or 0.36 to 0.54 μg B₁₂, contributing another ~20 to 30% RDA for a typical 9 year old child with CF and PI. The contribution of B₁₂ from PERT was not included in the total B₁₂ intake calculations for this report. Acid reduction medications (Histamine H₂ receptor antagonists and/or proton pump inhibitors) were used by 65% of study subjects. Inhaled tobramycin was used by 29% of study subjects, and oral antibiotics by 36% of study subjects. Both inhaled and oral antibiotics were used by 49% of study subjects. There were no differences in acid reduction medication, inhaled, oral or total antibiotic use between B₁₂ status groups.

Children in the Hi-B₁₂ group were significantly older, had poorer BMIZ and lung function than those with RR-B₁₂ (Table 1). Homocysteine was lower and cysteine was higher in Hi-B₁₂ group, and methionine did not differ by group. Total intake of B₁₂ was significantly higher in the Hi-B₁₂ compared to RR-B₁₂ group, 13 versus 10 times higher than the RDA, respectively. This was driven by the increased supplement-based B₁₂ intake (10 times the RDA in the Hi-B₁₂ group versus 6 times the RDA in the RR-B₁₂ group). Furthermore, there were no group differences in ietary B₁₂ intake between B₁₂ status groups. Study subjects with Hi-B₁₂ were significantly more likely to be taking > 1000 %RDA supplemental B₁₂ than those with RR-B₁₂. Supplemental intakes of both B₆ and folic acid were also significantly higher in the Hi- B₁₂ group. B₁₂ groups did not differ by sex, genotype or PERT use (data not shown).

Significant predictors of B₁₂ status were tested using multiple logistic regression analysis and the best model is shown in Table 3. From the multiple logistic regression model, higher supplemental B₁₂ intake and older age significantly increased risk for Hi-B₁₂ status, while higher FEV₁% predicted reduced risk, with these predictors combined explaining 18% of the variance in serum B₁₂ status (P<0.001). Compared to subjects receiving <500 %RDA of supplemental B₁₂, there is a 2.3 fold increased risk for Hi-B₁₂ for subjects for every additional 500 %RDA, reaching a 6.9 fold increase for those receiving ≥1500 %RDA B₁₂ from supplement-based B₁₂ intake. BMIZ was inversely associated with B₁₂ status in the simple model, but was no longer significant in the multivariable model. Other potential predictors were tested (for example, the use of acid reduction medications and antibiotic use) and did not add significantly to the model.

Table 3. Multiple Logistic Regression Model Predicting High Serum B₁₂ Status Group.

		Hi-B12 Status Group
n=103	Odds ratio	SE	Z	P	95% CI	Pseudo-R2
				<0.001		0.18
B12 supplement intake group, %RDA1	2.26	0.67	2.78	0.005	1.27-4.03
Age, > 12 years	3.38	1.85	2.23	0.026	1.16-9.87
Sex, female	2.34	1.11	1.80	0.072	0.93-5.93
FEV1, % predicted	0.98	0.01	−2.18	0.029	0.96-0.99
Constant	0.30	0.47	−0.77	0.441	0.01-6.55

¹B₁₂ supplementation intake categories, %RDA: category 1 = 0-499; category 2 = 500-999; category 3 = 1000-1499; category 4 = ≥1500

For the assessment of possible cyanide-related toxicity with sustained high cyanocobalamin supplement intake, whole blood cyanide was analyzed in seven subjects. Cyanide was below detectible concentrations (< 5μg/dL) for six subjects, and 8 μg/dL in one subject.

DISCUSSION

Serum B₁₂ concentrations were greater than the age and sex reference ranges for more than half of the subjects, and was mostly driven by supplement-based B₁₂ intake. Total B₁₂, B₆, and folic acid intakes were generally greater in subjects with CF than those observed in children of similar age and sex from the general population²³. These high B₁₂ intake and serum values were unexpected in this population of patients.

In healthy individuals approximately 50% B₁₂ intake is absorbed, with approximately 98% by active transport at the terminal ileum and 2% via passive intestinal diffusion. Absorbed B₁₂ is bound to transcobalamin (TC) II in the circulation and at the tissue level, and to TC-I in the serum and in the enterohepatic circulation¹. Perturbations in B₁₂ digestion and absorption have been reported in CF, including altered glycosylation of intrinsic factor (IF)², hyperacidity, IF hypersecretion²⁴, and loss of pancreatic function (poor duodenal alkalinization and protease insufficiency with incomplete R binder degradation)². Patients with meconium resections may have impaired B₁₂ absorption and enterohepatic circulation²⁵. However, despite these potential perturbations in B₁₂ digestion and absorption, alkalinization of the duodenal fluid⁵, and providing PERT ⁵ or trypsin²⁶, normalized B₁₂ absorption in subjects with CF and PI. B₁₂ concentrations measured after pancreatin treatment (pancreatic extract predecessor to modern PERT) in subjects with CF and PI were normal or elevated²⁷. Early studies documented abnormal absorption of crystalline B₁₂ and impaired Schilling’s tests. With modern PERT and B₁₂ intake from food and B₁₂ as cobalamin rather than the crystalline form, B₁₂ metabolism is likely similar to healthy children and younger adults. B₁₂ deficiency has not been described in the CF literature since 1973⁶. It is unclear if B₁₂ supplementation is required in patients with CF and PI under current standards of care.

In the current report, we determined B₁₂ intake from the diet and from vitamin supplements to understand the contributions of sources to intake and serum concentrations. Dietary intake of B₁₂ were similar to that reported in healthy children, and both had dietary intake of about three times the RDA²³. CF specific vitamins contain approximately the B₁₂ dose found in multivitamins designed for older adults who have increased risk of B₁₂ deficiency and pernicious anemia. In addition to B₁₂ in the diet and vitamin supplements, PERT products contain B_12. There was a relationship of age with serum B_12, and CF vitamin supplement-based intake increases with age. Often between 8 and 12 years of age, the vitamin pill intake is increased from one CF specific vitamin per day to two per day in order to increase the fat soluble vitamin intake²⁸. The amount of B₁₂ and other vitamins in the product also increases. Age associations are observed with other clinical and nutritional outcomes in CF, specifically, negative associations with FEV₁% and BMIZ⁴. In part this age-based intake multivitamin intake increase may contribute to the inverse relationship of both FEV₁ and BMIZ with serum B₁₂ observed in this cohort.

The finding that B₁₂ concentrations were inversely related to homocysteine concentrations likely reflects the 1-methyl metabolic pathway (homocysteine is converted to methionine) or to the transulfuration pathway (homocysteine is converted to cysteine). Homocysteine and methylmalonic acid are used for diagnosing subclinical B₁₂ deficiency^{29, 30}.

Serum B₁₂ may be elevated with inflammation and was postulated as a possible etiology for elevated concentrations in subjects with CF in early studies²⁷. Elevated serum B₁₂ was noted in subjects with CF and was associated with both elevated serum transcobalamin II and normal unsaturated B₁₂ binding capacity. These investigators postulated the findings were related to hepatic dysfunction, recurrent pulmonary infection or increased turnover of myeloid cells. Elevated B₁₂ levels can occur in chronic myelogenous (leukemic) conditions, renal failure and liver disease. We evaluated blood markers for inflammation, renal function, hepatic function and complete blood counts with differentials. High sensitivity-CRP concentration and white blood cell count (WBC) did not differ between B₁₂ status groups. There were no subjects with a myeloproliferative disorder or renal disease. Liver enzymes were mildly elevated as expected in a group of subjects with CF and PI and were not associated with B₁₂ status. There were no associations with hemoglobin, hematocrit, platelet count, liver enzymes, total protein, albumin, BUN or creatinine with serum B₁₂ (data not shown). In this study there was no indication that serum B₁₂ was acting as an acute phase reactant, or that it was related to other diagnoses.

The consequences of lifelong high dose supplement-based B₁₂ and sustained elevated serum concentrations are unknown. In the non-CF medical setting, elevated serum B₁₂ is an indication for a diagnostic work up to rule out serious disease^{16, 31}. Our data demonstrate that with current CF care, patients take CF-specific vitamins containing high doses of B₁₂ and many patients will likely have elevated serum B₁₂. Thus, the elevated B₁₂ cannot serve as a sign of potential serious disease. Supplement-based B₁₂ is in the form of cobalamin (cyanocobalamin). Cyanide is released intracellularly and is rapidly cleared and cyanide toxicity is not expected. While elevated pulmonary secretion cyanide levels have been reported in CF, these likely result from Pseudomonas and Burkholderia secretions³². In the current study, only one of seven subjects had a detectable cyanide level, and this was within the laboratory reference range. From these limited data, the risk of systemic cyanide exposure is likely low.

Of interest, FEV₁ differed between B₁₂ status groups and persisted in the multiple logistic regression models after adjusting for age. In CF, FEV₁ declines with age; B₁₂ dose increases with age, and the upper limit of serum B₁₂ reference range declines with age. Other factors that may be related to the inversed relationship between B₁₂ and FEV₁ include the observation that sicker patients may take more medication including supplements (more prescriptions and/or better adherence to prescribed treatment). Further study is required to better understand mechanisms and clinical significance of the B₁₂ and FEV₁ inversed correlation.

It is unclear if there are clinical benefits of B₁₂ intake beyond that to sustain a normal B₁₂ status. Scambi et al³³ reported improvements in phospholipid docosahexaenoic acid (DHA) status in young children with CF with daily 5-methyltetrahydrofolate (7.5 mg) and B₁₂ (0.5 mg) supplementation in a 24 week intervention. The phospholipid DHA improvement was thought to be a result of the interaction of folate, B₁₂, phospholipid DHA and 1-methyl metabolic pathway³³. The intakes of B₁₂ and B₆ were orders of magnitude greater than recommended for healthy people. There are currently no specific CF-specific B-vitamin intake recommendations different than those for the general population. There is no known toxicity of vitamin B₁₂^12,13. Chronic supplement based B₆ intake >1000 μg/d may increase the risk for peripheral neuropathy¹³. An adverse effect of high folic acid intake has the potential to mask co-incident B₁₂ deficiency and associated pernicious anemia and progressive neurological damage¹³. Folic acid intakes were not unusually high in our sample of children.

In summary, total B₁₂ intake (diet, supplement, PERT) was high, and largely due to high supplement-based B₁₂ intake. The B₁₂ and B₆ vitamin intake was higher than recommended, and provided no known benefit. Lifelong high B₁₂ intake will result in sustained elevated serum B₁₂. The associated risks or benefits of prolonged high B₁₂ intake and elevated serum B₁₂ in people with CF is unknown. Studies are needed to determine the B₁₂ dose that supports concentrations and status within the reference ranges and to evaluate possible B₁₂-related clinical outcomes across the increasing life span of people with CF. The inverse correlation between B₁₂ and FEV₁ merits further investigation.

ABBREVIATIONS AND ACRONYMS

Adj HAZ

Height-for-age Z score after height adjustment for mid-parent height

AI

Adequate Intake

BMI

Body mass index

BMIZ

BMI-for-age Z score

CF

Cystic fibrosis

CHOP

Children’s Hospital of Philadelphia

CTRC

Clinical Translational Research Center

DHA

Docosahexaenoic acid

DRI

Dietary Reference Intake

EER

Estimated Energy Requirement

FEV₁

Forced expiratory volume at one second

Hi-B₁₂

Serum B₁₂ above reference range for age and sex

HAZ

Height-for-age Z score

RR-B₁₂

Serum B₁₂ within reference range for age and sex

hsCRP

High sensitivity C-reactive protein

PERT

Pancreatic enzyme replacement therapy

PI

Pancreatic insufficiency

RDA

Recommended Dietary Allowance

WAZ

Weight-for-age Z score