If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Requests for reprints should be addressed to Lawrence R. Solomon, MD, Section of Palliative Care, Department of Medicine, Yale University School of Medicine and Smilow Cancer Hospital, WWW-403; 333 Cedar Street, P.O. Box 208021, New Haven, CT 06520-8021.
Cobalamin (B12) deficiency can lead to irreversible neurocognitive changes if unrecognized. Screening involves measurement of serum cobalamin levels, but the sensitive metabolic indicators of cobalamin deficiency, methylmalonic acid (MMA) and homocysteine (HCys), may be normal when cobalamin values are low and elevated when cobalamin values are normal. Because cobalamin is inactivated by oxidation, the relationship between these metabolites and comorbidities associated with increased oxidative stress (oxidant risks) in subjects with low and low-normal cobalamin levels was studied.
Methods
A retrospective record-review was conducted of community-dwelling adults evaluated for cobalamin deficiency during a 12-year period with serum cobalamin values in the low (≤ 200 pg/mL; n = 49) or low-normal (201-300 pg/mL; n = 187) range and concurrent measurement of MMA.
Results
When “No” oxidant risk was present, elevated MMA (>250 nmol/L) and HCys (>12.1 μmol/L) values occurred in 50% and 30% of subjects, respectively (P <.01). In contrast, when “Three or More” oxidant risks were present, mean MMA and HCys values were significantly higher, and elevated MMA and HCys values occurred in 84% and 78% of these subjects, respectively (P ≤.012). Pharmacologic doses of cyanocobalamin significantly decreased metabolite values in ≥ 94% of treated subjects.
Conclusion
In subjects with low or low-normal cobalamin values, metabolic evidence of cobalamin deficiency is more frequent when 3 or more oxidant risks are present. Thus, defining a low serum cobalamin level to screen for cobalamin deficiency may be a “moving target” due to the variable presence and severity of often subtle, confounding clinical conditions in individual subjects.
The incidences of elevated methylmalonic acid and homocysteine values, metabolic markers of cobalamin deficiency, were similar in subjects with low (≤ 200 pg/mL) and low-normal (201-300 pg/mL) cobalamin levels.
•
The incidences of high metabolite values were 1.5- to threefold higher in subjects with 3 or more comorbidities associated with increased oxidative stress than in subjects with no oxidant risks.
•
Methylmalonic acid values were more frequently increased than homocysteine values.
Cobalamin (ie, vitamin B12) deficiency is common in vegetarians, the elderly, and subjects using such frequently prescribed medications as proton pump inhibitors, H2-blockers and metformin.
If unrecognized, this disorder can lead to irreversible neurocognitive dysfunction. Measurement of serum cobalamin (Cbl) is the initial test usually performed to screen for Cbl deficiency. However, cut-off points suggested vary widely between 200 and 500 pg/mL.
Moreover, even at Cbl levels ≤ 200 pg/mL, the lowest cut-off point commonly used to define Cbl deficiency, many subjects have neither clinical evidence of Cbl deficiency (ie, megaloblastic anemia or neurocognitive disorders) nor elevated levels of the Cbl-dependent metabolites, methylmalonic acid (MMA) and homocysteine (HCys) (biochemical indicators of Cbl depletion).
Cobalamin deficiency in general practice. Assessment of the diagnostic utility and cost-benefit analysis of methylmalonic acid determination in relation to current diagnostic strategies.
It is of note then that Cbl is readily inactivated by oxidation and that elevated MMA and HCys levels in subjects with normal Cbl values have been related to the presence of comorbidities associated with increased oxidative stress (oxidant risks).
Because Cbl values between 201 and 300 pg/mL are often present in subjects with both metabolically and clinically significant Cbl deficiency, a retrospective study was performed to determine if metabolic changes consistent with Cbl deficiency also were related to the presence of oxidant risks in subjects with low (≤ 200 pg/mL) and low-normal (201-300 pg/mL) Cbl values.
The laboratory reference range for Cbl was 201-1100 pg/mL (Quest Diagnostics, Madison, New Jersey). Because Cbl values between 201 and 300 pg/mL are often present in subjects with both metabolically and clinically significant Cbl deficiency due to Cbl malabsorption, Cbl values ≤ 200 pg/mL were defined as “low,” while Cbl values of 201-300 pg/mL were defined as “low-normal.”
When values were obtained on more than one occasion within a 6-week period, the lowest Cbl value and the highest metabolite values were used for analysis.
Subjects
A retrospective review of the medical records of all community-dwelling subjects evaluated by the author for Cbl deficiency in a primary care setting between August 1, 1993 and June 30, 2005, with the measurement of both serum Cbl and MMA levels, was conducted as previously described.
Subjects were screened because of the presence of clinical findings consistent with Cbl deficiency or because of the presence of disorders known to lead to Cbl depletion.
Cbl values were “low” in 49 subjects and “low-normal” in 187 subjects. HCys values were obtained in 204 of these 236 individuals (86%).
This study conforms to the principles of the Declaration of Helsinki of 1975 as revised in 2008, and the institutional human investigation committee determined that further review was not required.
Identification of Oxidant Risks
All active medical conditions present in each subject were identified, and a Medline search was performed to determine which disorders were associated with increased oxidative stress as defined by the presence of increased oxidative byproducts systemically, including malondialdehyde and F2-isoprostanes (as indices of lipid oxidative damage); carbonylated proteins (as an index of protein oxidative damage); 8-hydroxy-2′-deoxyguanosine (as an index of DNA oxidative damage); and reduced and oxidized glutathione levels (as a general index of the redox state).
At least one oxidant risk was present in 152 subjects (64%), including advanced age (≥ 70 years; n = 61); hypertension (n = 58); cigarette abuse (n = 35); alcohol abuse (n = 25); diabetes mellitus (n = 24); malignancy (n = 20); mild-moderate renal insufficiency (creatinine = 1.4-2.4 mg/dL) (n = 10); chronic infections (n = 9); medication-dependent asthma (n = 8); rheumatologic disorders (n = 8); pregnancy (n = 7); iron deficiency (n = 7); hepatitis (n = 4); neurodegenerative disorders (n = 2); sickle cell disease (n = 2); chronic pancreatitis (n = 2); congestive heart failure (n = 1); inflammatory bowel disease (n = 1); recent myocardial infarction (n = 1); hyperthyroidism (n = 1); and an unexplained high sedimentation rate of 115 mm/h (n = 1). Subjects were then divided into 4 groups: “No” oxidant risks (n = 84); “One” oxidant risk (n = 70); “Two” oxidant risks (n = 37); and “Three or More” (“Three+”) oxidant risks (n = 45; includes 35 subjects with 3 risks and 10 subjects with 4 risks).
Cobalamin Treatment
Patients were offered treatment if they had clinical findings known to be associated with Cbl deficiency or if MMA values were elevated. Because abnormal HCys values are less specific for Cbl deficiency, treatment was usually not offered to subjects with isolated elevated HCys values.
Patients were treated with cyanocobalamin 2 mg per day orally or 1 mg intramuscularly 3 times a week for 2 weeks, weekly for 8 weeks and monthly thereafter. Metabolites were re-measured 1-3 months after beginning Cbl therapy. A response to Cbl therapy was defined as either a fall in the metabolite level to a value within the normal reference range (ie, ≤ 250 nmol/L for MMA and ≤ 12.1 μmol/L for HCys) or a decrease in the metabolite by more than 1 SD greater than its mean intraindividual variability as previously determined for this population (ie, >116 nmol/L for MMA and >3.6 μmol/L for HCys).
Biochemical indicators of B vitamin status in the US population after folic acid fortification: results from the National Health and Nutrition Examination Survey 1999-2003.
Thus, geometric means, 2-tailed Student t tests and paired t tests using log-transformed data, linear regression analyses, and χ2 analyses were determined using StatPlus:mac (release 5.7, 2009, AnalystSoft, Vancouver, BC, Canada). A P value <.05 was considered significant.
Results
Patient Characteristics
Age, sex, race, and the distribution of oxidant risks were similar in the two Cbl groups (Table 1). Subjects with “low” Cbl values had higher mean MMA values than subjects in the “low-normal” group, but the incidences of elevated MMA values were not significantly different in the two Cbl populations. Both mean HCys values and the incidences of elevated HCys values were the similar in subjects with “low” and “low-normal” Cbl levels.
P <.001 vs the mean MMA value in subjects with Cbl = 201-300 pg/mL.
[105-4911]
278 (187) [101-1384]
MMA >250 nmol/L (%)
32 (65%)
101 (54%)
HCys (μmol/L)
12.3 (41) [4.4-61.4]
11.2 (163) [3.0-60.7]
HCys >12.1 μmol/L (%)
18 (44%)
67 (41%)
Values for age are arithmetic means ± 1 SD, while values for Cbl, MMA, and HCys are geometric means for the number of subjects shown in parentheses with the range of values for each parameter shown in brackets.
When “No” oxidant risks were present, mean MMA values were not significantly different in subjects with “low” or “low-normal” Cbl levels (Figure 1A). The presence of “One” or “Two” oxidant risks did not affect mean MMA values in either Cbl group. However, when “Three+” oxidant risks were present, mean MMA levels were significantly higher than when “No” or “One” oxidant risk was present within both Cbl populations. Mean MMA levels were also higher in subjects with “low” Cbl values than in subjects in the “low-normal” Cbl group when either “Two” or “Three+” oxidant risks were present. Overall, there was a significant inverse linear relationship between Cbl and MMA in each oxidant risk group, but this relationship was more than fourfold greater in subjects with “Three+” oxidant risks than in those with “No” oxidant risks (Figure 2A).
Figure 1Mean MMA and HCys values relative to Cbl levels and presence of oxidant risk factors. Values are geometric means for MMA (A) and HCys (B) for the number of subjects shown in parentheses. ∗P = .012 and P = .0054 vs MMA in subjects with Cbl ≤ 200 pg/mL and “No” and “One” oxidant risk, respectively. †P = .0010 and P <.001 vs MMA in subjects with Cbl = 201-300 pg/mL and “No” and “One” oxidant risk, respectively. §P = .013 and .055 vs HCys in subjects with Cbl ≤ 200 pg/mL and “No” and “One” oxidant risk, respectively. ¶P <.001 vs HCys in subjects with Cbl = 201-300 pg/mL and either “No” or “One” oxidant risk. ∗∗P = .0072 and P <.001 vs HCys in subjects with Cbl = 201-300 pg/mL and “No” and “One” oxidant risk, respectively. Cbl = cobalamin; HCys = homocysteine; MMA = methylmalonic acid.
Figure 2Linear regression analyses of the relationships between Cbl and MMA or HCys relative to the number of oxidant risks present. Linear regression analyses between Cbl and MMA (A) and HCys (B) were performed in subjects with “low” and “low-normal” Cbl values and either “No” or “Three+” oxidant risks. Because only 2 subjects in the entire study group had Cbl values < 100 pg/mL (both with “One” oxidant risk and Cbl values of 58 and 86 pg/mL, respectively), regression lines were not extended below a Cbl value of 100 pg/mL. Cbl = cobalamin; HCys = homocysteine; MMA = methylmalonic acid.
The pattern was the same when the incidences of high MMA values were considered, except that values were similar in the “low” and “low-normal” Cbl populations even when “Two” and “Three+” oxidant risks were present (Table 2).
Table 2Relationship of the Incidence of High Metabolite Values to the Number of Oxidant Risks
Incidences of high MMA and HCys values in subjects with Cbl levels ≤ 200 pg/mL were not significantly greater than those in subjects with Cbl levels of 201-300 pg/mL and the same number of oxidant risks (P ≥.24).
P = .0080 vs the incidence of high MMA values in subjects with Cbl levels ≤ 200 pg/mL and “No” oxidant risks; and P = .0034 vs the incidence of high MMA values in subjects with Cbl levels ≤ 200 pg/mL and “One” oxidant risk.
P = .0034 vs the incidence of high MMA values in subjects with Cbl levels of 201-300 pg/mL and “No” oxidant risks; and P <.001 vs the incidence of high MMA values in subjects with Cbl levels of 201-300 pg/mL and “One” oxidant risk.
Incidences of high MMA and HCys values in subjects with Cbl levels ≤ 200 pg/mL were not significantly greater than those in subjects with Cbl levels of 201-300 pg/mL and the same number of oxidant risks (P ≥.24).
P = .0078 vs the incidence of high HCys values in subjects with Cbl levels ≤ 200 pg/mL and “No” oxidant risks; and P = .0041 vs the incidence of high HCys values in subjects with Cbl levels ≤ 200 pg/mL and “One” oxidant risk.
P = .022 vs the incidence of high HCys values in subjects with Cbl levels of 201-300 pg/mL and “No” oxidant risks and P = .0034 vs the incidence of high HCys values in subjects with Cbl levels of 201-300 pg/mL and “One” oxidant risk.
∗ Incidences of high MMA and HCys values in subjects with Cbl levels ≤ 200 pg/mL were not significantly greater than those in subjects with Cbl levels of 201-300 pg/mL and the same number of oxidant risks (P ≥.24).
† P = .0080 vs the incidence of high MMA values in subjects with Cbl levels ≤ 200 pg/mL and “No” oxidant risks; and P = .0034 vs the incidence of high MMA values in subjects with Cbl levels ≤ 200 pg/mL and “One” oxidant risk.
‡ P = .0034 vs the incidence of high MMA values in subjects with Cbl levels of 201-300 pg/mL and “No” oxidant risks; and P <.001 vs the incidence of high MMA values in subjects with Cbl levels of 201-300 pg/mL and “One” oxidant risk.
§ P = .0078 vs the incidence of high HCys values in subjects with Cbl levels ≤ 200 pg/mL and “No” oxidant risks; and P = .0041 vs the incidence of high HCys values in subjects with Cbl levels ≤ 200 pg/mL and “One” oxidant risk.
‖ P = .022 vs the incidence of high HCys values in subjects with Cbl levels of 201-300 pg/mL and “No” oxidant risks and P = .0034 vs the incidence of high HCys values in subjects with Cbl levels of 201-300 pg/mL and “One” oxidant risk.
¶ P <.001 vs the incidence of high HCys values in subjects with Cbl levels of 201-300 pg/mL and either “No” oxidant risks or “One” oxidant risk.
Mean HCys values were not statistically different when subjects in the two Cbl populations with the same number of oxidant risks were compared (Figure 1B). Mean HCys levels were also similar in subjects with “No” or “One” oxidant risk within both Cbl populations, but were significantly higher when “Three+” oxidant risks were present. Overall, there was an inverse relationship between Cbl and HCys when “Three+” oxidant risks were present, but not when “No” oxidant risks were present (Figure 2B).
Similarly, the incidences of high HCys values were significantly greater in subjects with “Two” or “Three+” oxidant risks than in those with either “No” or “One” oxidant risk in both Cbl populations (Table 2). Moreover, when “No” oxidant risk was present in subjects with “low” or “low-normal” Cbl values, the incidence of high HCys values (30%) was significantly lower than the incidence of high MMA values (50%; P = .0093) (Table 2). The incidence of high HCys values (22%) was also significantly lower than the incidence of high MMA values (43%) when only one oxidant risk was present (P = .013). However, when “Two” or “Three+” oxidant risks were present, the incidences of high HCys (57% and 78%) and high MMA values (59% and 84%) were similar (P ≥.48).
Pattern of Increased Metabolite Values
Both MMA and HCys were measured in 204 subjects in the two Cbl populations, and at least one metabolite was elevated in 137 of them (67%). Isolated elevations in MMA values were significantly more frequent in subjects with “No” or “One” oxidant risk than in subjects with “Two” or “Three+” oxidant risks, while combined elevations in both metabolites occurred significantly more frequently in subjects with “Three+” oxidant risks (Figure 3). Thus, MMA values were elevated more frequently than HCys values when “No” or “One” oxidant risk was present, but the incidences of elevated values of both metabolites were similar when “Two” or “Three+” oxidant risks were present.
Figure 3Relationship of the pattern of increased MMA and HCys values to the presence of oxidant risk factors. Values plotted are the percent of those subjects in whom both metabolites were measured and in whom an increased value of MMA or HCys was noted. ∗P = .025 and P <.001 vs the incidences of isolated high MMA values in subjects with “Two” or “Three+” oxidant risks, respectively, by χ2 analyses. †P <.001 vs the incidence of high MMA values in subjects with “No or One” oxidant risks; and P = .025 and P <.001 vs the incidences of high HCys values in subjects with “Two” and “Three+” oxidant risks, respectively. HCys = homocysteine; MMA = methylmalonic acid.
Overall, 81 of the 133 subjects with high MMA values were both treated with Cbl and had posttreatment values obtained (61%). Significant responses were noted in 78 of these 81 subjects (96%). Similarly, 51 of the 84 subjects with high HCys values were evaluable for the effects of Cbl therapy (61%), and significant responses were noted in 48 of them (94%). Response rates and posttreatment values for both metabolites were not significantly different regardless of the number of oxidant risks present (Table 3).
Table 3Effect of Cbl Therapy of Subjects With Elevated MMA and HCys Values
Metabolite
Oxidant Risks
Number Elevated
Treated (%)
Responders
Pretherapy
Posttherapy
P-Value
MMA (nmol/L)
None
42
30 (71%)
30 (100%)
484
173
<.001
One
30
16 (53%)
14 (88%)
445
199
<.001
Two
23
15 (65%)
14 (93%)
433
185
<.001
Three+
38
20 (53%)
20 (100%)
516
174
<.001
HCys (μmol/L)
None
23
16 (70%)
16 (100%)
15.1
9.1
<.001
One
12
6 (50%)
6 (100%)
16.1
10.1
.0013
Two
20
11 (55%)
9 (82%)
15.6
10.6
<.001
Three+
28
18 (64%)
17 (94%)
19.4
11.9
.0037
Response criteria are defined in the Methods. Pretherapy and Posttherapy metabolite values are geometric means for the number of treated subjects. “P” values were derived from paired t tests.
Cobalamin deficiency in general practice. Assessment of the diagnostic utility and cost-benefit analysis of methylmalonic acid determination in relation to current diagnostic strategies.
While genetic variants may alter the metabolic effectiveness of circulating Cbl by either decreasing the level of haptocorrin, the inactive binder of serum Cbl (ie, “false” low Cbl levels), or decreasing the effective level of holotranscobalamin, the active transport form of Cbl (ie, “false” normal Cbl levels), they are uncommon.
An audit of holotranscobalamin (“Active” B12) and methylmalonic acid assays for the assessment of vitamin B12 status: application in a mixed patient population.
In the present study, it was of note that neither mean metabolite values nor the incidences of high metabolite values were significantly greater in subjects with “low” Cbl values than in those with “low-normal” Cbl values when “No” oxidant risks were present (Figure 1; Table 2). However, MMA and HCys values were significantly higher when 3 or more oxidant risks were present in both Cbl populations (Figure 1; Table 2). These findings are similar to those previously described in subjects with Cbl values well within the normal reference range.
Significantly, the incidences of elevated values for both metabolites were the same when Cbl values were in the “low-normal” range as when they were frankly “low”, even when adjusted for the number of oxidant risks present (Table 1, Table 2). Thus, Cbl deficiency should not be excluded as a cause for hematologic or neurocognitive abnormalities in subjects with Cbl values greater than the usual cut-off value of 200 pg/mL.
MMA values were also more frequently increased than HCys values in subjects with “No” or “One” oxidant risk, but not in those with “Three+” oxidant risks (Table 2, Figure 3). Because MMA accumulation reflects decreased activity of a mitochondrial Cbl-dependent enzyme, while HCys accumulation reflects decreased activity of a cytoplasmic Cbl-dependent enzyme, and because mitochondrial processes are more sensitive to oxidative damage, this observation is also consistent with a role for oxidative stress as a determinant of metabolic Cbl deficiency.
Many clinical disorders associated with increased oxidative stress are stable over time, while others vary in severity depending on their natural history and the effectiveness of treatment. Thus, metabolic evidence of Cbl deficiency associated with oxidative inactivation of Cbl would also be expected to vary with the clinical status of the individual. Interestingly then, longitudinal follow-up of elevated MMA values in apparently healthy adults showed values to be stable in 39%, decreased in 42%, and increased in 16%.
Although high MMA values in this study did not predict the development of clinical abnormalities, exposure of 7 subjects with low Cbl values (≤ 200 pg/mL) and 2 subjects with low-normal Cbl values (227 and 312 pg/mL) to nitrous oxide, a drug known to oxidize Cbl, resulted in acute precipitation of neurologic dysfunction.
Similarly, Cbl-responsive neurologic disorders have also been shown to be associated with the presence of pro-oxidant disorders, and the increased oxidative stress associated with acute myocardial infarction has been linked to acute increases in urinary MMA excretion.
Thus, clinically overt Cbl deficiency may develop when susceptible individuals are exposed to additional oxidant stimuli. The findings in the current study also raise the possibility that higher and more frequent doses of Cbl than usually recommended may be needed to correct Cbl deficiency due to increased oxidative stress, and that fully reduced forms of Cbl (eg, methylcobalamin) may be more effective than the partially reduced forms of Cbl currently prescribed (ie, cyanocobalamin and hydroxocobalamin).
Oral pharmacologic doses of cobalamin may not be as effective as parenteral cobalamin therapy in reversing hyperhomocysteinemia and methylmalonic acidemia in apparently normal subjects.
This study was limited in that it was a retrospective analysis, with patients selected due to a suspicion of Cbl deficiency based on clinical findings or predisposing risk factors. Moreover, because the severity of oxidative stress varies between and within the disorders identified as oxidant risks, and because the presence of other oxidant risks may not have been recognized, this study was also limited by the absence of measurement of biomarkers of oxidative stress. Therefore, prospective studies with direct measures of oxidative stress are needed to confirm these observations. However, it is of note in this regard that a prospective study of 18 subjects with schizophrenia (a “pro-oxidant” disorder) and low-normal Cbl levels, found elevated levels of urinary MMA to be directly correlated with increased values of the oxidant marker malondialdehyde in erythrocyte membranes.
Altered red cell membrane compositions related to functional vitamin B12 deficiency manifested by elevated urine methylmalonic acid concentrations in patients with schizophrenia.
It is concluded that, in subjects with low or low-normal Cbl values, metabolic evidence of Cbl deficiency is more frequent when disorders marked by increased oxidative stress are present, and that HCys is a less reliable indicator of metabolic Cbl deficiency than MMA when the number of oxidant risks are limited. Thus, defining a low serum Cbl level to use as a cut-off value in screening for Cbl deficiency may be a “moving target” due to the variable presence and severity of often subtle, confounding clinical conditions both across different populations and within individual subjects. Because clinical manifestations of Cbl deficiency may develop or progress when subjects with metabolic Cbl deficiency are exposed to additional oxidant risks, measurement of MMA and HCys may be of value even when screening Cbl values are in the “normal” or “low-normal” range, and correction of elevated metabolite values with Cbl therapy should be considered pending further studies.
Cobalamin deficiency in general practice. Assessment of the diagnostic utility and cost-benefit analysis of methylmalonic acid determination in relation to current diagnostic strategies.
Biochemical indicators of B vitamin status in the US population after folic acid fortification: results from the National Health and Nutrition Examination Survey 1999-2003.
An audit of holotranscobalamin (“Active” B12) and methylmalonic acid assays for the assessment of vitamin B12 status: application in a mixed patient population.
Oral pharmacologic doses of cobalamin may not be as effective as parenteral cobalamin therapy in reversing hyperhomocysteinemia and methylmalonic acidemia in apparently normal subjects.
Altered red cell membrane compositions related to functional vitamin B12 deficiency manifested by elevated urine methylmalonic acid concentrations in patients with schizophrenia.
00694-4&id=gr1.jpg){kind=link}
00694-4&id=gr2.jpg){kind=link}
00694-4&id=gr3.jpg){kind=link}