| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Articles |
1
Laboratory of Clinical Biochemistry, Haukeland University Hospital, N-5021 Bergen, Norway, and Division for General Practice, University of Bergen, Norway.
2
Department of Pharmacology, University of Bergen,
Armauer Hansens Hus, N-5021 Bergen, Norway.
3
Fürst Medical Laboratory, Soeren Bulls vei 25,
N-1051 Oslo 10, Norway.
a Author for correspondence. Fax 0047 55 97 3115; e-mail sverre.sandberg{at}haukeland.no.
 |
Abstract |
|---|
 Top Abstract IntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
Among laboratory tests for the diagnosis of cobalamin deficiency, determination of s-cobalamin is most commonly used. However, the diagnostic efficiency of s-cobalamin is too low . This can partly be explained by the fact that the major fraction of cobalamin in serum is bound to haptocorrin, which is not available for uptake in most cells (10)(11). Approximately 20% is bound to transcobalamin II. Only the transcobalamin II-cobalamin complex can be utilized by peripheral cells through receptor-mediated uptake (12)(13). In addition, s-cobalamin may be influenced by changes in the binding protein concentration. Thus, total s-cobalamin is a relatively poor indicator of bioavailable cobalamin (14). Therefore, tests for other cobalamin markers have been developed, such as the holo-transcobalamin II concentration in serum (15), the deoxyuridine suppression test, or the functional markers methylmalonic acid (MMA) and total homocysteine (tHcy) (16).
Among the functional tests, determination of serum MMA (s-MMA) has received particular attention. This is because of the stability of MMA in blood (17) and the small sample volume requirements and high diagnostic accuracy of the test. s-MMA is assumed to be a better indicator of intracellular cobalamin status than s-cobalamin, and s-MMA has been proposed as a complement or replacement for s-cobalamin measurements (17). However, methods for MMA measurement are costly and cumbersome, and commercial assays are not available. This hampers its widespread routine use.
Here, we report on the diagnostic strategies used to assess cobalamin deficiency in general practice, especially the influence of s-cobalamin assay results on final diagnosis and the decision to supplement patients. We also estimated the increase in diagnostic accuracy obtained by additional determination of s-MMA, and indicate the s-cobalamin range, where this may be justified, based on cost-benefit analyses.
|
Patients and Methods |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
Approximately 6 weeks after FML had sent the results of cobalamin analysis, the general practitioners received a questionnaire to obtain information on why they had requested s-cobalamin measurement. The categories provided in the questionnaire were anemia; macrocytosis; psychiatric symptoms (specified as depression, dementia, or psychosis); neurologic symptoms (unspecified); and diffuse symptoms, screening, and others. Other questions asked whether additional analyses were requested, whether the general practitioner believed that the patient was cobalamin deficient, and whether the patient had received any treatment. General practitioners who did not respond to the first questionnaire received a new questionnaire after 6 weeks.
A total of 316 (84%) out of 376 questionnaires were returned. Five
were excluded from all further analyses because the doctors were not
general practitioners. Nine patients were excluded because the
questionnaires were incompletely filled out. In addition, we did not
ask for clinical data in 78 cases where s-cobalamin was ordered as a
confirmation test of either an earlier s-cobalamin determination or as
therapy control in patients on cobalamin supplementation. Thus, 224
patients were qualified for additional data analysis. In addition, 15
patients had to be excluded from parts of the study because there was
not enough serum left for measurement of serum concentrations of
creatinine and MMA. Likewise, 12 women and 7 men with serum creatinine
>115 and 125 µmol/L, respectively, were excluded from parts of the
study where s-MMA was used (18)(19)(20). Table 1
shows the demographic characteristics of the study population.
|
sample handling and biochemical analyses
In this study, the blood was collected and the serum prepared at
the general practitioners' offices. The serum samples were then mailed
to FML. After FML had performed the requested blood analyses, the serum
samples were stored at -20 °C. s-Creatinine, s-tHcy, and s-folate
were later measured at the Laboratory of Clinical Biochemistry,
Haukeland Hospital, Bergen, Norway, whereas s-MMA was analyzed at the
Department of Pharmacology, University of Bergen.
s-Cobalamin and s-folate were measured by RIA (Diagnostic Product Corp.). The reference range of s-cobalamin was 170–700 pmol/L, and the between-day CV was 3–7%, depending on the s-cobalamin concentration. The reference limit of s-folate was >5.0 nmol/L, and the between-day CV was 3–7%, depending on the s-folate concentration.
Serum creatinine was measured by the Jaffe method on an Axon® (Bayer Instruments Corp.). The reference ranges for serum creatinine were 55–115 and 60–125 µmol/L for women and men, respectively. The between-day CV was 2.5–3.5%. In 15 patients, there was not sufficient serum for creatinine determination, and these were excluded from part of the analyses.
Serum tHcy, which includes free and protein-bound Hcy forms, was determined by a modification of an automated procedure based on derivatization with monobromobimane, followed by HPLC and fluorescence detection (21)(22). The between-day CV was ~3%. The reference range for serum tHcy is 0–15 µmol/L, and results above 15.0 µmol/L are considered increased (23).
s-MMA was measured by capillary electrophoresis (24). The between-day CV was 5–10% for MMA concentrations within a range of 0.12–0.57 µmol/L. The reference range for s-MMA is 0.05–0.26 µmol/L. The cutoff point for diagnosing functional cobalamin deficiency was set to 0.376 µmol/L, which corresponded to 3 SD above the mean of healthy controls (22)(25)(26)(27)(28). Lower values were referred to as "normal".
statistics and data analyses
Data analyses were performed using multivariate stepwise logistic
regression analyses (SPSS, Ver. 4.0 for Macintosh). From this we obtain
the log odds and could remodel the equation to give the posttest
probability of an event. In the logistic regression analyses, different
dependent and independent variables were used.
|
Results and Discussion |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
In the present study, we investigated the diagnostic strategies among 224 general practitioners, each contributing one patient with a serum cobalamin <300 pmol/L. The patients belonged to five different categories defined according to their cobalamin concentration. This allowed us to investigate the different diagnostic approaches as a function of the cobalamin concentration. We first registered the indications that the general practitioner listed for requesting the serum cobalamin assay. We then investigated how the general practitioner responded to the serum cobalamin assay result as judged by (a) the subsequent request of additional biochemical tests; (b) their ability to set a definite diagnosis; and (c) the initiation of cobalamin therapy.
We performed a retrospective comparison of the s-MMA as the reference standard with the initial serum cobalamin assay result, the ability of the general practitioner to set a definite diagnosis, and the initiation of cobalamin supplementation. On the basis of these results, we were able to evaluate the diagnostic benefit of including MMA as an additional marker of cobalamin deficiency.
indications for requesting s-cobalamin
The indications for requesting s-cobalamin are summarized in Table 2
. Two or more indications were frequently reported. Anemia was
an indication for requesting s-cobalamin in ~30% of patients with
s-cobalamin values <170 pmol/L. This percentage decreased at higher
s-cobalamin concentrations. Macrocytosis was an uncommon reason for
cobalamin testing, even at low cobalamin concentrations. In almost 50%
of the patients with s-cobalamin <190 pmol/L, the general
practitioners did not report hematological, neurologic, or psychiatric
findings or symptoms.
|
primary and ancillary tests
Hemoglobin was requested together with s-cobalamin for 70–90% of
the patients. s-Iron, s-total iron binding capacity, or s-ferritin was
requested for 60–80%, thyroid function tests were requested for
30–50%, and serum or red cell folate was requested for ~30% of the
patients. s-Lactate dehydrogenase, s-bilirubin, or the Schilling test
were seldom ordered as initial or ancillary tests (data not shown).
Cobalamin retesting was the most frequently used follow-up test in
these patients (Table 3
). With s-cobalamin in the range of 170–189 pmol/L, 66% of the
general practitioners ordered a new test within 6 weeks. Examination of
blood smears, gastroscopy, or measurement of s-antibodies against
intrinsic factors or parietal cells were only performed in a few
patients.
|
influence of s-cobalamin concentrations on the clinical
diagnosis
We investigated whether the physicians had made a definite
diagnosis regarding the cobalamin status of their patients. The
alternatives given were "deficient", "not deficient", and
"uncertain". As shown in Table 4
, the category uncertain was the diagnosis for >50% of
the patients with s-cobalamin between 140 and 219 pmol/L and 41% of
the patients with s-cobalamin <140 µmol/L. In contrast, the majority
of general practitioners were able to make a definite diagnosis of not
deficient or deficient in patients belonging to the highest and lowest
cobalamin intervals. Only 2–9% of the patients with s-cobalamin
between 170 and 299 pmol/L were regarded as deficient, and <10% of
the patients with s-cobalamin <170 pmol/L were diagnosed as not
deficient (Table 4
).
|
In a logistic regression analysis, we used the physician's diagnosis
as the response variable (deficient = 1, not deficient = 0)
and indications for requesting s-cobalamin (Table 2
), s-cobalamin, and
the patient's age and sex as independent variables. In this model,
s-cobalamin (decreasing values in intervals of 10 pmol/L) and anemia
(no = 0; yes = 1) significantly predicted the clinical
diagnosis deficient, with odds ratios of 1.34 [95% confidence
interval (CI), 1.21–1.48] and 3.95 (95% CI, 1.08–14.5),
respectively. Patients classified as uncertain were not included in
this model.
evaluation of the diagnostic strategies
By definition, 2.5% of healthy people have values below the lower
reference limit of a quantitative laboratory test (25). If
there is prior clinical selection of the subjects to be tested, the
percentage is expected to be higher. However, we found that only 2.8%
of routinely measured s-cobalamin values were <170 pmol/L (Table 1
).
This may be explained by the indications for ordering s-cobalamin in
our survey (Table 2
). Patients with anemia and/or macrocytosis have a
relatively high pretest probability of cobalamin deficiency
(26)(29)(30)(31); however, these indications
represent a minority in our survey results. An additional explanation
might be the use of s-cobalamin as a control test during cobalamin
supplementation. Furthermore, in the majority of cases, s-cobalamin was
requested for screening purposes or for examination of diffuse
symptoms, which is also indicated by the frequent request of other
biochemical tests together with s-cobalamin. Thus, s-cobalamin is most
likely used as a case-finding test for cobalamin deficiency in general
practice (32).
The high portion of subjects categorized as uncertain diagnosis (Table 4
) may indicate that general practitioners are aware of the low
predictive value of s-cobalamin. However, there could be a bias towards
this category because general practitioners may prefer to categorize
their patients as uncertain diagnosis to avoid making a false-negative
diagnosis. The number of patients with uncertain diagnoses
increased around the lower reference limit of s-cobalamin. This
observation is corroborated by the frequent cobalamin retesting in
patients with low and low-normal values (Table 3
). In spite of this
considerable diagnostic uncertainty, ancillary tests such as the mean
corpuscular volume, a blood smear, or the intrinsic factor
antibody are seldom used.
serum concentrations of vitamins and metabolic markers
Different reference standards for functional cobalamin deficiency
have been presented. The deoxyuridine suppression test (dUST) is
theoretically sound (27)(33), but is expensive
and labor-intensive, and therefore impractical for large-scale use.
Moelby et al. (32) used an abnormal Schilling test and/or
megaloblastic bone marrow morphology, which could not be explained by
folate deficiency, as the reference standard. Increased s-MMA and/or
s-tHcy or a substantial decrease in these metabolites after
cobalamin injections have been used by other investigators
(19)(28)(34).
In 1993, s-MMA was not available as a routine analysis in Norway, and tHcy was little known and not available at FML. The physicians, therefore, made their diagnoses without knowledge of the s-MMA or tHcy values for their patients.
The distributions of increased s-MMA, tHcy, and low s-folate at
different s-cobalamin concentrations are shown in Table 5
. As expected, tHcy was higher at low s-cobalamin concentrations
and decreased as a function of the cobalamin concentration. In
contrast, s-folate showed no relationship with s-cobalamin, and
<11.5% of the patients had s-folate <5 nmol/L throughout the
s-cobalamin strata (Table 5
).
|
The relationship between s-MMA and tHcy is depicted in Fig. 1
. tHcy was increased in a substantial number of patients with
s-MMA values within the reference range, whereas only six
patients with reference values of tHcy (<15 µmol/L) had
increased s-MMA (Fig. 1
). This may reflect the lower specificity of
s-tHcy as an indicator of cobalamin deficiency (35). In
accordance with these data and with procedures advocated by others
(19)(20)(34)(36)(37),
we therefore used s-MMA >0.376 µmol/L (3 SD above the mean of the
reference range) as the reference standard in this study.
|
assessment of diagnostic accuracy of s-cobalamin
The s-cobalamin concentration was the basis for patient selection
in this study. We therefore had to correct for the frequency
distribution of s-cobalamin values when calculating the sensitivities
and specificities of s-cobalamin. This correction was done on the basis
of the frequency distribution of the total s-cobalamin analyses at FML
in 1993 (Table 1
).
We then evaluated the diagnostic sensitivity and specificity of
s-cobalamin 
170 pmol/L, using s-MMA >0.376 µmol/L as the reference
standard. The sensitivity of s-cobalamin was 0.40 (95% CI,
0.22–0.58), and the specificity was 0.98 (95% CI, 0.976–0.983). In
these calculations, we have assumed that there are no
cobalamin-deficient patients with s-cobalamin >300 pmol/L.
These results, which showed low sensitivity for s-cobalamin, are in line with other studies. Norman et al. (38) discovered that 51% (18 of 35) of elderly people with increased urinary MMA had s-cobalamin below the lower reference limit. Pennypacker et al. (28) found that 7 (37%) of 19 geriatric outpatients with s-MMA >0.376 µmol/L had s-cobalamin <148 pmol/L, and Nilsson et al. (39) estimated the sensitivity of s-cobalamin <150 pmol/L to be 41% in a psychogeriatric population. In surviving members of the original Framingham elderly population, only 12 (20%) of 59 patients with increased s-MMA and a s-creatinine value within the reference range had s-cobalamin <148 pmol/L (20).
The estimated diagnostic accuracy of a test is always influenced by the selection of the study population. The specificity will usually be lowest in populations with high prevalence of other/similar diseases, and the sensitivity will be highest in populations with advanced disease (38). Accordingly, we would expect our estimate of specificity to be high and our sensitivity to be low compared with a group of inpatients.
The overall prevalence of cobalamin deficiency in the patients having their s-cobalamin tested at FML was estimated to 2.96%, when increased s-MMA (>0.376 µmol/L) was used as the reference.
predictors of high s-mma
In multiple logistic regression analyses, we used indications for
requesting s-cobalamin (Table 2
), s-cobalamin, s-creatinine, and
patient age and sex as independent variables. In this model, increased
s-MMA was significantly predicted by low s-cobalamin (decreasing values
in intervals of 10 pmol/L), high age, and female sex (coded = 1),
with odds ratios of 1.34 (95% CI, 1.22–1.63), 1.12 (95% CI,
1.08–1.17), and 3.12 (95% CI, 1.09–9.1), respectively.
A similar relationship between s-MMA, s-cobalamin, age, and sex was found by Pennypacker et al. (28). Norman et al. (38) found increased urinary MMA in 5.3% of women compared with 1.9% of men. They found no statistically significant association between age and urinary MMA concentrations. However, their study population consisted only of persons >65 years of age.
An independent effect of age and sex on the probability of functional cobalamin deficiency at the same concentration of s-cobalamin requires substantiation.
clinical diagnosis and therapeutic conclusion in relation to
concentrations of mma and cobalamin
The relationship between the general practitioners' clinical
diagnoses and the s-MMA and s-cobalamin concentrations is illustrated
in Fig. 2
. There is a considerable discrepancy between the general
practitioners' diagnosis of "cobalamin-deficient" and the number
of patients with increased MMA (Fig. 2A
). This is especially true for
higher cobalamin concentrations. On the other hand, the agreement
between the clinical diagnosis not deficient and s-MMA values within
the reference range is very good (Fig. 2C
). Only one patient with s-MMA
>0.376 µmol/L was diagnosed as not deficient. Nine percent of
patients, whom the general practitioners had classified as uncertain,
had increased s-MMA (Fig. 2B
).
|
The decision by a general practitioner to start cobalamin
supplementation was also made when the diagnosis was uncertain. In this
case, low s-cobalamin seemed particularly influential because cobalamin
supplementation was given to patients in the uncertain category almost
exclusively when they had s-cobalamin <170 pmol/L (Table 6
).
|
In the group classified as deficient by the general practitioners, two patients were advised to change their diets, two were followed up by other general practitioners, one refused any treatment, and one died before therapy was started. The rest of the patients received supplementation with cobalamin. For the estimates below, all patients classified as deficient by the general practitioners were included in the group of patients receiving cobalamin supplements. When supplementation was used as the response indicator and given the assumption that no patients with s-cobalamin >300 pmol/L were supplemented, the clinical decision to supplement the patient had a sensitivity of 0.51 (95% CI, 0.32–0.69) and a specificity of 0.98 (95% CI, 0.976–0.983).
To compare the diagnostic efficiency of clinical judgement
(supplementation or not) with the s-cobalamin assay, again using s-MMA
as the reference standard, we performed a logistic regression analysis
and calculated "posttest" probabilities. We first used
supplementation as the response variable in an analysis with
s-cobalamin as the independent variable (Fig. 3
, inset). As expected, as s-cobalamin values decreased, the
probability that a patient would receive a supplement gradually
increased. We next used high s-MMA as the response variable and then
analyzed the following three groups separately: patients receiving
cobalamin supplementation, patients not receiving supplementation, and
all patients. The probability of true cobalamin deficiency (defined by
increased MMA) is shown separately for the three groups in Fig. 3
.
|
The importance of clinical judgement can be estimated by the difference
in the probability of cobalamin deficiency in the supplemented and
nonsupplemented groups. For example, in patients with a s-cobalamin
value of 100 pmol/L, the probability that the patient had a cobalamin
deficiency was ~0.59. In the supplemented and nonsupplemented groups,
the corresponding numbers were 0.73 and 0.31, respectively (Fig. 3
).
diagnostic benefit of s-mma determination
Our findings indicate that the clinical judgment of general
practitioners has a higher diagnostic accuracy than s-cobalamin used
without clinical information. However, the clinical diagnostic
efficiency is still not high. Therefore, we wanted to estimate the
diagnostic benefit of using s-MMA determination. The gain in diagnostic
accuracy can be expressed as the discrepancy between the clinical
diagnosis and the result of the s-MMA determination.
As an example, we can study a case where the s-cobalamin is 100 pmol/L.
For this example, we will consider s-MMA >0.376 nmol/L (and
s-creatinine within the health-related reference range) as synonymous
with cobalamin deficiency. The probability that the patient will
receive a supplement is 0.72 (Fig. 3
, inset). Among the subjects
receiving supplements , the probability of having increased s-MMA is
0.73 (Fig. 3
, main panel). Thus, the fraction correctly supplemented
(of all patients with an s-cobalamin value of 100 pmol/L) is 0.52
(0.72 x 0.73). The fraction supplemented without having a
deficiency is 0.20 (0.72 - 0.52). The fraction not supplemented
is 0.28 (1 - 0.72; Fig. 3
, inset), with a fraction of 0.69
(1 - 0.31; Fig. 3
, main panel) having s-MMA within the reference
interval. Thus, the fraction of patients correctly not supplemented is
0.19 (0.28 x 0.69), and the fraction wrongly not supplemented,
i.e., subjects who should have been supplemented, is 0.09
(0.28 - 0.19). Of all patients with an s-cobalamin value of 100
pmol/L, a fraction of 0.60 had increased s-MMA and a fraction of 0.40
had s-MMA within the reference range. The use of s-MMA as the reference
standard therefore implies a fractional increase in true-positive
diagnoses of 0.08 (0.60 - 0.52) and a fractional increase in
true-negative diagnoses of 0.21 (0.40 - 0.19). This fractional
increase in "true" diagnoses is shown in Fig. 4
A.
|
In our model, we know the fraction of patients with s-MMA >0.376
nmol/L (A) for different cobalamin concentrations. If s-MMA
is not used as the reference standard, but as a test with a certain
sensitivity (se) and specificity (sp) for
diagnosing cobalamin deficiency, samples with s-MMA >0.376 will
include samples from patients with true-positive (TP) and
false-positive (FP) diagnoses. Similarly, s-MMA 0.376
nmol/L (B) will include patients with true-negative
(TN) and false-negative (FN) diagnoses. If we use
the following (40) formulas:
 |
 | (1) |
by calculations and rearrangements, we find that:
 |
 | (2) |
These equations were used to calculate the fractions of
true-positive, false-negative, true-negative, and false-positive
diagnoses for different sensitivities and specificities for s-MMA. Fig. 4B
shows the calculated fractional diagnostic difference in
true-negative and true-positive diagnoses between s-MMA and the
"clinical diagnosis" (supplementation or not) with the MMA test
sensitivity set at 0.95 and the specificity set at 0.97.
The diagnostic performance of metabolite assays (s-MMA and s-tHcy) was reviewed extensively in a recent monograph (41). The sensitivity and specificity of s-MMA are >0.90 in most studies.
cost-benefit estimates of s-mma measurements
A simple cost-benefit analysis was made to find the s-cobalamin
interval where s-MMA should be determined. In this analysis, we
compared the costs and benefits of the additional use of s-MMA in
diagnosing cobalamin deficiency:
 | (3) |
 | (4) |
We defined in absolute values the benefit of giving supplements or
not giving supplements to a deficient patient as equal (a).
In comparison, the cost of treating and the benefit of not treating
nondeficient patients are also defined as equal (b). The
values for w in Eqs. 3
, and 4
are the different ratios of
a to b. The relative benefit between the
conventional diagnostic procedures and the use of s-MMA for the
diagnosis of cobalamin deficiency is obtained by plotting the ratio of
Eq. 3
to Eq. 4
. When this ratio is <1, s-MMA should be requested. When
the ratio is >1, s-MMA should not be requested. The ratio when the
costs of the analysis are set to 5% of b and the
sensitivity and specificity for s-MMA as the test are 100% is shown in
Fig. 5
A. We can see that the test should be requested in most
cases up to s-cobalamin values of 250 pmol/L. By using the Eqs. 1
, and 2
, we can simulate the changes of relative benefit of the MMA-test
giving the assay different sensitivities and specificities. With a
sensitivity of 95% and a specificity of 97% for s-MMA, which is
presumably a realistic assumption (35), s-MMA should be
requested if a patient has s-cobalamin values between 60 and 220
pmol/L. The conclusion is more or less independent of w
(Fig. 5B
). With a sensitivity of 90% and a specificity of 95%, s-MMA
should be requested for s-cobalamin values between 90 and 200 pmol/L
(Fig. 5C
). Even if the costs of the s-MMA-assay are changed to 20% of
the false positives (b), the effects on the conclusions of
our model are minor (data not shown), and s-MMA should be requested at
s-cobalamin values between 90 and 190 pmol/L. Thus, the model is rather
robust for small changes in the diagnostic accuracy of s-MMA, as well
as for alterations in the cost-benefit weights and different estimates
of the costs of the MMA analysis.
|
|
Conclusions |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
Our data emphasize the poor diagnostic utility of low and low-normal s-cobalamin assay results and call for more sensitive and specific markers of cobalamin deficiency. s-MMA meets these criteria, and if used, our data suggest routine measurement in patients when s-cobalamin is >60–90 pmol/L and <200–220 pmol/L, depending on the diagnostic accuracy of s-MMA.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionPatients and MethodsResults and DiscussionConclusionsReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  H.-A. Lakso, P. Appelblad, and J. Schneede Quantification of Methylmalonic Acid in Human Plasma with Hydrophilic Interaction Liquid Chromatography Separation and Mass Spectrometric Detection Clin. Chem., December 1, 2008; 54(12): 2028 - 2035. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Schmedes and I. Brandslund Analysis of methylmalonic Acid in plasma by liquid chromatography-tandem mass spectrometry. Clin. Chem., April 1, 2006; 52(4): 754 - 757. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Gilfix Vitamin B12 and homocysteine Can. Med. Assoc. J., November 22, 2005; 173(11): 1360 - 1360. [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Solomon Cobalamin-responsive disorders in the ambulatory care setting: unreliability of cobalamin, methylmalonic acid, and homocysteine testing Blood, February 1, 2005; 105(3): 978 - 985. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Refsum, A. D. Smith, P. M. Ueland, E. Nexo, R. Clarke, J. McPartlin, C. Johnston, F. Engbaek, J. Schneede, C. McPartlin, et al. Facts and Recommendations about Total Homocysteine Determinations: An Expert Opinion Clin. Chem., January 1, 2004; 50(1): 3 - 32. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J.G. Ray and H.J. Blom Vitamin B12 insufficiency and the risk of fetal neural tube defects QJM, April 1, 2003; 96(4): 289 - 295. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. S. Morris, P. F. Jacques, I. H. Rosenberg, and J. Selhub Elevated Serum Methylmalonic Acid Concentrations Are Common among Elderly Americans , J. Nutr., September 1, 2002; 132(9): 2799 - 2803. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Matern and M. J. Magera Mass Spectrometry Methods for Metabolic and Health Assessment J. Nutr., May 1, 2001; 131(5): 1615S - 1620. [Abstract] [Full Text]
|
|
|
|
|
|
B. J. Bolann, J. D. Solli, J. Schneede, K. A. Grottum, A. Loraas, M. Stokkeland, A. Stallemo, A. Schjoth, R. B. Bie, H. Refsum, et al. Evaluation of Indicators of Cobalamin Deficiency Defined as Cobalamin-induced Reduction in Increased Serum Methylmalonic Acid Clin. Chem., November 1, 2000; 46(11): 1744 - 1750. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. J. Magera, J. K. Helgeson, D. Matern, and P. Rinaldo Methylmalonic Acid Measured in Plasma and Urine by Stable-Isotope Dilution and Electrospray Tandem Mass Spectrometry Clin. Chem., November 1, 2000; 46(11): 1804 - 1810. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. G. Klee Cobalamin and Folate Evaluation: Measurement of Methylmalonic Acid and Homocysteine vs Vitamin B12 and Folate Clin. Chem., August 1, 2000; 46(8): 1277 - 1283. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Pfeiffer, S. J. Smith, D. T. Miller, and E. W. Gunter Comparison of Serum and Plasma Methylmalonic Acid Measurements in 13 Laboratories: An International Study Clin. Chem., December 1, 1999; 45(12): 2236 - 2242. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||
, patients with increased s-creatinine; 
, patients with low
s-folate; 
, patients with s-creatinine and s-folate within the
reference range.
), all patients (——–), and
nonsupplemented patients (
