| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Clinical Chemistry, Turku University Central Hospital, FIN-20521 Turku, Finland.
2 Härkätie Health Centre, FIN-21421 Lieto, Finland.
3 Department of Clinical Biochemistry, Aarhus University Hospital at Skejby, DK-8200 Aarhus N, Denmark.
4 Institute of Clinical Medicine, General Practice, University of Turku, FIN-20041 Turku, Finland.
5 Satakunta Central Hospital, FIN-28500 Pori, Finland.
aAuthor for correspondence. Fax 358-2-2613920; e-mail saila.loikas{at}tyks.fi.
 |
Abstract |
|---|
 Top Abstract IntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
Methods: The reference population consisted of 303 individuals 22–88 years of age, without disease or medication affecting Cbl or homocysteine metabolism. In elderly individuals (
65 years), normal Cbl status was further confirmed by total homocysteine (tHcy; <19 µmol/L) and methylmalonic acid (MMA; <0.28 µmol/L) concentrations within established reference intervals. HoloTC in Cbl deficiency was studied in a population of 107 elderly individuals with normal renal function. The Cbl deficiency was graded as potential (total Cbl 
150 pmol/L or tHcy 19 µmol/L), possible (total Cbl 150 pmol/L and either tHcy 19 µmol/L or MMA 0.45 µmol/L), and probable (tHcy 19 µmol/L and MMA 0.45 µmol/L).
Results: The intra- and between-assay imprecision (CV) for the holoTC RIA were 4–7% and 6–8%, respectively. A 95% central reference interval for serum holoTC was 37–171 pmol/L. All participants (n = 16) with probable Cbl deficiency, 86% of those with possible, and 30% of those with potential Cbl deficiency had holoTC below the reference limit (<37 pmol/L). The holoTC correlated with total Cbl (rs = 0.80; P <0.0001) and inversely with MMA (rs = -0.52; P <0.0001). HoloTC concentrations were significantly (P = 0.01) higher in women than in men.
Conclusions: The new holoTC RIA is precise and simple to perform. Low holoTC is found in individuals with biochemical signs of Cbl deficiency, but the sensitivity and specificity of low holoTC in diagnosis of Cbl deficiency need to be further evaluated.
|
Introduction |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
60 min (7)(8). The holoTC concentration may be reduced by impaired generation of holoTC either because of Cbl malabsorption or because of depletion of Cbl reserves in general Cbl deficiency. Tissue requirements for Cbl and hepatic and renal impairment may also affect the holoTC concentration in blood (9)(10)(11). The clinical usefulness of holoTC measurements has not been thoroughly evaluated, probably because there have not been methods suitable for routine use (10). The majority of the methods described measure holoTC indirectly as the difference between total Cbl and the amount of haptocorrin-bound Cbl. Because the concentrations of these variables are high, a small holoTC value may not be precise (4)(5)(12). Only one in-house method that directly quantifies holoTC has been described previously (6), but it is too laborious and imprecise for routine use.
Recently, two immunologic methods for measuring holoTC have been reported (13)(14). The HoloTC RIA method described by Ulleland et al. (14) is the first commercially available method for measurement of holoTC. HoloTC in the serum sample is separated by magnetic microspheres coated with monoclonal anti-human TC antibodies; Cbl bound to holoTC is then released and measured by a competitive binding RIA standardized with recombinant human holoTC. In the method described by Nexø and coworkers (13)(15), the Cbl-binding apoproteins in serum are precipitated with magnetic beads coated with Cbl and the holoTC remaining in the supernatant is measured by ELISA.
In the present study, we evaluated the suitability of the HoloTC RIA for routine use in a clinical laboratory. We determined the reference values for the method in adult and elderly Finnish populations and assessed whether low holoTC concentrations are associated with other biochemical signs of Cbl deficiency [decreased serum total Cbl and increased serum methylmalonic acid (MMA) and plasma total homocysteine (tHcy)].
|
Participants and Methods |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
65 years). The Lieto Study is a population-based health survey of an unselected elderly population living in this area (16). All elderly Lieto residents 65 years of age or over (n = 1596) were invited: 273 refused, 63 died before data collection, and 1260 (82%) participated. The data of the Lieto Study were collected by the Department of General Practice of the University of Turku and the Department of Clinical Chemistry of the Turku University Central Hospital in cooperation with the municipal health center of Lieto during a period from March 1998 to December 1999. The Lieto Study has been approved by the Ethics Committee of the Turku University Central Hospital. All participants gave written informed consent. The healthy elderly individuals selected for the holoTC reference values (n = 226) and those selected for evaluating holoTC in relation to other biochemical markers of Cbl deficiency (n = 107) were chosen from among this large study population (Fig. 1
).
|
elderly individuals for assessment of reference values
The group of elderly for assessment of reference values was selected among participants of the Lieto Study population (Fig. 1
). The exclusion criteria were any medication known to affect Cbl, folate, or Hcy metabolism; smoking; alcoholism; excessive coffee consumption (nine or more cups per day); previously diagnosed anemia or Cbl or folate deficiency; cardiovascular disease; dementia; renal failure; gastric or intestinal diseases that could cause malabsorption; liver or pancreatic disorders; type I or type II diabetes with treatment other than diet; Parkinson disease; polyneuropathy; lymphoproliferative diseases; or any autoimmune disease except treated hypothyroidism. In addition, 65 participants were excluded because adequate samples for plasma tHcy measurements were not available. Of the remaining 310 participants, 63 were excluded because of hyperhomocysteinemia (tHcy 15 µmol/L) as a marker of potential Cbl or folate deficiency. The individuals with the lowest one-third of holoTC results had serum MMA measured. Of these, 21 were excluded for having MMA above the upper reference limit (0.28 µmol/L), which is compatible with possible Cbl deficiency.
The remaining 226 participants (18% of the total Lieto Study population) constituted the healthy elderly reference group (93 men and 133 women). They had a mean age of 71 years (range, 65–88 years). Of the reference group, 51 (23%) received no medication. The average number of drugs used was two (range, zero to nine drugs). Only 16 individuals (7%) took five or more drugs. Fourteen (6%) individuals had no previously diagnosed diseases. The average number of previously diagnosed diseases was 4 (range, 0–12).
adult participants for assessment of reference values
The individuals in the nonelderly adult reference group were volunteers recruited among the laboratory personnel and their spouses in the Turku area. The participants met the following criteria: typical Finnish diet, no Cbl- or folate-containing vitamin supplementation, no chronic diseases, and no medications that influence Cbl, folate, or tHcy metabolism. The use of other vitamin supplementation or oral contraceptives was not an exclusion criterion, nor was smoking or coffee consumption. The data for this group were collected during March and April 2001. Serum total Cbl, serum holoTC, and complete blood counts were measured. Of the 84 volunteers enrolled, 1 was excluded because of a vegetarian diet and 3 were excluded because of missing hematologic laboratory values. The remaining 80 participants were apparently healthy Finnish adults (22 men and 58 women) with a mean age of 45 years (range, 22–62 years).
holoTC assessment in presumed Cbl deficiency
Participants with a markedly increased plasma tHcy concentration (19 µmol/L) or a low serum total Cbl concentration (150 pmol/L) were considered most likely to have Cbl deficiency. There were 183 (15%) individuals who fulfilled one or both criteria in the Lieto Study population (Fig. 1
). Serum MMA was determined in these selected individuals to confirm Cbl deficiency.
To avoid the influence of renal insufficiency on the concentrations of the biochemical markers for Cbl deficiency (tHcy and MMA), a serum cystatin C (cysC) concentration of 1.3 mg/L was used as the cutoff limit to exclude individuals with impaired glomerular filtration rate. On the basis of this definition, 76 of 183 individuals (42%) had impaired glomerular filtration rate, leaving 107 individuals in the study group for assessment of holoTC in Cbl deficiency.
The likelihood of Cbl deficiency in these individuals was graded as potential [low serum total Cbl (150 pmol/L) or high plasma tHcy (19 µmol/L)], possible [low serum total Cbl (150 pmol/L) and either high plasma tHcy (19 µmol/L) or high serum MMA (0.45 µmol/L)], or probable [high plasma tHcy (19 µmol/L) and high serum MMA (0.45 µmol/L)]. The cutoff limit of 0.45 µmol/L for serum MMA was calculated from the within-person variation (13%) and the upper reference limit (0.28 µmol/L) of the method (17)(18).
specimens
Venous blood samples were drawn after an overnight fast, with light stasis and the individual sitting. The blood count was analyzed from EDTA-anticoagulated samples within 4 h after sampling. Heparin-anticoagulated samples were placed immediately on ice and centrifuged (2110g for 10 min at 4 °C) within 1 h to obtain plasma. The samples in tubes without anticoagulant were allowed to clot at room temperature, and the serum was separated by centrifugation (2110g for 10 min) within 1 h. The serum samples from elderly participants were stored at -70 °C for 2–3 years before the holoTC assay. The samples from adult participants were stored at -20 °C and were analyzed within 2.5 months. Samples were not thawed and refrozen before holoTC assay.
laboratory analyses
The complete blood counts were measured on an Advia 120 hematology analyzer (Bayer Corporation). Serum total Cbl was analyzed with a competitive protein binding assay (AutoDelfia; Wallac; CV = 2% at a mean concentration of 238 pmol/L). The plasma creatinine concentration was determined with the Jaffe method on a Hitachi 917 Automatic Analyzer (Roche Diagnostics GmbH; CV = 2% at a mean concentration of 118 µmol/L). Serum cysC was measured with an immunonephelometric assay on a Behring nephelometer (Dade Behring; CV = 3% at a mean concentration of 1.7 mg/L). Plasma tHcy was measured with fluorescence polarization assay on an IMx system (Abbott Laboratories; CV = 2% at a mean concentration of 13 µmol/L). Serum MMA was determined with stable-isotope-dilution capillary gas chromatography-mass spectrometry [Ref. (17); CV = 6% at a mean concentration of 0.38 µmol/L].
Serum holoTC was measured directly with a new commercial RIA (HoloTC RIA; Axis Shield ASA) (14). Briefly, 400 µL of serum was diluted with 400 µL of 0.1 mol/L phosphate-buffered saline, and magnetic microspheres coated with anti-TC antibodies were added. Samples were incubated for 1 h on a roller mixer at room temperature. After the incubation, the holoTC attached to microspheres was precipitated on a magnetic rack, the supernatant was discarded, and the precipitate was washed with phosphate-buffered saline. The holoTC concentration was determined with a RIA standardized with recombinant human holoTC, using intrinsic factor as a binder and 57Co-labeled vitamin B12 as a tracer. The calibrators that were used to construct the calibration curve were processed identically to the samples. All samples were analyzed in duplicate.
statistical analyses
Imprecision results are presented as intra- and between-assay CVs, mean concentration, and SD. Differences between groups were tested with the Kruskal-Wallis nonparametric test for quantitative data and with the 
2 test for categorical data. A P value <0.05 was considered significant. The reference intervals for holoTC were calculated with a parametric method, after log transformation because the holoTC values were positively skewed, and presented as the central 95% interquantile interval with 95% confidence limits. For measuring associations, the Spearman rank-order correlation coefficients were calculated.
Data were analyzed using the SAS System for Windows, release 8.02 (SAS Institute, Inc.).
|
Results |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
.
|
The lower limit of the reportable range was considered the concentration of the lowest calibrator, i.e., 20 pmol/L. This limit was chosen because the precision of the low patient sample pool was acceptable (CV = 14% at a mean of 25 pmol/L), whereas the precision of the very low patient sample pool was poor (CV = 32% at a mean of 5 pmol/L). The greater imprecision at lower concentrations justifies the measurement of holoTC in duplicate in routine work. The concentration of the highest calibrator (320 pmol/L) was considered a reasonable upper limit of the reportable range. Linearity was not tested.
stability of holoTC during storage
The stability of holoTC during storage was tested in nine samples with holoTC concentrations within the reportable range (range, 36–210 pmol/L). The samples were stored at -70 °C for 7–16 months. There was no statistically significant difference between initial concentrations and those measured after storage at -70 °C (mean of differences, 4 pmol/L; P = 0.56).
reference values
There was no statistically significant difference between the age groups in serum holoTC concentrations (mean, 84 vs 87 pmol/L; P = 0.35), total Cbl concentrations, or hematologic indices. Thus, the adult and elderly groups were combined for determination of the reference interval. HoloTC concentrations were significantly (P = 0.02) higher in women than in men (mean, 90 vs 79 pmol/L). At lower concentrations (100 pmol/L), however, there was no statistically significant gender difference. Because only low concentrations are relevant with regard to the diagnosis of Cbl deficiency, we did not calculate separate reference intervals for men and women. After the exclusion of 3 values more than 3 SD from mean as outliers, the reference population included 303 individuals. Their characteristics are summarized in Table 2
.
|
The holoTC concentration range was 25–254 pmol/L, with a geometric mean of 79 pmol/L. The 95% central reference interval was 37–171 pmol/L and was calculated after log transformation of the data. The 95% confidence interval for the lower reference limit was 35–38 pmol/L, and that for the upper reference limit was 164–179 pmol/L. Exclusion of nonelderly adults (<65 years) did not change the reference interval.
holoTC in Cbl deficiency
In the whole Lieto Study population (n = 1055), 10% (109 of 1055) of participants had serum holoTC concentrations below the reference limit (<37 pmol/L). There was a significant correlation between serum holoTC and total Cbl (rs = 0.80; P <0.0001; Fig. 2
). Both total Cbl and holoTC correlated inversely with plasma tHcy (rs = -0.31; P <0.0001 and rs = -0.32; P <0.0001, respectively). Serum MMA was not measured in the whole Lieto population. The holoTC concentration was significantly (P <0.0001) higher in women than in men (mean concentration, 95 vs 77 pmol/L).
|
In the study group of 107 individuals for assessment of holoTC in Cbl deficiency, 48% (51 of 107) had holoTC below the reference limit (<37 pmol/L; Table 3
). Both holoTC and total Cbl correlated inversely with MMA (rs = -0.52; P <0.0001 and rs = -0.47; P <0.0001, respectively). There was no correlation between tHcy and holoTC, but we found a weak positive correlation between total Cbl and tHcy (rs = 0.22; P = 0.02). When the likelihood of Cbl deficiency was graded as potential, possible, or probable (Fig. 1
), the frequencies of low holoTC in these groups were 30% (23 of 77), 86% (12 of 14), and 100% (16 of 16), respectively (Fig. 3
).
|
|
In the probable Cbl deficiency group, serum total Cbl was low (150 pmol/L) in 81% (13 of 16) of individuals, and all of them had low holoTC. The difference was not statistically significant (P >0.05). In the potential Cbl deficiency group, 43% (10 of 23) of individuals with low holoTC and high tHcy had total Cbl within the reference interval.
In the healthy elderly group (n = 224), 4 individuals (2%) had low holoTC. When we compared holoTC concentrations in different likelihood groups and healthy elderly, there was a statistically significant difference between all groups (P <0.0001 between others, and P = 0.004 between possible and probable Cbl deficiency groups), with holoTC concentrations being lowest in the probable group and highest in the healthy elderly group.
|
Discussion |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
Ulleland et al. (14) reported that the working range (CV <20%) of the HoloTC RIA method was at least 5–370 pmol/L, but we consider the limit of functional sensitivity to be closer to the concentration of the lowest calibrator, i.e., 20 pmol/L. Because there is less clinical interest for exact values below the lower reference limit (37 pmol/L), we suggest that the concentration of 20 pmol/L be set as the limit for the reportable range in clinical use. The precision of the HoloTC RIA is good for a manual method for values within the reference interval and acceptable between 20 and 320 pmol/L.
In our study, the central 95% reference interval for serum holoTC in 303 healthy adults and elderly was 37–171 pmol/L (geometric mean, 79 pmol/L). These values are consistent with the values obtained previously by methods that separate TC from haptocorrin (4)(6) and also by a new method described by Nexø et al. (13) in which apoproteins are first separated from holoproteins and then the TC fraction of the holoproteins is directly measured by ELISA.
Surprisingly, however, our reference values were higher than those reported by Ulleland et al. (14) with the same method (24–157 pmol/L; mean, 61 pmol/L). The population in which these reference values were assessed consisted of 105 apparently healthy volunteers 20–80 years of age. This difference between their results and ours may arise from the smaller population in their study or the different methods for calculating the reference interval. The selection of the reference group may also cause the difference. Ulleland et al. (14) described their reference group as being apparently healthy but gave no further specifics; thus, individuals with undiagnosed Cbl deficiency may, in contrast to our reference group, have been included, thus decreasing the lower reference interval. The reference group in our study was selected carefully to exclude individuals with any conditions or medications that might cause Cbl deficiency. Potentially Cbl-deficient individuals were excluded from our elderly population on the basis of increased tHcy and MMA concentrations. The reliability of this selection method is further corroborated by the narrow confidence interval (35–38 pmol/L) we obtained for the lower reference limit (37 pmol/L).
Although there was no statistically significant difference in holoTC concentrations between the age groups below and above 65 years in our study, there are some obvious risks in combining these groups for determination of the reference interval. One risk is that the samples were stored differently: samples from the adult group were stored at -20 °C for no more than 2.5 months before holoTC analysis, whereas samples from the elderly group were stored for 2–3 years at -70 °C. To our knowledge, no studies on the stability of holoTC during storage have been published. According to our preliminary studies, no change occurred in holoTC concentrations during 16 months of storage at -70 °C. Therefore, and because none of the samples had been thawed before the holoTC assay, we consider it reasonable to combine the adult and the elderly reference value populations regardless of the differences in storage.
In addition, the elderly reference value group was selected carefully not to include any individuals with signs of impaired Cbl metabolism. Because Cbl deficiency is known to be more common among elderly than younger, nonelderly adults, the healthy adult participants were not as intensively investigated but were selected not to have diseases, medications, or diets that could influence Cbl metabolism. There is no physiologic reason to expect different distributions of holoTC concentrations in healthy adult and elderly populations. As predicted, we did not detect statistically significant differences between these groups for serum holoTC or total Cbl concentrations or for hematologic indices. We thus consider it reasonable to combine the healthy adult and elderly groups to achieve clinically relevant reference values for serum holoTC concentrations.
Interestingly, the mean holoTC concentration was significantly higher in women than in men in our reference group. The same phenomenon was evident in the whole elderly population of the Lieto Study. Other studies have not reported similar sex differences for holoTC concentrations, but serum total TC has been reported to be higher in women older than 50 years than in women younger than 50 years (15). Importantly, however, the difference between sexes was evident only at concentrations close to the upper reference limit, which is irrelevant in the context of diagnosing Cbl deficiency.
The proportion of low holoTC concentrations in the whole elderly population was 10%, which equals the prevalence of Cbl deficiency defined by other criteria reported previously in elderly populations (19)(20). Earlier, Herbert(21) reported low holoTC concentrations in 35% of 150 elderly people 65–95 years of age.
There is, at the moment, no gold standard or true reference method to diagnose subtle Cbl deficiency, which makes evaluation of the clinical usefulness of holoTC and the estimation of sensitivity and specificity problematic. In this study, we aimed to assess whether low holoTC concentrations are congruent with other biochemical signs of Cbl deficiency. We found that all individuals with probable Cbl deficiency, defined as a plasma tHcy concentration 19 µmol/L and serum MMA concentration 0.45 µmol/L, also had low holoTC, which is in accordance with previous studies (5)(6)(13). The effect of even mild renal impairment was eliminated by excluding individuals with increased cysC. The proportion of individuals with low holoTC decreased with decreasing likelihood of Cbl deficiency, being less than one-third in individuals with only one other positive biochemical marker for Cbl deficiency and only 2% in healthy individuals. Taken together, we found low holoTC in individuals with Cbl deficiency defined with conventional biochemical markers.
In our study, it was not possible to adequately compare holoTC and total Cbl measurements and determine which test is better in identifying individuals with Cbl deficiency because we used low total Cbl concentration as an inclusion criterion when selecting individuals with Cbl deficiency. There was no statistically significant difference between the proportions of pathologic holoTC and total Cbl values in the group with a probable Cbl deficiency. In addition, it is important to observe that there were individuals with low holoTC and high tHcy but serum MMA and total Cbl concentrations within the respective reference intervals. This may indicate poor specificity of low holoTC for Cbl deficiency, which has also been reported in a few previous studies (6)(9). On the other hand, these individuals may be in the early stages of developing a Cbl deficiency when metabolic changes have not yet become evident, although this cannot be taken for granted without a follow-up study. Moreover, if low holoTC concentrations are explained primarily by impaired absorption of Cbl, even temporary dietary restrictions or short-term use of drugs that cause malabsorption may decrease the values (9).
In conclusion, we have verified and expanded previous findings that the new HoloTC RIA method is reliable and simple to use, which makes it suitable for routine laboratory work. Low holoTC was found in individuals with biochemical signs of Cbl deficiency. However, further studies assessing Cbl malabsorption and the influence of Cbl intake on holoTC are needed to confirm the clinical utility and specificity of holoTC in diagnosing early Cbl deficiency.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionParticipants and MethodsResultsDiscussionReferences |
|---|
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  M. Tettamanti, M. T. Garri, A. Nobili, E. Riva, and U. Lucca Low Folate and the Risk of Cognitive and Functional Deficits in the Very Old: The Monzino 80-plus Study J. Am. Coll. Nutr., December 1, 2006; 25(6): 502 - 508. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Refsum, C. Johnston, A. B. Guttormsen, and E. Nexo Holotranscobalamin and Total Transcobalamin in Human Plasma: Determination, Determinants, and Reference Values in Healthy Adults Clin. Chem., January 1, 2006; 52(1): 129 - 137. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. A. Dhonukshe-Rutten, M. van Zutphen, L. C. de Groot, S. J. Eussen, H. J Blom, and W. A van Staveren Effect of supplementation with cobalamin carried either by a milk product or a capsule in mildly cobalamin-deficient elderly Dutch persons Am. J. Clinical Nutrition, September 1, 2005; 82(3): 568 - 574. [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]
|
|
|
|
|
|
R. Obeid, H. Schorr, R. Eckert, and W. Herrmann Vitamin B12 Status in the Elderly as Judged by Available Biochemical Markers Clin. Chem., January 1, 2004; 50(1): 238 - 241. [Full Text] [PDF]
|
|
|
|
|
|
Z. Lloyd-Wright, A.-M. Hvas, J. Moller, T. A.B. Sanders, and E. Nexo Holotranscobalamin as an Indicator of Dietary Vitamin B12 Deficiency Clin. Chem., December 1, 2003; 49(12): 2076 - 2078. [Full Text] [PDF]
|
|
|
|
|
|
A. McCaddon, P. Hudson, C. McCracken, R. Ellis, and A. McCaddon Biological Variation of Holo-Transcobalamin in Elderly Individuals Clin. Chem., September 1, 2003; 49(9): 1542 - 1544. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
, plasma tHcy concentration <19 µmol/L; *, plasma tHcy concentration 