| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Technical Briefs |
a address for correspondence: Dept. of Lab. Med., Univ. of Connecticut Health Center, MC-2235, 263 Farmington Ave., Farmington, CT 06030-2235: fax 860-679-2154, e-mail makowski{at}nso1.uchc.edu
Good evidence exists that increased maternal plasma/red blood cell folate concentrations is associated with fewer neural tube defects in newborns (1)(2). This finding has led to the suggestion that women pre- and periconception maintain folate adequacy—the concentration of which remains controversial (3)(4)(5). Equally important yet generally overlooked is the variety of analytical methods used for folate measurement. For example, review of data obtained from ~1400 clinical laboratories participating in the 1995 College of American Pathologists (CAP) analytical survey (6) yields striking differences in plasma folate measurement. In three challenges during 1995 (K-A, K-B, and K-C), laboratories were asked to determine folate concentration in nine plasma samples (red blood cell folate was not measured). Survey data were grouped by methodology and statistical analysis was performed. Only results obtained from at least 20 laboratories reporting were considered for this study (>90% data).
Large differences in mean plasma folate concentration were observed
when a manual radioisotopic immunoassay was compared with an automated
fluorescent method (Fig. 1
A). Of the nine samples evaluated during this survey period, the
most discrepant (comparison of mean) plasma folate results were seen in
six samples analyzed by these two methods throughout the concentration
range tested (marked with an asterisk, Fig. 1A
). Differences in the
nine folate results obtained by these methods ranged from 144% to
264% and placed first (6), second (1), or
third (2) in degree of discrepancy when compared with all
methods cited in the survey (Table 1
). Comparison of results for mean plasma folate concentration
obtained with the manual vs automated method revealed considerable bias
(Fig. 1B
) [e.g., an automated plasma folate concentration of 5 µg/L
(adequate) would be considered marginal or inadequate (2.5 µg/L) by
manual method]. Average CV was not, however, substantially different
for the automated and manual method: 9.6% (range 4.7–12.2%) and
10.5% (range 8.2–16.6%), respectively. Also notable is the trend in
the number of laboratories using these methods (Fig. 1C
). The manual
RIA method was most cited in K-A, whereas the automated method was most
cited in K-C.
|
|
The shift in plasma folate methodology may reflect decreased reliance on radioisotopes in general as well as marketplace vagaries. It is reasonable that the dramatic changes in the healthcare industry have prompted the clinical laboratory to choose automated cost-saving methods. Although the advantages of analytical precision and accuracy can certainly be debated (7)(8)(9), it is evident that clinicians must be aware of these seldom-explored issues as they affect patient care. Establishment of method-dependent reference intervals will minimize this type of discrepancy. Future resolutions may require improved assay standardization and use of an international reference material.
Footnotes
Div. of Clin. Chem., Dept. of Lab. Med., Univ. of Connecticut School of Med., Farmington, CT 06030
References
The following articles in journals at HighWire Press have cited this article:
|
|
 |
|
  E. Nakano, F. A. Taiwo, D. Nugent, H. R. Griffiths, S. Aldred, M. Paisi, M. Kwok, P. Bhatt, M. H. E. Hill, S. Moat, et al. Downstream effects on human low density lipoprotein of homocysteine exported from endothelial cells in an in vitro system J. Lipid Res., March 1, 2005; 46(3): 484 - 493. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
