Fsnep.ucdavis.edu

Supplemental Material can be found at: Erythrocyte Folate and Its Response to Folic Acid Supplementation Is
Assay Dependent in Women1,2
Andrew J. Clifford,3 Elizabeth M. Noceti, Amy Block-Joy, Torin Block,* and Gladys Block† Department of Nutrition, University of California, Davis, CA 95616; *Block Dietary Data Systems, 2634 LeConte Avenue, Berkeley, CA 94709; and †University of California, Berkeley School of Public Health,Berkeley, CA 94720 Optimizing folate status requires continued monitoring of erythrocyte (RBC) folate and folate intake.
The accuracy of RBC folate assays remains a concern. Therefore, we measured RBC folate with 4 different assays,examined the interassay correlations, and compared RBC folate with folate intake as measured by an abbreviatedfolate-targeted food/supplement screener. The screener had 21 questions (19 diet, 2 supplement) and measuredusual and customary intakes of dietary folate equivalents (DFEs). Our design was a 4 ⫻ 2 ⫻ 2 factorial, 4 assaysin pregnant and nonpregnant women before and after each group received a folic acid supplement (1814 nmol/d)for 30 – 60 d. Folate assays included L. casei, chemiluminescence, GC-MS, and radioassay (RA). Baseline RBC folate levels ranked low to high by assay (mean ⫾ SE) were as follows: 1155 ⫾ 44 nmol/L (L. casei) ⬍ 1390 ⫾ 43nmol/L (chemiluminescence) ⬍ 1531 ⫾ 39 nmol/L (GC-MS) ⬍ 1727 ⫾ 55 nmol/L (RA) (P ⬍ 0.0001). Supplemen-tation raised RBC folate levels (mean ⫾ SE) as follows: 138 ⫾ 63 nmol/L (chemiluminescence) ⬍ 267 ⫾ 64 nmol/L(GC-MS) ⫽ 285 ⫾ 75 nmol/L (L. casei) ⬍ 351 ⫾ 87 nmol/L (RA). Pregnant women had higher RBC folate thannonpregnant women using chemiluminescence and RA. Interassay correlations (r) ranged from 0.4679 to 0.8261 (P ⬍ 0.001). Correlations of RBC folate with folate intake ranged from 0.2676 to 0.4622 (P ⬍ 0.0004). We concludethat RBC folate levels are assay dependent, as is the definition of optimized status; there continues to be a needfor an accurate assay of RBC folate. RBC folate correlated with total folate intake using a folate-targetedfood/supplement screener.
J. Nutr. 135: 137–143, 2005.
KEY WORDS: RBC folate analytical method low-income women folate-targeted food screener Adequate intake of folate is important to minimize the risk Women who fall below 185% of the Federal Poverty Level of neural tube defects (NTDs)4 (1,2), cardiovascular disease (FPL ⱕ 1.85) are an important target population group for (3), and some cancers (4). An adequate folate intake by improving folate status. They may not fully appreciate the women of child-bearing age is especially important because it importance of folate (8). The correlation between folate in- can raise RBC folate levels, and has been shown to reduce the take and RBC folate concentrations has not been adequately risk of NTDs by ⬃70% (5). The RBC folate level is rated a studied in this population; we were prompted to measure their good biomarker for nutritional status of this important nutri- folate status using their RBC folate levels and to examine its ent because of its correlation with liver folate, a significant correlation with their folate intake. The intake was measured tissue store (6). Cell accumulation of folate may be key to with a new abbreviated folate-targeted food/supplement fre- ameliorating NTD susceptibility with increased folate intake quency screener that was developed for this study.
(7). Therefore, the strength of the relations between folate Plasma and serum folate concentrations are commonly ac- intake and biomarkers of folate status such as RBC folate is of cepted to reflect recent dietary folate intake. Although eryth- rocyte (RBC) folate is considered to be a better indicator ofbody stores and hence nutritional status (9), there is consid- erable uncertainty about the reliability of the analytical meth- The semiquantitative Block Dietary Folate Equivalents screener is available with the online posting of this paper at www.nutrition.org.
ods for RBC folate (10,11). An interlaboratory comparison of 2 Supported by Grant/Cooperative Agreement (RO1 8928) from the Centers popular analytical methods for folate revealed large differences for Disease Control and Prevention (CDC) and National Institutes of Health DK among laboratories analyzing a common set of serum and 45939. The contents are the sole responsibility of the authors and do not neces-sarily represent the official views of the CDC. The microbiologic assays were whole-blood specimens (10). Results of the interlaboratory conducted in the Clinical Nutrition Research Unit, at the University of California, comparison prompted the recent development of promising Davis, which is supported by Grant DK 35747 from NIH.
new hyphenated methods that included LC-MS and LC- To whom correspondence should be addressed.
MS-MS for plasma and serum folate (12–15) and GC-MS for 4 Abbreviations used: DFE, dietary folate equivalents; FPL, federal poverty RBC folate (16 –18). Like the microbiologic assay methods, level; NF, natural food; NHANES, National Health and Nutrition Examination the new hyphenated methods can be internally and externally Survey; NTD, neural tube defect; PCV, packed cell volume; RA, radioassay; SFA,synthetic folic acid; WBC, white blood cell; WIC, Women, Infants, and Children.
0022-3166/05 $8.00 2005 American Society for Nutritional Sciences.
Manuscript received 4 August 2004. Initial review completed 8 September 2004. Revision accepted 1 October 2004.
CLIFFORD ET AL.
Here we present a feasibility study in which we compared a vitamin tablets (with the same amount of folic acid/tablet, Longs GC-MS assay to 3 other common assays for RBC folate. Also, Drug Store Brand) to match the pregnant group. In addition to folic we examined the correlation between RBC folate levels and acid, each tablet also supplied vitamin B-12, 4 ␮g (3 nmol); vitamin folate intake as measured by an 18-item folate-targeted food/ B-6, 2.6 mg (11 ␮mol); and iron, 27 mg (490 ␮mol). All women were supplement Dietary Folate Equivalents (DFE) screener devel- instructed to take 1 prenatal tablet/d until their 2nd clinic visit. A$25 gift certificate to a local supermarket (for food only) was given to oped for this study. The time required to complete the DFE each woman at the end of the visit.
screener (6 –12 min) makes it an attractive alternative to The 2nd clinic visit occurred 30 – 60 d after supplementation first longer FFQs and for epidemiologic investigations focused on began. Only 3 of 38 nonpregnant women were lactating at this visit.
folate status. If further research demonstrates the validity of All women again completed the 1-page folate-targeted food/supple- the DFE screener when it is self-administered, as opposed to ment intake screener, and blood was drawn again from each woman administered by clinic staff via interview, this instrument may as described above. All women were queried about compliance with prove useful as a classification tool for clinical settings.
supplement use, i.e., how often they took/missed taking the supple-ment. The overall compliance was determined from the 1-page folate-targeted food/supplement intake screener, interviews, and queries,.
Pregnant women were asked about their start date for receiving WIC SUBJECTS AND METHODS
benefits. At the end of the visit, another $25 gift certificate to a localsupermarket (for food only) was given to each woman.
The University of California Davis Institutional Review Board Folate-Targeted Semiquantitative Block DFE screener.
approved the study, and it was conducted according to Good Clinical semiquantitative Block Dietary Folate Equivalents (DFE) screener Practice guidelines and the Declaration of Helsinki, version 1989.
was designed to measure usual and customary intake of DFEs and was Informed consent was discussed and obtained from each woman optimized for use in low-income populations.1 The questions on the before study participation. Women were excluded for any history of screener were identified by examining the food records reported by serious medical conditions or use of medications that would interfere individuals ⱖ 17 y old in the National Health and Nutrition Exam- with folate metabolism (dilantin, phenytoin, primidone, metformin, ination Survey (NHANES) 1999 –2000. To develop the screener, all jn.nutrition.org sulfasalazine, triamterene, or methotrexate). Intermittent users of a 4368 food items and beverages in that dataset were divided into 152 multivitamin supplement (1–2 times/wk) were included, whereas food groups (questions). Each question represented a similar type of prenatal-vitamin supplement users were excluded. All pregnant food item (for example, pinto, red, black, or refried beans, separately, women were enrolled at the time of their first prenatal visit and were or as part of another food such as a "burrito," were coded together as not taking any prenatal vitamins.
beans). The folate contributed as DFE by each food in the NHANES Subject enrollment. Enrollment occurred from March through
dataset was calculated by multiplying, for each NHANES respondent, September 2003. Pregnant women (n ⫽ 29), and nursing or non- the mass of the food consumed by its folic acid concentration, and pregnant low-income women (n ⫽ 39) who were at least 18 y old, and then summing across all of the respondents. To calculate DFEs from who were eligible or who had begun the Women, Infants, and the NHANES data, the estimated synthetic folic acid (SFA) content Children (WIC) program during the preceding 2 wk, were enrolled in of each food was multiplied by 1.7, the bioavailability correction this study. All study instruments (consent, 1-page folate-targeted food factor provided by the Institute of Medicine (21). Food groups (ques- screener) were translated into Spanish, back translated into English, tions) were then ranked according to their contribution to total DFE and administered by bilingual research staff.
The pregnant women were enrolled primarily from 6 local com- intake; 19 questions, representing foods contributing the top 60% of munity clinics in Sacramento, CA. The clinics were providers of the U.S. DFE intake, were included in the Block DFE screener. The total county Comprehensive Perinatal Services Program (CPSP) and DFE consumed by each woman enrolled in our study was calculated Medi-Cal programs. CPSP and Medi-Cal are statewide government as the cumulative sum of DFEs contributed by each food group on the programs that provide comprehensive prenatal care to low-income screening questionnaire. Folate values for the questions on the DFE women. To be eligible for these programs, a family's income must be screener were identified by analysis of the USDA Nutrient Database ⱕ185% of the Federal Poverty Level (FPL ⱕ 1.85), i.e., gross for Standard Reference, Release 15 (August, 2002). Synthetic folic monthly income of $1400-$5000 for a family unit of 1 to 8 persons, acid values used for the computation of DFEs from vitamin supple- respectively. We targeted pregnant women who were scheduled for ments were identified by reference to label claims on commonly their first obstetrics appointment because they were most likely to available dietary supplements.
have not started using supplements. We also targeted women who Blood drawing and analysis. Blood was drawn from each woman
were newly recruited into the WIC program. WIC is a federal grant into 4 separate 4-mL tubes that contained spray-dried K -EDTA and program that provides vouchers for supplemental foods, nutrition one SST tube for vitamin B-12. The tubes were inverted gently to education, and prenatal counseling to low-income women. WIC food prevent clotting. All EDTA tubes for whole-blood folate were vouchers are special checks for purchasing health-promoting foods promptly wrapped in aluminum foil (to exclude light) and immedi- such as milk, juice, eggs, peanut butter, and so forth. The WIC ately stored on dry ice. One EDTA tube and the SST tube were stored authorized food list of July 2003 targeted foods rich in vitamins A and at ⬃5°C during transit (⬃2 h) to the University of California Davis, C, protein, and fiber, but not folate. Women enrolled in this federal Sacramento Medical Center (UCDMC) Pathology Department Lab- program have been shown to improve their nutritional status and oratory where complete blood count, packed cell volume (PCV), and pregnancy outcome (19,20). Community healthcare practitioners serum vitamin B-12 (SimulTRAC Radioassay Kit, Vitamin refer pregnant women to the WIC program to initiate nutrition B-12[57Co]/Folate[125I] from ICN Diagnostics) measurements were made. Also, one frozen tube of whole blood was analyzed for folate The nonpregnant women were enrolled from a local Woman's using chemiluminescence technology (Bayer Diagnostics ADVIA Center. This Center is a nonprofit organization in Sacramento, CA Centaur, Bayer Corporation Diagnostics Division) at the UCDMC that serves low-income women and their children by providing them Pathology Department Laboratory. The remaining 3 EDTA tubes a daily meal, and referrals to local state agencies for food, housing, were stored on dry ice during transit (⬃2 h) to the laboratories where and shelter.
they were stored at ⫺70°C for RBC folate analysis using GC-MS Clinic visits. Women enrolled at the 1st clinic visit (baseline
(16 –18), Lactobacillus casei (22), and radioassay (RA; SPNB Dual values) were questioned about their general medical history, dietary Count 125I, Diagnostic Products) methods. The L. casei and RA habits, length of time on WIC, and demographics. Each woman assays were conducted at the UCD-Clinical Nutrition Research Unit completed a 1-page folate-targeted food/supplement intake screener, (UCD-CNRU). The Chemiluminescence, L. casei, and RA assays and blood was drawn as described below. Pregnant women began were performed on whole-blood lysates (prepared with ascorbic acid).
taking a standard over-the-counter prenatal vitamin supplement (800 Folate concentrations of whole blood were divided by the PCV ␮g or 1814 nmol folic acid/tablet) as prescribed by their healthcare (fraction) and expressed as nmol folate/L erythrocytes. A pool of provider. All nonpregnant women received one bottle of 100 prenatal freshly drawn blood was promptly divided into small aliquots and RBC FOLATE LEVELS BY FOUR ASSAYS AND FOLATE INTAKE stored ⫺70°C; aliquots from the pool were analyzed with each batch women were in wk 18 ⫾ 2 of gestation and they were currently of study samples to serve as a control monitor for day-to-day varia- not receiving WIC benefits. The pregnant group had fewer tion. Finally, all RBC folate assays were performed within 3 wk from children (P ⫽ 0.0140) than the nonpregnant group, but the the date the blood was drawn.
difference would likely disappear when the pregnant group was Calculations and data analysis.
The data were checked for normality, and descriptive statistics were calculated for each response as old as the nonpregnant group.
variable. The data were analyzed by repeated-measures ANOVA.
As expected, the pregnant women had a lower PCV (P The 3-way interaction (assay ⫻ supplementation ⫻ pregnancy status) ⬍ 0.0001), RBC count (P ⬍ 0.0001), blood hemoglobin (P was not significant; thus, only main effects (Table 1) and the 2-way ⬍ 0.0001), and plasma vitamin B-12 levels (P ⫽ 0.0749) interactions, assay ⫻ supplement and assay ⫻ pregnancy status (Figs.
compared with nonpregnant women. Also, the pregnant 1, 2) are presented. Regression analysis of RBC folate levels by the 4 women had a higher white blood cell (WBC) count (P assay methods and by folate intake were also calculated. All statistical ⫽ 0.0073) than the nonpregnant women. Folic acid supple- analyses were conducted with StatView 5.0.1 software program (Aba- mentation did not significantly affect PCV, WBC count, RBC cus Concepts).
count, hemoglobin, or plasma vitamin B-12. The study popu-lation was normal.
RBC folate levels were dependent on assay method and Of the 30 pregnant women, 24 completed the study as did supplement (Table 1, Fig. 1). The overall mean concentra- Downloaded from
29 of the 38 nonpregnant women. Those who did not com- tions (main effects of assay method ranked low to high) were plete the study were lost to follow-up for not keeping their as follows: 1278 ⫾ 39 nmol/L by L. casei ⬍ 1450 ⫾ 32 nmol/L scheduled clinic appointments, termination of pregnancy, or by chemiluminescence ⬍ 1646 ⫾ 34 nmol/L by GC-MS moving out of the area. Of the 30 pregnant women, 6 were ⬍ 1877 ⫾ 46 nmol/L by RA (P ⬍ 0.001). In addition, the Caucasian, 8 were African American, 14 were Latina His- magnitude of the increases in RBC folate (nmol/L) in response panic, 2 were other (Native American); 17 were single and 13 to supplementation (ranked low to high) were as follows: 138 jn.nutrition.org were married; 4 lived alone, 24 lived with others, and 2 were ⫾ 63 nmol/L by chemiluminescence ⬍ 267 ⫾ 64 nmol/L by homeless. Of the 38 nonpregnant women, 8 were Caucasian, GC-MS ⫽ 285 ⫾ 75 nmol/L by L. casei ⬍ 351 ⫾ 87 nmol/L 15 were African American, 13 were Latina Hispanic, and 2 by RA. RBC folate levels were dependent on assay method were other (Asian); 26 were single and 12 were married; 6 and pregnancy status (Table 1, Fig. 2) with pregnant women
lived alone, 28 lived with others, and 4 were homeless. The having a higher RBC folate concentration only when analyzed pregnant and nonpregnant women had similar education lev- by chemiluminescence and RA.
els, 11 ⫾ 0.3 y (mean ⫾ SE), also similar to that of women The Block DFE screener had 21 questions (19 diet, 2 participating in the NHANES III (11.3 ⫾ 0.06 y, mean ⫾ SD) supplement) and measured usual and customary intake of for which the data were collected between 1988 and 1994 (23).
DFEs. It included the 19 questions that account for 60% of Other characteristics of the study participants are summa- total folate intake as DFEs in the U.S. diet, and 2 questions rized in Table 1. The pregnant women were younger than the
concerning vitamin supplements. The women enrolled in our nonpregnant women by approximately a decade. The pregnant study completed the screener in 6 –12 min.
Characteristics, hematology, erythrocyte folate, and folate intake of women in the study1 Pregnant vs.
Baseline vs.
Number of children, n Weeks pregnant, n WBC, 1 ⫻ 103/mm3 RBC, 1 ⫻ 106/mm3 Hemoglobin, g/L Plasma vitamin B-12, pmol/L RBC folate, nmol/L Food NF, ␮g/d Food SFA, ␮g/d Food NF ⫹ SFA, DFE/d Supplement SFA, ␮g/d Total SFA, ␮g/d Total, DFE/d 1 Values are means ⫾ SEM.
2 The estimated SFA content of each food was multiplied by 1.7, the bioavailability correction factor provided by the IOM (21).
CLIFFORD ET AL.
nmol folic acid for 30 – 60 d. Also, we measured folate intakewith a 1-page folate-targeted food/supplement frequencyscreener designed specifically for our study (Block DFEscreener), and determined whether folate intake was corre-lated with RBC folate.
Maternal RBC folate ⬎ 906 nmol/L, measured with Quan- taphase IR Folate Radioassay (5), is considered optimal forprevention of folate-responsive neural tube defects (28).
When we used chemiluminescence, GC-MS, and RA meth-ods, only 5, 1, and 3 women, respectively, had RBC folate ⬍ 906 nmol/L. However, when we used L. casei, 24 women had RBC folate ⬍ 906 nmol/L. A goal of Healthy People 2010 is to increase RBC folate of nonpregnant women to 500nmol/L (29) and only the L. casei assay identified women (n ⫽ 3) with RBC folate ⬍ 500 nmol/L in our study. Another research group (30) used the L. casei assay to determine RBCfolate status in a similar population of women residing in Downloaded from California postfortification and not taking any supplements.
Their respective levels for socioeconomically disadvantaged Effect of the assay used and folate supplementation on and advantaged women were 1172 ⫾ 342 nmol/L and 1387 RBC folate concentrations in women (n ⫽ 68) before (baseline) and after ⫾ 329 nmol/L, levels similar to ours. Our mean RBC folate they consumed ⬃600 ␮g/d (1360 nmol/d) SFA supplements for 30 – 60 levels and those of Caudill et al. (29) both far exceeded 362 d. Baseline levels and the increase upon supplementation were assay nmol/L (9), a concentration that is widely considered an jn.nutrition.org acceptable cutoff value.
Only 1 subject in our present study had an RBC folate level The intake of natural food (NF) folate was the same for the ⬍ 362 nmol/L; the lowest individual RBC folate values were pregnant and nonpregnant women (Table 1). Because the 333 nmol/L by L. casei, 653 nmol/L by chemiluminescence, pregnant group had a higher intake of SFA from fortified foods 806 nmol/L by RA, and 894 nmol/L by GC-MS. Of the (breakfast cereals) than the nonpregnant group (P ⫽ 0.0269), women in our study, 65% had RBC folate ⬎ 906 nmol/L, there was a trend (P ⫽ 0.0644) for them to have a higher which compares well with the value of 78% reported by intake of total food folate (NF ⫹ SFA). Differences in the food another research group (30). Using the L. casei method, the (NF ⫹ SFA) intake did not differ between the baseline and mean RBC folate concentration in the Framingham Offspring supplementation periods (P ⫽ 0.5742).
Cohort study after fortification was 1020 nmol/L (31), which The overall compliance with taking the folic acid supple- agreed with L. casei-measured baseline values in Table 1. Our ment was as follows: 64% of women reported taking the 30- to 60-d supplementation period was chosen to equal or supplement every day; 21% reported taking it 4 – 6 d/wk; 7.5% exceed the recommended 4-wk period of folic acid use before reported taking it 1–3 d/wk; and 7.5% reported that they had conception to improve folate status (21,32,33) and/or the stopped taking it 30 – 60 d before giving birth. This level of 4-wk period used in another study (34).
compliance resulted in the supplement providing 596 ⫾ 56 Since 1998, women capable of becoming pregnant have and 680 ⫾ 39 ␮g/d (1351 ⫾ 127 and 1542 ⫾ 88 nmol/d) of been advised to consume 907 nmol folic acid/d from fortified SFA to pregnant and nonpregnant women (Table 1).
foods and supplements, in addition to dietary folate (21). The The interassay correlation coefficients (r) for RBC folate ranged from 0.4679 (GC-MS vs. chemiluminescence) to
0.8261 (L. casei vs. RA) (Fig. 3). The r-value for RA vs.
GC-MS was 0.7063, the 4th highest of 6.
Relations between RBC folate levels (using 4 analytical methods) and folate intake as DFEs ranged from 0.2676 with
chemiluminescence to 0.4622 with RA (Fig. 4). The r-value
for folate intake vs. GC-MS-assayed RBC folate was 0.4343 (P
⬍ 0.0001), the 3rd highest of 4. Although the r-values were significant, only 16% (a modest portion) of the variation in adependent variable was explained by the independent variablewhen r ⫽ 0.40.
Because RBC folate reflects intracellular and tissue folate stores, it was chosen as the preferred indicator of folate nutri-tional status in establishing the new dietary reference intakes(21). At the same time, there are concerns over the accuracyof existing analytical methods for RBC folate (10,11) and onlya few groups (16,17,25–27) report within-laboratory compar-isons of several RBC folate assays. Therefore, we used 4 dif- Effect of the assay used and pregnancy status on RBC ferent methods to measure RBC folate levels in small popula- folate concentrations in pregnant (n ⫽ 30) and nonpregnant (n ⫽ 38) tions of pregnant and nonpregnant women (FPL ⱕ 1.85) women. Pregnant women had higher RBC folate than nonpregnant before and after each received a daily supplement of 1814 women when measured with chemiluminescence and RA.
RBC FOLATE LEVELS BY FOUR ASSAYS AND FOLATE INTAKE Correlations among RBC folate concentrations measuredin the women (n ⫽ 120) by chemilumi-nescence, L. casei, GC-MS, and RA.
The r values ranged from 0.4679 forGC-MS vs. chemiluminescence (A) to0.8261 for L. casei vs. RA (F). S , S , and S represent the standard errors of the intercept, slope, and estimate, re-spectively.
fortified foods and supplements together provided the pregnant creases in RBC folate (nmol/L) upon supplementation had a and nonpregnant groups with 939 ⫾ 55 and 950 ⫾ 64 ␮g/d lower P-value with chemiluminescence (P ⬍ 0.0171) than (2129 ⫾ 125 and 2154 ⫾ 145 nmol/d) of SFA, respectively with the RA assay (P ⬍ 0.0001), whereas the magnitude of the (Table 1). These quantities of SFA almost matched those
rise with the L. casei and GC-MS assays did not differ (Table recommended since 1998 (21).
1, Fig. 1). When the protein-binding assays were employed, Participants in the present study consumed ⬃600 ␮g/d pregnant women had higher RBC folate levels than nonpreg- (1360 nmol/d) supplement (SFA) and experienced (mean) nant women (Table 1, Fig. 2), suggesting that these assays may increases in RBC folate of 138 –351 nmol/L, depending on the be hormone sensitive. Finally, RBC folate, as determined by L. assay used. Had we doubled the duration of the supplementa- casei was correlated with the concentration measured by GC- tion period to match the ⬃120-d lifespan of RBCs, and/or MS with an r ⫽ 0.7185, the 4th highest of 6 (Fig. 3, Panel B).
increased the amount of the SFA supplement, we might have Therefore, it seems that the L. casei and GC-MS methods may achieved a higher rise in RBC folate in response to supple- offer hope for accurate RBC folate determinations that can mentation. The rise in RBC folate remained linear with no serve as guides to optimal folate status.
evidence of saturation as total folate intake increased irrespec- The RBC folate as determined by RA was correlated with tive of the assay method (Fig. 3); therefore, the plateau level the concentration measured by GC-MS with an r ⫽ 0.7063 of RBC folate remains unknown. Further investigation is war- (Fig. 3, Panel C); this did not differ from the r ⫽ 0.7010 value ranted to establish a reference range for RBC folate levels to reported by another group (26) who compared a number of prevent anemia and another to minimize the risk of more commercially available protein-binding assay kits.
serious defects (35).
The Block DFE screener was effective in assessing dietary The protein binding folate assays (chemiluminescence and intake of folate because the DFEs measured by the screener RA) are facile compared with the L. casei and GC-MS assays, were positively (and modestly) correlated with the results of all which require several steps in sample preparation (16,17,36).
4 folate assay methods (r ⫽ 0.26 – 0.46) and the r-values were Among the protein-binding assays, the magnitude of the in- significant (P ⬍ 0.0001) for all measures (Fig. 4). Yen et al.
CLIFFORD ET AL.
Correlations between total folate intake and RBC folate con-centration measured by GC-MS (A), L.
casei (B), chemiluminescence (C), andRA (D) in all women (n ⫽ 120). The rvalues ranged from 0.2676 with chemi-luminescence (C) to 0.4622 with RA(D). The abbreviated folate targetedfood/supplement screener was effec-tive in assessing folate intake. S , S , and S represent the standard errors of the intercept, slope, and estimate, re- (37) found a similarly modest correlation (r ⫽ 0.354, P randomised controlled trial of periconceptional multivitamin supplementation.
0.06) when they compared folate intake measured by 7 d of Final report. Arch. Gynecol. Obstet. 255: 131–139.
3. Boushey, C. J., Beresford, S. A., Omenn, G. S. & Motulsky, A. G.
"focused recalls" with plasma folate levels. They also found no A quantitative assessment of plasma homocysteine as a risk factor for vascular significant correlations between the Fred Hutchinson Cancer disease. Probable benefits of increasing folic acid intakes [see comments]. J. Am.
Research Center FFQ and plasma folate (37). Comparisons Med. Assoc. 274: 1049 –1057.
between correlations reported by Yen et al. (37) and those 4. Jennings, E.
Folic acid as a cancer-preventing agent. Med.
Hypotheses 45: 297–303.
presented here for the Block DFE screener should be inter- 5. Brown, J. E., Jacobs, D. R., Jr., Hartman, T. J., Barosso, G. M., Stang, preted with caution, because the 2 dietary instruments have J. S., Gross, M. D. & Zeuske, M. A.
Predictors of red cell folate level in different purposes. The instrument of Yen et al. (37) was women attempting pregnancy. J. Am. Med. Assoc. 277: 548 –552.
6. Wu, A., Chanarin, I., Slavin, G. & Levi, A. J.
Folate deficiency in designed to measure short-term folate intake and to compare it the alcoholic–its relationship to clinical and haematological abnormalities, liver with plasma folate, whereas the Block DFE screener was de- disease and folate stores. Br. J. Haematol. 29: 469 – 478.
signed to measure usual, long-term folate intake and to com- 7. Molloy, A. M.
Is impaired folate absorption a factor in neural pare it with RBC folate, a good biomarker of long-term folate tube defects? Am. J. Clin. Nutr. 72: 3– 4.
8. Kloeblen, A. S.
Folate knowledge, intake from fortified grain nutritional status.
products, and periconceptional supplementation patterns of a sample of low- The Block DFE screener was designed to capture foods income pregnant women according to the Health Belief Model. J. Am. Diet Assoc.
contributing the top 60% of folate intake (as DFE) in the 99: 33–38.
9. Sauberlich, H. E.
Folate status of U.S. population groups. In: United States, and was not intended to capture a point esti- Folate in Health and Disease (Bailey, L. B., ed.), pp. 171–194. Marcel Dekker, New mate that would reflect 100% of folate intake. Longer FFQ, however, are often designed to capture 80 –90% of nutrient 10. Gunter, E. W., Bowman, B. A., Caudill, S. P., Twite, D. B., Adams, M. J.
intake. The Block DFE screener produces lower estimates than & Sampson, E. J.
Results of an international round robin for serum and whole-blood folate. Clin. Chem. 42: 1689 –1694.
a full-length FFQ, or than the truth, by design. The purpose 11. Wright, A. J., Finglas, P. M. & Southon, S.
Erythrocyte folate was to create an instrument that would rank subjects well, analysis: a cause for concern? Clin. Chem. 44: 1886 –1891.
while being brief. Block et al. (38,39) showed that shorter 12. Pawlosky, R. J., Flanagan, V. P. & Pfeiffer, C. M.
of 5-methyltetrahydrofolic acid in human serum by stable-isotope dilution high- questionnaires such as the DFE screener rank subjects similarly performance liquid chromatography-mass spectrometry. Anal. Biochem. 298: to full-length FFQ.
We conclude that RBC folate and its response to folic acid 13. Nelson, B. C., Pfeiffer, C. M., Margolis, S. A. & Nelson, C. P.
supplementation was assay dependent; thus, method-specific Affinity extraction combined with stable isotope dilution LC/MS for the determi-nation of 5-methyltetrahydrofolate in human plasma. Anal. Biochem. 313: 117– RBC folate ranges may be necessary in evaluating published data at this time and setting cutoff values. Further comparisons 14. Pfeiffer, C. M., Fazili, Z., McCoy, L., Zhang, M. & Gunter, E. W.
among assay methods are warranted to identify the method Determination of folate vitamers in human serum by stable-isotope-dilution tan-dem mass spectrometry and comparison with radioassay and microbiologic that most accurately determines the actual RBC folate value.
assay. Clin. Chem. 50: 423– 432.
The folate screener was effective in assessing dietary folate 15. Nelson, B. C., Pfeiffer, C. M., Margolis, S. A. & Nelson, C. P.
intake. The adequacy of the folate nutritional status (of our Solid-phase extraction-electrospray ionization mass spectrometry for the quan- subjects) reflects the fortified diets.
tification of folate in human plasma or serum. Anal. Biochem. 325: 41–51.
16. Dueker, S. R., Lin, Y., Jones, A. D., Mercer, R., Fabbro, E., Miller, J. W., Green, R. & Clifford, A. J.
Determination of blood folate using acid extraction and internally standardized gas chromatography-mass spectrometrydetection. Anal. Biochem. 283: 266 –275.
Prevention of neural tube defects: results of the 17. Lin, Y., Dueker, S. R., Jones, A. D. & Clifford, A. J.
Medical Research Council Vitamin Study. MRC Vitamin Study Research Group.
processing solid phase extraction protocol for the determination of whole blood Lancet 338: 131–137.
folate. Anal. Biochem. 301: 14 –20.
2. Czeizel, A. E., Dudas, I. & Metneki, J.
Pregnancy outcomes in a 18. Lin, Y., Dueker, S. R. & Clifford, A. J.
Human whole blood folate RBC FOLATE LEVELS BY FOUR ASSAYS AND FOLATE INTAKE analysis using a selected ion monitoring gas chromatography with mass selective Objectives, Maternal, Infant and Child Health. http:/www.health.state.mo.us/GL- detection protocol. Anal. Biochem. 312: 255–257.
Request/MCH/hp2010.html [accessed 07/30/04].
19. Swensen, A. R., Harnack, L. J. & Ross, J. A.
Nutritional assess- 30. Caudill, M. A., Le, T., Moonie, S. A., Esfahani, S. T. & Cogger, E. A.
ment of pregnant women enrolled in the Special Supplemental Program for Folate status in women of childbearing age residing in Southern Califor- Women, Infants, and Children (WIC). J. Am. Diet Assoc. 101: 903–908.
nia after folic acid fortification. J. Am. Coll. Nutr. 20: 129 –134.
20. Owen, A. I. & Owen, G. M.
Twenty years of WIC: a review of 31. Choumenkovitch, S. F., Jacques, P. F., Nadeau, M. R., Wilson, P. W., some of the effects of the program. J. Am. Diet Assoc. 97: 777–782.
Rosenberg, I. H. & Selhub, J.
Folic acid fortification increases red blood 21. Food and Nutrition Board, Institute of Medicine cell folate concentrations in the Framingham study. J. Nutr. 131: 3277–3280.
Standing Committee on the Scientific Evaluation of Dietary References Intakes 32. Czeizel, A. E. & Dudas, I.
Prevention of the first occurrence of and Its Panel on Folate, Other B Vitamins and Choline and Subcommittee on neural-tube defects by periconceptional vitamin supplementation. N. Engl.
Upper Reference Levels of Nutrients. In: 1998 Dietary Reference Intakes: Thiamin, J. Med. 327: 1832–1835.
Riboflavin, Niacin, Vitamin B6, Folate, Vitamin B12, Panthothenic Acid, Biotin, and Recommendations for the use of folic acid to reduce the number of cases of spina bifida and other neural tube defects. Morb.
Choline. National Academy Press, Washington, DC.
Mortal. Wkly. Rep. 41: 1–7.
The USDA Nutrient Database for Standard Refer- 34. Brouwer, I. A., van Dusseldorp, M., Thomas, C. M., Duran, M., Hautvast, J. G., Eskes, T. K. & Steegers-Theunissen, R. P.
Low-dose folic acid 23. O'Broin, S. & Kelleher, B.
Microbiological assay on microtitre supplementation decreases plasma homocysteine concentrations: a randomized plates of folate in serum and red cells. J. Clin. Pathol. 45: 344 –347.
trial. Am. J. Clin. Nutr. 69: 99 –104.
24. Ford, E. S. & Bowman, B. A.
Serum and red blood cell folate 35. Green, N. S.
Folic acid supplementation and prevention of birth concentrations, race, and education: findings from the third National Health and defects. J. Nutr. 132: 2356S–2360S.
Nutrition Examination Survey. Am. J. Clin. Nutr. 69: 476 – 481.
36. Tamura, T.
Microbiological assay of folates. In: Folic Acid 25. Shane, B.
Folate assay: a comparison of radioassay and mi- Metabolism in Health and Disease. Contemporary Issues in Clinical Nutrition, Vol. Downloaded from crobiological methods. Clin. Chim. Acta 100: 13–19.
13 (Picciano, M. F., Stokstad, E.L.R. & Gregory, J. F., III, eds.), pp. 121–137.
26. Brown, R. D., Jun, R., Hughes, W., Watman, R., Arnold, B. & Kronenberg, Wiley-Liss, New York, NY.
Red cell folate assays: some answers to current problems with 37. Yen, J., Zoumas-Morse, C., Pakiz, B. & Rock, C. L.
radioassay variability. Pathology 22: 82– 87.
intake assessment: validation of a new approach. J. Am. Diet. Assoc. 103: 27. Owen, W. E. & Roberts, W. L.
Comparison of five automated 991–1000; discussion 1000.
serum and whole blood folate assays. Am. J. Clin. Pathol. 120: 121–126.
38. Block, E., Hartman, A. M. & Naughton, D.
A reduced dietary 28. Daly, L. E., Kirke, P. N., Molloy, A., Weir, D. G. & Scott, J. M.
questionnaire: development and validation. Epidemiology 1: 58 – 64.
Folate levels and neural tube defects, implications for prevention. J. Am. Med.
39. Cummings, S. R., Block, G., McHenry, K. & Baron, R. B.
Eval- jn.nutrition.org Assoc. 274: 1698 –1702.
uation of two food frequency methods of measuring dietary calcium intake. Am. J.
29. National Center for Health Statistics. Healthy People 2010. Summary of Epidemiol. 126: 796 – 802.

Source: http://fsnep.ucdavis.edu/publications/research-articles/research-articles/J.Nutr.2005Clifford13743.pdf