Clinical Utility of Serum and Red Blood Cell Analysis
By Cynthia Balion, PhD, FCACB and Bhushan M. Kapur, PhD, FRSC, FACB, FCACB
Researchers first documented an association between pregnant women’s folate levels and neural tube defects (NTD) in newborns in the 1960s. Today, low folate is a well-established risk factor for NTDs, and in 1998, the Food and Drug Administration mandated fortification of cereal grain products in the U.S. to ensure that women who are unaware of their pregnancies do not compromise their fetuses during the critical stage when the neural tube closes. The primary driving force behind this policy was the recognition that periconceptional folate supplementation, in addition to normal dietary folate intake, significantly reduces the incidence of NTDs. In fact, many countries now have either mandatory or voluntary flour fortification regulations to reduce the risk of NTDs in newborns.
Since the fortification of grains with folate began, there has been a substantial rise in folate concentration in the general population. So much so that low folate levels are now rare. Although folate levels in pregnant women are critical to the health of newborns, the majority of laboratory test orders for folate are made to investigate anemia along with vitamin B12 deficiency in nonpregnant patients.
This article will describe the metabolism of folate, the causes, clinical effects and prevalence of folate deficiency, as well as evidence about diagnostic thresholds and the clinical utility of folate testing.
Folic acid (pteroylmonoglutamate) acts as a carrier of one-carbon units in a variety of metabolic reactions (Figure 1). This class of vitamins is essential for the synthesis of nucleic acids, thymidylate, neurotransmitters, phospholipids, and hormones. Folate is also integral to the de-novo generation of methionine, which is required for genomic and nongenomic methylation reactions.
Humans lack the enzymes to synthesize folate, so dietary intake is necessary. Nutritionists estimate that the body stores 10–100 mg of folate, with 5–10 mg sufficient for about 4 months of normal metabolism.
Rich sources of dietary folate include green leafy vegetables, fruits, dairy products, cereals, yeast, and animal proteins. Cooking, however, destroys most folate in food. The dietary reference intake (DRI) recommendation for adults in the U.S. is 400 µg/day, and other countries have similar recommendations. For children, the recommended DRI is lower, but higher amounts are recommended for women during pregnancy and lactation. Natural folate is 50% bioavailable compared to 85% in fortified foods and almost 100% when taken as a supplement.
Causes of Folate Deficiency
The causes of folate deficiency fall into four categories: 1) nutritional deficiency; 2) inadequate absorption (e.g., malabsorption syndromes, gastrectomy); 3) increased requirements (e.g., pregnancy, dialysis); and 4) drug interference. Drugs that are metabolized via a one-carbon pathway such as methotrexate and 5-fluorouracil used to treat cancer may result in a functional folate deficiency because they inhibit key enzymes in the folate pathway; methotrexate inhibits dihydrofolate reductase (DHFR) and 5-fluoruracil inhibits thymidylate synthase (TS). Reduced absorption of folate also can occur from drugs such as metformin, cholestyramine, anticonvulsants, and H2 blockers, but only if dietary folate intake is low.
Clinical Effects of Folate Deficiency
Many experimental, observational, and randomized clinical trial (RCT) studies have investigated the association between folate levels and cardiovascular disease, cancer, and cognition. Overall, there is insufficient evidence to conclude that folate plays a causal role in these chronic diseases. The two primary pathophysiological consequences in which folate deficiency has been shown to play an important role are megaloblastic anemia and NTDs.
The hallmark of folate deficiency, as well as the most reliably quantifiable indicator, is the hematological manifestation of anemia. When folate levels are inadequate for cellular metabolism, megaloblastic changes occur in bone marrow and other tissues with rapidly dividing cells. Defective synthesis of DNA results in the production of larger erythrogenic precursors (macrocytes, macroovalocytes) compared to mature red blood cells and hypersegmented polymorphonuclear leukocytes. These cellular changes also occur in vitamin B12 deficiency. Although folate supplementation can reverse vitamin B12-deficiency anemia, it does not treat, and may even worsen, neurological disease associated with vitamin B12 deficiency.
Folate Deficiency in Pregnancy
Pregnancy imposes a unique requirement for additional folate due to the increase in blood volume and active cell proliferation rates critical for fetoplacental growth and development. Furthermore, the rate of folate catabolism progressively increases during pregnancy, reaching a peak in the third trimester at the time of maximal fetal growth. This increase appears to be due to accelerated breakdown of the vitamin during this period of enhanced cellular biosynthesis.
NTDs result from failure of the neural tube to close during early embryogenesis. Seventy-five percent of NTD-affected pregnancies end in miscarriage or stillbirth. The incidence of this lethal or severely debilitating birth defect varies considerably by geography, socioeconomic status, and ethnicity.
Researchers postulate that NTDs result from a multifactorial process, possibly the combined effects of several genes and environmental factors. Two RCTs in the early 1990s established the role of folate in preventing NTD. It is now well-established that periconceptional supplementation with folic acid can reduce the risk of babies being born with NTDs.
Prevalence of Folate Deficiency
A public health initiative launched in 1998 that mandated fortification of cereal grain products with folic acid in the U.S., Canada, and Costa Rica immediately reduced the prevalence of NTDs in those countries by 26%, 46%, and 35%, respectively. The success of this strategy led Chile, South Africa, and Australia to mandate fortification. Currently, the food safety agencies in New Zealand, the U.K., Ireland, and the Netherlands are advocating the same strategy.
Data from the National Health and Nutrition Examination Surveys (NHANES) 1988–2004 demonstrated the change in serum folate and red blood cell (RBC) folate during this period (Figure 2). The huge upward shift in the first survey after dietary fortification began (1999–2000) was followed by slightly lower peak values in the following two surveys. In fact, the data showed substantial changes in the prevalence of both serum folate and RBC folate deficiencies post-fortification, with serum folate deficiency defined as <3 ng/mL or <6.8 nmol/L and RBC folate deficiency defined as <140 ng/mL or <318 nmol/L. In the 1988–1994 survey, 15.6% of individuals had serum folate deficiency, while 30.5% were RBC folate-deficient. By the time of the 2003–2004 survey, these numbers had fallen to 0.5% and 4.0%, respectively. Participants age 60 and older were the least likely to be deficient: only 0.1% had insufficient serum folate levels, and just 2.1% lacked enough RBC folate.
The NHANES study of folate levels in the general population used Institute of Medicine (IOM) threshold values derived from studies that used microbiological assays for folate measurement and compared them to test results obtained by the Quantaphase (Bio-Rad Laboratories) method. When the bias of the Quantaphase assay—which is about 35% lower than the microbiological assay—was used to adjust these thresholds, the percentage of individuals classified as having serum or RBC folate deficiency across all post-fortification survey periods fell to <0.05% and < 0.5%, respectively.
Similarly, a Canadian community laboratory using the Quantaphase assay reported that serum folate deficiency (<1.5 ng/mL or <3.4 nmol/L) fell from 0.52% of the population studied before dietary fortification began to 0.22% post-fortification, while RBC folate deficiency (<95 ng/mL or <215 nmol/L) dropped from 1.78% to 0.41%. Analysis of 19,885 RBC folate tests performed from 2004 to 2006 by another Canadian laboratory, the Hamilton Regional Laboratory Medicine Program, found that only 0.3% of samples were below the adjusted threshold value of 229 ng/mL (520 nmol/L) using the folate assay on the Roche Diagnostics MODULAR ANALYTICS E170 (unpublished data).
Frequency Distribution of Serum and Red Blood Cell Folate
Frequency distribution of serum and red blood cell (RBC) folate among the entire population of the U.S. according to the National Health and Nutrition Examination Surveys spanning 1988–1994, 1999–2000,2001–2002, and 2003–2004.
Source: Pfeiffer, C. M et al. Am J Clin Nutr 2007;86:718–727. Used with permission.
Unmetabolized Folic Acid
In addition to increased folate levels, researchers have also found more circulating unmetabolized folic acid (UMFA) in blood. In NHANES 2001–2002, 38% of the 1,330 participants ≥60 years had detectable UMFA (4.4 ± 0.6 nmol/L). This UMFA amounted to about 6% of all measured serum folates. The presence of this form of folate indicates that the capacity of DHFR to convert the oxidized form of folate to the reduced form has been exceeded. UFMA only comes from supplements and fortified food; it is not formed from natural sources of folate.
Daily folate intake of about 300 µg is sufficient for UMFA to be detected in serum. Although evidence is lacking on whether UMFA has any specific biological affect, some data is emerging on this topic. For example, researchers have reported that UMFA and low vitamin B12 are associated with poorer cognitive function, highlighting a possible neuropsychiatric risk associated with UMFA.
Measurement of Serum and RBC Folate
Rather than a single molecular entity, folates comprise a group of molecules that are derivatives of folic acid. The most predominant form of folate in blood is 5-methyltetrahydrofolate (5MTHF), comprising 82–93% of total measured folates. Assays also measure other folates including UMFA, tetrahydrofolate (THF), 5,10-methenyl-THF, and 5-formyl-THF (Figure 1).
Before researchers developed protein-binding assays in the 1970s, labs used microbiological assays to measure folates. Today, several liquid chromatography-mass spectrometry methods are available, but most are not used for clinical measurements.
Most methods used in clinical laboratories are automated, nonisotopic methods that rely on folate-binding protein. Several factors affect these protein-binding assays. For example, folate-binding protein has greater affinity for polyglutamates (PGA) than monoglutamates, with PGA binding more strongly to 5MTHF. The amount of protein in the assay and the pH also affect binding. Dilution linearity can sometimes be a problem, too. In contrast to microbiological assays, protein-binding assays do not cross-react with antibiotics or folate analogs such as MTX or leucovorin (folinic acid).
While serum folate levels fluctuate significantly with diet, results of RBC folate assays reflect more closely tissue folate stores. Following dietary deprivation of folate, serum levels decline within 3 weeks, but RBC folate levels remain the same for 3–4 months. Although the actual assay is the same, RBC folate assays include a preanalytical treatment of the sample. The folate level obtained from the assay is a calculated number that also takes into account the patient’s hematocrit value. Table 1 provides a comparison of the characteristics of serum and RBC folate assays.
Comparison of Characteristics for RBC Folate and Serum Folate Measurements
|Indicator of tissue stores
Serum or plasma
||Dependent on recent dietary intake|
|Processing and storage
Whole blood hemolysates less stable than intact whole blood
> 90% recovery at 4°C up to 4 days
> 80% recovery at 22°C up to 4 days
Ambient thawing of whole blood can cause significant loss
< 10% at 4°C for less than 1 week
Plasma less stable than serum
Sensitive to freeze/thaw cycles in a frost-free freezer, but no loss up to 3 freeze/thaw cycles (exposed to ambient temperature for 1 hour)
||Hemolysis – off line
|CV – within person
|CV – between person
3.9 / 6.0
5.4 / 12.0
|Between method difference3
||Very good (5% of macrocytosis cases have normal serum folate)
||Slightly higher than for serum folate due to increased labor costs
|1 NHANES 1999-2002 (National Health Statistics Reports, Number 21; March 1, 2010)|
2 NHANES / Westgard Variation Database
3 CAP Survey 2009 K-B (K-08) and K-C (K-11, FOL-05, FOL-06), absolute method difference compared to all method mean (%).
4 Based on microbiological assays. No comparative studies with a reference standard available with current immunoassay methods.
It is also important to note that disagreements in the values obtained from the same specimen using assays from different manufacturers are due in part to the ability to detect the various forms of folate and the type of calibrators used. Figure 3 shows both serum and RBC folate results from a College of American Pathologists proficiency survey. The wide variations are most likely due to the difference in antibodies to different antigens the manufacturers use (glutamates). Currently, folate assays are not standardized; however, using a common reference standard would reduce the variability between assays. For this reason, it is important to interpret folate results from different laboratories and published studies with caution.
Relative Differences Between Commercial Methods Measuring Serum and RBC Folate
Data adapted from CAP Proficiency Survey 2009 using selected samples representing high and low concentrations for RBC folate and serum folate. The all methods mean is given in parentheses.
Click for Table 2
Commercial Assays for Folate
Diagnosis of Folate Deficiency
In 1998, the IOM reviewed the literature and chose as indicators of adequate folate status a value of 3ng/mL (7 nmol/L) for serum folate and 140 ng mL (305 nmol/L) for RBC folate. Noteworthy is the fact that these values were not associated with hematological manifestations or DNA changes related to folate deficiency.
While there is little consensus about folate levels in the general population, the most cited RBC folate concentration for prevention of NTDs in pregnant women is 900 nmol/L (400ng/mL). This level was determined in a large, case-control study in Ireland that characterized the dose response relationship between RBC folate and the risk of NTD. At this concentration the odds ratio was 1.0, meaning that the risk of a women having a baby with or without an NTD was the same.
Even with this guidance, clinicians often find patients’ folate test results confusing because laboratories report the values in different ways. Some laboratories report only reference intervals, others only thresholds values, and some a mix, depending on whether the result is from a RBC folate test or serum folate test. Furthermore, the actual values vary widely, with some laboratories reporting age thresholds and others reporting thresholds or intervals for borderline, normal, and excess amounts of folate. These inconsistent reporting methods hinder the clinical utility of the tests.
Clinical Utility of Folate Testing
Clinicians routinely order folate tests for evaluation of macrocytic anemia. This clinical practice is well-entrenched and continues despite the low prevalence of folate deficiency. In 2001, Cleveland Clinic investigators sought to determine the frequency of folate deficiency, whether there was a difference in clinical indication for folate testing between patients with and without folate deficiency, and whether folate replacement therapy was given in those who were deficient. The study found that just 1.7% of patients (74 of 4,315) had serum folate values <2.8 ng/mL (6.4 nmol/L). They found no difference in clinical indication for folate testing between groups, including macrocytic anemia, non-macrocytic anemia, macrocytosis/no anemia, dementia/delirium/neuropathy, and malabsorption/malnutrition. Of the 63 folate-deficient patients for whom medical records data were available, 24 were given folate replacement therapy. Overall, the authors concluded that the clinical utility of measuring folate is questionable due to the low prevalence of folate deficiency, limited value of the clinical indication to do the test, and poor response to low folate values.
Other studies have come to similar conclusions, including one focused on an underserved population. This 2004 study at San Francisco General Hospital found 0.51% (9 out of 159 patients) with RBC folate values to be <160 ng/ml (363 nmol/L). Only three patients had macrocytic anemia compared to 67 with macrocytic anemia who had folate values above this threshold.
However, studies like these have had little impact on reducing test orders. In an attempt to reduce unnecessary testing, the Hamilton Regional Laboratory Medicine Program sent a newsletter to physicians informing them about the rarity of folate deficiency. Three years later, there was no change in the number of tests ordered, so the laboratory decided to restrict testing by requiring a consultation with a biochemist or haematologist. Surprisingly, there was little difficulty in implementing this approach, although some issues, such as removing the test from panels, communicating the new policy to all laboratory sites within the system, and discussing the new policy with certain clinical areas like pediatric gastroenterology, required further attention. Within a month, the number of requests dropped by >99%, translating to a significant cost savings.
In general, study results over the last 12 years following folic acid fortification of cereal grains clearly indicate that measuring folate in patients suspected of being deficient is no longer necessary. Folate deficiency is now rare. Consequently, laboratories that perform these measurements need to reconsider how this service is offered. For example, it may make sense to send out folate tests. At the very least, laboratories should consider method biases when providing interpretive comments along with test values.
Today, folate deficiency is now uncommon in the U.S. and should not be considered for differential diagnosis for anemia. Patients who abuse alcohol, are severely malnourished, have malabsorption issues, or are receiving chemotherapy may be at a higher risk for folate deficiency. However, even in these high-risk groups the likelihood of folate deficiency is low. Evaluation of folate is important, however, in patients with persistent unexplained macrocytic anemia, and folate deficiency remains a concern during pregnancy and pre-conception.
Folate in Health and Disease, edited by Lynn Bailey. Boca Raton, FL: CRC Press, 2010.
Bar-Oz B, Koren G, Nguyen P, Kapur BM. Folate fortification and supplementation—are we there yet? Reprod Toxicol 2008;25:408–12.
Berry RJ, Bailey L, Mulinare J, Bower C. Fortification of flour with folic acid. Food Food Nutr Bull 2010;31:S22–35.
Daly LE, Kirke PN, Molloy A, Weir DG, et al. Folate levels and neural tube defects. Implications for prevention. JAMA 1995;274:1689–702.
Food and Nutrition Board, Institute of Medicine, Dietary Reference Intakes: Thiamin, riboflavin, niacin, vitamin B6, folate, vitamin B12, pantothenic acid, biotin, and choline. National Academy Press, Washington, DC, 1998.
Kaferle J, Strzoda CE. Evaluation of macrocytosis. Am Fam Physician 2009;79: 203–8.
Latif T, His ED, Rybicki LA, Adelstein DJ. Is there a role for folate determination in current clinical practice in the USA? Clin Lab Haematol 2004;26:379–83.
Owen WE, Roberts WL. Comparison of five automated serum and whole blood folate assays. Am J Clin Pathol 2003;120:121–6.
Pfeiffer CM, Johnson CL, Jain RB, Yetley EA, et al. Trends in blood folate and vitamin B12 concentrations in the United Sates, 2988-2004. Am J Clin Nutr 2007;86:718–27.
von Kuster K, Shojania AM. Ordering folate assays is no longer justified for investigation of anemias, in folic acid fortified countries. BMC Research Notes 2010;2:22.
Cynthia Balion, PhD, FCACB, is a clinical biochemist in the of Department of Laboratory Medicine at Hamilton General Hospital and associate professor in the Department of Pathology and Molecular Medicine at McMaster University in Hamiliton, Ontario.
Bhushan M. Kapur, PhD, FRSC, FACB, FCACB, is an associate scientist in the Sunnybrook Research Institute, Sunnybrook Health Sciences Center and assistant professor Department of Laboratory Medicine and Pathobiology, Faculty of Medicine, at the University of Toronto, Ontario.