EDTA in dried blood spots leads to false results in neonatal endocrinologic screening.
Materials and Methods
Extraction and butylation of amino acids and acylcarnitines during the standard procedure (2) led to butylation of EDTA in 4 positions. We tested 190 000 newborn samples for EDTA, using the mass transition 517.1-188.1 for EDTA measurement in multiple reaction mode. The limit of detection was 0.1 g/L. In the same samples, TSH and 17-OHP were measured according to the instruction manuals of the dissociation-enhanced lanthanide fluorescence immunoassay (DELFIA[R]) test kits Neonatal 17a-hydroxyprogesterone and Neonatal hTSH (PerkinElmer). Independent quantification of 17-OHP was achieved by use of an MS/MS method for steroids (3). In addition, 3 types of experiments were performed with whole blood samples from healthy adult donors: (a) The blood was spiked with human TSH (Calbiochem/Merck) and EDTA (Merck) at different concentrations and spotted on Whatman 903 filter paper (Whatman). (b) Blood samples containing 2 different concentrations of TSH were spotted either directly on filter paper or using EDTA-coated plastic capillaries (Microvette[R] CB 300, Sarstedt). (c) Blood samples (2.7 mL, 1 mL, and 0.5 mL) spiked with human TSH were filled into plastic devices designed for collection of 2.7 mL EDTA-blood for hematological tests (S-MonovetteZ[R]) and spotted on filter paper.
More detailed information on the methods can be found in the Data Supplement that accompanies the online version of this Brief Communication at http://www.clinchem.org/content/vol54/issue3.
EDTA was found in 138 of 190 000 samples. None of these were false negative for TSH, as demonstrated by the results of control analysis of second samples. In 27 samples, EDTA caused results that were in the pathological range for 17-OHP and were shown to be false positive by analysis of a second sample. In 21 cases MS/MS quantification of 17-OHP and other steroids in the first sample confirmed the negative screening results. Two examples of normal steroid profiles of EDTA-contaminated samples compared with samples from a premature newborn and a patient affected by congenital adrenal hyperplasia are shown in Table 1 in the online Data Supplement.
When EDTA was detected in a sample, information about the blood collection procedure was obtained from the hospital or midwife. This information revealed three reasons for EDTA contamination: (i) Venous blood was collected in devices prepared with EDTA and then spotted on the filter paper. EDTA was evenly distributed within these samples. (ii) Blood was taken by heel prick and transferred to the filter cards via capillary devices coated with EDTA. Irreproducible 17-OHP and EDTA results were found in different spots and even throughout one single spot. (iii) After a hematological microcontainer prepared with EDTA was filled, venous blood was spotted directly onto the filter card. The top of the needle touched the container wall and EDTA was transferred to the filter paper card. Again, irreproducible EDTA results were found in different spots and even throughout one single spot.
INFLUENCE OF EDTA ON DELFIA TEST RESULTS
Results are shown in Fig. 1. The occurrence of false 17-OHP values depended on the concentration of EDTA. Concentrations <1.5 g/L did not have any impact, but concentrations >2.0 g/L produced test results above the cutoff of 60 nmol/L. The results in samples spiked with 20 mU/L and 65 mU/L TSH, respectively, also depended on EDTA concentration. Concentrations <3 g/L did not interfere significantly with the results. Above this value, samples with 20 mU/L TSH measured lower than the cutoff of 15 mU/L. Very high TSH concentrations were also decreased but remained positive up to EDTA concentrations of 4 g/L.
BLOOD COLLECTION PROCEDURES
Blood was collected successively in 5 drops of about 50-60 AL in the same EDTA coated capillary and transferred one after the other to filter paper. Most of the EDTA was found in the first spot, resulting in falsely increased 17-OHP. A pathological TSH value in a comparable experiment, however, was not affected. Analyses of multiple punches from the first blood spot in this experiment showed highly variable 17-OHP results. Values from 73 nmol/L to >2000 nmol/L in 5 different punches were obtained, and EDTA concentrations varied from 1.3-2.4 g/L.
Filling the capillary with 300 [micro]L blood at once and transferring it to 5 blood spots on filter paper resulted in a maximum concentration in the last blood spot, for which an extremely high false positive result for 17OHP was measured.
The same experiment using a sample spiked with TSH in a highly pathological concentration of 180 mU/L again did not show any effect. Only when the blood volume was decreased to about 150 [micro]L did the TSH result become false negative in the last of 3 spots.
For blood from EDTA blood collection devices, the results are summarized in Table 1. In samples from devices filled correctly with 2.7 mL blood, the results for TSH and 17-OHP were not influenced significantly. When only 1 mL of blood was added, a true concentration of 20 mU/L TSH gave results lower than the cutoff of 15.0 mU/L. The true-negative result for 17-OHP was turned into a false-positive one. Further reduction to 0.5 mL blood turned highly pathological TSH values into false negatives and true-negative 17-OHP concentrations into extremely pathological values.
[FIGURE 1 OMITTED]
EDTA is a widely used anticoagulant in blood collection. Therefore contamination of dried blood samples with EDTA can be expected to occur in newborn screening samples.
Our results show that EDTA in concentrations achieved by correct use of devices for hematological testing does not lead to false results in tests for TSH and 17-OHP. These blood samples contain 1.2-2.0 g/L EDTA, according to manufacturer information. No significant influence of anticoagulants on immunofluorometric test results had been reported previously (3). In other reports, the use of EDTA-blood in a time-resolved immunofluorometric assay is explicitly recommended (4), or the addition of EDTA to the sample buffer is reported to decrease unspecific background signals (5). These data and our own results indicate that, if containers for hematological testing are filled with the correct amount of blood, the test results will not be falsified. For some immunofluorometric assays, however, EDTA and other anticoagulants in blood samples have been reported to influence test results (1). Under everyday conditions, many different blood collection procedures and device applications are used. It is common, for example, for blood collection devices not to be filled completely, because only 300-400 [micro]L of blood are needed for newborn screening. Under such conditions very high concentrations of EDTA may occur, leading to false immunofluorometric results. Fortunately, the 17-OHP test is affected more strongly by EDTA than the test for TSH. If EDTA is distributed evenly throughout the whole sample, a false-negative congenital hypothyreosis screening is unlikely to occur without a false-positive congenital adrenal hyperplasia screening.
EDTA may be distributed unevenly throughout the filter paper card, even throughout a single blood spot, especially when EDTA-coated capillary devices are used. In this situation, repetition of immunofluorometric tests with material from a single test card may produce variable results, and MS/MS EDTA measurement may not always detect such contamination. It is always possible that the filter paper sample used for MS/MS analysis will be free of measurable EDTA although the one used for a lanthanide-labeled test is highly contaminated.
In our study, EDTA was detected in 0.07% of test cards. In 27 cases this led to a false newborn screening result and the screening had to be repeated on a control sample. EDTA measurement in dried blood samples can directly decrease the number of invalid samples received; in our study the reporting of EDTA results had a significant educational effect. Obstetrical wards sending contaminated samples repeatedly learned to avoid incorrect application of capillaries or other devices. We therefore recommend newborn screening laboratories which already use immunofluorometric screening methods and MS/MS to add this inexpensive and simple measurement to their routine. Irreproducible and implausible pathological 17-OHP results can be explained and, most importantly, false screening results for congenital adrenal hyperplasia and for congenital hypothyreosis can be avoided.
Grant/funding Support: None declared.
Financial Disclosures: None declared.
(1.) Kohek M, Leme C, Nakamura IT, De Oliveira SA, Lando V, Mendonca BB. Effects of EDTA and sodium citrate on hormone measurements by fluorometric (FIA) and immunofluorometric (IFMA) methods. BMC Clin Pathol 2002;23:2.
(2.) Chace DH, Kalas TA, Naylor EW. Use of tandem mass spectrometry for multianalyte screening of dried blood specimens from newborns. Clin Chem 2003;49:1797-817.
(3.) Janzen N, Peter M, Sander S, Steuerwald U, Terhardt M, Holtkamp U, Sander J. Newborn screening for congenital adrenal hyperplasia: additional steroid profile using liquid chromatography-tandem mass spectrometry. J Clin Endocrinol Metab 2007;92:2581-9.
(4.) Parra MD, Tuomola M, Cabezas-Herrera J, Ceron JJ. Use of a time-resolved immunofluorometric assay for determination of canine C-reactive protein concentrations in whole blood. Am J Vet Res 2005;66:62-6.
(5.) Maentausta 0, Menjivar M, Vihko R. Time-resolved immunofluorometric assay of 17 beta-hydroxysteroid dehydrogenase in plasma. Clin Chem 1991;37:1412-5.
(6.) Zherdeva VV, Chudinov AV, Savitskii AP. Lanthanide fluorescent immune analysis for determination of hormone thyroxine in dry blood spots [Abstract; in Russian]. Biomed Khim 2003;49: 70-9.
Ute Holtkamp,  * Jeanette Klein,  Johannes Sander,  Michael Peter,  Nils Janzen,  Ulrike Steuerwald,  and Oliver Blankenstein 
 Screening-Labor Hannover, Hannover, Germany.  Newborn Screening, Charite Children's Hospital, Berlin, Germany; * address correspondence to this author at: Screening-Labor Hannover, Postbox 91 10 09, D-30430 Hannover, Germany. Fax +40 5108 92163 19.
Table 1. Influence of filling grade of EDTA-coated blood-collection devices on measured EDTA and 17-OHP concentrations. EDTA Measured 17-OHP Measured TSH MS/MS, concentration, concentration, Sample g/L nmol/L (a) mU/L (b) Blood sample without 0.0 2.5 18.8 58.5 EDTA Device (2.7 mL) filled 0.9 2.7 20.6 52.4 correctly with 2.7 mL blood Device (2.7 mL) filled 2.6 168.0 12.9 32.9 with 1 mL blood Device (2.7 mL) filled 5.5 1675.0 1.4 10.3 with 0.5 mL blood (a) Cutoff: 60 nmol/L. (b) Cutoff: 15 mU/L.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Brief Communications|
|Author:||Holtkamp, Ute; Klein, Jeanette; Sander, Johannes; Peter, Michael; Janzen, Nils; Steuerwald, Ulrike;|
|Date:||Mar 1, 2008|
|Previous Article:||A new ELISA for use in a 3-ELISA system to assess concentrations of VEGF splice variants and [VEGF.sub.110] in ovarian cancer tumors.|
|Next Article:||An unusual case of severe hypertriglyceridemia and splenomegaly.|