Annals of Clinical & Laboratory Science 34:67-74 (2004)
© 2004 Association of Clinical Scientists
Development and Validation of a Third Generation Allergen-Specific IgE Assay on the Continuous Random Access IMMULITE® 2000 Analyzer
Thomas M. Li,
Tom Chuang,
Susan Tse,
Debra Hovanec-Burns and
A. Said El Shami
Diagnostic Products Corporation, Los Angeles, California
Address correspondence to Thomas M Li, Ph.D., Diagnostics Products Corp., 5700 West 96th Street, Los Angeles, CA 90045-5597, USA; tel 310 645 8200, ext 2217; fax 310 645 9999; e-mail tli{at}dpconline.com. To request a reprint, the catalog number of this article is ZD121.
 |
Abstract
|
|---|
In vitro determination of allergen-specific IgE (sIgE) represents an important aid in the diagnosis and treatment of allergy. Improvements in laboratory methodology – the development of second, and now third generation assays for sIgE – have brought about major advances in speed, convenience, performance, and in the standards for judging performance. In this study, following the NCCLS I/LA20-A guidelines, we evaluated the analytical performance of a quantitative chemiluminescent enzyme immunoassay for sIgE using the continuous random access IMMULITE® 2000 system. Defining features of this "third generation" sIgE assay include a true zero calibrator with a detection limit and functional sensitivity of 0.1 and 0.2 kU/L, respectively. The use of liquid allergens allows for complete automation, fast binding kinetics between IgE and the natural allergenic protein conformations, and a time-to-first-result of 65 min. Stable reagents and the low nonspecific signal associated with the liquid allergens and centrifugal wash technique permit extension of the measuring range to 0.1–100 kU/L, based on lot-specific, factory-calibrated master curves standardized to the WHO 75/502 reference standard. The assay demonstrated good precision and linearity over its measuring range. Relative to a first generation RIA (mRAST, from Hycor), clinical sensitivity, specificity, and concordance were 88%, 92%, and 90%, respectively (n = 812). Quantitative comparisons to a second generation assay yielded a linear regression relationship of IMMULITE 2000 = 0.99 (Pharmacia FEIA) + 1.99 kU/L, r = 0.859 (n = 169).
(received 19 October 2003; accepted 27 October 2003)
Keywords: immunoglobulin E, radioallergosorbent test, diagnostic accuracy, NCCLS protocols
|
Introduction
|
|---|
Allergy is a hypersensitivity reaction initiated by an immunological mechanism [1]. For patients suffering from immunoglobulin E (IgE)-mediated allergy, laboratory-based serological analyses for IgE antibody can be performed, not only as primary confirmation for the diagnosis, but also to identify offending allergens, thus enabling patients to avoid allergens that might trigger allergic reactions –asthma attacks, for example.
Originally described in 1967 by Wide et al [2], the radioallergosorbent test (RAST) was the first assay reported for detection of allergen-specific IgE (sIgE) antibodies. The Phadebas RAST (Pharmacia & Upjohn Diagnostics, Kalamazoo MI) used radioiodinated polyclonal antihuman IgE as signal detection antibody, and the results were reported in arbitrary "Phadebas RAST" units/ml of anti-birch IgE, with a positive/negative cutoff set at 0.35 in these units. In the modified RAST (mRAST) system, which utilizes allergens immobilized on paper discs and two overnight incubations, there is a single calibrator; results are reported in class scores and categorized as positive or negative [3]. The mRAST system yields increased clinical sensitivity, but at the cost of impaired clinical specificity [4,5,6], because only the scoring system is changed, but not the actual detection limit. A quantitative mRAST (Turbo MP, from Hycor) with multiple calibrators was reported by Li et al (American College of Allergy, Asthma and Immunology, 2001 Annual Meeting, Orlando, FL, P-86).
Nonisotopic, second generation RAST-type assays for sIgE [7,8] employ enzyme substrates and photometry or fluorimetry for signal detection. They are quantitative enzyme immunoassays using batch analyzers, with a measuring range of 0.35–100 kU/L, and require approximately 3 hr to complete. They have several advantages over first generation assays, including greater ease of use and a time-to-first-result measured in hr rather than days [9]. Unlike the arbitrary calibration of first generation assays, these assays use calibrators traceable to the WHO IgE reference standard 75/502; nevertheless, they use (nominally) the same cutoff, 0.35 kU/L, for distinguishing positive and negative results. Because the lowest actual calibrator is 0.35 kU/L, which is also true for the quantitative Turbo MP assay, explicit results below this concentration cannot be reported without extrapolation to a "virtual zero calibrator," that is based on an assumed, rather than measured, dose-response relationship [10].
Responding to perceived clinical need for low assay results reported in the mRAST class system, Diagnostic Products Corporation, Pharmacia, and Hycor have developed the Extended Classification System (ECS), Alternate Scoring Method (ASM), and Turbo Multi-Point System (Turbo MP), respectively, to approximate the mRAST classification. In ASM and Turbo MP, a certain fraction of the signal of the lowest (0.35 kU/L) calibrator serves to classify samples as positive or negative, and to assign mRAST class scores to the positive results. The ASM and Turbo MP both rely on extrapolation below their lowest calibrator [10], in contrast to the ECS, which uses a true zero calibrator.
The aim of the present study was to evaluate the analytical performance of a third generation sIgE assay, using the NCCLS I/LA20-A guidelines [13], and to compare it to representative first and second generation assays. The third generation sIgE assay studied is a chemiluminescent enzyme immunoassay for the quantitation of sIgE in human serum on the continuous random access IMMULITE® 2000 System [11,12]. This assay has the following characteristics: a substantially broader working range of 0.1 to 100 kU/L with a full calibration curve, including a true zero calibrator; the ability to report results both quantitatively, in kU/L (WHO 75/502), and in standard or extended classes; a functional sensitivity of 0.2 kU/L; the use of liquid allergens, barcoding, and other automation features to reduce labor and human error; and a total assay time of 65 min. The lot-specific, factory-calibrated master curve is highly stable and the manufacturer’s recommendations call for 2-point recalibration ("adjustment") only once every 2 weeks.
|
Materials and Methods
|
|---|
The IMMULITE 2000 assay (Cat. no. 2KUNG, Diagnostic Products Corp, Los Angeles, CA) utilizes an enzyme-enhanced chemiluminescent enzyme immunoassay for the quantification of sIgE. Streptavidin-coated bead, biotinylated liquid allergen, and patient sample are incubated for 30 min. After a spin wash, an alkaline phosphataselabeled monoclonal antibody specific for human IgE is added and another 30-min incubation follows. The bead is washed again and the enzyme label is measured with a chemi-luminescent substrate (phosphate ester of adamantyl dioxetane).
In addition to NCCLS guideline I/LA20-A for evaluations of sIgE assays, we followed the NCCLS EP5-A guideline for precision, which describes an experimental design – 20 days, 2 runs/day, 2 replicates/run – yielding estimates of within-run and "total" (interassay) CVs. Precision profiling was applied to the EP5-A CVs to estimate functional sensitivity, defined as the lowest concentration with an interassay CV 
20%.
For comparison studies, we used a manual RIA (mRAST, from Hycor Biomedical Inc (Garden Grove, CA) and a fluoroimmunometric version of the CAP System (Pharmacia & Upjohn Diagnostics, Kalamazoo MI). The former employs a radioisotopic tracer and a single calibrator. The latter has seven classes, with class 0 including all results below 0.35 kU/L and class 6 including all results 
100 kU/L. All assays were performed according to the manufacturers’ instructions in the package inserts (ie, Diagnostic Products Corporation’s IMMULITE 2000 package insert, Pharmacia’s UniCAP package insert, and Hycor’s RIA mRAST package insert).
|
Results
|
|---|
Detection limit.
This was established for the IMMULITE 2000 by assaying replicates of the zero calibrator, using 3 instruments and 20 replicates/instrument. In each of the 3 experiments, the mean and SD were calculated for the counts of the zero calibrator replicates. The counts 2 SD above the mean were translated into concentrations via the calibration curve, yielding an estimate of the assay’s detection limit in kU/L, in accord with the classic NCCLS-recommended experimental design for this performance measure [13].
The process is illustrated in Fig. 1
. All 3 estimates were <0.1 kU/L, the IMMULITE 2000 assay’s conservatively stated detection limit. Note, for contrast, that a second generation assay for sIgE, because it lacks a zero calibrator, must treat the concentration of its lowest calibrator, ie, 0.35 kU/L, as the assay’s limit of detection or quantitation [14]. This important parameter cannot be determined experimentally for such an assay.
View larger version (17K):
[in this window]
[in a new window]
|
Fig. 1. Low end of a representative calibration ("dose-response") curve for the IMMULITE 2000 sIgE assay, illustrating the results of a detection limit experiment, with the counts (kilo counts/sec, kcps) 2 SD above the average for 20 replicates of the assay’s zero calibrator (vertical axis) translated through the calibration curve to the corresponding concentration value (horizontal axis).
|
|
Precision and functional sensitivity.
Precision characteristics of the IMMULITE 2000 sIgE assay were determined according to the NCCLS EP5-A guideline in 2 studies, performed 12 mo apart. The EP5 design yields estimates of within-run CVs and "total" CVs for each sample. The latter correspond more closely to the interassay CVs relevant in actual practice: the CVs that are routinely monitored by clinical laboratories via internal QC procedures.
Table 1 summarizes the within-run CVs observed for the IMMULITE 2000. These meet the NCCLS I/LA20-A performance targets for within-run imprecision, namely < 10% at "low," "medium," and "high" concentrations. Also tabulated are CVs at "very low" concentrations – down to 0.2 kU/L, which is below the range where second generation assays can yield quantitative results without extrapolation. Fig. 2 shows "total" (interassay) CVs generated in the 2 EP5 experiments. These meet NCCLS I/LA20 A performance targets for "total" imprecision, <15% at "medium" or "high" concentrations and <20% at "low" concentrations, ie, close to 0.35 kU/L.
View larger version (15K):
[in this window]
[in a new window]
|
Fig. 2. Total ("interassay") CVs from two separate NCCLS EP5 precision studies of the IMMULITE 2000 sIgE assay, fitted with a precision profile. The allergens included D1, D2, E1, E5, F13, F14, G6, H2, K82, M2, M6, and T3, as well as an IgE control. The relationship between extended classes (ECS) and sIgE concentration (in kU/L) is indicated along the horizontal axis. A horizontal dotted line indicates the 20% interassay CV level used to define "functional sensitivity."
|
|
Also shown in Fig. 2
is a curve fitted to the EP5 "total" CVs. This "precision profile" – which represents the interassay CVs for the IMMULITE 2000 sIgE assay as a function of concentration across its working range – indicates a "functional" sensitivity of 0.2 kU/L for the assay. Although originally developed for TSH [15], the concept of functional sensitivity and its operational definition in terms of a 20% interassay CV have been widely adopted for other analytes, as a means of indexing the concentration range over which an assay can yield clinically useful results. Accordingly, the functional sensitivity of an assay is defined as the lowest concentration at which it yields an interassay CV of 20%. This concept has gained acceptance as an objective indication of the practical lower limit of an assay’s usefulness; it is more relevant than indices based on within-run CVs or detection limits.
Dilutional linearity.
Fig 3 shows the linearity-under-dilution performance of a representative second generation sIgE assay on samples and calibrator [16]. The dilution factors used were between 1 and 100, extending down only to 0.5 kU/L. Figs. 4 and 5 show results for the IMMULITE 2000 sIgE assay, illust-rating the broader range of dilutional linearity that can be achieved by a third generation assay. The dilution factors were between 1 and 1000, extending down to 0.1 kU/L, the assay’s conservatively stated detection limit (which is well below the standard 0.35 kU/L cutoff).
View larger version (27K):
[in this window]
[in a new window]
|
Fig. 3. Dilutional linearity for a representative second generation sIgE assay, using both samples and calibrators. From Yunginger et al [17]. Reproduced with permission of the publisher (Elsevier).
|
|
View larger version (44K):
[in this window]
[in a new window]
|
Figs. 4 and 5. Dilutional linearity for the IMMULITE 2000, a third generation sIgE assay. From top to bottom: F8, M2, D3, I1, W1, K82 (Fig. 4) and Calibrator, G2, E1, T6, F33 (Fig. 5).
|
|
Interference.
Serum samples (N = 7) containing varying amounts of sIgE (1.4 to 64 kU/L) for E1, E5, T3, G6, G2, M6 and D1 were spiked with increasing amounts of bilirubin, hemoglobin, or triglycerides. Spiked and unspiked samples were then assayed by the IMMULITE 2000. Bilirubin (up to 20 mg/dl), hemoglobin (up to 500 mg/dl), and triglycerides (up to 3000 mg/dl) did not have any significant effect on the sIgE results.
Crossreactivity.
Subclasses of naturally occurring immunoglobulins (IgA, IgD, IgG, and IgM) were spiked into the IMMULITE 2000 sIgE assay’s zero calibrator. (All dilutions were at least 1 in 20.) With % crossreactivity defined as 100 times the apparent concentration, divided by the amount added, crossreactivity with IgA, IgD, IgG, and IgM proved <0.001% in each case, as recommended in the NCCLS I/LA20 A guidelines.
Nonspecific binding.
Three samples containing high total IgE levels (greater than 1000 kU/L) were tested with 22 different allergens, including 2 dusts, 1 animal, 5 foods, 5 molds, 3 trees and 6 weeds. The IMMULITE 2000 sIgE results were all below 0.1 kU/L, the assay’s detection limit.
Method comparisons.
The IMMULITE 2000 sIgE assay yields quantitative results in kU/L, but these can be translated into either standard or extended class scores. Fig. 6 shows a comparison of the IMMULITE 2000 sIgE assay and the manual Hycor mRAST in terms of extended class scores. Forty-eight serum samples from atopic individuals containing different levels of antibodies to a variety of allergens were obtained from a serum bank. There were a total of 812 paired results, representing 66 allergens, including 5 animals, 1 dust, 10 foods, 7 grasses, 4 insects, 2 mites, 8 molds, 15 trees, and 14 weeds. Relative to the mRAST, clinical sensitivity, specificity, and concordance were 88%, 92%, and 90%, respectively. All of the paired results agreed within ±2 classes, most (90%) agreed within ±1 class, and over two-thirds (69%) of the samples yielded identical class scores. The comparison demonstrates good agreement between the automated IMMULITE 2000 sIgE and the manual Hycor mRAST in terms of class scores, despite the methodological differences between the two assay systems.
View larger version (35K):
[in this window]
[in a new window]
|
Fig. 6. Comparison of the IMMULITE 2000 to the manual Hycor mRAST, a first generation sIgE assay, in terms of extended class scores.
|
|
In another study, the IMMULITE 2000 sIgE assay was compared – in quantitative terms, rather than classes – to the Pharmacia FEIA, a second generation assay. Forty-nine serum samples from atopic donors in another serum bank were tested for 17 different allergens – including D1, D2, E1, E2, E5, F1, F3, F4, F14, F76, F77, F78, G6, K82, M6, T3, and W6 – for a total of 169 paired results between 0.35 and 100 kU/L by both methods. Linear regression analysis yielded the following relationship: IMMULITE 2000 = 0.995 (Pharmacia FEIA) + 1.986 kU/L, r = 0.859, n = 169. Fig 7 shows that the paired results are fairly evenly distributed above and below the "line of identity." The comparison demonstrates good agreement between the automated IMMULITE 2000 sIgE and the Pharmacia FEIA in terms of concentration (kU/L), despite methodological differences.
View larger version (25K):
[in this window]
[in a new window]
|
Fig. 7. Comparison of the IMMULITE 2000 to the Pharmacia FEIA, a second generation sIgE assay, in terms of kU/L. The diagonal line represents the "line of identity."
|
|
|
Discussion
|
|---|
At present, 3 manufacturers offer sIgE assays with claims that the tests are quantitative (Diagnostic Products Corporation’s IMMULITE 2000 package insert, Pharmacia’s UniCAP package insert, and Hycor’s Turbo mRAST package insert). All use calibration traceable to the WHO IgE reference standard 75/502.
The first generation RAST assay requires 2 overnight incubations, and the second generation FEIA provides results in 3 hr, whereas the third generation IMMULITE 2000 sIgE assay has a time-to-first-result of about 65 min.
Second generation assays lack a true zero calibrator: the lowest calibrator has a concentration of 0.35 kU/L. Although the Turbo MP does permit the reporting of explicit results below 0.35 kU/L, these are obtained by extrapolation, from percentage-of-signal at 0.35 kU/L, down to a concentration range not supported by the calibration curve (Fig. 8). If a second generation assay simply reports such results as <0.35 kU/L, then the assay reverts, in effect, from quantitative to ASM class-based reporting of results obtained by extrapolation, below the concentration of its lowest calibrator. With the recent emphasis on early detection of antibodies against allergens [17], explicit, calibration curve-based information on sIgE concentrations less than 0.35 kU/L may be of clinical interest because allergen avoidance can significantly reduce both the degree of sensitization and the development of allergic symptoms [18].
View larger version (13K):
[in this window]
[in a new window]
|
Fig. 8. Illustration of extrapolation and assumed (virtual) zero calibrator and cutoff response in second generation sIgE assays (such as Turbo MP and ASM). Note that the IMMULITE 2000 sIgE assay uses two calibrators below 0.35 kU/L, including a true zero calibrator.
|
|
The third generation sIgE assay studied in this paper overcomes the limitations exhibited by first and second generation sIgE assays, including the lack of a true zero calibrator, experimentally undefined assay detection limit, and poor precision and accuracy [19]. The third generation IMMULITE 2000 sIgE assay has immunoassay design features such as liquid phase allergens, monoclonal detection antibody, enzyme-enhanced chemiluminescent signal detection; use of an actual (rather than a virtual) zero calibrator, allowing explicit results as low as 0.1 kU/L to be reported by interpolation (rather than extrapolation) from the calibration curve; a detection limit of 0.1 kU/L; a functional sensitivity of 0.2 kU/L; dilutional linearity down to 0.1 kU/L; and "user convenience" and efficiency features such as continuous random access automation and a time-to-first-result of approximately one hr. Many advantages of this assay are derived from its detection system based on enzyme-enhanced chemiluminescence, which is orders of magnitude more sensitive than other conventional methods of signal detection [20].
Table 2 lists the characteristics of the 3 generations of sIgE assays. Benefits of the fully automated, third generation IMMULITE 2000 sIgE assay include an extended working range; a lower, well-defined detection limit; improved precision in the neighborhood of the standard 0.35 kU/L cutoff; and reductions in allergen and sample setup time with a reduced time-to-first-result of 65 min, instead of multiple hr or days.
Although the NCCLS I/LA20-A guidelines [13] were issued in 1997, the current study represents, to our knowledge, the first published report in which these guidelines have been used for complete validation of the analytical performance of an sIgE assay. Evaluation parameters detailed in NCCLS I/ LA20-A include linear range, within-run (intraassay) imprecision, total (interassay) imprecision, detection limit, dilutional linearity, nonspecific binding, correlation to other methodologies, calibration curve stability, interferences, and crossreactivity. The IMMULITE 2000 assay meets all the performance targets set for the NCCLS I/LA20-A evaluation criteria. The clinical performance of this sIgE assay in comparison with skin testing will be reported elsewhere (Ollert M et al, in preparation), as will interlaboratory and intermethod comparisons (Fu PC, Zic V, in preparation), and a multicenter precision profile study (Oosterom R, in preparation).
|
References
|
|---|
- Johansson SG, Hourihane JO, Bousquet J, Bruijnzeel-Koomen C, Dreborg S, Haahtela T, Kowalski ML, Mygind N, Ring J, van Cauwenberge P, van Hage-Hamsten M, Wuthrich B. A revised nomenclature for allergy. Allergy 2001;56:813–824.[Medline]
- Wide L, Bennich H, Johansson SG. Diagnosis of allergy by an in-vitro test for allergen antibodies. Lancet 1967;2:1105–1107.[Medline]
- Williams PB, Dolen WK, Koepke JW, Selner JC. Immunoassay of specific IgE: use of a single point calibration curve in the modified radioallergosorbent test. Ann Allergy 1992;69:48–52.[Medline]
- Practice Standard Committee. Skin testing and radio-allergosorbent testing (RAST) for diagnosis of specific allergens responsible for IgE-mediated diseases. J Allergy Clin Immunol 1983;72:515–517.
- Bunstein IL. The proceedings of the task force on guidelines for standardizing old and new technologies used for the diagnosis and treatment of allergic diseases. J Allergy Clin Immunol 1988;82:487–526.[Medline]
- Ownby DR. Tests for IgE antibody. In: Allergy, Asthma, and Immunology from Infancy to Adulthood (Burnman CW, Pearlman DS, Shapiro GG, Busse WW, Eds), 3rd ed, Saunders, Philadelphia 1996; pp 144–156.
- Hamilton RG, Adkinson NF Jr. Clinical laboratory assessment of IgE-dependent hypersensitivity. J Allergy Clin Immunol 2003;111(Suppl 2):S687–S701.[Medline]
- Kelso JM, Sodhi N, Gosselin VA, Yunginger JW. Diagnostic performance characteristics of the standard Phadebas RAST, modified RAST, and Pharmacia CAP system versus skin testing. Ann Allergy 1991;67:511–514.[Medline]
- Plebani M, Bernardi D, Basso D, Borghesan F, Faggian D. Measurement of specific immunoglobulin E: inter-method comparison and standardization. Clin Chem 1998;44:1974–1979.[Abstract/Free Full Text]
- Dolen WK. IgE antibody in the serum: detection and diagnostic significance. Allergy. 2003;58:717–723.[Medline]
- Li TM, Chuang T, Tse S, Hovanec-Burns D, El Shami AS. Development of a third generation allergen-specific IgE assay on the continuous random access IMMULITE 2000 analyzer. Clin Chem 2003;49(S6):A20.
- Tse S, Chuang T, Li TM, Hovanec-Burns D, El Shami AS. Analytical performance evaluation of the allergen-specific IgE assays on the IMMULITE 2000 System. Allergy 2003;58 (Suppl 74):364.
- Matsson P, Hamilton RG, Adkinson NF Jr, Esch R, Homburger HA, Maxim P et al. Evaluation methods and analytical performance characteristics of immunologic assays for human immunoglobulin E (IgE) antibodies of defined allergen specificities. National Committee for Clinical Laboratory Standards (NCCLS), Wayne, PA, Approved guideline I/LA20, 1997;17:24.
- Yman L. Allergy. In: The Immunoassay Handbook (Wild D, Ed) 2nd ed, Nature Publ, London, 2001; pp 664.
- Spencer CA, Takeuchi M, Kazarosyan M, MacKenzie F, Beckett GJ, Wilkinson E. Interlaboratory/intermethod differences in functional sensitivity of immunometric assays of thyrotropin (TSH) and impact on reliability of measurement of subnormal concentrations of TSH. Clin Chem 1995;41:367–374.[Abstract/Free Full Text]
- Yunginger JW, Ahlstedt S, Eggleston PA, Homburger HA, Nelson HS, Ownby DR, et al. Quantitative IgE antibody assays in allergic diseases. J Allergy Clin Immunol 2000;105:1077–1084.[Medline]
- Sasai K, Furukawa S, Muto T, Baba M, Yabuta K, Fukuwatari Y. Early detection of specific IgE antibody against house dust mite in children at risk of allergic disease. J Pediatr 1996;128:834–840.[Medline]
- Hide DW, Matthews S, Tariq S, Arshad SH. Allergen avoidance in infancy and allergy at 4 years of age. Allergy 1996;51:89–93.[Medline]
- Williams PB, Barnes JH, Szeinbach SL, Sullivan TJ. Analytic precision and accuracy of commercial immuno-assays for specific IgE: establishing a standard. J Allergy Clin Immunol 2000;105:1221–1230.[Medline]
- Kricka LJ. Principles of immunochemical techniques. In: Tietz’s Textbook of Clinical Chemistry (Burtis CA, Ashwood ER, Eds), 3rd ed. Saunders, Philadelphia, 1999; pp 205–225.
This article has been cited by other articles:
|
|
|
|
 
K. J. Kontis, A. Valcour, A. Patel, A. Chen, J. Wang, J. Chow, and N. Nayak
Correlation of the Turbo-MP RIA with ImmunoCAP FEIA for Determination of Food Allergen-Specific Immunoglobulin E.
Ann. Clin. Lab. Sci.,
December 1, 2006;
36(1):
79 - 87.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. Ollert, S. Weissenbacher, J. Rakoski, and J. Ring
Allergen-Specific IgE Measured by a Continuous Random-Access Immunoanalyzer: Interassay Comparison and Agreement with Skin Testing
Clin. Chem.,
July 1, 2005;
51(7):
1241 - 1249.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
M. H. Sohn, S.-Y. Lee, K. E. Lee, and K.-E. Kim
Comparison of VIDAS Stallertest and Pharmacia CAP Assays for Detection of Specific IgE Antibodies in Allergic Children
Ann. Clin. Lab. Sci.,
January 1, 2005;
35(3):
318 - 322.
[Abstract]
[Full Text]
[PDF]
|
|
|
Top
Abstract
Introduction
