| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Letters to the Editor |
1 Department of Clinical Neuroscience Karolinska Institute, Stockholm
2 External Quality Assurance in Laboratory Medicine in Sweden (EQUALIS AB), Uppsala, Sweden
aAddress correspondence to this author at: Alcohol Laboratory L7:03, Karolinska University Hospital Solna, SE-171 76, Stockholm, Sweden, Fax +46-8-51771532, e-mail anders.helander{at}ki.se
To the Editor:
Measurement of carbohydrate-deficient transferrin (CDT) can reveal alcohol-related changes in the serum transferrin glycosylation pattern. CDT is a more alcohol-specific indicator than liver-function tests and is used for identification and follow-up of chronic high alcohol consumption(1). Therefore stable calibration of the assays is very important. Comparison of CDT results between methods has often been hindered by method-dependent discrepancies in the definition of the measurand (the transferrin glycoforms covered) and the way results are expressed. With some methods there has been an increased risk for false-positive results(2). The lack of CDT standardization prompted initiation of a working group under the International Federation of Clinical Chemistry and Laboratory Medicine, whose aim was to define the measurand, select and validate a reference method, and work out procedures for the production of reference materials. The first recommendation was that the fraction of disialotransferrin to total transferrin (%disialotransferrin) should be the primary target for CDT testing, with HPLC as the candidate reference method(3).
The performance of individual laboratories and agreement of different methods can be determined through external quality assessment (EQA). A Swedish EQA scheme for CDT has been run by EQUALIS (External Quality Assurance in Laboratory Medicine in Sweden) since 1996. Each year, 10 samples (human serum pools without preservative) whose target %disialotransferrin values are set by selected expert laboratories that use HPLC(3) are distributed to the participants.
In-house or commercial HPLC assays are the most common CDT assays (n = 19) in the EQUALIS EQA scheme. HPLC methods involve separation of iron-saturated transferrin glycoforms by anion-exchange chromatography, and quantification by selective absorbance of the iron-transferrin complex at approximately 460 nm(4). The disialotransferrin fraction is calculated as relative peak area using baseline integration, in agreement with the recommendations of the CDT standardization work(3). The second most common method (n = 15) is the N Latex CDT direct immunonephelometric assay (Siemens, formerly Dade Behring). The N Latex CDT assay uses a monoclonal antibody that recognizes transferrin glycoforms lacking 1 or 2 complete N-glycans (i.e., disialo-, monosialo-, and asialotransferrin) and on a simultaneous transferrin immunoassay(5). The %CDT value (fraction of disialo-, monosialo-, and asialotransferrin to total transferrin) is calculated automatically.
In a multicenter evaluation of N Latex CDT(5), the %CDT results correlated well with the %disialotransferrin results by HPLC, but owing to the different measurands the numerical values were not interchangeable. For healthy controls, the upper 97.5th percentile for %disialotransferrin by HPLC was approximately 1.7%(4) compared with approximately 2.35% for %CDT by N Latex CDT(5). A discrimination limit of 2.5% (99th percentile) is proposed in the N Latex CDT package insert, and is commonly applied in clinical practice. However, starting in 2006, the relation between the HPLC and N Latex CDT values in the EQUALIS EQA surveys has gradually changed. At %disialotransferrin values around 2% by HPLC, the corresponding %CDT results by N Latex CDT were on average 0.4% higher in April 2006, but roughly identical in September 2007 (Fig. 1A
). This change was observed over the entire measuring range (Fig. 1B
).
|
Four serum pools with %disialotransferrin target values of 1.02%, 1.75%, 2,61%, and 3.52% that were stored frozen since their use in a Swedish CDT harmonization program in 2002–2003 were reanalyzed by HPLC in November 2007. The resulting values were almost identical to the original ones (1.00%, 1.83%, 2.58%, 3.53%; r2 = 0.997), confirming that the HPLC target values were stable over the study period.
The change in the relation between HPLC and N Latex CDT values thus seems to be related to a change in the calibration of N Latex CDT when new reagent batches were released. During the study period, at least 6 lots have been in use. The results indicate that the manufacturer-recommended cutoff limit for N Latex CDT at 2.5% is currently too high and will produce false-negative clinical results (patients with harmful drinking habits remain undetected and untreated). Based on the comparison with the HPLC results in the EQUALIS EQA scheme, the current 99th percentile for %CDT by N Latex CDT should be approximately 1.9%–2.0% (Fig. 1B
).
According to international directives, manufacturers of in vitro diagnostic products are required to provide reference intervals and control the conformity of each batch before placement on the market. However, the present observation highlights the need for laboratories to verify the cutoff limit for every new batch of N Latex CDT, and the value of EQA schemes. For an improved standardization of CDT, a reference method and reference materials are needed.
Acknowledgments
Grant/Funding Support: None declared.
Financial Disclosures: None declared.
Acknowledgments: The assistance of Elisabeth Nilsson, CDT scheme organizer at EQUALIS, is highly appreciated.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
