Diurnal Variability and in Vitro Stability of Carbohydrate-deficient Transferrin

Pharmacia & Upjohn Diagnostics AB, Alcohol Related Diseases, S-75182 Uppsala, Sweden
^a Author for correspondence. Fax 46-18 16 63 89; e-mail ragnhild.brandt{at}eu.pnu.com.

To the Editor:

Transferrin isoforms with pI values 5.7, known as carbohydrate-deficient transferrin (CDT), are present in increased concentrations in the serum of patients with current alcohol abuse. CDT has been found to be a specific and sensitive marker for detection and monitoring of high and continual alcohol consumption (1). The clinical performance of CDT has been studied using various techniques for analysis but also using various conditions for serum sample collection and sample storage. The biological (non-alcohol-influenced) variation over time has been found to be low (2). We have studied the diurnal variability of CDT and the effect of sample storage conditions in the commercial CDTect assays.

Blood was collected by venipuncture and allowed to clot at room temperature. The serum samples were separated by centrifugation, and 1-mL portions were aliquoted into polypropylene tubes (Sarstedt). CDT was measured by duplicate determinations in CDTect^® RIA or CDTect EIA (Pharmacia & Upjohn Diagnostics AB). Results are expressed as units per liter. In an evaluation performed on 76 serum samples the assays gave comparable results and showed good correlation, with a correlation coefficient of 0.99 and a regression line of EIA = 0.72 0.90 x RIA. The within-assay and between-assay coefficients of variation for the RIA were 7.2% and 9.3%, respectively; for the EIA, they were 6.5% and 9.3%, respectively.

To study the diurnal variation, we drew blood from 10 healthy in-house volunteers (CDT, 8–29 units/L) four times during a 24-h period, at 0800–0900, 1200–1300, 1600–1700, and 2100–2200. No restrictions regarding food intake were given. The serum samples were stored at -20 °C until analysis. The CDT concentrations from the different sampling times were calculated as percentages of the daily mean for each person. For the whole group, the mean values were 97%, 106%, 95%, and 102%, respectively. The variation between the sampling times was not statistically significant when analyzed using a linear model.

The effects of freezing and repeated freezing and thawing were investigated. Serum samples were collected from 26 subjects (9 females and 17 males; CDT, 11–62 units/L). Each sample was divided into two tubes, one stored at -20 °C and the other at 2–8 °C until analysis 1 day later. The mean ratio (frozen/unfrozen) was 98% (range, 86–112%). For the study of repeated freezing and thawing, serum samples from 17 subjects (CDT, 9–74 units/L) were used. Each sample was divided into two tubes and stored at -20 °C. One tube was subjected to 10 freeze-thaw cycles during a 2-year period. The ratio (thawed 10 times/thawed once) was 101% (range, 88–109%).

The stability of CDT in serum at 2–8 °C and 32 °C for a limited period of time and at -20 °C for a long period of time was studied.

First, serum samples from nine subjects (CDT, 9–24 units/L) were aliquoted into four tubes and frozen at -20 °C. Three of the tubes were transferred to 2–8 °C 72 h, 48 h, and 24 h before analysis. The fourth tube was thawed immediately before analysis and considered as the reference. CDT on all variants was determined in the same analytical run. Mean values for 24-, 48-, and 72-h storage were 103%, 98%, and 93%, respectively. One of the nine sera showed a decrease of >15% when stored for 72 h.

Next, serum samples from 20 subjects (CDT, 13–34 units/L) were divided into four tubes, frozen, and transferred to 32 °C at various times before analysis. Mean ratios for samples stored for 24, 48, and 72 h against serum thawed in immediate conjunction to the CDT determination were 102%, 96%, and 96%, respectively. The CDT values in three sera decreased >15% when stored for 72 h at this higher temperature. None increased >15%.

In a stability study of CDTect RIA, four serum samples (15, 17, 35, and 70 units/L) were included and stored at -20 °C. The stability was followed for 32 months (Table 1 ). Three of the samples were still available after 8 years. CDT concentrations at that time were, on average, 97% of the initial values.

In view of these results, we conclude that the diurnal variability of CDT is low. Thus, it does not seem to be necessary to restrict sample collection to a special time of the day.

We have also found a very good stability of CDT in most serum samples. Freezing and repeated freezing and thawing did not significantly influence the results. In general, CDT also showed high stability in thawed serum when stored both in the refrigerator at 2–8 °C and at high room temperature, 32 °C. However, one cannot exclude the possibility that individual samples may exhibit poor stability because of endogenous enzymes or bacterial contamination.

The finding of practically no change in CDTect values for sera stored >8 years at -20 °C points to a very good stability of the analyte in frozen serum. This may be of great importance for many clinical studies where the samples are stored for long periods of time before analysis.

References

1. Stibler H. Carbohydrate-deficient transferrin in serum: a new marker of potentially harmful alcohol consumption reviewed. Clin Chem 1991;37:2029-2037. [Abstract/Free Full Text]
2. Borg S, Helander A, Voltaire-Carlsson A, Högström-Brandt A-M. Detection of relapses in alcohol-dependent patients using carbohydrate-deficient transferrin: improvement with individualized reference levels during long-term monitoring. Alcohol Clin Exp Res 1995;19:961-963. [ISI][Medline] [Order article via Infotrieve]