Immunoturbidimetric determination of serum transferrin on a Kone Progress autoanalyser

The Kone Progress, a multiparametric discrete analyser, was used to determine serum transferrin with a kit supplied by Kone. Assays recommended by the French Society of Clinical Chemistry were performed in order to assess the suitability of the test. Repeatability was assessed using serum pools with low (L), medium (M) and high (H) concentrations of transferrin. The coeffcients of variation (CV) were 5.4, 3.2 and 2.0% respectively for 30 determinations (within-batch). Reproducibility on 15 consecutive days (between-batch) was also satisfactory (CV for L = 7.3%, M = 6.3% and H = 3.8%). There was no serum-to-serum contamination. Results correlated closely with those obtained using radial immunodiffusion (RID) (r = 0.942) and total iron-binding capacity (r = 0.954)for 90 determinations. Transferrin measurement by immunoturbidimetry on the Kone Progress emerges as a well-suited, rapid and inexpensive alternative to other time-consuming (RID) and sophisticated (laser immunonephelemeter) techniques.


Materials and methods
Immunoturbidimetry was performed using the transferrin test kit (Kone, Finland) containing the anti-transferrin anti-serum and polyethylene glycol to enhance the immunoprecipitation. The apparatus (Progress, Kone, Finland) [9] was used under the conditions described in The non-linear calibration curve was obtained using the protein calibration set (supplied by Kone) containing transferrin at five known concentrations (from to 8 g/l). Samples of serum and calibrators were diluted at 1/11 before use.
The measurement of serum transferrin is recognized as an indispensable parameter in the exploration of iron metabolism and its measurement is recommended to replace total iron binding capacity (TIBC), the old indirect method, for the evaluation oftransferrin in serum [1]. Furthermore, serum transferrin is considered to be a good marker of chronic and semi-acute malnutrition [2] and is a useful parameter in following the nutritional status of patients receiving nutritional support [3]. Radial  In this work the,efficiency of the measurement of serum transferrin has been evaluated by immunoturbidimetry using a discrete analyser, the Kone Progress, routinely used for biochemical assays in the Laboratoire de Biochimie A, at the H6pital Saint Antoine.
This work is dedicated to Marie-Thg Gaubert .

Results
Within-run and between-run precision was tested with pools of serum from patients at the hospital: low (L), medium (M)and high (H)transferrin concentrations (different pools were used for within-run and betweenrun assays). The within-run assay consisted of 30 consecutive measurements for each pool; and the between-run of measurements on 15 consecutive days. Results are given in table 2.
Sample dispenser carry-over assessment was performed measuring (10 times) a sample with low transferrin (L1), and (10 times) a sample with high transferrin (HI) and then performing the sequence H2HL2L six times where H2 is the sample H1 measured in first position, H is the   The method comparison was performed using 90 patient sera, which were analysed by TIBC determination, RID and immunoturbidimetry (IT) (results expressed in g/l). Results are presented in figure and in table 3.
[6]: in this work, as in our own, there is a systematic and constant difference in values (intercept different from 0, slope near 1). In the same way, immunoturbidimetry gives lower values than laser nephelometry [4]. These differences requires that the values obtained be compared to normal values for a given technique: normal values of serum transferrin range from 2"50 to 3"50 g/1 using immunoturbidimetry [4 and 7].

Discussion
Transferrin has been assessed indirectly by the measurement of TIBC for some years. However, there is now a general consensus [1] that advances in clinical biochemistry make the direct determination of transferrin preferable to TIBC. The most common methods are IDR and laser nephelometry; but the former takes 48 hours and the latter requires expensive, dedicated apparatus. Immunoturbidimetry thus appears as an attractive alternative, especially when automatized on a routine analyser. Immunoturbidimetric determination of transferrin on the Kone Progress proved to be accurate, reliable and inexpensive. It is interesting to note that immunoturbidimetry gave significantly lower values than RID (T 27"4; p < 0"001). There is no clear explanation for this difference but this has previously been reported