Search form

OLV Ziekenhuis Waregem - klinisch laboratorium

Contact details
Traineeship proposition

Abstract Bachelor project 2017-2018: Evaluation of the Atellica NEPH 630 nephelometer for protein chemistry 

Determination of proteins in serum, urine and cerebrospinal fluid make it possible to make an estimation of the physiological state of a patient and can therefore be important for diagnosis or follow-up of several diseases. At the department of Laboratory Medicine at the Hospital of Waregem, these important biomarkers are measured on the nephelometer IMMAGE ® 800 (Beckman Coulter, CA, USA), an old platform. A suitable nephelometer for its replacement is currently under evaluation, the Atellica NEPH 630 ® (Siemens, Erlangen, Germany).

The aim of this essay is to evaluate the Atellica NEPH 630® nephelometer and compare it to the IMMAGE 800® for a large range of assays.

Precision, bias and allowable total error are evaluated with control materials on the Atellica NEPH 630. Furthermore, patient serum, urine and cerebrospinal fluid samples are collected for determination on IMMAGE 800® and Atellica NEPH 630®. Parameters under evaluation are alpha-1-antitrypsin, alpha-1-microglobulin, albumin in serum, urine and CSF, antistreptolysin-O, Bèta-2-microglobulin, C1 esterase inhibitor, Complement 3 and 4, ceruloplasmin, haptoglobin, immunoglobulin A, G, M and immunoglobuline IgG2, 3 and 4 and rheumatoid factor. 

Imprecision experiments on patient material (within-run), manufacturer’s control materials (within- and between-run) and third-party control materials (between-run) show good results as evaluated by several criteria for imprecision, bias and total error. Only one parameter, ceruloplasmin, exceeded total error specifications for manufacturer’s control materials because of bigger contributing bias (limited on-board stability). Method comparison results show overall good correlation (Rs>0.95) except for 4 parameters mainly due to limited sample numbers. Passing-Bablok regression analysis is acceptable although multiple proportional and systematic statistically significant biases were to be noted. To be highlighted is the difference for IgG subclasses 3 and 4 with major proportional biases to be noted. When consulting literature, these differences are to be expected because of historically different calibration materials. Cross table summaries of the patient samples evaluated showed no major discrepancies for all parameters, except for IgG subclasses 3 and 4 which is well known.

In all, differences between the IMMAGE 800® and Atellica NEPH 630® are within limits for nephelometric protein chemistry which makes the Atellica NEPH 630 ® suitable for routine practice. A possible switch for IgG subclasses is to be done with caution as major differences in standardization limits transferability and shows the need for reagent specific age-dependent reference intervals.


Vijfseweg 150


Line Coucke
Maxime De Sloovere
Via Map