Blood lead and cadmium determination: Results of the Italian external quality assessment scheme

Abstract

The procedures adopted in the Italian external quality assessment scheme (EQAS) for blood Pb (B-Pb) and Cd (B-Cd) determination, including the preparation of control materials, are described. Each scheme involves the use of internal quality control materials and the participation in periodical external quality assessment trials. All control materials are prepared at the Laboratory of Clinical Biochemistry, Italian National Institute of Health (Istituto Superiore di Sanità, ISS), as the reference centre. Computerized procedures have been adopted whenever possible, i.e. sample randomization, data transfer and treatment. The analytical performance of the participating laboratories, evaluated from the results obtained in ten years of activity, is reported. A total of eight phases has been carried out between 1983 and 1993. Over the course of these ten years, a positive trend was observed for B-Pb towards a smaller bias and a lower percentage of laboratories with poor performance. In the eighth phase, taking into account the results from all participants, the mean absolute inaccuracy for B-Pb was 32 μg/l (mean Pb concentration in control samples, 165 μg/l). The percentage of laboratories providing results within established acceptability limits for at least 80% of the examined samples (good performers) increased with time from 28% in the first phase to 48% in the last phase. The laboratories providing less than 50% of acceptable results (poor performers) decreased from 35% in the first phase to 22% in the last phase. As regards B-Cd, mean absolute inaccuracy decreased from 0.96 μg/l to 0.55 μg/l between the first phase and the sixth phase. The percentage of good performers increased from 45% to 62% and that of poor performers decreased from 31% to 5%. From the seventh phase a large group of new laboratories agreed to participate in the EQAS for B-Cd and the inaccuracy rose to 0.75 μg/l (mean concentration: 2.7 μg/l) and the percentage of poor performers rose to 24%, whereas the percentage of good performers remained almost unchanged.