Examen De Evaluación

CLIN. CHEM.

32/10, 1884-1 889 (1986)

External Quality Assessment
David
G. Bullock,1’2

of Assays of Lead in Blood
and

Norman

J. Smith,3

Thomas

P. Whltehead2
After several years’ experience with the U.K. External Quality Assessment Scheme (UKEQAS) for General Clinical Chemistry (formerly the National Quality Control Scheme) at Birmingham (7, 8), our two laboratoriesinitiated jointly a similar national scheme for lead in blood.4 Distribution of specimens commenced in 1973, and the scheme has since been extended to include participants outside the U.K. Here we describe the history and operation of the scheme, present some of the results obtained, and comment on the current state of the art. Materials

The operation and results of the United Kingdom External QualityAssessment Scheme for Lead in Blood, which currently has about 140 participants within and outside the U.K. are described. The choice of specimens, scheme design, and scoring of performance are discussed, as is the validity of the consensus values used as a basis for the scoring. There has been a continued improvement in performance as assessed by this program, some of which appears to have beenstimulated by the introduction of Variance Index scoring. The relative performance of the methods for assaying lead in blood is also examined. of lead in blood is important in diagnosing suspected toxicity, detecting excessive industrial or environmental exposure, and monitoring the cases found by such testing or screening procedures. Its reliability in all these situations, particularly with theincreasing emphasis in the U.K. on blood lead values in occupational health enforcement (1), obviously requires that the results produced in all laboratories must be both precise (reproducible) and accurate (free from bias). Otherwise, incorrect clinical decisions might be made and data from multicenter studies could be misinterpreted. Despite this, the literature contains few reports ofinterlaboratory surveys, and even fewer of regular external quality-assessment schemes; the scanty data that have been published indicate that the situation is not satisfactomy. Berlin and his colleagues (2,3) have carried out surveys in Europe of lead, mercury, and cadmium assays with blood and urine specimens. In 1972 the results for blood lead determined in 22 laboratories ranged from 0.7 to 4.1 imol/L(median of 2.4 moI/L); a larger continuing study (52 participants) in 1974 showed a between-laboratory CV of 43% at a median of 1.1 p.moJ/L (range 0.5-4.2 pmoliL). In Scandinavia, Paulev et al. (4) reported CVs of about 12 and 17% at lead concentrations of 1.0 and 2.9 pmol/L among five participants in 1977. The Centers for Disease Control in the U.S., which has carried out surveys at aboutthree-monthly intervals since 1974 as part of a proficiency testing program, reported a similar performance in 1979 (5), with typical CVs of 22% for lead at 2.5 pmol/L. In addition, the most commonly used methods were shown, as expected, to be positively biased (in comparison with a definitive method) at low lead concentrations and negatively biased at high concentrations; mean results were accuratearound 2.0 mo1fL. More recently Saltzman (6) has collated data for 1979-1983 from three schemes in North America, yielding CVs of 10, 13, and 19% for 28, 54, and 237 laboratories at mean lead concentrations around 2.5 imol/L.
patients with 1Wo!fson Chemistry,

and Methods
Selection and Preparation

The

assay

Specimen

At the inception of the scheme, human blood was chosen as thespecimen. Specimens are prepared at the Health and Safety Executive Occupational Medicine and Hygiene Laboratories (HSE), which arranged with a local hospital to provide one unit of blood every two weeks to meet the requirements of the scheme. The blood is given by volunteer donors who understand the purpose for which it is to be used. The blood is collected into an acid-washed bottle with dipotassium...