[Biological limits of exposure: evolution of interpretative and methodologic criteria]

Biological monitoring is an extremely efficient investigation tool in evaluation of exposure to exogenous substances, in both occupational and environmental settings. However, although biological tests have been widely acclaimed as the ideal approach to or as a completion of environmental measurements, they have had difficulty in becoming firmly established due to a number of limitations of various nature. Besides the numerous uncertainties in the set-up phase, which depend on the choice of indicators or criteria for their use, difficulties are still incurred in making a univocal interpretation of the results. These difficulties are due to the limited pharmaco-kinetic information available on the various substances and to the wide variability in individual biological response. The possibility of extrapolating Biological Exposure Limits (BELs) from the corresponding TLV-TWA, by means of calculations based on the experimental regressions observed between internal and external dose indices, has up to now been considered a dubious operation from a formal point of view and merely indicative for practical purposes. This paper examines the possibility of establishing BELs that are fixed taking account of the influence of the various biological variables, thus permitting a more correct, objective and generalised use of biological indicators in current practice. From the regression function and relative tolerance limits, intended as a range of values within which regression values can be expected to be found with a probability that can be fixed a priori, it is possible to calculate 3 BEL values for each environmental TLV-TWA concentration. The question of which of these different values should be selected as a BEL for practical purposes must be solved on the basis of a series of observations related to the type of investigation being performed and the context in which the results are to be used. It should be emphasized that each BEL value has its own particular sensitivity and specificity, i.e., the possibility of correctly classifying the exposure conditions in relation to the environmental values. If sensitivity and specificity values are known, it is possible to calculate the predictive value of the biological measurements compared to the environmental measurements; if numerical data on these parameters are available, account can be taken of the variability of the results in a rigourously scientific manner.(ABSTRACT TRUNCATED AT 400 WORDS)