An International Study on Feasibility of Estimating Polychlorinated Biphenyls by Using Specific Polychlorinated Biphenyl Congeners

The feasibility of using specific PCB congeners was studied in two parts: a questionnaire to obtain information on the methodology used in various countries for quantification of PCBs and an international check sample program to evaluate the accuracy of the specific PCB congener approach. Information from the questionnaires indicated the use of a wide variety of PCB quantification procedures and considerable interest in using specific PCB congeners. In the check sample program, participants were requested to analyze two mixtures of Aroclor 1242 and 1254 in different proportions, and a human milk extract. The results indicated that when individual congeners were present at less than approximately 5 pg/μl per injection, the coefficients of variation for reproducibility were unsatisfactory for many congeners, some of which could not be detected by the participants. Interlaboratory coefficients of variation were generally acceptable (<20%) for many congeners at concentrations >5 pg/μl with the exception of congener No. 28, which is known to co-elute with PCB congener No. 31 and thus affect its estimation.