Comparison of PCB congeners and pesticide levels between serum and milk from lactating women

Samples of blood and milk were obtained from lactating women participating in the New York State Angler study. A total of seven women gave one blood a nd one milk sample at time intervals between blood and milk collection diff erent for each woman. The time between samples varied from 3 to 318 days. O ne subject provided a second milk sample 219 days after the first milk samp le, The samples were analyzed for 69 PCB congeners, DDE (a metabolite of DD T), Mirex, and hexachlorobenzene (HCB), Lipid content was determined by gra vimetric analysis. The congener profiles in serum and milk were similar for each individual but different among all subjects. The sum of the concentra tions of the congeners present above the limit of detection was used to est imate the total PCB concentration that was in the range of 2.6 to 5.8 ng/g of serum and 3.5 to 14.1 ng/g of milk. The ratio of serum to milk concentra tions varied from 0.18 to 1.66 with a mean of 0.65+/-0.49 showing no consis tency among individuals prior to adjusting the data for lipid content. The total PCB levels normalized for lipid content were 320 728 ng/g of serum li pid and 239 428 ng/g of milk lipid. The range of the lipid adjusted serum/m ilk ratio was 1.1 to 2.8 and the mean+/-SD serum/milk ratio was 1.9+/-0.5, The ranges of lipid adjusted serum concentration of DDE, HCB, and Mirex wer e 95 to 591, 8 to 48, and 3 to 29 ng/g lipid, respectively, The ranges of l ipid adjusted milk concentration of DDE, HCB, and Mirex were 90 to 577, 11 to 22, and 1 to 10 ng/g lipid, respectively. For DDE, HCB, and Mirex, the m eans of the individual lipid adjusted serum to milk ratios were 1.5+/-0.7 2 .5+/-1.5, and 5.3+/-4.6, respectively. Considerable differences were found among lipid adjusted concentrations of these environmental pollutants in se rum and milk samples from the same individual. This suggests that body burd en estimates in lactating women using different matrices may not be equival ent even when lipid adjusted values are used. (C) 1999 Academic Press.