Sister chromatid exchange and micronucleus frequency in human lymphocytes of 1,650 subjects in an Italian population: I. Contribution of methodological factors

The influence of several methodological factors on mean values of sister chromatid exchanges (SCEs) and micronuclei (MN) in peripheral lymphocytes of 1,650 subjects was analyzed. Donors belonged to a general healthy population living in Pisa and in two nearby small cities: Cascina and Navacchio (Ca-Na). Blood samples were collected over a period of 29 months and processed in three different laboratories of the some institute. Slides were analyzed by several scorers. Our data showed that lymphocyte proliferation indexes (PIs) and baseline mean values of SCEs were affected mainly by sampling period. This factor accounted for a percentage ranging from roughly 10% (Pisa) to 20% (Ca-Na) of total SCE variance and from roughly 10% (Pisa) to 13% (Ca-Na) of total PIs variance. A marginal effect was attributable to the different laboratories involved (maximum 3% for SCEs and 7% for PIs). The sampling period variable included many sources of variability such as culture media batches, fetal calf serum, PHA, BrdUrd, and seasonality. MN counts revealed a more marked dependence on processing laboratories. This factor accounted for a percentage of roughly 10% (Pisa and Ca-Na) of total variance, while the sampling period was marginally effective (about 1-4% of total variability). Because laboratories were equipped and supplied with the same materials and consumables and technicians were rotated constantly, the only variable ascertained was represented by the three different models of CO2 incubators used for lymphocyte culturing. When "month" and "incubator" variables were considered jointly, experimental variability accounted for 15-20% of total variance, both for PIs and mean values SCEs and MN. The variability due to slide scoring was reduced by assigning each slide to five different scorers and matching low with high scorers in each group. Present data show that when the study is performed under these controlled conditions, about 20% of total interdonor variability can be explained by experimental or seasonal factors