Variation in the human lymphocyte sister chromatid exchange frequency as a function of time: results of daily and...

The variation in lymphocyte sister chromatid exchange (SCE) frequency was investigated in healthy nonsmokers who were not taking any medication. Two separate studies were undertaken. In the first, blood was drawn from four women twice a week for 8 weeks. These donors recorded the onset and termination of menstruation and times of illness. In the second study, blood was obtained from two women and two men for 5 consecutive days on two separate occasions initiated 14 days apart. One donor participated in both studies. Analysis of the mean SCE frequencies in each study indicated that significant temporal variation occurred in each donor, and that more variation occurred in the longer study. Some of the variation was found to be associated with the menstrual cycle. Peaks in the SCE frequency were observed at times that corresponded approximately to ovulation and to the beginning and end of menstruation. In the daily study, most of the variation appeared to be random, but occasional day-to-day changes occurred that were greater than those expected by chance. To determine how well a single SCE sample estimated the pooled mean for each donor in each study, we calculated the number of samples that encompassed that donor's pooled mean within 1 or more standard errors. For both studies, about 75% of the samples encompassed the pooled mean within 2 standard errors. An analysis of high-frequency cells (HFCs) was also undertaken. The results for each study indicate that the proportion of HFCs, compared with the use of Fisher's Exact test, is significantly more constant than the means, which were compared by using the t-test. These results coupled with our previous work suggest that HFC analysis may be the method of choice when analyzing data from human population studies.