Performance of cytology and human papillomavirus testing in relation to the menstrual cycle

Cervical smears prepared around the time of menses have been linked to unsatisfactory specimens and false negative results; however, it is unclear whether liquid-based cytology is similarly affected and data relating date of last menstrual period (LMP) to human papillomavirus (HPV) DNA testing are conflicting. Accordingly, we evaluated liquid-based cytology and HPV test results using Hybrid Capture 2 and PCR by LMP (days 0-10; 11-21; 22-28). We studied 5060 participants in ALTS, the Atypical Squamous Cells of Undetermined Significance (ASCUS) Low Grade Squamous Intraepithelial Lesion (LSIL) Triage Study. On average, women had 3.4 examinations (median 4, range 1-5) during a 2-year period of observation permitting an examination of intra-individual variation in cytology and HPV by LMP. Although uncommon, unsatisfactory cytology specimens were most likely on days 0-10. For satisfactory specimens, the frequency with which cytologic categories were reported varied by time since LMP, although differences were modest and did not affect the chance of abnormal cytology or its severity among women diagnosed with CIN2+. The frequency of positive HC2 tests did not vary with date of LMP. Among HPV infected women, independent of eventual diagnosis and the number of viral genotypes present, mid-cycle specimens yielded the highest frequency of LSIL cytologic interpretations and the highest HPV load; however, the magnitude of these effects were small. Intraindividual correlations of cytology or HPV by LMP were generally weak. We conclude that mid-cycle specimens yield slightly higher HPV DNA loads and slightly increased LSIL interpretations, but the clinical impact is marginal. Standardizing collection times would slightly improve interpretation of trends in HPV load. Finally, these data are consistent with the view that the biological properties of the HPV-infected cervix vary with the date of the LMP.