We read with interest the article by Gideon and Zaharopoulos regarding diagnosis of cytomegalovirus (CMV) in cervical smear material. ' We wish to describe three additional cases.
The table summarizes the clinical and cytologic findings in our three patients. In two cases, there was no history of immunosuppression. Each smear contained rare, enlarged cells, either in sheets or lying singly, with one dark, round intranuclear inclusion surrounded by a clear zone or "halo" (Fig. 1). Within the cytoplasm of several of these cells, multiple granular eosinophilic inclusions were noted (Fig. 2).
Historically, CMV inclusions were detected in biopsy specimens from the endocervix of asymptomatic women.213 In 1977, Vesterinen et al. detected antibodies to CMV in 7 1 % of asymptomatic pregnant women, while no cytomegalic inclusions were noted on their corresponding cervical smears.
Reports of CMV cytopathic effect in cervical smears have appeared more frequently in recent years. 5-7 The well-recognized association of CMV with AIDS may lead to further increases in the frequency of viral inclusions in the uterine cervix. Whether or not identification of CMV in cervical material from an otherwise healthy-appearing woman should translate into a recommendation that HIV lesting be considered is not clear at this time.