The trusted source for
healthcare information and
Abstract & commentary
Synopsis: The finding of normal endometrial cells on a postmenopausal Pap smear is not associated with endometrial neoplasia.
Source: Gomez-Fernandez CR, et al. Obstet Gynecol 2000;96:874-878.
The purpose of this article was to determine if the presence of normal endometrial cells on a routinely obtained Papanicolaou (Pap) smear was more common in women with endometrial neoplasia than in those without. Gomez-Fernandez and colleagues reviewed cytology material from all women diagnosed at their institution with endometrial hyperplasia or endometrial adenocarcinoma between 1990 and 1998. Two percent of these women had normal endometrial cells on their smears.
The comparison group consisted of women who had an endometrial biopsy that did not show neoplasia and who had a Pap smear within 12 months of the endometrial biopsy. Five percent of the women in this comparison group had normal endometrial cells on cytology.
Gomez-Fernandez et al concluded that the presence of normal endometrial cells in cytology material in postmenopausal women is not a reason for further evaluation.
Comment by Kenneth L. Noller, MD
The Bethesda System (TBS) recommends that cytopathologists report normal endometrial cells on cervical cytology specimens from postmenopausal women. Some past studies have suggested that finding such cells is associated with a greater likelihood of endometrial neoplasia than in women who do not have these cells present. This has led to various recommendations, the most common one being that postmenopausal women with endometrial cells on their smears have an endometrial biopsy.
This article suggests that the finding of normal endometrial cells on a postmenopausal cytology specimen means nothing. Gomez-Fernandez et al strongly suggest that women with such findings not have further evaluation. While their study generally supports this conclusion, there are some problems with the study methodology, some of which probably cannot be overcome by any type of study design.
Because Gomez-Fernandez et al wanted to have an endometrial sample from the comparison (control) group, it is not at all surprising that virtually all of the women in the comparison group had abnormal uterine bleeding of some type. Therefore, the results of this study are not generalizable to the entire female population. There are several ways to design a study that would circumvent this problem, but it would need to be much larger and would be much more costly and time consuming.
However, a little common sense applied to the results of this study makes the findings quite useful. Since only 2% of women with endometrial cancer or hyperplasia had normal endometrial cells present on their cytology specimens, it is clear that these normal cells are not a good marker for neoplastic disease. It would be nice to know what the prevalence of such cells is in the general postmenopausal population, but that information cannot be gleaned from this study. Nevertheless, even when endometrial hyperplasia is known to be present, normal endometrial cells are found on only 2% of samples. Therefore, I do agree with Gomez-Fernandez et al that we should be less aggressive about our evaluation of postmenopausal women with normal endometrial cells.
One piece of information was hidden in the article. That is, when atypical endometrial cells were present there was a high rate of endometrial adenocarcinoma or hyperplasia. Therefore, action always needs to be taken when the cells are seen.