Are annual Pap smears in the over-50s valuable? --- HealthandAge

December 1, 2008


	RSS

Women's Health Center


	[ Health Centers > Women's Health > RELATED ARTICLE ]

Are annual Pap smears in the over-50s valuable?

Summarized by Robert W. Griffith, MD
January 22, 2001 (Reviewed: February 18, 2003)

Introduction

There is no disputing the benefits of cervical smears for detecting early cancer of the cervix and uterus; indeed, the Pap smear is established as an important regular screening test for women in their reproductive years. However, women are postmenopausal for a third of their lives, and some people have recommended discontinuing screening from age 65 onwards, in view of the relatively low incidence of important cervical disease over 50. The likelihood that a positive test result is actually a "false-positive" increases as the actual disease incidence decreases, and it may be that Pap smears in older women are indeed unnecessary. A recent study has attempted to answer this question.

Method

Cervical smears were collected prospectively as part of the Heart and Estrogen/progestin Replacement Study (HERS). This was a randomized, double-blind, placebo-controlled comparison of an oral estrogen/progestin combination with placebo. 2,763 women under 80 were assigned to receive the hormones or placebo for an average of 4.1 years, with cervical smears performed annually.

Women were identified among the collective who had a normal cervical smear at entry with an abnormal smear at year one or year two. Abnormal was defined as having one or more of the following four characteristics:

atypical squamous cells of undetermined significance (ASCUS)
atypical glandular cells of undetermined significance (AGCUS)
low-grade squamous intraepithelial lesion
high-grade squamous intraepithelial lesion

Follow-up was done on all women with normal baseline smears and abnormal smears at year one or two, using data from exams done under the study protocol and results of questionnaires. Results of any additional diagnostic tests or procedures were obtained and evaluated.
The final histological status of these women was determined as either:

normal (atrophy, cervicitis or inflammation permitted)
low-grade histological condition (e.g. grade I cervical intraepithelial neoplasia)
high-grade histological condition (e.g. grades II to III cervical or vaginal intraepithelial neoplasia)
unknown status

Additional information collected covered the known risk factors for cervical neoplasia (current smoking, parity, age at first intercourse, recent sexual activity, and sexually transmitted diseases).

Results

A total of 2,561 women with normal baseline smears were registered. Their mean age was 66.7 years. After one year, 78 of these women (3.0%) had an abnormal smear. An additional 32 of 2,346 women still in the study at two years (1.4%) had an abnormal smear at two years. Thus the incidence for new cytological abnormalities in the two years following a normal smear was 110 per 4,895 person-years (2.25%).

Follow-up information was gathered on the 110 women with abnormalities in the first two years. Two were lost to further study. The additional procedures or interventions done were repeat Pap smears (112 instances), colposcopy (33), cervical or vaginal biopsy (30), endocervical curettage (35), endometrial biopsy (8), D&C (4), loop excision (7), and cone biopsy (2).

Final diagnoses were obtained for all these subjects except for the two lost to study and 5 others - three with ASCUS and two with low-grade squamous intraepithelial lesions, all 5 of whom had at least one normal smear subsequently.

Most of the final diagnoses were normal (94 of 110 women, 85.5%). Low-grade histological changes were seen in 6 women, and high-grade changes in two - one had cervical, the other vaginal, intraepithelial neoplasia grade I to II. A further subject had endometrial hyperplasia without atypia.

As none of the 78 women with abnormal smears at one year were abnormal on final diagnosis, the positive predictive value for any cervical smear abnormality in the first year following a normal smear was a 0%. Only one of the 110 women with abnormal smears at years one or two had a final diagnosis of an important cervical lesion, so that the positive predictive value for any cervical smear abnormality in the two years following a normal smear was 0.9%. The incidence of an important cervical histological finding within two years of a normal smear was therefore one per 4,895 person-years (0.02%).

Women in the hormone treatment group had a non-significant increased incidence of cytological abnormality over the two-year period, largely because of an increase in ASCUS. There was no statistically significantly increased incidence of low- or high-grade lesions at final diagnosis in the hormone-treated subjects.

Comment

This study shows that important histologic cervical lesions are uncommon in previously screened postmenopausal women. As many as 2.3% of women within two years of a normal smear had an abnormality on repeat smears; however, all but one of these (0.02%) were false-positives. Thus women having smears at one or two yours after a normal smear were many times more likely to have a false-positive than a true-positive smear.

The implications are obvious. Over-screening is costly, but, more important, false-positives can lead to symptoms of anxiety, low self-esteem and even depression. In this study, 231 additional interventions were needed to detect one cervical intraepithelial grade I to II neoplastic lesion.

The authors conclude that these results can be generalized to postmenopausal women with an intact cervix who have had a recent normal cervical smear. Such women do not require repeat Pap smears for at least two years. This obviously does not apply to postmenopausal women who have never been screened, or have not been not screened for at least two years. If women request a Pap smear within that period, they should be made aware of the likelihood of a false-positive result rather than receiving a clinically important finding.

Source

The positive predictive value of cervical smears in previously screened postmenopausal women: The Heart and Estrogen/progestin Replacement Study (HERS). GF. Sawaya, D. Grady, K. Kerlikowske, Ann Internal Med, 2000, vol. 133, pp. 942--950