Study finds women screened for cervical cancer one, two and three years after normal test results ha

By Leslie Harris

Women who undergo Pap smear tests to screen for cervical cancer one, two or
three years after having had normal test results had about the same low rate of
significant cellular abnormalities, a University of California, San Francisco
study has found. The finding suggests some women may not need to be screened
annually, as is common practice among many U.S. physicians.

“We were interested in determining optimal screening strategies in women with
recent normal Pap smears.  We set out to determine if screening outcomes
differed in woman who returned one, two or even three years later after a
normal smear,” said George Sawaya, MD, lead author and UCSF assistant professor
of obstetrics, gynecology and reproductive sciences. “Our assumption was that
the longer you wait, the more likely you are to have an important abnormality.
We actually didn’t find that. We found that it didn’t make a difference if
women came back for a Pap smear one, two or three years after they had a normal
smear. That was against conventional wisdom.”

The study results are published in the August issue of Obstetrics & Gynecology.

Over the last 50 years, the incidence of cervical cancer has declined by more
than 70 percent in the U.S. The widespread availability of Pap smears—used to
examine the cervix for pre invasive cells that can turn cancerous if untreated
(called dysplasia)—helped to bring about the drop in cervical cancer cases.
About 50 million smears are done in the U.S. each year and most are normal,
according to the study.

Published guidelines suggest women with a history of normal smears may be
screened less often than annually. However, it is unclear when the next smear
should be performed, according to the study. Other guidelines suggest women may
be screened every three years.

The study included 124,805 women at community based clinics throughout the U.S.
screened for cervical cancer within three years of a normal smear. The women
were examined through the National Breast and Cervical Cancer Early Detection
program, sponsored by the Centers for Disease Control.

Incidence of cellular abnormalities were defined as: atypical squamous cells of
undetermined significance (ASCUS), low grade squamous intraepithelial lesions
(SIL), high grade SIL and a category called suggestive of squamous cell cancer.
Atypical squamous cells of undetermined significance and low grade
intraepithelial lesions usually do not indicate pre cancerous conditions,
Sawaya said. The other two classifications are more serious signs of possible

The study found that clinically- important, abnormal Pap smear results are
uncommon in women who wait up to three years to be screened for cervical cancer
following a normal Pap smear. Over a three year period following a normal
smear, the incidence of new smears interpreted as high grade SIL or worse was
66 out of 10,000 women who were 30 years old and younger; 22 out of 10,000 for
women between the ages of 30 to 49; 15 out of 10,000 for those ages 50 to 64;
and 10 out of 10,000 for women who were 65 years old or older.

Age adjusted incidence rates of high grade SIL or worse were similar for women
screened 9-12 months (25/10,000); 13-24 months (29/10,000); and 25-36 months
(33/10,000) after a normal smear.

“The bottom line is when we are making decisions on the clinical level, we need
to inform patients better about the true benefits and risks of screening,”
Sawaya said. “If some women are advised by their physicians to extend screening
intervals beyond annual, these women should realize the risk of having a bad
outcome is very, very small.  In fact, less screening of low-risk women may be
better since the chance of false-positive testing in these women may be much
higher than the chance of having significant cervical disease. We often do not
inform patients of these benefits and risks since they have not been well
quantified in the past.”

Age adjusted incidence rates of ASCUS, the most common abnormality, did not
change. But the incidence of smears interpreted as low grade SIL rose as time
from the normal smear increased. These low grade abnormal smears comprised more
than 95 percent of abnormalities following a normal smear, according to the
study. The results are often clinically insignificant, but can prompt further
tests and procedures. Such actions can cause unfounded patient anxiety,
according to the study.

“There are some risks associated with screening in terms of false-positive
results,” Sawaya said. “In preventive medicine, we want to enhance the health
of the healthy, and this is no mean feat. We need to be careful that we don’t
turn a healthy person into a patient because of false-positive test results.”

The highest rate of abnormalities occurred in the group of women 30 years old
and younger. As age advanced, the incidence of all types of abnormalities
decreased. For example, 1,000 women aged 65 years or older had to be screened
to find one case of a highly abnormal Pap smear result after a normal smear,
Sawaya said.
“We can’t draw any absolute conclusions from this,” he said. “But it gives us
some critical pieces of the puzzle to help us determine optimal screening
intervals following a normal result and identifying an age at which we may be
able to recommend that screening end, given previously normal test results.”

Cervical cancer accounts for about 15,700 new cancer cases in the United States
each year and 4,900 deaths, according to the National Institutes of Health. One
of the main causes of cervical cancer is the human papillomavirus (HPV), a
common sexually transmitted infection. But most women with HPV do not develop
cervical cancer, Sawaya said. Half of the women who get cervical cancer in the
United States have never had a Pap smear, he said. 

“Rescreening women with a recent normal smear addresses but a small part of the
problem of cervical cancer in the U.S.,” Sawaya said.

The other study authors are Karla Kerlikowske, MD, UCSF assistant professor in
residence, general internal medicine at the San Francisco Veteran’s Affairs
Medical Center; Nancy C. Lee, MD, from the Centers of Disease Control and
Prevention; Ginny Gildengorin, PhD, UCSF statistician, department of
obstetrics, gynecology and reproductive sciences; and A. Eugene Washington, MD,
Msc, UCSF chair and professor of obstetrics, gynecology and reproductive

The Agency for Health Care Policy and Research funded the study.