Cervical cytology.qxd

PRACTICE
BULLETIN
(Replaces Committee Opinion Number 152, March 1995) Cervical Cytology
Screening
This Practice Bulletin wasdeveloped by the ACOG Com- Although cervical cancer was the leading cause of cancer death in American women as recently as the 1930s, both the incidence and mortality from cervical cancer have decreased by almost one half since the early 1970s, largely as a result of widespread screening with the Pap test (1–3). However, the annual inci- dence rate has remained at approximately 8 cases per 100,000 women over the aid practitioners in makingdecisions about appropriate past few years (4). New technology for performing cervical cytology is evolving rapidly, as are recommendations for classifying and interpreting the results. The purpose of this document is to provide a review of the best available evidence on screening for cervical cancer. Specific equipment and techniques for performing cervical cytology and interpretation of the results are discussed elsewhere. cedure. Variations in practicemay be warranted based on theneeds of the individual patient, Background
Value of Cervical Cytology
Although the incidence and mortality from cervical cancer have decreasedsubstantially in the past several decades among women in the United States,cervical cancer remains the third most common gynecologic malignancy (2,5). In countries where cytologic screening is not widely available, cervicalcancer remains common. Worldwide, it is the second most common canceramong women, the third most common cause of cancer-related death, and themost common cause of mortality from gynecologic malignancy (3, 6, 7).
When cervical cytology screening programs have been introduced intocommunities, however, marked reductions in cervical cancer incidence havefollowed (7–9).
Cervical cytology screening is, in many respects, the ideal screening test (8). Cervical cancer has a defined premalignant phase of many years, whichallows repeated tests to significantly reduce the impact of individual false-neg-ative test results. Cervical cytology is inexpensive and is readily accepted among American women. In 1998, 79% of women aged • Care should be taken to avoid contaminating the 18 years and older had cervical cytology screening in the preceding 3 years (10). Treatment is effective in reducing • If testing for sexually transmitted diseases is indicat- the chance of progression to invasive disease.
ed, cell collection for cervical cytology should be Despite effective screening measures and treatment, it is estimated that 50% of the women who receive cervi- • Ideally, the entire portio of the cervix should be vis- cal cancer diagnoses each year have never had cervical cytology screening. Another 10% had not been screenedwithin the 5 years before diagnosis (11). Thus, one • Routine swabbing of the discharge from the cervix approach to reducing the incidence and mortality of may result in cytologic samples of scant cellularity cervical cancer would be to increase screening rates among women who currently are not screened or under- • In an effort to reduce air-drying artifact, the speci- men should be transferred and fixed as quickly aspossible.
Addressing Errors in Cervical Cytology
When performing cervical cytology by standard In some cases, cervical cancer is undetected despite a preparation, a single slide, combining both the endocer- recent screening test because of errors in sampling, inter- vical and ectocervical samples, or two separate slides can pretation, or follow-up. Sampling errors occur when dys- be used. The most important consideration is rapid fixa- plastic cells on the cervix are not transferred to the slide; tion. If liquid-based preparations are used, rapid immer- errors of interpretation are attributed to lack of recogni- sion in liquid media is equally important.
tion of abnormal cells in the laboratory. These two sourcesof false-negative test results are associated with 30% of New Screening and Interpretation
the new cases of cervical cancer each year (12, 13).
Devices
The problem of errors in interpretation is com- Many methods to refine and improve cervical cytology pounded by inconsistency among cytologists. When have been proposed (17). In the 1980s, new devices were results of monolayer cytology specimens were reviewed developed for enhancing the collection of exfoliated cells by quality control pathologists, only negative and low- from the cervix. These included nylon brushes for sam- grade squamous intraepithelial lesion (LSIL) readings pling the endocervix and “broom” sampling devices, had greater than 50% consistency (14). Most revised which simultaneously sample both the ectocervix and results were downgraded to lesser diagnoses. Of those endocervix. These devices have been shown to increase reported as atypical squamous cells of undetermined sig- the amount of cells captured from the transformation nificance (ASCUS), 39% were downgraded to negative zone and to increase the amount of dysplastic cells col- on further review. Of those originally interpreted as high- lected when compared with cotton-tipped applicators and grade squamous intraepithelial lesions (HSIL), 50% were wooden Ayre’s spatulas (18, 19). In 1996, the U.S. Food reinterpreted as LSIL, ASCUS, or negative.
and Drug Administration (FDA) approved the first of two currently available liquid-based thin-layer cytology Amendments (CLIA), passed in response to claims of preparations for cervical screening. In addition, automat- poor or absent quality control practices in U.S. cytology ed, computer-based technologies have been marketed laboratories, limited the number of cervical cytology that use digitally scanned images to facilitate primary tests a technician could read each day to a maximum of screening and the CLIA-mandated rescreening of cervi- 100. In addition, CLIA mandated that each laboratory cal cytology tests that have negative results.
rescreen at least 10% of the cervical cytology tests thathave negative results (15).
Cytologic Reporting
Techniques of Cervical Cytology
The nomenclature for reporting cervical cytology resultshas undergone several changes since the publication of Sampling involves collecting exfoliated cells from the the original Papanicolaou system. The Bethesda System ectocervix and endocervical canal and transferring them of reporting is the most widely used system in the United to a glass microscope slide or into a liquid transport medi- States (20). First proposed in 1988, it was revised in 1991 um for review. Patient preparation and proper provider and again in 2001 (21–23). The most important changes technique can help optimize the collection of cells: in the 2001 revised terminology are listed as follows (23): • Cells should be collected before the bimanual • Specimen adequacy—Slides are to be reported as “satisfactory” or “unsatisfactory” for interpretation.
“Satisfactory, but limited by …” is no longer report- more aggressive (26, 27). The 2001 terminology ed as a separate category under the heading “speci- subdivides atypical glandular cells by cell type, men adequacy.” The presence or absence of an ie, atypical endocervical cells, atypical endometrial endocervical or transformation zone component is cells, or atypical glandular cells not otherwise spec- described in the narrative portion of the laboratory ified. Although the subdivision of “favor neoplastic” report, as are other quality indicators, such as partly is maintained in the 2001 reporting system, favor obscuring inflammation or blood. If a slide is cate- reactive is not. In addition, because sufficient cyto- gorized as unsatisfactory, the reason should be spec- logic criteria exist to designate endocervical adeno- ified. If abnormalities are found on an otherwise carcinoma and adenocarcinoma in situ, these two unsatisfactory slide, it will, by definition, be consid- findings are reported when identified.
ered satisfactory for interpretation.
• Low-grade squamous intraepithelial lesions—As in • Negative for intraepithelial lesion or malignancy— the original terminology, the 2001 nomenclature This designation should be used for slides with no combines cytologic findings of CIN 1 (mild dyspla- cytologic evidence of neoplasia. This category sia) and those consistent with human papillomavirus includes findings previously designated as “benign (HPV) infections into the category LSIL (22, 28, 29).
cellular changes.” When specific organisms are iden- • High-grade squamous intraepithelial lesions—The tified (eg, Trichomonas vaginalis, Candida species, 2001 nomenclature maintains the category of HSIL, shift in flora suggestive of bacterial vaginosis, bacte- which combines CIN 2 and CIN 3 (moderate dys- ria consistent with Actinomyces species, and cellular plasia, severe dysplasia, and carcinoma in situ).
changes consistent with herpes simplex virus), they Although the natural history of CIN 2 lies between are reported and categorized as “negative for intraep- CIN 1 and CIN 3, the virology of CIN 2 is more like ithelial lesion or malignancy.” Other nonneoplastic CIN 3 than CIN 1 in its likelihood of representing findings, including reactive cellular changes associ- aneuploidy and monoclonal proliferation with a sin- ated with inflammation, radiation, or an intrauterine device, as well as glandular cells posthysterectomyor atrophy, also may be included in this category.
• The absence of endocervical cells or a transforma- Endometrial cells found in a woman aged 40 years tion zone component on the cervical cytology sam- or older will be listed under this category, but the ple may reflect that the transformation zone was not finding of endometrial cells will not be reported rou- well sampled. This finding is common in pregnant tinely if noted in a woman younger than 40 years.
women and in postmenopausal women in whom thetransformation zone has receded onto the canal.
• Atypical squamous cells—The epithelial abnormal- Data conflict as to whether the lack of these cells is ity ASCUS has been replaced by “atypical squamous associated with an increase in squamous intraep- cells” (ASC) with the subcategories “atypical squa- ithelial lesions. Women with this finding whose mous cells of undetermined significance” (ASC-US) recent cervical cytology test results have been nor- and “atypical squamous cells cannot exclude HSIL” mal without intervening findings of ASC-US or (ASC-H). The modifier of “favor reactive” was elim- worse may be monitored by repeat cervical cytology inated. The category ASC-H was introduced to screening in 1 year. Others, including those with include those cytologic changes suggestive of HSIL incompletely evaluated abnormal test results, in- but lacking sufficient criteria for definitive interpre- completely visualized cervix, immunocompromised tation. The literature suggests ASC-H should repre- status, and poor prior screening, should have repeat sent 5–15% of the total pool of ASC but would have cervical cytology screening within 6 months. Preg- a significantly higher predictive value for diagnosing nant women lacking endocervical cells or transfor- cervical intraepithelial neoplasia (CIN) of grades 2 mation zone component should have repeat cervical • Atypical glandular cells—This term designates cells exhibiting atypia that are of glandular rather thansquamous origin and replaces the term “atypical Natural History of Cervical Neoplasia
glandular cells of undetermined significance.” The Infection with HPV is a necessary factor in the develop- finding of atypical glandular cells on cytology is ment of cervical neoplasia; however, most HPV-infected more likely to be associated with both squamous women will not develop significant cervical abnormali- and glandular abnormalities than is ASC-US, and ties (7, 29, 31–33). The infection is easily transmitted the workup required of atypical glandular cells is during sexual intercourse. Most women, especially younger women, have an effective immune response that ly vulnerable to this infection during adolescence when clears the infection or reduces the viral load to unde- squamous metaplasia is most active. Human papillo- tectable levels in an average of 8–24 months (32, 34–36).
mavirus infections are commonly acquired by young Factors that determine which HPV infections will devel- women (34, 35), but, in most, they are cleared by the op into squamous intraepithelial lesions have been poor- immune system within 1–2 years without producing neo- ly identified. Cigarette smoking may be a co-factor, and plastic changes. The risk of neoplastic transformation a compromised immune system appears to play a role in increases in those women whose infections persist (35, 41). Most cervical squamous intraepithelial lesions do Despite decades of study, the natural history of cer- not progress to cervical cancer (29, 38, 39). The small vical intraepithelial lesions is still not completely under- proportion of women who do develop invasive squamous stood. The once widely held concept that low-grade cancer generally do so over many years, and the transi- lesions are necessary precursors to the high-grade lesions tion from CIN to cancer takes longer in younger women that, in turn, may progress to invasive cancer has been (29). Cervical cancer screening in adolescents within the questioned as the sole pathogenesis (32, 33, 37). It has first 3 years after initiation of sexual intercourse is not been observed, for example, that many women present likely to result in the identification of HSIL or cancer. In with CIN 2 or CIN 3 without prior CIN 1 lesions. Foci of addition, earlier onset of screening may increase anxiety, CIN 1 and CIN 3 with different HPV types have been morbidity, and expense from increased follow-up proce- reported in the same cervical lesion, which raises the pos- dures. Furthermore, squamous cell cervical cancer is sibility that concomitant development of different grades exceedingly rare in the first two decades of life (4).
of CIN may occur (37). A few cases of invasive cancer of Therefore, it seems reasonable to begin cervical cancer the cervix have been reported despite continuous and screening approximately 3 years after initiation of sexual intercourse, but no later than age 21 years. Recognizing Multiple longitudinal studies have attempted to doc- the time course in the progression of CIN and the unpre- ument rates of “progression” and “regression” of CIN. A dictable nature of follow-up in younger women, cyto- review of the literature since 1950 reported that 57% of logic screening may be initiated earlier at the discretion patients with CIN 1 and 32% with CIN 3 undergo spon- taneous regression (38). However, the same review reported that 1.7% of all patients with CIN of any grade What is the optimal frequency of cervical
progress to invasive cancer, ranging from 1% for CIN 1 cytology screening?
to more than 12% for CIN 3. Progression from CIN 3 toinvasive cancer has been reported in up to 36% of cases The optimal number of negative cervical cytology test (29). A review of 30 years of the literature calculated results needed to reduce the false-negative rate to a min- pooled rates of progression from LSIL and HSIL to inva- imum has not been demonstrated (3, 42). Several studies sive cancer to be 0.15% and 1.44%, respectively, over have shown that in an organized program of cervical can- 24 months (39). In that analysis, 47% of LSIL and 35% cer screening, annual cytology examinations offer little of HSIL regressed to normal during the 2-year observa- advantage over screening performed at 2- or 3-year inter- tion period. Conclusions from reviews of multiple natu- vals (43–45). These studies showed minimal difference ral history studies must be interpreted with caution. The in the acquisition of cervical cancer or HSIL at screening studies included in these reviews used varying diagnostic intervals of 1, 2, or 3 years in women who had at least criteria (biopsy or cytology or both), populations, and one prior normal screening result and who were enrolled duration of follow-up. Moreover, they did not account for in health care programs that provided and monitored cer- the poor reproducibility inherent in both cytologic and Several practical considerations must be examined before biennial or triennial screening can be adopted as anational standard. Published studies have assumed a pro- Clinical Considerations and
gram of cervical cancer screening and follow-up. In thecurrent U.S. practice climate, a woman’s care provider Recommendations
may change frequently, as employment and insurance When should screening begin?
carriers change. Consequently, the physician may beunable to determine a woman’s screening history—ie, the Cervical neoplasia develops in susceptible individuals in date of her last cervical cytology test, frequency and response to a sexually transmitted infection with a high- results of her prior tests, or prior abnormal test results risk type of HPV (28, 29, 31, 40). The cervix is especial- and their management. Patients are frequently inaccurate in recalling the timing and results of recent screening, When is it appropriate to recommend discon-
more often underestimating the time elapsed and incor- tinuing screening?
rectly recalling abnormal results as normal (46–49). Inaddition, the high false-negative rate of cytology screen- Although the rate of new-onset cervical cancer plateaus ing remains a concern, as does the relatively poor repro- at age 65 years in U.S. women in general, among certain ducibility inherent in cervical cytology (14). Performing subsets—most notably, African-American women—the multiple screening tests at regular intervals remains the incidence increases steadily across the age spectrum (2, best way to ensure existing premalignant cervical disease 7). Most new cases are seen in unscreened or infrequent- has been ruled out before extending the interval between ly screened women. It is difficult to set an upper age limit screenings. This is especially true for young women who for cervical cancer screening. Postmenopausal women have a high likelihood of acquiring a high-risk type of screened within the prior 2–3 years have been shown to have a very low risk of developing abnormal cytology(53, 54).
There is room to individualize screening frequency The American Cancer Society recommends that in a woman who is known to have a negative history and screening may be discontinued at age 70 years in low- several recent annual cervical cytology tests. The chance risk women (5). The U.S. Preventive Services Task Force such a patient will develop CIN 2 or CIN 3 is extremely has set age 65 years as the upper limit of screening (55).
low, and screening at intervals of every 2–3 years is a An older woman who is sexually active and has had mul- safe, cost-effective approach. It is important to educate tiple partners may be at lower risk for new-onset CIN patients about the nature of cervical cytology, its limita- than a younger woman because of her decreased rate of tions, and the rationale for prolonging the screening metaplasia and less accessible transformation zone; how- interval. Physicians also should inform their patients ever, she is still at some risk for acquiring HPV and CIN.
that annual gynecologic examinations are still appropri- A woman with a previous history of abnormal cytology ate even if cervical cytology is not performed at each also is at risk; women in both of these categories should continue to have routine cervical cytology examinations.
Annual cytology screening should be recommended Primary vaginal cancer represents a very small frac- for women younger than 30 years. Women aged tion of gynecologic malignancies (5). The vaginal 30 years and older who have had three consecutive cer- mucosa lacks a transformation zone. Women who have vical cytology test results that are negative for intraep- had a hysterectomy and have no history of CIN are at ithelial lesions and malignancy may be screened every very low risk of developing vaginal cancer. Cytologic 2–3 years. Certain risk factors have been associated with screening in this group has a low rate of diagnosing an CIN in observational studies; women with any of the abnormality and a very low positive predictive value. In following risk factors may require more frequent cervi- a study of 9,610 Pap tests performed among women who had a hysterectomy for benign indications an average of • Women who are infected with human immunodefi- 19 years previously, only 1.1% had cytologic abnormali- ties. Biopsies of these women showed no vaginal intra-epithelial neoplasia grade 3 or cancer (54). Continued • Women who are immunosuppressed (such as those routine vaginal cytology examinations in such women are not cost-effective and may cause anxiety and • Women who were exposed to diethylstilbestrol in overtreatment. Thus, women who have had a total hys- terectomy and have no prior history of high-grade CINmay discontinue screening.
Women infected with HIV should have cervical Women who had high-grade cervical intraepithelial cytology screening twice in the first year after diagnosis lesions before hysterectomy can develop recurrent and annually thereafter (22, 50). Women treated in the intraepithelial neoplasia or carcinoma at the vaginal cuff past for CIN 2 or CIN 3 or cancer remain at risk for per- several years postoperatively (56, 57). Women who have sistent or recurrent disease and should continue to be had a hysterectomy and have a history of CIN 2 or CIN screened annually (51, 52). Women with previously nor- 3—or in whom a negative history cannot be document- mal cervical cytology results whose most recent cervical ed—should continue to be screened annually until three cytology sample lacked endocervical cells or transfor- consecutive satisfactory negative cervical cytology mation zone components and those with partly obscur- results are obtained. Routine screening may then be dis- ing red or white blood cells should be rescreened in continued. A woman who has had three consecutive sat- isfactory negative examinations following treatment for CIN 2 or CIN 3 before she had a hysterectomy also may 26% to 103% more cases of HSIL than the conventional method (59–63). True-positive rates documented with Before considering whether a woman who has had a biopsy were improved with the use of liquid-based cytol- hysterectomy should continue regular cytology screen- ogy in some but not all studies (60–64).
ing, the provider should be sure the woman’s cervical Although liquid-based thin-layer cervical cytology cytology history is accurate. The history should confirm appears to have increased sensitivity for detecting cancer that she had benign findings at the time of hysterectomy precursor lesions over the conventional method, the and that her cervix was removed as part of the hysterec- degree to which sensitivity is increased is unknown.
tomy. However, when a woman’s past cervical cytology Equally important, the difference in specificity between and surgical history are not available to the physician, the liquid-based and conventional tests has not been deter- screening recommendations may need to be modified.
mined. Although an increase in sensitivity will permit ear-lier detection of cancer precursor lesions, any decrease in How do the various methods of cervical cytol-
specificity can result in increased cost and morbidity from ogy compare in terms of effectiveness?
false-positive diagnoses. The conventional test, althoughdisappointing in its documented sensitivity, has proved Cervical cytology is the basis of the most effective and effective in reducing cervical cancer rates where screen- cost-effective cancer screening program ever implement- ing programs exist. Liquid-based products can be effec- ed. Cervical cytology, however, is not a diagnostic test tive in population screening as well. Their reported (1). The sensitivity of cervical cytology recently has been increase in sensitivity may make them especially useful in reported to be lower than the previously estimated women who are screened infrequently. Providers select- 60–85% (29). A recent comprehensive review of the lit- ing a cervical cytology method should consider the screen- erature evaluated the accuracy of screening cervical ing history of their patient, the cost of the test, and the cytology in screened populations with a low prevalence possible effects of false-negative or false-positive results.
of cervical disease (42). For inclusion in this review, a study was required to have sufficient verification of both Is the recommended frequency of screening
negative and positive cervical cytology to calculate sen- affected by the method of screening?
sitivity and specificity. Only three studies met the inclu-sion criteria to evaluate the standard preparation for The American Cancer Society recommends that women cervical cytology at a threshold of ASCUS or worse and younger than 30 years undergo cervical cancer screening estimate its ability to diagnose CIN 1 or more severe annually if the conventional method is used or every lesions. At these thresholds, the standard preparation had 2 years if a liquid-based method is used (5). However, a sensitivity of 51% and a specificity of 98%. The authors there are very limited data to support this approach. The also calculated performance measures based on nine recommendation of biennial cytology using the liquid- studies that permitted evaluation at the cytologic thresh- based method discounts the possibility of false-negative old of LSIL. The mean sensitivity was 47%, and speci- results, a consideration with both liquid-based and con- ventional methodologies. Moreover, the increased sensi- Studies comparing the accuracy of liquid-based thin- tivity of liquid-based methods over conventional methods layer cervical cytology with the standard preparation is small with studies showing overlapping confidence have used 1 of 2 study designs. The split sample design intervals. According to FDA-required labeling, the prepares the specimen by first placing cells on a glass ThinPrep technique may be marketed as better able to slide for a standard preparation, then suspending the detect LSIL and HSIL than the conventional Pap test, and remaining cells in liquid medium for liquid-based cytol- the SurePath technique may be marketed as equivalent to ogy. This design has the potential to falsely decrease the sensitivity of the liquid-based preparation. The direct-to- When is HPV testing appropriate?
vial technique, however, prepares the entire specimen forliquid-based cytology but compares a screened popula- Although it is estimated that up to 100% of women with tion with historic controls. Although most studies have histologic CIN 2 or CIN 3 will test positive for a high- included confirmation of positive test results with col- risk type of HPV, many women harbor the virus in their poscopy and biopsy, which allows an estimate of sensi- lower genital tracts without showing cytologic or histo- tivity, few have used sufficient verification of negative logic changes (31, 32, 34, 40, 65). Currently, only one cervical cytology to determine specificity. With both product, Hybrid Capture II, is FDA-approved for testing study designs, liquid-based cytology diagnosed from for cervical HPV DNA. It assesses exfoliated cervical 36% to more than 200% more cases of LSIL and from cells for the presence of 1 or more of 13 high- and inter- mediate-risk HPV types. Although the test appears to be older. This new indication for the use of HPV DNA test- very sensitive, rare cross-reactivity with low-risk HPV ing was based on information from several large studies types and HPV types of undetermined significance has (71, 75–78). These studies demonstrated that women been reported. The clinical implications of this finding aged 30 years and older who had both negative cervical cytology test results and negative high-risk type HPV- Its utility has been well demonstrated for the pri- DNA test results were at extremely low risk of develop- mary triage of cervical cytology tests read as ASC-US ing CIN 2 or CIN 3 during the next 3–5 years. This risk (23, 67–70). In this setting, high-risk HPV DNA testing was much lower than the risk for women who had only has been shown to have a sensitivity ranging from 78% cytology and tested negative. Because the FDA approval to 96% for the detection of CIN 2 or CIN 3, with rates of for the use of HPV DNA as a primary screening modal- referral for colposcopy ranging from 31% to 56%. The ity was based on clinical study data, whether the combi- use of “reflex” HPV testing has been recommended as a nation of virus screening and cytology will perform convenient and cost-effective approach to evaluating equally well when applied to population-based screening ASC-US (68, 71, 72). The technique involves collecting a sample for high-risk HPV DNA testing at the same Any woman aged 30 years or older who receives time as cervical cytology screening and evaluating it negative test results on both cervical cytology screening only if the cytology is read as ASC-US. Reflex HPV test- and HPV DNA testing should be rescreened no more fre- ing may be done by testing from residual preservative if quently than every 3 years. The combined use of these liquid-based cytology is used or by performing a sepa- modalities has been shown to increase sensitivity but rate HPV DNA test at the same time as cervical cytology also decrease specificity and increase cost. However, it and storing it for use if ASC-US is the result.
has been estimated that the increase in screening interval High-risk HPV DNA test results would be expected will offset the cost of this new screening regimen (79).
to be positive when cervical cytology results indicate It is important to note that the FDA approval for use HSIL, so the test has little utility in this setting. Likewise, of this approach is only for the panel of high-risk HPV up to 83% of women with LSIL diagnosed by cervical types. In addition, the combination of cytology and HPV cytology have been shown to be positive for high-risk DNA screening should be restricted to women aged HPV types, thus limiting the usefulness of the test in this 30 years and older because transient HPV infections are setting as well (73). Because HPV is more prevalent in common in women younger than 30 years, and a positive younger women and the rate of CIN 2 and CIN 3 increas- test result may lead to unnecessary additional evaluation es with age, it has been suggested that HPV DNA testing and treatment. Routine testing using cytology alone might be a more selective test in older women (68).
remains an acceptable screening plan.
However, stratifying results by age demonstrated onlyminimal differences in the sensitivity of HPV DNA test-ing when used as a triage test for ASCUS results (74).
Summary of
The rate of referral to colposcopy decreased with age,however, from 68% in women younger than 29 years to Recommendations
31% for women aged 29 years and older (74).
Another clinical setting in which HPV DNA testing The following recommendations are based on
may be useful is in the secondary triage of women with good and consistent scientific evidence (Level A):
a cytologic diagnosis of ASC-US, ASC-H, or LSIL in whom colposcopy is negative or biopsy fails to reveal Annual cervical cytology screening should begin CIN. A protocol of follow-up in 1 year with HPV DNA approximately 3 years after initiation of sexual testing has been suggested as an alternative to repeat intercourse, but no later than age 21 years.
cytology in this group, with repeat colposcopy for those Women younger than 30 years should undergo When cervical cytology and HPV DNA test-
Women aged 30 years and older who have had three ing are used together, can women be
consecutive negative cervical cytology screening screened less frequently?
test results and who have no history of CIN 2 or CIN3, are not immunocompromised and are not HIV The FDA has recently approved the combination of cer- infected, and were not exposed to diethylstilbestrol vical cytology and HPV DNA testing for primary screen- in utero may extend the interval between cervical ing for cervical cancer for women aged 30 years and cytology examinations to every 2–3 years.
Evidence-based data indicate both liquid-based and References
conventional methods of cervical cytology areacceptable for screening.
1. Janicek MF, Averette HE. Cervical cancer: prevention, diagnosis, and therapeutics. CA Cancer J Clin 2001;51: Women who have undergone hysterectomy with removal of the cervix for benign indications and 2. Annual report to the nation on the status of cancer, who have no prior history of CIN 2 or CIN 3 or 1973–1999: supplemental materials. Bethesda (MD): worse may discontinue routine cytology testing.
National Cancer Institute. Available at http://seer.cancer.
gov/reportcard/supplemental.html. Retrieved January 22, The following recommendations are based on lim-
ited and inconsistent scientific evidence (Level B):
3. Sawaya GF, Brown AD, Washington AE, Garber AM.
Clinical Practice. Current approaches to cervical-cancer Women previously treated for CIN 2 or CIN 3 who screening. N Engl J Med 2001;344:1603–7. (Level III) have completed their posttreatment follow-up should 4. Reis LA, Eisner MP, Kosary CL, Hankey BF, Miller BA, be monitored annually until at least three consecutive Clegg L, et al, editors. SEER cancer statistics review, negative cytology screening results are documented.
1973–1999. Bethesda (MD): National Cancer Institute;2002. Available at http://seer.cancer.gov/csr/1973_1999/.
The use of a combination of cervical cytology and HPV DNA screening is appropriate for women aged 5. American Cancer Society. Cancer facts & figures 2003.
30 years and older. If this combination is used, Atlanta (GA): ACS; 2003. Available at http://www.cancer.
women who receive negative results on both tests o rg / d ow n l o a d s / S T T / C A F F 2 0 0 3 P W S e c u r e d . p d f .
Retrieved January 22, 2003. (Level II-3) should be rescreened no more frequently than every3 years.
6. World Health Organization. The world health report 2002: reducing risks, promoting healthy life. Geneva: Women who have undergone hysterectomy with WHO; 2002. Available at http://www.who.int/whr/2002/ removal of the cervix and have a history of CIN 2 or download/en/. Retrieved January 22, 2003. (Level II-3) CIN 3 should continue to be screened annually until 7. Herrero R. Epidemiology of cervical cancer. J Natl three consecutive negative vaginal cytology test Cancer Inst Monogr 1996;(21):1–6. (Level III) 8. Eddy DM. Screening for cervical cancer. Ann Intern Med The following recommendations are based primar-
9. van der Graaf Y, Klinkhamer PJ, Vooijs GP. Effect of pop- ily on consensus and expert opinion (Level C):
ulation screening for cancer of the uterine cervix inNijmegen, The Netherlands. Prev Med 1986;15:582–90.
Physicians should consider individualization in determining the time to begin screening, the interval 10. U.S. Department of Health and Human Services. Healthy people 2010. Vol 1–2. 2nd ed. Washington, DC: U.S.
between cervical cytology examinations, the age at Government Printing Office; 2000. (Level III) which cervical cytology testing is no longer needed, 11. Cervical cancer. NIH Consens Statement 1996;14:1–38; and the testing methodology to be used. In addition to considering risk factors for cervical cancer, the 12. Sawaya GF, Grimes DA. New technologies in cervical provider ideally should be able to determine the cytology screening: a word of caution. Obstet Gynecol patient’s past screening history and reliably monitor 13. Shingleton HM, Patrick RL, Johnston WW, Smith RA.
Evidence is inconclusive to establish an upper age The current status of the Papanicolaou smear. Ca CancerJ Clin 1995;45:305–20. (Level III) limit for cervical cancer screening. If screening isdiscontinued, risk factors should be assessed during 14. Stoler MH, Schiffman M. Interobserver reproducibility of cervical cytologic and histologic interpretations: realistic the annual examination to determine if reinitiating estimates from the ASCUS-LSIL Triage Study. JAMA Yearly testing using cytology alone remains an 15. Centers for Disease Control and Prevention. Regulations for implementing the Clinical Laboratory Improvement Amendments of 1988: a summary. MMWR Recomm Rep Regardless of the frequency of cervical cytology screening, women should be counseled that annual 16. Hild-Mosley KA, Lindblade JA, Julian TM. The manage- examinations, including pelvic examination, are still ment of cervical mucus in obtaining a Papanicolaou smear. J Low Genital Tract Dis 1997;1:1–3. (Level I) 17. Cervical cytology screening. ACOG Technology 31. Walboomers JM, Jacobs MV, Manos MM, Bosch FX, Assessment No. 2. American College of Obstetricians Kummer JA, Shah KV, et al. Human papillomavirus is a and Gynecologists. Obstet Gynecol 2002;100:1423–7. necessary cause of invasive cervical cancer worldwide. J 18. Hutchinson M, Fertitta L, Goldbaum B, Hamza M, 32. Schiffman MH, Brinton LA. The epidemiology of cervi- Vanerian S, Isenstein L. Cervex-Brush and Cytobrush.
cal carcinogenesis. Cancer 1995;76:1888–901. (Level III) Comparison of their ability to sample abnormal cells 33. Kiviat N. Natural history of cervical neoplasia: overview for cervical smears. J Reprod Med 1991;36:581–6. and update. Am J Obstet Gynecol 1996;175:1099–104.
19. Martin-Hirsch P, Lilford R, Jarvis G, Kitchener HC.
Efficacy of cervical-smear collection devices: a systemat- 34. Ho GY, Bierman R, Beardsley L, Chang CJ, Burk RD.
ic review and meta-analysis. Lancet 1999;354:1763–70.
Natural history of cervicovaginal papillomavirus infection in young women. N Engl J Med 1998;338:423–8. (Level II-2) 20. The 1988 Bethesda System for reporting cervical/vaginal cytological diagnoses. National Cancer Institute Work- 35. Moscicki AB, Shiboski S, Broering J, Powell K, Clayton L, Jay N, et al. The natural history of human papillo-mavirus infection as measured by repeated DNA testing in 21. Broder S. From the National Institutes of Health. JAMA adolescent and young women. J Pediatr 1998;132: 22. Kurman RJ, Solomon D. The Bethesda System for report- 36. Woodman CB, Collins S, Winter H, Bailey A, Ellis J, ing cervical/vaginal cytologic diagnoses: definitions, cri- Prior P, et al. Natural history of cervical human papillo- teria, and explanatory notes for terminology and specimen mavirus infection in young women: a longitudinal cohort adequacy. New York: Springer-Verlag; 1994. (Level III) study. Lancet 2001;357:1831–6. (Level II-2) 23. Solomon D, Davey D, Kurman R, Moriarty A, O’Connor 37. Park J, Sun D, Genest DR, Trivijitsilp P, Suh I, Crum CP.
D, Prey M, et al. The 2001 Bethesda System: terminology Coexistence of low and high grade squamous intraepithe- for reporting results of cervical cytology. JAMA 2002; lial lesions of the cervix: morphologic progression or mul- tiple papillomaviruses? Gynecol Oncol 1998;70:386–91.
24. Sherman ME, Tabbara SO, Scott DR, Kurman RJ, Glass AG, Manos MM, et al. “ASCUS, rule out HSIL”: cyto- 38. Ostor AG. Natural history of cervical intraepithelial neo- logic features, histologic correlates, and human papil- plasia: a critical review. Int J Gynecol Pathol 1993;12: lomavirus detection. Mod Pathol 1999;12:335–42. (Level II-2) 25. Sherman ME, Solomon D, Schiffman M. Qualification of 39. Melnikow J, Nuovo J, Willan AR, Chan BK, Howell ASCUS. A comparison of equivocal LSIL and equivocal LP. Natural history of cervical squamous intraepithe- HSIL cervical cytology in the ASCUS LSIL Triage Study.
lial lesions: a meta-analysis. Obstet Gynecol 1998;92: Am J Clin Pathol 2001;116:386–94. (Level II-2) 26. Veljovich DS, Stoler MH, Andersen WA, Covell JL, Rice 40. Bosch FX, Manos MM, Munoz N, Sherman M, Jansen LW. Atypical glandular cells of undetermined signifi- AM, Peto J, et al. Prevalence of human papillomavirus in cance: a five-year retrospective histopathologic study. Am cervical cancer: a worldwide perspective. International J Obstet Gynecol 1998;179:382–90. (Level III) Biological Study on Cervical Cancer (IBSCC) StudyGroup. J Natl Cancer Inst 1995;87:796–802. (Level II-2) 27. Kinney WK, Manos MM, Hurley LB, Ransley JE.
Where’s the high-grade cervical neoplasia? The impor- 41. Ho GY, Burk RD, Klein S, Kadish AS, Chang CJ, Palan P, tance of minimally abnormal Papanicolaou diagnoses.
et al. Persistent genital human papillomavirus infection as Obstet Gynecol 1998;91:973–6. (Level III) a risk factor for persistent cervical dysplasia. J NatlCancer Inst 1995;87:1365–71. (Level II-2) 28. Lorincz AT, Reid R, Jenson AB, Greenberg MD, Lancaster W, Kurman RJ. Human papillomavirus infec- 42. Agency for Health Care Policy and Research. Evaluation tion of the cervix: relative risk associations of 15 com- of cervical cytology. Evidence Report/Technology mon anogenital types. Obstet Gynecol 1992;79:328–37. Assessment no 5. Rockville (MD): AHCPR; 1999.
AHCPR Publication no. 99-E010. (Meta-analysis) 29. Mitchell MF, Tortolero-Luna G, Wright T, Sarkar A, 43. Screening for squamous cervical cancer: duration of low Richards-Kortum R, Hong WK, et al. Cervical human risk after negative results of cervical cytology and its papillomavirus infection and intraepithelial neoplasia: a implication for screening policies. IARC Working Group review. J Natl Cancer Inst Monogr 1996;(21):17–25.
on evaluation of cervical cancer screening programmes.
Br Med J (Clin Res Ed) 1986;293:659–64. (Level II-3) 30. Davey DD, Austin RM, Birdsong G, Buck HW, Cox JT, 44. Sawaya GF, Kerlikowske K, Lee NC, Gildengorin G, Darragh TM, et al. ASCCP patient management guide- Washington AE. Frequency of cervical smear abnormali- lines: pap test specimen adequacy and quality indicators.
ties within 3 years of normal cytology. Obstet Gyncol J Low Genital Tract Dis 2002;6:195–9. (Level III) 45. Eddy DM. The frequency of cervical cancer screening.
Papanicolaou test in screening for and follow-up of cervi- Comparison of a mathematical model with empirical data.
cal cytologic abnormalities: a systematic review. Ann 46. Boyce JG, Fruchter RG, Romanzi L, Sillman FH, Maiman 59. Diaz-Rosario LA, Kabawat SE. Performance of a fluid- M. The fallacy of the screening interval for cervical based, thin-layer Papanicolaou smear method in the clini- smears. Obstet Gynecol 1990;76:627–32. (Level II-2) cal setting of an independent laboratory and an outpatient 47. Mamoon H, Taylor R, Morrell S, Wain G, Moore H.
screening population in New England. Arch Pathol Lab between self-reported survey and registry-derived Pap test 60. Carpenter AB, Davey DD. ThinPrep Pap Test: perform- rates. Aust N Z J Public Health 2001;25:505–10. ance and biopsy follow-up in a university hospital. Cancer 48. Eaker S, Adami HO, Sparen P. Reasons women do not 61. Papillo JL, Zarka MA, St John TL. Evaluation of the attend screening for cervical cancer: a population-based ThinPrep Pap test in clinical practice. A seven-month, study in Sweden. Prev Med 2001;32:482–91. (Level II-2) 16,314-case experience in northern Vermont. Acta Cytol 49. Sawyer JA, Earp JA, Fletcher RH, Daye FF, Wynn TM.
Accuracy of women’s self-report of their last Pap smear.
62. Hartmann KE, Nanda K, Hall S, Myers E. Technologic Am J Public Health 1989;79:1036–7. (Level II-2) advances for evaluation of cervical cytology: is newer bet- 50. 1999 USPHS/IDSA guidelines for the prevention of ter? Obstet Gynecol Surv 2001;56:765–74. (Level III) opportunistic infections in persons infected with human 63. Hutchinson ML, Zahniser DJ, Sherman ME, Herrero R, immunodeficiency virus. U.S. Public Health Service Alfaro M, Bratti MC, et al. Utility of liquid-based cytol- (USPHS) and Infectious Diseases Society of America ogy for cervical carcinoma screening: results of a popula- (IDSA). MMWR Recomm Rep 1999;48(RR-10):1–59, tion-based study conducted in a region of Costa Rica with a high incidence of cervical carcinoma. Cancer 1999;87: 51. Baldauf JJ, Dreyfus M, Ritter J, Cuenin C, Tissier I, Meyer P. Cytology and colposcopy after loop electrosur- 64. Vassilakos P, Schwartz D, de Marval F, Yousfi L, Broquet gical excision: implications for follow-up. Obstet Gynecol G, Mathez-Loic F, et al. Biopsy-based comparison of liq- uid-based, thin-layer preparations to conventional Pap 52. Flannelly G, Langhan H, Jandial L, Mana E, Campbell M, smears. J Reprod Med 2000;45:11–6. (Level II-2) Kitchener H. A study of treatment failures following large 65. Herrero R, Hildesheim A, Bratti C, Sherman ME, loop excision of the transformation zone for the treatment Hutchinson M, Morales J, et al. Population-based study of of cervical intraepithelial neoplasia. Br J Ostet Gynaecol human papillomavirus infection and cervical neoplasia in rural Costa Rica. J Natl Cancer Inst 2000;92:464–74.
53. Sawaya GF, Grady D, Kerlikowske K, Valleur JL, Barnabei VM, Bass K, et al. The positive predictive value 66. Peyton CL, Schiffman M, Lorincz AT, Hunt WC, of cervical smears in previously screened postmenopausal Mielzynska I, Bratti C, et al. Comparison of PCR- and women: the Heart and Estrogen/Progestin Replacement hybrid capture-based human papillomavirus detection Study (HERS). Ann Intern Med 2000;133:942–50. systems using multiple cervical specimen collection strategies. J Clin Microbiol 1998;36:3248–54. (Level II-2) 54. Pearce KF, Haefner HK, Sarwar SF, Nolan TE. Cyto- 67. Manos MM, Kinney WK, Hurley LB, Sherman ME, pathological findings on vaginal Papanicolaou smears Shieh-Ngai J, Kurman RJ, et al. Identifying women with after hysterectomy for benign gynecologic disease. N cervical neoplasia: using human papillomavirus DNA Engl J Med 1996;335:1559–62. (Level II-3) testing for equivocal Papanicolaou results. JAMA 1999;281:1605–10. (Level II-2) 55. U.S. Preventive Services Task Force. Screening for cervi- cal cancer. Rockville (MD): Agency for Healthcare 68. Wright TC Jr, Lorincz A, Ferris DG, Richart RM, Research and Quality; 2003. Available at http://www.ahrq.
Ferenczy A, Mielzynska I, et al. Reflex human papillo- gov/clinic/3rduspstf/cervcan/cervcanwh.pdf. Retrieved mavirus deoxyribonucleic acid testing in women with abnormal Papanicolaou smears. Am J Obstet Gynecol1998;178:962–6. (Level II-2) 56. Kalogirou D, Antoniou G, Karakitsos P, Botsis D, Papadimitriou A, Giannikos L. Vaginal intraepithelial 69. Shlay JC, Dunn T, Byers T, Baron AE, Douglas JM Jr.
neoplasia (VAIN) following hysterectomy in patients Prediction of cervical intraepithelial neoplasia grade 2–3 treated for carcinoma in situ of the cervix. Eur J Gynaecal using risk assessment and human papillomavirus testing in women with atypia on Papanicolaou smears. ObstetGynecol 2000;96:410–6. (Cost-effectiveness analysis) 57. Sillman FH, Fruchter RG, Chen YS, Camilien L, Sedlis A, McTigue E. Vaginal intraepithelial neoplasia: risk factors 70. Bergeron C, Jeannel D, Poveda J, Cassonnet P, Orth G.
for persistence, recurrence, and invasion and its manage- Human papillomavirus testing in women with mild cyto- ment. Am J Obstet Gynecol 1997;176:93–9. (Level II-2) logic atypia. Obstet Gynecol 2000;95:821–7. (Level II-1) 58. Nanda K, McCrory DC, Myers ER, Bastian LA, 71. Wright TC Jr, Cox JT, Massad LS, Twiggs LB, Wilkinson Hasselblad V, Hickey JD, et al. Accuracy of the EJ. 2001 Consensus Guidelines for the management of women with cervical cytological abnormalities. JAMA2002;287:2120–9. (Level III) The MEDLINE database, the Cochrane Library, andACOG’s own internal resources and documents were used 72. Kim JJ, Wright TC, Goldie SJ. Cost-effectiveness of alter- to conduct a literature search to locate relevant articles pub- native triage strategies for atypical squamous cells of lished between January 1985 and May 2003. The search undetermined significance. JAMA 2002;287:2382–90.
was restricted to articles published in the English language.
Priority was given to articles reporting results of original 73. Human papillomavirus testing for triage of women with research, although review articles and commentaries also cytologic evidence of low-grade squamous intraepithelial were consulted. Abstracts of research presented at sympo- lesions: baseline data from a randomized trial. The sia and scientific conferences were not considered adequate Atypical Squamous Cells of Undetermined Signifi- for inclusion in this document. Guidelines published by or- cance/Low-Grade Squamous Intraepithelial Lesion Triage ganizations or institutions such as the National Institutes of Study (ALTS) Group. J Natl Cancer Inst 2000;92: Health and the American College of Obstetricians and Gy- necologists were reviewed, and additional studies were 74. Sherman ME, Schiffman M, Cox JT. Effects of age and located by reviewing bibliographies of identified articles.
human papilloma viral load on colposcopy triage: data When reliable research was not available, expert opinions from the randomized Atypical Squamous Cells of from obstetrician–gynecologists were used.
Undetermined Significance/Low-Grade Squamous Intra- Studies were reviewed and evaluated for quality according epithelial Lesion Triage Study (ALTS). J Natl Cancer Inst to the method outlined by the U.S. Preventive Services Task 75. Petry KU, Menton S, van Loenen-Frosch F, De Carvalho Evidence obtained from at least one properly de- Gomes H, Holz B, Schopp B, et al. Inclusion of HPV-test- ing in routine cervical cancer screening for women above II-1 Evidence obtained from well-designed controlled 29 years in Germany: results for 8466 patients. Br J II-2 Evidence obtained from well-designed cohort or case–control analytic studies, preferably from more 76. Belinson J, Qiao YL, Pretorius R, Zhang WH, Elson P, Li L, et al. Shanxi Province Cervical Cancer Screening II-3 Evidence obtained from multiple time series with or Study: a cross-sectional comparative trial of multiple without the intervention. Dramatic results in uncon- techniques to detect cervical neoplasia. Gynecol Oncol trolled experiments also could be regarded as this 77. Schiffman M, Herrero R, Hildescheim A, Sherman ME, Opinions of respected authorities, based on clinical Bratti M, Wacholder S, et al. HPV DNA testing in cervi- experience, descriptive studies, or reports of expert cal cancer screening: results from women in a high-risk province of Costa Rica. JAMA 2000;283:87–93. (Level I) Based on the highest level of evidence found in the data, 78. Sherman ME, Lorincz AT, Scott DR, Wacholder S, Castle recommendations are provided and graded according to the PE, Glass AG, et al. Baseline cytology, human papillo- mavirus testing, and risk for cervical neoplasia: a 10-year Level A—Recommendations are based on good and consis- cohort analysis. J Natl Cancer Inst 2003;95:46–52. Level B—Recommendations are based on limited or incon- 79. Mandelblatt JS, Lawrence WF, Womack SM, Jacobson D, Yi B, Hwang YT, et al. Benefits and costs of using HPV Level C—Recommendations are based primarily on con- testing to screen for cervical cancer. JAMA 2002;287: Copyright August 2003 by the American College of Obste-tricians and Gynecologists. All rights reserved. No part of thispublication may be reproduced, stored in a retrieval system, ortransmitted, in any form or by any means, electronic, mechan-ical, photocopying, recording, or otherwise, without prior writ-ten permission from the publisher.
Requests for authorization to make photocopies should bedirected to Copyright Clearance Center, 222 Rosewood Drive,Danvers, MA 01923, (978) 750-8400.
The American College of Obstetricians and Gynecologists
409 12th Street, SW, PO Box 96920
Washington, DC 20090-6920
Cervical cytology screening. ACOG Practice Bulletin No. 45.
American College of Obstetricians and Gynecologists. Obstet Gynecol2003;102:417–27.

Source: http://www.thinprep.com/pdfs/acog_practice_bulletin.pdf