Colposcopic screening of women with atypical Papanicolaou smears

A retrospective study was done on 575 patients with atypical Papanicolaou smears obtained during a ten-year period (1974-83) at Bellevue Hospital-New York University Medical Center. The smears were classified into three categories. The first consisted of 463 patients (81%) with inflammatory atypia specifically treated for Trichomonas, Monilia, Chlamydia, Gardnerella and atrophic vaginitis. Of them, 162 patients (35% of the group) had persistent inflammatory atypia 90 days after the beginning of therapy and were subjected to colposcopic examination. The second group consisted of 86 patients (15%) with squamous atypia. That group underwent colposcopic examination. The third group, 26 patients (4%), had endocervical atypia and underwent colposcopy and endocervical curettage. Overall, of 162 patients with persistent inflammatory atypia, 36 (22%) were found to have cervical intraepithelial neoplasia (CIN); of 86 with squamous atypia, 60 (70%) had CIN; and of 26 with endocervical atypia, 15 (58%) had CIN. Fifty-eight (28%) of all the patients with CIN had lesions of severity greater than CIN 1. It appears that all patients with persistent inflammation or squamous or endocervical atypia would benefit from colposcopic screening.     

Abnormal Papanicolaou smears after earlier smears with atypical squamous cells

From November 1984 to January 1986, 138 women were seen in our cervical clinic with atypical squamous cells in a referring cervical Papanicolaou smear. We obtained a second smear and performed a careful pelvic examination, microscopic examination of the vaginal secretions and endocervical culture for the gonococcus. The time interval between the smear showing atypical squamous cells and the one performed in our clinic ranged from 1 to 12 months. The Papanicolaou smear from 59 (50%) of the 117 women who had it repeated within four months returned abnormal as compared to 5 (24%) of the 21 repeated after four months (P = .04). Of 41 women with persistent atypical squamous cells, 10 (24%) had a clinically detectable vaginal infection, while only 3 (4%) of the 74 women with a normal repeat smear had an infection (P less than .01). These findings suggest an association between cervical cytology showing atypical squamous cells and vaginal infections. Nevertheless, this cytologic diagnosis should not be dismissed lightly since a high percentage of women with atypical squamous cells on cytology are found to have dysplasia on colposcopically directed biopsies. The most appropriate method of evaluating these patients has yet to be determined. If the clinician elects to follow these patients with serial Papanicolaou smears, the ideal interval between smears needs to be defined. Our data suggest that an interval of four to six months would be the most efficient.     

Aggressive evaluation for atypical squamous cells in Papanicolaou smears

A retrospective study was done on women who had atypical Papanicolaou smears and were referred for immediate colposcopy. The smears were obtained during January 1985 to March 1989 at Edwards Air Force Base, California. Excluded from the evaluation were abnormal Papanicolaou smears with hyperkeratosis, parakeratosis and koilocytotic atypia suggestive of human papillomavirus (HPV) infection. The evaluation included colposcopy, colposcopically directed biopsies, endocervical curettage and repeat Papanicolaou smears. A total of 101 patients were included in the study. Cervical intraepithelial neoplasia (CIN) was seen in 29.7% (30 patients): 12.9% (13) CIN I, 12.9% (13) CIN II and 3.9% (4) CIN III. Carcinoma was seen in 3.9% (4) of the patients: 2.9% (3) was carcinoma in situ, and 0.99% (1) was invasive squamous cell carcinoma, stage IIb. HPV and dysplastic lesions were seen together in 19.8% (20) of the patients. HPV was seen alone in 45% (46). Twenty-one patients (20.8%) had no apparent lesions on colposcopy, although one developed microinvasive keratinizing squamous cell carcinoma within 36 months of colposcopy. Many significant lesions can go undetected for extended periods of time in women with atypical Papanicolaou smears, resulting in delayed management. Referral for immediate colposcopy is advocated strongly.     

CO2 laser vaporization in the treatment of cervical human papillomavirus infection in women with abnormal Papanicolaou smears.

In a randomized study, we have evaluated the treatment of cervical human papillomavirus (HPV) lesions by CO2 laser vaporization. Fifty patients with abnormal Papanicolaou smears and histological evidence of cervical HPV infection associated or not with cervical intraepithelial neoplasia (CIN) grade I were randomized to either a treatment or a control group. The cervical swabs were obtained every 3 months in both groups and examined for HPV type 16 DNA by the polymerase chain reaction. After a follow-up period of 12 months no significant differences were found between the laser treatment and the control groups in relation to the disappearance of the abnormal Papanicolaou smear. Two patients in the treatment group and 3 in the control group had a conization because of development of CIN I or aggravation of the concomitant CIN found at the initial visit. The percentage of women who demonstrated HPV in their cervical smears at 12 months' follow-up was identical in the two groups, supporting the hypothesis that HPV is a persistent infection during which the virus is widespread in the vaginal epithelium.     

Comparative utility of repeat Papanicolaou smears, cervicography, and colposcopy in the evaluation of atypical Papanicolaou smears

In an attempt to establish the significance and management of the atypical Papanicolaou smear, 97 patients with atypical Papanicolaou smears were each evaluated with a repeat Papanicolaou smear, cervicography, and colposcopy. In the detection of significant lesions, cervicography was more sensitive than a repeat smear, but less so than colposcopy. Forty-two percent of the colposcopically detected lesions would have gone undetected by repeat Papanicolaou smears, compared with 11% by cervicography. However, Papanicolaou smears were more specific than cervicography (55 versus 29%). The cost per case detected using cervicography for triage was equal to that using follow-up Papanicolaou smears, but was a third higher than referring all patients directly to colposcopy. Merely using repeat smears in patients with atypical Papanicolaou smears may result in nondetection of many significant lesions, especially in populations where follow-up is poor.     

Human papillomavirus triage for young women with atypical squamous cells of undetermined significance

OBJECTIVE: Human papillomavirus testing is a cost-effective strategy for the management of atypical squamous cells of undetermined significance. Young women have a high prevalence of transient human papillomavirus infections and low incidence high-grade cervical lesions, which may limit the usefulness of human papillomavirus testing in this cohort. We sought to estimate the usefulness of human papillomavirus testing for young women with atypical squamous cells of undetermined significance. METHODS: A retrospective study of women with atypical squamous cells of undetermined significance was undertaken. Reflex human papillomavirus results and pathologic follow-up were evaluated. Age-stratified rates of human papillomavirus positivity, rates of high-grade dysplasia, and sensitivity and specificity were estimated. RESULTS: A total of 1,290 women with atypical squamous cells of undetermined significance cytology were identified. The rate of human papillomavirus positivity decreased with age from 55% for those aged 25 years or younger to 12% for women aged older than 50 years (P < .001). The cumulative rate of high-grade lesions increased with age; 12% for patients aged 25 years or younger compared with 24% for women aged older than 50 years (P = .05). A negative human papillomavirus test more effectively excluded high-grade disease in the young women. No high-grade lesions were detected in the human papillomavirus negative women aged 25 years or younger compared with 3.8% of those aged older than 50 years (P = .04). The sensitivity for detection of high-grade disease was higher in women aged younger than 25 years compared with those aged older than 50 years (100% compared with 50%), whereas specificity was lower (14% compared with 44%). CONCLUSION: Given the high prevalence of human papillomavirus and low occurrence of high-grade lesions in young women with atypical squamous cells of undetermined significance, a human papillomavirus-based triage strategy will result in the referral of a large number of women for colposcopy and may limit its cost-effectiveness. LEVEL OF EVIDENCE: III.     

Managing atypical squamous cells of undetermined significance in Papanicolaou smears

OBJECTIVE: To assess strategies using repeated conventional Pap smear and human papillomavirus (HPV) DNA testing, alone or in combination, for identifying women with concomitant cervical intraepithelial neoplasia 2 and 3 (CIN 2/3) in women with atypical squamous cells of undetermined significance (ASCUS) in their Pap smears. STUDY DESIGN: A total of 360 women cytologically diagnosed with ASCUS were referred for colposcopy and underwent a repeat Pap smear, a biopsy when necessary and HPV testing using three different modes of detection of high-oncogenic-risk HPV types: 1, first-generation Hybrid Capture test (HC-1) (Digene Diagnostics, Gaithersburg, Maryland); 2, second-generation Hybrid Capture test (HC-2); and 3, polymerase chain reaction (PCR). RESULTS: Nineteen patients (5.3%) had histologic CIN 2/3. The sensitivity and specificity of the repeat Pap smear alone for the detection of CIN 2/3 were 73.7% and 62.9%, respectively, when referring all women with a repeat Pap smear using an ASCUS-positive threshold. The proportion of women referred for colposcopy was 39.0%. When HPV testing for high risk was used for identification of women with histologic CIN 2/3, sensitivity and specificity were, respectively, 68.4% and 85.9% for HC-1, 89.5% and 73.9% for HC-2 and 89.5% and 59.0% for PCR. The rate of referral for colposcopy of these three modes of HPV testing was 16.9%, 29.4% and 44.0%, respectively. The sensitivity and specificity for identification of women with concomitant CIN 2/3 using a combination of repeat cytology showing a low grade squamous intraepithelial lesion or high grade squamous intraepithelial lesion and/or a test positive for high-oncogenic-risk HPV group were, respectively, 94.7% and 73.2% when used in combination with HC-2. The referral rate of women for colposcopy of this combined strategy was 30.4%. CONCLUSION: As compared to the strategy using abnormal repeat Pap smear alone, those using high-risk HPV testing with Hybrid Capture showed statistically significantly higher specificities and lower proportions of women with ASCUS referred for colposcopy. In particular, a promising strategy would be to refer for colposcopy only women with repeat Pap smears showing squamous intraepithelial lesion and/or those positive for high-risk HPV detected by Hybrid Capture testing.     

Human papillomaviruses in women with a history of abnormal Papanicolaou smears and in their male partners

Human papillomavirus infection of the genital tract was identified by the filter in situ hybridization test. Exfoliated cervical cells were tested separately for the prevalence of human papillomavirus 6/11 and 16/18. Human papillomavirus deoxyribonucleic acid (DNA) was identified in 70 and 92% of specimens of U.S. and West German women, respectively, who showed concurrent cytologic and colposcopic abnormalities, and in 50 and 54% of women, respectively, who showed neither cytologic nor colposcopic abnormalities at the time of examination. In the cytologic categories of condyloma, mild to moderate dysplasia (cervical intraepithelial neoplasia I/II), and severe dysplasia-carcinoma in situ (cervical intraepithelial neoplasia III), the overall DNA detection rate of human papillomavirus 6/11 and 16/18 varied between 75 and 83%; but human papillomavirus 16/18 was recovered relatively more frequently from the more severe lesions. Forty-eight West German women were monitored cytologically over a period of three to 24 months; progression to carcinoma in situ (cervical intraepithelial neoplasia III) was correlated with initial isolation of human papillomavirus 16/18. The vagina and vestibule were found to be frequent sites of human papillomavirus infection with the same virus type as in the cervix. In an investigation of male partners of 40 human papillomavirus-positive women, human papillomavirus was identified in exfoliated cells from 26; in 19 instances, the males harbored the same human papillomavirus types as their female partners.     

Human papillomavirus DNA detection and histological findings in women referred for atypical glandular cells or adenocarcinoma in situ in their Pap smears

OBJECTIVE: To evaluate the association between high-risk human papillomavirus (HPV) DNA detection and histological diagnosis in women referred for atypical glandular cells (AGC) or adenocarcinoma in situ (AIS) at Pap smear. METHODS: In this cross-sectional study, 146 women referred for AGC (124), AGC with high-grade squamous intraepithelial lesion (HSIL) (15), or AIS (7) were tested for HPV DNA using Hybrid Capture II (HC II). All women underwent colposcopic examination, and cervical biopsy was performed for 95 patients. Fifty-one women referred due to AGC with normal colposcopy and normal second Pap smear were scheduled for control visits every 4 months. RESULTS: The overall prevalence of HPV DNA was 38%. HPV DNA was detected in 93% of the women with HSIL associated with AGC and in 71% of women with AIS Pap smear, being significantly higher when compared with the prevalence (29%) in women with AGC alone. Forty-five women (30.8%) had clinically significant histological lesions (CIN 2 or worse). High-risk HPV DNA was detected in only 16% of the women without significant abnormalities in biopsy, in contrast to 96% of those who had CIN 2 or CIN 3 and 75% of women with AIS. Eighty-five percent of women with invasive cervical carcinoma (squamous or adenocarcinoma) tested positive for HPV DNA. HPV DNA detection was significantly associated with histological diagnosis of CIN 2 or worse, with an odds ratio (OR) = 51.8 (95% CI 14.3-199.9). CONCLUSION: HPV DNA detection was strongly associated with the severity of cervical lesion (CIN 2 or worse) in women referred for AGC or AIS in their Pap smear. These data implicate the use of HPV testing in triage of women with AGC Pap smears.     

Cervical chlamydia trachomatis and mycoplasmal infections in women with abnormal Papanicolaou smears

In a series of 2,346 Papanicolaou-stained smears from women undergoing routine gynaecological examination, 39 showed cytomorphological signs of inflammation suggesting Chlamydia trachomatis infection (Papanicolaou class II or III). The 39 smears were studied microbiologically by the direct-immunofluorescence test and cell culture to see whether chlamydial infection correlated with the presence of Mycoplasma hominis and Ureaplasma urealyticum. The results were compared with the cytological and colposcopic findings. C. trachomatis was cultured in 56.41% of the 39 smears, and isolated by the direct-immunofluorescence test in 51.28%. M. hominis was detected in 35.89% and U. urealyticum in 25.54%. Though all three organisms coexisted in 10.25% of the smears, C. trachomatis and M. hominis in 15.38%, C. trachomatis and U. urealyticum in 2.56%, no valid conclusions could be drawn from their association. The study did, however, indicate that vacuolated cells and cells with "bubbly" cytoplasm are common also to other infections seen in PAP-test smears and do not necessarily warrant a diagnosis of C. trachomatis, but that Gupta-type intracellular inclusion bodies do.     

Significance of atypical glandular cells of undetermined significance on ThinPrep Papanicolaou smears

OBJECTIVE: The aim of this study was to assess the incidence and risk factors predictive of significant histopathologic findings among women with atypical glandular cells of undetermined significance (AGCUS) on ThinPrep Papanicolaou smears. METHODS: ThinPrep smears with AGCUS obtained between 1997 and 1999 were reviewed. Patients' charts were reviewed and patients' characteristics, follow-up information, and colposcopy and biopsy results were recorded. Pathologic slides were reviewed. The demographic features of women with favor reactive smears were compared with those with favor neoplasia and risk factors predictive of significant histopathologic findings (high-grade squamous intraepithelial lesion, endometrial hyperplasia, and cervical or endometrial cancers) were calculated. RESULTS: The rate of diagnosis of AGCUS was 0.65%. Eighty-four patients with follow-up information were identified. The demographic features of women with smears favor neoplasia (n = 43) were similar to those with smears favor reactive (n = 41). The rates of incidence of any dysplasia or cancer and significant histopathologic findings were 32.1 and 22.6%, respectively, and were higher among women with smears favor neoplasia than among women with smears favor reactive (41.9% versus 22.0%, P = 0.051, and 34.9% versus 9.8%, P = 0.006, respectively). The subtype of cytology was the only factor that predicted significant histopathologic findings (odds ratio = 5.0, 95% confidence interval 1.6, 15.6, P < 0.010). CONCLUSIONS: In women with AGCUS on ThinPrep smears, significant histopathologic findings were found in 34.9% versus 9.8%, depending on the subtype of the smear (favor neoplasia vs reactive). Further studies are needed to validate the cytologic criteria for subtyping AGCUS smears and base management of women with AGCUS cytology on the subtype of the smear.     

Human papillomavirus in cervical smears taken from women wearing an intrauterine contraceptive device

Summary Cervical cells were screened for human papillomavirus (HPV) infections by in situ filter hybridization in 108 patients wearing intrauterine contraceptive devices. The cells were tested for the presence of HPV 6, 11, 16, 18. DNA. In 86 cases (79.6%) no reactions were obtained with³² P-labeled HPV DNA probes. 22 smears (20.3%) were positive for HPV: type 6 and 11 sequences were demonstrated in 17 cases (3 of them carried both). 4 smears hybridized with the probe for oncogenic HPV 16 and one with that for oncogenic type 18. All these five patients had normal smears and no abnormality on colposcopy.     

Endocervical glandular atypia in Papanicolaou smears.

The importance of endocervical glandular atypia in a cervicovaginal Papanicolaou smear has not been fully investigated. Between July 1988 and June 1989, 21,930 cervicovaginal smears were reviewed by the Massachusetts General Hospital Cytopathology Laboratory. One hundred smears with endocervical atypia were identified, an incidence of 0.46%. Follow-up was available on 63 cases: Seven had negative follow-up smears for at least 2 years, 15 had negative biopsies, seven had endocervical polyps, two had endometrial hyperplasia, eight had mild dysplasia, five had moderate dysplasia, six had severe dysplasia, six had squamous carcinoma in situ, five had adenocarcinoma in situ, and two had invasive adenocarcinoma. Twelve women's smears showed endocervical atypia with features suggestive of reactive atypia; three of these had dysplasia. Twenty-six (41%) of the Papanicolaou smears with endocervical atypia had coexisting squamous atypia or dysplasia. We conclude that endocervical atypia may be associated with substantial cervical disease in as many as half of cases.     

A comparison of the human papillomavirus test and Papanicolaou smear as a second screening method for women with minor cytological abnormalities

BACKGROUND: Of the estimated one million Papanicolaou (pap) smears performed annually in Sweden, about 4% show any degree of abnormality. Approximately, 1% of these cases contain moderate or severe atypia (high-grade squamous intraepithelial lesions) and the rest contain low-grade atypia. Recommendations for the management of minor abnormalities vary in various parts of Sweden. Generally, a second Pap smear is obtained 4-6 months after the first one showing low-grade atypia. The aim of this study is to compare the sensitivity of human papilloma virus (HPV)-DNA testing for the detection of cervical intraepithelial neoplasia (CIN) 2-3 with that of a second Pap smear in women, who had low-grade atypia in their first Pap smear. METHODS: Women with low-grade atypia in the Stockholm area, detected at a population-based cytology screening, were enrolled. A repeat Pap smear, HPV test, and colposcopically directed biopsies were obtained. For the detection of HPV, Hybrid Capture II (HC II) was used. RESULTS: The HPV-DNA test was positive in 66% of the 177 participating women. The sensitivity of the second Pap smear and HPV-DNA test to detect CIN 2-3 was 61 (95% CI = 45-74) and 82% (95% CI = 67-91), respectively. The positive and negative predictive values of HPV testing were 27 (95% CI = 18-35) and 89% (95% CI = 80-97), respectively. CONCLUSIONS: In Sweden, a second Pap smear is often obtained for the follow-up of women with low-grade atypia. The results of our study show that compared to the second Pap smear, HPV testing with HC II is a more sensitive method for detecting high-grade lesions.     

The isolated finding of histiocytes in Papanicolaou smears from postmenopausal women

The effectiveness of uncovering uterine disease by drawing special attention to increased histiocytes on Papanicolaou smears from postmenopausal patients was evaluated during a 22-month period. Of the 95,593 screened, 478 had increased histiocytes. Half the patients did not have specific follow-up during the study period. One-third had repeated Papanicolaou smears, which were usually unremarkable and free of increased histiocytes. Seventy-eight patients had subsequent endometrial sampling, hysterectomy or both; 41 of these samples failed to disclose pathology. Thirty-seven cases did yield abnormal tissue. However, when the clinical records of those patients were analyzed, it appeared that most had postmenopausal bleeding or an abnormal physical finding. Of five cases in which the finding of increased histiocytes apparently led to uncovering significant uterine disease, two had endometrial cells reported on the Papanicolaou smear in question. There remained only three cases in which the isolated presence of increased histiocytes led to the discovery of disease; they included one instance of adenomatous endometrial hyperplasia and two of endometrial polyp. We conclude that the isolated finding of increased histiocytes in the absence of postmenopausal bleeding or endometrial cells in a postmenopausal patient is a poor indicator of uterine disease. Such a finding, on its own, is insufficient justification for recommending a diagnostic tissue biopsy.     

Self-reported Papanicolaou smears and hysterectomies among women in the United States

OBJECTIVE: To evaluate the potential overuse of Papanicolaou smears among women who have had a hysterectomy. METHODS: We analyzed two surveys of US women aged 18 years or older, the Behavioral Risk Factor Surveillance System (1992-1997) and the National Health Interview Survey (1993-1994), and one survey of US hospitals (National Hospital Discharge Survey, 1980-1997). We examined the number of women who have had a hysterectomy who had a recent (within 3 years) Papanicolaou smear. We also examined trends in the proportions and rates of hysterectomies by diagnoses and type of procedure that potentially could require a Papanicolaou smear. RESULTS: From the Behavioral Risk Factor Surveillance System, an estimated 21.2% of US women have had a hysterectomy. Among women who have had a hysterectomy, 78.3% had a recent Papanicolaou smear. Among those reporting no hysterectomy, 82.1% had a recent Papanicolaou smear. Estimates from the National Health Interview Survey were similar. From the National Hospital Discharge Survey, an estimated 6.7% to 15.4% of women with a history of hysterectomy would require a subsequent Papanicolaou smear because they had a diagnosis related to cervical neoplasia or because they had undergone a supracervical hysterectomy. For an estimated 10.6-11.6 million of the 12.5 million women who had a hysterectomy and a recent Papanicolaou smear, that test could be considered unnecessary. CONCLUSION: Continued Papanicolaou screening of women without an intact uteri may result in excessive use of resources in time and money with minimal impact on decreasing cervical cancer.     

Incidence of cervical cancer in perimenopausal and postmenopausal women detected by Papanicolaou smears.

This paper concerns itself with the prevalence of abnormal Papanicolaou smears in women 40 years of age or older. They are considered in two groups, i.e. perimenopausal (40 through 49) and postmenopausal (50 or more). In the perimenopausal group, the incidence of abnormal smears was higher than in a central population of women of all age groups. Abnormal smears were less than half as common in the postmenopausal group. Invasive cancer was four times greater in the perimenopausal group and two times greater in the postmenopausal group.     

Management of abnormal Papanicolaou smears in pregnancy.

All pregnant patients with abnormal cervical cytologic smears were evaluated as outpatients in the colposcopy clinic. Of the 131 patients, 84% had colposcopically directed biosies. Initial biopsy diagnosis of severe dysplasia/carcinoma in situ was noted in 16% of patients with class II smears and 43% of patients with class III smears. Two patients underwent conization and 2 had modified conizations because they failed to meet our criteria for adequate colposcopic evaluation. Of the 110 patients available for postpartum follow-up, all had either the same or a lesser degree of cervical intraepithelial neoplasia. Invasive carcinoma was not detected in any patient who underwent subsequent hysterectomy or conization. No abortions or premature labors could be directly attributed to the procedure.     

Human papillomavirus testing and screening

Cervical cancer is a largely preventable disease through the detection, treatment and follow-up of its precursors. Traditionally, this has been accomplished through screening women with cervical cytology, and referring women with abnormal cytology for colposcopy, histological sampling and treatment. In organized programmes that achieve wide coverage of the target population at frequent intervals, this approach has resulted in a considerable reduction in cervical cancer. Recently, however, the development of reliable and reproducible tests for the detection of human papillomavirus (HPV) infection of the cervix (which is now accepted to be causally associated with the development of almost all cervical cancers) has led to the evaluation of HPV de-oxyribonucleic acid (DNA) testing as either an alternative or adjunctive test to cytology for the detection of cervical cancer and its precursors. There is now a large body of data supporting the clinical utility of HPV DNA testing for the prevention of cervical cancer, particularly in the settings of primary screening of women older than 30 years, in the triage of women with equivocal cytology and for the follow-up of women post-treatment.