Volume 205, Issue 1 , Pages 64.e1-64.e7, July 2011
Patients with cervical cancer: why did screening not prevent these cases?
Article Outline
Objective
The objective of the study was to assess the screening history of women with cervical cancer and review normal cervical smears 5 years preceding the diagnosis.
Study Design
Cytological and histological results of 401 women treated for invasive cervical cancer between 1991 and 2008 at the Radboud University Nijmegen Medical Center were studied. Ninety-eight normal smears were reviewed.
Results
Of the 401 women, 269 (67%) received at least 1 invitation for the national screening program for cervical cancer (NCSP). One- third fell outside the target age of the NCSP. Seventeen percent never responded to the invitation(s). Twenty-one percent had 1 or more normal smears within 5 years preceding the diagnosis. After review, only 39% of those smears were reviewed as a normal smear.
Conclusion
Half of the women with cervical cancer were never screened because of the limited target age range or nonattendance. Twenty-one percent had a normal smear within 5 years preceding the diagnosis, caused by interpretation and/or sampling errors.
Key words: cervical cancer, cervical smear, screening
A national screening program for cervical cancer (NCSP) was introduced in The Netherlands in 1988. Women between the ages of 35 and 53 years were invited for a cervical smear every 3 years. The NCSP was reorganized in 1996 to improve the effectiveness and efficiency of the screening. Therefore, the age range of the target population was extended to 30-60 years, and the screening interval increased from 3 to 5 years. From 1990 to 2006, the incidence and mortality rate for cervical cancer in The Netherlands decreased by 28% and 42%, respectively.1, 2, 3
Despite organized screening in The Netherlands, around 600-700 women are still diagnosed with cervical cancer annually.1 Factors influencing these numbers are the target age range of the screening program, the frequency of screening, and nonattendance to the screening program.4 Important factors in women diagnosed with cervical cancer despite adequate screening are problems with sampling, interpretation, and consistent clinical follow-up.5 However, a small number of rapidly developing cancers (interval cancers) cannot be ruled out.
The Papanicolaou smear has been a remarkably effective tool in cancer prevention. However, cytological smears have been associated with a significant rate of sampling and/or interpretation errors, resulting in a sensitivity of approximately 60-70%.6, 7, 8 False-negative results may have a great impact because cervical abnormalities might be missed and therefore left untreated. Smears within normal limits (WNL) preceding cervical cancer and confirmed by cytological review may be either an indication for a sampling error or suggestive for a rapidly progressive form of cervical cancer. A study of WNL smears preceding cervical cancer may show to what extent improving interventions might be possible.9
The aim of this study was to determine how many women diagnosed with cervical cancer in the Nijmegen region had been screened according to the national guidelines and, if screened properly, why screening did not prevent these cases?
Materials and Methods
Data collection
Between 1991 and 2008, 421 women were diagnosed with invasive cervical cancer and treated at the Radboud University Nijmegen Medical Center (RUNMC), Nijmegen, The Netherlands. Data concerning cervical cytology and histology were retrieved from the Dutch nationwide registry of histo- and cytopathology (PALGA).10 This database has national coverage from 1991 onward, enclosing all surgical specimens and cervical smears ever taken from each patient, both by the general practitioner and medical specialists. Twenty patients who were initially diagnosed in a foreign country or had incomplete data were excluded, leaving 401 patients opting for evaluation.
The cytological results were registered according to the diagnostic categories of Papanicolaou and/or CISOE-A (Composition, Inflammation, Squamous epithelium, Other and endometrium, Endocervical columnar epithelium, Adequacy of the smear), which is easily translatable to the Bethesda 2001 (TBS 2001) classification.11 Only cytological smears performed within the framework of the screening program were registered. Diagnostic smears leading to the diagnosis of cervical cancer, performed as a consequence of symptoms, were not included. Analysis has focused on smears within the screening program, prior to the diagnosis.
Patient and tumor characteristics (eg, stage/type) were obtained from the hospital medical records. All histological types were included in our analysis to properly evaluate the current screening program. According to the date of birth, we assessed whether women had ever been invited for the national screening program. Women within the age category of the national screening program constituted our target group for further study. They had been invited at least once.
Women who were younger or older than the target cohorts were not further analyzed with respect to their screening history because the aim of the study was to evaluate women who were eligible for cervical cancer screening.
In the collected data, women were scored as regularly screened, irregularly screened, or never screened. A patient was scored as regularly screened when screened according to the guidelines of the national screening program on cervical cancer (responded to all invitations of the national screening program, every 5 years from the age of 30 years until 60 years). A deviation in screening interval of 10% was accepted. Patients were scored as never screened if they never had a smear taken (nonresponders). Otherwise, the patients were scored as irregularly screened. This group responded to at least 1 of the invitations of the screening program but not to all, or the frequency and/or timing was not adequate (interval longer than 5 years).
All normal cervical smears within 5 years before the diagnosis of cervical cancer were requested from pathology laboratories around the country. All smears were made anonymous before sending them to 2 expert cytotechnologists (CTs) (M.R.J.S.-v.d.P. and J.E.M.V.) for review. Both CTs scrutinized the smears for abnormal cells with full knowledge of the study, using a review protocol, noting the CISOE-A classification, the quantity and nature of the cells, and the extent of certainty about their judgment.12 The aim was not to repeat a normal screening setting but to scrutinize the smears for abnormal cells to distinguish interpretation errors from sampling errors/progressive forms of cancer.
After an extensive search for abnormal cells, all cervical smears were provided with a new diagnosis, according to the 2001 Bethesda System.11 In case the judgments of the 2 CTs were discordant, they reassessed the smear together and came to concordance.
Statistical analysis
The Statistical Package for Social Sciences (SPSS version 16.0; SPSS, Chicago, IL) was used to perform the analyses. Patient characteristics and tumor characteristics were compared using nonparametrical tests (χ2 and Mann-Whitney). We considered P ≤ .05 as statistically significant.
The study was exempt from institutional review board approval because data were gathered retrospectively and subsequently made anonymous.
Results
Of the 401 eligible women, 269 (67%) had been invited at least once for participation in the national screening program (target cohorts). A number of 87 women (22%) exceeded the age limit to be invited for the NCSP (older than target cohorts), and 45 (11%) were younger than the starting age of the NCSP when diagnosed with cervical cancer (younger than target cohorts) (Figure 1).
FIGURE 1.
Women with invasive cervical cancer
Flow diagram of total 401 women with invasive cervical cancer at the RUNMC in the period 1991-2008, including reviews of the normal smears within 5 years before the diagnosis.
ASCUS, atypical squamous cells of undetermined significance; HSIL, high-grade squamous intraepithelial lesion; LSIL, low-grade squamous intraepithelial lesion; RUNMC, Radboud University Nijmegen Medical Center; WNL, within normal limits.
de Bie. Failures of cervical cancer prevention. Am J Obstet Gynecol 2011.
The youngest group showed 36% nonsquamous cell carcinoma's compared with 16% of the oldest group (P = .011). The nonsquamous cell carcinomas in the youngest group mainly consisted of adenocarcinomas. They accounted for 26.7% of the total carcinomas in the youngest group, compared with 19.7% in the target cohort and 11.5% in the oldest group. In the oldest group, 54% had advanced stage (IIB or greater) disease, compared with 18% in the youngest and 11% in the target group (P < .001). Women of the target cohort who were regularly screened had in 13.1% stage 1A disease vs 7.4% for those who were irregularly screened. This difference was not statistically significant (P = .123). Patient and tumor characteristics are shown in Table 1.
TABLE 1. Patient and tumor characteristics of women with invasive cervical cancer at the RUNMC in the period 1991-2008e
| Characteristic | Younger than target cohorts (I) (%)a | Target cohorts (II) (%)b | Older than target cohorts (III) (%)c | Total (%) | P value | |||
|---|---|---|---|---|---|---|---|---|
| Regularly screened | Irregularly screened | I vs III | I vs II | II vs III | ||||
| Median age, y (range) | 29 (21–35) | 37(30–60) | 43(30–60) | 72(54–84) | 42(21–84) | |||
| Histological type | ||||||||
| Squamous cell carcinoma | 29(64.4) | 74(69.2) | 134(82.7) | 73(83.9) | 310(77.3) | .011 | .063 | .190 |
| Nonsquamous cell carcinoma | 16(35.6) | 33(30.8) | 28(17.3) | 14(16.1) | 91(22.7) | |||
| Adenocarcinoma | 12(26.7) | 28(26.2) | 25(15.4) | 10(11.5) | 75(18.8) | |||
| Adenosquamous carcinoma | 3(6.7) | 3(2.8) | 3(1.9) | 1(1.1) | 10(2.5) | |||
| Other | 1(2.2) | 2(1.8) | 0 | 3(3.4) | 6(1.4) | |||
| FIGO stage | ||||||||
| Stage IA | 1(2.2) | 14(13.1) | 12(7.4) | 1(1.1) | 28(7.0) | .001d | .624d | <.001d |
| Stage IB, IIA | 36(80.0) | 88(82.2) | 125(77.2) | 46(52.9) | 295(73.5) | |||
| Advanced stage IIB or greater | 8(17.8) | 5(4.7) | 25(15.4) | 40(46.0) | 78(19.5) | |||
| Total | 45(11.2) | 107(26.7) | 162(40.4) | 87(21.7) | 401(100.0) | |||
aYounger than the age limit to be invited for the screening program; |
bTarget age of the screening program, currently 30-60 years; |
cOlder than the age limit to be invited for the screening program; |
dStage IIA or less vs stage IIB or greater. |
Of the 269 women of the target cohorts, 107 (40%) were screened according to the guidelines of NCSP (cytological smear every 5 years from the age of 30 years to 60 years). Of the remaining invited patients, 68 (25%) were never screened (nonresponders) and 94 (35%) were irregularly screened (Figure 1). Of the 107 regularly screened women, 36 were diagnosed with cervical cancer when they responded to their first invitation for cervical screening.
Of the 201 women who ever participated in the NCSP, 85 (42%) had at least 1 normal cervical smear within the 5 years before the diagnosis of cervical cancer and therefore within the screening interval. Because some women had more than 1 normal smear within the interval, these 85 women had a total of 136 normal smears preceding the diagnosis.
Of these smears, 105 (77%) were retrieved from 17 different pathology laboratories. The other 31 smears were not retrieved because 1 pathology laboratory denied collaboration because the smears were older than the restricted storage period of 10 years and therefore destroyed, or they were not traceable. Seven AutoCyte PREP smears (BD Diagnostics/Tripath, Inc, Burlington, VT) were excluded because the selected CTs lacked experience with this method and the selection of a third CT would not contribute to the equality of the study. Of the remaining 98 smears, 87 (89%) were conventional smears and 11 (11%) were liquid-based smears (Thin Prep; Hologic, Inc, Marlborough, MA). In total, 38 (39%) smears were considered to be normal after review and 23 (24%) were judged as unsatisfactory for evaluation.
Reasons for unsatisfactory smears were few epithelial cells, too many leucocytes, and too much blood or a combination of these. These smears were initially incorrectly classified as WNL instead of repeated after 6 weeks. In total, 18 smears (18%) were reviewed as atypical squamous cells of undetermined significance (ASC-US) or low-grade squamous intraepithelial lesion (LSIL). These women would otherwise have been advised to have a repeat smear in 6 months. Nineteen smears were reviewed as high-grade squamous intraepithelial lesion (HSIL) or more severe and should therefore have been referred to the gynecologist immediately. For the review results, see TABLE 2, TABLE 3.
TABLE 2. Review of WNL cervical smears within 5 years before the diagnosis of cervical cancer
| Review | Liquid based, n (%) | Conventional, n (%) | Total, n (%) |
|---|---|---|---|
| Unsatisfactory smear | 2(18) | 21(24) | 23(24) |
| WNL smear | 7(64) | 31(36) | 38(39) |
| ASC-US/LSIL smear | 2(18) | 16(18) | 18(18) |
| HSIL smear or greater | 0 | 19(22) | 19(19) |
| Total | 11(100) | 87(100) | 98(100) |
TABLE 3. Unsatisfactory smears previously diagnosed as WNL smears
| Reasons for unsatisfactory smears | Unsatisfactory smears, n |
|---|---|
| A (too much blood) | 1 |
| B (too many leucocytes) | 6 |
| C (few epithelial cells) | 8 |
| D (badly fixated) | 1 |
| A plus B | 1 |
| A plus C | 2 |
| B plus C | 3 |
| A plus B plus C | 1 |
| Total | 23 |
When the total 401 women with cervical cancer were considered, 33% were diagnosed outside the target age range of the NCSP, and 17% never responded to the invitation of the NCSP. These women were never screened. Another 17% were screened, but the last smear before the diagnosis of cervical cancer exceeded the screening interval of 5 years. Thirteen percent were incorrectly diagnosed with a normal smear within 5 years preceding the diagnosis of cervical cancer. Eight percent had a normal smear (confirmed by review), suggesting a sampling error or a rapidly progressive tumor. The remaining 12% were either diagnosed when they attended their first screening round (9%) or developed cervical cancer despite follow-up of abnormal smears by cytology and/or colposcopy (3%) (Figure 2). It was not possible to trace where appropriate follow-up, evaluation (eg, colposcopy) or treatment (eg, loop electrosurgical excision procedure, cone biopsy, or cryotherapy) went wrong. It was estimated that these follow-up smears were either misinterpreted or the colposcopic evaluation was insufficient.
FIGURE 2.
Women with cervical cancer, subdivided into categories
Diagram of the total 401 women with cervical cancer, subdivided into categories of explanations why screening did not prevent these cases.
de Bie. Failures of cervical cancer prevention. Am J Obstet Gynecol 2011.
Comment
This relatively large study, from a centrally organized program, with a pathology database that has full national coverage makes these data useful alongside other studies.13, 14, 15 This study shows that 33% of women with cervical cancer are diagnosed outside the target age range of the NCSP. Additionally, 17% of the patients within the target age range never responded to any of the invitations. Therefore, half of the women were never screened before the diagnosis of cervical cancer. An additional 9% were diagnosed after responding to the invitation of their first screening round. Seventeen percent were screened at least once but had their last smear longer than 5 years before the diagnosis of cervical cancer. Twenty-one percent had a normal smear preceding the diagnosis of cervical cancer, suggesting an interpretation or sampling error or a combination of those.
Of the 401 women included in this study, 45 (11%) and 87 (22%) women were, respectively, younger or older than the target cohorts to be invited for cervical cancer screening. Another 9%, diagnosed after their first invitation, were due to too-late initiation of screening. This is in accordance with the study of Bos et al,16 in which they describe that in The Netherlands around 30% of the women with cervical cancer never had been invited for cervical cancer screening because of their age. However, the proportion of women older than the target cohort will be reduced in the future because this cohort will vanish in time.
In The Netherlands, the age range (30–60 years) is rather limited as compared with other countries in which some screening programs start inviting at the age of 20 years and continue until the age of 65 years.17 Extending the age range may lead to better prevention of cervical cancer and to earlier diagnoses of earlier stage of disease. However, despite extending the age range, there is still a risk of missing adenocarcinomas in the youngest group because the sensitivity of cervical cytology for the detection of adenocarcinomas is significantly lower than that for squamous carcinomas.18, 19, 20, 21
Despite the NCSP with 5 year invitations, 25% of the invited patients never responded to any of the invitations before they were diagnosed with cervical cancer. Another 35% of the women responded to at least 1 of the invitations but not to all or not in time. This stresses the fact that, in accordance with previous studies,16, 22, 23, 24, 25, 26, 27 underscreening is the most important modifiable risk factor for cervical cancer. Nonattendance has been reported previously as being due to fear, embarrassment, a male doctor, and having a too-busy schedule.28 Efforts should be undertaken to increase the coverage of screening.29
Consistent with previous studies, we found a substantial number of women with a WNL smear preceding the diagnosis cervical cancer.23, 26 The fact that cervical cancer can occur despite a history of regular screening suggests the need for further studies of primary screening tests.23 According to the reviews of the WNL smears in this study, 38% indicated interpretation errors. Another 39% of the smears was confirmed WNL by review and suggestive for sampling errors or a fast carcinogenesis (rapidly progressive cancer) after the smear was taken.
Nonsquamous cancers might be those rapidly progressive cancers and are often located in the endocervix. These nonsquamous cancers are therefore not easily detected by cytological screening. The percentage of nonsquamous cancers was slightly higher (34%) in the group of women with a WNL smear within 5 years preceding the diagnosis of cervical cancer, compared with the total group (22.7%).
Remarkably, one-fourth of the reviewed smears were unsatisfactory for evaluation. The cervical lesion might have been detected if these smears would have been repeated after 6 weeks, according to the guidelines. The majority (89%) of these unsatisfactory smears were conventional smears. This might suggest that conventional smears are more likely to be unsatisfactory than liquid-based cytology, as previously reported.30, 31 However, this suggestion completely depends on the distribution of conventional smears vs liquid-based smears in the screening population. During the study period, the use of solely conventional cytology slowly shifted toward a combination of conventional and liquid-based methods.
The sensitivity of cytology for the detection of cervical lesions may be increased by improving the technical standards for smears and their reading.5, 32, 33, 34, 35 Recently it has been shown that improving the sensitivity of the initial screening step by using an additional high-risk human papillomavirus (hrHPV) test might further reduce the proportion of false-negative results.7, 36, 37 The nonsquamous cervical cancers in this study, which were mainly adenocarcinomas, are detectable as well by hrHPV testing.
Additionally, cervicovaginal self-samples appear to be as reliable as physician-taken samples.38, 39 Screening by self-samples performed in the privacy of women's own homes might provide a better attendance than screening by samples taken by a physician or other health care providers.40, 41 A self-sampling method for hrHPV performed by the woman herself might be a possible way for current nonresponders to lower the threshold to attend and increase the coverage of screening.40, 42
Although hrHPV testing has a higher sensitivity than cytological screening to detect cervical abnormalities, it has a lower specificity. It is important to search for a balance between harms and benefits. Triage strategies for women who are hrHPV positive, for example, by repeat hrHPV testing, subsequent cytology, or the use of biomarkers, increases specificity and prevents overtreatment.
However, concerning this study, we have to consider that the NCSP started in 1988 and women diagnosed just after the beginning of the national program could not have optimally benefited from screening. After all, the sensitivity of the screening program is higher due to a repetition of a moderate sensitive test. Furthermore, the CTs reviewed all smears with full knowledge of the study. The purpose of this study was to scrutinize the smears for abnormal cells to distinguish interpretation errors, sampling errors, or progressive forms of cancer. The aim was not to repeat a normal screening setting. Therefore, the sensitivity of the detection of abnormal smears in this study is not completely comparable with that in a screening setting.12
A limitation of this study is that the study population was drawn from a single medical center, although the RUNMC renders referral medical services for 1,500,000 people and therefore practically reflects the general population. A long inclusion period is required to achieve a large cohort in a single center. However, this study does provide us the opportunity to detect where improvements are achievable in this chain of events and which part of the cervical cancer incidence is attributable to limited detection of (pre-)malignancies of cervical cancer.
Another limitation is that women with cervical cancer stage IA are not necessarily referred to the RUNMC and often treated at regional hospitals, as a result of which our study population contains relatively fewer stage IA cervical cancers. Additionally, the RUNMC is a national referral center for radical trachelectomy for small stage IBI cervical cancers. A maximum of 30 women were referred with this indication during the study period.
In conclusion, new cases of cervical cancer occur because of the following: (1) nonattendance to the NCSP, (2) a narrow target age range of 30-60 years, and (3) limited sensitivity of the screening test caused by sampling and/or interpretation errors. Therefore, the target age range of the current NCSP in The Netherlands may need to be reevaluated, and shortening the screening interval might decrease the percentage of WNL smears before the diagnosis of cervical cancer.
The incidence and mortality rate from cervical cancer can be reduced further by interventions to increase the participation rate of the NCSP and by improving the sensitivity of the screening test. Using additional hrHPV testing and/or hrHPV self-sampling for nonresponders are possibilities to reach this goal.41
References
- . http://www.rivm.nlAccessed Jan. 10, 2010
- . Cervical cancer mortality in young women in Europe: patterns and trends. Eur J Cancer. 2000;36:2266–2271
- . Mass screening programmes and trends in cervical cancer in Finland and The Netherlands. Int J Cancer. 2008;122:1854–1858
- . Clinical practice (Current approaches to cervical-cancer screening). N Engl J Med. 2001;344:1603–1607
- Accuracy of the Papanicolaou test in screening for and follow-up of cervical cytologic abnormalities: a systematic review. Ann Intern Med. 2000;132:810–819
- . Long-term protective effect of high-risk human papillomavirus testing in population-based cervical screening. Br J Cancer. 2005;92:1800–1802
- Human papillomavirus testing in primary screening for the detection of high-grade cervical lesions: a study of 7932 women. Br J Cancer. 2001;84:1616–1623
- . Meta-analysis of Pap test accuracy. Am J Epidemiol. 1995;141:680–689
- . Invasive cervical cancer audit: a relative increase in interval cancers while coverage increased and incidence declined. BJOG. 2009;116:845–853
- Pathology databanking and biobanking in The Netherlands, a central role for PALGA, the nationwide histopathology and cytopathology data network and archive. Cell Oncol. 2007;29:19–24
- The 2001 Bethesda System: terminology for reporting results of cervical cytology. JAMA. 2002;287:2114–2119
- . False negatives in focused rescreening of Papanicolaou smears: how frequently are ‘abnormal' cells detected in retrospective review of smears preceding cancer or high-grade intraepithelial neoplasia?. Arch Pathol Lab Med. 1997;121:273–276
- Screening-preventable cervical cancer risks: evidence from a nationwide audit in Sweden. J Natl Cancer Inst. 2008;100:622–629
- Invasive cervical cancer audit: why cancers developed in a high-risk population with an organised screening programme. BJOG. 2010;117:736–745
- Pathways to diagnosis of cervical cancer: screening history, delay in follow up, and smear reading. BJOG. 2007;114:398–407
- . Nonattendance is still the main limitation for the effectiveness of screening for cervical cancer in the Netherlands. Int J Cancer. 2006;119:2372–2375
- Overview of important cervical cancer screening process values in European Union (EU) countries, and tentative predictions of the corresponding effectiveness and cost-effectiveness. Eur J Cancer. 2000;36:2177–2188
- . Papanicolaou smear sensitivity for the detection of adenocarcinoma of the cervix: a study of 49 cases. Cancer. 2001;93:8–15
- . Adenocarcinoma in situ of the cervix. Cancer. 2002;96:330–337
- . The rising incidence of adenocarcinoma relative to squamous cell carcinoma of the uterine cervix in the United States—a 24-year population-based study. Gynecol Oncol. 2000;78:97–105
- Significant decrease of adenocarcinoma in situ not reflected in cervical adenocarcinoma incidence in The Netherlands, 1989-2003. Br J Cancer. 2008;98:165–167
- Screening history of women with cervical cancer: a 6-year study in Aarhus, Denmark. Br J Cancer. 2008;98:1292–1294
- The screening histories of women with invasive cervical cancer in Connecticut. Am J Public Health. 1995;85:791–794
- . The screening history of women with cervical cancer in the Rotterdam area. Eur J Epidemiol. 2000;16:641–645
- . Review of the screening history of Alberta women with invasive cervical cancer. CMAJ. 1997;157:513–519
- . Papanicolaou smear history and diagnosis of invasive cervical carcinoma among members of a large prepaid health plan. Cancer. 2000;88:2283–2289
- . Differences in screening history, tumour characteristics and survival between women with screen-detected versus not screen-detected cervical cancer in the east of The Netherlands, 1992-2001. Eur J Obstet Gynecol Reprod Biol. 2008;139:204–209
- Uptake of cervical cancer screening in The Netherlands is mainly influenced by women's beliefs about the screening and by the inviting organization. Eur J Public Health. 2007;17:178–185
- . Effects of HPV detection in population-based screening programmes for cervical cancer; a Dutch moment. Gynecol Oncol. 2006;100:451–454
- Accuracy of liquid based versus conventional cytology: overall results of new technologies for cervical cancer screening: randomised controlled trial. BMJ. 2007;335:28
- . Cytologic detection of cervical abnormalities using liquid-based compared with conventional cytology: a randomized controlled trial. Obstet Gynecol. 2008;112:1327–1334
- Effect of study design and quality on unsatisfactory rates, cytology classifications, and accuracy in liquid-based versus conventional cervical cytology: a systematic review. Lancet. 2006;367:122–132
- . Can we change the way we screen? (The ThinPrep Imaging System). Cancer. 2004;102:340–344
- . Liquid-based cervical cytology. Cancer. 2003;99:263–271
- . Comparison of conventional Papanicolaou smears and a fluid-based, thin-layer system for cervical cancer screening. Obstet Gynecol. 1997;90:278–284
- POBASCAM, a population-based randomized controlled trial for implementation of high-risk HPV testing in cervical screening: design, methods and baseline data of 44,102 women. Int J Cancer. 2004;110:94–101
- Management of women who test positive for high-risk types of human papillomavirus: the HART study. Lancet. 2003;362:1871–1876
- Diagnostic accuracy of self collected vaginal specimens for human papillomavirus compared to clinician collected human papillomavirus specimens: a meta-analysis. Sex Transm Infect. 2005;81:207–212
- . Are self-collected samples comparable to physician-collected cervical specimens for human papillomavirus DNA testing? (A systematic review and meta-analysis). Gynecol Oncol. 2007;105:530–535
- Human papillomavirus testing on self-sampled cervicovaginal brushes: an effective alternative to protect nonresponders in cervical screening programs. Int J Cancer. 2007;120:1505–1510
- HPV testing on self collected cervicovaginal lavage specimens as screening method for women who do not attend cervical screening: cohort study. BMJ. 2010;340:c1040
- Primary screening for high risk HPV by home obtained cervicovaginal lavage is an alternative screening tool for unscreened women. J Clin Pathol. 2002;55:435–439
Cite this article as: de Bie RP, Vergers-Spooren HC, Massuger LFAG, et al. Patients with cervical cancer: why did screening not prevent these cases? Am J Obstet Gynecol 2011;205:64.e1-7.
PII: S0002-9378(11)00236-5
doi:10.1016/j.ajog.2011.02.046
© 2011 Mosby, Inc. All rights reserved.
Volume 205, Issue 1 , Pages 64.e1-64.e7, July 2011
