The cytologic diagnosis of adenocarcinoma in situ of the cervix uteri and related lesions. II. Microinvasive adenocarcinoma

Criteria for the cytologic diagnosis of microinvasive adenocarcinoma of the cervix have not been previously established. Such cytologic criteria were evolved through the detailed analysis of cervical smears from 40 histologically confirmed cases. The cellular features of cervical adenocarcinoma in situ (AIS) were always associated with microinvasion. Syncytia of glandular cells, small cells in very crowded sheets and papillary groupings of cells, when seen in conjunction with AIS, were suggestive of microinvasion. Dissociation of cells was common. Nuclear pleomorphism with an irregular chromatin pattern and inconspicuous-to-prominent nucleoli was frequently present. In some cases, a tumor diathesis was seen in the smear background. Using these criteria, our predictive accuracy for diagnosing microinvasive cervical adenocarcinoma is improving steadily and now approaches 50%. Ongoing investigation of these cases must include a diagnostic come biopsy to further improve the predictive accuracy for this lesion.     

Atypical glandular cells of undetermined significance on cervical smears. A study with cytohistologic correlation.

OBJECTIVE: The incidence of endocervical adenocarcinoma has increased steadily over the past two decades. Since the Bethesda System was introduced, the diagnosis of atypical glandular cells of undetermined significance (AGUS) has also risen and now accounts for 0.46-1.83% of all cervical (Pap) smears. The purpose of this study was to evaluate the significance of a diagnosis of AGUS using cytohistologic correlation. STUDY DESIGN: A retrospective review of archival material from 1993 through 1996 identified 64 patients who had smears diagnosed as AGUS and had a subsequent surgical biopsy. The smears were reviewed and cytologic features analyzed and correlated with the histologic diagnosis. RESULTS: On biopsy, 3 (5%) of the 64 cases showed endocervical adenocarcinoma in situ (AIS) (1 case with invasive adenocarcinoma also), 14 (22%) had a benign glandular lesion (endocervical polyp, tubal metaplasia, microglandular hyperplasia, reactive changes), 35 (54%) had squamous intraepithelial lesion (SIL) (15 diagnosed on the original smear), and 12 (19%) had no abnormality. Among the cytologic criteria evaluated, feathering (P = .01), palisading (P < .001) and chromatin clearing (P = .002) were shown to have a significant association with the histopathologic diagnosis of AIS/adenocarcinoma. These features were also useful in distinguishing AIS/adenocarcinoma from SIL and benign glandular changes from AIS/adenocarcinoma but not benign/reactive glandular changes from SIL. CONCLUSION: A diagnosis of AGUS correlated with a clinically significant lesion in the majority of cases. Squamous dysplasia (SIL) was the most common lesion identified. The presence of feathering, nuclear palisading and chromatin clearing increased the likelihood of a histologic diagnosis of AIS/adenocarcinoma.     

Comparative cytologic features of adenocarcinoma in situ of the uterine cervix

The cytologic findings in 30 cases of adenocarcinoma in situ (AIS) and related lesions of the cervix were compared with those in 13 cases of cervical invasive adenocarcinoma and 8 cases of cervical nonneoplastic conditions that mimicked AIS cytologically. Although there was considerable overlap, the presence of large cells with irregular nuclei and uneven chromatin distribution in smears containing no normal endocervical cells helped to distinguish invasive adenocarcinoma from AIS. The presence of "feathering," rosettes, mitotic figures and very crowded nuclei with scant cytoplasm and without cilia helped distinguish AIS from benign conditions.     

Atypical glandular cells of undetermined significance in cervical smears after conization. Cytologic features differentiating them from adenocarcinoma in situ

OBJECTIVE: To evaluate the cytologic features of atypical glandular cells of undetermined significance (AGUS) smears following conization through a comparison with adenocarcinoma in situ (AIS) smears. STUDY DESIGN: Fifty cervical smears, diagnosed as AGUS based on groups of crowded glandular cells that raised the possibility of AIS, from 38 patients who had conization and 24 AIS smears, histologically confirmed, from 17 patients were reviewed. Subsequent follow-up biopsies or hysterectomies in 38 patients were evaluated. RESULTS: Nuclear atypia was a more reliable feature than architectural structure in differentiating postcone effect from AIS on cytology. The predominant cytologic features of the postcone smears were crowded glandular cells with a high nuclear/cytoplasmic ratio and relatively small, hyperchromatic nuclei with rather finely granular and uniformly dispersed chromatin, less distinct nuclear membranes, less frequent mitosis and presence of endometrial-type stromal cells in the background. The architecture of the crowded cells in the postcone smears was sometimes similar to that of AIS. No AIS or high grade squamous intraepithelial lesion histology was encountered in follow-up biopsies or hysterectomy specimens. CONCLUSION: The cytologic features distinguishing AGUS from AIS may be helpful in identifying the postcone effect. Since it is important to avoid miscalling the postcone effect as AIS, it is recommended that one check for a previous history of a cone biopsy.     

Androgen insensitivity syndrome (AIS): emotional reactions of parents and adult patients to the clinical diagnosis of AIS and its confirmation by androgen receptor gene mutation analysis

The emotional reactions of parents and adult patients on disclosure of the clinical diagnosis of androgen insensitivity syndrome (AIS) and its later confirmation by gene mutation analysis were assessed. A semistructured interview and three questionnaires were used. Parents came from 18 different families with a total of 20 children (15 complete AIS, 5 partial AIS), 19 raised as girls, 1 as a boy. Ten adult women with complete AIS came from six families. The short-term reaction upon the clinical diagnosis was in the majority of both parents and adult patients associated with shock, grief, anger, and shame and in the mothers and adult patients with guilt. Emotional reactions were more long-lasting in mothers and adult patients than in fathers. The confirmation by DNA analysis did not alter the actual feelings of both parents. Adolescents with AIS should be informed completely - but in a step-by-step way - about their condition, since adult patients indicated that they had suffered from being not at all or misinformed about AIS in their adolescence.     

Morphology of adenocarcinoma in situ and microinvasive adenocarcinoma of the uterine cervix. A cytologic and ultrastructural study

Adenocarcinoma in situ (AIS) and microinvasive adenocarcinoma of the uterine cervix and normal endocervical columnar epithelium were studied by cytology, morphometry and electron microscopy to identify differentiating features and to ascertain the cellular origin of cervical adenocarcinoma. Smears from AIS showed the characteristic cytology, consisting of glandular rosettes, palisading and crowded sheets; most nuclei had a relatively uniform oval shape. Smears from microinvasive adenocarcinoma showed more crowded sheets, with enlarged, round and irregular-shaped nuclei and prominent oval nucleoli. These nuclear features were confirmed by the morphometric results. Ultrastructurally, reserve cells in the normal tissues contained tonofibers and secretory granules and showed squamous and adenomatous features. The ultrastructural features of microinvasive adenocarcinoma were similar to those of well-differentiated invasive adenocarcinoma. The cells from both contained tonofibers and secretory granules. These findings suggested that the reserve cell is the cell of origin for cervical adenocarcinoma.     

Interlaboratory reproducibility of atypical glandular cells of undetermined significance: a national survey

OBJECTIVE: The aim of this study was to evaluate the inter-laboratory reproducibility for atypical glandular cells (AGC) (The Bethesda System (TBS) 2001) of the laboratories involved in the screening programmes in Italy. METHODS: A set of 35 selected slides were circulated among 167 laboratories involved in local population-based cervical screening programmes. Each laboratory provided one single diagnosis per smear. The smears were read blind to the original diagnosis and to the diagnoses provided by other laboratories. A 'majority' diagnosis was defined for each case and assumed as the reference standard. The diagnosis provided from each laboratory was compared with the majority diagnosis. RESULTS: According to the majority report the 35 slides in the set were classified as negative in nine cases, AGC in eight, adenocarcinoma in eight, and squamous lesion or squamous + glandular lesion in 10. The crude agreement between all pairs of laboratories was 49.43%. K-values were 0.46, 0.21, 0.34, 0.36 and 0.32 for negative, AGC/AIS (adenocarcinoma in situ of endocervix), AdenoCa, Sq/Sq + Gl and all reporting categories respectively. Concordance according to overall K was moderate to substantial in 77% of the participating laboratories. CONCLUSIONS: The present study shows that the AGC category is not easily reproducible. The data confirmed the importance, in a screening scenario, of AGC/AIS diagnoses, but also presented difficulties in differentiating between the two diagnoses. In addition to the results obtained from the circulation of the slides, laboratories which had annually a low number of cervical smears were able to gain experience focused on particular morphological pictures.     

Increased diagnostic accuracy of atypical glandular cells in cervical liquid‐based cytology using cell blocks

E. K. J. Risse, J. P. Holierhoek, E. M. Meijer‐Marres, E. Ouwerkerk‐Noordam and M. E. Boon Increased diagnostic accuracy of atypical glandular cells in cervical liquid‐based cytology using cell blocks Objective: The purpose of this study was to reduce the number of diagnoses of atypical glandular cells (AGC). Residual material from the cervical ThinPrep® samples (Hologic, Marlboruogh, MA, USA) was used for cell blocks (CB) and immunohistochemistry (IHC). Methods: In 2007 there were 87 patients (0.12% of tests) with AGC on liquid‐based cytology (LBC) in the Leiden Cytology and Pathology Laboratory (LCPL) using the Bethesda System 2001 (TBS). CB with IHC was used for 26 of these cases. The vials still containing the brush (Cervex‐Brush^® Combi) were placed in a shaker for 10 minutes to dislodge the material trapped between the bristles. The residual sampling fluid was used to prepare paraffin sections (Shandon Cytoblock^®) stained with Papanicolaou and immunostaining. Results: Four of five cases with AGC not otherwise specified (NOS) were diagnosed with CB/IHC as benign mimics (endometrium, tubal metaplasia, follicular cervicitis, microglandular hyperplasia) and one of four with AGC‐favour neoplasia (FN) (endocervical polyp). In one of five cases with AGC‐NOS and in two of seven with AGC‐FN, CIN3 was found on subsequent histological biopsy. Of six cases diagnosed as adenocarcinoma in situ (AIS) on LBC with CB/IHC the diagnosis was confirmed in four; one was adenocarcinoma and one glandular atypia. Of eight cases diagnosed as adenocarcinoma on cytology and CB/IHC, the diagnosis was confirmed in three. The other five cases were found to be one each of AIS, squamous cell carcinoma, CIN3, CIN2 with glandular atypia, and cervical endometriosis. Conclusions: By reducing the number of benign mimics of AGC, we achieved a high proportion (16/26; 61.5%) of neoplastic or preneoplastic lesions (glandular or squamous) on histological outcome potentially avoiding colposcopy. Histological biopsy verification by the gynaecologist is needed for final diagnosis of AGC‐FN, AIS and adenocarcinoma.     

Microglandular hyperplasia of the uterine cervix. Cytologic diagnosis in cervical smears

OBJECTIVE: To identify cytomorphologic features specific to microglandular hyperplasia (MGH) in cervical cytologic smears. STUDY DESIGN: Twenty-four cervical smears from 24 patients obtained before the histologic diagnosis of MGH made on colposcopically directed biopsies during a period of two years (1995-1997) were evaluated. RESULTS: Of cases with MGH, 13/24 (54%) showed the presence of bidimensional or tridimensional cellular clusters made up of cubic or cylindrical glandular cells with vacuolated cytoplasm; cells with dense cytoplasm, basaloid in appearance, corresponding to immature squamous metaplasia; and subcylindrical reserve cells with small, round nuclei and scant cytoplasm. Clusters showed microlumina or fenestrated spaces, preserved polarity and absence of nuclear peripheral dispersion. In the control group these cellular clusters were found in 6/100 (6%). Statistical analysis (chi 2) showed a strong, highly significant association (P < 0.001) of the cytologic parameters selected and the histologic diagnosis of MGH. CONCLUSION: Until now, no specific cytologic parameters were described for MGH. This study underscored the value of cytomorphologic parameters described for typical cellular clusters showing microlumina or fenestrated spaces with shared party walls and an admixture of glandular cells, and immature squamous metaplastic and subcylindrical reserve cells in the cytologic diagnosis of MGH.     

Utility of imprint cytology for early presumptive diagnosis in clinically suspicious cervical cancer

OBJECTIVE: To determine the utility of imprint cytology (IC) in providing an early presumptive diagnosis of clinically suspected cervical carcinoma. STUDY DESIGN: A total of 219 clinically suspicious cervical cancer cases underwent Pap test, punch biopsy and IC at the same sitting. Correlations were performed between these diagnostic modalities to determine the sensitivity and specificity of IC in diagnosis of cervical cancer. RESULTS: The overall accuracy of IC in detecting cervical cancers was 96.2%. About 78% of squamous cell carcinomas (SCC), 60% of adenocarcinomas and 100% of small cell carcinoma could be accurately typed on imprints. Twelve malignant lesions were diagnosed on IC among 26 unsatisfactory biopsies. Although there was no false positive result, 3.5% false negative diagnoses were given on IC. The sensitivity and specificity of imprint smear cytology to detect malignancy was 96.2% and 100%. Agreement between imprint cytology and Pap smear diagnosis of malignancy was 95.3%. kappa Statistics revealed excellent agreement between imprints and biopsies and between imprints and Pap smears in diagnosis of malignant lesions. CONCLUSION: IC can be used as an adjunctive technique for an early and reliable preliminary presumptive diagnosis of cancer of the uterine cervix.     

Molecular analysis of the androgen receptor gene in 52 patients with complete or partial androgen insensitivity syndrome: a collaborative study.

In patients with androgen insensitivity syndrome (AIS), RFLP study of the androgen receptor gene made it possible to analyze whether deletions or mutations could be responsible for abnormalities in androgen responsiveness. We studied RFLPs of DNA from 25 46,XY patients with partial AIS (PAIS), defined as a concentration of androgen receptor in genital-skin fibroblasts less than 340 fmol/mg DNA, and DNA from 27 46,XY patients with complete AIS (CAIS) with no detectable androgen receptor site. DNA samples were digested with BamHI, EcoRI, HindIII and TaqI restriction enzymes and hybridized with three cDNA probes covering the three domains of the androgen receptor. When we had the maternal and an unaffected brother's DNA, we analyzed the two androgen receptor gene polymorphisms described, the HindIII and the exon 1 CAG repeat polymorphisms, in order to distinguish the two maternal X chromosomes, and to detect carriers of AIS. We did not find any large deletion among the 52 patients. We observed a heterozygous mother in 3 of 14 families studied with the HindIII polymorphism, and in 12 of 25 families using the exon 1 CAG repeat polymorphism. This study suggests that in AIS, abnormalities in androgen receptor response could be related to point mutations or microdeletions rather than to gross structural alterations of the androgen receptor gene. Furthermore, unless the point mutation has been described, exon 1 and HindIII polymorphism studies would enable the identification of carriers in 50% of families, and the prenatal diagnosis of AIS.     

Cervical cytology screening history of women diagnosed with adenocarcinoma in situ of the cervix: a case-control study

OBJECTIVE: To determine whether women diagnosed with adenocarcinoma in situ (AIS) of the cervix are as well screened as healthy control women. STUDY DESIGN: A case-control study was performed (307 cases, 1,228 controls) within a statewide registry. Cases consisted of women diagnosed with AIS on histology between 1995 and 2001. Screening histories were compiled from registry records. Variables of interest included number of previous negative Pap smears, time interval since last negative Pap smear, median time between previous negative Pap smears, proportion of negative Pap smears with an endocervical component and history of cervical abnormality. RESULTS: Conditional logistic regression showed that cases and controls did not differ significantly in the number of previous negative smears or in the median time between previous negative smears. Decreasing time since last negative smear was protective against AIS, with controls more likely to have had a recent negative smear. There was no difference in the proportion of previous negative smears with an endocervical component between cases and controls; however, some complex temporal relationships were observed in the regression analysis. CONCLUSION: Women who are diagnosed with AIS have a screening history similar to that of healthy, control women. The findings are consistent with the concept that AIS is predominantly a screening-detected disease.     

Distribution of human papillomavirus types in different histological subtypes of cervical adenocarcinoma

Little information is available regarding distribution of HPV types in different histological subtypes of adenocarcinoma (AC). Thus, in this study we examined the frequency of high-risk (hr) HPV types in AC, adenocarcinoma in situ (AIS) and adenosquamous carcinoma (ADSQ). A total of 102 cases of primary cervical adenocarcinoma (26 AIS and 76 invasive AC) obtained from pathology files from 1995-2006 were histologically subtyped. Our results demonstrated that endocervical type occupied the major subtype of AC (22/66) followed by ADSQ (17/66) where as in the group of AIS endocervical type (12/23) was followed by intestinal type of AIS (7/23). Successful DNA extraction was obtained in 89 samples; 81 out of 89 (91.0%) tested positive for HPV DNA. The prevalence of HPV DNA in AIS, AC and ADSQ was 91.3% (21/23), 90.9% (60/66) and 94.1% (16/17), respectively. We found HPV 18 type to be the most predominant type in AIS (11/21) and AC (17/60) followed by HPVof undeternmined type in AIS (3/21) and HPV 16 in AC (9/60) as the sole viral type. HPV 18 was most frequently detected type in all histological subtypes of AIS and AC. We have detected HPV DNA in all 5 samples of clear cell carcinoma (CCC), although other studies have reported a highly variable prevalence of HPVDNA in CCC. The most prevalent HPV type in ADSQ was HPV-16 followed by HPV 33 as single type. The observed overall predominance of HPV 18 in AIS (chi(2) = 6.109, p< or = 0.025) and AC (chi(2) = 8.927, p< or =0.01) as well as of HPV 16 in ADSQ (chi(2) = 10.164, p < or = 0.01) was statistically significant. Our data revealed statistically significant predominance of single hrHPV infections in AIS (16/21; chi(2) = 11.523, p < 0.001) and AC (37/60; X2 = 6.533, p < 0.025) whereas multiple hrHPV infections were more abundant in AC comparing to AIS (23/81and 5/81, respectively; chi(2) = 13.989, p< or =0.001).     

Cell block as an adjunct to conventional Papanicolaou smear for diagnosis of cervical cancer in resource-limited settings

OBJECTIVE: To assess the utility of indigenously prepared cell blocks (CBs) as an adjunct to a conventional smear test in providing a reliable diagnosis of clinically suspicious cervical cancer in resource-limited settings. METHODS: Eighty-six clinically suspicious cervical cancer cases underwent a conventional smear test, CB preparation from residual cellular samples and biopsies at the same sitting. Correlations were performed between these modalities in order to derive the sensitivity and specificity of the CB technique to diagnose cervical cancer. OBSERVATION & RESULTS: Out of 86 clinically suspicious cervical cancers, 72 (83.7%), 70 (81.4%) and 67 (77.9%) cases were diagnosed as malignant on tissue biopsies, CBs and smears respectively. CB-biopsy agreement in the diagnosis of malignancy was feasible in 87.5% of the cases while CB-Pap smear agreement was feasible in 92.5% of the cases. Sensitivity and specificity of CB preparation to diagnose malignancy was 92.5% and 100%, respectively, when the smear was taken as the reference test (excluding the unsatisfactory smears). When biopsy was taken as the gold standard, the sensitivity and specificity of CBs were 87.5% and 100% respectively (excluding the unsatisfactory biopsies). In 8/19 cases where the smear diagnoses were either unsatisfactory or atypical squamous cells/atypical glandular cells, CBs picked up malignant lesions. CONCLUSION: CBs prepared from the residual cellular sample of conventional cervical scrapes augment the sensitivity of the smear test. When used as an adjunct to the smear, CBs aid in providing a reliable diagnosis of cervical cancer in the majority of the clinically suspected cases and thus the biopsy load can be reduced significantly in resource-poor settings.     

Endocervical glandular cell abnormalities in conventional cervical smears: evaluation of the performance of cytomorphological criteria and HPV testing in predicting neoplasia

Objective:  To analyse the correlation between cytomorphological criteria in smears with atypical glandular cells (AGC) or adenocarcinoma in situ (AIS) and human papillomavirus (HPV) reflex test results with different neoplastic histological diagnoses, particularly to distinguish between glandular and squamous neoplasia.
Methods:  A series of 155 women with glandular abnormalities in their conventional cervical smears was included: 106 with AGC, 35 with AGC associated with high-grade squamous intraepithelial lesion (HSIL) and 14 with AIS. Two reviewers evaluated 35 cytomorphological criteria and hybrid capture II (HCII) was performed in all cases. Colposcopy was carried out in all cases and biopsy in 126/155. For statistical purposes, predictive values and odds ratio (OR) were calculated, followed by chi-square automatic interaction detection.
Results:  Histology detected 56 cases of squamous and 17 of glandular intraepithelial or invasive neoplasia. Predictive values of the papillary groups and feathering criteria for glandular neoplasia were, respectively, 80.0% and 73.3%. Feathering was the criterion with the highest OR for distinguishing glandular from squamous neoplasia and also for distinguishing between glandular and non-neoplastic diagnosis. Rosettes and pseudostratified strips did not perform as well. Multivariant Classification and Regression Trees analysis identified feathering as the best criterion for distinguishing between glandular, squamous and non-neoplastic diagnoses regardless of HPV status.
Conclusions:  Feathering was the best criterion for predicting glandular neoplasia.     

电子阴道镜在诊断妊娠期阴道出血的价值

探讨应用电子阴道镜在寻找妊娠期阴道出血原因方面的价值。方法:对我院2004年5月-2006年11月64例非产科原因孕期阴道流血的孕妇,进行电子阴道镜检查及颈管分泌物培养和镜下定位活检。结果:阴道镜下发现宫颈赘生物12例,其中2例坏死,10例明显充血水肿;宫颈炎伴活动性出血2例,宫颈炎无活动性出血18例;正常宫颈16例;宫颈上皮内瘤样病变图象15例;宫颈分泌物培养阳性23例。结论:妊娠期不规则阴道出血原因复杂,阴道镜检查能较快排除与查明宫颈及阴道病变所导致的出血。     

PCR diagnosis of HPV in cervical biopsies of CIN and invasive neoplasia formerly diagnosed as HPV negative

OBJECTIVE: To evaluate whether the polymerase chain reaction (PCR) enhances the diagnosis of human papillomavirus (HPV) in biopsies of the uterine cervix with cervical intraepithelial neoplasia (CIN) or invasive neoplasia. STUDY DESIGN: Samples of 71 paraffin-embedded cervical tissue blocks from patients seen in the period 1997-1998 were analyzed. Samples were selected according to age (18-60 years old) and an active sexual life and divided in to 3 groups: test (samples with CIN or invasive neoplasia and a negative HPV diagnosis), positive controls (samples with CIN or invasive neoplasia and a positive HPV diagnosis) and negative controls (samples without CIN or invasive neoplasia and a negative HPV diagnosis). Samples were subjected to DNA extraction and PCR for HPV detection. RESULTS: PCR analysis matched the colposcopic and cytopathologic diagnoses in the positive and negative controls. However, 77% of samples in test group were HPV positive. CONCLUSION: CIN, an invasive neoplasm, is associated with the presence of HPV. Colposcopy and cytopathology are efficient but not sufficient to identify HPV. Thus, despite the high cost, PCR can be used as an additional examination, in women with cervical lesions.     

Observer variation in histopathological diagnosis and grading of cervical intraepithelial neoplasia.

To assess the variability among histopathologists in diagnosing and grading cervical intraepithelial neoplasia eight experienced histopathologists based at different hospitals examined the same set of 100 consecutive colposcopic cervical biopsy specimens and assigned them into one of six diagnostic categories. These were normal squamous epithelium, non-neoplastic squamous proliferations, cervical intraepithelial neoplasia grades I, II, and III, and other. The histopathologists were given currently accepted criteria for diagnosing and grading cervical intraepithelial neoplasia and asked to mark their degree of confidence about their decision on a visual linear analogue scale provided. The degree of agreement between the histopathologists was characterised by kappa statistics, which showed an overall poor agreement (unweighted kappa 0.358). Agreement between observers was excellent for invasive lesions, moderately good for cervical intraepithelial neoplasia grade III, and poor for cervical intraepithelial neoplasia grades I and II (unweighted kappa 0.832, 0.496, 0.172, and 0.175, respectively); the kappa value for all grades of cervical intraepithelial neoplasia taken together was 0.660. The most important source of disagreement lay in the distinction of reactive squamous proliferations from cervical intraepithelial neoplasia grade I. The histopathologists were confident in diagnosing cervical intraepithelial neoplasia grade III and invasive carcinoma (other) but not as confident in diagnosing cervical intraepithelial neoplasia grades I and II and glandular atypia (other). Experienced histopathologists show considerable interobserver variability in grading cervical intraepithelial neoplasia and more importantly in distinguishing between reactive squamous proliferations and cervical intraepithelial neoplasia grade I. It is suggested that the three grade division of cervical intraepithelial neoplasia should be abandoned and a borderline category introduced that entails follow up without treatment.     

Contextual analysis and intermediate cell markers enhance high-resolution cell image analysis for automated cervical smear diagnosis.

Until now, efforts to automate cervical smear diagnosis have focused on analyzing features of individual cells. In a complex specimen such as that obtained from a cervical scrape, diagnostically significant cells may not be adequately represented or may elude detection by the automated technology. An approach is needed that extracts additional quantitative information from cervical smears beyond what the cell-by-cell approach can provide. A new methodology, contextual analysis, was developed to extract global quantitative information about cells, cell clusters, and background debris. This pilot study was designed to compare the efficacy of contextual analysis with high-resolution, single cell analysis and the analysis of intermediate cell markers. Thirty-four samples prepared as monolayers and stained with the Feulgen-Thionin/Congo Red stain were measured. Contextual analysis alone was able to classify 91% of the smears correctly; single cell analysis classified 94% of the cells correctly; and the intermediate cell analysis correctly identified the smear diagnosis for 84% of the cells. When all three analysis methods were combined into a simple smear level classifier, the overall smear classification accuracy was improved over those obtained using the three methodologies alone.