CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / Objetivos: avaliar o desempenho da citologia oncÃtica convencional e da captura de hÃbridos II na detecÃÃo de lesÃes cervicais neoplÃsicas e prÃ-neoplÃsicas. Sujeitos e MÃtodos: foram recrutadas aleatoriamente 1685 mulheres, da demanda espontÃnea de postos de saÃde da rede pÃblica, em cinco municÃpios do estado do CearÃ. As pacientes, apÃs assinarem termo de consentimento, responderam a um questionÃrio prÃ-elaborado e, a seguir, foram submetidas à coleta de material para CO, CH II e à realizaÃÃo de colposcopia, que, sendo positiva, levou à imediata biÃpsia dirigida das Ãreas anormais. Os dados foram digitados no programa Microsoft Excel 2000 e analisados no SPSS-for Windows, versÃo 10.0. O desempenho da CO e CH II foram calculados atravÃs da sensibilidade, especificidade, dos valores preditivos positivo e negativo e dos respectivos intervalos de confianÃa de 95%. Considerou-se, para anÃlise, como padrÃo ouro negativo, o resultado da colposcopia negativo ou resultado negativo no exame histopatolÃgico e, como padrÃo ouro positivo, o resultado positivo do histopatolÃgico. Avaliaram-se dois pontos de corte distintos: qualquer achado prÃ-neoplÃsico e neoplÃsico do colo uterino e achados de lesÃes intra-epiteliais de alto grau ou cÃncer. Resultados: 56 mulheres (3,4%) apresentaram atipias celulares na CO, sendo a CH II positiva em 315 (19%). Embora 337(20,32%) mulheres tenham sido positivas em um dos testes, somente 19(1,1%) foram positivas nos dois. Entre as 150 que tiveram colposcopia positiva somente em 53 foram encontradas lesÃes no exame histopatolÃgico, sendo a prevalÃncia estimada de 3,2% para qualquer lesÃo e de 0,4% para lesÃes de alto grau/cÃncer. Considerando o ponto de corte o achado de qualquer lesÃo prÃ-neoplÃsica ou neoplÃsicas, a sensibilidade encontrada para a CO e a CH II foi de 30,2% e de 71,7%, respectivamente. A especificidade dos testes mencionados foi de 97,5% e de 82,7%. O VPP e VPN da CO foram de 28,6% e de 97,7%, respectivamente. Jà o VPP e VPN da CH foram 12,1% e 98,9%. Considerando o ponto de corte lesÃes de alto grau ou cÃncer, temos: sensibilidade e especificidade da CO de 28,6% e de 99,9%, enquanto os VPP e VPN foram de 54,8% e de 99,7%, respectivamente. A CH II alcanÃou 100% de sensibilidade e 81,3% de especificidade. Os VPP e VPN ficaram em 2,2% e 100%. ConclusÃo: o teste de detecÃÃo do DNA-HPV pela CH II foi mais sensÃvel, porÃm menos especÃfico que a CO. Quando associado à CO, melhora significativamente a detecÃÃo das lesÃes cervicais, principalmente as de alto grau e cÃncer. Para este grupo de lesÃes, a CH II isolada apresentou melhor especificidade sem perda da sensibilidade, mostrando-se um bom teste para o rastreamento primÃrio. / Objective: to compare the usual Pap smear (Papanicolaou) and the Hybrid Capture II tests in detecting cervical intraepithelial neoplasia in women of Ceara State. Subjects and Methods: 1685 women were enrolled from routine practice in five municipalities of the main Cearà State Health Regions. The whole study was explained to the volunteers, who accepted to participate by signing an informed consent form. The study procedures included filling a questionaire and a cervical sample collection, done by a physician, for cytology and HPV-DNA Hybrid Capture, followed by a complete colposcopic evaluation with directed biopsy if necessary. Data were analyzed in Statistical Package for Social Sciences - SPSS - for Windows 10.0. The accuracy of both tests â Pap smear and Hybrid Capture II - was evaluated by using the sensitivity, specificity, positive predictive value, negative predictive value and the respective 95% confidence intervals. The negative colposcopic examination or negative histological result were considered gold standard for negative results. Positive histological results were considered gold standard for positive results. Results: 56 women (3,4%) had abnormal pap smear. Hybrid Capture tests were positive in 315 women (19%). Despite 337 (20,32%) tests had positive results for one of the two tests, only 19 (1,1%) were positive in both tests. Lesions were detected in 53 women among those 150 considered positive in colposcopic examination. The prevalence for any lesion was estimated in 3,2% and for high grade lesions and cancer in 0,4%. Using the cut-off point as the finding of any cervical lesion, the sensitivity of pap smear and HC II was 30,2% and 71,7%, respectively. The specificity for pap smear and HC II was 97,5% and 82,7%, respectively. The positive and negative predictive value for pap smear was 28,6% and 97,7%, respectively. The positive and negative predictive value for HC II was 12,1% and 98,9%, respectively. By using the cut-off value as high grade cervical lesions and cancer, the sensitivity and specificity for pap smear were 28,6% and 99,9%, respectively, and the positive predictive value and negative predictive value for the same test were 54,8% and 99,7%. The sensitivity and specificity for HC II were 100% and 81,3%, respectively, as well as 2,2% and 100% for positive and negative predictive value. Conclusions: hybrid Capture II test was more sensitive than pap smear, however Hybrid Capture II test was less specific than pap smear. When both tests were used together for detecting cervical lesions the results improved significantly, mainly high grade lesion and cancer. For this group of lesions, HC II alone, presented better specificity, without loss of the sensitivity, apparently itâs a good test for primary sceening.

Conselho Nacional de Desenvolvimento CientÃfico e TecnolÃgico / CoordenaÃÃo de AperfeiÃoamento de Pessoal de NÃvel Superior / A infecÃÃo pelo Papilomavirus humano à a principal causa do carcinoma de colo uterino. No Cearà o cÃncer cervical à o segundo mais freqÃente entre as mulheres, quando se excetua o cÃncer de pele. à uma doenÃa de fÃcil prevenÃÃo, haja vista que quando se dispÃe de um serviÃo bem estruturado a mortalidade à reduzida em atà 80% dos casos. O problema à que muitas mulheres nÃo tÃm acesso a testes de varredura, cujos motivos estÃo relacionados a diversas barreiras culturais e geogrÃficas, bem como a baixa oferta e/ou inexistÃncia de serviÃos. O Cearà conta com cerca de 10.000 agentes comunitÃrios de saÃde, o que representa uma possibilidade de se levar um teste de autocoleta atà estas mulheres, considerando, todavia, que este teste tivesse uma sensibilidade e especificidade aceitÃvel e, assim, atendendo rapidamente um quantitativo significativo de mulheres que nunca fizeram o exame de prevenÃÃo ou estÃo em intervalo recomendado para repeti-lo. Em face ao exposto, este estudo tem por objetivo comparar a sensibilidade e especificidade da Captura HÃbrida de espÃcime cÃrvico-vaginal para HPV-DNA por autocoleta domiciliar em condiÃÃes reais, com a Captura HÃbrida para HPV-DNA coletado por mÃdico ginecologista em consultÃrio e nas condiÃÃes ideais. Os dados foram coletados no perÃodo de agosto a dezembro de 2002, junto a 878 mulheres de cinco cidades cearenses. Estas foram submetidas, na mesma semana, aos testes de Captura HÃbrida por autocoleta, citologia, Captura HÃbrida coletada pelo mÃdico ginecologista, colposcopia e histopatolÃgico, quando necessÃrio. Das 878 mulheres participantes 815 foram consideradas negativas pelo padrÃo ouro, 54 foram consideradas positivas - baixo grau e 9 foram consideradas alto grau - carcinoma. Nos resultados das amostras para Captura HÃbrida HPV-DNA da autocoleta e da coleta mÃdica, obtiveram-se que em 546 casos, ambos os testes concordaram como negativos, em 216 casos concordaram como positivos, em 35 casos houve discordÃncia com a coleta mÃdica positiva e em 81 casos houve discordÃncia com a autocoleta positiva. Os percentuais de sensibilidade encontrados da citologia, Captura HÃbrida por coleta mÃdica e Captura HÃbrida autocoleta foram, respectivamente, 18,3%, 63,3% e 66,7%. Os percentuais de especificidade verificados da citologia Captura HÃbrida coleta mÃdica e Captura HÃbrida autocoleta foram, respectivamente, 98,0%, 73,0% e 68,7%. Comparando-se a concordÃncia entre ambas as coletas de Captura HÃbrida, obteve-se o coeficiente de Kappa (K=0,693) com um erro padrÃo de 0,026 embora existe diferenÃa significativa da prevalÃncia detectada por ambos os testes, teste de McNemar (p<0,001). Avaliando-se as Ãreas da curva ROC para ambos os testes, mostrou-se coleta mÃdica Ãrea=0,738 com coeficiente de 95% de certeza - o intervalo de confianÃa IC=[0,673;0,802] e autocoleta Ãrea= 0,670 com 95% de certeza - o intervalo de confianÃa IC=[0,597;0,742] para detecÃÃo do Papilomavirus. Concluiu-se haver boa concordÃncia entre os resultados obtidos pela autocoleta de espÃcime para realizaÃÃo da Captura HÃbrida HPV-DNA com a coleta mÃdica. / Human Papillomavirus infection is the mean cause of the most of cervical cancers. In Cearà this type of cancer is the second in frequency among the women, when the skin cancers are excluded. Cervical cancer is one the most preventable. Where well structured programs exist the mortality has declined, and in some cases in about 80% .The mean problem is that cervical cancer screening is not fully utilized among groups of women, especially those without access or because there are no services offered or when services exist there are many other barriers, since cultural aspects to geographic barriers. In Cearà exist well structured Health communitarian agents program, which we estimate in about 10.000 agents that cover fully the necessities of population in their areas and the all Cearà territory. With this program we can carry on one screening program by self-sampling if this test were acceptable and had a good sensitivity and specificity. By these communitarian agents we could insert in screening program all women who never underwent to pap smear or other type of screening test. The mean objective of this work is to determine whether testing of self-collected vaginal specimen for Human Papillomavirus has the same accuracy of sampling collected by physicians . In order to evaluate this one Cross-sectional observational study was done between August and December of 2002, 878 women from five municipalities were enrolled and the tests were done in the same week, the women started by doing self-sampling at home and after that they were undergone in physicianâs clinics to the others examination in following sequence cytology, hybrid capture HPV-DNA, colposcopy and when were necessary biopsy. OF 878 women that participated in this study , 815 were considered negative by the gold standard ,54 were considered positive low grade and 9 were high grade/ carcinoma . Of 878 samples to HPV-DNA , there was negative concordance to both test in 546 samples, there was positive concordance in 216 samples, there was discordance in 35 samples where the physician collect were positive and finally there was discordance in 81 samples where the self-collect were positive. The results of sensitivity to cytology, Hybrid Capture by physicians, Hybrid Capture by self-sampling were, respectively 18%, 63,3% and 66,7%. The prevalence estimated by the gold standard were 7,2% in this sample. The results of specificity to cytology, hybrid capture by physicians, hybrid capture by self-sampling were, respectively 98%, 73% and 68,7%. There was significant difference between the results of HPV-DNA self-collected and collected by the physicians, McNemar test p<0,001. When we compare the concordance through the Kappa index we have obtained k=0,693 with stand error of 0,026. Compared the areas obtained by the ROC curve as follows area=0,738 that represent the achievement of physician collected with IC=[0,673;0,802] with 95% confidence interval to detect HPV in sample, area=0,670 that represent the achievement of self-collected sample to HPV with the 95% of confidence interval IC=[0,597;0,742].We concluded that Hybrid Capture by self-collected vaginal sample is as good as Hybrid Capture collected by the physicians, and there was good concordance between these tests.