Objetivo: Avaliar a concordância entre o exame anatomopatológico transoperatório de congelação (TO) e o diagnóstico histológico no exame anatomopatológico convencional (AP-conv) nas massas anexiais, divididas em grupos conforme seu tamanho e suas características morfológicas na ultrassonografia da pelve, para especificar fatores ultrassonográficos preditores de erro no TO. Os diagnósticos do TO nos grupos foram comparados com os AP-conv de tumores benignos, borderline e malignos. Métodos: Estudo transversal com avaliação retrospectiva em 302 pacientes com diagnóstico ultrassonográfico de massas anexiais, submetidas a procedimento cirúrgico no Hospital de Clínicas de Porto Alegre. Estas foram divididas em oito grupos, conforme as características morfológicas ultrassonográficas e o tamanho tumoral. Grupo 1: tumores uniloculares ≤ 10 cm; grupo 2: tumores líquidos septados ≤ 10 cm; grupo 3: tumores heterogêneos ≤ 10 cm; grupo 4: tumores sólidos ≤ 10 cm; grupo 5: uniloculares > 10 cm; grupo 6: líquidos septados > 10 cm; grupo 7: heterogêneos > 10 cm; e grupo 8: sólidos > 10 cm. O resultado diagnóstico do TO foi então comparado com o diagnóstico histológico final no AP-conv. Resultados: A concordância diagnóstica variou entre os grupos. Nos 33 casos do grupo 1, houve 100% de concordância (Kappa 1) entre o TO e o AP-conv. No grupo 2, com 32 casos, também houve 100% de concordância, assim como nos seis casos do grupo 8 (Kappa 1). No grupo 3, com 90 casos, a concordância diagnóstica também foi ótima (Kappa 0,898), com dois casos discordantes (2,22%): um diagnóstico benigno no TO que se confirmou borderline no AP-conv e outro benigno no TO que se confirmou maligno no AP-conv. O grupo 4, com 24 casos, apresentou uma discordância (4,17%) de benigna no TO e maligna no AP-conv (Kappa 0,869). No grupo 5, houve concordância em 93% dos 15 casos, com uma discordância (6,67%) no diagnóstico: benigno no TO e maligno no AP-conv; não foi possível calcular o Kappa neste grupo. Dos 39 casos do grupo 6, 89,74% tiveram o diagnóstico concordante, com duas discordâncias (5,13%): benignos no TO foram borderline no AP-conv; uma discordância (2,57%): borderline no TO foi benigno no AP-conv; e outra discordância (2,57%): borderline no TO foi maligno no AP-conv (Kappa 0,591). O grupo 7, com 63 pacientes, teve concordância em 55 casos (87,30%), com oito casos discordantes (12,70%): dos seis benignos (9,52%) no TO, três foram borderline (4,76%) e três malignos (4,76%) no AP-conv; dos dois borderline (3,18%) no TO, um foi benigno (1,59%) e um maligno (1,59%) no AP-conv (Kappa 0,776). Conclusão: O TO tem uma concordância com o AP-conv que varia de ótima em tumores císticos a moderada em tumores multiloculados com mais de 10 cm. Nosso estudo apresenta limitações por ser um estudo retrospectivo, além de não haver sido o mesmo patologista quem avaliou todas as peças. Mas a estratificação das massas anexiais em grupos, de acordo com seu tamanho e características morfológicas na ultrassonografia, é um bom método para avaliação pré-operatória de massa anexiais, sabendo-se que, nas lesões císticas septadas ou com componentes sólidos maiores que 10 cm, a concordância do TO com o AP-conv é moderada. Portanto, devemos estar cientes que, em tumores maiores de 10 cm com componente sólido, o erro diagnóstico do TO aumenta. Assim, nesses casos, o patologista e o cirurgião deverão estar atentos para um correto diagnóstico e um planejamento adequado do tratamento, evitando, com isso, o subtratamento ou o sobretratamento da paciente. / Objective: To assess agreement between intraoperative frozen section (IFS) and final histopathology (HPE) for anatomic pathology examination of adnexal masses stratified according to size and morphological characteristics on pelvic ultrasonography and define sonographic predictors of diagnostic error of IFS. IFS classification of masses as benign, borderline, or malignant was compared to final diagnoses after HPE. Methods: Cross-sectional study with retrospective assessment of 302 patients with a sonographic diagnosis of adnexal masses that underwent surgical treatment at Hospital de Clínicas de Porto Alegre. Patients were divided into eight groups according to mass size and sonographic morphology as follows: Group 1, unilocular tumors ≤10 cm in size; Group 2, septated cystic tumors ≤10 cm in size; Group 3, heterogeneous tumors ≤10 cm in size; Group 4, solid tumors ≤10 cm in size; Group 5, unilocular tumors >10 cm in size; Group 6, septated cystic tumors >10 cm in size; Group 7, heterogeneous tumors >10 cm in size; and Group 8, solid tumors >10 cm in size. The diagnostic findings of IFS were then compared with the final histopathologic diagnosis. Results: Diagnostic agreement varied among groups. In Groups 1 (33 cases), 2 (32 cases), and 8 (6 cases), there was 100% agreement between IFS and HPE (Kappa = 1.0). In Group 3 (90 cases), agreement was excellent (Kappa = 0.898), with only two divergences (2.22%): one mass classified as benign on IFS that was borderline on HPE and another initially classified as benign that later proved malignant on HPE. In Group 4 (24 cases), Kappa was 0.869, with one divergence (4.17%), again a mass classified as benign on IFS which proved malignant on HPE. In Group 5, there was agreement in 93% of 15 cases, with one divergence (6.67%) in diagnosis: benign on IFS and malignant on HPE. Kappa could not be calculated for this group. Of the 39 cases in Group 6, there was agreement in 89.74%, with two masses classified as benign on IFS later deemed malignant on HPE (5.13%); one borderline on IFS diagnosed as benign on HPE (2.57%); and one borderline on IFS and diagnosed as malignant on HPE (2.57%) (Kappa = 0.591). In Group 7 (63 patients), there was agreement in 55 cases (87.30%), with eight divergences (12.70%): of six masses deemed benign on IFS (9.52%), three (4.76%) were diagnosed as borderline and three (4.76%) as malignant on HPE; of two masses deemed borderline on IFS (3.18%), one was later deemed benign (1.59%) and one diagnosed as malignant (1.59%) on HPE (Kappa = 0.776). Conclusion: Agreement between IFS and HPE ranged from excellent (for cystic masses) to moderate (for multilocular tumors larger than 10 cm). Limitations of this study include its retrospective design and the fact that not all surgical specimens were examined by the same pathologist. Nevertheless, stratification of adnexal masses by sonographic morphology and size is a good method for preoperative assessment, with the knowledge that agreement between IFS and HPE is only moderate for septated cystic or heterogeneous lesions larger than 10 cm. Therefore, clinicians should bear in mind that the diagnostic error of IFS is higher for adnexal masses >10 cm in size with a solid component. In these cases, particular care is required from both the pathologist and surgeon for proper diagnosis and treatment planning, thus avoiding undertreatment or overtreatment.

Objetivo: Avaliar a concordância entre o exame anatomopatológico transoperatório de congelação (TO) e o diagnóstico histológico no exame anatomopatológico convencional (AP-conv) nas massas anexiais, divididas em grupos conforme seu tamanho e suas características morfológicas na ultrassonografia da pelve, para especificar fatores ultrassonográficos preditores de erro no TO. Os diagnósticos do TO nos grupos foram comparados com os AP-conv de tumores benignos, borderline e malignos. Métodos: Estudo transversal com avaliação retrospectiva em 302 pacientes com diagnóstico ultrassonográfico de massas anexiais, submetidas a procedimento cirúrgico no Hospital de Clínicas de Porto Alegre. Estas foram divididas em oito grupos, conforme as características morfológicas ultrassonográficas e o tamanho tumoral. Grupo 1: tumores uniloculares ≤ 10 cm; grupo 2: tumores líquidos septados ≤ 10 cm; grupo 3: tumores heterogêneos ≤ 10 cm; grupo 4: tumores sólidos ≤ 10 cm; grupo 5: uniloculares > 10 cm; grupo 6: líquidos septados > 10 cm; grupo 7: heterogêneos > 10 cm; e grupo 8: sólidos > 10 cm. O resultado diagnóstico do TO foi então comparado com o diagnóstico histológico final no AP-conv. Resultados: A concordância diagnóstica variou entre os grupos. Nos 33 casos do grupo 1, houve 100% de concordância (Kappa 1) entre o TO e o AP-conv. No grupo 2, com 32 casos, também houve 100% de concordância, assim como nos seis casos do grupo 8 (Kappa 1). No grupo 3, com 90 casos, a concordância diagnóstica também foi ótima (Kappa 0,898), com dois casos discordantes (2,22%): um diagnóstico benigno no TO que se confirmou borderline no AP-conv e outro benigno no TO que se confirmou maligno no AP-conv. O grupo 4, com 24 casos, apresentou uma discordância (4,17%) de benigna no TO e maligna no AP-conv (Kappa 0,869). No grupo 5, houve concordância em 93% dos 15 casos, com uma discordância (6,67%) no diagnóstico: benigno no TO e maligno no AP-conv; não foi possível calcular o Kappa neste grupo. Dos 39 casos do grupo 6, 89,74% tiveram o diagnóstico concordante, com duas discordâncias (5,13%): benignos no TO foram borderline no AP-conv; uma discordância (2,57%): borderline no TO foi benigno no AP-conv; e outra discordância (2,57%): borderline no TO foi maligno no AP-conv (Kappa 0,591). O grupo 7, com 63 pacientes, teve concordância em 55 casos (87,30%), com oito casos discordantes (12,70%): dos seis benignos (9,52%) no TO, três foram borderline (4,76%) e três malignos (4,76%) no AP-conv; dos dois borderline (3,18%) no TO, um foi benigno (1,59%) e um maligno (1,59%) no AP-conv (Kappa 0,776). Conclusão: O TO tem uma concordância com o AP-conv que varia de ótima em tumores císticos a moderada em tumores multiloculados com mais de 10 cm. Nosso estudo apresenta limitações por ser um estudo retrospectivo, além de não haver sido o mesmo patologista quem avaliou todas as peças. Mas a estratificação das massas anexiais em grupos, de acordo com seu tamanho e características morfológicas na ultrassonografia, é um bom método para avaliação pré-operatória de massa anexiais, sabendo-se que, nas lesões císticas septadas ou com componentes sólidos maiores que 10 cm, a concordância do TO com o AP-conv é moderada. Portanto, devemos estar cientes que, em tumores maiores de 10 cm com componente sólido, o erro diagnóstico do TO aumenta. Assim, nesses casos, o patologista e o cirurgião deverão estar atentos para um correto diagnóstico e um planejamento adequado do tratamento, evitando, com isso, o subtratamento ou o sobretratamento da paciente. / Objective: To assess agreement between intraoperative frozen section (IFS) and final histopathology (HPE) for anatomic pathology examination of adnexal masses stratified according to size and morphological characteristics on pelvic ultrasonography and define sonographic predictors of diagnostic error of IFS. IFS classification of masses as benign, borderline, or malignant was compared to final diagnoses after HPE. Methods: Cross-sectional study with retrospective assessment of 302 patients with a sonographic diagnosis of adnexal masses that underwent surgical treatment at Hospital de Clínicas de Porto Alegre. Patients were divided into eight groups according to mass size and sonographic morphology as follows: Group 1, unilocular tumors ≤10 cm in size; Group 2, septated cystic tumors ≤10 cm in size; Group 3, heterogeneous tumors ≤10 cm in size; Group 4, solid tumors ≤10 cm in size; Group 5, unilocular tumors >10 cm in size; Group 6, septated cystic tumors >10 cm in size; Group 7, heterogeneous tumors >10 cm in size; and Group 8, solid tumors >10 cm in size. The diagnostic findings of IFS were then compared with the final histopathologic diagnosis. Results: Diagnostic agreement varied among groups. In Groups 1 (33 cases), 2 (32 cases), and 8 (6 cases), there was 100% agreement between IFS and HPE (Kappa = 1.0). In Group 3 (90 cases), agreement was excellent (Kappa = 0.898), with only two divergences (2.22%): one mass classified as benign on IFS that was borderline on HPE and another initially classified as benign that later proved malignant on HPE. In Group 4 (24 cases), Kappa was 0.869, with one divergence (4.17%), again a mass classified as benign on IFS which proved malignant on HPE. In Group 5, there was agreement in 93% of 15 cases, with one divergence (6.67%) in diagnosis: benign on IFS and malignant on HPE. Kappa could not be calculated for this group. Of the 39 cases in Group 6, there was agreement in 89.74%, with two masses classified as benign on IFS later deemed malignant on HPE (5.13%); one borderline on IFS diagnosed as benign on HPE (2.57%); and one borderline on IFS and diagnosed as malignant on HPE (2.57%) (Kappa = 0.591). In Group 7 (63 patients), there was agreement in 55 cases (87.30%), with eight divergences (12.70%): of six masses deemed benign on IFS (9.52%), three (4.76%) were diagnosed as borderline and three (4.76%) as malignant on HPE; of two masses deemed borderline on IFS (3.18%), one was later deemed benign (1.59%) and one diagnosed as malignant (1.59%) on HPE (Kappa = 0.776). Conclusion: Agreement between IFS and HPE ranged from excellent (for cystic masses) to moderate (for multilocular tumors larger than 10 cm). Limitations of this study include its retrospective design and the fact that not all surgical specimens were examined by the same pathologist. Nevertheless, stratification of adnexal masses by sonographic morphology and size is a good method for preoperative assessment, with the knowledge that agreement between IFS and HPE is only moderate for septated cystic or heterogeneous lesions larger than 10 cm. Therefore, clinicians should bear in mind that the diagnostic error of IFS is higher for adnexal masses >10 cm in size with a solid component. In these cases, particular care is required from both the pathologist and surgeon for proper diagnosis and treatment planning, thus avoiding undertreatment or overtreatment.

Objetivo: Avaliar a concordância entre o exame anatomopatológico transoperatório de congelação (TO) e o diagnóstico histológico no exame anatomopatológico convencional (AP-conv) nas massas anexiais, divididas em grupos conforme seu tamanho e suas características morfológicas na ultrassonografia da pelve, para especificar fatores ultrassonográficos preditores de erro no TO. Os diagnósticos do TO nos grupos foram comparados com os AP-conv de tumores benignos, borderline e malignos. Métodos: Estudo transversal com avaliação retrospectiva em 302 pacientes com diagnóstico ultrassonográfico de massas anexiais, submetidas a procedimento cirúrgico no Hospital de Clínicas de Porto Alegre. Estas foram divididas em oito grupos, conforme as características morfológicas ultrassonográficas e o tamanho tumoral. Grupo 1: tumores uniloculares ≤ 10 cm; grupo 2: tumores líquidos septados ≤ 10 cm; grupo 3: tumores heterogêneos ≤ 10 cm; grupo 4: tumores sólidos ≤ 10 cm; grupo 5: uniloculares > 10 cm; grupo 6: líquidos septados > 10 cm; grupo 7: heterogêneos > 10 cm; e grupo 8: sólidos > 10 cm. O resultado diagnóstico do TO foi então comparado com o diagnóstico histológico final no AP-conv. Resultados: A concordância diagnóstica variou entre os grupos. Nos 33 casos do grupo 1, houve 100% de concordância (Kappa 1) entre o TO e o AP-conv. No grupo 2, com 32 casos, também houve 100% de concordância, assim como nos seis casos do grupo 8 (Kappa 1). No grupo 3, com 90 casos, a concordância diagnóstica também foi ótima (Kappa 0,898), com dois casos discordantes (2,22%): um diagnóstico benigno no TO que se confirmou borderline no AP-conv e outro benigno no TO que se confirmou maligno no AP-conv. O grupo 4, com 24 casos, apresentou uma discordância (4,17%) de benigna no TO e maligna no AP-conv (Kappa 0,869). No grupo 5, houve concordância em 93% dos 15 casos, com uma discordância (6,67%) no diagnóstico: benigno no TO e maligno no AP-conv; não foi possível calcular o Kappa neste grupo. Dos 39 casos do grupo 6, 89,74% tiveram o diagnóstico concordante, com duas discordâncias (5,13%): benignos no TO foram borderline no AP-conv; uma discordância (2,57%): borderline no TO foi benigno no AP-conv; e outra discordância (2,57%): borderline no TO foi maligno no AP-conv (Kappa 0,591). O grupo 7, com 63 pacientes, teve concordância em 55 casos (87,30%), com oito casos discordantes (12,70%): dos seis benignos (9,52%) no TO, três foram borderline (4,76%) e três malignos (4,76%) no AP-conv; dos dois borderline (3,18%) no TO, um foi benigno (1,59%) e um maligno (1,59%) no AP-conv (Kappa 0,776). Conclusão: O TO tem uma concordância com o AP-conv que varia de ótima em tumores císticos a moderada em tumores multiloculados com mais de 10 cm. Nosso estudo apresenta limitações por ser um estudo retrospectivo, além de não haver sido o mesmo patologista quem avaliou todas as peças. Mas a estratificação das massas anexiais em grupos, de acordo com seu tamanho e características morfológicas na ultrassonografia, é um bom método para avaliação pré-operatória de massa anexiais, sabendo-se que, nas lesões císticas septadas ou com componentes sólidos maiores que 10 cm, a concordância do TO com o AP-conv é moderada. Portanto, devemos estar cientes que, em tumores maiores de 10 cm com componente sólido, o erro diagnóstico do TO aumenta. Assim, nesses casos, o patologista e o cirurgião deverão estar atentos para um correto diagnóstico e um planejamento adequado do tratamento, evitando, com isso, o subtratamento ou o sobretratamento da paciente. / Objective: To assess agreement between intraoperative frozen section (IFS) and final histopathology (HPE) for anatomic pathology examination of adnexal masses stratified according to size and morphological characteristics on pelvic ultrasonography and define sonographic predictors of diagnostic error of IFS. IFS classification of masses as benign, borderline, or malignant was compared to final diagnoses after HPE. Methods: Cross-sectional study with retrospective assessment of 302 patients with a sonographic diagnosis of adnexal masses that underwent surgical treatment at Hospital de Clínicas de Porto Alegre. Patients were divided into eight groups according to mass size and sonographic morphology as follows: Group 1, unilocular tumors ≤10 cm in size; Group 2, septated cystic tumors ≤10 cm in size; Group 3, heterogeneous tumors ≤10 cm in size; Group 4, solid tumors ≤10 cm in size; Group 5, unilocular tumors >10 cm in size; Group 6, septated cystic tumors >10 cm in size; Group 7, heterogeneous tumors >10 cm in size; and Group 8, solid tumors >10 cm in size. The diagnostic findings of IFS were then compared with the final histopathologic diagnosis. Results: Diagnostic agreement varied among groups. In Groups 1 (33 cases), 2 (32 cases), and 8 (6 cases), there was 100% agreement between IFS and HPE (Kappa = 1.0). In Group 3 (90 cases), agreement was excellent (Kappa = 0.898), with only two divergences (2.22%): one mass classified as benign on IFS that was borderline on HPE and another initially classified as benign that later proved malignant on HPE. In Group 4 (24 cases), Kappa was 0.869, with one divergence (4.17%), again a mass classified as benign on IFS which proved malignant on HPE. In Group 5, there was agreement in 93% of 15 cases, with one divergence (6.67%) in diagnosis: benign on IFS and malignant on HPE. Kappa could not be calculated for this group. Of the 39 cases in Group 6, there was agreement in 89.74%, with two masses classified as benign on IFS later deemed malignant on HPE (5.13%); one borderline on IFS diagnosed as benign on HPE (2.57%); and one borderline on IFS and diagnosed as malignant on HPE (2.57%) (Kappa = 0.591). In Group 7 (63 patients), there was agreement in 55 cases (87.30%), with eight divergences (12.70%): of six masses deemed benign on IFS (9.52%), three (4.76%) were diagnosed as borderline and three (4.76%) as malignant on HPE; of two masses deemed borderline on IFS (3.18%), one was later deemed benign (1.59%) and one diagnosed as malignant (1.59%) on HPE (Kappa = 0.776). Conclusion: Agreement between IFS and HPE ranged from excellent (for cystic masses) to moderate (for multilocular tumors larger than 10 cm). Limitations of this study include its retrospective design and the fact that not all surgical specimens were examined by the same pathologist. Nevertheless, stratification of adnexal masses by sonographic morphology and size is a good method for preoperative assessment, with the knowledge that agreement between IFS and HPE is only moderate for septated cystic or heterogeneous lesions larger than 10 cm. Therefore, clinicians should bear in mind that the diagnostic error of IFS is higher for adnexal masses >10 cm in size with a solid component. In these cases, particular care is required from both the pathologist and surgeon for proper diagnosis and treatment planning, thus avoiding undertreatment or overtreatment.

Les masses kystiques latérales du cou (MKLC) bénignes et malignes sont difficiles à différencier cliniquement. L’utilité des modalités d’imagerie et de prélèvement doit être clarifiée. Une revue rétrospective de cas entre 2010 et 2016. Les données d’imagerie ont été récoltées et plusieurs variables propres à la masse furent analysées. Les rapports de cytoponction à l’aiguille fine (CAAF), de la biopsie au trocart (BT) et des examens extemporanés (EE) ont été analysés. La sensibilité, la spécificité, la valeur prédictive positive (VPP) et la valeur prédictive négative (VPN) pour prédire la malignité ont été calculées pour toutes les variables comparées entre les masses kystiques bénignes et malignes. Aucune variable d’imagerie n’a pu différencier les masses kystiques bénignes de malignes. La sensibilité de la CAAF est plus basse que celle de la BT (59% vs 83%; p=0.036) et de l’EE (59% vs 93%; p=0.01). L’EE a une meilleure VPN que la CAAF (92% vs 40%; p<0.001) et que la BT (92% vs 50%; p=0.062). La VPP et la spécificité étaient similaires dans tous les groupes. Les cliniciens ne peuvent pas se fier uniquement à l’imagerie pour différencier les masses bénignes des masses malignes. Vu sa VPP adéquate (92%), la CAAF devrait être utilisée initialement pour tous les patients avec une MKLC. Si la CAAF s’avère négative, la BT devrait être utilisée vu sa meilleure sensibilité. Un examen extemporané devrait toujours suivre une BT négative vu la faible VPN de la BT. Un résultat positif à l’une des trois modalités de prélèvement indique la présence de malignité. / Benign and malignant lateral cystic neck masses (LCNM) are difficult to distinguish clinically. The usefulness of imaging and sampling modalities in clarifying the diagnosis remains unclear. Retrospective review of cases between 2010 and 2016. Imaging data was reviewed and the variables pertaining to the mass were assessed including the following: size, nodal level, fat stranding, extracapsular spread, calcifications, vascularity, necrosis and standardized uptake value. Sampling reports of fine-needle aspiration (FNA), core-needle biopsy (CNB) and frozen section (FS) were also assessed. Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) for predicting malignancy were calculated for all variables and compared between benign and malignant cystic neck masses. Ultrasound was used in 47.2% and CT-Scan in 90.5% of patients. No variables on imaging could definitely differentiate benign from malignant LCNM. FNA had a lower sensitivity then CNB (59% vs 83%; p=0.036) and FS (59% vs 93%; p=0.01). FS had a better NPV when compared to FNA (92% vs 40%; p<0.001) and CNB (92% vs 50%; p=0.062). Specificities and PPV were similar among all groups. Clinicians cannot rely on imaging to differentiate benign from malignant LCNM. Given its adequate PPV (92%), FNA should be used initially on lateral cystic neck masses. Because of its high sensitivity, CNB should be considered if FNA is not diagnostic of malignancy. FS should always follow a CNB not indicative of malignancy, because of the low NPV. Any result diagnostic of malignancy on either FNA, CNB or FS strongly indicates presence of malignancy.

The sentinel lymph node biopsy is part of recommended surgical staging guidelines in patients with early stages of cervical cancer. High success rates of bilateral detection of SLN are achieved in sites with adequate experience with this procedure. The sentinel lymph node biopsy without systematic pelvic lymph node dissection is currently considered inadequate procedure for stages IB to IIA of the disease. One of the benefits of sentinel lymph node detection is extensive histopathological examination using the ultrastaging protocol enabling detection of small metastases (i.e. micrometastases). At the moment, there is lack of evidence about oncological safety of sentinel lymph node biopsy which might replace systematic lymph node dissection in the future. Prognostic significance of micrometastases is also controversial due to the lack of data about their potential presence in non-sentinel lymph nodes in cases with negative sentinel lymph nodes. This dissertation deals with the concept of sentinel lymph node biopsy in the cervical cancer and focuses on several topics. We have shown that the presence of micrometastasis is associated with significant negative impact on patients' prognosis on the largest retrospective cohort of patients ever published. Only 67% of patients with micrometastasis have...