Manuel Maria Ianieri1, Tommaso Staniscia2, Giovanni Pontrelli3, Attilio Di Spiezio Sardo4, Federica Stella Manzi5, Maristella Recchi6, Marco Liberati7, Marcello Ceccaroni3. 1. Department of Obstetrics and Gynecology, Gynecologic Oncology and Minimally Invasive Pelvic Surgery, International School of Surgical Anatomy, Sacred Heart Hospital, Negrar, Verona, Italy. Electronic address: manuel_dorian@yahoo.it. 2. Department of Medicine and Ageing, University "G. d'Annunzio" of Chieti-Pescara, Italy. 3. Department of Obstetrics and Gynecology, Gynecologic Oncology and Minimally Invasive Pelvic Surgery, International School of Surgical Anatomy, Sacred Heart Hospital, Negrar, Verona, Italy. 4. Department of Obstetrics and Gynecology, University of Naples "Federico II", Naples, Italy. 5. School of Medicine and Health Science, University "G. d'Annunzio" of Chieti-Pescara, Italy. 6. Unit of Obstetrics and Gynecology, Cittadella Hospital, Padova, Italy. 7. Obstetrics and Gynecology Unit, "SS. Annunziata" Hospital, Chieti, Italy; University "G. d'Annunzio" of Chieti-Pescara, Italy.
Abstract
STUDY OBJECTIVE: To develop a new hysteroscopic morphologic scoring system that helps physicians, especially those who have less experience, to make a differential diagnosis among normal endometrium (NE), endometrial hyperplasia, and endometrial carcinoma. DESIGN: A retrospective study (Canadian Task Force Classification II). SETTING: An office hysteroscopy service. PATIENTS: A total of 435 endometrial biopsies were included in the study: 201 NE, 160 endometrial hyperplasia without atypia (EH), 30 atypical endometrial hyperplasia (AEH), and 44 endometrial cancer (EC). INTERVENTIONS: The authors retrospectively evaluated all videos of diagnostic hysteroscopies performed before endometrial biopsies to note endometrial morphologic parameters suggestive of pathology. Principal significant variables were selected by means of the chi-square test (p < .05) and integrated into an ordinal multivariate analysis. Through the estimate of the beta coefficient, a score was obtained to be appointed to each of the selected variables, and characteristic intervals of each of the endometrial lesions were created. MEASUREMENTS AND MAIN RESULTS: The scoring system showed a sensitivity and specificity of 71.1% and 80%, 48.7% and 82.5%, 63.3% and 90.4%, and 95.4% and 98.2% regarding NE, EH, AEH, and EC, respectively. The positive predictive values and negative predictive values, respectively, were 76.8% and 80% for NE, 62% and 73.5% for EH, 32.7% and 97% for AEH, and 85.7% and 99.5% for EC. CONCLUSIONS: The proposed scoring system showed good diagnostic performance, especially in relation to endometrial cancer, and may represent a useful diagnostic tool, mainly for operators with less experience.
STUDY OBJECTIVE: To develop a new hysteroscopic morphologic scoring system that helps physicians, especially those who have less experience, to make a differential diagnosis among normal endometrium (NE), endometrial hyperplasia, and endometrial carcinoma. DESIGN: A retrospective study (Canadian Task Force Classification II). SETTING: An office hysteroscopy service. PATIENTS: A total of 435 endometrial biopsies were included in the study: 201 NE, 160 endometrial hyperplasia without atypia (EH), 30 atypical endometrial hyperplasia (AEH), and 44 endometrial cancer (EC). INTERVENTIONS: The authors retrospectively evaluated all videos of diagnostic hysteroscopies performed before endometrial biopsies to note endometrial morphologic parameters suggestive of pathology. Principal significant variables were selected by means of the chi-square test (p < .05) and integrated into an ordinal multivariate analysis. Through the estimate of the beta coefficient, a score was obtained to be appointed to each of the selected variables, and characteristic intervals of each of the endometrial lesions were created. MEASUREMENTS AND MAIN RESULTS: The scoring system showed a sensitivity and specificity of 71.1% and 80%, 48.7% and 82.5%, 63.3% and 90.4%, and 95.4% and 98.2% regarding NE, EH, AEH, and EC, respectively. The positive predictive values and negative predictive values, respectively, were 76.8% and 80% for NE, 62% and 73.5% for EH, 32.7% and 97% for AEH, and 85.7% and 99.5% for EC. CONCLUSIONS: The proposed scoring system showed good diagnostic performance, especially in relation to endometrial cancer, and may represent a useful diagnostic tool, mainly for operators with less experience.