Artur Ludwin1, Inga Ludwin2, Marek Kudla3, Jan Kottner4. 1. Department of Gynecology and Oncology, Jagiellonian University, Krakow, Poland; Ludwin & Ludwin Gynecology, Private Medical Center, Krakow, Poland. Electronic address: ludwin@cm-uj.krakow.pl. 2. Department of Gynecology and Oncology, Jagiellonian University, Krakow, Poland; Ludwin & Ludwin Gynecology, Private Medical Center, Krakow, Poland. 3. Department of Oncological Gynecology, Gynecology and Obstetrics, Medical University of Silesia, Katowice, Poland. 4. Department of Dermatology and Allergy, Charité-Universitätsmedizin Berlin, Berlin, Germany.
Abstract
OBJECTIVE: To estimate the inter-rater/intrarater reliability of the European Society of Human Reproduction and Embryology/European Society for Gynaecological Endoscopy (ESHRE-ESGE) classification of congenital uterine malformations and to compare the results obtained with the reliability of the American Society for Reproductive Medicine (ASRM) classification supplemented with additional morphometric criteria. DESIGN: Reliability/agreement study. SETTING: Private clinic. PATIENT(S): Uterine malformations (n = 50 patients, consecutively included) and normal uterus (n = 62 women, randomly selected) constituted the study. These were classified based on real-time three-dimensional ultrasound single volume transvaginal (or transrectal in the case of virgins, 4 cases) ultrasonography findings, which were assessed by an expert rater based on the ESHRE-ESGE criteria. The samples were obtained from women of reproductive age. INTERVENTION(S): Unprocessed three-dimensional datasets were independently evaluated offline by two experienced, blinded raters using both classification systems. MAIN OUTCOME MEASURE(S): The κ-values and proportions of agreement. RESULT(S): Standardized interpretation indicated that the ESHRE-ESGE system has substantial/good or almost perfect/very good reliability (κ >0.60 and >0.80), but the interpretation of the clinically relevant cutoffs of κ-values showed insufficient reliability for clinical use (κ < 0.90), especially in the diagnosis of septate uterus. The ASRM system had sufficient reliability (κ > 0.95). CONCLUSION(S): The low reliability of the ESHRE-ESGE system may lead to a lack of consensus about the management of common uterine malformations and biased research interpretations. The use of the ASRM classification, supplemented with simple morphometric criteria, may be preferred if their sufficient reliability can be confirmed real-time in a large sample size.
OBJECTIVE: To estimate the inter-rater/intrarater reliability of the European Society of Human Reproduction and Embryology/European Society for Gynaecological Endoscopy (ESHRE-ESGE) classification of congenital uterine malformations and to compare the results obtained with the reliability of the American Society for Reproductive Medicine (ASRM) classification supplemented with additional morphometric criteria. DESIGN: Reliability/agreement study. SETTING: Private clinic. PATIENT(S): Uterine malformations (n = 50 patients, consecutively included) and normal uterus (n = 62 women, randomly selected) constituted the study. These were classified based on real-time three-dimensional ultrasound single volume transvaginal (or transrectal in the case of virgins, 4 cases) ultrasonography findings, which were assessed by an expert rater based on the ESHRE-ESGE criteria. The samples were obtained from women of reproductive age. INTERVENTION(S): Unprocessed three-dimensional datasets were independently evaluated offline by two experienced, blinded raters using both classification systems. MAIN OUTCOME MEASURE(S): The κ-values and proportions of agreement. RESULT(S): Standardized interpretation indicated that the ESHRE-ESGE system has substantial/good or almost perfect/very good reliability (κ >0.60 and >0.80), but the interpretation of the clinically relevant cutoffs of κ-values showed insufficient reliability for clinical use (κ < 0.90), especially in the diagnosis of septate uterus. The ASRM system had sufficient reliability (κ > 0.95). CONCLUSION(S): The low reliability of the ESHRE-ESGE system may lead to a lack of consensus about the management of common uterine malformations and biased research interpretations. The use of the ASRM classification, supplemented with simple morphometric criteria, may be preferred if their sufficient reliability can be confirmed real-time in a large sample size.