Ettore Cicinelli1, Amerigo Vitagliano2, Alka Kumar3, Riccardo Bassil Lasmar4, Stefano Bettocchi1, Sergio Haimovich5. 1. 2nd Unit of Obstetrics and Gynecology, Department of Biomedical and Human Oncologic Science, Policlinico University of Bari, Bari, Italy. 2. Department of Women and Children's Health University of Padua, Padua, Italy. Electronic address: amerigovitagliano.md@gmail.com. 3. Hysteroscopic Surgery Division, Women's Health Centre, Jaipur, India. 4. Department of Maternal and Child Health, Federal Fluminense University, Brazil. 5. Department of Obstetrics and Gynecology, Hospital del Mar, Autonomous University of Barcelona, Barcelona, Spain.
Abstract
OBJECTIVE: To develop a consensus on the diagnostic criteria for chronic endometritis (CE) at hysteroscopy (HSC), and to evaluate these proposed criteria in a randomized-controlled observer study. DESIGN: Systematic review of studies evaluating the diagnostic accuracy of HSC in CE diagnosis; Delphi consensus on hysteroscopic diagnostic criteria for CE; randomized-controlled observer study to evaluate the reproducibility of the proposed diagnostic criteria. SETTING: Not applicable. PARTICIPANT(S): Experts from different countries were involved in the systematic review and contributed to the Delphi consensus. Physicians from different countries were involved in the observer study. INTERVENTION(S): After reaching consensus on the diagnostic criteria, the Delphi poll created a questionnaire including 100 hysteroscopic pictures (50 from women with CE [domain 1] and 50 from women without CE [domain 2]), with a single question per picture (Answer_A: suggestive of CE; answer B: not suggestive of CE). A total of 200 physicians were invited to take part in the observer study. Before completing the questionnaire, physicians were randomized to receive a description of the diagnostic criteria (group A) or no such information (group B). MAIN OUTCOME MEASURE(S): The primary outcome was to compare the questionnaire scores for the two groups of observers. The secondary outcome was to assess the interobserver agreement in the diagnosis of CE in each group. RESULT(S): A total of 126 physicians completed the questionnaire (62 in group A and 64 in group B). Observers in group A obtained higher total scores compared with those in group B (P<.001). Specifically, group A showed higher mean score in domain 1 (P<.001), but not in domain 2 (P=.975). A substantial agreement was found among observers in group A (intraclass correlation coefficient [ICC] 0.78), whereas a fair agreement was found among observers in group B (ICC 0.40). CONCLUSION(S): This randomized-controlled observer study found a positive impact of our criteria on physicians' ability to recognize CE.
RCT Entities:
OBJECTIVE: To develop a consensus on the diagnostic criteria for chronic endometritis (CE) at hysteroscopy (HSC), and to evaluate these proposed criteria in a randomized-controlled observer study. DESIGN: Systematic review of studies evaluating the diagnostic accuracy of HSC in CE diagnosis; Delphi consensus on hysteroscopic diagnostic criteria for CE; randomized-controlled observer study to evaluate the reproducibility of the proposed diagnostic criteria. SETTING: Not applicable. PARTICIPANT(S): Experts from different countries were involved in the systematic review and contributed to the Delphi consensus. Physicians from different countries were involved in the observer study. INTERVENTION(S): After reaching consensus on the diagnostic criteria, the Delphi poll created a questionnaire including 100 hysteroscopic pictures (50 from women with CE [domain 1] and 50 from women without CE [domain 2]), with a single question per picture (Answer_A: suggestive of CE; answer B: not suggestive of CE). A total of 200 physicians were invited to take part in the observer study. Before completing the questionnaire, physicians were randomized to receive a description of the diagnostic criteria (group A) or no such information (group B). MAIN OUTCOME MEASURE(S): The primary outcome was to compare the questionnaire scores for the two groups of observers. The secondary outcome was to assess the interobserver agreement in the diagnosis of CE in each group. RESULT(S): A total of 126 physicians completed the questionnaire (62 in group A and 64 in group B). Observers in group A obtained higher total scores compared with those in group B (P<.001). Specifically, group A showed higher mean score in domain 1 (P<.001), but not in domain 2 (P=.975). A substantial agreement was found among observers in group A (intraclass correlation coefficient [ICC] 0.78), whereas a fair agreement was found among observers in group B (ICC 0.40). CONCLUSION(S): This randomized-controlled observer study found a positive impact of our criteria on physicians' ability to recognize CE.