J Johns1, N Greenwold, S Buckley, E Jauniaux. 1. Elizabeth Garrett Anderson Hospital, Academic Department of Obstetrics and Gynaecology, University College London Hospital, London, UK.
Abstract
OBJECTIVE: To examine the relationship between ultrasound and histological features in the screening for molar changes in missed miscarriage. METHODS: A prospective cohort study was conducted on all missed miscarriages, with features suspicious of molar pregnancy, on transvaginal ultrasound and/or on histological examination over a 5-year period. All cases of molar pregnancy diagnosed histologically were examined and cross-referenced with cases diagnosed on ultrasound and with the supplementary report from the regional referral center. When available, maternal serum beta-human chorionic gonadotropin (hCG) levels were recorded. RESULTS: Fifty-one cases of suspected molar pregnancy were referred to the regional center for further histological opinion and follow-up, and five cases were subsequently excluded from the final analysis because of the diagnosis of hydropic abortion (HA). In 33 cases a molar pregnancy was suspected at the initial scan. Of these, 26 (78.8%) were confirmed on histology, resulting in a 56% detection rate using ultrasound alone. In 15 cases hCG results were available, of which nine were greater than two multiples of the median. CONCLUSIONS: The diagnosis of both complete (CHM) and partial (PHM) hydatidiform moles in first-trimester miscarriages is difficult. hCG is significantly higher in both CHM and PHM and, in conjunction with transvaginal ultrasound, could provide the screening test required to enable clinicians to counsel women more confidently towards non-surgical methods of management of their miscarriage, where histopathological examination is not available.
OBJECTIVE: To examine the relationship between ultrasound and histological features in the screening for molar changes in missed miscarriage. METHODS: A prospective cohort study was conducted on all missed miscarriages, with features suspicious of molar pregnancy, on transvaginal ultrasound and/or on histological examination over a 5-year period. All cases of molar pregnancy diagnosed histologically were examined and cross-referenced with cases diagnosed on ultrasound and with the supplementary report from the regional referral center. When available, maternal serum beta-human chorionic gonadotropin (hCG) levels were recorded. RESULTS: Fifty-one cases of suspected molar pregnancy were referred to the regional center for further histological opinion and follow-up, and five cases were subsequently excluded from the final analysis because of the diagnosis of hydropic abortion (HA). In 33 cases a molar pregnancy was suspected at the initial scan. Of these, 26 (78.8%) were confirmed on histology, resulting in a 56% detection rate using ultrasound alone. In 15 cases hCG results were available, of which nine were greater than two multiples of the median. CONCLUSIONS: The diagnosis of both complete (CHM) and partial (PHM) hydatidiform moles in first-trimester miscarriages is difficult. hCG is significantly higher in both CHM and PHM and, in conjunction with transvaginal ultrasound, could provide the screening test required to enable clinicians to counsel women more confidently towards non-surgical methods of management of their miscarriage, where histopathological examination is not available.