Background
Histopathology is the reference standard for diagnosing hydatidiform molar pregnancy (HMP).

Aim
To assess the performance of pre‐operative transvaginal ultrasound (TVS) to predict HMP.

Materials and Methods
A retrospective diagnostic accuracy study was performed on women who had both TVS and histopathological examination of uterine curettings between January 2011–February 2017. TVS diagnosis of partial MP (PMP) included assessment of fetal parts and/or empty gestational sac with small cystic spaces adjacent to gestational sac. TVS diagnosis of complete MP (CMP) included assessment of complex, echogenic intra‐uterine mass(es) containing multiple small cystic areas. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), positive and negative likelihood ratios (PLR/NLR) for TVS were reported. Reference standard was histology.

Results
There were 4917 consecutive women who underwent TVS; 1636 underwent surgical curettage for miscarriage. Forty out of 4917 (0.6%) had HMP histologically; 25/40 (62.5%) had suspected HMP and 15/40 (37.5%) did not; 24/40 (60.0%) had CMP of which 19/24 (79.1%) were suspected on TVS and 5/24 (20.8%) were not; 16/40 (40.0%) had PMP of which 6/16 (37.5%) were suspected on TVS and 10/16 (62.6%) were not. The sensitivity, specificity, PPV, NPV, PLR and NLR of pre‐operative TVS to predict HMP were: 60.0, 99.1, 63.2, 99.0% 68.4, 0.4; for CMP: 79.2, 99.8, 86.3, 99.6%, 421.7, 0.2; for PMP: 37.5, 99.3, 35.3, 99.4%, 54.3, 0.6.

Conclusion
Histology remains the gold standard for diagnosing HMP. TVS is an acceptable diagnostic tool for HMP. When there are no ultrasound features of HMP, almost all these women will not have an underlying HMP.