Partial mole and triploidy: screening patients with first-trimester spontaneous abortion.

OBJECTIVE

To evaluate the role of histopathology in the categorization of women at risk of trophoblastic disease after early pregnancy failure associated with triploidy.

METHOD

A retrospective study of histopathologic findings on 587 first-trimester spontaneous abortions for which both histologic and karyotype results were available. The incidence of chromosomal abnormalities and placental molar changes and the proportion of agreement for histologic diagnosis of triploidy were calculated.

RESULTS

An abnormal chromosome complement was found in 241 (41.1%) cases, including 75 (31.1%) with trisomy, 71 (29.5%) with triploidy, 60 (24.9%) with monosomy X, and 35 with other abnormalities. Molar transformations were found macroscopically in 20 triploidies, in six spontaneous abortions with a normal karyotype, in one trisomy, in one monosomy X, and in one tetraploidy. There was one complete hydatidiform mole. Complete agreement between two investigators was seen in 48 (67.6%) triploidy cases. Inter- and intra-observer degree of agreement for histologic diagnosis of triploidy was good to excellent. The sensitivity of histology ranged between 87.3 and 94.4%, the specificity between 81.7 and 85.9%, the positive predictive value between 83.1 and 86.1%, and the negative predictive value between 86.8 and 93.8%.

CONCLUSION

Triploidy is associated with molar changes less often in the first trimester of pregnancy than in the second or third trimester. Therefore, most triploid spontaneous abortions escape detection on the basis of ultrasound or macroscopic examination. The use of standardized criteria for detection by microscopic examination is both accurate and reproducible and should play a pivotal role in screening for women at risk of persistent gestational trophoblastic disease.