Urine β-core fragment, a potential screening test for ectopic pregnancy and spontaneous abortion

The incidence of ectopic pregnancy in the United States has risen 6-fold in the last three decades. It now accounts for about 2% of reported pregnancies. Tests are now needed to identify ectopic pregnancy before it is clinically evident. We evaluated human chorionic gonadotropin beta-core fragment as a test to predict ectopic pregnancy and spontaneous abortion. Urine samples were collected from women with in vitro fertilized pregnancies, 2 1/2-5 weeks after embryo transfer. Fifty samples were collected from those later shown to have normal intrauterine pregnancies, samples from 13 women subsequently found (at 5-9.3 weeks) to have ectopic gestations, and 15 from those with impending spontaneous abortion. beta-Core fragment levels were determined by immunoassay, and results normalized to creatinine concentration. Median beta-core fragment levels at 2 1/2-3, 3-4, and 4-5 weeks after embryo transfer, were 6.7, 91 and 737 mu g/g for unaffected pregnancies, 1.0, 5.9 and 0.6 mu g/g for impending ectopic pregnancies (0.15, 0.065 and 0.0008, multiples of the unaffected pregnancy median, MoM), and 0.75, 6.8 and 12 mu g/g for impending spontaneous abortions (0.11, 0.07 and 0.016 MoM). A gestation-linked curve was modeled to discriminate unaffected pregnancy from impending ectopic gestation or spontaneous abortion. Plotted beta-core fragment levels were below this curve in 12 of 13 (92%) women with impending ectopic pregnancy, in 10 of 15 (67%) with spontaneous abortion outcome, and in 2 of 50 (4%) with intrauterine pregnancy and term outcome. Measurement of urine beta-core fragment at 2 1/2-5 weeks after embryo transfer (4 1/2-7 weeks after last menstrual period) might be useful for identifying failing pregnancies. Over three quarters (predictive value positive 76%) of those with low beta-core fragment levels have ectopic pregnancy or spontaneous abortion. On the contrary, 95% (predictive value negative) of those with normal range test values may be predicted to have a nonfailing term pregnancy. Diagnosis of ectopic pregnancy could be confirmed by transvaginal ultrasound, and ectopic pregnancy terminated early by nonsurgical methods, with minimal mortality or fertility loss. Major fetal defects that cause spontaneous abort pregnancies may also be recognized by transvaginal ultrasound. In such cases, chorionic villous sampling or possibly termination may be considered.