The role of a single free beta-human chorionic gonadotrophin measurement in the diagnosis of early pregnancy failure and the prognosis of fetal viability.

Abstract

The prospective controlled study investigated the concentrations of free beta-human chorionic gonadotrophin (HCG) subunit in 554 women with a singleton intrauterine or tubal pregnancy. They presented with vaginal bleeding and/or abdominal pain in the first 18 weeks of pregnancy. The control group comprised 156 women with musculoskeletal pain and no vaginal bleeding. Their pregnancies continued to term. The study group comprised 398 women (141 threatened-continuing pregnancies, 37 threatened-miscarriages, 185 non-continuing pregnancies and 35 tubal pregnancies). Free beta-HCG concentrations were significantly lower in the non-continuing, threatened-miscarriage and tubal pregnancy groups [mean 4.62, 6.50 and 4.27 ng/ml respectively; 95% confidence interval (CI) 3.75-5.69, 4.46-9.48 and 2.92-6.2 respectively] than in the control and threatened-continuing groups (mean 41.61 and 48.22 ng/ml respectively; 95% CI 34.53-50.13 and 42.03-55.32 respectively) (P < 0.001 in all cases). A cut-off value at 20 ng/ml was found to differentiate between the 'viable' (control and threatened-continuing) and the 'abnormal' (non-continuing, threatened-miscarriage and tubal) pregnancies, with 88.3% sensitivity and 82.6% positive predictive value. An excellent diagnostic and prognostic usability of free beta HCG was confirmed by a receiver operating characteristic curve plot. In conclusion, a single serum free beta-HCG measurement taken in early pregnancy is valuable in the immediate diagnosis of early pregnancy failure and the long-term prognosis of viability.