Measuring early pregnancy loss: laboratory and field methods

Abstract

We intensively studied 30 women attempting pregnancy in order to lay groundwork for larger studies of early pregnancy loss. These women collected first morning urine specimens for up to 6 months after discontinuing use of birth control. Urine specimens were successfully collected for 98% of the woman-days in the study. Three assays for human chorionic gonadotropin (hCG) were performed on each urine specimen. An immunoradiometric assay (IRMA) specific to the carboxyterminal peptide of the hCG beta-chain proved to be more sensitive and more specific than two radioimmunoassays (RIAs). Using the IRMA, we found four cases in which hCG rose and fell over successive days, consistent with early pregnancy loss. For three of these four cases, the level of hCG was too low to be detectable with the RIAs. Among the control group of five women with tubal ligations, there was no detectable hCG above threshold with the IRMA. Thus, the enhanced sensitivity and specificity of the IRMA allows very early pregnancy losses to be identified that would otherwise be undetectable. Furthermore, its effectiveness with small quantities of first morning urine makes the IRMA a useful tool for epidemiologic studies.