Predictive Value of Sequential βHCG in IVF Pregnancy

Overview

Early diagnosis of pregnancy with its localization and evolution has always been one of the major objectives of gynecology and obstetrics, even more so in Artificial Reproductive Technology (ART) centers. The pivotal test is the βHCG assay. Various protocols have been proposed over the years, including single assessment and serial assays. Several studies in the past years have tried to define a cut-off predictive of a successful pregnancy. Abnormal levels of βHCG are associated with biochemical pregnancies, non-viable pregnancies and ectopic pregnancies (EP). The efficacy of a single serum βHCG test to predict EP is low and a significant amount of time and resources are spent diagnosing it. In recent studies, better sensitivity was obtained from the ratio of two successive time points of βHCG concentration, with better specificity instead from regression models. These proposed models however lack validation and require further improvement.

Fertility clinics follow different protocols for the measurement of βHCG. Typically the initial serum measurement is performed 10-12 days after blastocyst transfer or 12-14 days after transfer at cleavage stage. Often a second serum measurement of hCG is performed at 48h since an increase of at least 50% is known to be a good predictor of ongoing pregnancy. The doubling time was first described in natural pregnancies where the rate of βHCG rise was reported to be at least 53% in two days, with a median of 50% increase at day 1 and 124% at day 2. Following the first detection and rise of serum βHCG, it is possible to predict earlier than with transvaginal ultrasound non-viable pregnancies, ectopic pregnancies, biochemical pregnancies and spontaneous abortion or reassure the couple when these values are representative of an ongoing pregnancy (OP). Conversely, however, many studies have found different thresholds of βHCG to be representative of OP, with many women under the cut-off value ending up having a normal pregnancy. The variability within the threshold expresses the need for a better biological marker or cut-off value. Implementing patient characteristics in a model to redefine and personalize the cut-off is necessary to improve pregnancy detection and management.

Conditions