The use of novel biochemical markers in predicting spontaneously resolving 'pregnancies of unknown location'

Bckground: 'Pregnancies of unknown location' (PULs) include viable and failing intrauterine and extrauterine pregnancies. The aim of this study was to evaluate the role of novel biochemical markers in the prediction of spontaneous resolution of PULs. Methods: Serum samples were taken at the first visit to the pregnancy unit for measuring the traditional markers beta-hCG and progesterone, and for inhibin A, inhibin pro-alpha C-related immunoreactivity (inhibin pro-alpha C-RI) and insulin-like growth factor-binding protein 1 (IGFBP-1). Follow-up was continued until the pregnancy had resolved, the location of the pregnancy and viability was determined or treatment was required. Outcomes were dichotomized into 'spontaneous resolution' and 'other outcome' categories. Results: One-hundred and nine cases of PUL were included in the data analysis. Spontaneous resolution occurred in 70% and a further scan was required in 30% to reach a diagnosis. Levels of progesterone and inhibin A were significantly lower (both P < 0.001) and levels of IGFBP-1 significantly higher (P = 0.02) in the pregnancies that spontaneously resolved than in those pregnancies that required further intervention. In decision tree analysis, the novel markers were less useful than progesterone and beta-hCG in predicting spontaneously resolving PULs. Inhibin pro-alpha C-RI and IGFBP-1 were not useful in the prediction of spontaneously resolving PULs. Inhibin A is more predictive than beta-hCG alone, but serum progesterone is the best single marker and progesterone and hCG together continues to be the best way of predicting spontaneously resolving PULs. Conclusions: These novel biochemical markers are not clinically useful in predicting spontaneously resolving PULs.