Introduction
Ectopic pregnancy (EP) occurs when fertilized ovum locate outside the uterine cavity, and its incidence rate is about 1.5% to 2% in pregnancy1. The death toll of ectopic pregnancy can account for more than 6% of maternal mortality, which is the leading cause of maternal death in early pregnancy2, 3. According to a research in Scotland4, nearly half of the patients had to visit the clinicians more than three times to be diagnosed for ectopic pregnancy. Moreover, it is estimated that 136, 400 pounds sterling has been spend yearly on the diagnosis of EP in pregnancy of unknown location(PUL). PUL refers to pregnancy test positive, but the first transvaginal ultrasound (TVS) has not found evidence of pregnancy in both intrauterine and extrauterine, accounting for about 5% to 42% of all early pregnant women5. Among them, ectopic pregnancy, accounted for 6% to 20% of all PUL patients,required surgery or chemotherapy and had relatively higher mortality comparing with the rest of PUL patients.
The elevated serum hCG and progesterone are the most widely used indications for early pregnancy. Some protocols use a single biomarker to distinguish patients with different outcomes by setting certain threshold, while others use established mathematical formulas. In European countries and regions, the logistic regression model M4, established by George Condous et al6, is now the most widely used formula for predicting the outcomes of PUL. But the sensitivity for EP was too low for the model to be used in clinical practice in the US7. This suggests that different definitions of pathology and disease management processes may lead to distinct accuracy of prediction protocols in specific regions. Thus, there is no international consensus on how to manage women with a pregnancy of PUL. In this study, we will plot the SROC curve and calculate the area under the curve(AUC)by meta-analysis of studies, jointly contrasting the sensitivity and specificity of individual protocols. We will introduce the concepts of average production utility and marginal benefit assessing the feasibility of individual protocols, and attempt to describe the impact of the complexity of the protocols.