TABLE/FIGURE CAPTION LIST
Table 1: Comparison of binary test performance of the
sFlt-1/PlGF ratio cut-off of 38 criteria and its combination with the
prognostic prediction tool using all the included population,
independently from sFlt-1/PlGF ratio value. Pregnancies with an
sFlt-1/PlGF ratio of 38 or below have been considered as negative for
the prognostic prediction tool. 95% CI: 95% confidence interval. DR:
detection rate. FPR: false positive rate. PPV: positive predictive
value. NPV: negative predictive value. PLR: positive likelihood ratio.
Figure 1: Inclusion-exclusion flow-chart.
Figure 2: Receiver operating characteristics (ROC) curve for
median risk predictions of early-onset pre-eclampsia leading to delivery
within one week in 123 women (213 samples) with suspected early-onset
pre-eclampsia and an sFlt-1/PlGF ratio above 38, using a linear mixed
model based on individual markers and gestational age (GA). Significant
differences between AUC estimates of the sFlt-1 MoM model (solid line)
and the PlGF MoM model (dashed line) were observed, referred to as A
p-value (p< 0.001). While no differences were observed between
the NT-proBNP (dotted line) and sFlt-1 MoM models, referred to as B
p-value.
Figure 3: Receiver operating characteristics (ROC) curve for
median risk predictions of early-onset pre-eclampsia leading to delivery
within one week in 123 women (213 samples) with suspected early-onset
pre-eclampsia and an sFlt-1/PlGF ratio above 38. The curves show the
predictive performance of generalized linear mixed (GLM) models that
combine gestational age (GA), and sFlt-1/PlGF ratio (left panel) and GA
and sFlt-1 MoM (right panel) with NT-proBNP (solid line) and without
NT-proBNP (dashed line). Vertical dashed red line indicates 5% false
positive rate. Addition of NT-proBNP to both models significantly
increases their predictive ability.