Evidence for the Triage PlGF Test:
The PELICAN study was a prospective, multicentre, observational study
investigating the diagnostic accuracy of PlGF in diagnosing
pre-eclampsia.12 This study showed that low PlGF had
high sensitivity (0.96; 95% CI 0.89 – 0.99) and high negative
predictive value (NPV 0.98; 0.93 – 0.995) in diagnosing pre-eclampsia
necessitating delivery within 14 days, in women with suspected
pre-eclampsia before 35 weeks’ gestation.12 The area
under the receiver operating characteristic curve (AUC) for PlGF was
greater than all other commonly used tests (systolic blood pressure,
diastolic blood pressure, alanine transaminase, urate, dipstick
proteinuria) for making a diagnosis in women presenting with suspected
pre-eclampsia. Low PlGF was classified as
<5th centile according to predetermined
normal ranges.9 However, for implementation in
clinical practice a threshold of <100pg/ml was identified,
independent of gestation, with test performance similar to the
5th centile and retaining the same high sensitivity
and NPV. When multiples of median were used for analysis, this reduced
the predictive power of low PlGF (unpublished data). The study was
conducted across seven sites in the United Kingdom and recruited 625
women; 346 developed pre-eclampsia, of whom 176 developed pre-eclampsia
prior to 35 weeks’ gestation. Therefore, this demonstrates high test
performance across a broad range of clinical settings and population
demographics.
PlGF testing has been investigated in the PETRA trial, a prospective
observational study recruiting 1,112 women from 24 centres across North
America.14 This demonstrated that low PlGF
concentration was strongly correlated with time to delivery in women
with suspected pre-eclampsia before 35 weeks’ gestation (753 women). The
median time to delivery was 10 days for low PlGF (100pg/ml or lower) and
2 days for very low PlGF (less than 12pg/ml). The sensitivity for
diagnosing pre-eclampsia and delivery within 14 days was 92.5% and the
NPV was 90.3%. This lower NPV may reflect the higher prevalence of
pre-eclampsia in the study population, with 71% of participants (538
women) diagnosed with preterm pre-eclampsia. A secondary analysis of
this study has also been reported.15 This found a
significant association between low and very low PlGF and composites of
perinatal and maternal adverse outcomes. The sensitivity and NPV of low
PlGF for adverse neonatal outcomes were 95.8% and 99.2%, and for
adverse maternal outcomes 86.8% and 98.1%. They conclude that women
with abnormal PlGF are significantly more likely to suffer adverse
neonatal and maternal outcomes, and that PlGF is useful for risk
stratification.
PlGF testing has been evaluated in the PARROT trial, a multi-centre
stepped-wedge cluster-randomised controlled trial.16The trial design involved random and sequential transition of maternity
units (representing clusters) from concealed to revealed PlGF testing,
alongside a simple clinical management algorithm. 1023 women with
suspected preterm pre-eclampsia were enrolled from 11 maternity units
across the UK. The time to diagnosis was reduced from 4.1 to 1.9 days,
with a significant reduction in a composite of severe maternal adverse
outcomes from 5.4% to 3.8% (aOR 0.32, 95% CI 0.11 – 0.96). There was
no difference in gestational age at delivery or perinatal adverse
outcomes. The study strengths include its broad inclusion criteria and a
population who were diverse from an ethnicity and sociodemographic
perspective, from multiple sites, thereby enhancing generalisability of
the findings to the real-world setting. Therefore, this study provided
novel evidence that PlGF testing proves a valuable diagnostic adjunct in
suspected pre-eclampsia, allowing targeted surveillance for those at
highest risk of adverse outcomes.
Initial studies using economic modelling demonstrated that PlGF-based
testing may afford a cost-saving of between £330 and £1032 per woman
tested.17, 18 In a more recent study describing their
cost-effectiveness analysis of the PARROT trial, Duhig and colleagues
found a total cost-saving of £149 per woman, based on £70 per Triage
PlGF test.19 Given that there were 646,794 births in
England in 2017 and 10% of pregnant women have suspected pre-eclampsia,
with 30% of these presenting prior to 37 weeks’ gestation, PlGF testing
could be performed in approximately 38,800 women per year. This
represents a potential cost saving of £2,891,196 each year in England.
These cost-savings are driven by a reduction in outpatient attendances
for those with a normal, low risk result. There was an increased cost
seen associated with inpatient admissions in those at higher risk.
Despite this, overall PlGF testing was found to be cost saving, but the
cost-effectiveness will depend on the cost of the test being performed.
The authors conclude that their more conservative cost-saving is due to
improved clinical risk stratification leading to appropriate
redistribution of resources, rather than an overall reduction of
resource use anticipated by hypothetical analyses.19