Strengths and limitations
To our knowledge, this is the first cohort study evaluating the
predictive potential of GCF-PLAP concentrations for early-pregnancy risk
assessment of PE. These data are consistent with and extend our previous
observations99. PLAP concentrations in GCF were
significantly greater than those measured in paired plasma samples. In
women who subsequently developed PE, GCF-PLAP concentrations were 3- to
6-fold greater than those measured in matched plasma samples. These
findings suggest that potential biomarkers of obstetrical diseases can
be concentrated in GCF, highlighting the opportunity to use placental
biomarkers measured in GCF to improve the performance of prediction
models.
Currently, there is considerable debate about the use multivariate
algorithms for the prediction of PE, and most obstetric societies do not
recommend their use as an screening strategy in routine obstetrical care78,88–94, arguing the lack of solid evidence of
external validation and/or randomization of many of such existing
models84,128,129. In fact, many societies only
recommend the use of maternal risk factors to identify those women who
may be at increased risk of PE. Nevertheless, this screening approach
based on identification of clinical risk factors has several
limitations, especially considering their limited detection
capabilities. When only clinical assessment of risk factors is used to
identify patients at risk of PE, a minor proportion of PE cases are
detected (approximately 30 - 40% of preterm PE and 20 to 35% of late
PE cases)130–133, with very high rates of
false-positive screening results. In contrast, uterine artery
pulsatility index with or without the combination of several maternal
risk factors, measured at 11 to 14 weeks of, increases the detection
rates of early PE (<34 weeks of gestation) to about 40% to
65% at a fixed false positive rate of 5%, and with an overall
sensitivity for early PE of nearly 50%134–152. The
most recently developed predictive algorithms, using different
combinations of maternal risk factors, biophysical variables like mean
maternal blood pressure, uterine artery pulsatility index, and maternal
plasmatic biomarkers like placental growth factor and/or
pregnancy-associated plasma protein A at 11-14 weeks of gestation, have
consistently demonstrated detection rates of preterm PE over nearly 70%
at false-positive rates of 10%153–167. The results
obtained in this study further support the use of multiparametric
algorithms for improving the prediction of PE. Future studies are needed
to confirm our results and to address the predictive capabilities of
CGF-PLAP concentration alone and/or in combination with more variables
and risk factors, such as the uterine artery Doppler value, to further
increase the performance of the algorithm.
Regarding the link between periodontal disease and the risk of PE, our
study did not confirm the association between periodontal diagnoses and
PE that had been previously described in the
literature108–112. These results, however, should be
interpreted with caution given that the present study was powered to
determine the association between GCF-PLAP concentrations and PE but not
the association between PE and periodontal diagnosis. Moreover, this
link has not been confirmed in all populations120,121.
In fact, in a cohort study conducted in 1,562 pregnant women from
Argentina122, no significant association between
periodontal disease and PE was identified. In addition, in our study,
periodontal disease was assessed during early pregnancy. It is known
that periodontal disease usually worsens during
pregnancy123–125 and that its evaluation at a later
stage of pregnancy may be more related to the development of PE.