It is with interest that I read the recent paper of Gallo et al (from the Kypros Nicolaides group at the Harris Birthright Research Centre for Fetal Medicine [HBRCFM], King’s College, London), in your journal. I have great respect for the work that this group has done regarding the first trimester in the prediction of early onset preeclampsia (PE).

That being said, with regard to the midtrimester prediction of PE, I have some concerns.

The HBRCFM group (Yu et al) has had success by defining an abnormal uteroplacental circulation using the mean pulsatility index (PI) but data from midtrimester uteroplacental Doppler flow studies, including ours, found that looking at the more abnormal uterine artery flow velocity waveforms (UAFVW) was of most value in predicting PE.

The placenta is less localized in the first trimester than in the second trimester and the UAFVW from the nonplacental side, in the second trimester, tends to be more abnormal. Thus using mean (or “true” ) PI may have less predictive ability than that from the side with the most abnormal UAFVW.

The uteroplacental circulation continues to evolve over the midtrimester with the early diastolic notch in the UAFVW being commonly found at 14 weeks but uncommon at 24 weeks in normal pregnancy ; and the presence of a notch in the midtrimester is useful in predicting PE.

Most have found that in the midtrimester prediction of PE either resistance index or systolic/diastolic ratio were of more value than PI.

Of interest is that the recently published UTOPIA study confirmed that using mean PI from midtrimester UAFVW was poor at predicting late-onset PE (≥34 weeks’ gestation) with a sensitivity of only 38.4%.

The HBRCFM group use “PI” when they are really referring to mean PI, which can become confusing.

I hope that these comments are more clarifying than critical and may allow for the better ability to predict PE in the midtrimester.