A definition of preeclampsia that incorporates the assessment of maternal, fetal, and uteroplacental status would optimize the identification of pregnancies at risk of complications at term gestational age. This definition would include “carrying forward” angiogenic test results from 35 to 36 weeks of gestation to term gestational age. Would this approach still be useful if testing is performed earlier or at a routine midgestation scan and the result is used to inform the diagnosis of preeclampsia that developed thereafter?
This study aimed to evaluate whether fetoplacental assessment at a 19- to 23-week scan could be “carried forward” to contribute to the classification of preeclampsia and improve the detection of women and fetuses at risk of adverse outcomes associated with hypertension.
In this prospective cohort study of singleton pregnancies at 2 maternity hospitals in England (October 2011 to March 2020), women attending a routine hospital visit at 19 to 23 weeks of gestation underwent an assessment that included history, ultrasonographic estimated fetal weight, Doppler measurements of the pulsatility index in uterine arteries, and serum placental growth factor. Preeclampsia was defined according to various definitions: (1) traditional, based on new-onset proteinuria at ≥20 weeks of gestation; (2) 2013 American College of Obstetricians and Gynecologists; (3) 2018 International Society for the Study of Hypertension in Pregnancy maternal factor; (4) 2018 International Society for the Study of Hypertension in Pregnancy maternal-fetal factor (death or growth restriction), based on ultrasound scans at the 19 0/7 to 23 6/7 week of gestation (an estimated fetal weight of <3rd percentile or estimated fetal weight between the 3rd and 10th percentiles with a uterine artery pulsatility index of >95th percentile); and (5) 2021 International Society for the Study of Hypertension in Pregnancy maternal-fetal factor plus placental growth factor (with abnormal placental growth factor defined as an estimated fetal weight of <5th percentile for gestational age). The detection rates for outcomes of interest (ie, severe maternal hypertension, major maternal morbidity, perinatal mortality or major neonatal morbidity, neonatal intensive care unit admission ≥48 hours, and birthweight of <3rd percentile) ascertained by health record review were compared using the chi-square test. A P value of <.05 was considered statistically significant.
Among 40,241 singleton pregnancies, preeclampsia incidence varied by definition, from lows of 2.6% (traditional) and 3.0% (American College of Obstetricians and Gynecologists) to a high of 3.8% (International Society for the Study of Hypertension in Pregnancy maternal-fetal factor plus placental growth factor). The International Society for the Study of Hypertension in Pregnancy maternal-fetal factor plus placental growth factor definition (vs the traditional) best identified women who developed adverse outcomes: severe hypertension (detection rate: 70.6% vs 52.8%; P <.001), major maternal morbidity (detection rate: 100% vs 87.5%; P =.027), perinatal mortality or major morbidity (detection rate: 84.6% vs 69.5%; P =.004), neonatal intensive care unit admission ≥48 hours (detection rate: 76.6% vs 63.2%;, P =.0002), and birthweight of <3rd percentile (detection rate: 81.3% vs 61.9%; P <.0001]. The detection rates improved, going from the American College of Obstetricians and Gynecologists definition to the International Society for the Study of Hypertension in Pregnancy maternal-fetal factor plus placental growth factor definition, for severe hypertension (11.4%; P =.003), perinatal mortality or major morbidity (10.6%; P =.03), neonatal intensive care unit admission ≥48 hours (8.6%; P =.01), and birthweight of <3rd percentile (16.2%; P <.001). However, going from the International Society for the Study of Hypertension in Pregnancy maternal-fetal factor definition to the International Society for the Study of Hypertension in Pregnancy maternal-fetal factor plus placental growth factor definition, the detection of fetuses with a birthweight of <3rd percentile improved by 7.0% ( P =.01), but no other improvement was seen for severe hypertension (1.7%; P =.33), major maternal morbidity (0%), perinatal mortality or major morbidity (4.0%; P =.20), and neonatal intensive care unit admission ≥48 hours (3.2%; P =.17).
The criteria for uteroplacental dysfunction (including placental growth factor) from the 19- to 23-week assessment can be used in the assessment of women who are later suspected of having PE, to best identify pregnancies at risk of adverse outcomes.
Preeclampsia (PE) is the hypertensive disorder of pregnancy associated with the greatest risk of pregnancy complications. Its identification is an antenatal care priority that is the basis of blood pressure measurement at each antenatal visit.
Why was this study conducted?
This study aimed to evaluate whether ultrasound and angiogenic marker results at a 19- to 23-week ultrasound scan can be used later in pregnancy when a diagnosis of preeclampsia (PE) is suspected, to improve the detection of women and fetuses at risk of adverse outcomes.
A definition of PE that included maternal end-organ and uteroplacental dysfunctions—abnormalities in placental perfusion or function—best identified women and fetuses at increased risk of adverse outcomes, at preterm and term gestational ages. This definition was particularly true when the findings from routine second-trimester ultrasonographic assessment and placental growth factor (PlGF) were used to inform a diagnosis of PE when hypertension subsequently developed.
What does this add to what is known?
Our study data has provided further evidence for the use of a PE definition that includes maternal end-organ dysfunction (including but not restricted to proteinuria) and uteroplacental dysfunction, at any point from 20 weeks of gestation. Importantly, the data showed that 19- to 23-week results of fetoplacental ultrasound and PlGF testing can be subsequently used by clinicians to evaluate pregnant women with hypertension suspected for PE.
To date, there is international consensus that PE should be defined “broadly,” incorporating not only proteinuria but also other relevant end-organ manifestation. This approach reflects the systemic nature of the disease and its multifaceted syndrome and aims to optimize the identification of mothers and fetuses at risk of complications. Most national and international pregnancy hypertension guidelines now include relevant maternal symptoms, signs, and abnormal laboratory test results in their PE definitions. However, although the International Society for the Study of Hypertension in Pregnancy (ISSHP) endorses the inclusion of uteroplacental dysfunction—abnormalities in placental perfusion or function—in the definition of PE, the inclusion in other guidelines is variable. Interestingly, 1 guideline includes abnormal fetal heart rate or oligohydramnios; few guidelines include placental abruption, abnormal umbilical artery Doppler, or fetal death; and approximately 75% of guidelines include fetal growth restriction (FGR) in their PE definitions. In their 2021 guidance, the ISSHP has included angiogenic imbalance as a manifestation of uteroplacental dysfunction. Although 3 other guidelines recommend angiogenic marker testing either to “rule out” , or “rule in” PE when suspected, abnormalities in these markers are not part of their formal PE definitions. Of note, the American College of Obstetricians and Gynecologists (ACOG) pregnancy hypertension guidance includes none of these uteroplacental manifestations, although they are acknowledged to be associated with PE.
Previously, we have shown that a definition of PE that incorporates uteroplacental assessment optimizes the identification of women and fetuses at risk of complications at term gestational age. However, these findings were based on an evaluation, including angiogenic marker testing, at a 35- to 36-week fetal assessment that is not routine. Of note, up to one-third of PE occurred preterm and before 35 to 36 weeks of gestation. As both ultrasound and angiogenic markers are specialized tests, we evaluated whether their performance at a routine 19- to 23-week assessment could be “carried forward” to be used as diagnostic criteria for PE if hypertension subsequently developed, to optimize the identification of women and fetuses at risk of adverse outcomes associated with hypertension. We hypothesized that ultrasonographic and angiogenic marker results from 19 to 23 weeks of gestation when “carried forward” would increase the detection rate of subsequent preterm or term PE and better identify women and fetuses at risk of adverse outcomes.
Study design and participants
The study data were derived from a prospective screening study for adverse obstetrical outcomes in women attending routine pregnancy care at 19 0/7 to 23 6/7 weeks of gestation at King’s College Hospital and Medway Maritime Hospital in the United Kingdom between October 2011 and March 2020. The women gave written informed consent to participate in the study, which was approved by the National Health Service Research Ethics Committee.
In brief, at the routine hospital visit at 19 0/7 to 23 6/7 weeks of gestation, women underwent (1) recording of maternal demographic characteristics and medical history, (2) an ultrasound examination for fetal anatomy and growth, (3) measurement of the mean of the left and right uterine artery pulsatility indices (PIs) using transvaginal or transabdominal color Doppler ultrasound, , and (4) measurement of the mean arterial pressure using validated automated devices and a standardized protocol. The fetal head circumference, abdominal circumference, and femur length were measured, and the estimated fetal weight (EFW) was calculated using the Hadlock formula, identified as the most accurate model by systematic review. Serum placental growth factor (PlGF) was measured by BRAHMS Kryptor compact PLUS (Thermo Fisher Scientific, Hennigsdorf, Germany), DELFIA Xpress system (PerkinElmer Life and Analytical Sciences, Waltham, MA), or Cobas e411 (Roche Diagnostics, Penzberg, Germany). Gestational age was determined by the measurement of the fetal crown-rump length at 11 to 13 weeks of gestation or the fetal head circumference at 19 to 24 weeks of gestation. ,
Data related to pregnancy outcomes were abstracted from electronic hospital maternity records or those of the women’s general medical practitioners. The obstetrical records of all women with chronic hypertension or pregnancy-associated hypertension were examined to determine the diagnosis of PE or gestational hypertension (GH).
The diagnosis of GH was based on the finding of new-onset hypertension developing after 20 weeks of gestation in previously normotensive women. The diagnosis of chronic hypertension was based on a prepregnancy history of such hypertension at <20 weeks of gestation. Hypertension was defined as a systolic blood pressure (BP) of ≥140 mm Hg or diastolic BP of ≥90 mm Hg, on at least 2 occasions, at least 4 hours apart.
The inclusion criteria for this analysis were singleton pregnancies delivering a nonmalformed live birth or stillbirth at ≥24 0/7 weeks of gestation. We excluded pregnancies with aneuploidy or major fetal abnormalities.
Diagnosis of preeclampsia
Here, 5 different definitions of PE were considered (Supplemental Table). The traditional definition of PE was based on new-onset proteinuria (ie, ≥300 mg per 24 hours, protein-to-creatinine ratio of ≥30 mg/mmol, or ≥2+ on urinary dipstick testing).
The broader definitions of PE, from both the ACOG and ISSHP definitions, , were based on evidence of maternal end-organ dysfunction (ACOG and ISSHP definitions) or uteroplacental dysfunction (ISSHP definition only). Neither broad definition requires the presence of proteinuria. In defining maternal end-organ dysfunction, we included only quantitative measures of renal, hepatic, or hematologic dysfunction, reliably documented in clinical care notes. The ACOG definition of PE was based on the development of at least 1 of the following: new-onset proteinuria, serum creatinine of >97 μmol/L (in the absence of underlying renal disease), serum transaminases more than twice the upper limit of normal (.e, ≥65 IU/L for our laboratory), platelet count of <100,000/μL, headache or visual symptoms, or pulmonary edema. The ISSHP definition of PE was examined according to its maternal (ISSHP-M) and uteroplacental (ISSHP-MF) components. The ISSHP-M definition was based on at least 1 of the following: new-onset proteinuria, serum creatinine of ≥90 μmol/L (in the absence of underlying renal disease), serum transaminases of >40 IU/L, platelet count of <150,000/μL, or neurologic complications (ie, altered mental status, blindness, stroke, clonus, severe headaches, or persistent visual scotomata); the criteria of altered mental status and clonus were not available. The ISSHP-MF definition included all criteria of ISSHP-M, with the addition of fetal death or FGR; FGR was defined according to ultrasound findings at 19 0/7 to 23 6/7 weeks of gestation, defined as an EFW of <3rd percentile or an EFW between the 3rd and 10th percentiles together with the uterine artery PI multiple of the median (MoM) of >95th percentile. The ISSHP-MF+PlGF definition included all criteria of the ISSHP-MF definition, with the addition of low serum PlGF MoM of <5th percentile at 19 to 23 weeks of gestation.
The maternal and perinatal outcomes of interest were as follows: severe maternal hypertension, a composite of maternal death or major morbidity, a composite of perinatal death or major neonatal morbidity, neonatal intensive care unit (NICU) admission ≥48 hours, and birthweight (BW) of <3rd percentile.
Severe maternal hypertension was defined as systolic BP ≥160 mm Hg and/or diastolic BP ≥110 mm Hg. Major maternal morbidity was defined as one or more of the following: eclampsia, blindness, stroke, myocardial ischemia, pulmonary edema, elevated liver enzymes, hepatic hematoma, low platelet count, or acute kidney injury; these were based on the core maternal outcome set in PE, except for outcomes that were not available (ie, liver rupture, postpartum hemorrhage, ICU admission, and intubation and ventilation other than for childbirth), exclusion of placental abruption (which was defined clinically and underreported), and addition of myocardial ischemia (based on the Delphi-derived Preeclampsia Integrated Estimate of Risk score , ).
Perinatal death was defined as stillbirth or neonatal death prior to hospital discharge. Major neonatal morbidity was defined as one or more of the following, as indicated in the BadgerNet Neonatal discharge summary: ventilation (ie, need for continuous positive airway pressure or nasal continuous positive airway pressure or intubation), respiratory distress syndrome (ie, the need for surfactant and ventilation), brain injury (ie, hypoxic-ischemic encephalopathy, intraventricular hemorrhage grade ≥2, or periventricular leukomalacia), sepsis (based on positive blood cultures), anemia treated with blood transfusion, or necrotizing enterocolitis requiring surgical intervention. The BW percentile for gestational age was determined using the Fetal Medicine Foundation fetal and neonatal weight charts. Perinatal outcomes covered the core perinatal outcome set in PE, except for neonatal seizures.
Descriptive analysis was undertaken for (1) baseline data of assessment at 19 0/7 to 23 6/7 weeks of gestation and subsequent pregnancy outcomes for the study population overall, (2) the contribution of the components contributing to a diagnosis of PE among relevant women, and (3) pregnancy outcomes according to different definitions of PE (and the related impact on GH). Continuous variables were summarized by median and interquartile range, and categorical variables were summarized by number (percentage). The chi-squared test was used to compare the detection rate for adverse maternal and perinatal outcomes for each of the broad definitions of PE, relative to the traditional one; this was undertaken for PE overall and preterm (delivery at <37 weeks of gestation) and term PE. A P value of <.05 was considered statistically significant.
There were 40,241 pregnancies evaluated at visits at 19 0/7 to 23 6/7 weeks of gestation and included in this analysis.
Table 1 shows that, on average, women were in their early 30s and overweight, with 25% of women considered obese. Most women were White, with a substantial minority of Black race. Few women (<10%) were cigarette smokers. Medical history was usually unremarkable, with few women reporting chronic hypertension (most of whom were treated with antihypertensive therapy), pregestational diabetes mellitus, or rheumatic disease. Just over half of women were parous, with 2.5% of parous women having had previous PE. Few women reported that their mothers had suffered from PE. Almost all conceptions were natural, following an interpregnancy interval of just under 3 years, when relevant. Few women were on aspirin for PE prevention, the prescription of which was guided entirely by routine clinical care.
|Age (y)||31.9 (27.9–35.5)|
|Body mass index (kg/m 2 )||26.2 (23.5–30.0)|
|South Asian||1923 (4.8)|
|East Asian||784 (1.9)|
|Cigarette smoker||3016 (7.5)|
|Chronic hypertension||425 (1.1)|
|On antihypertensive medication||354 (83.3)|
|SLE or antiphospholipid antibody syndrome||85 (0.2)|
|Diabetes mellitus (type 1 or 2)||354 (0.9)|
|Parous without previous PE||20,300 (50.4)|
|Parous with previous PE||987 (2.5)|
|Mother had PE||1451 (3.6)|
|Interpregnancy interval (y)||2.7 (1.7–4.7)|
|Assisted by use of ovulation drugs||295 (0.7)|
|In vitro fertilization||1513 (3.8)|
|On aspirin for PE prevention||1339 (3.3)|
|Gestational age at screening (wk)||21.6 (21.1–22.0)|
|Screening markers for PE at 19 0/7 to 23 6/7 wk|
|Mean arterial pressure (mm Hg)||85.7 (80.8–90.7)|
|Uterine artery PI MoM||1.0 (0.8–1.2)|
|Uterine artery PI MoM>95th percentile||2022 (5.0)|
|PlGF MoM||1.0 (0.7–1.4)|
|PlGF MoM<5th percentile||2112 (5.2)|
|Gestational age at birth (wk)||39.9 (39.0–40.7)|
|Preterm birth||2319 (5.8)|
|Induction of labor||8244 (20.5)|
|Vaginal delivery||29,020 (72.1)|
|Spontaneous vaginal delivery||23,128 (57.4)|
|Cesarean delivery||11,221 (27.9)|
|Perinatal mortality or major morbidity a||983 (2.4)|
|Neonatal death||18 (0.04)|
|Brain injury||48 (0.1)|
|Neonatal intensive care unit admission≥48 h||2662 (6.6)|
|BW<3rd percentile b||1923 (4.8)|
a Major neonatal morbidity was defined as one or more of the following: ventilation, RDS, brain injury, sepsis, anemia, or NIC
b The BW percentile for gestational age was determined using the Fetal Medicine Foundation fetal and neonatal weight charts.
The 19- to 23-week assessment occurred at a median of 21.6 weeks; at which point, approximately 5% of women had abnormal uterine artery Doppler readings or abnormal PlGF ( Table 1 ).
Birth occurred at a median of approximately 40 weeks, following induction for approximately 20% of women and by cesarean delivery for almost 30% of women ( Table 1 ). Perinatal mortality was mainly because of stillbirth, and neonatal morbidity was because of respiratory problems, for which just under 7% of neonates overall required prolonged NICU admission. Just under 5% of neonates were born with a BW of <3rd percentile.
Table 2 presents the elements of the PE definitions, for women with new-onset hypertension (n=2188) or chronic hypertension (n=425). Most commonly, women with hypertension satisfied the maternal diagnostic criteria for PE based on proteinuria, particularly among women with new-onset hypertension (44.7%). Less often, abnormal routine laboratory tests defined PE in women with hypertension, particularly related to low platelet count of <150×10 9 /L (10.6% of women) or elevated liver enzymes (11.4% of women). Almost 10% of women with new-onset hypertension had uteroplacental dysfunction based on low PlGF at 19 0/7 to 23 6/7 weeks of gestation, with about half that rate (approximately 5%) among women with chronic hypertension. More women satisfied the more liberal ISSHP (vs more conservative ACOG) criteria for thrombocytopenia (231 [10.6%] vs 38 [1.7%], respectively) and elevated liver enzymes (249 [11.4%] vs 141 [6.4%], respectively). Few women met the criteria for PE based on gestational or chronic hypertension with headache (77 [3.5%]) or visual symptoms (5 [1.2%]).
|Characteristic||New-onset hypertension (n=2188)||Chronic hypertension (n=425)|
|Proteinuria a||979 (44.7)||81 (19.1)|
|Maternal symptoms b|
|Headache||77 (3.5)||5 (1.2)|
|Visual symptoms||42 (1.9)||2 (0.5)|
|Maternal signs c|
|Eclampsia||13 (0.6)||0 (0)|
|Myocardial ischemia||0 (0)||0 (0)|
|Pulmonary edema||4 (0.2)||0 (0)|
|Abnormal maternal laboratory tests d|
|Platelet count<150×10 9 /L||231 (10.6)||41 (9.6)|
|Platelet count<100×10 9 /L||38 (1.7)||2 (0.5)|
|Serum creatinine≥90 μmol/L||79 (3.6)||28 (6.6)|
|Serum creatinine>97 μmol/L||51 (2.3)||19 (4.5)|
|AST or ALT of >40 IU/L||249 (11.4)||45 (10.6)|
|AST or ALT of ≥65 IU/L||141 (6.4)||17 (4.0)|
|Intrauterine fetal death||10 (0.5)||3 (0.7)|
|EFW<3rd percentile||58 (2.7)||16 (3.8)|
|EFW between the 3rd and 10th percentiles with uterine artery PI MoM>95th percentile||32 (1.5)||3 (0.7)|
|PlGF MoM <5th percentile||201 (9.2)||22 (5.2)|