First trimester SHARP1 and second-trimester uterine artery Doppler to predict preeclampsia

Abstract Objectives The objective of this study was to identify the predictive value of the first-trimester serum SHARP1 level and the second-trimester uterine artery Doppler in singleton pregnancy for the prediction of preeclampsia. Methods A prospective study including singleton pregnancy presenting at an antenatal clinic, King Chulalongkorn Memorial Hospital, Department of Obstetrics and Gynecology, Faculty of Medicine, Chulalongkorn University from 2019–March 2020 was conducted. Serum SHARP1 was collected at the gestational age (GA) of 11–13+6 weeks, and transabdominal uterine artery Doppler ultrasound was performed at GA of 18–24 weeks. Serum SHARP1 level and uterine artery pulsatility index (PI) were combined to calculate the predictive value for preeclampsia detection. Results 288 pregnant women were enrolled in the first trimester, but only 249 participants completed the study. Thirteen patients had preeclampsia (5.2%), which three cases (1.2%) had early-onset preeclampsia. The median serum SHARP1 level in the first trimester of pregnant women with preeclampsia was lower than the normal pregnancy group (1392 pg/ml vs. 1941 pg/ml, p = 0.046). The second-trimester uterine artery PI and prevalence of early diastolic notching were higher in the preeclampsia group than in the normal pregnancy group (p = 0.029 and p = 0.001, respectively). When the first-trimester serum SHARP1 level is combined with the second-trimester uterine artery PI, the sensitivity, specificity, PPV, and NPV for preeclampsia prediction were 84.6%, 47.5%, 8.2%, and 98.3%, respectively. Conclusions This study demonstrated that serum SHARP1 level in the first trimester combined with the uterine artery PI in the second trimester had good sensitivity to predict preeclampsia.


Introduction
Preeclampsia, one of the most common obstetrical complications, is a hypertensive disorder with proteinuria that results in a high healthcare burden.The global incidence is 2-8% and leads to a high mortality rate and comprises 10-15% of deaths from all causes of obstetrical mortality [1].It also affects both maternal and neonatal morbidity and mortality [2].The annual incidence of preeclampsia at our institute in the past five years was 4.9% [3].
The pathogenesis of preeclampsia has been widely studied.One of the most accepted mechanisms is an abnormal invasion of trophoblastic cells to the uterine spiral vessels [4].In a normal pregnancy, fetal trophoblastic cells invade the vascular endothelial cells to increase the diameter of uterine spiral arteries to increase blood flow to the placenta.However, in preeclampsia, this invasion occurs incompletely and leads to an impairment of placental blood flow.Various inflammatory mediators are produced in response to placental ischemia and hypoxia and cause endothelial dysfunction and systemic inflammation [2,4].
Previous studies have been conducted to identify predictors for early preeclampsia detection, using clinical history, maternal factors, serum biomarkers, and Doppler ultrasound [5][6][7].In 2012, Tal found that the level of hypoxia-inducible factor (HIF)-1a increases in various kinds of ischemic tissue, including the placenta [8].HIF-1a production, in response to placental ischemic, induces the production of transforming growth factor-b3, which inhibits the trophoblast invasion [9].
Accordingly, this mechanism worsens the degree of placental ischemia and hypoxia.From these findings, HIF-1a might be used as a screening biomarker to predict preeclampsia among pregnant women.Since HIF-1a is affected by many factors that limit its specificity, serum enhancer-of-split and hairy-related protein 1 (SHARP1), which plays an important role in promoting the degradation of HIF-1a, has been introduced to detect the occurrence of preeclampsia with better sensitivity and specificity than HIF-1a [10][11][12][13].
SHARP1, known as the basic helix-loop-helix family, member E41 (BHLHE41) and differentially expressed in chondrocytes 2 (DEC2), is produced from the Bhlhe41 gene on chromosome 12p.It is the adaptive protein that is found in many human tissues in response to tissue hypoxia.Montagner et al. in 2012, found that SHAPR1 promotes the degradation of HIF-1a.Unlike HIF-1a, SHARP1 works in both normal and low oxygen levels and does not depend on von Hippel-Lindau protein (pVHL) or ubiquitination pathways [11].Ethersoy et al. in 2016, found that the SHARP1 level was significantly lower in pregnant women with preeclampsia than in normal pregnant women [14].
Previously, there was a study using a combination of the serum SHARP1 level and the uterine artery Doppler in singleton pregnancy at 11-13 þ6 weeks of gestation to predict preeclampsia [15].Uterine artery Doppler in the second trimester has more accuracy for the prediction of preeclampsia than in the first trimester [16].Therefore, this study aimed to identify the predictive value of the first-trimester serum SHARP1 levels and the second-trimester uterine artery Doppler ultrasound in singleton pregnancy for predicting preeclampsia.

Materials and methods
A prospective observational study was conducted.After the Research Ethics Committee's approval, pregnant women visiting the antenatal clinic at King Chulalongkorn Memorial Hospital, Department of Obstetrics and Gynecology, Faculty of Medicine, Chulalongkorn University from March 2019-March 2020 were recruited.The inclusion criteria were singleton pregnant women between the age of 20-45 years old and gestational age 11-13 þ6 weeks.The pregnant women who received aspirin or any anticoagulant drugs before recruitment or had fetal abnormality (chromosomal or structural) were excluded.
The definition of preeclampsia is a blood pressure of at least 140/90 mmHg after 20 weeks of gestation, measured on two occasions at least 4 h apart, with proteinuria of at least 300 mg/24 h or at least 2þ on urine dipstick test or UPCI !0.3 or in the absence of proteinuria, new-onset hypertension with the new onset of any of the following: thrombocytopenia, renal insufficiency, impaired liver function, pulmonary edema or new-onset headache unresponsive to medication and not accounted for by alternative diagnoses or visual symptoms [17].Fetal growth restriction was defined as fetuses with an estimated fetal weight that was less than the 10th percentile for gestational age [18].
The primary outcome of this study was to identify the predictive value of the first-trimester serum SHARP1 level and second-trimester uterine artery Doppler, to predict preeclampsia in singleton pregnancy.The secondary outcome was to identify the sensitivity and specificity of using either first-trimester serum SHARP1 levels or second-trimester uterine artery Doppler alone in the prediction of preeclampsia.Additionally, the association of first-trimester serum SHARP1 level and second-trimester uterine artery Doppler for other pregnancy complications, such as preterm delivery, fetal growth restriction, and gestational diabetes were evaluated.

Sample collection of serum SHARP1
At 11-13 þ6 weeks of gestation, blood pressure was measured by validated automated devices after at least 5 min of rest in the seated position.Maternal weight before pregnancy and height were measured to calculate the body mass index (BMI).Then, blood samples were collected by venipuncture into non-heparinized tubes before centrifugation at 2500 rpm for 10 min and stored at À80 C until assayed.According to the manufacturer's recommendation, an enzymelinked immunosorbent assay (ELISA) kit for the human basic helix loop helix family (SHARP1) (Shanghai Korain Biotech Co, Ltd, Shanghai, China), a sandwich enzyme immunoassay was used to measure maternal serum SHARP1 levels.With a 450 nm wavelength absorbance and a microplate reader, the degree of enzymatic turnover was measured.The required concentration of SHARP1 in the assay is 0.029 ng/ml, with the ranges of intra-and inter-assay smaller than 10%.

Uterine artery Doppler evaluation
At 18-24 weeks of gestation, the ultrasonographic machines (GE Voluson E10 GE Medical Systems, Zipf, Austria) with a convex abdominal probe AB 2-7 MHZ were used to obtain uterine artery flow velocity waveforms.The uterine artery was visualized by placing the probe longitudinally and medially at the lower lateral quadrant of the abdomen.The color Doppler was used to identify the uterine artery when it crossed the external iliac artery.The place to collect the sample volumes was at the level of the crossing-over.The angle of insonation between the Doppler waveform and the vessel was maintained close to 0 degrees (<30 degrees) with a peak systolic velocity of more than 60 cm/s [19].At least three flow velocity waveforms from each uterine artery were recorded.Pulsatility index (PI) was measured, and the mean PI was calculated.The mean PI that was above the 95th percentile for each gestation was defined as an abnormal uterine artery Doppler.

Statistical analysis
For statistical analysis, SPSS version 22.0 was used and the data was presented with mean ± standard deviation (SD), and median (interquartile range, IQR).Chisquare test was used for comparing categorical data, an independent t-test, and a Mann-Whitney U test for comparing continuous data.The 95th percentile uterine artery PI was calculated according to gestational age.The cutoff value of serum SHARP1 levels was calculated using the receiver operating characteristic curve.p-value < 0.05 was considered statistically significant.

Results
A total of 288 pregnant women were recruited for the study.Of that, 39 were excluded due to loss of followup in the second trimester (n ¼ 30) and pregnancy termination (7 cases with abortion/dead fetus in utero, 1 fetus with trisomy 21, and 1 fetus with congenital pulmonary airway malformation and hydrops fetalis).Finally, 249 patients completed the study.Thirteen cases developed preeclampsia (5.2%), three of whom (1.2%) were patients with early-onset preeclampsia.Pregnant women with preeclampsia had higher mean arterial pressure and more total weight gain than normal pregnant women.Pregnant women with preeclampsia had a significantly higher rate of preterm delivery than normal pregnant women (Table 1).Cesarean section was high in both groups due to cephalopelvic disproportion and previous cesarean section.
Median serum SHARP1 level in the first trimester of pregnant women with preeclampsia was lower than the normal pregnancy group (1392 pg/ml vs 1941 pg/ml, p ¼ 0.046).There was a significant difference in terms of the mean uterine artery PI between women with and without preeclampsia (1.26 vs 0.97, p ¼ 0.029).Detection of the early diastolic notch was more common in the preeclampsia group with a prevalence of 38.5% in the preeclampsia group and 5.9% in the normal pregnancy group (p ¼ 0.001) (Table 2).
The optimum cutoff value for serum SHARP1 level at the time of measurement was 1.0 multiple of the median (MoM), according to the receiver operating characteristic curve (AUC 0.673; p ¼ 0.036; 95%CI 0.564-0.783)(Figure 1).The sensitivity and specificity of serum SHARP1 levels during the first trimester for predicting the development of preeclampsia were 84.6% and 50%, respectively.AUC was 0.673 (95%CI 0.564-0.783).
Multivariate analysis with the addition of blood pressure, weight gain, and uterine artery notching, which were significantly different in the univariate analysis, was performed.Only uterine artery notching was significantly different.Thus, combining uterine artery notching to examine prediction accuracy (Table 3).The sensitivity and specificity were 92.3%, 44.9%,and AUC was 0.686 (95%CI 0.567-0.806).
An abnormal value of first-trimester serum SHARP1 level and an abnormal second trimester uterine artery Doppler were not associated with fetal growth restriction, preterm delivery, and gestational diabetes (Table 4).

Discussion
Preeclampsia is one of the most common complications in pregnancy, which causes maternal and neonatal mortality and morbidity.Many studies have tried to identify markers to predict preeclampsia in the early trimester.This study demonstrated that the combination of the first-trimester serum SHARP1 levels with the second-trimester uterine artery PI had good sensitivity in predicting preeclampsia.This finding was consistent with the study by Prakansamut et al. conducted in 2019 [15], which found that combined serum SHARP1 levels and uterine artery PI in the first trimesters showed good sensitivity in predicting preeclampsia.Therefore, a combination of serum SHARP1 levels and the second-trimester uterine artery PI showed a promising outcome that might help in screening pregnant women who had a high risk of developing preeclampsia and needed close monitoring.Our results differed from those of previous reports of failure to improve overall PE detection [6,20,21].This discrepancy might be due to the difference of serum markers and gestational age at inclusion.This study demonstrated that the combination of the first-trimester serum SHARP1 levels with the second-trimester uterine artery PI had good sensitivity in predicting preeclampsia.This finding was consistent with previous studies which found a combination of maternal serum markers with Doppler of the uterine artery increased the sensitivity for predicting preeclampsia [22,23].This might be due to the combination test improves the prediction of preeclampsia.
This study also examined the association of first-trimester serum SHARP1 level and the second-trimester uterine artery Doppler for other pregnancy complications, such as preterm delivery, fetal growth restriction, and gestational diabetes.The pathophysiology of fetal growth restriction included impaired trophoblastic invasion [18].However, the result of the study    did not find an association between serum SHARP1 level and the second-trimester uterine artery Doppler for other pregnancy complications.This study also found that in a pregnancy complicated with preeclampsia, the mean uterine artery PI in the second trimester was higher than in normal pregnancy groups as well as the prevalence of early diastolic notching.These findings were similar to the meta-analysis by Cnossen et al. in 2008 [16].
The uterine artery PI above the 95th percentile had a high specificity for predicting preeclampsia (95.8%), and it had a low sensitivity (23.1%).The findings were consistent with the previous meta-analysis by Cnossen et al. [16].They found that an increased UAPI in the second trimester had 42% (25-58%) sensitivity for the prediction of preeclampsia in low-risk patients, whereas the presence of any notching produced a higher sensitivity of 74% (60-87%).Unlike the study by Dash et al., they used an abnormal uterine artery PI greater than 1.32 [24].They found a high sensitivity of 92.6% and specificity of 84.7% for predicting preeclampsia in the second trimester.The reason may be due to the differences in the populations studied, gestational age at examination, and the definition of abnormal flow velocity waveform.Thus, the second trimester is the best time to evaluate the uterine artery Doppler [16].
The advantage of SHARP1 was that SHARP1 which plays an important role in promoting the degradation of HIF-1a, has been introduced to detect the occurrence of preeclampsia with better sensitivity and specificity than HIF-1a [10][11][12][13].The disadvantage of SHARP1 was that the cost-effectiveness and the availability of this investigation especially in low-resource settings were not evaluated.
The strength of this study is that it is the first study to evaluate serum SHARP1 levels in the first trimester combined with uterine artery Doppler in the second trimester for preeclampsia prediction and the study is a prospectively collected cohort.The limitation was that uterine artery Doppler was performed during the second trimester.However, this may be useful for pregnant women who cannot be performed firsttrimester uterine artery Doppler.Identification of high-risk pregnant women can improve pregnancy outcomes, either by frequent surveillance or the consideration of starting aspirin.Another limitation was that there were few cases of early-onset preeclampsia.Further studies with larger sample sizes of patients with early-onset preeclampsia should be conducted.

Conclusion
This study demonstrated that serum SHARP1 in the first trimester combined with the uterine artery PI in the second trimester has good sensitivity to predict preeclampsia.

Disclosure statement
No potential conflict of interest was reported by the authors.  of Medicine, Chulalongkorn University for their helpful suggestions and assistance.The authors would also like to thank Ms. Walailak Thongthab and Ms. Natnicha Houngham for their technical assistance.

Table 1 .
Demographic data, and maternal and neonatal outcomes of women with preeclampsia compared with women without preeclampsia.

Table 2 .
Serum SHARP1 level and uterine artery Doppler findings in women with preeclampsia and without preeclampsia.

Table 3 .
Chulalongkorn University.The authors wish to thank the staff and nurses of the Division of Maternal-Fetal Medicine, Department of Obstetrics and Gynecology, Faculty Predictive value of serum SHARP1 level and uterine artery Doppler for preeclampsia.PI: pulsatility index: MoM: multiple of median: CI: confidence interval: PPV: positive predictive value: NPV: negative predictive value: LR: likelihood ratio.

Table 4 .
The association of abnormal serum SHARP1 level and uterine artery Doppler for other outcomes FGR: fetal growth restriction: GA: gestational age: GDM: gestational diabetes mellitus.