Accepted: 2 March 2022 | Published Online 5 April 2022 DOI: 10.1111/1471-0528.17151 RESEARCH ARTICLE Fetal Medicine Angiogenic factors for planning fetal surveillance in fetal growth restriction and small-­for-­gestational-­age fetuses: A prospective observational study Erika Bonacina | Manel Mendoza | Alba Farràs | Pablo Garcia-­Manau | Berta Serrano | Ivan Hurtado | Raquel Ferrer-­Oliveras | Lidia Illan | Mireia Armengol-­Alsina | Elena Carreras Maternal Fetal Medicine Unit, Department of Obstetrics, Hospital Universitari Vall Abstract d'Hebron, Universitat Autonoma de Barcelona, Barcelona, Spain Objective: The aim of this study was to assess the added value of the soluble fms-­like tyrosine kinase-­1 (sFlt-­1) and placental growth factor (PlGF) ratio for adjusting the Correspondence Manel Mendoza and Erika Bonacina, periodicity of ultrasound examinations in early-­onset fetal growth restriction (FGR) Maternal Fetal Medicine Unit, Department of Obstetrics, Hospital Universitari Vall and small for gestational age (SGA). d'Hebron, Universitat Autònoma de Design: A prospective, observational study. Barcelona, Passeig de la Vall d'Hebron, 119-­ 129, 08035 Barcelona, Spain. Setting: Tertiary referral hospital. Emails: mmendoza@vhebron.net and erika. bonacina@gmail.com Population: One hundred and thirty-­four single pregnancies with ultrasonographic Funding information estimated fetal weight (EFW) below the 10th centile between 20+0 and 31+6 weeks of None. gestation with antegrade umbilical artery flow. Methods: The time from Doppler and sFlt-­1/PlGF assessment to delivery was re- corded and classified into four ranges: <1, <2, <3 and <4 weeks. Main outcome measures: Sensitivity (Sn), specificity (Sp), positive predictive value (PPV) and negative predictive value (NPV) of sFlt-­1/PlGF values to predict the time to delivery. Results: In the SGA cohort, the NPV calculated for an sFlt-­1/PlGF cut-­off value of 38 was 100% for delivery before 3 weeks, and 98% for delivery before 4 weeks after diag- nosis (95% CI 0.89–­1.00). In the FGR cohort, the NPV calculated for an sFlt-­1/PlGF cut-­off value of 38 was 100% for delivery before 2 weeks after diagnosis (95% CI 0.92–­ 1.00). By contrast, more than 50% of cases with an sFlt-­1/PlGF value of >85 required an elective delivery before 1 week. Conclusions: sFlt-­1/PlGF values in early-­onset SGA and FGR are predictive of the time to delivery and could be used for planning fetal surveillance, by reducing the frequency of ultrasound in cases with sFlt-­1/PlGF < 38 and by providing closer fol- low-­up in cases with sFlt-­1/PlGF >85. K EY WOR DS fetal Doppler, fetal growth restriction, fetal surveillance, PlGF, sFlt-­1, small for gestational age Linked article: This article is commented on by Cecilia Villalaín, pp. 1878 in this issue. To view this minicommentary visit https://doi.org/10.1111/1471-0528.17172. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2022 The Authors. BJOG: An International Journal of Obstetrics and Gynaecology published by John Wiley & Sons Ltd. 1870 | wileyonlinelibrary.com/journal/bjo  BJOG. 2022;129:1870–1877. ANGIOGENIC FACTORS FOR FETAL SURVEILLANCE IN FGR/SGA FETUSES    | 1871 Tweetable abstract: sFlt-­1/PlGF values in early-­onset SGA/FGR could be used in ­addition to Doppler for planning fetal surveillance. 1 | I N T RODUC T ION congenital defects, PE and stillbirth at the time of diagno- sis. FGR was defined as an EFW below the third centile, or Small fetuses are usually defined as those with an ultrasound EFW between the third and 10th centiles accompanied by a estimated fetal weight (EFW) below the 10th centile.1 Fetal UA pulsatility index (PI) above the 95th centile, a cerebro- growth restriction (FGR) is a failure to achieve biological placental ratio below the fifth centile, middle cerebral ar- growth potential and is usually accompanied by fetal/placental tery PI below the fifth centile and/or UA PI above the 95th Doppler anomalies, whereas fetuses deemed small for gesta- centile, and SGA was defined as an EFW of between the tional age (SGA) are usually considered constitutionally small third and 10th centiles with no fetal or maternal Doppler and have normal Doppler results.2,3 Along with pre-­eclampsia anomalies.1,13–­15 PE was defined as the new onset of high (PE), FGR and some SGA might be caused by a degree of placen- blood pressure (systolic blood pressure ≥ 140 mmHg and/or tal dysfunction, associated with poorer perinatal outcomes.4,5 diastolic blood pressure ≥ 90 mmHg), the worsening of pre- Adequate surveillance and planned delivery for SGA and FGR vious high blood pressure added to new-­onset proteinuria fetuses can reduce the rate of stillbirth and improve pregnancy (protein/creatinine ratio of >300), the worsening of previ- outcomes;6 however, optimal management is challenging as ous proteinuria, or to at least one of the following signs and it requires multiple ultrasounds for fetal Doppler assessment, symptoms: cerebral or visual symptoms, elevation of liver which represents a healthcare burden and arouses anxiety and enzymes to twice the normal concentration, platelet count stress in patients. The management of FGR and SGA varies of <100 000/μl, serum creatinine concentrations of >1.1 mg/ considerably internationally, as there is no clear consensus on dl or pulmonary oedema.16 the adequate frequency of ultrasound examinations.7 In early-­ Maternal demographic characteristics and medical history, onset cases (<32 weeks of gestation) with antegrade flow in the ultrasound assessment of fetal anatomy, EFW and serum con- umbilical artery (UA),8 most guidelines advise performing centrations of PlGF and sFlt-­1 were recorded at diagnosis. EFW Doppler assessment at least weekly for FGR and every 2 weeks was measured by fetal biparietal diameter, head circumference, for SGA.7 abdominal circumference and femur length,17 and EFW cen- Soluble fms-­like tyrosine kinase-­1 (sFlt-­1) and placental tiles were calculated according to customised birthweight stan- growth factor (PlGF) are angiogenic-­related biomarkers of dards for a Spanish population.18 PlGF and sFlt-­1 were both placental dysfunction.9 These markers have been demon- measured in pg/ml by means of fully automated Elecsys as- strated to predict adverse maternal and perinatal outcomes says on an electrochemiluminescence immunoassay platform in PE,10 and correlate positively with both FGR/SGA severity (Cobas E analysers; Roche Diagnostics, Basel, Switzerland). and speed of disease progression, which lead to shorter times Gestational age (GA) was determined by fetal crown–­rump to delivery.11,12 However, their ability to assist with decision length measurement at 11+0–­13+6 weeks of gestation.19 making around the frequency of fetal surveillance has not Specialists were not blinded to angiogenic factor levels; been investigated. nevertheless, the timing and mode of delivery were based The aim of this study was to assess the added value of the in all cases on GA, Doppler findings, conventional visual sFlt-­1/PlGF ratio for adjusting the periodicity of ultrasound cardiotocography (CTG) interpretation, and maternal signs examinations in early-­onset FGR and SGA. and symptoms following the current hospital protocols, regardless of the results of the sFlt-­1/PlGF measurement.1 According to that protocol, elective delivery was recom- 2 | M ET HODS mended at >40 weeks of gestation for SGA, >37 weeks of gestation for FGR with antegrade UA flow, >34 weeks of This is a prospective observational study carried out at Vall gestation for FGR with AEDF and >30 weeks of gestation for d'Hebron University Hospital, Barcelona, Spain, between FGR with REDF. CTG indications for elective delivery were July 2017 and July 2019. The local ethics committee (CEIC-­ fetal heart rate sinusoidal tracing or absent fetal heart rate VHIR PR[AMI]349/2016) approved the study protocol. variability, accompanied by recurrent late decelerations, re- Patients were not actively involved in the research. Single current variable decelerations or bradycardia.20 pregnancies with an ultrasonographic estimated fetal weight (EFW) below the 10th centile between 20+0 and 31+6 weeks of gestation were invited to participate. Women were enrolled 2.1 | Statistical analysis after providing their written informed consent. The exclu- sion criteria were UA absent end-­diastolic flow (AEDF), The Strengthening the Reporting of Observational Studies in UA reversed end-­diastolic flow (REDF), abnormal ductus Epidemiology (STROBE) statement was followed for report- venosus flow, known fetal chromosomal abnormalities, ing the results.21 Categorical data were reported as frequency 1872 |    BONACINA et al. TA BL E 1 Baseline characteristics of the study population and percentage and continuous variables were reported as median and interquartile range (IQR). The time from Doppler SGA (n = 48) FGR (n = 86) and sFlt-­1/PlGF assessment to delivery was recorded and clas- Maternal age (years) 33.0 (30.0–­35.0) 31.0 (27.0–­35.7) sified into four ranges: <1, <2, <3 and <4 weeks. The associa- Pre-­pregnancy BMI 23 (20.6–­25.3) 24.2 (21.6–­27.9) tion between sFlt-­1/PlGF values above the cut-­off values and (kg/m 2) time to delivery was investigated by calculating the sensitivity Smoking 5 (10.4%) 15 (17.4%) (Sn), specificity (Sp), positive predictive value (PPV) and neg- Obstetric history ative predictive value (NPV). Kaplan–­Meier survival curves were constructed for the analysis of time to delivery after the FGR in previous 9 (18.7%) 7 (8.1%) pregnancy diagnosis of early-­onset FGR for sFlt-­1/PlGF cut-­off values of 38, 85, 110, 201 and 655. The two-­sided statistical significance SGA in previous 1 (2.1%) 3 (3.5%) pregnancy level was set at p < 0.05. The R Commander package in R 2.3-­1 was used for statistical analysis. Stillbirth in previous 1 (2.1%) 1 (1.2%) pregnancy Previous pre-­eclampsia 4 (8.3%) 1 (1.2%) 3 | R E SU LT S Maternal history Chronic hypertension 1 (2.1%) 0 During the study period, 134 consecutive women were in- Pre-­pregnancy diabetes 2 (4.2%) 1 (1.2%) vited to take part in the study and all agreed to participate. Chronic kidney disease 1 (2.1%) 0 Among them, 48 (35.8%) had a fetus that was SGA and 86 SLE 0 0 (64.2%) had FGR. All FGR with abnormal Doppler results Antiphospholipid 0 3 (3.5%) were associated with UA increased pulsatility. The median syndrome GA at diagnosis of SGA and FGR was 26.5 weeks (IQR 24.0–­ Ethnicity 29.0 weeks) and 25.6 weeks (IQR 23.0–­29.3 weeks), respec- tively. The demographic and clinical characteristics of the White 40 (83.3%) 78 (90.7%) studied cohorts are shown in Table 1. The median GA at Black 4 (8.3%) 2 (2.3%) delivery was 38.0 weeks (IQR 37.0–­39.0 weeks) for SGA and Asian 1 (2.1%) 0 37.0 weeks (IQR 34.2–­37.0 weeks) for FGR, with low EFW Southeast Asian 3 (6.2%) 6 (7%) centile being the most common indication in both groups Other 0 0 (61.7% in SGA and 38.4% in FGR). More details regarding Mode of conception pregnancy outcomes are presented in Table 2. In the SGA cohort, the median interval between diagnosis Spontaneous 47 (97.9%) 82 (95.3%) and delivery was 63.0 days (IQR 47.2–­75.5 days). All women IVF 1 (2.1%) 3 (3.5%) with a fetus recorded as SGA had an sFlt-­1/PlGF ratio of Insemination 0 1 (1.2%) below 38 at diagnosis and the time to delivery was longer Gestational age (weeks) at 26.5 (24.0–­29.0) 25.6 (23.0–­29.3) than 3 weeks in all cases. Thus, sFlt-­1/PlGF < 38 at diagnosis diagnosis had an NPV of 100% for the need for elective delivery before UA PI >95th centile 0 11 (12.8%) 3 weeks. More details are presented in Figure 1. UtA PI >95th centile 0 58 (67.4%) In the FGR cohort, at diagnosis 67 women (77.9%) had an EFW <3rd centile 0 44 (51.1%) sFlt-­1/PlGF value of <38, four (4.7%) had an sFlt-­1/PlGF value Pre-­eclampsia at diagnosis 0 0 of 38–­85, 15 (17.4%) had an sFlt-­1/PlGF value of >85, 13 (15.1%) had an sFlt-­1/PlGF value of >110, eight (9.3%) had an sFlt-­1/ sFlt-­1/PlGF 3.95 (2.11–­7.53) 7.80 (2.89–­28.9) PlGF value of >201 and six (7.2%) had an sFlt-­1/PlGF value of <38 48 (100%) 67 (77.9%) >655. The overall median interval to delivery was 61.0 days >85 0 15 (17.4%) (IQR 41.2–­87.5 days). The interval to delivery was analysed >110 0 13 (15.1%) by sFlt-­1/PlGF groups and was significantly shorter in cases >201 0 8 (9.3%) with higher sFlt-­1/PlGF values. The median (IQR) intervals >655 0 6 (7.0%) were: 70.0 (48.5–­91.5) and 28.0 (5.0–­53.0) days in cases with Mean uterine artery 0 58 (67.4%) sFlt-­1/PlGF values of <38 or ≥38, respectively (p < 0.001); 70.0 Doppler >95th centile (50.3–­89.8) and 6.0 (5.0–­35.5) days in cases with sFlt-­1/PlGF values of ≤85 or >85, respectively (p < 0.001); 70.0 (49.8–­90.5) Note: Continuous data are given as median and interquartile range; categorical data are given as frequency and percentage. and 6.0 (5.0–­28.0) days in cases with sFlt-­1/PlGF values of Abbreviations: BMI, body mass index; EFW, estimated fetal weight; FGR, fetal ≤110 or >110, respectively (p < 0.001); 67.0 (46.8–­89.3) and growth restriction; IVF, in vitro fertilisation; PlGF, placental growth factor; sFlt-­1, 5.0 (2.9–­23.0) days in cases with sFlt-­1/PlGF values of ≤201 soluble fms-­l ike tyrosine-­k inase-­1; SGA, small for gestational age; SLE, systemic or >201, respectively (p < 0.001); and 63.5 (45.0–­89.0) and 2.9 lupus erythematosus; UA PI, umbilical artery pulsatility index; UtA PI, mean uterine artery pulsatility index . (2.0–­5.0) days in cases with sFlt-­1/PlGF values of ≤655 or ANGIOGENIC FACTORS FOR FETAL SURVEILLANCE IN FGR/SGA FETUSES    | 1873 TA BL E 2 Pregnancy outcomes of the study population TA BL E 2 (Continued) SGA (n = 48) FGR (n = 86) SGA (n = 48) FGR (n = 86) Neonatal weight (g) 2607.5 2160.0 EFW 30 (62.5%) 33 (38.4%) (2215.0–­2941.2) (1650.0–­2555.0) Other 8 (16.7%) 7 (8.1%) GA at delivery (weeks) 38.0 (37.0–­39.0) 37.0 (34.2–­37.0) Note: Continuous data are given as median and interquartile range; categorical data Delivery <37 weeks 5 (10.4%) 34 (39.5%) are given as frequency and percentage. Delivery <34 weeks 1 (2.1%) 16 (18.6%) Abbreviations: BPD, bronchopulmonary dysplasia; CTG, cardiotocography; EFW, estimated fetal weight; NEC, necrotising enterocolitis; NICU, neonatal intensive Delivery <30 weeks 0 7 (8.1%) care unit; OBICU, obstetric intensive care unit; RDS, respiratory distress syndrome; TOP, termination of pregnancy. Days from diagnosis to 63.0 (47.2–­75.5) 61.0 (41.25–­87.5) delivery Adverse pregnancy outcomes >655, respectively (p < 0.001). Kaplan–­Meier curves are pre- Placental abruption 1 (2.1%) 4 (4.6%) sented in Figure 2. In the group with an sFlt-­1/PlGF value of Pre-­eclampsia 1 (2.1%) 17 (19.8%) <38, there was no need for elective delivery before 2 weeks in Mild 1 (2.1%) 4 (4.65%) any case, and one case (1.5%) required an elective delivery before 4 weeks. A sFlt-­1/PlGF cut-­off value of 38 at diagnosis Severe 0 13 (15.1%) had an NPV of 100% for the need for elective delivery before HELLP 0 2 (2.3%) 2 weeks and an NPV of 98% for delivery before 4 weeks. By syndrome contrast, more than 50% of cases with sFlt-­1/PlGF >85 re- Eclampsia 0 1 (1.2%) quired elective delivery before 1 week, rising to 100% in cases Stillbirth 0 2 (2.3%) with sFlt-­1/PlGF >655. More details are presented in Table 3 Caesarean for non-­ 3 (6.2%) 19 (22.1%) and Figure 1. reassuring CTG NICU admission of 9 (18.7%) 36 (41.9%) ≥48 h 4 | DISC US SION Days in NICU 18.5 (7.25–­48.5) 17.0 (9.2–­39.0) Neonatal death 0 2 (2.3%) 4.1 | Main findings RDS 3 (6.2%) 16 (18.6%) BPD 0 7 (8.1%) This study provides evidence that the sFlt-­1/PlGF ratio can be useful for classifying the severity of FGR and to assess Sepsis 1 (2.1%) 3 (3.5%) the interval to delivery among cases with similar baseline Retinopathy 0 0 severity and prognosis, regarding Doppler and ultrasound (stages III–­I V) findings. An sFlt-­1/PlGF ratio cut-­off value of 38 has an NPV NEC 1 (2.1%) 2 (2.3%) of 100% for ruling out the need for elective delivery before Intraventricular 1 (2.1%) 0 3 weeks for fetuses that are SGA and before 2 weeks for fe- haemorrhage tuses with FGR. By contrast, the sFlt-­1/PlGF cut-­off value of grade III or IV 85 is associated with a shorter time to delivery, requiring an Periventricular 0 0 elective delivery before 1 week in more than 50% of cases. leukomalacia Apgar score of <7 at 2 (4.2%) 11 (12.8%) 5 minutes 4.2 | Strengths and limitations Artery pH of ≤7.0 0 3 (3.5%) SGA/FGR severity at delivery The main strengths of this study are its prospective design SGA 42 (87.5%) 0 in a relatively large cohort of early-­onset FGR and SGA. Antegrade UA flow 5 (10.4%) 76 (88.4%) Additionally, this study provides valuable information that UA AEDF 1 (2.1%) 6 (7.0%) could be used for tailoring the frequency of fetal scans based UA REDF 0 4 (4.6%) on sFlt-­1/PlGF values. In Spain, most centres have the sFlt-­1/ PlGF value available for use in clinical practice for women Elective delivery 41 (85.4%) 67 (77.9%) with suspected PE. According to the PROGNOSIS trial pro- Pre-­eclampsia 0 8 (9.3%) tocol,10 FGR is considered a PE-­related finding; therefore, Fetal Doppler 1 (2.1%) 9 (10.5%) the determination of sFlt-­1/PlGF is indicated in such cases. CTG 1 (2.1%) 5 (5.8%) We acknowledge some limitations of this study. First, Abruption 1 (2.1%) 3 (3.5%) computerised CTG was not available at our centre; thus, our Stillbirth 0 2 (2.3%) results may not be applicable to cases with such a tool avail- able for fetal monitoring. Second, the sFlt-­1/PlGF results 1874 |    BONACINA et al. F I G U R E 1 Bar chart showing the frequencies of SGA and FGR cases with delivery at <1, <2, <3 and <4 weeks after diagnosis, according to soluble fms-­like tyrosine kinase-­1/placental growth factor value at diagnosis. FGR, fetal growth restriction; PlGF, placental growth factor; sFlt-­1, soluble fms-­like tyrosinekinase-­1; SGA, small for gestational age F I G U R E 2 Kaplan–­Meier graphs showing time from diagnosis to delivery in pregnancies with early-­onset fetal growth restriction, according to soluble fms-­like tyrosine kinase-­1/placental growth factor values of <38 or ≥38 (A), ≤85 or >85 (B), ≤110 or >110 (C), ≤201 or >201 (D) and ≤655 or >655 (E), at diagnosis were not blinded to the investigators, which could have in- 4.3 | Interpretation fluenced the fetal surveillance protocol; nevertheless, the indication for delivery was based strictly on ultrasound and Previous studies have shown that high sFlt-­1/PlGF values in CTG findings in all cases. Third, chromosomal anomalies FGR and PE reveal an antiangiogenic status, which is caused and congenital defects were excluded in this study; thus, our by the underlying placental dysfunction.22 The sFlt-­1/PlGF results are not applicable to such cases. Fourth, our results ratio has demonstrated that it is useful for predicting preg- might be of great value for tailoring follow-­up and prenatal nancy complications associated with placental insufficiency, counselling; however, they cannot be used for planning de- and particularly for ruling out PE in pregnancies with FGR livery. Finally, only 15 cases had an sFlt-­1/PlGF value of >85, or other clinical signs and symptoms of PE.10 Recently, sev- which could limit the external validity of our findings. eral studies have shown that the sFlt-­1/PlGF ratio values in ANGIOGENIC FACTORS FOR FETAL SURVEILLANCE IN FGR/SGA FETUSES    | 1875 T A B L E 3 Sensitivity, specificity, positive predictive value and negative predictive value of different soluble fms-­like tyrosine kinase-­1/placental growth factor cut-­off values used to predict delivery at <1, <2, <3 and <4 weeks FGR with antegrade umbilical artery flow (n = 86) sFlt-­1/PlGF Time to delivery Sn (95% CI) Sp (95% CI) PPV (95% CI) NPV (95% CI) 38 (n = 67) <1 week 1 (0.52–­1) 0.86 (0.76– ­0.93) 0.42 (0.2– ­0.66) 1 (0.92–­1) <2 weeks 1 (0.52–­1) 0.86 (0.76– ­0.93) 0.42 (0.2– ­0.66) 1 (0.92–­1) <3 weeks 0.89 (0.52–­1) 0.86 (0.76– ­0.93) 0.42 (0.2– ­0.66) 0.98 (0.92–­1) <4 weeks 0.9 (0.55–­1) 0.87 (0.77– ­0.93) 0.47 (0.24– ­0.71) 0.98 (0.92–­1) 85 (n = 15) <1 week 1 (0.52–­1) 0.9 (0.82– ­0.96) 0.53 (0.27– ­0.79) 1 (0.92–­1) <2 weeks 0.89 (0.52–­1) 0.91 (0.82– ­0.96) 0.53 (0.27– ­0.79) 0.99 (0.92–­1) <3 weeks 0.9 (0.55–­1) 0.92 (0.84– ­0.97) 0.6 (0.32– ­0.84) 0.99 (0.92–­1) <4 weeks 0.9 (0.55–­1) 0.92 (0.84– ­0.97) 0.6 (0.32– ­0.84) 0.99 (0.92–­1) 110 (n = 13) <1 week 1 (0.52–­1) 0.94 (0.86– ­0.98) 0.61 (0.32– ­0.86) 1 (0.93–­1) <2 weeks 1 (0.52–­1) 0.94 (0.86– ­0.98) 0.61 (0.32– ­0.86) 1 (0.93–­1) <3 weeks 1 (0.52–­1) 0.94 (0.86– ­0.98) 0.61 (0.32– ­0.86) 1 (0.93–­1) <4 weeks 0.9 (0.55–­1) 0.95 (0.87– ­0.98) 0.69 (0.39– ­0.9) 0.99 (0.93–­1) 201 (n = 8) <1 week 0.75 (0.35– ­0.97) 0.97 (0.91–­1) 0.75 (0.35– ­0.97) 0.97 (0.91–­1) <2 weeks 0.75 (0.35– ­0.97) 0.97 (0.91–­1) 0.75 (0.35– ­0.97) 0.97 (0.91–­1) <3 weeks 0.67 (0.3– ­0.92) 0.97 (0.9–­1) 0.75 (0.35– ­0.97) 0.96 (0.89– ­0.99) <4 weeks 0.6 (0.26– ­0.88) 0.97 (0.9–­1) 0.75 (0.35– ­0.97) 0.95 (0.87–­0.99) 655 (n = 6) <1 week 0.5 (0.07– ­0.93) 1 (0.93–­1) 1 (0.09–­1) 0.97 (0.91–­1) <2 weeks 0.5 (0.07– ­0.93) 1 (0.93–­1) 1 (0.09–­1) 0.97 (0.91–­1) <3 weeks 0.4 (0.05– ­0.85) 1 (0.93–­1) 1 (0.09–­1) 0.96 (0.89– ­0.99) <4 weeks 0.33 (0.04– ­0.77) 1 (0.93–­1) 1 (0.09–­1) 0.95 (0.88– ­0.99) Abbreviations: CI, confidence interval; NPV, negative predictive value; PlGF, placental growth factor; PPV, positive predictive value; sFlt-­1, soluble fms-­l ike tyrosine-­k inase-­1; Sn, sensitivity; Sp, specificity. early-­onset FGR rise as Doppler severity increases, which fetal surveillance, as they have shown to be highly associated leads to shorter times to delivery, lower birthweight z-­scores with the interval to delivery. and greater risks of adverse perinatal outcomes.12,23 Another For these reasons, our study has important clinical impli- study in a cohort of 122 cases of early-­onset FGR with ante- cations as it has shown excellent performance for detecting grade UA flow showed that 35.6% of cases with sFlt-­1/PlGF pregnancies at a higher risk of fetal distress, which require values of >85 and none of the cases with sFlt-­1/PlGF values more frequent scans, but also cases at a lower risk of com- of ≤85 required an elective delivery within 1 week.11 These plications that could benefit from a less strict surveillance findings are similar to our findings here; however, in that protocol. According to our results, FGR cases with sFlt-­1/ study no other cut-­off values were investigated, and nor were PlGF values of >85 should be closely monitored for the early the NPV and PPV values used for predicting and excluding detection of fetal or maternal complications. By contrast, in the need for elective delivery in subsequent weeks. pregnancies with a sFlt-­1/PlGF value of <38 it might be pos- Predicting the time between diagnosis and delivery in sible to extend the interval between scans to up to 3 weeks SGA and FGR fetuses is challenging, as the progression of for SGA and up to 2 weeks for FGR with antegrade UA flow, Doppler deterioration can vary widely;24 thereby, it is dif- which would significantly reduce the number of fetal ultra- ficult to establish from Doppler findings alone the optimal sounds, thereby lowering parental anxiety and the burden interval between ultrasounds in the management of these on the healthcare system. fetuses. In early-­onset cases with antegrade flow in UA, management recommendations vary considerably as there is no consensus on the adequate frequency of scans;7 nev- 5 | C ONC LUSION ertheless, most international guidelines advise that Doppler ultrasound should be performed at least weekly for FGR and sFlt-­1/PlGF values in early-­onset SGA and FGR are predic- every 2 weeks for fetuses that are SGA.7 In this study we show tive of the time to delivery and could be used for planning that sFlt-­1/PlGF values could be useful for individualising fetal surveillance by reducing the frequency of ultrasounds 1876 |    BONACINA et al. 3. Figueras F, Savchev S, Triunfo S, Crovetto F, Gratacos E. An integrated in cases with sFlt-­1/PlGF < 38 and by providing closer follow model with classification criteria to predict small-­for-­gestational-­ up in cases with sFlt-­1/PlGF >85. age fetuses at risk of adverse perinatal outcome. Ultrasound Obstet Gynecol. 2015;45:279–­85. AC K N OW L E D G E M E N T S 4. McIntire DD, Bloom SL, Casey BM, Leveno KJ. Birth weight in re- lation to morbidity and mortality among newborn infants. N Engl J We thank all the physicians who facilitated the recruitment Med. 1999;340:1234–­8. of individuals at the Hospital Universitari Vall d'Hebron and 5. Paules C, Youssef L, Rovira C, Crovetto F, Nadal A, Peguero A, et al. all the participants who agreed to take part in this study. Distinctive patterns of placental lesions in pre-­eclampsia vs small-­ for-­gestational age and their association with fetoplacental Doppler. C ON F L IC T OF I N T E R E S T Ultrasound Obstet Gynecol. 2019;54:609–­16. 6. Gardosi J, Madurasinghe V, Williams M, Malik A, Francis A. Manel Mendoza has received lecture fees from Roche Maternal and fetal risk factors for stillbirth: population based study. Diagnostics. The other authors report that they have no BMJ. 2013;346:f108. conflicts of interest associated with this work. Completed 7. McCowan LM, Figueras F, Anderson NH. Evidence-­based national disclosure of interests form available to view online as sup- guidelines for the management of suspected fetal growth restric- porting information. tion: comparison, consensus, and controversy. Am J Obstet Gynecol. 2018;218:S855–­68. 8. Savchev S, Figueras F, Sanz-­Cortes M, Cruz-­Lemini M, Triunfo S, AU T HOR C ON T R I BU T ION S Botet F, et al. Evaluation of an optimal gestational age cut-­off for EB, MM, PG-­M and BS had full access to all of the data in the definition of early-­and late-­onset fetal growth restriction. Fetal the study and take responsibility for the integrity of the data Diagn Ther. 2014;36:99–­105. and the accuracy of the data analysis. EB and MM conceived 9. Levine RJ, Maynard SE, Qian C, Lim K-­H, England LJ, Yu KF, et al. Circulating angiogenic factors and the risk of preeclampsia. N Engl J and designed the study. PG-­M, BS, AF, IH, RF-­O, LI and Med. 2004;350:672–­83. MA-­A contributed to the literature research. EB, MM, AF, 10. Zeisler H, Llurba E, Chantraine F, Vatish M, Staff AC, Sennström PG-­M, BS, LI and MA-­A contributed to data collection and M, et al. Predictive value of the sFlt-­1:PlGF ratio in women with sus- confirmation. EB, MM, PG-­M and BS contributed to data pected preeclampsia. N Engl J Med. 2016;374:13–­22. analysis, and EB, MM, PG-­M, BS, AF and EC contributed 11. Quezada MS, Rodríguez-­Calvo J, Villalaín C, Gómez-­A rriaga PI, Galindo A, Herraiz I. sFlt-­1/PlGF ratio and timing of delivery in to data interpretation. EB and MM drafted the article. EC early-­onset fetal growth restriction with antegrade umbilical artery made substantial revisions to the article. flow. Ultrasound Obstet Gynecol. 2020;56:549–­56. 12. Garcia-­ Manau P, Mendoza M, Bonacina E, Garrido-­ Gimenez C, E T H IC A L A PPROVA L Fernandez-­Oliva A, Zanini J, et al. Soluble fms-­like tyrosine kinase The Vall d'Hebron University Hospital Ethics Committee to placental growth factor ratio in different stages of early-­onset fetal growth restriction and small for gestational age. Acta Obstet Gynecol (CEIC-­VHIR PR[AMI]349/2016) approved the study proto- Scand. 2021;100:119–­28. col. Informed consent was obtained from all patients, which 13. Arduini D, Rizzo G. Normal values of Pulsatility index from fetal ves- was included in the patient’s medical record. sels: a cross-­sectional study on 1556 healthy fetuses. J Perinat Med. 1990;18:165–­72. DATA AVA I L A BI L I T Y S TAT E M E N T 14. Baschat AA, Gembruch U. The cerebroplacental doppler ratio revis- ited. Ultrasound Obstet Gynecol. 2003;21:124–­7. Data available on request from the authors. 15. Gómez O, Figueras F, Fernández S, Bennasar M, Martínez JM, Puerto B, et al. Reference ranges for uterine artery mean pulsatil- ORC I D ity index at 11-­41 weeks of gestation. Ultrasound Obstet Gynecol. Erika Bonacina https://orcid.org/0000-0002-8042-8247 2008;32:128–­32. Manel Mendoza https://orcid.org/0000-0002-3030-3833 16. American College of Obstetricians and Gynecologists, Task Force on Hypertension in Pregnancy. Hypertension in pregnancy. Report Alba Farràs https://orcid.org/0000-0003-2100-7693 of the American College of Obstetricians and Gynecologists' Pablo Garcia-­Manau https://orcid.org/0000-0002-2415-1626 Task Force on hypertension in pregnancy. Obstet Gynecol. Berta Serrano https://orcid.org/0000-0002-1264-9195 2013;122:1122–­31. Ivan Hurtado https://orcid.org/0000-0002-0084-1710 17. Hadlock FP, Harrist RB, Sharman RS, Deter RL, Park SK. Raquel Ferrer-­Oliveras https://orcid. Estimation of fetal weight with the use of head, body, and femur measurements –­a prospective study. Am J Obstet Gynecol. org/0000-0001-9942-7589 1985;151:333–­7. Mireia Armengol-­Alsina https://orcid. 18. Figueras F, Meler E, Iraola A, Eixarch E, Coll O, Figueras J, et al. org/0000-0002-2087-5176 Customized birthweight standards for a Spanish population. Eur J Elena Carreras https://orcid.org/0000-0003-3471-7248 Obstet Gynecol Reprod Biol. 2008;136:20–­4. 19. Robinson HP, Fleming JE. A critical evaluation of sonar ‘crown-­r ump length’ measurements. Br J Obstet Gynaecol. 1975;82:702–­10. R EFER ENCES 20. Macones GA, Hankins GDV, Spong CY, Hauth J, Moore T. The 1. Figueras F, Gratacós E. Update on the diagnosis and classification of 2008 National Institute of Child Health and Human Development fetal growth restriction and proposal of a stage-­based management workshop report on electronic fetal monitoring: update on defi- protocol. Fetal Diagn Ther. 2014;36:86–­98. nitions, interpretation, and research guidelines. Obstet Gynecol. 2. Gordijn SJ, Beune IM, Thilaganathan B, Papageorghiou A, Baschat 2008;112:661–­6. AA, Baker PN, et al. Consensus definition of fetal growth re- 21. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, striction: a Delphi procedure. Ultrasound Obstet Gynecol. Vandenbroucke JP, et al. The Strengthening the Reporting of 2016;48:333–­9. Observational Studies in Epidemiology (STROBE) statement: ANGIOGENIC FACTORS FOR FETAL SURVEILLANCE IN FGR/SGA FETUSES    | 1877 guidelines for reporting observational studies. Int J Surg. 2014;12:​ 1495–­9. S U PP ORT I N G I N F OR M AT ION 22. Maynard SE, Min J-­Y, Merchan J, Mondal S, Libermann TA, Morgan Additional supporting information may be found in the JP, et al. Excess placental soluble fms-­l ike tyrosine kinase 1 (sFlt1) online version of the article at the publisher’s website. may contribute to endothelial dysfunction, hypertension, and pro- teinuria in preeclampsia. J Clin Invest. 2003;111:649–­58. 23. Schlembach D, Wallner W, Sengenberger R, Stiegler E, Mörtl M, How to cite this article: Bonacina E, Mendoza M, Beckmann MW, et al. Angiogenic growth factor levels in ma- ternal and fetal blood: correlation with Doppler ultrasound Farràs A, Garcia-­Manau P, Serrano B, Hurtado I, et al. parameters in pregnancies complicated by pre-­eclampsia and Angiogenic factors for planning fetal surveillance in intrauterine growth restriction. Ultrasound Obstet Gynecol. fetal growth restriction and small-­for-­gestational-­age 2007;29:407–­13. fetuses: a prospective observational study. BJOG. 24. Turan OM, Turan S, Gungor S, Berg C, Moyano D, Gembruch U, et al. 2022;129:1870–­1877. https://doi. Progression of Doppler abnormalities in intrauterine growth restric- tion. Ultrasound Obstet Gynecol. 2008;32:160–­7. org/10.1111/1471-­0528.17151