The risk of late-onset fetal growth restriction and unfavorable neonatal outcomes among fetuses within the 10th-25th weight percentile: a prospective cohort study

This Prospective cohort study was conducted at the Maternal–Fetal Medicine Department of the Shahid Akbarabadi Hospital in Tehran from 21 May 2024 to 11 March 2026. This study complied with the Declaration of Helsinki and was performed according to ethics committee approval. The Ethics and Research Committee of the Iran University of Medical Sciences approved the study (IR.IUMS.FMD.REC.1403.098). The records of all patients who provided written informed consent for using their data for research purposes were analyzed. Statistical analysis was performed with patient anonymity following ethics committee regulations. The primary objective was to evaluate the risk of late-onset of FGR and unfavorable neonatal outcomes among fetuses within the 10th-25th weight percentile. The secondary objective was to assess the predictive ability of Doppler indices for negative outcomes in fetuses within the 10th-25th percentile weight range. Inclusion criteria comprised pregnancies without major maternal or fetal comorbidities at enrollment, singleton pregnancy, mothers aged 18 to 39 years, gestational age from 32 to 33 weeks determined by the first trimester ultrasound, birth weight appropriate for gestational age (10th to 90th percentile), and normal fetal anatomy as assessed by the second trimester scan. Uncertain gestational age, pre-existing maternal conditions such as diabetes (either overt or gestational), hypertension issues (chronic, gestational, preeclampsia), connective tissue disorders (lupus, antiphospholipid syndrome, etc.), blood disorders (sickle cell disease, thalassemia major, etc.), persistent kidney disease, chronic liver illness were among the exclusion criteria’s. Fetuses identified with fetal growth restriction according to the Delphi criteria during the first appointment were also excluded from the study. The research population consisted of expecting mothers attending the clinic who were at least 32 weeks of gestation, with their fetal weight falling between the 10th and 90th percentiles based on biometric ultrasound during the early stage of the third trimester of gestation. Mothers were classified into two categories according to fetal weight percentile: 10th-25th percentile group and > 25th–<90th percentile group. Though participants were not matched individually, both groups were chosen from the same institution within the same study timeframe and confined to the 32.0–33.0 weeks of gestational age range at inclusion to reduce bias related to timing or protocol. Gestational age was verified by the last menstrual period, supported by a first-trimester ultrasound (i.e., craniocaudal length) or an early second-trimester biparietal diameter. In cases of dating discrepancies gestational age was determined by the first trimester ultrasound. Both groups underwent biometric ultrasound at 32–33 weeks of gestation and received biometric and fetal vascular Doppler ultrasound three weeks afterwards. The primary Doppler analysis refers to the second examination. Every participant underwent scanning at the gestational age of 32.0–33.0 weeks of pregnancy on a Philips Affiniti 50 ultrasound machine utilizing a 5 MHz curvilinear transducer with Doppler capabilities. The participants had sonography while lying on their backs on the examination couch to assess the fetal biometric parameters (BPD, HC, AC, and FL) employing standard protocols set by the International Society of Ultrasound in Obstetrics and Gynecology, and the estimated fetal weight (EFW) was derived using the Hadlock 4 formula. The following Hadlock IV formula was used: Log10 BW = 1.326 − (0.00326 × AC × FL) + (0.0107 × HC) + (0.0438 × AC) + (0.158 × FL) [ 2 ]. All Doppler ultrasonography recordings were conducted without any fetal respiration or fetal movement. The analysis was conducted using the mean of three successive Doppler velocity waveforms from each vessel. To guarantee that blood circulation in the major vessels was consistent and devoid of aliasing, the optimal speed of the color Doppler was adjusted for high velocities. The high-pass filter was set to 50 Hz, and the energy output remained under 50 mW/cm2. The volume size of the sample was modified to correspond to the diameter of the vessel, ensuring it was fully covered. In addition to fetal biometric measurements, the machine automatically computed the following umbilical artery Doppler indices: Peak Systolic Velocity (PSV) in cm/s, End Diastolic Velocity (EDV) in cm/s, Pulsatility Index (PI), Resistive Index (RI), and Systolic/Diastolic Ratio (S/D Ratio). umbilical artery pulsatility index (UA-PI), middle cerebral artery (MCA-PI), uterine artery PI, and cerebroplacental ratio (CPR), indices were measured and documented for every fetus. All subjects were monitored until delivery, and the perinatal results were recorded. These results were obtained from the case files of mothers and infants. All ultrasound scans were conducted by a single maternal-fetal medicine (MFM) specialists (to minimize inter-observer differences) under the oversight of a consultant professor. Every ultrasound measurement was taken two times, and the mean value was noted (to minimize intra-observer variability). Beside the baseline variables including maternal age, parity, history of abortion, body mass index (BMI), number of alive children, fetuses were monitored for perinatal outcomes such as fetal weight, FGR, intrauterine fetal demise, oligohydramnios, meconium-stained liquor, LBW, cesarean delivery due to irregular fetal heart rate patterns, meconium-stained amniotic fluid, abnormal cord arterial blood gas (ABG), APGAR (Appearance, Pulse, Grimace, Activity, and Respiration) score under 7 at 5 min, neonatal intensive care unit (NICU) admission, respiratory distress syndrome (RDS), and neonatal mortality. Ultimately, the link between Doppler indices and negative perinatal and neonatal outcomes were analyzed statistically with the relevant statistical tests. The sample size was determined utilizing PASS 2021 software to assess two Receiver operating characteristic (ROC) charts with area under the curve (AUC) discrepancies of 0.95 and 0.87 for the designated diagnostic approach [ 19 ] and taking into account a ratio of 3 to 1 for fetuses above the 25th percentile versus those below the 25th percentile, while also factoring in a type I error rate of 0.05 and a type II error rate of 0.2, totaling 188 fetuses (comprising 47 fetuses below the 25th percentile and 141 fetuses above the 25th percentile). The analysis of the study was conducted using SPSS 20.0 software, incorporating the Kolmogorov-Smirnov test to verify normal distribution. Continuous variables were expressed as median or mean ± standard deviation and categorical variables as frequencies. The pertinent statistical analyses performed comprised students’ t-test, Mann–Whitney U and Pearson chi-squared test. Modified Poisson regression models with robust error variances were employed for binary fetal, neonatal, and obstetric outcomes to compute crude and adjusted risk ratios (RRs). A p-value of < 0.05 was established as the threshold for statistical significance in all conducted tests. ROC analysis, sensitivity, specificity, positive predictive value, and negative predictive value were utilized to assess the predictive performance of Doppler indices for negative outcomes in fetuses within the 10th-25th percentile weight range.

Based on the general characteristics of the study population, a total of 402 pregnancies were part of the analysis. The mothers had a median age of 29 years, with ages ranging from 18 to 39 years. The average BMI was 24.76 ± 10.90 kg/m². Regarding obstetric history, the median total pregnancies were 2, the median deliveries were 1, the median live births were 1, and the median abortions were 0. Based on the first ultrasound evaluation at 32–33 weeks of gestation, among 402 eligible participants, 320 (79.6%) belonged to the > 25th–<90th percentile group and 82 (20.4%) were part of the 10th-25th percentile group. The follow-up biometric ultrasound showed 7.46% of instances of FGR. No cases of oligohydramnios or intrauterine fetal demise were observed. Regarding Doppler indices, 43 cases (10.69%) exhibited abnormal UA-PI. Additionally, 23 cases (5.72%) exhibited abnormal MCA-PI, while 14 cases (3.48%) presented with abnormal mean uterine artery PI. Abnormal CPR index, defined as both abnormal UA-PI and abnormal MCA-PI occurring simultaneously, was noted in 4 cases (0.99%). (Table  1 ) The average weight at birth for the infants was 3137.97 ± 369.80 g. According to the infant’s weight, 27 cases (6.71%) were recognized as LBW. Regarding different neonatal and perinatal results, RDS was noted in 28 instances (7.0%), low Apgar scores occurred in 29 instances (7.21%), meconium-stained amniotic fluid was found in 44 instances (10.94%), and NICU admissions were recorded in 53 instances (13.18%). Additionally, 192 cases (47.76%) resulted in cesarean delivery, while 69 cases (17.16%) had cesarean section because of fetal distress. Two instances (0.49%) of neonatal mortality were identified, one linked to RDS and the other related to birth asphyxia. (Table  1 ) Table 1 Characteristics of the study population and frequency of the outcome measures Variable Study cohort ( n  = 402) Maternal age (years)* 29 (18–39) Body mass index (BMI) (kg/m²)** 24.76 ± 10.90 Gravidity* 2 (1–5) Parity* 1 (0–4) Living children* 1 (0–4) Abortion* 0 (0–2) Gestational age at baseline visit (weeks)* 32.57 (32.00-33.86) Gestational age at the time of delivery (weeks)* 39.14 (33.00-42.14) Days from visit to delivery* 45 (5–65) Abnormal Umbilical artery pulsatility index (PI) *** 43 (10.69) Abnormal fetal middle cerebral artery (MCA) pulsatility index (PI) *** 23 (5.72) Abnormal Mean uterine artery pulsatility index (PI) *** 14 (3.48) Abnormal Cerebroplacental ratio (CPR)*** 4 (0.99) Birth weight (g)** 3137.97 ± 369.80 Fetal growth restriction 30 (7.46) Low birth weight *** 27 (6.71) Respiratory distress syndrome (RDS) *** 28 (6.96) Low APGAR *** 29 (7.21) Meconium-stained liquor*** 44 (10.94) Neonatal intensive care unit (NICU) admission*** 53 (13.18) Cesarean delivery*** 192 (47.76) Cesarean for fetal distress*** 69 (17.16) Abnormal arterial blood gas (ABG) *** 23 (5.72) Neonatal mortality*** 2 (0.49) * Median and the min-max values; specifically, the central data point is presented alongside the smallest and largest values observed. ** Mean ± standard deviation. *** Number of cases and percentages Characteristics of the study population and frequency of the outcome measures * Median and the min-max values; specifically, the central data point is presented alongside the smallest and largest values observed. ** Mean ± standard deviation. *** Number of cases and percentages Table  2 illustrates the comparison among groups of the study population regarding clinical features, Doppler indices, and negative neonatal outcomes. As indicated in Table  2 , the gestational age at delivery for the 10th-25th percentile group was notably less than that of the > 25th– 25th–<90th percentile group, whose mean weight was 3250.64 ± 283.47 g. Abnormal UA-PI, abnormal MCA-PI, and abnormal CPR were more commonly observed in fetuses with a weight in the 10th-25th percentile range. Abnormal Mean uterine artery PI was not significantly different between the groups. The incidence of negative neonatal outcomes like FGR (25.60% vs. 2.81%), LBW (20.73% vs. 0.31%), RDS (13.41% vs. 5.31%), low APGAR scores (14.63% vs. 5.31%), abnormal ABG (13.41% vs. 3.75%), and NICU admission (28.04% vs. 9.37%) was higher in the 10th-25th percentile group compared to the > 25th– 25th–<90th percentile ( n  = 320) p -value Gravidity* 2 (1–5) 2 (1–5) 0.637 Parity* 0 (0–3) 1 (0–4) 0.279 Living children* 0 (0–3) 1 (0–4) 0.254 Abortion* 0 (0–2) 0 (0–2) 0.205 Gestational age at visit (weeks)* 32.57 (32.00-33.71) 32.57 (32.00-33.86) 0.315 Gestational age at delivery (weeks)* 38.43 (33.00-40.29) 39.29 (33.71–42.14) < 0.001 BMI (kg/m²)** 24.59 ± 3.59 24.80 ± 12.08 0.382 Birth weight (g)** 2699.63 ± 339.63 3250.64 ± 283.47 < 0.001 Abnormal Umbilical artery PI*** 15 (18.3) 28 (8.8) 0.022 Abnormal MCA PI*** 9 (11.0) 14 (4.4) 0.031 Abnormal Mean uterine artery PI*** 1 (1.2) 13 (4.1) 0.318 Abnormal CPR*** 3 (3.7) 1 (0.3) 0.028 FGR 21 (25.60) 9 (2.81) < 0.001 LBW *** 17 (20.73) 10 (0.31) < 0.001 RDS*** 11 (13.41) 17 (5.31) < 0.001 Low APGAR *** 12 (14.63) 17 (5.31) < 0.001 Abnormal ABG *** 11 (13.41) 12 (3.75) < 0.001 Meconium-stained liquor*** 8 (9.75) 36 (11.25) 0.851 NICU admission*** 23 (28.04) 30 (9.37) < 0.001 Cesarean delivery due to fetal distress*** 35 (42.68) 157 (49.06) 0.364 Neonatal mortality 1 (1.21) 1 (0.31) 0.098 * Median (Min-Max), ** Mean ± SD, *** n (%), PI: Pulsatility index; MCA: Middle cerebral artery; CPR: Cerebroplacental ratio; FGR: Fetal growth restriction; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas; LBW: Low birth weight Comparison of clinical features, Doppler indices, and adverse perinatal and neonatal outcomes between groups * Median (Min-Max), ** Mean ± SD, *** n (%), PI: Pulsatility index; MCA: Middle cerebral artery; CPR: Cerebroplacental ratio; FGR: Fetal growth restriction; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas; LBW: Low birth weight Modified Poisson regression models featuring robust error variances were employed to assess the presence and degree of a potential link between a 10th to 20th fetal growth centile and the adverse neonatal outcome. The regression models adjusted for covariates such as maternal age, gravidity, parity, BMI, gestational age at the time of delivery, and birth mode showed that the risk of FGR (aRR: 9.23, 95% CI: [7.11–11.78]), LBW, (aRR: 5.03, 95% CI: [3.57–6.97]), RDS (aRR: 1.98, 95% CI: [1.98–2.78]), and NICU admission (aRR: 2.87, 95% CI: [1.45–3.41]) in fetuses within the 10th-25th percentile weight range was greater compared to those in the > 25th-<90th percentile weight range (Table  3 ). Table 3 Modified poisson regression models for predicting adverse perinatal and neonatal outcomes Unfavorable outcome Risk Ratio, Crude Effect (95% CI) Risk Ratio, Adjusted Effect (95% CI) p -value FGR 12.28 (10.23–14.01) 9.23 (7.11–11.78) < 0.001 LBW 8.15 (5.95–10.12) 5.03 (3.57–6.97) < 0.001 RDS 2.75 (1.19–3.05) 1.98 (1.98–2.78) < 0.001 Low APGAR 3.05 (2.06–4.62) 2.75 (1.97–3.09) 0.087 Abnormal ABG 4.05 (2.81–5.33) 3.56 (3.02–4.05) 0.198 NICU admission 3.77 (1.98–4.94) 2.87 (1.45–3.41) < 0.001 FGR: Fetal growth restriction; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas; LBW: Low birth weight Modified poisson regression models for predicting adverse perinatal and neonatal outcomes FGR: Fetal growth restriction; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas; LBW: Low birth weight To examine whether the Doppler indices had any predictive value for adverse outcomes in 10th-25th percentile weight fetuses we applied the multivariate logistic regression analysis. First we divided the groups of a study to those with adverse perinatal and neonatal outcomes and those without it as a reference group. Then we used the sonographic data (UA-PI, MCA-PI, CPR, mean uterine artery PI) as predictors using forward selection of variables to determine independent predictors for the composite adverse neonatal outcome. We developed a composite adverse neonatal outcome if any cases of SGA at birth, cesarean delivery due to fetal distress, RDS, low APGAR scores at 5 min postpartum, abnormal ABG, NICU admissions, or neonatal death were noted. Based on the regression equation, it was noted that fetal Doppler indices was not able in predicting adverse neonatal outcomes among fetuses within the 10th-25th percentile of weight. (Table  4 ) Table 4 Multivariate logistic regression analysis of the sonographic parameters for predicting adverse neonatal outcomes Predictors Odds ratio 95% Confidence interval p -value CPR 12.37 8.02–14.98 0.231 MCA PI 2.73 0.91– 3.49 0.829 Umbilical artery PI 2.34 1.24–3.11 0.099 Mean uterine artery PI 0.028 0.003–0.118 0.787 PI: Pulsatility index; MCA: Middle cerebral artery; CPR: Cerebroplacental ratio; SGA: Small for gestational age; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas Multivariate logistic regression analysis of the sonographic parameters for predicting adverse neonatal outcomes PI: Pulsatility index; MCA: Middle cerebral artery; CPR: Cerebroplacental ratio; SGA: Small for gestational age; RDS: Respiratory distress syndrome; NICU: Neonatal intensive care unit; ABG: Arterial blood gas ROC analysis was also performed to determine the performance of the measured obstetric Doppler parameters in predicting the development of adverse neonatal outcomes. UA-PI (AUC 0.563, 95% CI: 0.456–0.627), CPR (AUC 0.512, 95% CI: 0.439–0.567), Mean uterine artery PI (AUC 0.506, 95% CI: 0.487–0.534), and MCA-PI (AUC 0.493, 95% CI: 0.466–0.532) had no predictive value for adverse neonatal outcomes. (Fig.  1 ) Fig. 1 The area under the curve of the receiver operating characteristic for prediction of the adverse parameters for predicting adverse neonatal outcomes based on the Doppler indices The area under the curve of the receiver operating characteristic for prediction of the adverse parameters for predicting adverse neonatal outcomes based on the Doppler indices Table 5 The predictive ability of Doppler indices for negative outcomes in fetuses within the 10th-25th percentile weight range Doppler parameter AUC (95% CI) Sensitivity (%) Specificity (%) Positive predictive value (%) Negative predictive value (%) Umbilical artery PI 0.56 (0.51–0.61) 21.2 91.4 32.6 85.5 MCA PI 0.49 (0.46–0.52) 4.5 94.0 13.0 83.4 Mean uterine artery PI 0.50 (0.48–0.53) 4.5 96.7 21.4 83.8 CPR 0.51 (0.49–0.53) 3.0 99.4 50.0 83.9 The predictive ability of Doppler indices for negative outcomes in fetuses within the 10th-25th percentile weight range As it shown in Table  5 , ROC analysis, sensitivity, specificity, positive predictive value, and negative predictive value were utilized to assess the predictive performance of Doppler indices for negative outcomes in fetuses within the 10th-25th percentile weight range. ROC analysis showed limited discriminatory performance of the binary Doppler classifications for predicting the composite adverse perinatal outcome. Abnormal umbilical artery PI demonstrated the highest AUC among the evaluated Doppler parameters (AUC 0.56, 95% CI 0.51–0.61), with a sensitivity of 21.2%, specificity of 91.4%, positive predictive value of 32.6%, and negative predictive value of 85.5%. Abnormal MCA PI showed an AUC of 0.49 (95% CI 0.46–0.52), with sensitivity and specificity of 4.5% and 94.0%, respectively. Abnormal mean uterine artery PI yielded an AUC of 0.506 (95% CI 0.48–0.53), with sensitivity of 4.5% and specificity of 96.7%. The abnormal CPR classification showed an AUC of 0.51 (95% CI 0.49–0.53), with very low sensitivity (3.0%) but high specificity (99.4%). Overall, none of the evaluated binary Doppler classifications demonstrated strong discriminatory ability for the composite adverse perinatal outcome.

Identifying growth disorders has enabled the detection of fetuses at risk for encountering negative perinatal events, facilitating their monitoring, and possibly enhancing fetal and neonatal results [ 1 ]. In standard obstetrics, fetal size is categorized as normal if it is at or above the 10th percentile and abnormal if it falls below the 10th percentile of estimated fetal weight (EFW) or abdominal circumference (AC) [ 2 ]. Fetal growth restriction (FGR) is typically defined as a fetal weight estimation below the tenth percentile for gestational age determined via prenatal ultrasound. The American College of Obstetricians and Gynecologists (ACOG) applies this criterion for FGR [ 3 ]. The revised definitions of Late FGR now go beyond fetuses in the 10th percentile for weight, including a decrease in growth rate as shown by a drop in weight or abdominal circumference by two quartiles, along with abnormal Doppler indices as part of the Doppler diagnostic guidelines [ 4 ]. However, distinguishing between a naturally small fetus and one that has growth restrictions due to medical issues remains a significant challenge, as some fetuses are naturally small due to their genetic growth potential [ 5 ]. Variations in fetal size are frequently inherent (e.g., parental dimensions, ethnicity, fetal gender). Approximately 30% are genuinely growth-restricted and face a higher likelihood of perinatal morbidity and mortality [ 6 ]. FGR can be divided into two primary categories, each possessing distinct traits, including onset at gestational age, ultrasound results, and pathological characteristics. FGR can be categorized based on the timing of onset: early-onset occurs before 32 weeks of gestation, while late-onset happens after 32 weeks [ 3 ]. Fetal growth restriction happens in as many as 10% of pregnancies and is the second leading cause of infant morbidity and mortality after premature birth. Besides its important perinatal effects, FGR also influences long-term health results [ 7 ]. The reported prevalence of FGR in Iran is 8% (ranging from 3% to 19%) based on various regions [ 8 , 9 ]. Nevertheless, while adverse perinatal outcomes are more frequent in fetuses with abnormal growth (i.e., EFW < 10th centile [ 10 ]), they also occur often in fetuses with appropriate-for-gestational-age, likely due to the presence of some placental dysfunction, even among those at higher EFW percentiles [ 8 – 11 ]. Data from the population indicates that among fetuses of normal weight for their gestational age, a lower percentile corresponds to an increased risk of mortality. Some studies indicate that fetuses within the 10th-25th percentile of weight percentile have a twofold increased risk of mortality [ 11 ]. Earlier reports highlighted that the likelihood of negative perinatal results, such as stillbirth, gradually diminishes as the EFW percentile rises. Nonetheless, no definitive percentile cutoff has been established that excludes this risk, complicating the monitoring of fetuses above the 10th percentile [ 12 , 13 ]. A span of 10th-90th percentiles has been historically regarded as appropriate-for-gestational age, with small for gestational age or large for gestational age frequently classified as < 10th or ≥ 90th percentiles, respectively. However, it is crucial to recognize that pathological fetal growth operates on a gradient, and varying percentile thresholds will yield different proportions of fetuses that are constitutionally small and not growth-restricted, and conversely for larger fetuses [ 14 ]. For example, research conducted on singleton births in the United Kingdom identified a higher risk of stillbirth and infant death for birthweights at or below the 25th percentile and at or above the 85th percentile, indicating that the frequently applied cut-off points of the 10th and 90th percentiles may overlook fetuses at risk for mortality [ 15 ]. Another investigation in the Netherlands revealed that the risk of perinatal mortality was greatest below the 2.3 percentile, followed by those in the 2.3 to < 5th and 5th to < 10th percentiles, with perinatal mortality exhibiting a “u-shape” pattern that reached its lowest point at the 80th to 84th percentiles for births occurring between 28 and 43 weeks [ 16 ]. These studies (though focused on birthweight) suggest that while extreme values correlate with elevated risks for perinatal mortality, there is a more consistent risk observed in the central portion of the distribution. This choice might vary in healthcare environments compared to its application in a public health framework for tracking and assessing the health and progress of populations. Ideally, the 25th percentile of a population reference would identify the fetuses most at risk for perinatal morbidity and mortality [ 14 – 16 ]. Since fetal growth restriction and low fetal weight present complications at any stage, the obstetric team aims to mitigate this risk by employing preventive, care, and diagnostic measures. Frequent ultrasound examinations throughout pregnancy to monitor fetal growth are regarded as a standard practice; however, because of the steep expense of ultrasounds, numerous mothers, especially in low-income countries, may find them unaffordable [ 17 , 18 ]. Therefore, a balance must be achieved between cost and benefit to guarantee that the best diagnostic methods are performed at the lowest expense and with the least number of ultrasounds, to prevent unwanted complications. In Iran, similar to many low-income nations, an ultrasound is conducted during the third trimester of pregnancy to assess fetal development. The absence of agreement on risk thresholds restricts the ability to compare studies and the execution of uniform monitoring strategies. The primary objective of this research was to assess negative neonatal outcomes in fetuses with an EFW falling between the 10th and 25th percentiles during the early third trimester (32.0 weeks of gestation), in comparison to those within the > 25th to < 90th percentile range, typically regarded as normal growth standards. Additional objectives involved assessing the progression toward FGR and low birth weight (LBW) among these groups. We proposed that fetuses within the 10th-25th percentile of weight would experience a considerably elevated risk of negative outcomes, highlighting the necessity for more research into this possibly at-risk group. We also aimed to examine the effectiveness of Doppler indices in identifying growth restriction in fetuses that appear to have appropriate weight for gestational age but fall within the lower quartile of the weight percentile, as well as the connection between these indices and perinatal outcomes. By promptly identifying growth restriction and at-risk fetuses, the necessity for ongoing monitoring and timely action is recognized to decrease perinatal mortality and morbidity. The secondary research question addresses whether an abnormal Doppler ultrasound within the 10th-25th percentile of birth weight can predict perinatal complications, and whether this test should be ordered. If that is the case, which Doppler index holds greater significance in predicting FGR and the associated complications?

The incidence of negative neonatal outcomes like FGR, LBW, RDS, low APGAR scores, and NICU admission was higher in fetuses in the 10th-25th percentile range. The occurrence of abnormal UA-PI, abnormal MCA-PI, and abnormal CPR was more prevalent in fetuses falling within the 10th-25th percentile weight group; nonetheless, fetal Doppler indices were unable to predict adverse neonatal outcomes for fetuses in this weight range. Upcoming extensive prospective studies should focus on categorizing fetuses with different percentile increases and incorporate a variety of populations and care environments to improve relevance.

This study aimed to investigate three main questions: (1) Is the incidence of adverse perinatal outcomes higher in fetuses within the 10th to 25th weight percentile than in those with > 25th–<90th percentile? (2) Does an unusual Doppler ultrasound in the 10th to 25th weight percentile indicate perinatal issues, and is it essential to request it? (3) Are abnormal Doppler parameters such as UA-PI, MCA-PI, CPR, and average uterine artery PI values more prevalent in fetuses within the 10th to 25th weight percentile compared to those in the > 25th–<90th percentile? In that scenario, which Doppler index is more useful in predicting neonatal adverse complications? According to the findings of our study involving 402 participants, 320 (79.6%) fell within the > 25th–<90th percentile range, whereas 82 (20.4%) were situated in the 10th-25th percentile range, the incidence of negative perinatal and neonatal outcomes like FGR (25.60% vs. 2.81%), LBW (20.73% vs. 0.31%), RDS (13.41% vs. 5.31%), low APGAR scores (14.63% vs. 5.31%), abnormal ABG (13.41% vs. 3.75%), and NICU admission (28.04% vs. 9.37%) was higher in the 10th-25th percentile group compared to the > 25th–<90th percentile group. Numerous studies have shown a significant connection between fetal and birth weight and negative pregnancy outcomes [ 20 – 22 ]. According to earlier research, the probabilities of stillbirth and neonatal death rose markedly by 2.6% (0.4%-4.6%) and 2.8% (1.0%-4.5%), respectively, for each 1 percentile decrease in fetal development. Additionally, the likelihood of having a SGA infant rose by 2.4% (2.2%-2.7%) for each percentile decrease. In contrast, diminished fetal growth was not linked to emergency cesarean sections for concerning fetal conditions or significant neonatal complications [ 23 ]. Another study indicated that the likelihood of negative neonatal outcomes rose notably when the fetal weight for gestational age fell below the 30th percentile. The ORs (95%CI) for mortality, necrotizing enterocolitis, bronchopulmonary dysplasia, severe retinopathy of prematurity, and sepsis in infants with fetal weight for gestational age within the 10th-30th percentile were 1.94 (1.56–2.41), 1.08 (0.79–1.47), 2.48 (2.03–3.04), 2.35 (1.63–3.39), and 1.39 (1.10–1.77), respectively [ 24 ]. Based on our findings, the risk of FGR (aRR: 9.23, 95% CI: [7.11–11.78]), LBW (aRR: 5.03, 95% CI: [3.57–6.97]), RDS (aRR: 1.98, 95% CI: [1.98–2.78]), and NICU admission (aRR: 2.87, 95% CI: [1.45–3.41]) in fetuses within the 10th-25th percentile weight range was greater compared to those in the > 25th- 25th-<90th percentile weight, while abnormal UA-PI, abnormal MCA-PI, and abnormal CPR were more commonly observed in fetuses with a weight in the 10th-25th percentile group, fetal Doppler indices was not able in predicting adverse neonatal outcomes. However, recent studies showed that abnormal Doppler parameters such as UA-PI, MCA-PI, and CPR were clearly corresponding with adverse perinatal outcomes in pregnancies experiencing FGR. Doppler sonography of the umbilical artery provides a non-invasive way to indirectly evaluate issues with the fetal and uteroplacental circulations, which are a significant factor in FGR. The Doppler indices such as systolic/diastolic ratio (S/D ratio), resistance index (RI), and PI have been shown to be associated with vascular resistance [ 25 ]. A meta-analysis of 128 studies (involving 47,748 women) revealed significant differences in reported sensitivities and specificities of each Doppler parameters in predicting adverse neonatal outcomes. According to their report, CPR surpassed UA-PI Doppler in predicting composite adverse outcomes and emergency delivery due to fetal distress. MCA-PI showed considerably poorer performance compared to UA-PI in predicting low APGAR scores and emergency deliveries due to fetal distress. CPR demonstrated superior performance compared to MCA-PI in predicting composite adverse outcomes and emergency delivery due to fetal distress [ 26 ]. The Doppler indices (S/D, PI, and RI) demonstrate differing levels of effectiveness in predicting the perinatal outcomes of FGR pregnancies. A study indicated a sensitivity of 75% and specificity of 73.8% in predicting adverse neonatal outcomes [ 27 ], similar to the research conducted by Chanprapaph [ 28 ] et al. (sensitivity of 66.7% and specificity of 78.85%) and Gyawali [ 29 ] et al. (sensitivity of 76% and specificity of 76.9%). Fleischer [ 30 ] and colleagues also found that the S/D ratio demonstrated a sensitivity of 78% for predicting FGR. In a similar manner, Wang [ 31 ] et al. reported that the sensitivity and specificity of the umbilical arterial S/D ratio for predicting FGR were 80.0% and 83.7%, respectively. However, according to our findings, ROC analysis to determine the performance of the measured obstetric Doppler parameters in predicting the development of adverse neonatal outcomes showed that UA-PI (AUC 0.56, 95% CI: 0.45–0.62), CPR (AUC 0.51, 95% CI: 0.43–0.56), mean uterine artery PI (AUC 0.50, 95% CI: 0.48–0.534), and MCA-PI (AUC 0.49, 95% CI: 0.46–0.53) had no predictive value for adverse neonatal outcomes. This discrepancy may arise from differences in the target groups, as our population comprises fetuses at normal weight percentiles (10th-25th percentile), whereas the other studies focused on FGR populations (EFW < 10%). Fetuses designated as “normal” based on their EFW [ 32 ] percentile may be susceptible to sentinel occurrences during labor or placental issues. Nonetheless, they might maintain a standard fetal weight since fetal dimensions merely indicate the transplacental transfer of nutrients [ 33 , 34 ]. Additionally, there is no EFW percentile beyond which positive results can be guaranteed. Our results highlight the necessity for additional investigation to clarify the spectrum of risk throughout the fetal weight distribution, especially concerning fetuses typically deemed “normal” but situated at the lower end of the percentile range. This research noted notably elevated rates of negative outcomes in the 10th–20th percentile group; nonetheless, fetal Doppler indices did not demonstrate any predictive capability for negative neonatal results in the 10th–20th percentile group. In addressing the key inquiry of our research, which asked, “Does abnormal Doppler ultrasound in the 10th to 25th weight percentile indicate perinatal complications, and is it essential to request it?“, our findings support further evaluation of Doppler surveillance in this population, preferably in larger multicenter prospective cohorts. Nonetheless, a biometric assessment for fetuses in the 10th to 25th weight percentile may be required. Recognizing fetuses at risk for FGR, LBW, or requiring neonatal intensive care can assist us in customizing the birth plan for patients, particularly for those lacking access to advanced birth centers [ 35 – 37 ]. Future comprehensive prospective research should aim at classifying fetuses with varying percentile increases and include diverse populations and care settings to enhance applicability. Our research has several strengths. This cohort study involved a group of fetuses for which maternal–fetal medicine specialists conducted the scans following standardized guidelines. The prospective aspect of the study design enabled us to observe the study participants. The sample size of our research was adequate when compared to earlier studies. It should be noted that although the sample size was determined using two anticipated ROC AUC values of 0.95 and 0.87, assuming a 3:1 ratio between the groups, the final population included 320 compared to 82 participants, yielding approximately 3.9:1, which could influence the analysis. Being a single-center cohort study, the findings may be affected by intervention bias, as perinatal outcomes might have been swayed by local follow-up protocols, management choices, and the judgment of the attending obstetricians. The study reports no cases of intrauterine fetal demise or oligohydramnios. Therefore, the study cannot provide meaningful estimates regarding these outcomes. Since several maternal conditions were excluded, potentially relevant factors such as maternal BMI, parity, socioeconomic characteristics, smoking status, and fetal sex may influence both fetal growth and neonatal outcomes, the result might be affected by the excluding such covarities.