A defined grey zone with in fetal growth curves for predicting adverse neonatal morbidity in fetuses being small for their gestational age at term in uncomplicated pregnancies

Background: “Small for gestational age” (SGA) is a term used to define an important risk factor for both neonatal morbidity and mortality. Our hypothesis suggests that adverse neonatal morbidity (ANM) in fetuses can occur when the birth weight is closer to 10th percentile. For example, although a fetus with a birth weight within the 11th or 12th percentile is appropriate for gestational age (AGA), it is difficult to clearly distinguish these fetuses from SGA fetuses for ANM; therefore we suggest defining a transition zone, or “grey zone”, for ANM. The aim of the present study was to examine ANM frequency in fetuses using this newly defined grey-zone percentile. Methods: This retrospective analysis comprised 7,817 pregnant women with uncomplicated pregnancies and single deliveries between 37 0/7 and 41 6/7 gestational weeks. The babies were divided into groups according to birth weight percentiles as follows: (1) SGA, (2) 10–20 percentile, and 21–90 percentile. The primary outcome was ANM, defined as any of the following: Apgar score <4 at 5 min; respiratory distress; mechanical ventilation; intraventricular hemorrhage, grade III or IV; necrotizing enterocolitis, stage 2 or 3; neonatal sepsis, stillbirth or neonatal death. Results: Demographic and obstetric characteristics of the mothers were similar among the groups. ANM rates were 10.7% in the SGA group, 6.8% in the 10–20 percentile group, and 2.1% in the 21–90 percentile group, a significant difference. ANM was 5-fold higher in the SGA group and 3.2-fold higher in the 10–20 percentile group than in the 21–90 percentile group. Delivery induction or augmentation, cesarean delivery for non-reassuring fetal heart rate or fetal distress, apgar score <4 at 5 min, mechanical ventilation, neonatal sepsis, stillbirth, or neonatal death significantly increased in the 10–20 percentile group compared

with those in the 21-90 percentile group. Conclusion: In uncomplicated pregnancies, ANM for SGA fetuses born at term are significantly worse than that for AGA fetuses.
Fetuses with a birth weight within the 10-20 percentile (grey zone) had a significant increased risk of ANM than those within the 21-90 percentile.

Background
Pregnant women are often seen who have serious concerns about the wellbeing of their babies because of clinically inadequate fetal growth or a fetus that is small for its gestational age (SGA) as confirmed by ultrasound. SGA is a complex and multifactorial condition and an important risk factor for both neonatal morbidity and mortality [1]. Gestational age at delivery, multiple gestation, presence of maternal comorbidity, hypertensive disease, and diabetes are the main determinants of adverse neonatal morbidity (ANM) in SGA fetuses [2][3][4][5][6]. In clinical practice, the majority of SGA fetuses born are below the 10 percentile in terms of gestational age. Lee et al. [7] have shown that in 2010, 29 million SGA fetuses were born at term in 138 developing countries.
The literature has clarified that SGA is related to ANM in uncomplicated term pregnancies [8][9][10][11]. The main hypothesis of our study was that when considering SGA, the birth weight between the 10th and 90th percentile is appropriate for gestational age (AGA), but that ANM can occur when the birth weight is nearer the 10th percentile. For example, although a fetus with a birth weight within11th or 12th percentile is defined as AGA, it is difficult to clearly distinguish these fetuses from those with SGA for ANM; therefore we suggest that there should be a transition zone (i.e.,a"grey zone") for increased ANM. Thus, the aim of the present study was to examine ANM frequency in fetuses within a newly defined grey-zone percentile.

Methods
This was a retrospective cohort study approved by the Ethics Committee of Erciyes University, Kayseri, Turkey (Decision no. 2019/283). The study was conducted at Kayseri City Hospital in accordance with the Declaration of Helsinki.
The study comprised 7,817 pregnant women who met the inclusion criteria and delivered at the Kayseri City Hospital between May 2018 and July 2019. The inclusion criteria were as follows: 1) pregnant women who delivered singletons between 37 0/7 and 41 6/7 weeks of gestation, 2) last menstrual period was used to determine gestational week, and 3) gestational age was calculated according to ultrasonographic measurements performed in the first trimester when the last menstrual period was unknown. The exclusion criteria were as follows: 1) pregnant women with multiple pregnancies; 2) preterm delivery before 37 weeks of gestation; 3) fetal chromosomal or congenital anomalies; or 4) tobacco, alcohol, or drug use. A pregnancy was considered complicated if a woman had any of the following: diabetes (pregestational or gestational), hypertensive disease of pregnancy (chronic hypertension, gestational hypertension, preeclampsia, or eclampsia), intrahepatic cholestasis of pregnancy, placenta previa, placental abruption, and nonobstetric morbidities. In the absence of any of these parameters, the pregnancy was defined as uncomplicated.
The 7,817 pregnant women were divided into three groups according to birth weight percentiles as follows: (1) SGA (n:390), (2) 10-20 percentile (n:750), and 21-90 percentile (n:6,677). SGA and other percentiles were determined using the Alexander growth curve for neonatal gestational age at delivery, birth weight, and sex [12]. Delivery induction or augmentation was preferred in the presence of oligohydramnios, anhydramnios, membrane rupture, or reduced fetal movements.
The new grey zone was defined as the fetal birth weight between the 10th and 20th percentile. To determine the cutoff value for the grey zone, we considered it appropriate to determine the sensitivity and specificity using the receiver operating characteristic curve; however, it was not possible to determine the fetuses individually in the form of 11,12,13,14,15 percentiles according to their birth weight. Therefore, we defined the grey zone as the 10-20 percentile range based on our clinical experience.
The primary outcome of study was the presence of ANM, which was defined as any of the following: Apgar score <4 at 5 min; respiratory distress syndrome; need for mechanical ventilation; intraventricular hemorrhage, grade III or IV; necrotizing enterocolitis, stage 2 or 3; neonatal sepsis, suspected or proved; confirmed seizure; stillbirth, or neonatal death. Each ANM parameter was previously defined in the study by Mendez-Figueroa et al. [8]. Stillbirth was defined as any fetal death occurring before or during labor, and neonatal mortality was defined as death after delivery or up to 28 d after birth. Maternal characteristics and ANM were compared among the groups.

Statistical analysis
To compare more than two groups, an analysis of variance followed by Tukey's posthoc test analyzed using Minitab 16 (MinitabInc., StateCollege, PA, USA) was used. To compare two groups, the Shapiro-Wilk test was used to determine the normality of the data, and the Levene's test was used to test the homogeneity of variance assumption. Values are expressed as the mean ± standard deviation. Parametric comparisons were made using the Student's t-test, and nonparametric comparisons were made using the Mann-Whitney U test. The difference among the groups was considered statistically significant when p <0.05.

Results
Of the 7,817 pregnant women with uncomplicated term pregnancies enrolled in the study, 390 neonates were in the SGA group (<10th percentile), 750 were in the 10-20 percentile group, and 6,677 were in the 21-90 percentile group. The demographic and obstetric characteristics of the mothers were compared and are provided in Table 1. Maternal age (p = 0.470), BMI<30 kg/m 2 rates (p = 0.486), nuliparity rates (p = 0.511), and previous cesarean delivery rates (p = 0.785) were similar among the groups. Table 2 shows the delivery outcomes and ANM results. The primary outcome assessed in the study was ANM, the rates of which were 10.7% in the SGA group, 6.8% in 10-20 percentile group, and 2.1% in 21-90 percentile group, which was a significant difference among the groups (p<0.001). ANM was 5-fold higher in the SGA group and 3.2-fold higher in the 10-20 percentile group than in the 21-90 percentile group. The gestational age at delivery was similar among the groups. The fetal birth weight was 2550±240g in SGA group, 2720±190 g in the 10-20 percentile group, and 3320±340g in the 21-90 percentile group, which was a significant difference among the groups (p<0.001). Although not as high as in SGA deliveries, delivery induction or augmentation, cesarean delivery for non-reassuring fetal heart rate or fetal distress, Apgar score <4 at 5 min, neonatal sepsis, stillbirth, or neonatal death significantly increased within the 10-20 percentile group compared with these in the 21-90 percentile group (p<0.001, p<0.001, p = 0.042, p<0.001, p<0.001, respectively).In addition, mechanical ventilation rates significantly increased in the 10-20 percentile group compared with those in the 21-90 percentile group (p = 0.001).

Discussion
The present study showed that in uncomplicated pregnancies, ANM for SGA fetuses born at term is significantly worse than that of AGA fetuses. In addition, fetuses with birth weights within the 10-20 percentile (grey zone) had a significantly increased risk of ANM compared to those within the 21-90 percentile, and this risk was significantly lower than that in SGA fetuses. Specifically, ANM was 5-fold higher in the SGA group and 3.2-fold higher in the 10-20 percentile group than that in the 21-90 percentile group.
In their multicenter prospective study, Mendez-Figueroa et al. [8] reported that composite neonatal outcomes were significantly higher in SGA newborns than in AGA newborns at term in uncomplicated pregnancies. Their study compared 5,416 SGA newborns with 44,595 AGA newborns for composite neonatal outcomes and their results showed that SGA newborns had 60% higher rates and 3-fold and 2-fold higher rates of stillbirth and neonatal mortality [8]. In another study, Chauhan et al.
[10] reported that hypoxic composite neonatal morbidity was significantly higher in SGA fetuses compared to that in AGA fetuses in uncomplicated term pregnancies.
Recently, the results of a large retrospective Australian study by Madden et al. [13] comprising 95,900 infants suggested that term SGA infants from low-risk women are at a significantly increased risk of neonatal mortality and morbidity. The study results showed that composite neonatal morbidity was 11.1% in the AGA group, 13.7% in the <10 percentile group, and 22.6% in the <5 percentile group [13].
The results of our study suggest that ANM for SGA fetuses born at term after uncomplicated pregnancies are significantly worse than that in AGA fetuses. These findings together with those of three recent publications [8, 10, 13] provide evidence of the ANM risks for SGA fetuses, even those born at term. Our study results also indicated that ANM was 5-fold higher in the SGA group and 3.2-fold higher in the 10-20 percentile group than that in the 21-90 percentile group. The ANM rates were 10.7% in the SGA group, 6.8% in the 10-20 percentile group, and 2.1% in the 21-90 percentile group. When reviewing similar studies that compared SGA with AGA fetuses [8, 10, 13], we found that ANM rates were lower in AGA fetuses, and that ANM rates in SGA fetuses were similar. It is possible to explain this situation by classifying those fetuses within the 10-20 percentile group separately from those within 20-90 percentile group.
The results of our study might contribute the following important indications in clinical practice:1) sound evidence from ultrasound examinations that detect SGA fetuses conflict with the results of other studies that show no benefit of these tests [14,15], while others have shown detection rates >50% [16][17][18]. In addition, Cochrane reviews have not confirmed any advantage to using either routine latepregnancy ultrasound or umbilical artery Doppler in low-risk populations [19,20]; however, there is evidence that shows that SGA is associated with ANM, even in a low-risk population, and that a late-pregnancy ultrasound will help to reduce the risk of ANM in SGA fetuses. It is possible to suggest that this evaluation is valid for fetuses within the 10-20 percentile; 2) our findings of increased ANM for SGA fetuses born at term support the American Congress of Obstetricians and Gynecologists (ACOG) guidelines and consensus that fetuses with restricted growth (weight in the < 10th percentile) without other risk factors should be delivered by 39.0 weeks [1]. Because our results showed an increased risk of ANM in the 10-20 percentile group, obstetricians might plan to deliver these fetuses at 39 weeks to reduce that risk.
Our study had both strengths and limitations. The strengths were its large sample size from a tertiary institution with clear evidence-based protocols that guided management. The main limitations were related to its retrospective nature and its focus within a single institution. In addition, newborns were divided into groups based on the Alexander growth curve instead of a customized growth curve because the latter has not consistently identified the characteristics of pregnancies with adverse outcomes and, more importantly, is not recommended by ACOG guidelines(1).

Conclusion
The present study showed that in uncomplicated pregnancies, the rates of ANM for SGA fetuses born at term are significantly higher than that for AGA fetuses. In addition, fetuses with birth weights within the 10-20 percentile (grey zone) had a significant increased risk of ANM than those with weights within the 21-90 percentile. Specifically, ANM was 5-fold higher in the SGA group and 3.2-fold higher in 10-20 percentile group than in the 21-90 percentile group.

Consent for publication
Participants gave their consent for publication.
Of all participants who were included in the study written informed consent was taken.

Availability of data and materials
The dataset used and analyzed during the current study is available from the corresponding author on reasonable request.

Competing interests
The authors declare that they have no competing interests.

Funding
This study was not funded by any specific grant from any funding agency in the public, commercial, or not-for-profit sectors.