Investigation of serum afamin concentration in pregnant women diagnosed with late fetal growth restriction or small for gestational age fetus

Abstract Objective To investigate the afamin concentration in the serum of pregnant women diagnosed with late fetal growth restriction (FGR) or small for gestational age (SGA) in the third trimester. Methods This prospective case-control study was conducted on 126 pregnant women, 42 of whom were diagnosed with late FGR in the third trimester, 43 were SGA, and 41 were healthy controls. The groups were compared in terms of maternal serum afamin concentrations. Results Three groups were similar in terms of demographic characteristics and gestational age at blood sampling for afamin (p < .05). The median afamin concentration was determined as 199 ng/mL in the late FGR group, 153 ng/mL in the SGA group, and 108 ng/mL in the control group (p = .000). In the post-hoc analysis, while maternal serum afamin concentrations were found to be significantly higher in the late FGR group and SGA group compared to the control group but, this significance could not be shown between the FGR group and the SGA group (p = .00001, p = .005, p = .137, respectively). In the ROC analysis, the optimal cutoff value of serum afamin concentration to predict late FGR was determined as 141 ng/mL, with a sensitivity of 66.6% and a specificity of 85.3%. Conclusions The serum afamin concentration in the third trimester was found to be higher in pregnant women with late FGR compared to the SGA and control groups. Although afamin is seen as a promising molecule in the clinical prediction of late FGR, this needs to be supported by large series of studies.


Introduction
Fetal growth during the intrauterine period depends on different factors, including the fetal genetic background, maternal nutrition, and adequate oxygen transfer and nutrients from the placenta to the fetus.Fetal growth restriction (FGR) means that the fetus does not reach its genetically predetermined growth potential before birth as a result of impaired placental functions [1].
When fetal growth retardation is detected, this condition was previously referred to as intrauterine fetal growth restriction (IUGR), but the definition became clear with the article '"Delphi procedure'" published in 2016 [2].According to this, in the absence of any fetal congenital anomaly, retardation in fetal development was classified as early FGR if it started before the 32nd gestational week, and as late FGR if it started at the 32nd gestational week or later.For early FGR, three solitary parameters as abdominal circumference (AC) < 3rd centile, estimated fetal weight (EFW) < 3rd centile, and absent end-diastolic flow in the umbilical artery, and also four contributory parameters as AC or EFW < 10th centile combined with a pulsatility index (PI) > 95th centile in either the umbilical or uterine artery were defined.For late FGR, two solitary parameters as AC or EFW < 3rd centile and four contributory parameters as EFW or AC < 10th centile, AC or EFW crossing centiles by > two quartiles on growth charts and cerebroplacental ratio < 5th centile or umbilical artery PI > 95th centile were defined [2].
Since fetuses with FGR are at high risk of adverse perinatal outcomes, it is important for clinical management to clearly distinguish FGR from small for gestational age (SGA) fetuses [3].If the fetal AC or EFW measurement is between the 3rd and 10th percentile but has normal uteroplacental and fetoplacental circulation, this fetus can be considered SGA [4].
SGA involves structurally small but mostly healthy fetuses and has a lower risk of abnormal perinatal outcomes [5].
Afamin was discovered in 1994 as the fourth member of the human albumin gene family, which includes albumin, a-fetoprotein, and vitamin D binding protein [6].In humans, the genes of these four proteins belonging to the albumin family are encoded in the 4q11-q22 chromosomal region.Afamin is a glycoprotein with a molecular weight of 87 kDa and 55% amino acid sequence identity to albumin, but unlike albumin, it is highly and complexly glycosylated [6].
Human plasma afamin is a specific binding protein for vitamin E. Afamin shows a specific binding affinity for both a-tocopherol and c-tocopherol, two of the most important forms of vitamin E. A maximum of 18 binding sites for vitamin E per afamin molecule have been estimated.It is thought that afamin is involved in the transport of vitamin E in body fluids due to its high binding capacity for vitamin E [7].
In a study by Dieplinger et al. serum afamin concentration was determined as 45-99 mg/L (median 68 mg/L) in healthy adults, independent of age and gender [8].It has been determined that afamin is expressed in different organs but, how much afamin is expressed in these tissues contributes to circulating afamin is still unclear [9].
In a different study conducted on uncomplicated pregnant women, it was shown that serum afamin concentration was 61.9 mg/L in the first trimester, 79.6 mg/L in the second trimester, and 98.6 mg/L in the third trimester.So, it was determined that maternal serum afamin concentration increased linearly throughout pregnancy, and after delivery median afamin concentrations were found to decrease to baseline values of 54.6 mg/L [10].
Atakul et al. investigated the afamin in the prediction of large for gestational age (LGA) fetuses in pregnant women with GDM.In the pregnant women with GDM, they found that maternal serum afamin concentration in the third trimester was significantly higher in pregnant women with LGA fetuses than in pregnant women with appropriate-for-gestational-age (AGA) fetuses.As a result of this study, the authors suggested that afamin concentrations may predict the risk of LGA babies independently of glycemic control status and that afamin may play a role in the pathogenesis of LGA fetuses [11].
In light of this information, we hypothesized that the afamin molecule is associated with fetal growth in the advanced weeks of pregnancy, therefore, maternal serum afamin concentration may be useful in predicting late FGR.Clinicians are generally more cautious in the management of pregnant women with early FGR as it is associated with more severe adverse perinatal outcomes.We think that the main problem in practice is the differential diagnosis and management of late FGR cases from SGA cases.In this context, we aimed to investigate the serum afamin concentrations in pregnant women diagnosed with late FGR and its usefulness in the differential diagnosis of late FGR.

Materials and methods
This prospective case-control study was conducted with 126 pregnant women who applied to the Gynecology and Obstetrics Clinic of Umraniye Training and Research Hospital, Istanbul, Turkey between September 2021 and July 2022 and were followed up and delivered in our hospital.In the pregnancy followups, 42 pregnant women who were diagnosed with late FGR after 32 weeks of gestation were included in the late FGR group, 43 pregnant women who were diagnosed with SGA were included in the SGA group, and 41 healthy pregnant women who had AGA fetuses in the third trimester formed the control group.Three groups were formed by matching maternal age, BMI, and the gestational week at blood sampling to avoid any confounding factor on maternal serum afamin concentrations.
Gestational age was calculated according to the last menstrual period and confirmed by fetal crown-rump length measured in the first trimester.Serial fetal biometric measurements and percentiles were recorded as the pregnancy follow-up of the participants until delivery was done in our clinic.Late FGR and SGA groups were formed using the criteria reported in the "Delphi procedure" reported in 2016 and the ISUOG Practice Guidelines published in 2020.Accordingly, cases with no congenital anomaly and occurring at or after 32 weeks of gestation, with fetal AC or EFW values below the 3rd percentile, and with at least two of the following four findings were accepted as late FGR; EFW or fetal AC below the 10th percentile, fetal AC or EFW percentiles cut > two quarters on growth charts and a cerebroplacental ratio below the 5th percentile or umbilical artery PI above the 95th percentile.SGA was diagnosed in pregnant women whose fetal AC or EFW values were between the 3rd and 10th percentiles according to the gestational week, and whose umbilical artery was Doppler velocimetry values and so cerebroplacental ratios were normal [2,4].
Pregnant women with multiple pregnancies, a history of pregestational or gestational systemic disease, known vascular disease, thrombophilia or autoimmune disease, congenital uterine anomaly, pregnant women using aspirin or low molecular weight heparin, and smokers were excluded from the study.Those who were diagnosed with FGR and additionally developed gestational hypertension, preeclampsia, or HELLP syndrome were not included in the study.Pregnant women with known chromosomal or structural abnormalities in themselves, their partners, or their fetuses were not included in the study.In addition, pregnant women who were in the high-risk group in fetal chromosomal anomaly screening tests were not included in the study.
Participants' age, BMI, and obstetric histories were recorded.Fetal biometric and umbilical artery Doppler velocimetry measurements were performed by the same obstetrician on the same ultrasound device (Hitachi Aloka F37 Ultrasound Device).
Approximately 5 mL of blood samples were drawn at any time of the day during the third trimester to investigate serum afamin concentrations in the participants.Blood samples taken into biochemistry tubes were kept at room temperature for about 20 min and then centrifuged at 4000 rpm for 10 min.After centrifugation, the supernatant was separated and stored at À80 degrees.
Afamin concentrations were measured with the Human Afamin Elisa Kit, using the Architect I2000 SR immunoassay analyzer (Abbott Diagnostics, Abbott Park, IL).For the Human Afamin Elisa Kit used in the study, the inter-measurement (between runs) value was 5-600 ng/mL, and the sensitivity was determined as 2.38 ng/mL.
The late FGR, SGA, and control groups were compared in terms of maternal serum afamin concentrations as the primary outcome of the study.
Statistical analysis of the study was performed with the Statistical Package for Social Sciences (SPSS) program version 21 (IBM, Armonk, NY).Descriptive statistics were presented as numbers and percentages for categorical variables, such as mean, median, minimum, maximum, and standard deviation (SD) values for numerical variables.Homogeneity was performed according to Levene's test and p > .05 was evaluated as homogeneous.Whether the distribution of continuous variables was close to normal or not was accepted as a normal distribution with the Kolmogorov Smirnov and Shapiro Wilk test p > .05 or within ±1.5 with the skewness and kurtosis test.For numerical variables, the One Way ANOVA Test was used when the normal distribution assumption was provided in the triple independent group comparisons, and the Kruskal Wallis Test was used when it was not provided.Nominal variables were analyzed with Pearson's Chisquare or Fisher's Chi-square test.The direction and level of the relationship between numerical variables were determined by Pearson correlation.The receiver operating curve (ROC) was used to determine the effectiveness of afamin in predicting late FGR and its significant threshold.Statistical significance was accepted as p < .05for all values.

Results
When the late FGR, SGA, and control groups were compared in terms of demographic characteristics, three groups were similar in terms of age, BMI, and parity (p ¼ .354,p ¼ .973,p ¼ .703,respectively) (Table 1).
When the three groups were compared in terms of fetal biometric measurements, fetal BPD, AC, FL, and EFW measurements were significantly higher in the control group than in the late FGR and SGA groups, as expected (p ¼ .000,for each).When the three groups were compared in terms of umbilical artery Doppler velocimetry values, umbilical artery PI and S/D values were found to be significantly higher in the late FGR group, while RI was similar in the three groups (p ¼ .000,p ¼ .000,p ¼ .96,respectively) (Table 2).
There was no difference between the three groups in terms of mode of delivery, 1st minute, and 5th minute Apgar scores (p ¼ .750,p ¼ .305,p ¼ .148,respectively).Gestational age at birth, birth weight, and height were significantly higher in the control group than in the late FGR and SGA groups (p ¼ .000,for all) (Table 3).There was no fetal loss in the intrapartum or postnatal period in any of the participants included in the study.
Three groups were compared in terms of maternal serum afamin concentrations.While the highest maternal serum afamin concentration was found in the late FGR group at 199 ng/mL, it was determined as 153 ng/mL in the SGA group and 108 ng/mL in the control group (p ¼ .000).In the post-hoc analysis performed to understand this significant difference, serum afamin concentrations in the late FGR group and SGA group were found to be significantly higher than in the control group (p ¼ .00001,p ¼ .005,respectively).Although serum afamin concentration was higher in the late FGR group than in the SGA group, this was not statistically significant (p ¼ .137)(Table 4).
According to Pearson Correlation Analysis, there was a significant and negative correlation between     5).ROC analysis was performed to determine the value of maternal serum afamin concentration in terms of predicting late FGR.AUC analysis of maternal serum afamin for late FGR estimation was 0.795 (p < .00001,95% CI ¼0.69 À 0.87).The optimal cutoff value for maternal serum afamin concentration was determined as 141 ng/mL with 66.6% sensitivity and 85.3% specificity (Figure 1).

Discussion
In this study, maternal serum afamin concentration in the third trimester was found to be higher in pregnant women whose pregnancy was complicated with late FGR compared to SGA and control groups.While this difference was significant in the control group, we think that it could not reach statistical significance in the SGA group due to the small number of participants.
In uncomplicated pregnancies, it was determined that the median afamin concentration, which was approximately 61.9 mg/L in the first trimester, increased to 98.6 mg/L in the third trimester and decreased to 54.6 mg/L after delivery [10].To evaluate the source of the linear increase in maternal serum afamin concentration in uncomplicated pregnancies, afamin expression in the placenta was investigated but afamin expression could not be demonstrated in both the first trimester and term placenta [10].Although the reason for the increase in maternal serum afamin concentration during pregnancy has not been clearly revealed, it is thought that the change in the hormonal status of the expectant mother stimulates the expression of the afamin gene in the liver.This increase may also be thought of as an effect of hormonal regulation (estrogen-induced) of hepatic synthesis of lipids and lipoproteins, which leads to physiological hyperlipidemia during pregnancy [12].Some studies in the literature have shown that serum afamin is strongly associated with metabolic syndrome parameters such as high BMI, glucose intolerance, dyslipidemia, and hypertension [13].In   THE JOURNAL OF MATERNAL-FETAL & NEONATAL MEDICINE light of this information, researchers began to investigate the relationship between afamin and pregnancyrelated diseases.K€ oninger et al. compared the first-trimester maternal serum afamin concentration in the group that developed preeclampsia (PE) during pregnancy and in normal healthy pregnant women.The median serum afamin concentration in the first trimester was found to be significantly higher in the group that developed PE compared to the control group (101.81mg/L, 86.40 mg/L, respectively) [14].In a different study, second-trimester maternal serum afamin concentration was evaluated in the prediction of PE.Second-trimester median maternal serum afamin concentration was found to be significantly higher in the group that developed PE in late pregnancy compared to the control group (145.3 ng/mL, 46.9 ng/mL, respectively) [15].
In a different study, Tramontana et al. investigated the relationship between first trimester maternal serum afamin concentrations and pregnancy-related complications.While first-trimester maternal serum afamin concentration was found to be significantly higher in women whose pregnancy was complicated with PE and gestational diabetes mellitus (GDM), compared to the healthy controls, afamin concentration was found to be similar in the group with IUGR or preterm delivery compared to healthy controls [16].In contrast to this study, we found that serum afamin concentration was significantly higher in the third trimester in the late FGR group compared to healthy controls.Although the maternal serum afamin concentration in the first trimester did not show a significant difference between the IUGR and control groups in the above-mentioned study, we think that the high afamin concentrations, which already increased as the gestational week progressed, reached a significant difference between the late FGR and control groups in the third trimester in our study.
Researchers also investigated the relationship between afamin and GDM.K€ oninger et al. evaluated the utility of afamin as an early marker in predicting GDM.In this study, both first and second-trimester maternal serum afamin concentrations were found to be significantly higher in patients with pathological OGTT results compared to patients without GDM [17].Similarly, in another study published in 2021, first-trimester maternal serum afamin concentration was found to be higher in patients who developed GDM during pregnancy compared to the group that did not develop GDM [18].In a study conducted by Wnag et al. in 2021, they developed a formulation that can be used to predict GDM in early pregnancy by using maternal serum afamin, maternal age, triglyceride, and platelet/lymphocyte ratio [19].
In a different study, Atakul et al. found that the serum afamin concentrations of pregnant women with GDM were higher than those without GDM, in line with other studies reported in the literature.Interestingly, the authors determined that serum afamin concentrations in pregnant women with LGA fetuses in the GDM group were significantly higher than those in the GDM group with AGA fetuses.As a result, they stated that afamin is associated with fetal growth independent of glycemic control and can be used to predict LGA fetuses in pregnant women with GDM [11].
At the beginning of our study, we hypothesized that the maternal serum afamin concentration in the group with late FGR would be lower than in the control and SGA groups.However, on the contrary, we found the highest maternal serum afamin concentration in the late FGR group.
Interestingly, the fact that afamin was detected to be high in maternal serum in both pregnant women with LGA fetuses in GDM and pregnant women diagnosed with late FGR makes the relationship between afamin and intrauterine fetal growth worth investigating.
To the best of our knowledge, this is the first study in the literature to examine maternal serum afamin concentration in the third trimester in pregnant women diagnosed with late FGR.The limitations of this single-center study are the small number of participants, the evaluation of serum afamin concentration only in the third trimester, and that it was not evaluated in early FGR cases.
In conclusion, third trimester serum afamin concentration was found to be higher in pregnant women whose pregnancy was complicated with late FGR compared to SGA and control groups.This is a preliminary study and we think that prospective studies with larger series are needed to determine whether maternal serum afamin is a useful molecule for routine use in the clinical prediction of late FGR.

Figure 1 .
Figure 1.ROC analysis for sensitivity, specificity, and positive, and negative predictive value of maternal serum afamin concentration in late fetal growth restriction.

Table 1 .
Comparison of groups in terms of demographic features.

Table 2 .
Comparison of the groups in terms of fetal biometric measurements and umbilical artery Doppler values.

Table 3 .
Comparison of the groups in terms of perinatal outcomes.
Ã One-way ANOVA, ÃÃ chi-square test, FGR: fetal growth restriction; SGA: small for gestational age.p1 represents a comparison between late FGR and SGA groups.p2 represents a comparison between late FGR and the control groups.p3 represents a comparison between SGA and the control groups.

Table 4 .
Comparison of the groups in terms of maternal serum afamin concentrations.One-way ANOVA, ÃÃ Kruskal Wallis Test, FGR: fetal growth restriction; SGA: small for gestational age.p1 represents a comparison between late FGR and SGA groups.p2 represents a comparison between late FGR and the control groups.p3 represents a comparison between SGA and the control groups.

Table 5 .
Correlation between maternal serum afamin concentrations and late FGR-related parameters.