Does Fetal Size Affect Umbilical Artery Pulsatility Index in Pregnancies Complicated by Gestational Diabetes?

Abstract

Objectives: Excessive fetal growth is the most common fetal complication associated with gestational diabetes (GDM), resulting in adverse short- and long-term outcomes. Our main objective was to evaluate the influence of excessive fetal growth on Doppler ultrasonographic measurements of the Umbilical Artery (UA) among women with GDM during the third trimester of pregnancy. Methods: A retrospective study among 472 women with GDM was conducted. UA-PI was measured by Doppler ultrasonography three different times during the third trimester of pregnancy at 28, 32, and 36 weeks. Pregnancies were grouped according to the fetal weight centile or birthweight in two groups: large for gestational age (LGA) group (>90th percentile or ≥4000 g at birth) and adequate for gestational age (AGA) group (<90th percentile or <4000 g at birth, not including the intrauterine growth restrictions). Results: In the LGA group (n = 57, 12.1%), women had higher BMI (p = 0.0001) and fasting blood glucose than the AGA group (97.08 ± 40.69 vs. 86.29 ± 39.58 mg/dL; p = 0.0550). They required insulin therapy more frequently to achieve glycemic control (63.2% vs. 34%, p = 0.0001). In LGA, UA-PI decreased progressively from 28 to 36 weeks (p = 0.0048). The most pronounced reduction occurred at 32 weeks (p = 0.0076). Conclusions: All fetuses from mothers with GDM had a significant and progressive decline in UA-PI during the third trimester of pregnancy. LGA fetuses showed lower UA-PI values compared with AGA fetuses. Since maternal hyperglycemia increases the risk of fetal overweight and GDM may represent a fetal vascular disorder, it therefore seems possible that in LGA fetuses, maternal hyperglycemia could influence the fetal vasculature.

1. Introduction

The pathophysiological processes underlying the unfavorable fetal outcomes associated with gestational diabetes (GDM) appear to be multifactorial and complex and not yet fully understood [1,2].

Excessive fetal growth is the most common fetal complication associated with GDM, resulting in adverse short- and long-term outcomes. Macrosomia, defined as neonatal weight over 4000 g, is an independent risk factor for acute fetal distress, shoulder dystocia, birth trauma, emergency cesarean section, perinatal morbidity, respiratory distress syndrome, and neonatal hypoglycemia [3,4,5].

Of concern, however, is that macrosomia is involved in “fetal programming”, which plays a seminal role in determining health trajectories across the lifespan [6]. From fetal life, a metabolic memory is established; this may be associated with the development of metabolic diseases later in life, including type II diabetes, obesity, hypertension, dyslipidemia, and cardiovascular disease [3,7].

Fetal weight appears to be related to the size of the placenta and its vascularization [8]. Placental weight tends to be heavier in diabetic pregnancies, and the placental-to-fetal weight ratio is higher than in normal pregnancies [8].

Furthermore, the architecture and vascularization of the placenta affect the Pulsatility Index (PI) values of the Umbilical Artery (UA).

Currently, fetal well-being in pregnancies affected by GDM and fetal macrosomia is evaluated using ultrasonographic biometry to estimate fetal weight, umbilical artery Doppler assessments, and fetal heart monitoring [9]. UA-PI is recognized as a valuable parameter for assessing fetal health and optimizing delivery timing, particularly in cases complicated by gestational hypertension, preeclampsia, and intrauterine growth restriction (IUGR). UA-PI is considered a powerful index to evaluate fetal well-being and to achieve an adequate timing of delivery, mostly in pregnancies complicated by gestational hypertension, preeclampsia, and intrauterine growth restriction (IUGR) [10].

However, its use in diabetic pregnancies has shown conflicting results, and a lack of predictability for adverse perinatal outcomes was generally found [11,12,13,14,15,16]. The ability of Doppler measurements to reflect maternal glycemic control is also controversial [14,16].

Multiple studies have examined variations in umbilical artery Doppler measurements in pregnancies affected by GDM or Type 1 diabetes, specifically in relation to macrosomia [17,18]. Findings indicate that LGA fetuses tend to have lower UA-PI values compared to controls. Additionally, a direct correlation has been observed between UA-PI and both birthweight and birthweight centile [17], suggesting that UA-PI may serve as a predictor of birthweight in women with GDM [19,20].

However, we do not have much data on what time in pregnancy this difference is established and what the trend of UA Doppler velocimetry has been since the diagnosis of GDM was made.

The primary objective of this study is to determine whether large for gestational-age (LGA) fetuses in pregnancies complicated by GDM show different UA Doppler PI values than adequate for gestational-age (AGA) fetuses. Our secondary objective is to evaluate the UA-PI trend during the third trimester of pregnancy at three different time points.

Macrosomia is a common complication of GDM, increasing the risk of birth trauma and metabolic disorders. While Doppler ultrasound is used to assess fetal well-being, its role in evaluating vascular changes in LGA fetuses remains unclear. This study examines UA-PI trends in GDM pregnancies, exploring whether excessive fetal growth alters fetal vascular resistance, which could improve prenatal monitoring and delivery planning.

We hypothesized that LGA fetuses would exhibit significantly lower UA-PI values than AGA fetuses due to increased fetal metabolic demands and placental vascular adaptations associated with excessive fetal growth. Prior research suggests that macrosomic fetuses, particularly in pregnancies complicated by GDM, experience changes in umbilical artery blood flow, likely driven by altered placental resistance and fetal hyperinsulinemia [11,12,13,14,15,16,17,18]. This may contribute to a more pronounced decline in UA-PI as gestation progresses.

2. Materials and Methods

A retrospective observational study was conducted at the high-risk pregnancy referral center of the Department of Gynecology and Obstetrics, University Hospital Maggiore della Carità, Novara, Piedmont, Italy, from January 2023 to February 2025.

We considered 472 pregnant women diagnosed with GDM. All participants were screened for GDM using a 75 g oral glucose tolerance test (OGTT) between 24 and 28 weeks of gestation. Women identified as high-risk for GDM—due to factors such as a history of GDM, BMI ≥ 30, or fasting plasma glucose (FPG) levels between 100 and 125 mg/dL in the first trimester—underwent early screening at 16–18 weeks. If their glucose levels were within the normal range, they repeated the OGTT at 24–28 weeks [21]. The IADPSG criteria were applied to establish the diagnosis of GDM [22].

After GDM diagnosis, patients were educated to self-monitoring their blood glucose. Nutritional counseling with the dietary plan was provided, together with lifestyle intervention advice. If the glycemic values were above the target recommended by the Italian National Guidelines in more than half of the measurements during the first 2 weeks of evaluation, insulin treatment was started [21].

All pregnancies were singleton and accurately dated by early ultrasonographic examination. Gestational age was calculated based on the last menstrual period and was confirmed in all cases with an ultrasound assessment at 11–12 weeks in which the crown-rump length (CRL) was measured. Maternal age was not a limiting factor for study inclusion, as we aimed to evaluate UA-PI trends across a representative population of pregnancies complicated by GDM. However, age-related factors, such as increased risk for GDM and fetal macrosomia in older mothers, were recorded and analyzed as potential confounders in our study population.

Patients with twin pregnancies, fetal aneuploidies, major fetal defects, fetal weight centile < 10th, and gestational age at delivery < 37 weeks (preterm) were excluded. Women were excluded from this study if they had significant preexisting maternal conditions, including insulin-dependent type 1 or type 2 diabetes, cardiac, renal, or liver disease, chronic infections, or autoimmune disorders such as antiphospholipid antibody syndrome or systemic lupus erythematosus. Other exclusion criteria included chronic hypertension requiring medication before pregnancy, chromosomal or genetic abnormalities, and a history of any type of cancer. Collected data included maternal age, ethnicity, body mass index (BMI), gestational weight gain, the onset of pregnancy (in vitro fertilization -IVF- or spontaneous), smoking habits during pregnancy, obstetric history including parity (nulliparous if no previous pregnancies with an outcome > 24 weeks gestation), previous gestational diabetes or macrosomia, gestational age at delivery, mode of delivery, birthweight percentile, neonatal birthweight, fetal sex, Apgar score, NICU admission. GDM management (diet only versus diet plus insulin treatment) was also recorded.

All ultrasound examinations and Umbilical Artery (UA) velocity waveform recordings were obtained with a Samsung HERA W9 and Samsung WS80 ultrasound equipment (Samsung Healthcare Global) equipped with 2–5 MHz transabdominal probes and using color and pulsed wave Doppler functions. All ultrasound scans and measurements, including Doppler assessment, were performed by operators with extensive experience in obstetric ultrasound and a certification of competence granted by The Fetal Medicine Foundation (FMF, King’s College Hospital, London, UK) to minimize inter-observer variability.

UA Doppler measurements were recorded from a free-floating loop of the umbilical cord, and the Doppler sample volume was placed over an artery and the vein. The Pulsatility Index (PI) was measured over at least three consecutive and uniform heart cycles [23,24]. UA-PI measurements were recorded three different times during the third trimester of pregnancy at 28, 32, and 36 weeks.

Fetal biometric measurements, including the abdominal circumference (AC), head circumference (HC), biparietal diameter (BPD), femur length (FL), assessment of amniotic fluid volume, and Doppler measurements (Middle Cerebral Artery, Ductus Venosus, and Uterine Arteries) were performed according to the ISUOG practice guideline [25].

LGA refers to infants whose birthweight exceeds the 90th percentile for growth at a specific gestational age [4].

Thus, pregnancies were grouped according to fetal weight centile or birthweight in two groups: large for gestational age (LGA) group (>90th percentile or ≥4000 g at birth) and adequate for gestational age (AGA) group (<90th percentile or <4000 g or).

Statistical analysis

Clinical data were extracted by reviewing the medical records of all women followed in our clinic in a standardized electronic database. UA Doppler measurements were extracted from the Astraia (Nexus/Astraia, Ismaning, Germany) software (version 29.2.1). Quantitative variables were calculated as the mean and standard deviation, whereas qualitative variables were calculated as absolute and relative frequencies. The Student’s t-test was used to compare the means of the two groups for normally distributed continuous variables. The Mann–Whitney U test was used when continuous variables did not follow a normal distribution. The chi-squared test or Fisher’s exact test, where appropriate, was used for comparisons of categorical variables. Statistical significance was set at a p-value < 0.05. All statistical analyses were performed using the Graph Pad Prism 6 software.

3. Results

We analyzed 472 pregnant women with GDM. The women were divided into two groups: 57 women (12.1%) delivered LGA babies (cases), while 415 women (87.9%) delivered AGA babies (controls). There were no significant differences between the two groups according to the maternal characteristics related to age and parity, as reported in

Table 1. Demographic, anamnestic, and obstetrical characteristics of study groups.

	LGA n = 57	AGA n = 415	p-Value
Age (years)	33.09 ± 5.56	33.54 ± 5.55	0.5604
Nationality – Italian – Immigrant	34 (59.7) 23 (40.3)	275 (66.3) 140 (33.7)	0.3729
Smoking	3 (5.3)	22 (5.3)	1.0000
Parity – Nulliparous – Multiparous	16 (28.1) 41 (71.9)	161 (38.8) 254 (61.2)	0.1443
Pregravidic BMI (kg/m2) BMI at delivery (kg/m2)	28.77 ± 6.78 32.58 ± 7.26	26.25 ± 5.34 29.15 ± 4.95	0.0013 0.0001
GWG (kg)	10.36 ± 4.89	7.84 ± 5.63	0.0014
Onset of pregnancy – Spontaneous – IVF	53 (93) 4 (7)	388 (93.5) 27 (6.5)	0.7797
Previous macrosomia	15 (26.3)	18 (4.3)	0.0001
Previous GDM	21 (36.8)	61 (14.7)	0.0001
GDM therapy – Diet – Diet plus insulin – Long-acting insulin – Rapid insulin analogue – Combined therapy	21 (36.8) 36 (63.2) 15 (41.7) 2 (5.6) 19 (52.7)	274 (66) 141 (34) 78 (55.3) 13 (9.2) 50 (35.5)	0.0001 0.1903 0.7397 0.0838
Fasting blood glucose I trimester (mg/dL)	97.08 ± 40.69	86.29 ± 39.58	0.0550
Fasting blood glucose at OGTT (mg/dL)	99.74 ± 47.96	94.35 ± 44.60	0.3972
Glycated hemoglobin (%)	5.4 ± 1.23	5.44 ±1.16	0.8114

Data are expressed as mean ± standard deviation or as absolute number (percentage). p < 0.05 was considered statistically significant (marked in bold). BMI: body mass index; GDM: gestational diabetes mellitus; GWG: gestational weight gain; IVF: in vitro fertilization; OGTT: oral glucose tolerance test.

. The LGA group had a mean age of 33.09 ± 5.56 years, while the AGA group had 33.54 ± 5.55 years. Nulliparous women were 28.1% in the LGA group versus 38.8% in the AGA group. No difference was reported regarding the onset of pregnancy between groups (Table 1).

Pregestational BMI and BMI at delivery were significantly higher in the LGA group than in the AGA group (pre-BMI = 28.77 ± 6.78 Kg/m² versus 26.25 ± 5.34 Kg/m², p = 0.0013; at delivery BMI = 32.58 ± 7.26 Kg/m² versus 29.15 ± 4.95 Kg/m², p = 0.0001). Similarly, gestational weight gain was also higher in the LGA group (+10.36 ± 4.89 Kg) than in the AGA group (+7.84 ± 5.63 Kg, p = 0.0014).

Among multiparous women, previous macrosomia and previous GDM were significantly more common in the case group (26.3% and 36.8%, respectively) than in the control group (4.3% and 14.7%, respectively, p = 0.0001) (Table 1).

In the LGA group, glycemic balance was most frequently achieved by adding insulin therapy to the diet (63.2% versus 34%, p = 0.0001). In contrast, in the control group, the diet alone was more frequently able to keep maternal glycemia under control (66% versus 36.8%, p = 0.0001). Among the different insulin regimens used (long-acting insulin, rapid insulin analog, or combined therapy), no difference between the groups was reported (Table 1).

Fasting blood glucose measured during the first trimester was higher in the LGA group (97.08 ± 40.69 mg/dL) than in the AGA group (86.29 ± 39.58 mg/dL) with a p-value = 0.0550. Fasting blood glucose at OGTT had no significant difference between the two groups (99.74 ± 47.96 mg/dL versus 94.35 ± 44.60 mg/dL; p = 0.3972). Glycated Hemoglobin was also similar; 5.4 ± 1.23% in LGA group and 5.44 ± 1.16% with a p-value = 0.8114) (Table 1).

Gestational age at delivery was similar between the two groups (38.23 ± 1.60 versus 38.62 ± 1.73 weeks; p = 0.1100). A higher percentage of cesarean deliveries was reported in the LGA group (45.6% versus 28%, p = 0.0086). However, there were no statistically significant differences in the rate of postpartum hemorrhage (14% versus 12.3%, p = 0.6718) (

Table 2. Delivery and neonatal outcomes.

	LGA n = 57	AGA n = 415	p-Value
Gestational age at delivery (weeks)	38.23 ± 1.60	38.62 ± 1.73	0.1100
Mode of delivery – VD – CS	31 (54.4) 26 (45.6)	299 (72) 116 (28)	0.0086
PPH	8 (14)	51 (12.3)	0.6718
Fetal sex – M – F	28 (49.1) 29 (50.9)	230 (55.4) 185 (44.6)	0.3967
Birthweight (g)	3882.63 ± 453.21	3161.35 ± 523.22	0.0001
Apgar score – 1st minute – 5th minute	8.30 ± 1.21 8.86 ± 0.93	8.30 ± 1.59 8.87 ± 0.78	0.9980 0.8944
Apgar score < 7	4 (7)	38 (9.1)	0.8045
NICU admission	5 (8.8)	36 (8.7)	1.0000

Data are expressed as mean ± standard deviation or as absolute number (percentage). p < 0.05 was considered statistically significant (marked in bold). CS: cesarean section; F: female; g: grams; M: male; NICU: neonatal intensive care unit; PPH, postpartum hemorrhage; VD: vaginal delivery.

). Birthweight was significantly higher in cases (3882.63 ± 453.21 g) versus controls (3161.35 ± 523.22 g; p = 0.0001), but fetal sex did not differ between the groups. Apgar score at 1st and 5th minute, Apgar score <7, and NICU admission also did not differ significantly between cases and controls (Table 2).

We recorded UA-PI at three different stages of pregnancy at 28, 32, and 36 weeks. We compared the different values within each group and then compared them between the two groups analyzed.

In the LGA group, the UA-PI value decreased significantly and progressively from the first monitoring at 28 weeks to 36 weeks (p = 0.0048). The most pronounced reduction occurred between 28 and 32 weeks (p = 0.0076). Between 32 and 36 weeks UA-PI decreased further but without reaching the statistically significant difference (

Table 3. UA-PI measurements during the third trimester of pregnancy, according to the fetal weight centile.

	LGA n = 57	AGA n = 415	p-Value
UA-PI at 28 ws	1.12 ±0.46	1.05 ± 0.44	0.2643
UA-PI at 32 ws	0.88 ± 0.48	0.97 ± 0.52	0.4353
UA-PI at 36 ws	0.87 ± 0.46	0.87 ± 0.47	0.9579
	0.0076 1 0.9096 2 0.0048 3	0.0115 1 0.0083 2 0.0001 3

Data are expressed as mean ± standard deviation. p < 0.05 was considered statistically significant (marked in bold). UA-PI: umbilical artery pulsatility index; ws: weeks. ¹ UA-PI P-value calculated from 28 to 32 weeks of gestation, respectively, in LGA and AGA groups. ² UA-PI P-value calculated from 32 to 36 weeks of gestation, respectively, in LGA and AGA groups. ³ UA-PI P-value calculated from 28 to 36 weeks of gestation, respectively, in LGA and AGA groups.

).

Also, in the AGA group, there was a significant decrease in UA-PI value in each period from 28 to 32 weeks (p = 0.0115), from 32 to 36 weeks (p = 0.0083), and from 28 to 36 weeks (p = 0.0001) (Table 3).

Over the course of pregnancy, UA-PI values were lower in the LGA group than in the AGA group. These differences were most evident at 32 and 36 weeks, although they did not reach statistical significance (Table 3, Figure 1).

In both groups, the decrease in UA-PI is significant between 28 and 32 weeks, but in the LGA group, this drop was even more pronounced (Figure 1).

Furthermore, we also subdivided the participants into two groups based on babies’ weight at birth (more than 4000 g or less) and measured the UA-PI values within these two groups (

Table 4. UA-PI measurements during the third trimester of pregnancy, according to birthweight.

	≥4000 g n = 29	<4000 g n = 442	p-Value
UA-PI at 28 ws	1.08 ± 0.38	1.06 ± 0.45	0.7961
UA-PI at 32 ws	0.88 ± 0.48	0.96 ± 0.52	0.4160
UA-PI at 36 ws	0.82 ± 0.42	0.88 ± 0.47	0.5207
	0.0771 1 0.6393 2 0.0167 3	0.0016 1 0.0192 2 0.0001 3

Data are expressed as mean ± standard deviation or as absolute number (percentage). p < 0.05 was considered statistically significant (marked in bold). ¹ UA-PI P-value calculated from 28 to 32 weeks of gestation, respectively, in macrosomic and normal weight babies. ² UA-PI P-value calculated from 32 to 36 weeks of gestation, respectively, in macrosomic and normal weight babies. ³ UA-PI P-value calculated from 28 to 36 weeks of gestation, respectively, in macrosomic and normal weight babies.

).

The macrosomic group was composed of 29 newborns (6.2%), and the non-macrosomic group was composed of 442 (93.8%). Results were similar to the previous categorization according to fetal weight centile (LGA versus AGA). In both groups, there was a gradual and significant decrease in UA-PI values from 28 to 36 weeks (Table 4; Figure 2). The macrosomic babies had lower UA-PI in all the stages of pregnancy than the leaner babies, although a significant difference was not reached (Table 4).

4. Discussion

Although there is no absolute consensus on the definition of LGA or macrosomia, fetal or neonatal weight is counted among the complications most frequently associated with gestational diabetes [4]. The prevalence of GDM and related fetal macrosomia has increased significantly in recent years, mainly due to the global increase in the prevalence of overweight, obesity, advanced maternal age over 35 years, and sedentary lifestyle [26].

This study demonstrated that LGA fetuses in GDM pregnancies exhibit lower UA-PI values compared to AGA fetuses. This finding remains consistent when considering only newborns with a birth weight of ≥4000 g. Additionally, the most significant decline in UA-PI among LGA fetuses was observed around 32 weeks of gestation. In addition, all fetuses from mothers with GDM had a significant and progressive decline in UA-PI during the third trimester of pregnancy. Increasing gestational age results in placental blood supplementation, increased umbilical artery blood volume, and reduced vascular resistance. This may explain the results obtained.

These data are in agreement with previous studies of pregnant women with type 1 diabetes or gestational diabetes, in which, similarly, LGA fetuses showed lower UA-PI values compared with normally growing fetuses [17,18]. However, the authors included only UA-PI measurements taken within 1 week before delivery [17,18].

These findings suggest an inverse correlation between UA-PI and neonatal birth weight [20,24].

Additionally, Gibbons et al. reported that UA-PI is more frequently affected in diabetic pregnancies compared to middle cerebral artery (MCA)-PI. This impairment showed a strong correlation with the severity of diabetes (p < 0.001), indicating that diabetes-related vascular changes may have a greater impact on the umbilical artery than on the MCA, potentially increasing the risk of adverse outcomes [27].

The placenta probably plays a key role, and the changes it undergoes when gestational diabetes occurs may explain these findings [28]. Placental inflammation, characterized by increased levels of inflammatory cytokines, has been observed in pregnancies complicated by GDM [28]. Insulin resistance is directly related to the inflammatory state, and exaggerated insulin resistance leads to an exaggerated inflammatory state [29].

It is also known that GDM is characterized by endothelial dysfunction, inflammation, and insulin resistance underlying the altered vascularization of the diabetic placenta [30,31,32].

There are several evidence documenting the effect of placental architecture and its vascularization on hemodynamic indices. In 1986, Singh et al. first described the changes in the umbilical cord caused by GDM [33]. Increased permeability of umbilical arteries due to the erosion and disruption of the endothelial lining of the vessels has been documented.

Vessel wall smooth muscle also undergoes fiber rupture and degeneration, thus leading to the dilatation of the umbilical vein. Wharton’s jelly showed an altered distribution pattern of its fibers, with wide gaps between them. These findings indicate that GDM negatively impacts the umbilical vessels and the connective tissue structure of Wharton’s jelly [33]. Weissman et al. observed that the umbilical cord is notably larger in fetuses of mothers with GDM compared to the general population, with the primary cause being an increased amount of Wharton’s jelly [34]. Additionally, fetal macrosomia is associated with a greater UA diameter [35]. This expansion in placental vasculature, along with a wider UA, may contribute to the observed reduction in flow impedance within the umbilical artery.

The endovascular deterioration seen in GDM may be linked to disruptions in the nitric oxide (NO) pathway [36]. In both normal-weight and obese pregnant women, NO production is reduced, leading to impaired endothelium-dependent vasodilation and altered insulin regulation, particularly after the second trimester. Some studies have reported a positive correlation between endothelial NO activity and umbilical artery PI, with lower NO levels corresponding to a more pronounced decline in PI [36].

Furthermore, in GDM, chronic low-grade inflammation and insulin resistance—frequently associated with obesity—contribute to reduced cytotrophoblast invasion and placental hypoxia from week 20 onward, resulting in decreased NO production and bioavailability within the fetoplacental circulation [37,38,39].

As pregnancy progresses, the physiological decline in NO levels, combined with obesity-related vascular damage and insulin dysregulation, may explain the observed reduction in umbilical artery PI. Additionally, macrosomic infants exhibit elevated serum triglyceride and total and free cholesterol levels compared to controls. Macrosomia is also associated with lower vitamin E levels, impaired superoxide dismutase activity, and increased serum thiobarbituric acid-reactive substances, suggesting heightened oxidative stress [40]. Furthermore, studies on humans and animals have shown that diabetes during pregnancy shifts the balance of Th1/Th2 cells to a protective Th2 phenotype, whereas, in macrosomic and obese offspring of diabetic mothers, the Th1/Th2 balance is shifted to a proinflammatory Th1 phenotype, which may confer to these animals a potential proinflammatory and “diabetogenic status”, as revealed by the hyperglycemia and hyperinsulinemia observed in adulthood [41].

In our cohort, a lower UA-PI at 32 weeks was linked to a higher likelihood of induction or emergency cesarean, likely due to placental vascular changes. However, UA-PI alone is not sufficient to predict labor outcomes, as maternal BMI, fetal growth velocity, and glycemic control also play key roles. Future research should integrate multiple clinical and Doppler parameters for better predictive models.

Given our study’s retrospective design, an optimized monitoring strategy for future research could be the following:

Early Ultrasound (24–26 weeks): Establish baseline Doppler indices.

Serial UA-PI (28, 32, 36 weeks): Track fetal vascular changes.

Middle Cerebral Artery Doppler (from 32 weeks): Assess fetal adaptation.

Biophysical Profile (36+ weeks): Detect fetal distress.

Our study has some limitations. Most importantly, the sample size is small for an accurate interpretation of the results.

In addition, we do not have simultaneous information on fetal venous hemodynamics nor on other fetal arterial districts (such as fetal brain hemodynamics). All Doppler ultrasound parameters were measured by only two experienced examiners to reduce the incorrect measurement. We intentionally excluded preterm or growth-restricted infants from this study to eliminate any bias that could be due to the presence of an abnormal increase in UA-PI in these fetuses.

Our findings suggest that placental function plays a critical role in the altered UA-PI values observed in LGA fetuses from GDM pregnancies. While we did not directly assess placental structure or function through Doppler studies of the uteroplacental circulation or histopathological analysis, the lower UA-PI in LGA fetuses may reflect increased placental perfusion and vascular remodeling associated with fetal overgrowth. Prior research indicates that GDM can lead to placental inflammation, endothelial dysfunction, and altered nitric oxide signaling, all of which can contribute to reduced vascular resistance in the umbilical artery [3,8,14,17,18,19,20,21]. Future studies should incorporate uterine artery Doppler, placental histopathology, and biochemical markers of placental function to better elucidate the relationship between placental changes and fetal vascular adaptation in GDM pregnancies.

Further research should analyze if, among multiparous women, those with prior macrosomia had a significantly higher risk of LGA in the current pregnancy compared to those with previous GDM but no macrosomia.

Moreover, further research should also focus on the maternal diet and its influence on glycemic control, fetal weight, and fetal well-being, including Doppler’s studies. Among the latter, it is of interest if any UPI values relate to a worst glycemic control, major risk of LGA, or negative fetal outcomes.

Another limitation of our study is the imbalance in sample sizes between the LGA and AGA groups (57 vs. 415). While this study is relatively well-powered, the smaller LGA sample may have reduced the statistical power to detect subtle differences in UA-PI trends, particularly at later gestational ages. This could explain why some differences in UA-PI, despite appearing clinically relevant, did not reach statistical significance. Future studies with larger LGA cohorts are needed to confirm these findings and assess the reproducibility of the observed trends.

On the other hand, the main strength of the present study is that we recorded and compared the UA-PI of LGA fetuses at three different time points during the third trimester of pregnancy. In addition, we focused on LGA infants by considering the percentile of birth weight in relation to gestational age. This is to reduce the bias of considering only birth weights above 4000 g, regardless of gestational age.

5. Conclusions

Our study shows that overgrowth in fetuses of mothers with GDM could be associated with reduced impedance in the umbilical artery flow, with a more significant reduction around 32 weeks gestational age.

Since maternal hyperglycemia increases the risk of macrosomia and GDM may represent a fetal vascular disorder, it, therefore, seems possible that in LGA fetuses, maternal hyperglycemia could have a worsening (or deleterious) effect on the fetal vasculature [42].

This underscores the importance of early diagnosis, strict glycemic and metabolic control, and an appropriate management of GDM to help reduce the potentially adverse effect of hyperglycemia on these fetuses and to optimize short and long-term health outcomes for mother and child [43,44].

Ultrasonography is a reliable, non-invasive, and cost-effective method for assessing fetal well-being. It remains the primary tool for fetal monitoring.

However, the significance of umbilical artery PI in evaluating LGA fetuses, particularly in pregnancies complicated by diabetes, is not yet fully understood. Clarifying its role alongside other ultrasound parameters would be crucial for improving counseling on the optimal timing and mode of delivery. Future research on vascular indices may offer valuable insights into the management of gestational diabetes by helping identify fetuses at a higher risk of adverse outcomes who could benefit from timely obstetric intervention.