Gestational Weight Gain and Adverse Pregnancy Outcomes in Women with Gestational Diabetes Mellitus and Obesity

Abstract

Background/Objectives: The association between gestational weight gain (GWG) and adverse outcomes in individuals with gestational diabetes mellitus (GDM) and obesity remains unclear. This study aimed to evaluate the relationship between total GWG and maternal, obstetric, and neonatal outcomes in patients with GDM, stratified by obesity class. Methods: This retrospective cohort included 695 pregnant individuals with GDM treated at a tertiary university hospital in Brazil between 2007 and 2021. GWG was categorized as insufficient, adequate, or excessive per National Academy of Medicine guidelines. Outcomes included maternal, obstetric, and neonatal events. Analyses were conducted for the entire cohort and stratified by obesity class (I and II/III), using multivariate regression models adjusted for maternal age, parity, and pre-pregnancy BMI. Results: The mean age was 33.6 (SD 5.7) years. GWG was insufficient in 33.2%, adequate in 28.2%, and excessive in 37.8%. Excessive GWG was associated with increased odds of cesarean delivery (OR 1.69; 95% CI 1.15–2.48) and large-for-gestational-age newborns (OR 3.29; 95% CI 1.61–6.46). As a continuous variable, GWG was positively associated with cesarean delivery (OR 1.04), LGA (OR 1.10), and birthweight (β = 0.02). Lower GWG was independently associated with reduced preeclampsia risk (OR 1.09 per kg). Insufficient GWG was not linked to increased risk of small-for-gestational-age newborns or other adverse outcomes and was associated with lower insulin requirement. Results remained consistent across obesity subgroups, except for cesarean delivery in class II/III obesity. Conclusions: In individuals with GDM and obesity, insufficient GWG was not associated with increased adverse outcomes, while excessive GWG was consistently linked to unfavorable maternal and neonatal risks. Stricter GWG control may be safe and beneficial in this population.

1. Introduction

Gestational weight gain (GWG), defined as the difference between pre-pregnancy weight and maternal weight just before delivery, is associated with maternal and fetal outcomes. GWG is a unique and complex biological phenomenon that supports the functions of growth and development. It results from the combined growth of the fetus, placenta, and amniotic fluid, along with maternal physiological adaptations. These include increased blood volume, enlargement of uterine and mammary tissues, accumulation of adipose stores, and fluid retention. The dynamic regulation of these components is driven by complex endocrine mechanisms. The placenta functions as a major endocrine organ, producing human placental lactogen, sex steroids, and inflammatory cytokines that progressively reduce maternal insulin sensitivity and modify energy metabolism to support fetal growth. In parallel, maternal hormones such as insulin, leptin, and thyroid hormones undergo marked changes, influencing nutrient partitioning, fat deposition, and fluid balance. Such physiological and endocrine processes provide a mechanistic basis for the associations observed between GWG and adverse maternal and fetal outcomes [1,2].

Given this biological foundation, deviations from the recommended GWG ranges carry meaningful clinical consequences. GWG below recommended levels has been linked to an increased risk of preterm birth and small-for-gestational-age (SGA) newborns, whereas excessive GWG has been associated with hypertensive disorders of pregnancy, fetal macrosomia, cesarean delivery, and other adverse conditions [3,4,5].

In 2009, the Institute of Medicine (IOM), now known as the National Academy of Medicine (NAM), published GWG guidelines recommending progressively lower weight gain targets for women with higher adiposity [1]. Although widely adopted by major international scientific societies, these guidelines have important limitations: they were derived from data predominantly from Caucasian populations and do not include specific recommendations for high-risk pregnancies, such as those complicated by gestational diabetes mellitus (GDM) or for different degrees of obesity [6].

For women with obesity (BMI > 30 kg/m²), the NAM recommends a total GWG ranging from 5 to 9 kg throughout pregnancy, without distinguishing between different obesity classes: class I (BMI 30–34.9 kg/m²), class II (BMI 35–39.9 kg/m²), and class III (BMI > 40 kg/m²), the latter being of particular concern given the progressively increased obstetric risks associated with higher BMI. The incidence of outcomes such as SGA, LGA (large for gestational age), prematurity, and other adverse events has shown conflicting results, especially in cases of more severe obesity [7]. Some authors have proposed more stringent GWG targets for women with class II and III obesity. One study comparing the NAM recommendations (5–9 kg) with a more restrictive range (0–4 kg) in this subgroup found no significant differences in the incidence of SGA or other maternal–fetal complications [8]. Similar findings have been reported by other authors, even in cases of weight loss during pregnancy, both among women with GDM and in other ethnic populations [9,10].

There is evidence that excessive GWG increases the risk of GDM-related complications, such as poor glycemic control, macrosomia, and hypertensive disorders. However, there is a lack of evidence regarding the most appropriate weight gain targets for women with GDM and class II or III obesity.

In this context, questions remain regarding the adequacy of current GWG recommendations, particularly for pregnant women with more advanced degrees of obesity and/or increased obstetric risk conditions such as GDM. The objective of the present study is to evaluate the association between GWG and adverse maternal–fetal outcomes according to the degree of pre-pregnancy obesity in women with GDM, considering the classification of GWG as insufficient, adequate, or excessive based on NAM guidelines. Outcomes assessed included maternal events (preeclampsia, gestational hypertension, cesarean delivery), neonatal anthropometric classifications (SGA, AGA, LGA), and other neonatal complications such as hypoglycemia, jaundice, respiratory distress, congenital malformations, prematurity, and NICU admission.

2. Materials and Methods

2.1. Study Population

This was a longitudinal study with data collected during prenatal care of high-risk pregnancies in women with gestational diabetes mellitus (GDM) at a tertiary care center (Diabetes Center of the Federal University of São Paulo–UNIFESP), between 2007 and 2021.

Inclusion criteria consisted of pregnant women diagnosed with GDM according to the criteria recommended by the World Health Organization (WHO) [11]. Exclusion criteria included conditions that could interfere with the interpretation of outcomes influenced by gestational weight gain, glycemic control, or that involved populations with distinct risk profiles for adverse obstetric outcomes. These were: pregnancy following bariatric surgery, twin pregnancy, maternal age under 18 years during pregnancy, use of oral corticosteroids during pregnancy, and absence of data on maternal pre-pregnancy BMI or GWG.

From the initial dataset (1000 pregnant women with diabetes mellitus monitored during pregnancy at the Diabetes Center between 2007 and 2021), we excluded 5 twin pregnancies, 2 cases with a history of bariatric surgery, and 56 due to missing data on pre-pregnancy BMI, resulting in 937 cases eligible for initial analysis. Considering that the diagnostic criteria for GDM used in 2007 differed from those currently recommended, we reanalyzed the patients’ laboratory results and included only those who met the current WHO diagnostic criteria for GDM. This standardization aimed to harmonize the study population with present-day guidelines and to exclude patients with more severe hyperglycemia consistent with overt diabetes mellitus. After this step, 156 cases were excluded, resulting in 781 records.

Among these, only seven patients had a pre-pregnancy BMI below 18.5 kg/m²—a very small number relative to the overall sample. These patients belong to a group associated with obstetric outcomes that differ from those observed in individuals with higher BMI, and they are subject to distinct gestational weight gain recommendations. Since the study aimed to evaluate outcomes related to overweight and obesity, these patients were excluded to ensure sample homogeneity, leaving 773 cases. Finally, an additional 78 records were excluded due to missing data on GWG, resulting in a final sample of 695 pregnancies included in the study (Figure 1).

2.2. Baseline and Outcome Measures

2.2.1. Exposure Variables

• GWG was calculated as the difference between self-reported pre-pregnancy weight, obtained at the beginning of prenatal care, and third-trimester weight, recorded at the last prenatal visit before delivery.
• GWG classification was based on NAM recommendations and categorized according to pre-pregnancy BMI as follows: Inadequate GWG: below recommended levels (Normal weight: <11.5 kg; Overweight: <7.0 kg; Obesity: <5.0 kg); Adequate GWG: within recommended ranges (Normal weight: 11.5–16.0 kg; Overweight: 7.0–11.5 kg; Obesity: 5.0–9.0 kg); Excessive GWG: above recommended levels (Normal weight: >16.0 kg; Overweight: >11.5 kg; Obesity: >9.0 kg).

2.2.2. Covariates

• GDM: The diagnosis followed the guidelines of the Brazilian Diabetes Society, based on the criteria established by the World Health Organization (WHO) [11]. These include: fasting plasma glucose ≥ 92 mg/dL and <126 mg/dL if measured during the first trimester, and/or altered values on the oral glucose tolerance test (with one or more abnormal values): fasting > 92 mg/dL (and <126 mg/dL); 1-h > 180 mg/dL (and <209 mg/dL); 2-h > 153 mg/dL (and <200 mg/dL) [12].
• BMI classification: The World Health Organization (WHO) criteria were used, categorized as follows: Normal weight: 18.5–24.9 kg/m²; Overweight: 25.0–29.9 kg/m²; Obesity class I: 30.0–34.9 kg/m²; Obesity class II: 35.0–39.9 kg/m²; Obesity class III: >40.0 kg/m² [13]. To increase statistical power, the last two categories (class II and III) were combined into a single group.
• Chronic Hypertension: Defined as systolic blood pressure (SBP) > 140 mmHg and/or diastolic blood pressure (DBP) > 90 mmHg prior to pregnancy or diagnosed before 20 weeks of gestation, or based on the patient’s self-reported history during follow-up [14].
• Macrosomia: Defined as birth weight greater than 4000 g.

2.2.3. Outcomes

• Preeclampsia: A condition characterized by systolic blood pressure (SBP) > 140 mmHg and/or diastolic blood pressure (DBP) > 90 mmHg, accompanied by proteinuria (>300 mg/24 h) or, in the absence of proteinuria, by signs of systemic involvement or target organ dysfunction such as thrombocytopenia, hepatic dysfunction, renal insufficiency, acute pulmonary edema, or clinical features indicating impending eclampsia or eclampsia [14].
• Gestational Hypertension: Systolic blood pressure (SBP) > 140 mmHg and/or diastolic blood pressure (DBP) > 90 mmHg diagnosed after 20 weeks of gestation in a previously normotensive patient, without proteinuria or additional criteria for preeclampsia [14].
• Cesarean Delivery: Birth recorded via abdominal route, regardless of indication (elective or emergency).
• Small for Gestational Age (SGA): Birth weight below the 10th percentile according to the WHO reference curve for gestational age.
• Appropriate for Gestational Age (AGA): Birth weight between the 10th and 90th percentiles according to the WHO reference curve for gestational age; includes newborns whose weight is within the expected range for their gestational age.
• Large for Gestational Age (LGA): Birth weight above the 90th percentile according to the WHO reference curve for gestational age.
• Neonatal Hypoglycemia: Plasma glucose level below 60 mg/dL within the first 72 h of life, which was the cutoff used by the service.
• Neonatal Jaundice: Clinical diagnosis or medical record documentation of jaundice in the newborn, confirmed by maternal report or discharge summary.
• Neonatal Respiratory Distress: Any respiratory condition characterized by clinical signs of respiratory difficulty occurring within the first 24 h of life, potentially requiring ventilatory support or supplemental oxygen therapy.
• Congenital Malformations: Structural anomalies identified through prenatal ultrasound or discovered after birth, requiring ongoing medical follow-up or surgical treatment.
• Prematurity: Birth occurring before 37 completed weeks of gestation.

2.3. Statistical Analysis

The sample was stratified into three GWG groups: inadequate, adequate, and excessive. ANOVA or chi-square tests were performed to compare groups for outcome variables and covariates. Continuous variables are presented as mean and standard deviation, while categorical variables are expressed as frequency and percentage. Possible associations between exposure variables and adverse maternal–fetal outcomes were analyzed using multiple regression models. Linear or logistic regressions, adjusted for potential confounders, were employed depending on the type of dependent variable (continuous or dichotomous). Comparative and regression analyses were also conducted stratified by obesity groups: class I and combined classes II and III.

To identify potential confounders in these associations, the method of Directed Acyclic Graphs (DAGs) was used. A DAG is a causal diagram based on a theoretical mathematical model, employed to visually represent the relationships among variables (exposure, outcome, and other associated variables). Its primary purpose is to identify the minimal sufficient adjustment set of variables necessary to avoid bias and overadjustment. In this theoretical model, the minimal adjustment set identified to estimate the total effect of GWG on maternal–fetal outcomes included Age, parity, and pre-pregnancy BMI. DAG models were created using the software DAGitty, version 3.0 (www.dagitty.net, accessed on 9 August 2025) and are presented in the Supplementary Material.

The analysis was performed using IBM^® Statistical Package for the Social Sciences^® Statistics, version 25.0 (IBM Corporation, New York, NY, USA, 2017). A p-value of less than 0.05 was considered statistically significant.

3. Results

The mean age of the study population was 33.6 years (SD 5.72), and the mean gestational age at the first prenatal visit was 25.1 weeks (SD 7.73). Among the 695 pregnancies evaluated in this study, 160 (23.0%) were normal weight, 234 (33.7%) overweight, 184 (26.5%) class I obesity, 76 (10.9%) class II obesity, and 41 (5.9%) class III obesity. There were 231 (33.2%) pregnant women with inadequate GWG, 200 (28.2%) with adequate GWG, and 263 (37.8%) with excessive GWG according to NAM recommended ranges based on pre-pregnancy BMI categories (normal weight, overweight, and obesity—regardless of obesity class). The GWG recommendation for obesity was considered the same across all obesity classes, and the distribution of GWG by pre-pregnancy BMI category is shown in Figure 2.

A comparative analysis of characteristics among the 695 pregnancies in women with GDM was conducted, categorized according to GWG based on NAM recommendations into three groups: Inadequate, Adequate, and Excessive, as shown in

Table 1. Maternal characteristics, laboratory tests, and obstetric outcomes in women with GDM according to GWG based on NAM recommendations.

Variable Analyzed	Insufficient GWG	Adequate GWG	Excessive GWG	p *	p **
Maternal characteristics and prenatal care data
Age at first prenatal visit, years	34.1 (5.6)	33.7 (6.0)	33.1 (5.6)	0.15	0.05
Two or more pregnancies. n (%)	189 (81.8)	155 (77.1)	208 (79.1)	0.48	0.47
GDM or macrosomia in prior pregnancy. n (%)	30 (13.3)	24 (12.8)	37 (14.7)	0.82	0.63
Preeclampsia in prior pregnancy. n (%)	24 (10.6)	22 (11.8)	40 (16.0)	0.18	0.08
Chronic hypertension. n (%)	22 (9.7)	17 (8.6)	42 (16.8)	0.01	0.02
Smoking. n (%)	21 (9.3)	30 (15.2)	61 (24.3)	<0.01	<0.01
Gestational age at GDM diagnosis, weeks	18.0 (8.0)	18.9 (8.5)	19.0 (8.4)	0.34	0.19
Insulin use during pregnancy. n (%)	56 (24.1)	63 (31.5)	101 (38.7)	<0.01	<0.01
Obstetric outcomes
Preeclampsia. n (%)	1 (0.5)	5 (2.6)	10 (4.1)	0.04	0.01
Gestational hypertension. n (%)	5 (2.3)	8 (4.2)	11 (4.5)	0.40	0.20
Cesarean delivery. n (%)	116 (51.8)	109 (54.8)	172 (67.2)	<0.01	<0.01
Neonatal and puerperium outcomes
Neonatal complications. n (%)	88 (40.2)	82 (41.0)	94 (38.4)	0.84	0.68
Birthweight, kg	3.0 (0.6)	3.2 (0.5) a	3.3 (0.5) a,b	<0.01	<0.01
Classification by birthweight				<0.01	<0.01
SGA. n (%)	27 (12.1)	19 (9.5)	17 (6.6)
AGA. n (%)	185 (83.0)	169 (84.9)	200 (77.5)
LGA. n (%)	11 (4.9)	13 (5.4)	51 (16.3)
Neonatal death. n (%)	6 (2.7)	0 (0.0)	2 (0.8)	NA	0.07
Congenital malformation. n (%)	11 (5.0)	9 (4.6)	8 (3.3)	0.63	0.36
Prematurity +. n (%)	24 (10.7)	23 (11.5)	20 (7.7)	0.34	0.25
Neonatal jaundice. n (%)	48 (21.8)	50 (25.0)	48 (19.5)	0.38	0.52
Neonatal respiratory distress. n (%)	16 (7.3)	12 (6.0)	24 (9.8)	0.32	0.30
NICU admission. n (%)	16 (7.3)	13 (6.5)	13 (5.3)	0.67	0.38
Neonatal hypoglycemia. n (%)	18 (8.2)	22 (11.1)	28 (11.4)	0.47	0.26
Fasting glucose 6–8 weeks postpartum, mg/dL	90.2 (9.8)	89.9 (10.2)	92.4 (9.5)	0.07	0.06
2 h-glucose (75 g OGTT) postpartum, mg/dL	108.6 (32.6)	105.4 (32.5)	112.7 (32.1)	0.20	0.28

Values are presented as mean (standard deviation) or n (%). p * = ANOVA used for continuous variables with Bonferroni correction when p < 0.05: a = vs. inadequate GWG; b = vs. adequate GWG; or chi-square test for categorical variables. Statistical significance was considered at p < 0.05. p ** = p for trend used for continuous variables or linear-by-linear association for categorical variables. GDM = Gestational diabetes mellitus; BMI = Body mass index; NAM = National Academy of Medicine; SGA = Small for gestational age; AGA = Appropriate for gestational age; LGA = Large for gestational age; NICU = Neonatal intensive care unit. Gestational weight gain recommendations based on NAM-2009: 12.5–18 kg for underweight women (BMI < 18.5 kg/m²), 11.5–16 kg for normal weight (BMI 18.5–24.9 kg/m²), 7–11.5 kg for overweight (BMI 25.0–29.9 kg/m²), and 5–9 kg for women with obesity (BMI > 30.0 kg/m²). ⁺ Preterm birth is defined as delivery before 37 completed weeks of gestation.

.

According to GWG categories, higher prevalences were observed for previous hypertension [22 (9.7%) vs. 17 (8.6%) vs. 42 (16.8%), n (%), p = 0.01, linear-by-linear p = 0.02], smoking [21 (9.3%) vs. 30 (15.2%) vs. 61 (24.3%), n (%), p < 0.01, linear-by-linear p < 0.01], insulin use during pregnancy [56 (24.1%) vs. 63 (31.5%) vs. 101 (38.7%), n (%), p < 0.01, linear-by-linear p < 0.01], preeclampsia [10 (4.1%) vs. 5 (2.6%) vs. 1 (0.5%), n (%), p = 0.04, linear-by-linear p = 0.01], and cesarean delivery [116 (51.8%) vs. 109 (54.8%) vs. 172 (67.2%), n (%), p < 0.01, linear-by-linear p < 0.01] across the inadequate, adequate, and excessive GWG groups, respectively (Table 1).

Table 1 also shows newborn weight [3.0 (0.6) vs. 3.2 (0.5) vs. 3.3 (0.5), ANOVA p < 0.01, p for trend < 0.01] according to the inadequate, adequate, and excessive GWG groups, respectively. When categorizing this variable as SGA, AGA, and LGA, a significant difference was observed among the GWG groups (p < 0.01, linear-by-linear p < 0.01).

In Supplementary Table S1, these comparisons were stratified into two groups: class I obesity and classes II/III obesity. In the class I obesity group, a higher occurrence of cesarean delivery was observed [19 (46.3%) vs. 29 (54.7%) vs. 62 (70.5%), n (%), p = 0.02, linear-by-linear p < 0.01], as well as a greater need for insulin therapy during pregnancy [10 (24.4%) vs. 18 (34.0%) vs. 47 (52.2%), p < 0.01, linear-by-linear p < 0.01] across the inadequate, adequate, and excessive GWG groups, respectively. Among pregnant women with class II and III obesity, higher HbA1c levels were observed [5.4 (0.4) vs. 5.7 (0.4) vs. 5.6 (0.5), p = 0.04, linear-by-linear p = 0.04] (Supplementary Table S1) according to the inadequate, adequate, and excessive GWG groups, respectively.

In

Table 2. Association between GWG adequacy and maternal-neonatal outcomes: comparison of inadequate or excessive GWG versus adequate GWG, in the overall sample and in stratified by class I and class II/III obesity.

Outcome	Insufficient GWG vs. Adequate			Excessive GWG vs. Adequate
All participants	OR	95% CI	p	OR	95% CI	p
Preeclampsia	0.17	0.02–1.51	0.11	1.27	0.94–1.70	0.12
Gestational hypertension	0.54	0.17–1.68	0.29	1.09	0.43–2.78	0.85
Cesarean delivery	0.88	0.60–1.30	0.54	1.69	1.15–2.48	<0.01
SGA	1.30	0.70–2.42	0.40	0.67	0.34–1.32	0.25
LGA	0.88	0.38–2.09	0.78	3.29	1.61–6.46	<0.01
Neonatal complications	0.96	0.65–1.42	0.86	0.89	0.61–1.31	0.57
Class I obesity	OR	95% CI	p	OR	95% CI	p
Preeclampsia	---	----	----	1.94	0.38–10.0	0.42
Gestational hypertension	0.86	0.14–5.41	0.87	1.05	0.24–4.58	0.95
Cesarean delivery	0.71	0.31–1.62	0.42	1.97	0.97–4.01	0.06
SGA	4.08	1.01–16.55	0.05	0.78	0.17–3.62	0.75
LGA	0.63	0.11–3.63	0.61	1.89	0.58–6.21	0.29
Neonatal complications	1.14	0.50–2.61	0.75	0.71	0.35–1.45	0.35
Class II and III obesity	OR	95% CI	p	OR	95% CI	p
Preeclampsia	---	----	----	0.24	0.02–2.84	0.26
Gestational hypertension	0.63	0.08–4.79	0.65	1.00	0.17–5.89	1.00
Cesarean delivery	0.94	0.35–2.57	0.91	1.70	0.64–4.49	0.29
SGA	0.32	0.05–1.89	0.21	0.11	0.01–1.06	0.06
LGA	0.22	0.02–2.20	0.20	2.15	0.54–8.47	0.26
Neonatal complications	0.87	0.32–2.34	0.79	1.03	0.40–2.66	0.95

Values are expressed as odds ratios (OR) with 95% confidence intervals (95% CI). Statistical significance was considered at p < 0.05. SGA = Small for gestational age; LGA = Large for gestational age; GWG = Gestational weight gain; Class I obesity = pre-pregnancy BMI between 30.0–34.9 kg/m²; Class II and III obesity = pre-pregnancy BMI above 35.0 kg/m².

, we present the association of inadequate or excessive GWG compared to adequate GWG with various maternal–fetal outcomes of interest. Excessive GWG was directly associated with an increased occurrence of LGA (OR 3.29; 95% CI 1.61–6.46; p < 0.01) and cesarean delivery (OR 1.69; 95% CI 1.15–2.48; p < 0.01) compared to adequate GWG, with no differences observed in the occurrence of preeclampsia, hypertensive disorders of pregnancy, SGA, or neonatal complications. Inadequate GWG was not associated with any of these maternal–fetal outcomes when compared to adequate GWG in the analysis of all participants. We noted that when obesity groups I and II/III were evaluated separately, neither inadequate nor excessive GWG was associated with maternal–fetal outcomes compared to adequate GWG.

The outcome variables that showed statistically significant differences in the comparisons presented in Table 1 were further evaluated for associations with GWG as a continuous variable, using regression analyses adjusted for the variables identified through the DAG method (Supplementary Figures S1–S5: maternal age, parity, and pre-pregnancy BMI), as shown in

Table 3. Association of gestational weight gain (GWG) as a continuous variable with maternal and fetal outcomes. Results were obtained using logistic regression (for binary outcomes) and linear regression (for neonatal weight), presented in crude and adjusted models.

Variable Analyzed	Crude Model			Adjusted Model
A. Logistic regression	OR	95% CI	p	OR	95% CI	p
Preeclampsia	1.07	1.005–1.16	0.03	1.09	1.02–1.17	0.01
Cesarean delivery	1.04	1.01–1.06	<0.01	1.04	1.02–1.07	<0.01
SGA	0.98	0.94–1.03	0.48	0.97	0.92–1.02	0.19
LGA	1.09	1.05–1.14	<0.01	1.10	1.05–1.15	<0.01
B. Linear regression	ß	95% CI	p	ß	95% CI	P
Birth weight	0.02	0.01–0.02	<0.01	0.02	0.01–0.03	<0.01

The adjusted model included the variables: maternal age, parity, and pre-pregnancy BMI. SGA = Small for gestational age; LGA = Large for gestational age.

. In logistic regression, GWG as a continuous variable was associated with preeclampsia (OR = 1.09; 95% CI 1.02–1.17), cesarean delivery (OR = 1.04; 95% CI 1.02–1.07), and LGA (OR = 1.10; 95% CI 1.05–1.15), independently of maternal age, parity, and pre-pregnancy BMI. An association with SGA was observed in the unadjusted model but lost significance after adjustment. In linear regression, where birth weight (continuous) was the dependent variable, a direct association with GWG was found even after adjusting for maternal age, parity, and pre-pregnancy BMI (β = 0.02; 95% CI 0.01–0.03).

In

Table 4. Association of GWG as a continuous variable with maternal and fetal outcomes in obesity subgroups, based on regression models.

Variable Analyzed	Crude Model			Adjusted Model
Class I obesity
Logistic regression	OR	95% CI	p	OR	95% IC	p
Preeclampsia	1.07	0.98–1.16	0.13	1.05	0.96–1.16	0.26
Gestational hypertension	1.01	0.91–1.11	0.86	1.02	0.93–1.13	0.66
Cesarean delivery	1.07	1.02–1.13	<0.01	1.08	1.02–1.14	<0.01
SGA	0.90	0.81–0.99	0.03	0.90	0.81–1.00	0.05
LGA	1.08	1.02–1.16	<0.01	1.10	1.03–1.17	<0.01
Linear regression	ß	95% IC	P	ß	95% IC	p
Birth weight	0.14	0.03–0.24	<0.01	0.14	0.03–0.24	0.01
Class II and III obesity
Logistic regression	OR	95% IC	p	OR	95% IC	p
Preeclampsia	1.05	0.87–1.26	0.62	1.11	0.91–1.35	0.30
Gestational hypertension	1.04	0.93–1.16	0.52	1.01	0.90–1.13	0.84
Cesarean delivery	1.03	0.97–1.09	0.29	1.03	0.97–1.10	0.28
SGA	0.95	0.85–1.07	0.40	0.96	0.86–1.08	0.55
LGA	1.09	1.02–1.16	<0.01	1.10	1.03–1.18	<0.01
Linear regression	ß	95% IC	p	ß	95% IC	p
Birth weight	0.21	0.11–0.31	<0.01	0.20	0.10–0.31	<0.01

The adjusted model included the variables: maternal age and parity. SGA = Small for gestational age; LGA = Large for gestational age.

, we present the regression analyses from Table 3, stratified by obesity class. Among women with class I obesity, a higher GWG was associated with an increased likelihood of cesarean delivery (OR 1.08; 95% CI 1.02–1.14; p < 0.01), as well as a direct association between GWG and newborn weight (β = 0.14; 95% CI 0.03–0.24; p = 0.01), even after adjustment for maternal age and parity. In the group of women with class II and III obesity, GWG was associated with a higher likelihood of LGA (OR 1.10; 95% CI 1.03–1.17; p < 0.01) and higher neonatal weight (β = 0.20; 95% CI 0.10–0.31; p < 0.01) in both crude and adjusted models for maternal age and parity.

4. Discussion

This study aimed to assess the association between GWG, as defined by NAM guidelines, and maternal–fetal outcomes in women with GDM and varying degrees of obesity, based on the hypothesis that this specific population may benefit from tailored recommendations—particularly lower GWG targets for those with more severe obesity. The results showed that in this cohort of women with GDM, lower GWG—even when classified as inadequate according to NAM—was associated with a lower prevalence of preeclampsia and cesarean delivery. Additionally, no increase in the occurrence of SGA newborns was observed, while a lower prevalence of LGA was seen with decreasing GWG. When stratified by obesity class, the data showed that among women with class I obesity, lower GWG was associated with reduced cesarean rates, lower LGA prevalence, and lower birth weight, without an increase in SGA. Among those with more severe obesity (class II and III), lower GWG was also associated with lower odds of LGA and lower birth weight, again without a higher prevalence of SGA. These findings suggest that more restrictive GWG goals in this population may lead to favorable outcomes without necessarily increasing the risk of adverse effects.

Our results allow for an evaluation of GWG in this sample of women with GDM and varying degrees of obesity, in comparison with GWG recommendations for the general pregnant population, based on the impact on key outcomes—particularly LGA and SGA as relevant neonatal outcomes, as well as preeclampsia and cesarean delivery as major maternal events. These outcomes (LGA, SGA, preeclampsia, and cesarean delivery) will be analyzed individually below, in light of the observed data and their underlying pathophysiological context.

4.1. Large for Gestational Age–LGA

In our sample of pregnant women with GDM, higher GWG values were associated with increased birth weight and greater odds of LGA, even after adjustment for potential confounders, highlighting the negative impact of excessive weight gain on fetal growth. Furthermore, in stratified analyses by obesity class, this association remained significant both among women with class I obesity and those in the combined class II and III subgroup. These findings align with GWG recommendations for pregnant women [7] and are supported by scientific literature in women with GDM. Soarez de Souza et al., based on a Brazilian cohort of 545 pregnant women with GDM, also identified a higher risk of LGA among those with excessive GWG [15]. Similarly, Ferreira et al., in a retrospective multicenter study involving 4563 pregnant women with GDM and obesity in Portugal, demonstrated that GWG above 9 kg was associated with higher rates of LGA and macrosomia [16]. Monteiro et al., in a Portuguese cohort of 18,961 pregnant women with GDM, also showed that excessive GWG was linked to an increased risk of macrosomia and LGA [17]. Together with the present study, these data reinforce the role of excessive weight gain as an important risk factor for neonatal outcomes related to increased fetal growth.

Moreover, in our analysis, lower GWG values—as assessed continuously—were significantly associated with lower odds of LGA, independently of maternal age, parity, and pre-pregnancy BMI. This is consistent with findings by Gavard et al., who also observed an association between lower gestational weight gain and reduced risk of LGA newborns [18]. This finding supports the hypothesis that modest weight restriction during pregnancy may help mitigate the risk of excessive fetal growth in women with GDM and obesity.

Biologically, the associations described above may be explained by the metabolic and hormonal adaptations that occur in pregnancy among women with GDM and excessive GWG. As gestation advances, placental hormones such as human placental lactogen, placental growth hormone, estrogens, and progesterone progressively reduce maternal insulin sensitivity [2,19,20]. At the same time, greater adipose tissue expansion—common in women with higher pre-pregnancy BMI—increases fatty acid release and promotes low-grade inflammation [19,21]. This subclinical inflammatory milieu may further enhance placental nutrient transport and contribute to accelerated fetal growth, independently of maternal glycemia [22,23,24]. These combined changes raise the maternal supply of glucose and lipids, enhancing nutrient transfer to the fetus. This environment encourages higher fetal insulin secretion and accelerated growth, which are central features of LGA [20]. Conversely, keeping GWG within moderate limits may help temper these processes and reduce LGA risk without impairing fetal growth, provided that adequate monitoring is maintained [2].

4.2. Small for Gestational Age—SGA

Initially, a significant difference in the occurrence of SGA was observed among groups categorized by GWG (Table 1), suggesting a higher frequency of SGA newborns among women with GDM and lower weight gain. However, when comparing groups directly, the paired analysis presented in Table 2—using adequate GWG as the reference—did not identify a statistically significant association between inadequate GWG and SGA. Complementing these findings, the adjusted analysis treating GWG as a continuous variable (Table 3) also showed no association between lower weight gain and the risk of SGA, indicating that the initial trend did not persist after controlling for confounders.

In the subgroup analysis of the population with GDM stratified by obesity class (Table 4), no independent association was observed between lower GWG values and SGA outcomes for any obesity class.

Similar findings to ours were reported by the meta-analysis by Viecceli et al., which included 88,599 women with GDM and showed that inadequate GWG was associated with a reduced risk of LGA newborns, without a statistically significant increase in the frequency of SGA [25]. Complementing this perspective, Huang et al., in a retrospective case–control study of 1651 women with GDM, demonstrated that inadequate GWG was not an independent risk factor for this outcome, with preeclampsia and a history of SGA siblings being the main predictors of this condition [26]. Finally, consistent with our results, Gavard et al., in a large study including 66,010 women with obesity in the US (most with diabetes, primarily GDM), observed that GWG below NAM guidelines—or even some weight loss—was not associated with increased risk of SGA, regardless of obesity class [18].

Conversely, there are studies reporting conflicting findings, indicating an increased risk of SGA in pregnancies with inadequate GWG among women with GDM [17,18,25,26,27,28]. Shi et al., in a study of 11,168 pregnant women with GDM, showed that GWG below NAM recommendations, despite reducing the risk of LGA and macrosomia, increased the risk of SGA [29]. Monteiro et al., in the large study previously cited involving women with GDM, found that insufficient weight gain—especially among women with underweight or normal pregestational BMI—was associated with a higher risk of SGA, despite other associated benefits. However, the same study showed that in women with overweight or obesity, inadequate GWG was not associated with adverse outcomes, suggesting that more restrictive GWG targets may be safe in this group with higher obesity levels [17], which aligns with our findings.

These patterns can be explained by differences in maternal body composition and endocrine regulation during pregnancy. In women with obesity, lower GWG generally reflects limited additional fat accumulation in the maternal body rather than a restriction of essential substrates, particularly those required for fetal growth. Preexisting insulin resistance and expanded adipose tissue already provide substantial lipid reserves and a state of low-grade chronic inflammation, while placental hormones—such as human placental lactogen, placental growth hormone, estrogens, and progesterone—further reduce maternal insulin sensitivity and maintain glucose transfer to the fetus [2,20]. In this context, restricting GWG does not necessarily compromise fetal growth, as baseline maternal metabolic adaptations already ensure an adequate energy supply. Taken together, these mechanisms help explain why insufficient GWG did not translate into a higher risk of SGA in our cohort and reinforce the safety of more conservative GWG targets in women with GDM and obesity, provided that pregestational BMI is taken into account and appropriate surveillance is maintained.

4.3. Lower GWG in GDM: Evidence of Safety and Benefit

Given the results above, it is noteworthy that we did not observe harm in women with GDM, including those with overweight and obesity, who had inadequate weight gain during pregnancy. Large population-based studies corroborate our findings. Huang et al., in a large US cohort (n = 1,338,460), proposed more restrictive GWG ranges for pregnant women with GDM compared to NAM recommendations, with specific targets for each BMI category and progressively lower values as the degree of obesity increased [30]. Similarly, the Belgian study by Benhalima et al. (n = 13,060), in women with GDM, proposed ideal GWG ranges aimed at minimizing the risk of SGA and LGA, also recommending more conservative targets for overweight and obesity. These findings reinforce the need to individualize GWG targets in women with GDM, especially in the presence of obesity [27].

In this context, the present study contributes to this field of knowledge by being conducted in a Brazilian population characterized by ethnic diversity and admixture, thereby expanding the evidence in a middle-income country and providing data more representative of realities distinct from those observed in high-income countries.

Many studies evaluating GWG in women with GDM do not consider the different classes of pregestational obesity, as was done in our study. When analyzing by pregestational BMI subgroup, we observed that the lack of association between inadequate GWG and SGA persisted across all BMI categories, including women with normal BMI and overweight—subgroups traditionally described in the literature as more susceptible to the risk of this outcome with inadequate GWG.

Studies such as Goldstein et al., along with the NAM guidelines themselves, have shown a higher risk of SGA with inadequate GWG in low-risk pregnant women across all BMI classes, including those with normal or low BMI. However, these references used the general population as a reference and not specifically women with GDM [1,31]. More recently, the study by Guo et al. addressed this gap by evaluating a large cohort composed exclusively of pregnant women with GDM. This study also found an association between inadequate GWG and SGA, but only for those with normal BMI, with no significant association observed for overweight or obese groups, as seen in low-risk pregnancies [32]. This finding suggests that the presence of GDM may attenuate the relationship between inadequate GWG and SGA among women with higher BMI.

Our study advances this debate by not identifying a significant association even in the subgroup of pregnant women with normal BMI, reinforcing the hypothesis that, when there is adequate clinical follow-up with effective fetal monitoring (as in a specialized and multidisciplinary service such as in the present study), moderate weight restriction may not necessarily imply a higher risk of fetal growth restriction, even in women with lower BMI.

As previously discussed, maternal pregestational BMI, adiposity, and lipid–endocrine adaptations, together with placental hormones sustaining glucose supply, and the presence of low-grade maternal inflammation, already provide a background of metabolic adjustments that may mitigate the risk of growth restriction, even under more conservative GWG targets. Building on this, intrinsic pathophysiological mechanisms of GDM itself may also play a role in neutralizing the risk of SGA. Among speculative hypotheses, one can highlight the possibility of mild fetal hyperinsulinemia in response to episodes of mild maternal hyperglycemia, which can occur even in well-controlled GDM patients, ultimately stimulating fetal growth. Catalano et al. had already described this mechanism as a potential inducer of fetal growth [33], although there is no evidence that it specifically prevents the occurrence of SGA. Nevertheless, it is plausible to consider that in certain metabolic profiles, this dynamic may attenuate the negative impact of inadequate GWG.

In addition, other authors suggest that the hyperglycemic environment may promote placental angiogenesis through activation of pathways such as VEGF2 [34] and IGF-1 [35], potentially improving nutrient supply to the fetus. Although we lack direct confirming studies, these mechanisms may help explain the absence of association between SGA and insufficient GWG in pregnant women with GDM in our cohort. Another important factor potentially influencing this outcome is the frequent and intensified prenatal care provided to patients diagnosed with GDM, compared to those without high-risk pregnancies. In this context, our results suggest the possibility that the clinical context of GDM, with its metabolic and care-specific particularities, may influence the relationship between insufficient GWG and risk of SGA, even in women with lower BMI—a hypothesis that, although speculative, could be a subject for future research.

4.4. Other Outcomes

4.4.1. Preeclampsia

In the present study, we observed a lower prevalence of preeclampsia among pregnant women with GDM as GWG decreased, comparing the three groups (insufficient, adequate, and excessive). This association remained significant even after adjustment for potential confounders, using GWG as a continuous variable, indicating an independent and linear relationship between increased weight gain and the occurrence of preeclampsia. Similar findings were reported by Ferreira et al. in a study of pregnant women with obesity and GDM, which showed that insufficient GWG, according to NAM guidelines, was associated with a lower risk of gestational hypertension and preeclampsia [16]. Furthermore, Kiefer et al., who analyzed data from 9322 births of women with GDM or pregestational diabetes, also demonstrated that insufficient GWG was associated with a lower risk of adverse outcomes such as preeclampsia [36].

4.4.2. Cesarean Delivery

When evaluating the entire sample, we observed a direct relationship between higher GWG and increased cesarean delivery rates. This finding was confirmed in the adjusted analysis, showing a 69% higher chance of cesarean delivery among women with excessive GWG compared to those with adequate GWG. On the other hand, insufficient GWG was not associated with a lower chance of cesarean delivery (OR 0.88; 95% CI 0.60–1.30; p = 0.54). In the stratified analysis by obesity class, similar results were observed for women with class I obesity. Among women with class II and III obesity, no significant association between GWG and mode of delivery was identified. A possible explanation is that in cases of milder obesity, lower GWG may positively influence the clinical team’s decision to opt for vaginal delivery. In patients with more advanced obesity, GWG apparently does not exert enough influence to modify this decision, with higher pregestational BMI being a more determining factor in choosing cesarean delivery.

These results are consistent with the study by Shi et al. in women with GDM, which showed that GWG above the NAM guidelines was associated with a higher risk of cesarean delivery [29]. Similarly, Benhalima et al. reported that excessive GWG in women with GDM was linked to an increased risk of cesarean delivery, whereas insufficient GWG was not associated with this mode of delivery (p = 0.06) [27]. Some studies suggest that insufficient GWG may be a protective factor against cesarean delivery [16,28].

The data from the present study, combined with evidence from the literature, suggest that insufficient GWG tends to be a potentially protective factor against cesarean delivery in patients with GDM, reinforcing the hypothesis that, within safe limits, lower weight gain may represent a clinical advantage for this group of pregnant women.

4.4.3. Insulin Requirement During Pregnancy

Additionally, we observed that lower levels of GWG were associated with a reduced insulin requirement during pregnancy, both in the overall analysis and in the subgroup of women with class I obesity. These findings suggest that moderated weight gain may contribute to better glycemic control and less severe metabolic dysfunction in pregnant women with GDM.

These findings align with the results of Cheng et al., who analyzed 31,074 pregnant women with GDM and demonstrated that GWG below the NAM recommendations was associated with a higher likelihood of maintaining glycemic control with diet alone and a lower need for pharmacological therapy, while those with GWG above the recommendations had a greater requirement for insulin therapy [37]. Similarly, Park et al., investigating Korean pregnant women with GDM and overweight or obesity, showed that GWG below recommendations was also associated with reduced insulin requirement, without a significant increase in the incidence of SGA, even with an average gain of only 2.4 kg [10]. These data reinforce the hypothesis that GWG is directly related to the severity of glycemic disturbance during pregnancy and that stricter control may reduce the need for insulin without compromising maternal–fetal safety, as evidenced in our sample.

In summary, our findings reinforce the need for a critical reevaluation of current GWG targets in patients with GDM, especially in the presence of obesity. Although current guidelines are still based on general recommendations, the data presented here—together with the growing body of evidence in the literature—point to the potential benefit of more individualized approaches. Adopting stricter GWG targets, under proper monitoring, may represent an effective and safe strategy to improve metabolic control and maternal–fetal outcomes in this population.

It is important to highlight that GWG is a potentially modifiable variable. Strategies such as regular supervised physical activity and specialized nutritional counseling have proven effective both in weight control and in improving insulin resistance in pregnant women with GDM [38,39,40]. These interventions should be considered essential components of clinical management. Despite the potential benefits of stricter GWG targets, it is advisable that this approach be accompanied by fetal growth monitoring during prenatal care to prevent possible adverse effects on fetal development [9,15].

5. Strengths and Limitations

One strength of this study is the large number of pregnancies evaluated, including patients with GDM, along with analyses stratified by obesity class. A sample with these characteristics is both relevant and innovative, given that the classification proposed by the NAM was not originally developed for high-risk pregnant women, making the investigation of its applicability in this group particularly meaningful.

Our study has some limitations inherent to its retrospective design, including the reliance on data obtained from previously collected records (databases and medical charts), which may be subject to information bias, lack of standardization in certain variables (e.g., presence of neonatal jaundice), and missing relevant data for the complete evaluation of outcomes (e.g., lack of information on glycemic control through capillary glucose monitoring—considered the gold standard for glycemic management in patients with GDM). In addition, laboratory variables such as insulinemia and other direct markers of insulin resistance (e.g., HOMA-IR) were not consistently recorded, even in subgroups, which prevented their inclusion in the analysis and limited a more comprehensive characterization of this aspect (insulin resistance). Although the analysis was adjusted for several important confounding factors, unmeasured variables such as nutritional intake or physical activity levels may still have influenced both gestational weight gain and pregnancy outcomes.

It is also important to consider that the study population was drawn from a single tertiary care center, which may have reduced the occurrence of adverse outcomes due to access to specialized medical care, thereby limiting the generalizability of the findings. On the other hand, the standardized clinical management in this setting helps prevent potential biases related to variations in care, thus ensuring greater internal validity. As a future perspective, prospective studies could minimize these biases by allowing for more rigorous control of confounding variables.

Despite these limitations, our findings contribute to a better understanding of the influence of GWG in patients with GDM, reinforcing the importance of specialized clinical follow-up in medical practice.

6. Conclusions

This study contributes to the growing body of evidence that more restrictive GWG targets in patients with GDM may be safe and associated with improved maternal and fetal outcomes, as well as additional benefits such as better metabolic control. This association appears to be particularly evident in pregnant women with both GDM and obesity. The relationship between lower GWG and reduced insulin requirement during pregnancy, lower cesarean delivery rates, and lower occurrence of LGA, without an increase in SGA, supports the hypothesis that moderating weight gain during pregnancy plays a relevant role in the clinical management and outcomes of women with GDM across different degrees of obesity. However, it is essential that such interventions are not applied indiscriminately and that careful fetal monitoring is ensured—such as through serial obstetric ultrasounds to assess fetal growth—to avoid potential consequences of excessive restriction. Randomized clinical trials will be necessary to validate the safety and effectiveness of different GWG recommendations in women with GDM and obesity in clinical practice, in order to support a potential reassessment of current guidelines for this population.