Association of a high-protein and low-glycemic-index diet during pregnancy with offspring growth and obesity until the age of 18 years – a target trial emulation

doi:10.21203/rs.3.rs-4938319/v1

Download PDF

Article

Association of a high-protein and low-glycemic-index diet during pregnancy with offspring growth and obesity until the age of 18 years – a target trial emulation

https://doi.org/10.21203/rs.3.rs-4938319/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

Maternal pre-pregnancy BMI and excessive gestational weight gain (GWG) are associated with offspring obesity. Moreover, maternal dietary patterns, particularly protein intake and glycemic index, influence GWG and fetal development. This study aimed to investigate the association of a high-protein, low-glycemic-index (HPLGI) diet during pregnancy with offspring growth and obesity.

Methods

Using observational data from the Danish National Birth Cohort, a target trial emulation was conducted to investigate the association of an HPLGI diet during pregnancy on offspring growth up to 18 years of age. A cohort of 17,551 women who met the inclusion criteria were categorized into exposure groups based on their protein intake and glycemic index to emulate the groups in the APPROACH trial.

Results

Offspring outcomes were assessed at various ages and a linear mixed model was used including potential confounders. Offspring born to women following an HPLGI diet during pregnancy had 2.59 kg higher body weight and 0.72 kg/m² higher BMI at 18 years compared to those from the moderate-protein, moderate-glycemic-index (MPMGI) group. Maternal characteristics varied between the exposure groups, with the HPLGI women exhibiting 1.67 kg higher pre-pregnancy weight and 0.49 kg/m² BMI, with no differences in GWG.

Conclusion

Offspring born to women consuming an HPLGI diet during pregnancy had higher body weight and BMI at 18 years compared to offspring born to women on an MPMGI diet. These findings raise questions about the long-term implications of maternal dietary patterns on offspring health. Further research is needed to elucidate the mechanisms underlying these associations.

Health sciences/Risk factors

Health sciences/Health care/Nutrition

Health sciences/Health care/Public health/Epidemiology

maternal nutrition

target trial emulation

pregnancy

fetal programming

prevention of childhood obesity

Individuals with obesity are at higher risk for developing many non-communicable diseases such as type 2 diabetes, heart disease, and cancer [1]. In 2020, about 50% of adults and 19% of primary school-aged children in Denmark were classified as overweight or obese [2, 3]. Globally, the prevalence of childhood obesity has tripled since the 1990s, and children who are overweight are significantly more prone to becoming adults with obesity [2, 4]. Preventing childhood obesity may thus have widespread implications for non-communicable diseases in the whole population.

Pre-pregnancy overweight and excessive gestational weight gain (GWG) increase the risk of the offspring being born large for gestational age and developing overweight later in life [5]. Despite the recommendation for GWG below 9 kg for women with pre-pregnancy obesity, almost 60% of them exceed this threshold [6]. The characteristics of the maternal diet, including low dietary protein intake and low protein-to-carbohydrate ratio, appear to be associated with increased GWG, fetal adiposity, and fetal abdominal fat deposition [7–9]. Research in non-pregnant individuals exploring diets with varying protein-to-carbohydrate ratios and glycemic index (GI) has highlighted the efficacy of a high-protein diet with a low GI in preventing weight regain after weight loss [10]. In pregnant women, high GI diets are associated with an increased risk of excessive GWG, postnatal weight retention, and delivery of large-for-gestational-age infants, who are likely to become adults with elevated BMI [11–13]. Throughout pregnancy, the maternal diet serves as the sole source for the developing fetus. Nevertheless, conflicting results exist regarding the association between maternal protein intake and offspring birth weight, with some studies reporting a positive association [14] and others not [15, 16]. Additionally, studies on maternal protein intake during pregnancy and offspring later weight are also inconclusive [17, 18].

An intervention study—the APPROACH trail—targeting higher protein and lower GI intake among well-nourished overweight and obese pregnant women found significant effects on lower GWG (1.7 kg) [19] and a positive association between GWG and offspring BMI z-score at birth [20]. However, no effect of the intervention diet was found on offspring BMI until the age of 5 years but showed some indication of worse metabolic profile in the offspring compared to controls (ref – in progress). Here, we utilize data from the large Danish National Birth Cohort (DNBC) to conduct an emulation of the APPROACH trial. Our aim is to replicate the dietary regimens employed in the APPROACH trial for the purpose of investigating the long-term association of an HPLGI diet during pregnancy on offspring growth up until the age of 18 years within an observational setting of DNBC.

We first specified the protocol and developed a data analysis strategy for our target trial emulation. Key components are described in Table 1.

Table 1

Target Trial Emulation of the APPROACH dietary intervention study.
Protocol	APPROACH RCT	Target Trail Emulation
Eligibility criteria	Pre-pregnancy BMI 28–45 Older 18–45 years Singleton pregnancy No abuse of alcohol or drugs (> 14 units of alcohol per week) No critical or chronic diseases	Pre-pregnancy BMI 25–45 Older than 18 years Singleton pregnancy No abuse of alcohol or drugs (> 14 units of alcohol per week) Plausible total energy intakes (> 2,500 kJ/day and < 25,000 kJ/day)
Treatment strategy	High-protein (25–28 E%), and low-GI (GI < 55) ad libitum diet. Active Comparator: Moderate-protein (protein 18 E%) moderate-GI (no information on restricting glycemic load but guided following the principles of the New Nordic Diet (GI ~ 60)) ad libitum diet.	High-protein (≥ 18 E%) and low-GI (GI ≤ 55) ad libitum diet. Active Comparator: Moderate-protein (protein < 18 E%) moderate-GI (GI > 55–100) ad libitum diet.
Randomization	Participants were randomly assigned in clusters of 4–8 women to either the HPLGI diet or the MPMGI diet.	We assumed randomization conditional on the eligibility criteria, and the exposure diet, and adjusted for covariates: maternal pre-pregnancy BMI, SES, GWG, parity, offspring sex, and offspring gestational age.
Outcome	Primary outcome: Gestational weight gain Secondary outcome: Growth and development of fetus and child from birth to 9 years of age.	Primary outcome: Growth and development of fetus and child from birth to 18 years of age. Secondary outcome: Gestational weight gain
follow-up	Starts after the randomization in 2013 and ends at the last follow-up visit at 9 years of age, expected in 2025.	Starts after the dietary questionnaires in 1996 and 2003 and ends with the follow-up of the offspring at 18 years.
Causal contrast	Intention to treat effect; per-protocol effect.	Observational analog of per-protocol effect.
Analysis plan	Available case analyses of the follow-up of the offspring. The linear mixed model methodology with offspring age as a fixed effect and participant-specific ID as a random effect was used to assess the difference between the HPLGI diet and the MPMGI diet on offspring outcomes.	Same as the APPROACH trial + pre-assignment data including the eligibility criteria and thresholds for the exposure group and post-assignment confounders such as maternal pre-pregnancy BMI, SES, GWG, parity, offspring sex, and offspring gestational age.

Study design and setting

The DNBC is a large prospective cohort study, which enrolled almost 95,000 mother-offspring dyads between 1996 and 2003. During pregnancy, the women completed a food frequency questionnaire (FFQ) in gestational week (GW) 25 [21]. The women gave written informed consent on behalf of themselves and their offspring. The DNBC was conducted according to the guidelines in the Declaration of Helsinki, and the National Committee on Health Research Ethics in Denmark (H-4-2011-045) approved all procedures.

Participants

Among participating women, 69,807 had information on nutrient intake from the FFQ administered in GW 25 (Fig. 1). From these, we selected a sub-cohort based on the APPROACH randomized controlled trial (RCT) inclusion and exclusion criteria: BMI 28–45 kg/m²; singleton birth; older than 18 years; and no alcohol or drug abuse. Additionally, we excluded implausible total energy intakes (< 2,500 kJ/day and > 25,000 kJ/day) (~ 1.0%). All the inclusion criteria were met by 17,551 women (25.1%).

Modifications of the target trial protocol, to adapt the analyses to the observational setting within DNBC

We applied the design principles of the target trial methodology [22] to replicate the exposure groups of the APPROACH RCT and prolong the follow-up period for offspring growth until the age of 18 years. Nevertheless, the exposure as compared to the original trail protocol could not be replicated identically. Therefore, we made necessary modifications to the trial protocol (Table 1). Furthermore, we attempted to emulate randomization through the eligibility criteria and by adjusting for confounding variables.

Exposure assessment

A 360-item FFQ was used to collect information on maternal diet in GW 25. All frequencies were computed into times per day. To estimate the intake of each food item in grams, standard portion sizes were multiplied by the daily frequencies. The FFQ had three main components: food frequency, dietary supplements, and other information. The questionnaire was validated against 7-day food diaries and urine nitrogen excretion (n = 88) [23] and developed from the questionnaire used by the Danish Cancer Registry [24]. Specific GI values for each food-item in the FFQ were extracted from published data [25] and were calculated by multiplying available carbohydrates (in grams) for the food item with the GI of the specific food item, summing up, and dividing by the total intake of available carbohydrates. White bread was used as a reference.

Exposure groups

Before the authors had access to outcomes, two exposure groups were defined based on maternal protein intake and GI. According to the hypothesis of the APPROACH RCT, we defined a high-protein low-glycemic index (HPLGI) group as protein ≥ 18 percent of total energy intake (E%) and GI ≤ 55 and a moderate-protein moderate glycemic index (MPMGI) group as protein < 18 E% and GI > 55–100. The chosen GI cut-off was determined by the specified threshold in the APPROACH RCT, and the International Standard Organization categorization for low GI [26]. For the MPMGI group, the GI was set to range from 55 to 100 to facilitate differences in GI between the two groups. The protein intake cut-off was selected considering the Nordic nutrition recommendations (dietary protein 10–20 E%) [27], and the high-protein threshold was then set at ≥ 18 E%. The final exposure groups consisted of 372 mother-offspring dyads in the HPLGI group and 6,643 mother-offspring dyads in the MPMGI group (Fig. 1).

Outcome variables

GWG was self-reported during an interview 6 months postpartum. Offspring weight and length were obtained from The National Patient Registry at birth, 5 months of age, and 12 months of age. Moreover, offspring weight and height were obtained at ages 7 years, 11 years, 14 years, and 18 years by questionnaires. Age- and sex-specific BMI z-scores were calculated by using the WHO Child Growth Standards for offspring aged < 5 years and WHO Growth Reference for those ≥ 5 years [28, 29]. We further classified the offspring as overweight/obese by using the corresponding age- and sex-specific WHO cutoffs [i.e., 1-1.99 SD as overweight and ≥ 2 SD as obese] [30].

Covariate assessment

Based on telephone interviews in GW 12 and GW 30 information on the following set of covariates included in our analyses were recorded: Parity (nulliparous or not), socioeconomic status (based on the highest level of professional occupation within the couple and categorized as “high” including high- or medium-level professionals, or skilled workers, or “low” including unskilled workers and others,), pre-pregnancy BMI (calculated as self-reported pre-pregnancy weight in kilograms divided by height in meters squared and categorized into overweight: BMI 25-29.9 kg/m² and obese: BMI ≥ 30 kg/m²), and smoking during pregnancy (Daily < 15 cigarettes, Daily ≥ 15 cigarettes, Occasionally, or Non-smoker). Information on the offspring's gestational age at delivery (days) was extracted from medical records. Breastfeeding duration was obtained from questionnaires 6 months postpartum and defined as the duration of any breastfeeding, not distinguishing between exclusive and supplemented breastfeeding, in categories of ≤ 1, 1–4, 4–6, or ≥ 6 months.

Statistics

To ensure that the sample size was adequate for comparing two means with a two-sided test, a power calculation was conducted. With a study population of 369 offspring at birth in the HPLGI group and 6,561 offspring in the MPMGI group and an estimated SD for BMI z-score of 1.16,[31] the difference in BMI z-score between the groups needed to be at least 0.174 SD to be detected with an 80% probability at a p-value of 0.05.

Maternal characteristics are presented as mean (SD) for continuous variables and frequencies for categorical variables. Differences between the HPLGI group and the MPMGI group were analyzed using ANOVA. To assess the difference between the HPLGI and MPMGI groups in offspring outcomes during the first 18 years of life, we used a linear mixed model with offspring age (birth, 5 months, 12 months, 7 years, 11 years, 14 years, and 18 years) as a fixed effect and participant-specific ID as a random effect. We performed an unadjusted model and a model adjusted for maternal pre-pregnancy BMI, SES, GWG, parity, offspring sex, and offspring gestational age at birth as independent variables. The residuals of all models described above were investigated to assess normality. The Sidak correction was used to adjust the significance level for multiple comparisons over time within the group and presented as p-adjusted.

Study Participants

Among the 69,807 pregnant women, 17,551 (25.1%) met the inclusion/exclusion criteria. Among these, 372 women were assigned to the HPLGI diet group, and 6,643 women were assigned to the MPMGI diet group. Follow-up assessments of offspring growth were completed from birth to 18 years of age with a follow-up rate from 99% at birth to ~ 46% at 18 years of age (Fig. 1). Women in the HPLGI group had higher pre-pregnancy weight (1.67 kg, P = 0.007) and higher pre-pregnancy BMI (0.49 mg/m², P = 0.010) compared to those in the MPMGI group. Additionally, women who consumed an HPLGI diet had on average 2,774 kJ/day lower energy intake compared to those in the MPMGI group. GWG was equal among the two groups (P = 0.134) (Table 2).

Table 2

Maternal Characteristics
	N	HPLGI (N = 372)	N	MPMGI (N = 6643)	(HPLGI-MPMGI Estimate)	P-value
Maternal age at delivery	372	30.4 (4.06)	6643	30.5 (4.17)	-0.06 (-0.50; 0.37)	0.776
Pre-pregnancy weight (kg)	372	83.3 (11.6)	6643	81.6 (11.6)	1.67 (0.46;2.88)	0.007
Pre-pregnancy height (cm)	372	168.8 (6.33)	6643	168.5 (6.05)	0.36 (-0.27; 0.99)	0.266
Pre-pregnancy BMI (kg/m²)	372	29.2 (3.73)	6643	28.7 (3.56)	0.49 (0.11;0.86)	0.010
BMI categories						0.040
Overweight (%)	250	67.2	4801	72.3
Obese (%)	122	32.8	1842	27.7
GWG	291	12.88 (7.27)	5425	13.52 (7.03)	-0.64 (-1.47;0.20)	0.134
Parity (%)						0.144
0	182	48.9	2916	43.9
1	130	34.9	2618	39.4
2+	60	16.1	1108	16.7
Breastfeeding						0.981
≤ 1 months	67	46.2	1378	46.3
1–4 months	32	22.1	670	22.5
4–6 months	35	24.1	731	24.5
6 + months	11	7.6	200	6.7
Smoking (%)						0.574
Daily < 15 cigarettes	34	9.1	664	10.0
Daily 15 + cigarettes	8	2.2	135	2.0
Occasionally	49	13.2	726	10.9
None smoker	281	75.5	5118	77.0
Socioeconomic status (%)						0.016
High	315	84.9	5276	79.6
Low	56	15.1	1349	20.4
Energy intake (kJ)	372	7139 (1643)	6643	9912 (2174)	-2774 (-2998;-2549)	< 0.001
Protein intake (%)	372	19.73 (1.22)	6643	14.9 (1.75)	4.35 (4.17;4.53)	< 0.001
Fat intake (%)	372	30.5 (5.49)	6643	30.6 (5.74)	-0.18 (-0.78;0.41)	0.547
Carbohydrate (%)	372	49.8 (5.87)	6643	54.0 (5.60)	-4.24 (-4.82;-3.65)	< 0.001
Fiber (g/day)	372	16.5 (6.51)	6643	26.1 (7.72)	-9.52 (-10.3;-8.72)	< 0.001
Added sugar (g/day)	372	25.4 (11.7)	6643	45.8 (24,8)	-20.38 (-22.9;-17.8)	< 0.001
Glycemic index (pr. day)	372	42.8 (10.95)	6643	76.0 (13.02)	-33.19 (-34.54;-31.84)	< 0.001
GWG: Gestational weight gain, HPLGI; High-protein low-glycemic-index, MPMGI; moderate-protein moderate-glycemic-index, BMI; Body mass index. Data are presented as mean (standard deviation (SD)) for continuous variables or as numbers (%) for categorical variables.

Offspring outcomes

Offspring born to women who consumed an HPLGI diet during pregnancy had on average 0.17 less BMI z-scores at birth compared with offspring in the MPMGI group. However, no differences between the HPLGI group and the MPMGI group on offspring outcomes from birth to 14 years of age were found after adjusting the p-value for multiple comparisons. At 18 years of age, offspring born to women who consumed an HPLGI diet had on average 2.59 kg higher body weight (P < 0.001) compared with offspring born to women in the MPMGI group. This corresponded to 0.72 kg/m² higher BMI (P = 0.001) and 0.20 higher BMI z-score (P = 0.040). No differences were observed in height. Whereas body weight and BMI persisted among offspring from the HPLGI diet after using the Sidak correction to adjust the significance level for multiple comparisons, the significance disappeared for BMI z-score (P = 0.248) (Table 3, adjusted model). Moreover, using the WHO age- and sex-specific categories for underweight, normal weight, overweight, or obesity, we found no difference between the groups at different ages (Table 4).

Table 3

Offspring characteristics
	Unadjusted					Adjusted
	HPLGI	MPMGI	(HPLGI-MPMGI Estimate)	P-value	P-adjust	HPLGI	MPMGI	(HPLGI-MPMGI Estimate)	P-value	*P-adjust
Body weight, kg
Birth	3.7 (0.38)	3.7 (0.09)	-0.04 (-0.80;0.73)	0.921	1.000	3.5 (0.43)	3.7 (0.11)	-0.18 (-1.04;0.67)	0.673	1.000
5 months	7.9 (0.46)	7.9 (0.11)	0.01 (-0.93;0.94)	0.991	1.000	7.9 (0.52)	7.9 (0.13)	-0.08 (-1.11;0.96)	0.887	1.000
12 months	10.5 (0.47)	10.4 (0.11)	0.09 (-0.86;1.04)	0.851	1.000	10.4 (0.53)	10.4 (0.13)	-0.01 (-1.07;1.05)	0.986	1.000
7 years	25.9 (0.50)	25.9 (0.12)	-0.03 (-1.04;0.97)	0.950	1.000	26.0 (0.56)	25.9 (0.14)	0.03 (-1.09;1.14)	0.964	1.000
11 years	41.6 (0.54)	42.8 (0.13)	-1.21 (-2.29;-0.13)	0.028	0.181	41.5 (0.60)	42.4 (0.15)	-0.95 (-2.16;0.26)	0.122	0.599
14 years	58.8 (0.58)	58.5 (0.13)	0.27 (-0.90;1.43)	0.653	0.999	58.9 (0.64)	58.5 (0.15)	0.40 (-0.90; 1.65)	0.562	0.997
18 years	73.8 (0.55)	73.4 (0.13)	0.48 (-0.63;1.58)	0.397	0.971	76.1 (0.64)	73.5 (0.14)	2.59 (1.31;3.87)	< .0001	< 0.001
Body length/height, cm
Birth	52.5 (0.30)	52.6 (0.07)	-0.07 (-0.67;0.52)	0.812	1.000	52.4 (0.33)	52.5 (0.08)	-0.14 (-0.79;0.50)	0.663	1.000
5 months	68.5 (0.35)	68.3 (0.08)	0.18 (-0.53;0.89)	0.624	0.999	68.4 (0.39)	68.2 (0.10)	0.16 (-0.61;0.93)	0.685	1.000
12 months	77.5 (0.36)	77.6 (0.08)	-0.12 (-0.85;0.60)	0.736	1.000	77.5 (0.39)	77.6 (0.10)	-0.09 (-0.88;0.69)	0.819	1.000
7 years	126.5 (0.38)	126.1 (0.09)	0.44 (-0.32;1.19)	0.258	0.877	126.5 (0.41)	125.9 (0.10)	0.57 (-0.24;1.39)	0.168	0.724
11 years	151.3 (040)	151.5 (0.09)	-0.27 (-1.07;0.53)	0.504	0.993	151.1 (0.44)	151.0 (0.11)	0.03 (-0.84;0.91)	0.944	1.000
14 years	169.5 (0.43)	168.8 (0.10)	0.64 (-0.22;1.51)	0.144	0.663	169.5 (0.46)	168.7 (0.11)	0.77 (-0.15;1.69)	0.101	0.526
18 years	174.9 (0.41)	174.9 (0.09)	-0.04 (-0.86;0.79)	0.928	1.000	175.4 (0.47)	175.0 (0.11)	0.33 (-0.60;1.26)	0.484	0.990
BMI
Birth	13.16 (0.13)	13.28 (0.03)	-0.12 (-0.37;0.14)	0.364	0.958	13.12 (0.14)	13.30 (0.04)	-0.19 (-0.47;0.10)	0.204	0.798
5 months	13.15 (0.15)	13.32 (0.04)	-0.16 (-0.47;0.15)	0.304	0.921	13.11 (0.17)	13.33 (0.04)	-0.22 (-0.56;0.13)	0.224	0.830
12 months	13.17 (0.16)	13.31 (0.04)	-0.14 (-0.45;0.18)	0.399	0.972	13.11 (0.18)	13.32 (0.04)	-0.21 (-0.56;0.14)	0.238	0.851
7 years	13.13 (0.16)	13.34 (0.04)	-0.21 (-0.54;0.12)	0.203	0.796	13.10 (0.18)	13.36 (0.04)	-0.27 (-0.63;0.10)	0.153	0.688
11 years	18.08 (0.18)	18.51 (0.04)	-0.43 (-0.78;-0.08)	0.017	0.115	18.11 (0.20)	18.44 (0.05)	-0.33 (-0.73;0.07)	0.104	0.535
14 years	20.43 (0.19)	20.47 (0.04)	-0.04 (-0.42;0.34)	0.846	1.000	20.47 (0.21)	20.47 (0.05)	-0.01 (-0.42;0.41)	0.982	1.000
18 years	24.06 (0.18)	23.93 (0.04)	0.13 (-0.23;0.49)	0.487	0.991	24.64 (0.21)	23.92 (0.05)	0.72 (0.30;1.14)	0.001	0.005
BMI z-score (0-228 months)
Birth	-0.24 (0.06)	-0.14 (0.01)	-0.10 (-0.02;0.22)	0.095	0.501	-0.28 (0.07)	-0.12 (0.02)	-0.17 (-0.30;0.03)	0.015	0.102
5 months	-0.31 (0.07)	-0.23 (0.02)	-0.08 (-0.06;0.23)	0.263	0.882	-0.28 (0.08)	-0.24 (0.02)	0.04 (-0.20;0.12)	0.606	0.999
12 months	0.51 (0.07)	0.39 (0.02)	0.13 (-0.28;0.02)	0.084	0.460	0.52 (0.08)	0.38 (0.02)	0.14 (-0.02;0.30)	0.093	0.496
7 years	0.29 (0.08)	0.32 (0.02)	-0.03 (-0.12;0.19)	0.686	1.000	0.32 (0.09)	0.32 (0.02)	-0.00 (-0.17;0.17)	0.966	1.000
11 years	0.16 (0.08)	0.31 (0.02)	-0.15 (-0.02;0.31)	0.081	0.448	0.20 (0.09)	0.31 (0.02)	-0.11 (-0.29;0.08)	0.262	0.881
14 years	0.21 (0.09)	0.20 (0.02)	0.01 (-0.19; 0.17)	0.909	1.000	0.21 (0.10)	0.21 (0.02)	0.01 (-0.19;0.19)	0.940	1.000
18 years	0.60 (0.08)	0.54 (0.02)	0.06 (-0.23; 0.11)	0.481	0.990	0.75 (0.10)	0.54 (0.02)	0.20 (0.01:0.40)	0.040	0.248
Weight z-score (0-120 months)
Birth	0.67 (0.06)	0.75 (0.01)	-0.08 (-0.19;0.03)	0.150	0.479	0.65 (0.06)	0.76 (0.02)	-0.11 (-0.23;0.01)	0.068	0.247
5 months	0.51 (0.07)	0.57 (0.02)	-0.06 (-0.19;0.08)	0.416	0.883	0.56 (0.07)	0.56 (0.02)	0.00 (-0.14;0.14)	0.979	1.000
12 months	0.81 (0.07)	0.79 (0.02)	0.02 (-0.11;0.16)	0.753	0.996	0.85 (0.07)	0.77 (0.02)	0.08 (-0.07;0.22)	0.302	0.762
7 years	0.79 (0.07)	0.73 (0.02)	0.05 (-0.09;0.19)	0.465	0.918	0.81 (0.07)	0.72 (0.02)	0.09 (-0.06;0.24)	0.228	0.645
Length z-score (0-228 months)
Birth	1.58 (0.06)	1.63 (0.01)	-0.05 (-0.17;0.07)	0.435	0.982	1.61 (0.07)	1.64 (0.02)	-0.03 (-0.16;0.11)	0.755	1.000
5 months	1.33 (0.07)	1.33 (0.02)	0.00 (-0.14;0.15)	0.966	1.000	1.36 (0.08)	1.33 (0.02)	0.03 (-0.12;0.19)	0.649	1.000
12 months	0.80 (0.07)	0.91 (0.02)	-0.11 (-0.25;0.03)	0.133	0.631	0.85 (0.08)	0.90 (0.02)	-0.05 (-0.21;0.11)	0.558	0.994
7 years	0.96 (0.07)	0.86 (0.02)	0.11 (-0.04;0.26)	0.164	0.715	0.97 (0.08)	0.85 (0.02)	0.12 (-0.04;0.29)	0.127	0.653
11 years	0.83 (0.08)	0.79 (0.02)	0.04 (-0.11;0.20)	0.578	0.998	0.86 (0.09)	0.77 (0.02)	0.09 (-0.09;0.26)	0.312	0.941
14 years	0.93 (0.08)	0.88 (0.02)	0.05 (-0.12;0.22)	0.552	0.996	0.96 (0.09)	0.87 (0.02)	0.09 (-0.09;0.27)	0.336	0.950
18 years	0.84 (0.08)	0.90 (0.02)	-0.06 (-0.23;0.10)	0.446	0.984	0.84 (0.09)	0.90 (0.02)	-0.05 (-0.24;0.13)	0.606	0.997
Data are presented as estimated means (SE). The difference between groups is presented as the estimated means from the HPLGI-MPMGI with 95% confidence intervals (CI). HPLGI: high-protein low–glycemic index; MPMGI: moderate-protein moderate–glycemic index. The adjusted model was adjusted for pre-pregnancy BMI, GWG, parity, gestational age, socioeconomic status, and offspring sex. * Adjusted with the Sidak correction.

Table 4

– Obesity categories
	HPLGI	MPMGI	P-value
Birth			0.900
Underweight, n (%)	19 (5.15)	354 (5.40)
Normal weight, n (%)	344 (93.23)	6095 (92.89)
Overweight, n (%)	6 (1.63)	100 (1.52)
Obese, n (%)	0 (0.00)	12 (0.18)
5 months			0.800
Underweight, n (%)	17 (7.17)	250 (5.70)
Normal weight, n (%)	214 (90.30)	4025 (91.77)
Overweight, n (%)	5 (2.11)	93 (2.12)
Obese, n (%)	1 (0.42)	18 (0.41)
12 months			0.700
Underweight, n (%)	4 (1.75)	87 (2.04)
Normal weight, n (%)	208 (91.23)	3872 (90.98)
Overweight, n (%)	12 (5.26)	253 (5.94)
Obese, n (%)	4 (1.75)	44 (1.03)
7 years			0.400
Underweight, n (%)	3 (1.48)	62 (1.64)
Normal weight, n (%)	192 (94.58)	3489 (92.18)
Overweight, n (%)	8 (3.94)	195 (5.15)
Obese, n (%)	0 (0.00)	39 (1.03)
11 years			0.700
Underweight, n (%)	7 (4.02)	84 (2.65)
Normal weight, n (%)	156 (89.66)	2887 (91.10)
Overweight, n (%)	11 (6.32)	193 (6.09)
Obese, n (%)	0 (0.00)	5 (0.16)
14 years			0.800
Underweight, n (%)	2 (1.34)	61 (2.17)
Normal weight, n (%)	141 (94.63)	2633 (93.77)
Overweight, n (%)	6 (4.03)	105 (3.74)
Obese, n (%)	0 (0.00)	9 (0.32)
18 years			0.800
Underweight, n (%)	3 (1.82)	43 (1.32)
Normal weight, n (%)	148 (89.70)	2865 (88.24)
Overweight, n (%)	10 (6.06)	261 (8.04)
Obese, n (%)	4 (2.42)	78 (2.40)
Data are presented as numbers (%) of participants in each obesity category; underweight, normal weight, overweight, or obese. HPLGI: high-protein low–glycemic index; MPMGI: moderate-protein moderate–glycemic index.

This population-based cohort study found that offspring born to women who consumed an HPLGI diet during pregnancy exhibited higher body weight and BMI at 18 years of age compared to those born to women consuming an MPMGI diet during pregnancy. Moreover, offspring from the HPLGI group had a higher BMI z-score at 18 years but this difference disappeared after adjusting the p-value for multiple comparisons.

In the APPROACH RCT, offspring born to women who consumed an HPLGI diet during their pregnancy exhibited unfavorable effects in their lipid blood markers at 3 and 5 years of age compared with offspring born to women on an MPMGI diet (ref– in progress). Collectively, these observations suggest that while an HPLGI diet may have a positive impact on maternal outcomes such as limited GWG and reduced pregnancy complications[19], it may have an unfavorable impact on long-term offspring outcomes. However, differences in BMI z-scores between the two APPROACH groups did not reach statistical significance (ref– in progress), consistent with our findings on BMI z-scores in early childhood from the present study.

It is well documented that elevated maternal blood glucose levels are associated with excessive fetal growth, cord blood serum C-peptide, as well as other adverse pregnancy outcomes [32]. Implementing a low GI diet during pregnancy has been associated with reduced birth weight, decreased incidence of macrosomia, and lower risk of offspring being born large-for-gestational-age among women with a high risk of gestational diabetes mellitus [33]. The potential mechanisms underlying the benefits of a low-GI diet may include the attenuated rise of postprandial glucose levels, which in turn reduces hyperinsulinemia[34] and oxidative stress [35]. Studies examining the effects of low GI diets and high GI diets indicate that offspring born to women following a low-GI diet during pregnancy tend to have a lower ponderal index, a marker of adiposity at birth [36, 37]. Additionally, another study examined differences in offspring BMI z-score across GI quantiles of the maternal diet found that children born to women with higher GI intake (Q4: GI 111–1603) during pregnancy had higher BMI z-scores at 7 years compared to those in the lowest GI quantile (Q1: GI 6–63) [38]. These observations support the concept that exposure to elevated glucose levels in utero can influence fetal growth and development [39, 40]. Despite the apparent benefits of a low GI diet during pregnancy on offspring health, our study did not support these findings. One possible explanation could be the interaction between the high protein and low GI, leading to a masking effects. High-protein foods tend to have a lower GI due to their slower digestion and absorption rates. This could potentially counteract the expected benefits of low GI foods.

However, ensuring adequate maternal protein intake during pregnancy is important to ensure sufficient amino acid availability for fetal growth and development. Consequently, there is ongoing speculation regarding the impact of increased maternal protein consumption during pregnancy on fetal development. Both excessive and insufficient maternal protein intake can result in adverse pregnancy outcomes such as impaired fetal growth [41]. However, it is hypothesized that heightened exposure to protein during fetal development might impact the offspring's future protein needs and appetite regulation, potentially leading to calorie overconsumption in an attempt to meet the ‘upregulated’ protein needs, which can lead to weight gain [42]. Additionally, the overall protein intake during infancy (from birth to 2y) appears to be associated with later obesity outcomes during childhood or adolescence [43, 44]. This suggests that offspring with prolonged exposure to high protein during infancy, and potentially exposure to high protein in their early environment, may be associated with the development of overweight and obesity in the long term. In another Danish cohort, an association between protein intake (substituted for carbohydrate) during pregnancy and the risk of offspring being overweight at 19–21 years of age was found [18], consistent with our findings at 18 years in the DNBC. Thus, the possibility exists that the influence of protein exposure in utero on the risk of obesity in the offspring manifests later in life, potentially explaining the higher weight and BMI observed in our study at 18 years. On the other hand, adequate protein intake is needed for the regulation of muscle mass, which is a major component of fat-free mass. In the generation R study including 2,694 Dutch mother-offspring dyads, an association was found between maternal protein intake during pregnancy and higher offspring fat-free mass index, but not BMI or fat mass index at 6 years, measured by DEXA[17]. This highlights the limitation of BMI and BMI z-score as these metrics do not account for body composition, particularly when considering the impact of high protein diets on fat-free mass [45].

The generalizability of the findings from the DNBC may be limited due to lifestyle differences in the late 1990s, notably lower rates of obesity, and lower intakes of protein. In the present study, women in the HPLGI group had higher pre-pregnancy BMI, and significantly lower energy intake compared with women in the MLMGI group, while gaining the same weight during pregnancy. It is well-documented that higher BMI is associated with a higher tendency for underreporting [46]. If indeed the HPLGI group is underreporting their dietary intake, this could potentially introduce a bias regarding the percent contribution of protein to total energy. Furthermore, all dietary data were self-reported in an FFQ by the women in GW 25. To enhance accuracy, the most extreme measures of energy intake were excluded, thus reducing the magnitude of measurement error. Moreover, offspring height and weight measurements were self-reported at 7 years, 11 years, 14 years, and 18 years by questionnaires, but the extent of potential misreporting in weight may have been less apparent due to the varying time intervals between the initial exposure (during pregnancy) and subsequent assessments (from birth to 18 years of age). However, this does not negate the possibility that such factors could be linked to misreporting. Despite these limitations, this cohort is a unique source of valuable longitudinal data with repeated measurements that allowed us to use detailed dietary information collected during pregnancy and offspring markers of adiposity from birth to the age of 18 years, while controlling for potential confounders.

In summary, this population-based cohort study suggests that offspring born to women who consumed an HPLGI diet during pregnancy exhibited higher body weight and BMI at 18 years of age compared to those born to women consuming an MPMGI diet. Further research may be needed to explore these findings and clarify the potential mechanisms and implications of maternal diet during pregnancy on offspring health outcomes.

APPROACH

An optimized programming of healthy children

DNBC

Danish National Birth Cohort

FFQ

Food frequency questionnaire

Glycemic index

Gestational week

GWG

Gestational weight gain

HPLGI

High-protein low-glycemic-index

MPMGI

Moderate-protein moderate-glycemic-index

RCT

Randomized controlled trial

Data Availability Statement:

Data sharing is restricted to uphold the privacy protection of participants within the Danish National Birth Cohort. Data are available from The Danish National Birth Cohort Secretariat, e-mail: bsmb@ssi. dk, for researchers who meet the criteria for access to confidential data.

Declaration of interests:

The author(s) declare none.

Acknowledgement

The authors wish to thank the pregnant women and their children for participating in the studies. Additionally, gratitude is extended to general practitioners and midwifes who dedicated themselves to the project. Also a thanks to the statistician Anna Damkjær Laksafoss for support and assistance in the analyses of this manuscript.

Author contribution statement

The authors’ contributions were as follows – CSM, AAB, CM, FM, NRWG, SFO, and TIH conceptualized and contributed to the design of the study and reviewed and edited the manuscript; CSM conducted the data analysis and drafted the manuscript; CG had full access to the data in the study and was responsible for data management. CSM has the responsibility for the integrity of the data analysis. All authors read and approved the ﬁnal manuscript.

Funding

March of Dimes Birth Defects Foundation (6-FY-96-0240, 6-FY97-0553, 6-FY97-0521, 6-FY00-407), the European Union (QLK1-2000-00083), the Danish Medical Research Foundation (9601842 and 22-03-0536), the Health Foundation (11/263-96), and the Heart Foundation (96-2-4-83-22450) and the Innovation Fund Denmark (grant 09-067124) supported the collection, development, and elaboration of the dietary data. Financial support for the Danish National Birth Cohort was also obtained from the Danish National Research Foundation, the Pharmacy Foundation, and the Egmont Foundation.

Jensen MD, Ard JD, Comuzzie AG, et al (2014) 2013 AHA/ACC/TOS Guideline for the Management of Overweight and Obesity in Adults: HHS Public Access Preamble and Transition to ACC/AHA Guidelines to Reduce Cardiovascular Risk. Circulation 129:102–138
World Obesity Federation. https://data.worldobesity.org/country/denmark-55/#data_prevalence. Accessed 24 Jan 2024
WHO European Childhood Obesity Surveillance Initiative (COSI).
Simmonds M, Llewellyn A, Owen CG, Woolacott N (2016) Predicting adult obesity from childhood obesity: a systematic review and meta-analysis. https://doi.org/10.1111/obr.12334
Heslehurst Id N, Vieira R, Id ZA, Id HB, Slack E, Ngongalah L, Pemuid A, Rankinid J (2019) The association between maternal body mass index and child obesity: A systematic review and meta-analysis. PLoS Med 16:1–20
Nohr EA, Vaeth M, Baker JL, Sørensen TI, Olsen J, Rasmussen KM (2008) Combined associations of prepregnancy body mass index and gestational weight gain with the outcome of pregnancy (American Journal of Clinical Nutrition (2008) 87, (1750–1759)). American Journal of Clinical Nutrition 88:1705
Maslova E, Halldorsson TI, Astrup A, Olsen SF Dietary protein-to-carbohydrate ratio and added sugar as determinants of excessive gestational weight gain: a prospective cohort study. https://doi.org/10.1136/bmjopen-2014-005839
Blumfield ML, Hure AJ, MacDonald-Wicks LK, Smith R, Simpson SJ, Giles WB, Raubenheimer D, Collins CE (2012) Dietary balance during pregnancy is associated with fetal adiposity and fat distribution. American Journal of Clinical Nutrition 96:1032–1041
Goyenechea E, Holst C, van Baak MA, et al (2011) Effects of different protein content and glycaemic index of ad libitum diets on diabetes risk factors in overweight adults: the DIOGenes multicentre, randomized, dietary intervention trial. Diabetes Metab Res Rev 27:705–716
Larsen TM, Dalskov S-M, Baak M van, et al (2010) Diets with High or Low Protein Content and Glycemic Index for Weight-Loss Maintenance. N Engl J Med 363:2102
Moses RG, Luebcke M, Davis WS, Coleman KJ, Tapsell LC, Petocz P, Brand-Miller JC (2006) Effect of a low-glycemic-index diet during pregnancy on obstetric outcomes. American Journal of Clinical Nutrition 84:807–812
Knudsen VK, Heitmann BL, Halldorsson TI, Sorensen TIA, Olsen SF (2013) Maternal dietary glycaemic load during pregnancy and gestational weight gain, birth weight and postpartum weight retention: A study within the Danish National Birth Cohort. British Journal of Nutrition 109:1471–1478
Moses RG, Luebke M, Petocz P, Brand-Miller JC (2007) Maternal diet and infant size 2 y after the completion of a study of a low-glycemic-index diet in pregnancy. Am J Clin Nutr 86:1806
Cucó G, Arija V, Iranzo R, Vilà J, Prieto MT, Fernández-Ballart J (2006) Association of maternal protein intake before conception and throughout pregnancy with birth weight. Acta Obstet Gynecol Scand 85:413–421
Murrin C, Shrivastava A, Kelleher CC (2013) Maternal macronutrient intake during pregnancy and 5 years postpartum and associations with child weight status aged five. Eur J Clin Nutr 67:670–679
Halldorsson TI, Birgisdottir BE, Brantsæter AL, Meltzer HM, Haugen M, Thorsdottir I, Olafsdottir AS, Olsen SF (2021) Old Question Revisited: Are High-Protein Diets Safe in Pregnancy? Nutrients 13:1–12
Tielemans MJ, Steegers EAP, Voortman · Trudy, Jaddoe VW V, Rivadeneira F, Franco OH, Kiefte-De Jong JC (2017) Protein intake during pregnancy and offspring body composition at 6 years: the Generation R Study. Eur J Nutr 56:2151–2160
Maslova E, Rytter D, Bech BH, Henriksen TB, Rasmussen MA, Olsen SF, Halldorsson TI (2014) Maternal protein intake during pregnancy and offspring overweight 20 y later 1–3. Am J Clin Nutr 100:1139–1187
Geiker NWR, Magkos F, Zingenberg H, Svare J, Chabanova E, Thomsen HS, Ritz C, Astrup A (2022) A high-protein low-glycemic index diet attenuates gestational weight gain in pregnant women with obesity: the “An optimized programming of healthy children” (APPROACH) randomized controlled trial. Am J Clin Nutr 115:970–979
Mogensen CS, Zingenberg H, Svare J, Astrup A, Magkos F, Geiker NRW (2023) Gestational weight gain in women with pre-pregnancy overweight or obesity and anthropometry of infants at birth. Front Pediatr. https://doi.org/10.3389/FPED.2023.1142920
Olsen J, Melbye M, Olsen SF, et al (2001) The Danish National Birth Cohort - its background, structure and aim. Scand J Public Health 29:300–307
Hernán MA (2021) Methods of Public Health Research — Strengthening Causal Inference from Observational Data. New England Journal of Medicine 385:1345–1348
Mikkelsen TB, Osler M, Olsen SF (2006) Validity of protein, retinol, folic acid and n–3 fatty acid intakes estimated from the food-frequency questionnaire used in the Danish National Birth Cohort. Public Health Nutr 9:771–778
Overvad K, Jønneland AT, Haraldsdóttir J, Ewertz M, Jensen OM (1991) Development of a semiquantitative food frequency questionnaire to assess food, energy and nutrient intake in Denmark. Int J Epidemiol 20:900–905
Kaye Foster-Powell, Susanna HA Holt and JCB-M (2002) International table of glycemic index and glycemic load values: 2002. Am J Clin Nutr 76:5–56
Atkinson FS, Brand-Miller JC, Foster-Powell K, Buyken AE, Goletzke J (2021) International tables of glycemic index and glycemic load values 2021: a systematic review. Am J Clin Nutr 114:1625–1632
Blomhoff R, Andersen R, Arnesen EK, et al (2023) Nordic Nutrition Recommendations 2023. Nordic Nutrition Recommendations 2023. https://doi.org/10.6027/NORD2023-003
World Health Organization (2009) WHO AnthroPlus for personal computers Manual: Software for assessing growth of the world’s children and adolescents.
World Health Organization (2011) WHO Anthro for Personal Computers Manual.
World Health Organization (2006) Length/height-for-age, weight-for-age, weight-for-length, weight-for-height and body mass index-for-age Methods and development.
Dodd JM, McPhee AJ, Deussen AR, Louise J, Yelland LN, Owens JA, Robinson JS (2018) Effects of an antenatal dietary intervention in overweight and obese women on 6 month infant outcomes: follow-up from the LIMIT randomised trial. Int J Obes 42:1326–1335
Metzger BE, Lowe LP, Dyer AR, et al (2008) Hyperglycemia and Adverse Pregnancy Outcomes. N Engl J Med 358:1991–2002
Liu L, Liu Z, Duan B, Zhang Q, Zhou Z, Liu W (2023) Effects of a low glycemic index or low glycemic load diet on pregnant women at high risk of gestational diabetes: A meta-analysis of randomized controlled trials. Nutrition, Metabolism & Cardiovascular Diseases 33:2006–2018
Galgani J, Aguirre C, Díaz E (2006) Acute effect of meal glycemic index and glycemic load on blood glucose and insulin responses in humans. Nutr J. https://doi.org/10.1186/1475-2891-5-22
Hu Y, Block G, Norkus EP, Morrow JD, Dietrich M, Hudes M (2006) Relations of glycemic index and glycemic load with plasma oxidative stress markers. Am J Clin Nutr 6:70–84
Scholl TO, Chen X, Khoo CS, Lenders C (2004) The Dietary Glycemic Index during Pregnancy: Influence on Infant Birth Weight, Fetal Growth, and Biomarkers of Carbohydrate Metabolism. Am J Epidemiol 159:467–474
Louie JCY, Brand-Miller JC, Markovic TP, Ross GP, Moses RG (2010) Glycemic index and pregnancy: A systematic literature review. J Nutr Metab. https://doi.org/10.1155/2010/282464
Maslova E, Hansen S, Grunnet LG, et al (2019) Maternal glycemic index and glycemic load in pregnancy and offspring metabolic health in childhood and adolescence—a cohort study of 68,471 mother–offspring dyads from the Danish National Birth Cohort. Eur J Clin Nutr 73:1049–1062
PEDERSEN J, BOJSEN-MØLLER B, POULSEN H (1954) Blood sugar in newborn infants of diabetic mothers. Acta Endocrinol (Copenh) 15:33–52
Catalano PM, Hauguel-De Mouzon S (2011) Is it time to revisit the Pedersen hypothesis in the face of the obesity epidemic? Am J Obstet Gynecol 204:479–487
Herring CM, Bazer FW, Johnson GA, Wu G (2018) Impacts of maternal dietary protein intake on fetal survival, growth, and development. Exp Biol Med 243:525–533
Freire T, Clark X, Pulpitel T, Bell-Anderson K, Ribeiro R, Raubenheimer D, Crean AJ, Simpson SJ, Solon-Biet SM (2024) Maternal macronutrient intake effects on offspring macronutrient targets and metabolism. Obesity. https://doi.org/10.1002/oby.23995
Tang M (2018) Protein Intake during the First Two Years of Life and Its Association with Growth and Risk of Overweight. Int J Environ Res Public Health. https://doi.org/10.3390/ijerph15081742
Stokes A, Campbell KJ, Yu H-J, Szymlek-Gay EA, Abbott G, He Q-Q, Zheng M (2021) Protein Intake from Birth to 2 Years and Obesity Outcomes in Later Childhood and Adolescence: A Systematic Review of Prospective Cohort Studies. Nutr 12:1863–1876
Micke GC, Sullivan TM, McMillen IC, Gentili S, Perry VEA (2011) Protein intake during gestation affects postnatal bovine skeletal muscle growth and relative expression of IGF1, IGF1R, IGF2 and IGF2R. Mol Cell Endocrinol 332:234–241
Moran LJ, McNaughton SA, Sui Z, Cramp C, Deussen AR, Grivell RM, Dodd JM (2018) The characterisation of overweight and obese women who are under reporting energy intake during pregnancy. BMC Pregnancy Childbirth. https://doi.org/10.1186/s12884-018-1826-x

There is NO conflict of interest to disclose.

Download PDF

Review #1 received at journal
28 Oct, 2024
Reviewer #1 agreed at journal
16 Oct, 2024
Reviewers invited by journal
23 Aug, 2024
Editor assigned by journal
21 Aug, 2024
Submission checks completed at journal
21 Aug, 2024
First submitted to journal
20 Aug, 2024
Unknown event
20 Aug, 2024

You are reading this latest preprint version

Association of a high-protein and low-glycemic-index diet during pregnancy with offspring growth and obesity until the age of 18 years – a target trial emulation

Status:

Version 1

Abstract

Background

Methods

Results

Conclusion

Figures

Introduction

Methods

Study design and setting

Participants

Exposure assessment

Exposure groups

Outcome variables

Covariate assessment

Statistics

Results

Study Participants

Offspring outcomes

Discussion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1