Pregnancy presents a unique époque in life with considerable potential to influence not only maternal health but also the health of the next generation. Nutritional interventions in pregnancy can influence maternal, foetal and infant health. This chapter describes the findings of a large randomised controlled trial on the impact of a low glycaemic index diet on maternal and foetal health and discusses areas of future research in this important area.
Introduction
The large-for-gestational-age or macrosomic infant is predisposed to a variety of adverse obstetric and neonatal outcomes, and delivery of a large infant significantly increases the risk of birth complications for the mother . In the long term, infants who are at the highest end of the distribution for weight or body mass index (BMI) are more likely to be obese in childhood, adolescence and early adulthood than other infants , and are at risk of cardiovascular and metabolic complications later in life .
Both maternal weight and gestational weight gain (GWG) exert a significant influence on birthweight . Increased maternal BMI confers an elevated risk of delivering a heavier infant , and there is evidence to support strong associations between excessive weight gain and an increased birthweight .
Glucose is the main energy substrate for foetal growth . Birthweights in infants of diabetic mothers are increased, with up to 35% above the 95th percentile . Moreover, it is now apparent that there is a strong, continuous association between maternal glucose levels below those diagnostic of diabetes and the incidence of macrosomia and its inherent complications . Maternal diet, and particularly its carbohydrate (CHO) type and content, influences maternal blood glucose concentrations. However, different CHO foods produce different glycaemic responses. The glycaemic index (GI) was conceived by Jenkins in 1981 as a method for assessing the glycaemic responses of different CHOs . It has been shown that a low glycaemic diet blunts the mid and late pregnancy increase in insulin resistance typically seen in westernised societies . It is postulated that eating primarily high glycaemic CHO foods results in foeto-placental overgrowth and excessive maternal weight gain and leads to a predisposition to foetal macrosomia, while intake of low glycaemic CHOs predisposes to normal infant birthweight and normal maternal weight gain . Prior to the ROLO study, s randomised control trial of low glycaemic index diet in pregnancy to prevent macrosomia in 2012, the effect of introducing a low GI (LGI) diet in pregnancy to reduce the incidence of foetal macrosomia has not been subject to a large randomised control trial.
It is known that foetal macrosomia recurs in a second pregnancy 30–50% of the time and GWG influences this risk . Our objective was to conduct a randomised control trial of an LGI diet from early pregnancy in a group of women, all in their second pregnancies (secundigravid), having previously delivered an infant weighing >4000 g (ROLO study). Our hypothesis was that an LGI diet would reduce the recurrence of foetal macrosomia. The primary outcome measure was difference in birthweight and the secondary outcome was difference in GWG between the two groups .
A randomised control trial of LGI diet in pregnancy to prevent macrosomia (ROLO study)
This is a randomised control trial with institutional ethical approval and maternal written consent.
All secundigravid women who previously delivered a macrosomic infant weighing >4 kg were identified upon first contact with the hospital and recruited at first antenatal consultation. At this visit, women with any underlying medical disorders, including a previous history of gestational diabetes, those on any medications and those unable to give full informed consent, were excluded.
Following recruitment and written informed consent, patients were randomised into either the control or the intervention arm of the study. Randomisation was achieved using computer-generated allocations in a ratio of 1:1 contained in sealed opaque envelopes by the research midwife.
Dietary intervention
The control arm received routine antenatal care. As is standard practice at our institution, this does not involve any formal dietary advice or specific advice about GWG. Women randomised to the intervention group attended one dietary education session lasting 2 h in groups of 2–6 women with the research dietician. The mean gestational age of those attending the dietary session was 15.7 ± 3.0 weeks. Women were first advised on general healthy eating guidelines for pregnancy . The remainder of the education session focussed on the GI: its definition, concept and rationale for use in pregnancy. Women were encouraged to choose as many LGI foods as possible and to exchange high GI (HGI) CHOs for LGI alternatives. Women received written resources about LGI foods after the education session. The recommended LGI diet was eucaloric and advice was not given about recommendations for GWG. The research dietician met with the patients at 28 and 34 weeks of gestation for reinforcement of the LGI diet and for addressing any dietary queries the patients had.
At first antenatal consultation, all patients had their weight and height recorded and BMI calculated. Fasting blood glucose was measured and a mid-upper arm circumference recorded. Additional demographic data including smoking history and socioeconomic data were also recorded.
Maternal weight was recorded at each antenatal consultation.
At 28 weeks of gestation, repeat fasting blood glucose was measured and glucose challenge testing (GCT) 1-h post 50-g glucose load was performed. As per institutional policy, those with a GCT of 8.3 mmol/L or greater had formal glucose tolerance testing carried out to outrule gestational diabetes. Gestational diabetes was diagnosed when an abnormal GCT was followed by two or more abnormal values on a 3-h 100-g glucose tolerance test using the Carpenter and Coustan criteria . Where gestational diabetes was diagnosed, care continued in the multidisciplinary diabetic clinic.
Foetal biometry was assessed ultrasonographically at 34 weeks of gestation. This included a measurement of foetal anterior abdominal wall (AAW) width. This measurement has been used in both diabetic and non-diabetic populations as a marker of foetal adiposity . Foetal ultrasound was performed using a Voluson 730 Expert (GE Medical Systems, Germany) by one of the two blinded sonographers.
At delivery, infant birthweight, length and head circumference were recorded and infant ponderal index (100 mass in grams/height in cubic centimetres) calculated. Birthweight centiles corrected for maternal weight, height, parity, ethnicity, gestational age at delivery and infant gender were calculated using Gestation Network’s Bulk Calculator version 6.2.3 UK .
Dietary assessment
All women completed three 3-day food diaries, one prior to dietary intervention and one each in the second and third trimester of pregnancy. They collected information on typical meal pattern and food choices over 3 days and allowed for estimation of the GI. To assess compliance to the LGI diet, a questionnaire was given to patients in the intervention group at their 34-week antenatal visit. This was based on a 5-point Likert scale (1 being ‘I followed the recommended diet all of the time’ and 5 being ‘I followed the recommended diet none of the time’).
Sample size calculation based on a significance level set of 5% and power set at 90% calculated that 360 in each group would be required to detect a 0.25 SD difference in birthweight between the two groups, equivalent to a 102-g difference in the birthweight.
Results
Eight hundred were recruited and randomised. There was one stillbirth in the intervention arm of an infant at 39 weeks of gestation, weighing 2.9 kg; post-mortem confirmed Trisomy 21. The baseline characteristics are contained in Table 1 with no differences noted between the two groups. There was no difference in the dietary GI prior to intervention between intervention group and controls (57.3 ± 4.2 vs. 57.7 ± 4.0). Following the introduction of the LGI diet, the intervention group had a lower GI in second (56.1 ± 4.0 vs. 57.8 ± 3.7, p < 0.001) and third trimesters (56.0 ± 3.8 vs. 57.7 ± 3.9, p < 0.001). Almost 80% of the intervention arm reported following the LGI dietary advice either all or most of the time on the compliance questionnaire.
| Baseline characteristic | Intervention group | Control group |
|---|---|---|
| Age (years) | 32.0 ± 4.2 | 32.0 ± 4.2 |
| Weight (kg) | 73.8 ± 14.8 | 73.4 ± 13.7 |
| Height (cm) | 166.3 ± 6.4 | 168.9 ± 6.5 |
| BMI (kg/m 2 ) | 26.8 ± 5.1 | 26.8 ± 4.8 |
| Mid-upper arm circumference | 29.5 ± 3.6 | 29.5 ± 3.4 |
| Fasting glucose (mmol/L) | 4.5 ± 0.36 | 4.5 ± 0.38 |
| Previous birthweight (grams) | 4253 ± 261 | 4242 ± 236 |
| Smokers ( n =) | 17 | 12 |
Foetal macrosomia (birthweight >4000 g) recurred in 50.7% and 51.5% of the intervention and control groups respectively, p > 0.05. There was no difference between the two groups in mean unadjusted birthweight at delivery, mean birthweight centile or birthweight adjusted for maternal BMI, gestation at delivery and infant gender by multiple logistic regression analysis. Similarly, there was no difference between the two groups in infant ponderal index at birth (2.76 ± 3.8 vs. 2.75 ± 0.33, p > 0.05) ( Table 2 ).
| Intervention group | Control group | |
|---|---|---|
| Birthweight (grams) | 4034 ± 510 | 4006 ± 497 |
| Birthweight centile | 70.5 ± 25.6 | 72.8 ± 25.6 |
| Length at birth (cm) | 52.9 ± 2.7 | 52.6 ± 2.1 |
| Head circumference at birth (cm) | 35.8 ± 1.3 | 35.7 ± 1.5 |
| Birthweight difference from first pregnancy (g) | −214.2 ± 541 | −250.8 ± 512 |
| Estimated foetal weight at 34 weeks (g) | 2631 ± 326 | 2616 ± 368 |
| Foetal abdominal circumference at 34 weeks (mm) | 315.8 ± 16.7 | 315.6 ± 19.2 |
| Foetal AAW at 34 weeks (mm) | 5.0 ± 1.3 | 5.1 ± 1.2 |
| Fasting glucose at 28 weeks (mmol/L) | 4.45 ± 0.4 | 4.51 ± 0.6 |
| GCT at 28 weeks (mmol/L) | 6.47 ± 1.4 | 6.67 ± 1.7 |
| Cord blood glucose (mmol/L) | 4.17 ± 1.1 | 4.16 ± 1.2 |
| Weight gain at 24 weeks (kg) | 5.3 ± 2.7 | 5.5 ± 2.7 |
| Weight gain at 28 weeks (kg) | 7.1 ± 2.8 | 7.7 ± 3.0 a |
| Weight gain at 34 weeks (kg) | 10.1 ± 3.7 | 10.9 ± 3.9 a |
| Weight gain at 40 weeks (kg) | 12.2 ± 4.4 | 13.7 ± 4.9 a |
| Caesarean delivery rate (%) | 17.8 | 22 |
| Mean gestational age at delivery (days) | 282.5 ± 9.2 | 280.8 ± 10.3 a |
| Delivery <37 weeks | 3 | 8 |
| Delivery >41 weeks | 107 | 92 |
| Foetal loss ( n =) | 10 | 7 |
| Early pregnancy loss | 9 | 6 |
| Mid-trimester loss | 0 | 1 |
| Stillbirth | 1 | 0 |
a Two-tailed p -value of <0.05, considered statistically significant.
There was a significant difference in the secondary outcome between those in the intervention and those in the control arm of the study. Women who received dietary intervention had significantly less GWG when compared to those who received no dietary intervention. At 40 weeks of gestation, the mean GWG in the intervention arm was 12.2 kg compared to 13.7 kg in the controls ( p = 0.01) ( Table 2 ).
Women in the intervention arm of the study were significantly less likely to exceed GWG recommendations as outlined by the Institute of Medicine (IOM), 37.7% versus 47.9%, p = 0.01.
Overall, a significantly higher proportion of women in the control group had a GCT result of >7.8mmol/L at 28 weeks of gestation. Similarly, a higher proportion of women in the control arm had either a fasting glucose at 28 weeks of ≥5.1mmol/L or a GCT of >7.8 mmol/L than in the intervention. An equal number of women in each arm underwent formal glucose tolerance testing and there was no difference in the incidence of gestational diabetes according to either Carpenter and Coustan or the American Diabetes Association criteria . Cord blood glucose was similar between the two groups ( Table 3 ).
| Intervention group | Control group | |
|---|---|---|
| 28-week fasting glucose ≥5.1mmol/L | 7.5% (24/321) | 11.6% (41/352) |
| GCT >7.8 | 15.4% (54/350) | 21.3% 79/371 a |
| 28-week fasting glucose >5.1mmol/L or GCT>7.8mmol/L | 20.9% 67/320 | 28.4% 100/352 a |
| GCT >8.3mmol/L | 12% 42/350 | 14% 52/371 |
| Gestational diabetes Carpenter and Coustan criteria | 2% 7/350 | 2.4% 9/371 |
| Gestational diabetes American Diabetes Association criteria | 3.4% 12/350 | 4.9% 18/371 |
a Two-tailed p -value of <0.05, considered statistically significant.
There was no difference in the preterm delivery rate or caesarean delivery rate between the two groups.
Discussion
Our results have found that LGI diet in pregnancy has no effect on infant birthweight in a group at risk of foetal macrosomia. It does, however, have a significant positive effect on GWG and on maternal glucose intolerance.
In recent years, there has been increasing interest in the role of the GI in pregnancy, and, in particular, its potential to modulate foetal growth . Until now, the small number of intervention trials that examined the use of an LGI diet in pregnancy suggested that it might be useful in preventing foetal macrosomia. The earliest publication by Clapp et al. in 1997 included just 12 women , but found that those on an LGI diet during pregnancy not only had lower GWGs but also gave birth to infants with lower birthweights than those on a HGI diet. Moses et al. in 2006 reported a study of 70 women who were assigned alternately to receive dietary counselling that encouraged LGI CHO foods or high-fibre, moderate-to-high GI foods. The authors found a significantly higher prevalence of large-for-gestational-age infants in those on the HGI diet of 33% compared to just 3% in those on the LGI diet.
A recent systematic review of interventions in pregnancy showed that all interventions had some impact on GWG . Overall, there was a 1.4-kg reduction in GWG with any intervention compared to control. Dietary interventions resulted in the largest reduction in GWG with no impact on birthweight noted. Physical activity interventions were associated with a reduced birthweight of 60 g. Both dietary and physical activity interventions resulted in reduced risk of pre-eclampsia (0.74, 0.6–0.9) and shoulder dystocia (0.39, 0.22–0.7). The LIMIT study , published recently, examined the impact of a healthy lifestyle intervention (healthy diet and exercise intervention) in overweight and obese pregnancy. No differences were noted in the primary outcome of large-for-gestational-age infants (lifestyle advice 19% vs. standard care 21%). However, infants born to women following lifestyle advice were less likely to have birthweights over 4 kg (15% vs. 19%, p = 0.04). No differences were noted in rates of gestational diabetes (14% vs. 11%) or in GWG.
Clinical implications
The ROLO study is a sufficiently powered, randomised controlled study and we found no difference in absolute birthweight, birthweight centile or the incidence of macrosomia.
We did find that an LGI diet in pregnancy significantly reduced GWG, our secondary outcome. By 40 weeks of gestation, women in the diet arm had gained 1.5 kg less than those women who received no dietary intervention. Maternal weight gain during pregnancy has been independently linked to adverse obstetric outcomes . There are also potential maternal implications to excessive GWG, such as an increased operative delivery rate , a higher likelihood of post-partum weight retention and a predisposition to later obesity.
Additionally, we found that an LGI diet in pregnancy reduces the incidence of maternal glucose intolerance at 28 weeks of gestation. A significantly higher proportion of women in the control arm at 28 weeks had either a fasting glucose concentration of >5.1mmol/L or a GCT result of >7.8mmol/L. Despite this reduction in glucose intolerance, there was no difference between the two groups in the incidence of gestational diabetes. Since the publication of the hyperglycaemia and adverse pregnancy outcome (HAPO) study in 2008 , however, it is widely accepted that there is a clear association between glucose concentrations below those diagnostic of gestational diabetes and a number of adverse pregnancy outcomes.
While it appears that an LGI diet in pregnancy alone may not be sufficient to reduce macrosomia, it does offer significant maternal benefits. The use of an LGI diet in pregnancy is a simple, safe and effective measure to improve maternal glucose homeostasis and to reduce GWG .
Impact of LGI dietary advice on maternal dietary intake in the ROLO study
Three-day food diaries were collected in each trimester in the ROLO study to assess the impact of the LGI dietary advice on dietary intakes.
At baseline, there were no significant differences in energy or nutrient intake. Maternal GI was significantly reduced in the intervention group at trimesters 2 and 3. The numbers of women within the lowest quartile of GI increased from 37% in trimester 1 to 52% in trimester 3 ( p < 0.001) among the intervention group ( Table 4 ). Post dietary intervention, the intervention group reported significantly lower energy intake at both trimesters 2 and 3 (7.6 vs. 8.1 MJ, p < 0.05 ). They also had significantly higher intakes of dietary fibre (20.3 vs. 18.8 g, p < 0.01 ), vitamin A (949.6 vs. 896.8 μg, p < 0.05 ) and magnesium (265.8 vs. 249.3 mg, p < 0.001 ) at trimester 3 and higher protein intakes as percentage total energy intake ( p < 0.01).
| Trimester 1 | Trimester 2 | Trimester 3 | P c | ||||
|---|---|---|---|---|---|---|---|
| I | C | I | C | I | C | ||
| GI | 57.3 ± 4 | 57.7 ± 4 | 56.1 ± 4 | 57.8 ± 4 | 56 ± 3.7 | 57.7 ± 3.9 | <0.001 |
| p 0.25 b | p < 0.001 b | p < 0.001 b | |||||
| N (%) | N (%) | N (%) | |||||
| Quartile GI | |||||||
| 1 a | 37.4 | 32.2 | 50.2 | 30.5 | 51.5 | 31.2 | |
| 2 | 16.2 | 20.0 | 13.6 | 19.3 | 16.2 | 19.3 | |
| 3 | 23.4 | 19.3 | 18.3 | 22.1 | 17.9 | 23.9 | |
| 4 | 23.0 | 28.4 | 17.9 | 28.1 | 14.5 | 25.6 | |
| p 0.21 d | p < 0.001 d | p < 0.001 d | |||||
a Quartile 1 (GI values 49.7–55.9); quartile 2 (GI values 56.0–57.6); quartile 3 (GI values 57.7–59.9); quartile 4 (GI values 60.0–69.7).
b Differences in mean GI between groups across trimesters were compared using independent sample t-test.
c Differences in mean GI between trimesters 1 and 2 within the intervention group assessed by paired t -test.
d Differences in numbers of women within each quartile of GI between groups were compared using the χ 2 test.
Discussion
Mean maternal GI at baseline was 57 in both groups in the ROLO study. This is similar to the baseline GI in previous studies, including the European Prospective Investigation into Cancer and nutrition (EPIC) study carried out in five European countries . Moses et al. reported a baseline GI of 58 in their sample of 72 pregnant women . The mean GI of the American population is reported to be 57 . After the LGI dietary intervention, mean maternal dietary GI was significantly lower among the intervention group; however, the actual difference was modest (57.3 vs. 56.1). Previous studies have reported larger decreases in maternal GI post intervention; however, these interventions employed more intensive dietary education sessions, often at numerous time points during pregnancy . Reported energy intakes were lower in our intervention group and this has been found in other dietary intervention studies . Energy intake was approximately 100 kcal lower among the LGI group in the trial by Moses and colleagues. Overall, their energy intakes ranged from 1800 to 1900 kcal, which is similar to the energy intakes reported in our study a priori exclusion of under reporters.
Therefore, an LGI dietary intervention in early pregnancy significantly reduced maternal GI, increased dietary fibre intake, increased intake of wholegrain breads and cereals and reduced consumption of high-energy beverages, white breads and refined cereals in a group of women who previously delivered a macrosomic infant. Overall, beneficial effects were noted on maternal dietary intakes .
Impact of an LGI diet in pregnancy on markers of maternal and foetal metabolism and inflammation
Following the introduction of an LGI diet, we aimed to assess the effect of an LGI diet in pregnancy on maternal and foetal insulin resistance, leptin and markers of inflammation.
Therefore, a secondary analysis of 621 women was performed in ROLO study, a randomised control trial of LGI diet in pregnancy to prevent the recurrence of macrosomia. In early pregnancy and again at 28 weeks, fasting serum was analysed for insulin, leptin, tumour necrosis factor- α (TNF-α) and interleukin-6 (IL-6). Maternal insulin resistance was calculated using the homeostasis model assessment (HOMA) index. At delivery, cord blood concentrations of leptin, TNF- α and IL-6 were recorded. Foetal insulin resistance was assessed with cord blood C-peptide estimation.
Results are shown in Table 5 .
| Mann–Whitney U | Standard Error | Asymptotic Sig. (two-sided test) | ||
|---|---|---|---|---|
| First trimester | 43,224 | 2038 | 0.76 | |
| Leptin | 28 weeks | 38,973 | 1904 | 0.97 |
| Cord | 27,908 | 1500 | 0.78 | |
| First trimester | 42,901 | 2100 | 0.48 | |
| HOMA | 28 weeks | 38,650 | 1913 | 0.79 |
| Cord blood C-peptide | 29,297 | 1542 | 0.96 | |
| First trimester | 42,451 | 2053 | 0.81 | |
| TNF-α | 28 weeks | 39,123 | 1954 | 0.55 |
| Cord | 15,332 | 979 | 0.43 | |
| First trimester | 21,977 | 1238 | 0.96 | |
| IL-6 | 28 weeks | 21,818 | 1249 | 0.78 |
| Cord | 5170 | 436 | 0.53 | |
| Insulin change | 32,684 | 1551 | 0.04 | |
There was no difference between the two groups in either fasting insulin or the HOMA index at 28 weeks of gestation. Similarly, no difference was noted in cord blood C-peptide between those who did and those who did not receive LGI dietary advice in early pregnancy.
Women in the intervention arm of the study did have a lower overall rise in insulin concentrations from early pregnancy to 28 weeks of gestation ( Table 5 ). When the proportion of women within the highest versus the lowest quartile for the cohort was compared, 20% of women in the intervention arm were in the highest quartile for insulin change (28-week insulin minus insulin at booking) compared to 29% of the controls ( p = 0.02).
Discussion
We have found that an LGI diet in pregnancy has little effect on markers of maternal insulin resistance or inflammation in a cohort of euglycaemic women at risk of macrosomia. We did note an attenuated response to the normal increase in insulin resistance seen in pregnancy with advancing gestation in those who received the low glycaemic dietary advice, though the overall effect was limited .
A randomised control trial of LGI diet in pregnancy to prevent macrosomia (ROLO study)
This is a randomised control trial with institutional ethical approval and maternal written consent.
All secundigravid women who previously delivered a macrosomic infant weighing >4 kg were identified upon first contact with the hospital and recruited at first antenatal consultation. At this visit, women with any underlying medical disorders, including a previous history of gestational diabetes, those on any medications and those unable to give full informed consent, were excluded.
Following recruitment and written informed consent, patients were randomised into either the control or the intervention arm of the study. Randomisation was achieved using computer-generated allocations in a ratio of 1:1 contained in sealed opaque envelopes by the research midwife.
Dietary intervention
The control arm received routine antenatal care. As is standard practice at our institution, this does not involve any formal dietary advice or specific advice about GWG. Women randomised to the intervention group attended one dietary education session lasting 2 h in groups of 2–6 women with the research dietician. The mean gestational age of those attending the dietary session was 15.7 ± 3.0 weeks. Women were first advised on general healthy eating guidelines for pregnancy . The remainder of the education session focussed on the GI: its definition, concept and rationale for use in pregnancy. Women were encouraged to choose as many LGI foods as possible and to exchange high GI (HGI) CHOs for LGI alternatives. Women received written resources about LGI foods after the education session. The recommended LGI diet was eucaloric and advice was not given about recommendations for GWG. The research dietician met with the patients at 28 and 34 weeks of gestation for reinforcement of the LGI diet and for addressing any dietary queries the patients had.
At first antenatal consultation, all patients had their weight and height recorded and BMI calculated. Fasting blood glucose was measured and a mid-upper arm circumference recorded. Additional demographic data including smoking history and socioeconomic data were also recorded.
Maternal weight was recorded at each antenatal consultation.
At 28 weeks of gestation, repeat fasting blood glucose was measured and glucose challenge testing (GCT) 1-h post 50-g glucose load was performed. As per institutional policy, those with a GCT of 8.3 mmol/L or greater had formal glucose tolerance testing carried out to outrule gestational diabetes. Gestational diabetes was diagnosed when an abnormal GCT was followed by two or more abnormal values on a 3-h 100-g glucose tolerance test using the Carpenter and Coustan criteria . Where gestational diabetes was diagnosed, care continued in the multidisciplinary diabetic clinic.
Foetal biometry was assessed ultrasonographically at 34 weeks of gestation. This included a measurement of foetal anterior abdominal wall (AAW) width. This measurement has been used in both diabetic and non-diabetic populations as a marker of foetal adiposity . Foetal ultrasound was performed using a Voluson 730 Expert (GE Medical Systems, Germany) by one of the two blinded sonographers.
At delivery, infant birthweight, length and head circumference were recorded and infant ponderal index (100 mass in grams/height in cubic centimetres) calculated. Birthweight centiles corrected for maternal weight, height, parity, ethnicity, gestational age at delivery and infant gender were calculated using Gestation Network’s Bulk Calculator version 6.2.3 UK .
Dietary assessment
All women completed three 3-day food diaries, one prior to dietary intervention and one each in the second and third trimester of pregnancy. They collected information on typical meal pattern and food choices over 3 days and allowed for estimation of the GI. To assess compliance to the LGI diet, a questionnaire was given to patients in the intervention group at their 34-week antenatal visit. This was based on a 5-point Likert scale (1 being ‘I followed the recommended diet all of the time’ and 5 being ‘I followed the recommended diet none of the time’).
Sample size calculation based on a significance level set of 5% and power set at 90% calculated that 360 in each group would be required to detect a 0.25 SD difference in birthweight between the two groups, equivalent to a 102-g difference in the birthweight.
Results
Eight hundred were recruited and randomised. There was one stillbirth in the intervention arm of an infant at 39 weeks of gestation, weighing 2.9 kg; post-mortem confirmed Trisomy 21. The baseline characteristics are contained in Table 1 with no differences noted between the two groups. There was no difference in the dietary GI prior to intervention between intervention group and controls (57.3 ± 4.2 vs. 57.7 ± 4.0). Following the introduction of the LGI diet, the intervention group had a lower GI in second (56.1 ± 4.0 vs. 57.8 ± 3.7, p < 0.001) and third trimesters (56.0 ± 3.8 vs. 57.7 ± 3.9, p < 0.001). Almost 80% of the intervention arm reported following the LGI dietary advice either all or most of the time on the compliance questionnaire.
| Baseline characteristic | Intervention group | Control group |
|---|---|---|
| Age (years) | 32.0 ± 4.2 | 32.0 ± 4.2 |
| Weight (kg) | 73.8 ± 14.8 | 73.4 ± 13.7 |
| Height (cm) | 166.3 ± 6.4 | 168.9 ± 6.5 |
| BMI (kg/m 2 ) | 26.8 ± 5.1 | 26.8 ± 4.8 |
| Mid-upper arm circumference | 29.5 ± 3.6 | 29.5 ± 3.4 |
| Fasting glucose (mmol/L) | 4.5 ± 0.36 | 4.5 ± 0.38 |
| Previous birthweight (grams) | 4253 ± 261 | 4242 ± 236 |
| Smokers ( n =) | 17 | 12 |
Foetal macrosomia (birthweight >4000 g) recurred in 50.7% and 51.5% of the intervention and control groups respectively, p > 0.05. There was no difference between the two groups in mean unadjusted birthweight at delivery, mean birthweight centile or birthweight adjusted for maternal BMI, gestation at delivery and infant gender by multiple logistic regression analysis. Similarly, there was no difference between the two groups in infant ponderal index at birth (2.76 ± 3.8 vs. 2.75 ± 0.33, p > 0.05) ( Table 2 ).
Full access? Get Clinical Tree
