Study design
All patients provided written informed consent before participating in this randomized, 2-group parallel, controlled trial, which was approved by the Research Ethics Committee of the Research Institute for Endocrine Sciences of Shahid Beheshti University of Medical Sciences (IR.SBMU.ries.Rec.1394.92).
The study took place in Tehran from December 2016 to August 2018. We enrolled pregnant women from 8 health centers and randomly divided to intervention group with low TFAs daily intake during pregnancy and control group without any change in TFAs daily intake.
Recruitment
In the run-in phase of the study midwifes present at health centers 5 days in a week. The investigators informed the participant of all aspects pertaining to participation in the study and was screened all the patients for the eligibility criteria at the time of admission. Women’s` screening continued until the target population is achieved.
Study population
We assessed the eligibility of pregnant women attend to public health centers. Pregnant women who had a gestational age less than 8 weeks, singleton pregnancy and intention to receive prenatal care, 24 > BMI > 18 pre pregnancy, 18 ≤ age ≤ 35years, gravid under 4, abortion under or equal 2 were included. We excluded participants who had any previous history of chronic disease such as diabetes mellitus, any history of preeclampsia or gestational diabetes, any history of pre pregnancy hypertension and those who smoked or drank alcohol. The research coordinator described the study for pregnant women. Pregnant women were given suitable time to reflect on the information and their questions were answered and gave free and voluntary consent. A total of 1016 pregnant women were screened for eligibility. Among them 45 pregnant women not meeting inclusion criteria, 31 declined to participate and 28 participants for other reasons excluded from this trial. After these exclusions, 912 pregnant women were randomized in two group (intervention group = 455, control group = 457). Finally 800 pregnant women remained for the analysis (Fig. 1).
Randomization
This study was a randomized parallel dietary interventional study with a 1:1 allocation. Pregnant women were randomly allocated to one of two either group (low TFAs group and control group) in a 2 × 4 Latin Square design. Randomization was performed by person from health centers not involved in the study. The randomization table was kept in a locked file cabinet and in a password-protected computer in the midwifery group center.
Blinding
The outcomes assessor performing and analyzing data was blinded to the participant grouping.
Obstetric management
Women in both groups received routine prenatal care. Anthropometric measurements were performed with light clothing and without shoes. Weight and height were measured at the first visit in pregnancy (less than 8 weeks) according to the standard protocol. Pre pregnancy body mass index (BMI) was calculated by dividing the weight in kilograms by the square of height in meters on the basis of weights in pre pregnancy. Gestational age was ascertained by first day of last menstrual period and by ultrasound under 12 weeks of gestational age. Pregnant women had monthly obstetric visits up to 28 weeks of gestational age, biweekly visits from week 28 to week 36, and weekly visits until the end of the pregnancy. For both groups, prenatal data were obtained from the patients’ medical records.
Data collection and assessment
Data was collected by face to face interview and using the structured questionnaire. The questionnaire had two parts. The first part of it was included to socio demographic background: maternal age, maternal education, and maternal occupation. The information of this part was important, because it showed the socioeconomic condition of participants that these could effect on the child birth weight. The second part of questionnaire obtained maternal condition such as gestational age, number of pregnancy, prenatal care follow up, gestational age at the first care, and pre pregnancy body mass index. The information about dietary intake was collected by dietary food recall in three days (one holiday and two week days).
All interviews and data collections were conducted by trained nutritionists based in health care centers. They were given 3 days training on interview technique and nutrition data collection. 5 nutritionists and 1 midwife supervised study.
Study intervention
Participants assigned to one of the two groups, the intervention (n = 393) and the control (n = 407) group (Fig. 1). For all pregnant women individual dietary pattern were designed based on age, height, pre-pregnancy weight and physical activity. Indeed, Women in intervention group received individualized dietary pattern with TFA content under 1% of total daily energy intake, and in the control group, women received individualized dietary pattern without any focus on TFA content. In the intervention group, we replaced mono- and polyunsaturated fatty acids free of trans fatty acids in dietary pattern, such as olive oil, fish, nuts, and low fat (1.5%) dairy products, and participants were forbidden to consume the food rich of trans fatty acids such as any kind of fast foods, processed meat, and deep fried foods; but in the control group, women were allowed to consume any kind of dairy products and oil for cooking and routine dietary recommendation were advised. For dietary assessment and ensure compliance, we used 24_hours dietary recalls (24h_DRs). For women dietary assessment, expert nutritionists assessed dietary intake of the participants using three non - consecutive 24-hour recalls, one weekend day (Thursday or Friday) and 2weekdays. The personal interview has held in private and took about 20 minute. Portion sizes of meals are converted to grams by using household measures (19). Nutrient intakes are calculated according to the US Department of Agriculture and Iranian Food Composition Tables (20, 21). First recall interviews were performed at health center; second and third recalls were filled by call. The 24h_DR data were also collected at first prenatal care visit and the end of every trimester (13, 25, 35 weeks).All recalls were checked by investigators, and ambiguities were resolved with the women. Because the only available Iranian food composition table (FCT) analyzes a very limited number of raw food items and nutrients, we used the united states drug and food composition table (USDA FCT) as the main FCT; the Iranian FCT was used as an alternative for traditional Iranian food items, like kashk, which are not included in the USDA FCT. All pregnant women who participated in the study routinely received 1 mg folic acid daily in the first trimester (up12 week) of and received multivitamin minerals daily from the beginning of the second trimester to the end of pregnancy. From the beginning of the 16th week of the pregnancy, all participants received the 30 mg daily iron supplement.
Unannounced telephone calls were used both to encourage participants and to ascertain adherence to the dietary.
Primary research question
What is the effect of low trans fatty acids dietary pattern on the risk of LBW?
The primary endpoint was the rate of low birth weight, defined as those weighted were less than 2500 gram.
Outcome measurement
The weight of newborns was measured 60 minute after birth with a Seca scale. The scale was always checked and zeroed before weighting each child. Low birth weight newborns were those weighted under 2500 gram. All deliveries were conducted in a central hospital.
Obstetric management
Women in both groups received routine prenatal care. The participant’s BMI was calculated on the basis of pre pregnancy weights (maximum three months before pregnancy) and heights at the first prenatal care visit (under 8 weeks of gestation). Gestational age was calculated by first day of last menstrual period or by ultrasound less than 12 weeks of gestational age. They had monthly obstetric visits up to 28 week of gestational age, every two weeks visits from week 28 to 36week, and weekly visits until the end of the pregnancy. For both groups, antenatal and perinatal data were obtained from the patients’ medical records.
Statistical analysis
We calculated sample size based on α = 0.05, 1-β = 0.80, and using the following formula:
Accordingly, sample size in each group was 325 people, with the probability of miss to follow-up of 20%, yielded to 400 persons per group.
Quantitative data were expressed as mean ± standard deviation (SD), where applicable and quantitative data were presented as percentages. Independent sample T-tests and Chi-2 were used to compare variables among two groups, respectively. All conflicted variables were matched in both groups. Each variable’s normality was checked by a histogram chart and one-sample Kolmogorov-Smirnov test. Repeated measure was used for analysis of nutritional and obstetrical changes during follow-up. When performing the Cox proportional-hazards model, groups were considered as independent variables and LBW as a dependent variable. Cox regression analysis is basically done when participants are followed up over time. In this case the exposure (independent variable) is the intervention of low TFA diet and the outcome is the birth weight that followed up during pregnancy. Incident rates and hazard ratios (95% confidence interval) for LBW based on groups were calculated (Table 4). The potential confounders for each group were age, pre pregnancy BMI, gravid in the adjusted model. In this study we used the Kaplan–Meier plot, is a non-parametric statistic used to estimate the survival function from lifetime data. In medical research, it is often used to measure the fraction of patients living for a certain amount of time after treatment. In this study, the percentage difference in weight of newborns in two groups (intervention and control group) after the pregnant women intake diet was evaluated. All statistical tests were two sided, and a significant association was declared at P-value < 0.05(95% confidence interval). All statistical analyses were performed using statistical package for social science (version 21; SPSS Inc., Chicago, IL, USA).
Table 4
Hazard ratio (95% CI) for developing LBW based on dietary pattern ≤ 1% TFAs intake
Models
|
Hazard ratios
|
CI (95%)
|
p-value
|
Unadjusted Model
Multi variables Model*
|
0.65
0.64
|
0.44–0.96
0.44–0.95
|
0.03
0.02
|
*: groups, age, pre pregnancy BMI, gravid were adjusted |