Declaration of Savannah O'Neill

Case 3:18-cv-00050-JD Document 17-12 Filed 01/31/18 Page 1 of 62
LAW OFFICES OF YOLANDA HUANG

1
YOLANDA HUANG, SBN 104543
2 475 14th Street, Suite 500
Oakland, CA 94612
3 Telephone: (510) 329-2140
Facsimile: (510) 580-9410
4
5 DENNIS CUNNINGHAM, SBN 112910
115A Bartlett St.
6 San Francisco, CA 94110
7 Telephone: 415-285-8091
Facsimile: 415-285-8092
8
Attorneys for Plaintiffs
9
10 UNITED STATES DISTRICT COURT
11 FOR THE NORTHERN DISTRICT OF CALIFORNIA
12
13 JACLYN MOHRBACHER, ERIN ELLIS,
DOMINIQUE JACKSON, CHRISTINA
14 No. 3:18-cv-00050-JD
ZEPEDA, ALEXIS WAH, AND KELSEY
ERWIN, on behalf of themselves and others
15 similarly situated,
DECLARATION OF JENNY SCAFIDI IN
16 Plaintiffs, SUPPORT OF PLAINTIFFS’ MOTION FOR A
17 PRELIMINARY INJUNCTION
vs.
18
19 ALAMEDA COUNTY SHERIFF’S OFFICE, et
al.,
20 Defendants.
21
I , JENNY SCAFIDI, declare:
22 1. I make this declaration based upon my own true knowledge and if called to testify, I can and
23 will testify as stated herein.
24 2. I am a UCSF trained nurse midwife and women’s health nurse practitioner. I was awarded a
25 Masters of Science from the University of California, San Francisco, CA 2014; and a
26 Bachelors in Community Health Studies from the University of California, Santa Cruz, in
2009 and a certified midwife from Maternidad La Luz, El Paso, Tx in 2006. I have more
27
than ten years experience working in women’s reproductive health, providing prenatal,
28
intrapartum and primary care. I am currently employed as a Nurse Midwife at Sutter Davis
Page 1 of 9
Mohrbacher v. Alameda County Sheriff’s Office 3:18-cv-00050, Dec. J. Scafidi in Support; Motion for Preliminary Injunction
Hospital in Davis, California, providing prenatal care and managing births; and as a primary
1
medical provider to women of reproductive age at Planned Parenthood Vallejo.
2
1. A primary necessity for a healthy pregnancy is a sufficient caloric intake. The American
3
Dietetic Association recommends an optimal caloric intake range of 2500 to 2700 kcal per
4
day. Attached as Exhibit A is a true and correct copy of this report. Poor maternal
5 malnutrition has been related to adverse pregnancy outcomes. “Early fetal nutrition is
6 considered the most important factor in child health, even before birth.” Nutrients 2017, 9,
7 627; doi:10.3390/nu9060627
8 2. Deficiencies in nutrition has drastic negative impacts on the fetus and predisposes the infant
9 to chronic conditions in later life. The lack of key nutrients during crucial periods of fetal
10 development may lead to reprogramming within fetal tissues, and could potentially lead to
obesity, cardiovascular disease, issues with bone health, cognition issues, immunity function
11
issues and potentially diabetes. Attached as Exhibit B is a true and correct copy of an
12
Academy of Nutrition and Dietetics study on this subject with these conclusions.
13
3. During the second trimester, the issue is not just total calories, but the type of calories.
14
Having a diet reliant on carbohydrates with a low portion of protein, particularly animal
15 protein can still contribute to low birth weight. Attached as Exhibit C is a true and correct
16 copy of this article from the Journal of Nutrients.
17 4. I have reviewed the some of the declarations which describe the meals provided. I have not
18 had an opportunity to examine first hand. Based upon the declarations, I have compiled what
19 I understand to be an average day’s meals. Breakfast consisting of oatmeal, 8 ounces of milk
20 and a few baby carrots. Lunch and snack both consisting of peanut butter jelly sandwiches,
and baby carrots, and 8 ounces of milk. Dinner being a meat patty, some potatoes, 8 ounces
21
of milk and 2 1 inch cookies. Portion sizes are guestimates, based upon standard serving
22
sizes. Further, it is unknown whether the milk is low fat, 2% or whole milk. Using standard
23
sizes, and 2% milk I calculate, a daily calorie count would of around 2,200 calories, which is
24
on the low end, and somewhat inadequate for a singleton pregnancy, given a recommended
25 daily calorie intake of 2500 to 2700. This rather unvaried diet is also deficient in fruits and
26 vegetables, particularly dark leafy greens. The daily recommendation is 4 vegetables,
27 including dark greens, and 3 fruits per day, which the current diet does not meet. And this
28 diet is definitely not sufficient for twin pregnancies. Further, anyone allergic to peanuts would
Page 2 of 9
need substitutions, for the peanut butter sandwiches, which I understand substitutions were
1
not provided. Lastly, I am assuming some type of vitamin supplement is being provided. Not
2
having received confirmation of supplements, nor the composition of any such supplements, I
3
am concerned regarding whether this diet meets the needs of pregnant women for folic acid,
4
or the other necessary B vitamins. A quick review indicates that this diet is probably low in
5 trace minerals such as zinc and potassium and could be too high in poly-unsaturated fats given
6 the high concentration of peanut butter.
7 Breakfast: 8 ounces of cooked oatmeal (166 cal), 8 ounces of milk (2% = 122 cal) = 288
8 Lunch: peanut butter (3 T. 282 cal) & jelly(2 Tab 100 cal) , two pieces of bread, (wheat, per slice
9 78 cal) 4-5 baby carrots (14) , 8 ounces of milk (2% = 122 cal) = 674
10 Snack: peanut butter & jelly, two pieces of bread, 4-5 small carrot nibs, 8 ounces of milk = 674
Dinner: meat patty (4 ox beef = 123), potatoes (1 cup = 134) , 4-5 small carrot nibs (14) 8
11
ounces of milk (2% = 122 cal) , 1-2 1” cookies. (1 oz – 145) = 538
12
Total calories: 2174
13
5. These recommendations and evaluations are only generic, as adjustments must be made
14
dependent on the mother’s BMI, and also blood work analysis, and an evaluation of what
15 supplements are being provided.
16 6. The declarations also raise concerns about the treatment of a UTI in pregnancy. If not treated
17 UTIs in pregnancy can lead to kidney issues, miscarriage, premature labor, and/or low birth
18 weight. The medical provider was right to treat an asymptomatic UTI in pregnancy, but
19 apparently did not adequately explain to the women why she was treating. More concerning
20 is the fact that the medical provider prescribed a medication that was contraindicated in
pregnancy to a verified pregnant patient. Depending on the trimester of this patient and the
21
antibiotic, this could result in significant birth defects such as esophageal atresia,
22
diaphragmatic hernia and cleft lip or palate. In the declarations there was also mention of
23
patients being prescribed contraindicated medications for trichomonas. Metronidazole is the
24
first line treatment for trichomonas and is safe in pregnancy, but further information is
25 required to verify this was the medication appropriately prescribed. Another troubling
26 complaint among the declarations is that laundry is not being adequately sanitized and
27 cleaned. Bacteria on soiled underwear can lead to a UTI as well as bacterial vaginosis, which
28 both can lead to significant maternal and fetal health conditions if left untreated. It would
Page 3 of 9
behoove an institution with the responsibility for housing and caring for pregnant women to
1
take immediate steps to reduce the incidences of bacterial vaginosis through the simple
2
measure of insuring sanitized and well laundered underwear. There is no justification for a
3
failure to do so. Proper laundering is a much preferred approach to reducing the spread of
4
bacteria, as there are clear associations between the use of antibiotics and birth defects.
5 Attached as Exhibit D is a true and correct copy of a Health and Human Services paper on
6 this subject.
7 7. Many of the pregnant women raise concerns about vaginal bleeding. Approximately 25% of
8 all pregnant women experience vaginal bleeding during the first trimester. And half of those
9 who experience vaginal bleeding result in miscarriages. Therefore, it is important to properly
10 diagnose and treat the issue of vaginal bleeding. From the declarations, it does not appear that
the medical provider at Santa Rita is following necessary protocols to diagnose and then treat
11
the incidences of vaginal bleeding. Simple bedrest is not sufficient as a treatment, and that is
12
all that has been presented. Vaginal bleeding could be a symptom of an ectopic pregnancy
13
and should be properly diagnosed. Attached as Exhibit E is a true and correct copy of an
14
article from American Family Physician on first trimester bleeding.
15 I make this declaration under penalty of perjury under the laws of the State of California.
16 Executed in Vallejo, California.
17
Dated: Jan. 26, 2018 JENNY SCAFIDI
18
19
20
_/s/ Jenny Scafidi__________________________
21
22 Under N.D. Cal. Local Rule 5-1(i)(3), I attest that I obtained concurrence in the filing of this
23 document from Jenny Scafidi on January 26, 2018.
.
24 By: __/s/ Yolanda Huang______________
YOLANDA HUANG
25
26
27
28
Page 4 of 9
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
EXHIBIT A
23
24
25
26
27
28
Page 5 of 9
nutrients
Article
Impact of Second Trimester Maternal Dietary Intake
on Gestational Weight Gain and Neonatal
Birth Weight
Malshani L. Pathirathna 1,2, *, Kayoko Sekijima 1 , Mieko Sadakata 1 , Naoshi Fujiwara 3 ,
Yoshiyuki Muramatsu 1 and Kuruppu M.S. Wimalasiri 4
1 Department of Nursing, Graduate School of Health Sciences, Niigata University, 2-746 Asahimachi-dori,
Chuo-ku, Niigata 951-8518, Japan; sekijima@clg.niigata-u.ac.jp (K.S.); atom@clg.niigata-u.ac.jp (M.S.);
murayosi@clg.niigata-u.ac.jp (Y.M.)
2 Department of Nursing, Faculty of Allied Health Sciences, University of Peradeniya,
Peradeniya 20400, Sri Lanka
3 Department of Medical Technology, Graduate School of Health Sciences, Niigata University,
2-746 Asahimachi-dori, Chuo-ku, Niigata 951-8518, Japan; fujiwaranaoshi@gmail.com
4 Department of Food Science and Technology, Faculty of Agriculture, University of Peradeniya,
Peradeniya 20400, Sri Lanka; swarnaw@pdn.ac.lk
* Correspondence: malshanilakshika@gmail.com or malshanilakshika@pdn.ac.lk; Tel.: +81-70-3604-4661
Received: 26 May 2017; Accepted: 14 June 2017; Published: 17 June 2017
Abstract: Poor maternal nutrition is a major contributor to the high incidence of low birth weight
deliveries in developing countries. This study aimed to assess the impact of second trimester
maternal dietary intake on gestational weight gain and neonatal birth weight. A longitudinal study
was conducted in a tertiary care hospital in Sri Lanka. Participants were 141 pregnant women at
18–24 weeks gestation who were followed up until delivery. Maternal dietary intake was assessed
using a validated Food Frequency Questionnaire at 21.1 ± 1.8 gestational weeks. Gestational weight
gain was examined at the end of 28 weeks gestation and at the end of pregnancy. Energy and nutrient
intakes were calculated using NutriSurvey 2007 (EBISpro, Willstaett, Germany) nutrient analysis
software, modified for Sri Lankan foods. The mean total gestational weight gain of women with low
carbohydrate intake (229–429 g/day) was 2.2 kg less than that of women with moderate carbohydrate
intake (430–629 g/day) (95% confidence interval (CI) 0.428–4.083 kg; p = 0.016). Similarly, babies
of women with low carbohydrate intake were 312 g lighter compared with those of women with a
moderate carbohydrate intake (95% CI 91–534 g; p = 0.006). Our results suggest that second trimester
maternal carbohydrate intake has significant impacts on total gestational weight gain and neonatal
birth weight.
Keywords: low birth weight; maternal diet; macronutrient; gestational weight gain; nutrition;
Sri Lanka
1. Introduction
Despite a consistent decline in maternal and infant mortality, Sri Lanka continues to experience
crucial health problems among pregnant women, infants, and children. A significant problem is the
high percentage of low birth weight (LBW) deliveries. LBW is defined as weight at birth less than
2500 g regardless of gestational age. According to 2012 national health statistics, there were 55,557 LBW
deliveries in Sri Lanka, accounting for 16.3% of total live births [1].
Perinatal complications associated with LBW are mostly attributed to fetal prematurity, but may
also result from intrauterine growth restriction [2]. LBW increases the risk of infant mortality [3],
infectious diseases, inhibited growth, and inhibited cognitive development. Children with LBW
Nutrients 2017, 9, 627; doi:10.3390/nu9060627 www.mdpi.com/journal/nutrients

Nutrients 2017, 9, 627 2 of 12
are also more likely to suffer from chronic illnesses in later life [4]. When overall LBW rates are
examined, Sri Lanka performs better than other countries in the region that have a similar income level.
However, multiple causative factors are associated with Sri Lanka’s low mean birth weight, including
preterm deliveries, pregnancy complications, and increased interest in elective cesarean sections before
40 weeks gestation.
Maternal malnutrition is the main contributing factor underlying high LBW percentages in many
developing countries [5], and plays a major role in both maternal and child health. Poor maternal
nutrition has been related to adverse pregnancy outcomes. However, this association is complex,
influenced by many intrinsic and extrinsic factors, and often results from socio-cultural and behavioral
factors. Body mass index (BMI) and gestational weight gain (GWG) are indicators of maternal nutrition.
Understanding the association between maternal nutrition and neonatal birth weight may inform
development of nutritional interventions that improve neonatal birth weight to within normal limits,
improve long-term quality of life, and reduce the healthcare burden. Early fetal nutrition is considered
the most important factor in child health, even before birth. If a woman is undernourished in early
pregnancy, the fetal metabolism will be altered as fetal adaption to the undernourished utero-placental
environment takes place. Ultimately, this slows the overall fetal growth rate and increases the risk
for LBW. However, there are lack of human studies that clarify the best timing for optimum nutrition
during early pregnancy. In addition, during the first trimester of pregnancy, almost all women
experience loss of appetite, nausea and vomiting, making it difficult to obtain sufficient nutrition.
Therefore, second trimester maternal nutrition can be considered as an important turning point for
both the mother and fetus.
Sri Lankans have a unique dietary pattern. The staple meal is large serving of rice accompanied
by different vegetable, egg, meat or fish side dishes cooked with spices and (most often) coconut milk.
This rice and curry meal is commonly consumed as lunch, although it may also form breakfast and
dinner, depending on personal preferences and factors such as economic status. Traditional morning
and evening meals usually comprise a starchy staple such as string hoppers, hoppers, pittu or bread with
one or two curries. To date, no studies on the relationships among maternal dietary intake, gestational
weight gain, and neonatal birth weight have been conducted in Sri Lanka. This study aimed to assess
the influence of second trimester maternal dietary intake on GWG and neonatal birth weight, and to
explore the relationships among these three factors.
2. Materials and Methods
2.1. Design, Setting and Participants

This study used a longitudinal design, and was conducted in antenatal clinics at the General
Hospital Kurunegala, Sri Lanka from October 2015 to June 2016. A convenience sample of pregnant
women at 18–24 weeks gestation was included and they were followed up until delivery. Initial
recruitment consisted of 150 pregnant women.
2.2. Procedure
Women were invited to participate in the study through a recruitment poster displayed in
the hospital’s outpatient department. Pregnant women who were interested in participating were
requested to inform the researcher directly, in person or via telephone. Written information sheets
were provided to all interested women, including an explanation of the study purpose, data collection
methods, time taken for data collection and confidentiality of personal information.
Exclusion criteria were risk factors according to obstetric history (e.g., miscarriages/abortions,
multiple fetuses, pregnancy-induced hypertension, and gestational diabetes mellitus) and medical
history (e.g., psychiatric disorder or long term cardiac, renal, lung or gastrointestinal disease).
Impending neonates with a 5-minute Apgar score less than 5, and women who expected to deliver
their baby at another hospital were also excluded. Based on these criteria, there were seven exclusions
Nutrients 2017, 9, 627 3 of 12
by the researcher (spontaneous abortion n = 2, multiple fetuses identified at the 20-week ultra sound
scan n = 2, and maternal desire to deliver at another hospital n = 3) and two personal withdrawals,
resulting in nine exclusions between the recruitment process and the end of pregnancy.
Neonatal data were collected from hospital records after participating women had delivered
their babies. The researcher failed to find neonatal data for 14 neonates because of confusion with
hospital delivery registry records, as many similar names made it difficult to accurately locate study
participants. One neonate with a 5-minute Apgar score less than 5 was excluded at the end of data
collection. Data were collected for 126 neonates (Figure S1).
Socioeconomic and demographic data were collected by the researcher through interviews using
a pre-designed and pre-tested questionnaire. Maternal body weight and height were measured
using standard scales. Pre-pregnancy BMI was calculated by pre-pregnancy weight in kg (weight
at the first antenatal clinic visit, usually 6–8 weeks gestation) divided by height in m2 . Women’s
weight at their first antenatal clinic visit was obtained from individual pregnancy cards. Height was
measured at the time of study recruitment. World Health Organization (WHO) international BMI
cut-off points were used for BMI group categorization [6]. The Institute of Medicine (IOM) 2009
Re-examined Guidelines [7] were used to define total GWG categories: 12.5–18 kg for underweight
women, 11.5–16 kg for women of normal weight, 7–11 kg for overweight women, and 5–9 kg for obese
women. Maternal GWG was measured at two points: at the end of 28 weeks gestation and at the end
of pregnancy. GWG up to 28 weeks gestation was calculated by taking the positive difference between
the pre-pregnancy weight and that measured at 28 weeks gestation. Total GWG was calculated by
subtracting pre-pregnancy weight from delivery weight (38.8 ± 1.5 gestational weeks). Delivery
weight was obtained from each woman’s hospital records.
Participants’ dietary intake was assessed using a Food Frequency Questionnaire (FFQ) that was
developed and validated for Sri Lankan adults, including women [8,9]. Thriposha, a supplement of
pulses provided to pregnant women, was included in the FFQ under the pulses group. This is a nutrient
supplement pack that is given to all pregnant women through community maternity clinics to combat
protein, energy, and micronutrient deficiencies. It is a pre-cooked and ready-to-eat cereal legume-based
food. In total, 100 g of Thriposha contains 401.8 kcal of energy, 61.9 g of carbohydrate, 20.0 g of protein,
7.8 g of fat, 1700 IU of vitamin A, and 18 mg of iron. It is recommended to consume 50 g of Thriposha
each day during pregnancy. Participants were asked to complete the FFQ once in the second trimester
(21.1 ± 1.8 gestational weeks). The FFQ was self-administered, with support from the researcher
provided where necessary. Questions focused on the women’s usual dietary intake in the second
trimester. Food photographs were included in the questionnaire to facilitate understanding of portion
sizes. Dietary assessment aids were day-to-day standard household measurements (e.g., plate, cup,
glass and spoons of different sizes). Energy and nutrient intakes were calculated using NutriSurvey
2007 (EBISpro, Willstaett, Germany) nutrient analysis software, after modification for individual Sri
Lankan food items and recipes.
2.2.1. Nutrient Analysis Software Modification for Sri Lankan Food

Modification of the nutrient analysis software involved adding single food items to the original
software using information from the food composition tables of Sri Lanka [10], and the United States
Department of Agriculture (USDA) nutrient database [11]. It should be noted there is no updated Sri
Lankan nutritional database up to date. Macronutrient and micronutrient values for single food items
in the original software and those from the USDA database were changed to reflect local food using the
food composition tables of Sri Lanka. Available nutrient data from food packaging were used for local
food items such as processed foods, cookies and snacks. A standard local recipe book was used for
curry/mixed dishes [12]. All recipes were checked for validity by consulting a convenience sample of
10 participants from the study group, and the recipes were modified accordingly. To estimate portion
sizes for curry/mixed dishes, a procedure was adapted from a previous study [13] (Figure S2):
Nutrients 2017, 9, 627 4 of 12
• For curry/mixed dishes, ingredients were weighed to the nearest 1 g of edible portion using a
standard kitchen weighing scale (Tanita, no. 1155).
• Dishes were cooked according to the validated recipes.
• The final products were measured using standard household measurement utensils.
Recipes (curry/mixed dishes) were entered in to the modified software by entering single food
items according to the validated recipes. For each recipe, a cooking method was applied from the
available options in the software to approximate weight loss due to water evaporation from different
food preparation methods. The software then automatically calculated and standardized the nutrient
composition for 100 g of final product. One cup was considered 150 mL and one glass 200 mL.
The portion size of each food item in the FFQ was also measured and the software was modified
as necessary. After the final software modification, the FFQ was entered in the software with the
option of changing frequency of consumption to daily, weekly or monthly. Finally, participants’ FFQ
data were entered into the software and estimated energy, carbohydrate, and protein intakes per day
were calculated.
2.2.2. Calculation of Estimated Energy Requirement (EER)

The EER for the second trimester of pregnancy for each woman was calculated based on IOM
guidelines [14]:
EER (second trimester) = non-pregnant EER + 340 kcal (1)
EER (non-pregnant) = 354 (6.91 ⇥ age (years)) + physical activity coefficient ⇥

(2)
(9.36 ⇥ weight (kg) + 726 ⇥ height (m))
The value of active physical level (1.27) for women aged 19 years and older [14] was used as the
physical activity coefficient. All the pregnant women reported that they were engaging in normal
day-to-day activities and there were no women with prescribed activity limitations.
2.2.3. Recommended Dietary Allowance (RDA) of Protein

The RDA of protein during the second trimester for each woman was calculated as 1.1 g/kg/day,
using the women’s pre-pregnancy weight [15].
2.3. Data Analysis

All data were entered and rechecked in Microsoft Excel 2007. Data for energy and macronutrient
intakes were transferred from NutriSurvey 2007 to Minitab version 17 for statistical analysis.
Descriptive statistics were expressed as mean ± standard deviation (SD). Correlations between neonatal
birth weight, gestational weight gain, dietary intake and other continuous variables were evaluated
with Pearson’s correlation analysis. To test the effects of different levels of energy and macronutrient
intakes on gestational weight gain and neonatal birth weight, participating women were divided
into two groups by energy intake (1 = daily energy intake less than the EER; 2 = daily energy intake
greater than or equal to EER) and two protein intake groups (1 = less than the RDA of protein intake;
2 = greater than or equal to RDA of protein intake) based on estimated dietary intakes. As all women
were above the RDA for carbohydrate, they were grouped in three categories with same class interval:
1 = 229–429 g/day; 2 = 430–629 g/day; and 3 = 630–829 g/day. One-way analysis of variance (ANOVA)
was used to examine the differences in means of gestational weight gain, energy and nutrient intakes
and neonatal birth weight. Two general linear models were constructed to define independent factors
associated with total gestational weight gain and neonatal birth weight, controlling for confounders.
Total energy intake was not included in the general linear models to avoid multicollinearity, as the total
energy intake represents energy from carbohydrate, protein, and fat. For each analysis, 95% confidence
intervals were calculated, and p < 0.05 was considered statistically significant.
Nutrients 2017, 9, 627 5 of 12
2.4. Ethics
Informed written consent was obtained from all the subjects before they participated in the
study. The study was conducted in compliance with the principles of the Declaration of Helsinki, and
the protocol was approved by the Ethical Review Committee, Graduate School of Health Sciences,
Niigata University, Japan (No: 125); the Ethical Review Committee, Faculty of Allied Health Sciences,
University of Peradeniya, Sri Lanka; and the Institutional Ethical Review Committee, Teaching Hospital
Kurunegala, Sri Lanka (No: ERC/2015/06).
3. Results
3.1. Participants’Characteristics
The final sample included 126 healthy newborns. Of these, 22 (17.4%) were LBW babies. In total,
20.6% women were underweight when they became pregnant and 56.7% had normal BMI. Participants’
characteristics are summarized in Table 1.
Table 1. Participants’ characteristics (n = 141).
Variable Mean (SD) n (%)

Age (years) 28.8 (6.2) -
Gestational age (weeks) a 38.8 (1.5) -
Pre-pregnancy weight (kg) b 51.9 (10.2) -
Pre-pregnancy BMI (kg/m2 ) 22.1 (4.3) -
Parity
Primiparous - 47 (33.3)
Multiparous - 94 (66.7)
Presence of hyperemesis gravidarum c - 23 (16.3)
History of miscarriage or abortion - 38 (26.9)
History of LBW delivery - 28 (19.9)
Total energy intake (kcal/day) 2921.5 (687.7) -
Carbohydrate intake (g/day) 532.7 (133.8) -
Total protein intake (g/day) 71.2 (16.8) -
Fat intake (g/day) 45.8 (16.9) -
an = 126. b Measured at the first antenatal clinic visit, usually around 6–8 weeks gestation. c During first trimester.
BMI: body mass index; LBW: low birth weight. SD: standard deviation.
Of the 141 pregnant women, 138 returned a completed FFQ. Two incomplete questionnaires were
excluded from the final analysis. The mean EER was 2224.4 ± 138.6 kcal/day; 26 (19.1%) participants
were below the EER, whereas almost all women (100%) had a carbohydrate intake above the RDA
(equal to 175 g/day). The mean RDA of protein was 57.3 ± 11.3 g/day, and 81.6% women had a protein
intake above the RDA (calculated on an individual basis). Of the total energy intake, 72.8% was from
carbohydrate and 9.8% was from protein. The protein intake mainly comprised plant protein (75.7%
of total protein intake), with only 12.8% of the total protein intake being animal protein. The mean
animal protein intake was 9.1 g/day and the mean plant protein was intake 53.9 g/day. Women with
the lowest level of monthly household income showed the lowest total energy, carbohydrate, and
protein intakes, but this did not reach statistical significance. Women living in rural areas showed the
highest carbohydrate intake, but this did not significantly differ from women in urban and sub-urban
areas (Table S1).
3.2. Correlation among Maternal Parameters, GWG and Neonatal Birth Weight
The correlation between total GWG and neonatal birth weight was 0.194 (p = 0.046). When
the analysis was repeated with pre-pregnancy BMI and gestational age fixed, there was a moderate
positive correlation between total GWG and neonatal birth weight (r = 0.302, p = 0.002). Maternal height
(r = 0.283), pre-pregnancy weight (r = 0.247), and pre-pregnancy BMI (r = 0.340) were significantly
Nutrients 2017, 9, 627 6 of 12
correlated with total GWG (p < 0.05). Gestational age (r = 0.315), pre-pregnancy weight (r = 0.234),
and pre-pregnancy BMI (r = 0.187) were significantly correlated with neonatal birth weight (p < 0.05).
A strong positive correlation was observed between GWG up to 28 weeks gestation and total GWG
(r = 0.812, p < 0.001). None of the studied dietary factors (total energy, carbohydrate, protein, and fat
intakes) showed a significant correlation with total GWG or neonatal birth weight.
3.3. Effects of Maternal and Neonatal Characteristics on GWG

A univariate analysis was performed to test the effects of maternal and neonatal characteristics on
GWG. Of the factors assessed, only the pre-pregnancy BMI category showed a significant relationship
with GWG up to 28 weeks gestation (p < 0.001), whereas both the pre-pregnancy BMI category
(p < 0.001) and dietary protein intake category (p < 0.05) were significantly associated with total GWG
(Table 2).
The fitted general linear model (Table 3) showed significant effects of education level (p < 0.05),
pre-pregnancy BMI category (p < 0.001), and daily carbohydrate intake category (p < 0.05) on total GWG.
The respective effects of fat intake, parity, and dietary protein intake category were not significant
(p > 0.05). Women with an underweight pre-pregnancy BMI showed a higher mean total GWG
(3.8 kg) compared with women with normal pre-pregnancy BMI (p < 0.001). Similarly, women with
underweight pre-pregnancy BMI showed a 4.8 kg higher mean total GWG compared with women
who were overweight (p < 0.001) and a 5.4 kg higher gain than those who were obese (p < 0.001).
The mean total GWG for women with a carbohydrate intake of 229–429 kcal/day was 2.2 kg below
that of women with a carbohydrate intake of 430–629 kcal/day (p = 0.016).
3.4. Effects of Maternal and Neonatal Characteristics on Neonatal Birth Weight

The univariate analysis revealed that five of 12 factors (area of residence, history of LBW delivery,
total GWG category, total energy intake category, and carbohydrate intake category) were significantly
associated with neonatal birth weight (p < 0.05) (Table 4).
The general linear model (R2 (adjusted) = 13.3%) showed that pre-pregnancy BMI (p = 0.025),
gestational age (p < 0.001), parity (p < 0.05), and carbohydrate intake category (p < 0.05) had significant
effects on neonatal birth weight. The babies of urban mothers were 258 g lighter than those of rural
mothers; However, the difference was not significant (p = 0.062). The mean birth weight of babies of
primiparous mothers was 187.4 g below that of babies of multiparous mothers (p = 0.046). For the
carbohydrate intake category, the mean neonatal birth weight for category 1 (229–429 kcal/day) was
312 g below the mean birth weight of those in category 2 (430–629 kcal/day) (p = 0.006). Income level
had no significant effect on neonatal birth weight (p > 0.05) (Table 5).
3.5. Effects of Supplemental Foods on GWG and Neonatal Birth Weight

The mean Thriposha intake was 34.8 ± 29.3 g/day (range 0–200 g/day); 11.8% (16/136) of women
reported they were not consuming the supplement at all. There was no significant difference in total
GWG between women who consumed 0–49 g/day of Thriposha (9.0 ± 3.7 kg) and those who consumed
50 g/day (9.8 ± 3.9 kg) (p = 0.271). The mean neonatal birth weight was 2859.1 ± 488.4 g in mothers
who consumed 0–49 g/day of Thriposha, and 2906.3 ± 523.1 g in those who consumed 50 g/day.
There was no significant difference in mean neonatal birth weight between the two Thriposha groups
(F = 0.26, p = 0.608).
Nutrients 2017, 9, 627 7 of 12
Table 2. Gestational weight gain (GWG) by maternal and neonatal characteristics (ANOVA).
GWG up to 28 Weeks (n = 105) Total GWG (n = 119)
Variable Sub-category n 95% CI p-Value 95% CI p-Value
Mean (SD) Mean (SD)
All - 6.33 (2.85) 5.78–6.89 - 9.30 (3.72) 8.63–9.98 -
None/primary 16 5.72 (2.80) 4.20–7.24 7.34 (3.72) 5.51–9.17
Education level Secondary 99 6.46 (2.92) 5.85–7.07 0.519 9.64 (3.71) 8.90–10.37 0.072
Higher 2 5.13 (0.11) 1.84–8.42 8.60 (0.57) 3.42–13.78
<9000 5 6.67 (2.57) 3.33–10.00 9.08 (4.91) 5.75–12.41
9000–13,999 20 6.09 (2.62) 4.73–7.46 8.30 (3.44) 6.64–9.96
Income level
14,000–19,999 32 6.27 (3.35) 5.26–7.27 0.978 9.42 (3.23) 8.11–10.74 0.474
(LKR (Sri Lankan rupee))
20,000–31,999 45 6.52 (2.77) 5.60–7.45 9.23 (4.21) 8.12–10.34
32,000 16 6.04 (2.44) 4.30–7.79 10.70 (3.31) 8.77–12.61
Urban 9 4.53 (3.49) 2.40–6.66 7.44 (3.93) 4.97–9.92
Area of residence Sub-urban 52 6.68 (3.33) 5.86–7.50 0.167 9.28 (3.99) 8.25–10.31 0.306
Rural 55 6.18 (2.15) 5.37–7.00 9.52 (3.46) 8.52–10.52
Yes 22 5.33 (3.03) 4.14–6.52 8.04 (3.36) 6.48–9.59
History of LBW deliveries 0.062 0.077
No 97 6.60 (2.76) 5.99–7.21 9.60 (3.76) 8.85–10.33
History of miscarriage Yes 32 6.47 (2.75) 5.43–7.51 8.87 (3.75) 7.56–10.17
0.759 0.442
or abortion No 87 6.28 (2.91) 5.62–6.94 9.46 (3.77) 8.67–10.25
Yes 20 6.34 (3.21) 5.03–7.65 9.04 (4.20) 7.37–10.70
Hyperemesis gravidarum a 0.978 0.730
No 97 6.32 (2.80) 5.70–6.93 9.36 (3.67) 8.60–10.12
Primiparous 40 6.89 (2.39) 5.86–7.92 10.00(4.00) 8.84–11.16
Parity 0.208 0.146
Multiparous 79 6.11 (3.00) 5.46–6.76 8.95 (3.54) 8.12–9.77
Underweight 23 8.43 (2.94) 1 7.24–9.62 12.00(4.07) 1 10.58–13.43
Pre-pregnancy BMI Normal 69 6.18 (2.26) 2 5.51–6.85 9.10 (3.12) 2 8.27–9.92
category b <0.001 * <0.001 *
Overweight 20 5.73 (3.38) 2,3 4.60–6.86 7.89 (3.81) 2 6.37–9.42
Obese 7 2.76 (0.97) 3 0.44–5.08 6.50 (3.30) 2 3.92–9.08
<EER 21 5.57 (3.04) 4.32–6.82 8.05 (3.44) 6.44–9.67
Energy intake c 0.188 0.083
EER 93 6.51 (2.85) 5.87–7.16 9.62 (3.80) 8.86–10.39
229–429 25 5.97 (3.08) 4.78–7.15 8.03 (3.60) 6.56–9.50
Carbohydrate intake
430–629 58 6.41 (2.99) 5.60–7.23 0.798 10.00 (3.68) 9.03–10.96 0.089
(g/day)
630–829 31 6.45 (2.63) 5.29–7.61 9.16 (3.88) 7.84–10.49
<RDA 21 5.49 (3.21) 4.24–6.74 7.81 (3.56) 6.20–9.41
Protein intake d RDA 93 6.53 (2.80) 5.89–7.18
0.141
9.68 (3.74) 8.92–10.44
0.039 *
Male 53 6.44 (2.78) 5.63–7.25 9.23 (3.31) 8.25–10.21
Sex of the newborn 0.552 0.976
Female 53 6.10 (2.75) 5.28–6.91 9.21 (3.86) 8.23–10.19
a During first trimester. b Based on WHO international BMI cut-off values [6]. c EER for the pregnancy second trimester based on IOM 2009 guidelines [14]. d RDA based on IOM Food and
Nutrition Board 2002/2005 guidelines [15]. 1,2,3 Values with the same superscript Arabic numeral do not represent a significance difference. Compared using one-way ANOVA followed by
Tukey’s post-hoc test. “n” column represents the subject numbers corresponding to total GWG. * p < 0.05. RDA: recommended dietary allowance; WHO: World Health Organization; EER:
estimated energy requirement; CI: confidence interval; IOM: Institute of Medicine.
Nutrients 2017, 9, 627 8 of 12
Table 3. General linear model for total gestational weight gain (kg).
Variable in Model Coefficient 95% CI t p-Value

Constant 11.35 8.30–14.39 7.4 <0.001 *
Continuous variables
Fat intake 0.04 0.08–0.00 1.77 0.08
Categorical variables
Education (none/primary)—reference level
Education (secondary) 2.19 0.40–3.97 2.43 0.017 *
Education (higher) 1.29 6.24–3.65 0.52 0.605
Pre-pregnancy BMI category
(underweight)—reference level
Pre-pregnancy BMI category (normal) 3.84 5.49– 2.19 4.62 <0.001 *
Pre-pregnancy BMI category (overweight) 4.80 6.96– 2.64 4.41 <0.001 *
Pre-pregnancy BMI category (obese) 5.44 8.49– 2.40 3.55 0.001 *
Parity (primiparous)—reference level
Parity (multiparous) 0.93 2.27–0.41 1.38 0.171
Category of carbohydrate intake
(229–429 g/day)—reference level
Category of carbohydrate intake (430–629 g/day) 2.26 0.43–4.08 2.45 0.016 *
Category of carbohydrate intake (630–829 g/day) 1.60 0.49–3.7 1.52 0.132
Category of protein intake (<RDA)—reference level
Category of protein intake ( RDA) 0.42 1.68–2.52 0.4 0.691
R2 (adjusted) = 16.5%. n = 112. * p < 0.05.
Table 4. Neonatal birth weight by maternal and neonatal characteristics, ANOVA.
Birth Weight (g)

Variable Sub-category n 95% CI p-Value
Mean (SD)
All - 126 2874.6 (497) 2787.0–2962.2 -
None/primary 20 2831 (428) 2612.5–3049.5
Education level Secondary 100 2867.6 (508) 2769.9–2965.3 0.629
Higher 4 3091 (391) 2603–3580
<9000 5 3130 (432) 2702–3558
9000–13,999 22 2702 (499) 2498–2906
Income level (LKR) 14,000–19,999 31 2965 (573) 2793–3136 0.093
20,000–31,999 50 2803.3 (454) 2667.9–2938.7
32,000 16 3036.9 (352.8) 2797.6–3276.2
Urban 13 2754 (519) 1,2 2492–3016
Area of residence Sub-urban 54 2747.1 (461.9) 1 2618.0–2875.7 0.011 *
Rural 57 3011.4 (481.9) 2 2886.3–3136.5
History of LBW Yes 24 2684.2 (418.2) 2486.1–2882.2
0.036 *
deliveries No 102 2919.4 (505.2) 2823.3–3015.4
History of miscarriage or Yes 33 2898.9 (507.6) 2727.1–3070.8
0.744
abortion No 93 2865.9 (495.7) 2763.5–2968.3
Hyperemesis Yes 20 2800.5 (366.5) 2582.4–3018.6
gravidarum a 0.499
No 104 2882.1 (512.5) 2786.4–2977.7
Primiparous 45 2770.1 (430.5) 2624.7–2915.5
Parity 0.079
Multiparous 81 2932.6 (523.9) 2824.2–3041.0
Underweight 27 2771.5 (459.8) 2523.2–2899.8
Pre-pregnancy BMI Normal 70 2912.5 (442.5) 2795.5–3029.5
category b 0.231
Overweight 22 2895 (673) 2687–3104
Obese 7 3059 (464) 2689–3428
Within recommended 32 2912.8 (539.7) 1,2 2740.8–3084.8
Total GWG category Below recommended 71 2863.9 (458.5) 1 2748.4–2979.3 0.042 *
Over recommended 3 3600 (721) 2 3038–4162
Energy intake c <EER 24 2692 (533) 2491–2892
0.039 *
EER 97 2927.8 (487) 2828.0–2037.6
Nutrients 2017, 9, 627 9 of 12
Table 4. Cont.
Birth Weight (g)

Mean (SD)
229–429 30 2686.7 (498.1) 1 2508.5–2864.9
Carbohydrate intake
430–629 64 2957.7 (486) 2 2853.7–3097.7 0.033
(g/day)
630–829 27 2872.2 (503.6) 1,2 2684.4–3060.1
<RDA 24 2913 (601) 2708–3117
0.733
Sex of the newborn Male 61 2852.9 (477.3) 2726.5–2979.2
0.637
Female 65 2894.9 (517.7) 2772.5–3017.3
a b c
During first trimester. Based on WHO international BMI cut-off values. EER for the pregnancy second trimester
based on IOM 2009 guidelines [14]. d RDA based on IOM Food and Nutrition Board 2002/2005 guidelines [15].
1,2,3 Values with the same superscript Arabic numeral do not represent a significance difference. Sample sizes vary
slightly because of missing data. Compared using one-way ANOVA followed by Tukey’s post-hoc test. * p < 0.05.
Table 5. General linear model for neonatal birth weight (g).

Constant 2011 4349–327 1.71 0.091
Pre-pregnancy BMI 23.7 3.0–44.4 2.27 0.025 *
Gestational age 102.1 46.7–157.5 3.65 <0.001 *
Fat intake 2.83 8.21–2.54 1.05 0.298
Income (<9000 LKR)—reference level
Income (9000–13,999 LKR) 209 679–262 0.88 0.381
Income (14,000–19,999 LKR) 43 495–410 0.19 0.852
Income (20,000–31,999 LKR) 74 523–374 0.33 0.743
Income ( 32,000 LKR) 156 322–634 0.65 0.518
Area of residence (urban)—reference level
Area of residence (sub-urban) 36 235–308 0.27 0.791
Area of residence (rural) 258 14–529 1.88 0.062
History of LBW deliveries (yes)—reference level
History of LBW deliveries (no) 209 6.0–424.0 1.93 0.057
Parity (multiparous) 187.4 3.2–371.6 2.02 0.046 *
Category of carbohydrate intake (430–629 g/day) 312 91–534 2.8 0.006 *
Category of carbohydrate intake (630–829 g/day) 237 44–517 1.67 0.097
Category of protein intake ( RDA) 66 326–194 0.51 0.615
R2 (adjusted) = 13.3%. n = 118. * p < 0.05.
4. Discussion
The present study showed that 17.4% of deliveries were LBW babies, which was slightly above
the national LBW rates for 2014 [16]. This might be because data for the present study were collected
in a large tertiary care hospital that included more complicated pregnancies. In the present study,
the average total GWG was 9.3 ± 3.7 kg, which was slightly below that observed for Sri Lankan
women in a previous study [17]. However, this mean total GWG was below that recommended for
underweight and normal BMI women. We found a moderate positive correlation between total GWG
and neonatal birth weight. Therefore, total GWG may be considered a good predictor of neonatal
birth weight, this emphasizes the importance of appropriate GWG when it is sought to prevent LBW
deliveries. As there was a strong positive correlation between total GWG and GWG up to 28 weeks
gestation, GWG up to 28 weeks can be used as a meaningful predictor to optimize individualized
Nutrients 2017, 9, 627 10 of 12
antenatal care for women who show low weight gain. Pre-pregnancy BMI also showed a significant
association with neonatal birth weight, indicating the importance of maternal nutrition at the time a
woman becomes pregnant.
In the present study, all participating women had a carbohydrate intake during pregnancy above
the RDA. The mean energy intake was significantly higher than previous studies [18,19]. However,
consuming a large serving of rice and/or other starchy staple in all three main meals and the daily
consumption of supplemental food with a higher energy value are central to this high energy intake.
In addition, the concept of “eating for two” during pregnancy still prevails in Sri Lanka, even though
it has no scientific basis. In the present study, 18.4% of women had a protein intake during pregnancy
below the RDA, with main protein supply being plant protein. Although protein from animal sources
is of greater nutritional value because it contains all essential amino acids, the animal protein intake
of women in our sample was low compared with Western countries [20]. This may be explained
by the vegetarian trend of the younger generation in Sri Lanka because of religious and cultural
influences and the poor economic status. There is no separate RDA for fat intake during pregnancy,
meaning that fat is recommended to be 25–35% of the total calorie intake, as for the general population.
Therefore, in an average 2200 kcal diet, approximately 550 kcal should be fat (approximately 60 g).
The present study showed a mean fat intake of 45.8 ± 16.9 g per day, which was moderately below the
general recommendation.
Our study revealed a significant relationship between second trimester carbohydrate intake and
neonatal birth weight. Godfrey et al. found that birth weight was inversely related to carbohydrate
intake in early pregnancy [20], which is inconsistent with our results. However, gestational age at
the time of nutrient assessment in Godfrey et al.’s study [20] was around 15 weeks gestation, which
differed from the present study, making it difficult to conduct a fair comparison. We found that
moderate carbohydrate intake was associated with both the total GWG and neonatal birth weight.
Godfrey et al. [20] suggested that high carbohydrate intake in early pregnancy suppressed placental
growth (and thus fetal growth), as did low dairy protein intake in late pregnancy. However, it was not
possible to compare our present results with those of Godfrey et al. because we focused on dietary
intake in the second trimester. No significant difference in any background characteristic was apparent
between the high and moderate carbohydrate-intake groups. Although statistical significance was
not attained, the moderate-intake group contained a higher proportion (54.7%) of wealthier women
(monthly household income 2000 LKR); this may partially explain the higher GWG and birth weight
in this group. The present study showed that babies of rural mothers were heavier than those of
urban mothers, although this did not reach the level of significance. This might be attributable to the
higher carbohydrate consumption of rural women compared with urban women (Table S1). No direct
relationship was observed between second trimester protein intake and neonatal birth weight, which is
consistent with results shown by Godfrey et al. [20]. Our univariate analysis showed that women with
a total energy intake below the EER delivered neonates with significantly lower mean birth weight
compared with women who were above the EER. Although the majority of women in our study were
above the EER and RDA for selected macronutrients, individual dietary analysis showed an imbalance
in meal habits, for example, a high proportion of carbohydrate and low amount of other important
nutrients. In particular, mean fat intake was relatively low. Findings of the current study can be
explained by the importance of carbohydrate intake for total GWG, and thereby neonatal birth weight.
Provision of the Thriposha supplement for pregnant women aims to maintain satisfactory GWG
and reduce the LBW percentage. Although this supplement is distributed free of charge, the present
study indicated that the mean Thriposha intake was below the recommended intake of 50 g/day.
In addition, 11.8% of women reported that they were not consuming the Thriposha supplement
even though they received it from the community clinics. This may be because the supplement
was consumed by the entire family rather than the pregnant woman. However, in the present study,
supplement consumption showed no significant effect on either GWG or neonatal birth weight. Further
Nutrients 2017, 9, 627 11 of 12
large scale, nationwide studies are recommended to evaluate the effects of supplements distributed
free of charge by community clinics.
4.1. Limitations
There was a notable data loss between recruitment and the end of data collection, resulting in
a smaller sample size than expected. In addition, using a FFQ to collect dietary intake data might
have resulted in over/under estimation of actual intake. Despite these limitations, the present study
provides the first local estimates of energy and macronutrient consumption among pregnant women
in Sri Lanka.
4.2. Implications
The Sri Lanka Ministry of Health recommends that the total GWG should be based on the IOM
2009 revised guidelines. Over 15% of women with inadequate GWG delivered LBW infants, indicating
that although the IOM guidelines were developed for Americans with larger body frames, they
were useful in our setting for identifying women at risk of delivering LBW infants. Culturally and
economically competent health care is required to allow Sri Lankan women to achieve a desirable GWG;
the emphasis should be on a healthy diet and regular weight monitoring. Nutritional education should
be integrated into first-trimester antenatal education sessions, with a focus on meals meeting the dietary
requirements of pregnancy and featuring a variety of foods from all food groups. Individual nutritional
counseling should be provided to women at risk, especially those with an underweight pre-pregnancy
BMI, those exhibiting low GWG, and those on poor diets. We also found that almost all women
consumed more than the RDA of carbohydrates during pregnancy. Thus, culturally appropriate
dietary recommendations relevant in the Sri Lankan context should be essential components of
strategies seeking to prevent low birth weight.
5. Conclusions
Second trimester maternal carbohydrate intake shows significant relationships with total GWG
and neonatal birth weight. Maintaining a moderate level of carbohydrate intake (430–629 g/day)
during the second trimester of pregnancy may promote favorable total GWG and neonatal birth weight
in the Sri Lankan context. Our results highlight the need for primary healthcare providers to be vigilant
in assessing maternal dietary intake during the second trimester. Individualized education should be
provided about good sources of carbohydrates in meals for pregnant women who show weight gain
below the recommended levels.
Supplementary Materials: The following are available online at www.mdpi.com/2072-6643/9/6/627/s1,

Figure S1: Process of data collection, Figure S2: Example estimation of portion sizes for curry dishes (bean
curry), Table S1: Second trimester maternal energy and macro-nutrient intake by maternal and neonatal
characteristics; ANOVA.
Acknowledgments: The authors thank all women who participated in this study, obstetricians, special grade
nursing officers, and the staff nurses of the antenatal clinics for their support of this study. This study received no
specific grant from any public, commercial, or not-for-profit sector.
Author Contributions: M.L.P., K.S., M.S., N.F. and K.M.S.W. contributed to the conception and design of this
study. M.L.P. conducted the survey. M.L.P. and K.S. performed the statistical analysis. M.L.P. wrote the paper.
K.S., M.S., N.F. Y.M. and K.M.S.W. critically reviewed the manuscript and supervised the whole study process.
All authors read and approved the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Ministry of Health. Annual Health Bulletin; Ministry of Health, Nutrition and Indigenous Medicine: Colombo,
Sri Lanka, 2012.
Nutrients 2017, 9, 627 12 of 12
2. Patterson, R.M.; Prihoda, T.J.; Gibbs, C.E.; Wood, R.C. Analysis of birth weight percentile as a predictor of
perinatal outcome. Obstet. Gynecol. 1986, 68, 459–463. [PubMed]
3. Joseph, K.S.; Kramer, M.S. Recent trends in infant mortality rates and proportions of low-birth-weight live
births in Canada. CMAJ 1997, 15, 535–541.
4. Barker, D.J.P. Mothers, Babies and Health in Later Life, 2nd ed.; Churchill Livingstone: Edinburgh, UK, 1998.
5. Ramakrishnan, U. Nutrition and low birth weight: From research to practice. Am. J. Clin. Nutr. 2004, 79,
17–21. [PubMed]
6. World Health Organization. Global Database on Body Mass Index. Available online: http://apps.who.int/bmi/
index.jsp?introPage=intro-3.html (accessed on 10 April 2016).
7. Institute of Medicine. Weight Gain during Pregnancy: Reexamining the Guidelines (2009). Available
online: http://www.nationalacademies.org/hmd/~/media/Files/Report%20Files/2009/Weight-Gain-
During-Pregnancy-Reexamining-the-Guidelines/Report%20Brief%20-%20Weight%20Gain%20During%
20Pregnancy.pdf (accessed on 10 March 2017).
8. Jayawardena, R.; Swaminathan, S.; Byrne, N.M.; Soares, M.J.; Katulanda, P.; Hills, A.P. Development of a
food frequency questionnaire for Sri Lankan adults. Nutr. J. 2012, 11, 63. [CrossRef] [PubMed]
9. Jayawardena, R.; Byrne, N.M.; Soares, M.J.; Katulanda, P.; Hills, A.P. Validity of a food frequency
questionnaire to assess nutritional intake among Sri Lankan adults. SpringerPlus 2016, 5, 162. [CrossRef]
[PubMed]
10. Wimalasiri, W.R. Composition of Sri Lankan foods; Faculty of Dental Sciences, University of Peradeniya:
Peradeniya, Sri Lanka, 1999.
11. United States Department of Agriculture. USDA Food Composition Databases. Available online:
https://ndb.nal.usda.gov/ndb/ (accessed on 5 May 2016).
12. Dissanayake, C. Ceylon Cookery, 6th ed.; Stamford Lake (Pvt) Ltd.: Pannipitiya, Sri Lanka, 2005.
13. Jayawardena, R.; Thennakoon, S.; Byrne, N.; Soares, M.; Katulanda, P.; Hills, A. Energy and nutrient intakes
among Sri Lankan adults. Int. Arch. Med. 2014, 7, 34. [CrossRef] [PubMed]
14. Institute of Medicine. Weight Gain during Pregnancy: Reexamining the Guidelines (2009). Available online:
https://www.ncbi.nlm.nih.gov/books/NBK32812/ (accessed on 13 June 2016).
15. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol,
Protein and Amino Acids. Available online: http://www.nationalacademies.org/hmd/Home/Global/
News%20Announcements/%20DRI (accessed on 3 June 2016).
16. Ministry of Health, Nutrition and Indigenous Medicine. Annual Health Bulletin; Ministry of Health, Nutrition
and Indigenous Medicine: Colombo, Sri Lanka, 2014.
17. Abeysena, C.; Jayawardana, P. Body mass index and gestational weight gain in two selected Medical Officer
of Health areas in the Gampaha District. J. Coll. Community Physicians Sri Lanka 2011, 16, 30–34. [CrossRef]
18. Mathews, F.; Yudkin, P.; Neil, A. Influence of maternal nutrition on outcome of pregnancy: Prospective
cohort study. BMJ 1999, 319, 339–343. [CrossRef] [PubMed]
19. Kubota, K.; Itoh, H.; Tasaka, M.; Naito, H.; Fukuoka, Y.; Muramatsu, K.; Kobayashi, K.; Sugihara, K.;
Kanayama, N.; Hamamatsu Birth Cohort Study Team. Changes of maternal dietary intake, body weight
and fetal growth throughout pregnancy in pregnant Japanese women. J. Obstet. Gynaecol. Res. 2013, 39,
1383–1390. [CrossRef] [PubMed]
20. Godfrey, K.; Robinson, S.; Barker, D.J.P.; Osmond, C.; Cox, V. Maternal nutrition in early and late pregnancy
in relation to placental and fetal growth. BMJ 1996, 312, 410–414. [CrossRef] [PubMed]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 EXHIBIT B
23
24
25
26
27
28
Page 6 of 9
FROM THE ACADEMY

Position Paper
Position of the Academy of Nutrition and Dietetics:

Nutrition and Lifestyle for a Healthy Pregnancy Outcome
ABSTRACT POSITION STATEMENT

It is the position of the Academy of Nutrition and Dietetics that women of childbearing It is the position of the Academy of Nutrition
age should adopt a lifestyle optimizing health and reducing risk of birth defects, sub- and Dietetics that women of childbearing
age should adopt a lifestyle optimizing
optimal fetal development, and chronic health problems in both mother and child. health and reducing risk of birth defects,
Components leading to a healthy pregnancy outcome include healthy prepregnancy suboptimal fetal development, and chronic
weight, appropriate weight gain and physical activity during pregnancy, consumption of health problems in both mother and child.
a wide variety of foods, appropriate vitamin and mineral supplementation, avoidance of Components leading to healthy pregnancy
outcome include healthy prepregnancy
alcohol and other harmful substances, and safe food handling. Pregnancy is a critical weight, appropriate weight gain and physical
period during which maternal nutrition and lifestyle choices are major influences on activity during pregnancy, consumption of a
mother and child health. Inadequate levels of key nutrients during crucial periods of wide variety of foods, appropriate vitamin
fetal development may lead to reprogramming within fetal tissues, predisposing the and mineral supplementation, avoidance of
alcohol and other harmful substances, and
infant to chronic conditions in later life. Improving the well-being of mothers, infants, safe food handling.
and children is key to the health of the next generation. This position paper and the
accompanying practice paper (www.eatright.org/members/practicepapers) on the same
topic provide registered dietitian nutritionists and dietetic technicians, registered; other
professional associations; government agencies; industry; and the public with the
Academy’s stance on factors determined to influence healthy pregnancy, as well as an
overview of best practices in nutrition and healthy lifestyles during pregnancy.
J Acad Nutr Diet. 2014;114:1099-1103.
T
HIS POSITION PAPER PROVIDES choices are major influences on mother !25) and almost one third were obese
Academy of Nutrition and Di- and child health. Improving the well- (BMI !30).4 Overconsumption/over-
etetics members, other profes- being of mothers, infants, and children weight throughout the reproductive
sional associations, government is key to the health of the next genera- cycle are related to short- and long-
agencies, industry, and the public with tion. One in 33 babies (approximately term maternal health risks, including
the Academy’s stance on factors deter- 3%) is born with a birth defect2; in 2010, obesity, diabetes, dyslipidemia, and
mined to influence healthy pregnancy, as low-birth-weight (LBW) infants cardiovascular disease. Caloric excess
well as emerging factors. Women with comprised 8.1% of US births.3 Birth de-
does not guarantee adequate intake or
inappropriate weight gain, hyperemesis, fects and LBW are ranked first and sec-
nutrient status critical to healthy
multiple gestations, poor dietary patterns ond, respectively, among the 10 leading
(eg, disordered eating), or chronic disease causes of death in US infants in 2006.3 A pregnancy outcomes.5
should be referred to a registered dieti- woman’s chance of having a healthy To improve maternal and child health
tian nutritionist (RDN) for medical nutri- baby improves when she adopts healthy outcomes, women should weigh within
tion therapy. For specific practice behaviors, including good nutrition; the normal BMI range when they
recommendations, refer to the Academy’s recommended supplementation; and conceive and strive to gain within
practice paper on “Nutrition and Lifestyle avoidance of smoking, alcohol, and illicit ranges recommended by the Institute of
for a Healthy Pregnancy Outcome.”1 drugs before becoming pregnant.2 Medicine (IOM) 2009 pregnancy weight
guidelines.4 High rates of overweight
and obesity are common in population
TRENDS IMPACTING OBESITY AND GESTATIONAL subgroups already at risk for poor
PREGNANCY OUTCOMES DIABETES maternal and child health outcomes,
Birth Defects, Low Birth Weight, Prepregnancy body mass index (BMI) is compounding the need for interven-
and Viable Birth Trends an independent predictor of many tion.4 In addition to health risks,
Pregnancy is a critical period during adverse outcomes of pregnancy. The gestational weight gain beyond the
which maternal nutrition and lifestyle prevalence of obesity in women 12 to recommendation substantially in-
44 years of age has more than doubled creases risk of excess weight retention
since 1976. In 1999 to 2004, nearly two in obese women at 1 year postpartum.6
2212-2672/$36.00
thirds of women of childbearing age More information on obesity and preg-
http://dx.doi.org/10.1016/j.jand.2014.05.005
were classified as overweight (BMI nancy outcomes can be found in the
ª 2014 by the Academy of Nutrition and Dietetics. JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1099
FROM THE ACADEMY
“Position of the Academy of Nutrition maternal metabolic conditions may be anemia in pregnant women in industri-
and Dietetics and American Society for associated with neurodevelopmental alized countries is 17.4%,19 with approx-
Nutrition: Obesity, Reproduction, and problems, including autism and devel- imately 9% of adolescent girls and
Pregnancy Outcomes.” 7 opmental delays in children.12 Inade- women of childbearing age in the United
New diagnostic criteria for gesta- quate levels of key nutrients during States having inadequate stores of body
tional diabetes mellitus (GDM) are ex- crucial periods of fetal development iron.20 The high incidence of iron defi-
pected to increase the proportion of may lead to reprogramming within ciency underscores the need for iron
women diagnosed with GDM, with fetal tissues, predisposing the infant to supplementation in pregnancy. During
potentially 18% of all pregnancies chronic conditions in later life. Those the first two trimesters of pregnancy,
affected.8 Immediately after pregnancy, conditions include obesity, cardiovas- iron-deficiency anemia increases the risk
5% to 10% of women with GDM are found cular disease, bone health, cognition, for preterm labor, LBW, and infant mor-
to have diabetes, usually type 2. Women immune function, and diabetes.13 tality.18 Maternal and fetal demand for
who have had GDM have a 35% to 60% Maternal weight gain during preg- iron increases during pregnancy; this
chance of developing diabetes in the nancy outside the recommended range increase cannot be met without iron
next 10 to 20 years.8 RDNs can provide is associated with increased risk to supplementation.18
valuable guidance to women seeking maternal and child health.4 Although
assistance regarding optimal weight physiological responses to prenatal Folic Acid. Folic acid is recognized as
and healthy food selection before, dur- overnutrition result in poor health important before and during pregnancy
ing, and post pregnancy. Additional in- outcomes that emerge in childhood because of its preventive properties
formation and guidance is available in and adolescence, fetal undernutrition against neural tube defects. All women,
the Academy’s GDM Evidence-Based responses range from fetal survival to including adolescents, who are capable
Nutrition Practice Guideline.9 poor health outcomes emerging later of becoming pregnant should consume
in the offspring’s adult life.14 The IOM 400 mg/day folic acid from fortified
Hypertension and Preeclampsia recommends that more US women foods and/or dietary supplements, in
achieve gestational weight gain within addition to eating food sources of
Prevalence of chronic hypertension in
the range identified for their prepreg- folate.21 Pregnant women are advised to
pregnancy in the United States is esti-
nant BMI.4 Pregnant women benefit consume 600 mg dietary folate equiva-
mated to be as high as 5%. This is pri-
from eating a variety of foods to meet lents daily from all food sources. Dietary
marily attributable to the increased
nutrient needs and consuming suffi- folate equivalents adjust for the differ-
prevalence of obesity, as well as delay in
cient calories to support recommended ence in bioavailability of food folate
childbearing to ages when chronic hy-
weight gain. Details regarding recom- compared with synthetic folic acid. One
pertension is more common.10 Hyper-
mended energy requirements and rec- dietary folate equivalent is equal to 1 mg
tension in pregnancy can harm both
ommended weight gain during food folate, which is equal to 0.6 mg folic
mother and fetus, and women with
pregnancy can be found in the related acid derived from supplements and
chronic hypertension are more likely to
practice paper.1 fortified foods taken with meals.14
experience preeclampsia (17% to 25% vs
Women who have had an infant with a
3% to 5% in the general population).10
Energy Expenditure neural tube defect should consult with
Age, preconception weight and health
their health care provider regarding
status, access to timely and appropriate Physical activity during pregnancy
the recommendation to take 4,000 mg
health care, and poverty are some of the benefits a woman’s overall health. In a
folic acid daily before and throughout
numerous factors affecting maternal low-risk pregnancy, moderately intense
the first trimester of pregnancy.22 An
health and the likelihood of a healthy activity does not increase risk of LBW,
association between the lack of peri-
pregnancy. Referral to the RDN and/or preterm delivery, or miscarriage.15 Rec-
conceptual use of vitamins or supple-
social worker may assure appropriate reational moderate and vigorous phys-
ments containing folic acid with an
care will be available, given the afore- ical activity during pregnancy is
excess risk for birth defects due to dia-
mentioned factors that can influence associated with a 48% lower risk of hy-
betes mellitus23 highlights ongoing
maternal and fetal outcomes. perglycemia, specifically among women
research.
with prepregnancy BMI <25.16 A pre-
OPTIMIZING PREGNANCY natal nutrition and exercise program,
regardless of exercise intensity, has
Vitamin D. The function of vitamin D
OUTCOMES WITH HEALTHY during pregnancy for both mother and
LIFESTYLE CHOICES been shown to reduce excessive gesta-
fetus is not fully defined at present.24
tional weight gain and decrease weight
Evidence is building that maternal diet Although vitamin D supplementa-
retention at 2 months postpartum in
and lifestyle choices influence the tion during pregnancy has been sug-
women of normal prepregnant BMI.17
long-term health of the mother’s chil- gested as an intervention to protect
dren. Prepregnancy adherence to against adverse gestational outcomes,
healthful dietary patterns, including Appropriate and Timely Nutrient including LBW,25 the need, safety, and
the alternate Mediterranean Diet, Di- Supplementation effectiveness of vitamin D supplemen-
etary Approaches to Stop Hypertension Iron. Iron deficiency with resultant tation during pregnancy remains
(DASH), and alternate Healthy Eating anemia is the most prevalent micro- controversial.24 The IOM recommends
Index, have been associated with a 24% nutrient deficiency worldwide, affecting 600 IU per day of vitamin D to meet the
to 46% lower risk of GDM.11 Population- primarily pregnant or lactating women needs of most North American adults,
based research provides evidence that and young children.18 Iron-deficiency including pregnant women.26 Ongoing
1100 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS July 2014 Volume 114 Number 7
FROM THE ACADEMY
research suggests higher levels of sup- jejuni.32 Pregnant women should closely containing caffeine do not increase
plementation are safe and effective for adhere to food-safety recommendations the risk of congenital malformations,
improving maternal and infant vitamin outlined in the 2010 Dietary Guidelines miscarriage, preterm birth, or growth
D status.24 for Americans.21 Updated food-safety retardation.36
guidelines can be reviewed on the Food
Choline. Choline is an essential and Drug Administration at www. Hydration and Water Needs. Ade-
nutrient during pregnancy because of fda.gov/Food/ResourcesForYou/Health quate hydration is essential to healthy
its high rate of transport from mother Educators/ucm083308.htm. pregnancy, as a woman accumulates 6
to fetus. Maternal deficiency of choline to 9 L of water during gestation. The
can interfere with normal fetal brain Benefits and Concerns Regarding total water Adequate Intake for preg-
development. Although choline is Fish and Seafood Consump- nancy (including drinking water, bev-
found in many foods, the majority of tion. The nutritional value of seafood erages and food) is 3 L/day. This
pregnant women are not achieving the is particularly important during fetal includes approximately 2.3 L (approxi-
Adequate Intake for pregnancy of 450 growth and development, as well as in mately 10 cups) as total beverages.37
mg choline per day.27 early infancy and childhood.14 Intake of
n-3 fatty acids, particularly docosahex- Energy Drinks. An energy drink is any
Calcium. The Dietary Reference Intake aenoic acid, from at least 8 oz of seafood beverage that contains some form of
for calcium in pregnancy is equal to that per week for pregnant women is as- legal stimulant and/or vitamins added
of nonpregnant women of the same age sociated with improved infant visual to provide a short-term boost in en-
because of increased efficiency in cal- and cognitive development.14 Although ergy. These drinks may contain sub-
cium absorption during pregnancy and prenatal mercury exposure (!1 mg/g) stantial and varying amounts of sugar
maternal bone calcium mobilization.26 was found to be associated with a and caffeine, as well as taurine, carni-
Women with suboptimal intakes greater risk of attention-deficit hyper- tine, inositol, ginkgo, and milk thistle.
(<500 mg/day) may need additional activity disorder"related behaviors, Many of these have not been studied
amounts to meet both maternal and prenatal fish consumption of more than for safety during pregnancy. Ginseng,
fetal bone requirements.28 two servings per week was protective of another common ingredient, is not
those behaviors.33 RDNs and dietetic recommended for use during preg-
technicians, registered, can help pregnancy. The avoidance of energy drinks
Iodine. Iodine is required for normal
nant women balance the benefits of during pregnancy is advised.
brain development and growth; iodine
eating fish while avoiding high-mercury
deficiency worldwide is a growing
content seafood.
concern. During pregnancy, iodine re- Sugar-Sweetened Drinks. Sugar-
quirements increase, making mother sweetened beverages, including regu-
and developing fetus vulnerable.
Non-Nutritive Sweeteners. Although lar sodas, sport drinks, energy drinks,
calorie and blood glucose control are
Congenital hypothyroidism is associ- and fruit drinks, provide 35.7% of
acknowledged benefits of non-nutritive
ated with cretinism, and clinical hypo- added sugars in the US diet.14 Reduced
sweeteners, limited research addresses
thyroidism has been associated with consumption of sources of added
the safety of non-nutritive sweeteners
increased risk of poor perinatal out- sugars lowers the calorie content of the
on healthy pregnancy or in GDM.34
comes, including prematurity, LBW, diet without compromising nutrient
miscarriage, preeclampsia, fetal death, adequacy.
and impaired fetal neurocognitive
Alcohol. Alcohol should not be
consumed by pregnant women or
development.29 Recent national surveys Health Conditions Between and
those who may become pregnant.14
indicate a subset of pregnant and After Pregnancies
Drinking alcohol during pregnancy,
lactating US women may have mild to
especially in early pregnancy, may Maternal return to healthy weight sta-
moderately inadequate dietary iodine
result in behavioral or neurological tus postpartum can prevent future
intake.30 The IOM recommends an
defects in the offspring and affect a overweight and obesity.14 The 2010
iodine intake from dietary and supple-
child’s future intelligence. No safe level Dietary Guidelines for Americans forms
ment sources of 150 mg/day before
of alcohol consumption during preg- the basis for nutrition counseling for
conception, and 220 mg per day for
nancy has been established.14 postpartum women, and RDNs and di-
pregnant women.31
etetic technicians, registered, can assist
Caffeine. Caffeine half-life increases in women in achieving their prepreg-
Environmental and Dietary Issues pregnancy from 3 hours in the first nancy weight.21 Outside of weight
Foodborne Illness during Preg- trimester to 80 to 100 hours in late status, recent research has shown that
nancy. Pregnant women and their fe- pregnancy. Women who are pregnant or diet quality, dietary intake, and overall
tuses are at increased risk of developing trying to become pregnant are advised nutritional status can affect the risk of
foodborne illness because of the hor- by the American College of Obstetri- postpartum depression. An association
monal changes of pregnancy that lead to cians and Gynecologists35 to consume among n-3 fatty acids, serotonin
decreased cell-mediated immune func- no more than 200 mg of caffeine per transporter genotype, and postpartum
tion. Of greatest concern during preg- day—the approximate amount in one depression has been identified.38 Low-
nancy are Listeria monocytogenes, 12-oz cup of coffee. However, birth de- income women with depressive
Toxoplasma gondii, Brucella species, Sal- fects research indicates moderate or symptoms and life stressors are at risk
monella species, and Campylobacter high amounts of beverages and foods for low-prenatal diet quality, so
July 2014 Volume 114 Number 7 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1101
FROM THE ACADEMY
intensive dietary intervention before 9. Academy of Nutrition and Dietetics Evi- www.cdc.gov/ncbddd/folicacid/recomme
dence Analysis Library. Gestational diabetes ndations.html. Accessed September 24,
and during pregnancy may be needed
evidence-based nutrition practice guide- 2012.
to promote optimal health.39 The risk line. http://andevidencelibrary.com/topic. 23. Correa A, Gilboa SM, Botto LD, et al. Lack
of maternal and infant mortality and cfm?cat¼3733. Accessed December 6, 2013. of periconceptional vitamins or supple-
pregnancy-related complications can 10. Seely EW, Ecker J. Chronic hypertension ments that contain folic acid and diabetes
be reduced with increased access to in pregnancy. N Engl J Med. 2011;365(5): mellitus-associated birth defects. Am J
439-446. Obstet Gynecol. 2012;206(3):218.e1-e13.
quality interconception care.
11. Tobias DK, Zhang C, Chavarro J, et al. 24. Hollis BW, Johnson D, Hulsey TC,
Prepregnancy adherence to dietary pat- Ebeling M, Wagner CL. Vitamin D sup-
terns and lower risk of gestational dia- plementation during pregnancy: Double-
CONCLUSIONS betes. Am J Clin Nutr. 2012;96(2):289-295. blind, randomized clinical trial of safety
Pregnancy has been regarded as a 12. Krakowiak P, Walker CK, Bremer AA, and effectiveness. J Bone Miner Res.
et al. Maternal metabolic conditions 2011;26(10):2341-2357.
maternal phase with requisite additional and risk for autism and other neuro-
nutritional requirements; mounting ev- 25. Thorne-Lyman A, Fawzi WW. Vitamin D
developmental disorders. Pediatrics.
during pregnancy and maternal, neonatal
idence suggests that the prenatal period 2012;129(5):e1121-e1128.
and infant health outcomes: A systematic
constitutes a critical convergence of 13. Hanley B, Dijane J, Fewtrell M, et al. review and meta-analysis. Paediatr Peri-
Metabolic imprinting, programming and
short- and long-term factors affecting nat Epidemiol. 2012;26(suppl 1):75-90.
epigenetics—A review of present prior-
the lifelong health of mother and child. ities and future opportunities. Br J Nutr. 26. Institute of Medicine. Dietary Reference
The aim of prenatal nutrition is to sup- 2010;104(suppl 1):S1-S25. Intakes for calcium and vitamin D. 2010.
http://www.iom.edu/Reports/2010/Diet
port a healthy uterine environment for 14. McMillen IC, MacLaughlin SM, ary-Reference-Intakes-for-Calcium-and-
optimal fetal development while sup- Muhlhausler BS, Gentili S, Duffield JL,
Vitamin-D.aspx. Published November 30,
Morrison JL. Developmental origins of
porting maternal health.5 The ideal pre- 2010. Accessed September 4, 2012.
adult health and disease: The role of per-
natal diet should limit overconsumption iconceptional and foetal nutrition. Basic 27. Caudill MA. Pre- and postnatal health:
for the mother and prevent undernutri- Clin Pharmacol Toxicol. 2008;102(2):82-89. Evidence of increased choline needs. J Am
Diet Assoc. 2010;110(8):1198-1206.
tion for the fetus5; a healthy lifestyle in- 15. US Department of Health and Human
28. Hacker AN, Fung EB, King JC. Role of cal-
cludes regular physical activity and Services. Physical activity for women
cium during pregnancy: Maternal and
during pregnancy and the postpartum
avoidance of harmful practices. period. In: 2008 Physical Activity Guide- fetal needs. Nutr Rev. 2012;70(7):397-409.
lines for Americans. Washington, DC: 29. Obican SG, Jahnke GD, Soldin OP,
Office of Disease Prevention & Health Scialli AR. Teratology public affairs com-
References Promotion; 2008:41-42. http://www. mittee position paper: Iodine deficiency
1. Academy of Nutrition and Dietetics. Prac- health.gov/paguidelines/guidelines/default. in pregnancy. Birth Defects Res.
tice Paper of the Academy of Nutrition and aspx. Accessed September 25, 2012. 2012;94(part A):677-682.
Dietetics: Nutrition and lifestyle for a
16. Deierlein AL, Siega-Riz AM, Evenson KR. 30. Stagnaro-Green A, Abalovich M,
healthy pregnancy outcome. http://www.
Physical activity during pregnancy and Alexander E, et al. Guidelines of the
eatright.org/members/practicepapers/.
risk of hyperglycemia. J Womens Health. American Thyroid Association for the
Published July 1, 2014. Accessed May 22,
2012;21(7):769-775. diagnosis and management of thyroid
2014.
17. Ruchat SM, Davenport MH, Giroux I, et al. disease during pregnancy and post-
2. Centers for Disease Control and Preven- partum. Thyroid. 2011;21(10):1081-1125.
Nutrition and exercise reduce excessive
tion. Division of Birth Defects and Devel-
weight gain in normal-weight pregnant 31. Swanson C, Zimmermann M, Skeaff S,
opmental Disabilities. Birth defects.
women. Med Sci Sports Exerc. 2012;44(8): et al. Summary of an NIH workshop to
http://www.cdc.gov/ncbddd/birthdefects/
1419-1426. identify research needs to improve the
index.html. Accessed October 4, 2012.
18. Gautam CS, Saha L, Sekhri K, Saha PK. Iron monitoring of iodine status in the United
3. Centers for Disease Control and Preven- States and to inform the DRI. J Nutr.
deficiency in pregnancy and the ratio-
tion. FastStats: Births and natality. http:// 2012;142(6):1175S-1185S.
nality of iron supplements prescribed
www.cdc.gov/nchs/fastats/births.htm.
during pregnancy. Medscape J Med. 32. Dean J, Kendall P. Food safety during preg-
Accessed October 4, 2012.
2008;10(12):283-288. http://www.ncbi. nancy. 2012;9.372. Colorado State Univer-
4. Rasmussen KM, Yaktine AL, eds. Weight nlm.nih.gov/pmc/articles/PMC264404/. sity Extension. Food and Nutrition Series.
Gain During Pregnancy: Reexamining Accessed October 3, 2012. http://www.ext.colostate.edu/pubs/food
the Guidelines. Washington, DC: National nut/09372.pdf. Accessed December 5, 2012.
19. Khalafallah AA, Dennis AE. Iron deficiency
Academies Press; 2009. http://www.
anaemia in pregnancy and postpartum: 33. Sagiv SK, Thurston SW, Bellinger DC,
nap.edu/openbook.php?record_id¼12584
Pathophysiology and effect of oral versus Amarasiriwardena C, Korrick SA. Prenatal
&page¼R1. Accessed March 19, 2014.
intravenous iron therapy [published on- exposure to mercury and fish consump-
5. Shapira N. Prenatal nutrition: A critical line June 26, 2012]. J Pregnancy. 2012; tion during pregnancy and attention-
window of opportunity for mother and 2012:630519. http://dx.doi.org/10.1155/ deficit/hyperactivity disorder-related
child. Womens Health. 2008;4(6):639-656. 2012/630519. behavior in children. Arch Pediatr Adolesc
6. Vesco KK, Dietz PM, Rizzo J, et al. Exces- 20. US Department of Agriculture, US Med. 2012;166(12):1123-1131.
sive gestational weight gain and post- Department of Health and Human Ser- 34. Academy of Nutrition and Dietetics Evi-
partum weight retention among obese vices. 2010 US Dietary Guidelines Advi- dence Analysis Library. Pregnancy and
women. Obstet Gynecol. 2009;114(5): sory Committee. Part D. Section 2: nutrition—Non-nutritive sweeteners.
1069-1075. Nutrient adequacy. In: Report of the Di- http://andevidencelibrary.com/evidence.
7. Academy of Nutrition and Dietetics. Po- etary Guidelines Advisory Committee on the cfm?evidence_summary_id¼250587.
sition of the Academy of Nutrition and Dietary Guidelines for Americans, 2010. Accessed December 3, 2013.
Dietetics and American Society for Nutri- 7th ed. Washington, DC: US Government 35. American College of Obstetrics and Gy-
tion: Obesity, reproduction, and preg- Printing Office; 2010: D2-38. necology. ACOG Committee opinion no.
nancy outcomes. J Am Diet Assoc. 21. US Department of Agriculture, US 462: Moderate caffeine consumption
2009;109(5):918-927. Department of Health and Human Ser- during pregnancy. Obstet Gynecol.
8. Centers for Disease Control and Preven- vices. Dietary Guidelines for Americans, 2010;116(2 Pt 1):467-468.
tion. 2011 National Diabetes fact sheet: 2010. 7th ed. Washington, DC: US Gov- 36. Brent RL, Christian MS, Diener RM. Eval-
Gestational diabetes in the United States. ernment Printing Office; 2010. uation of the reproductive and develop-
http://www.cdc.gov/diabetes/pubs/estim 22. Centers for Disease Control and Preven- mental risks of caffeine. Birth Defects Res
ates11.htm. Accessed November 11, 2012. tion. Folic acid: Recommendations. http:// (Part B). 2011;92(2):152-187.
1102 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS July 2014 Volume 114 Number 7
FROM THE ACADEMY
37. Institute of Medicine. Dietary reference 38. Shapiro GD, Fraser WD, Séguin JR. 39. Fowles ER, Stang J, Bryant M, Kim SH.
intakes for water, potassium, sodium, Emerging risk factors for postpartum Stress, depression, social support, and
chloride, and sulfate. http://www.nap.edu/ depression: Serotonin transporter eating habits reduce diet quality in the
openbook.php?record_id¼10925&page¼ genotype and omega-3 fatty acid sta- first trimester in low-income women:
151. Published 2005. Accessed October 21, tus. Can J Psychiatry. 2012;57(11): A pilot study. J Acad Nutr Diet. 2012;
2012. 704-712. 112(10):1619-1625.
This Academy of Nutrition and Dietetics position was adopted by the House of Delegates Leadership Team on May 3, 2002 and reaffirmed on
June 11, 2006 and September 9, 2010. This position is in effect until December 31, 2018. Requests to use portions of the position or republish in
its entirety must be directed to the Academy at journal@eatright.org.
Authors: Sandra B. Procter, PhD, RD/LD, Kansas State University, Manhattan, KS; Christina G. Campbell, PhD, RD, Iowa State University, Ames, IA
(Lead Author).
Reviewers: Jeanne Blankenship, MS, RD (Academy Policy Initiatives & Advocacy, Washington, DC); Quality Management Committee (Melissa N.
Church, MS, RD, LD, Chickasaw Nutrition-Get Fresh! Program, Oklahoma City, OK); Sharon Denny, MS, RD (Academy Knowledge Center, Chicago,
IL); Public Health dietetics practice group (DPG) (Kathryn Hillstrom, EdD, RD, CDE, California State University, Los Angeles, CA); Vegetarian
Nutrition DPG (Reed Mangels, PhD, RD, LDN, FADA. University of Massachusetts, Amherst); Kathleen Pellechia, RD (US Department of Agriculture,
WIC Works Resource System, Beltsville, MD); Julie A. Reeder, PhD, MPH, CHES (State of Oregon WIC Program, Portland, OR); Tamara Schryver, PhD,
MS, RD (TJS, Communications LLC, Minneapolis, MN); Alison Steiber, PhD, RD (Academy Research & Strategic Business Development, Chicago, IL);
Women’s Health DPG (Laurie Tansman, MS, RD, CDN, Mount Sinai Medical Center, New York, NY).
Academy Positions Committee Workgroup: Cathy L. Fagen, MA, RD (Chair) (Long Beach Memorial Medical Center, Long Beach, CA); Ainsley M.
Malone, MS, RD, CNSC, LD (Mount Carmel West Hospital, Columbus, OH); Jamie Stang, PhD, MPH, RD, LN (Content Advisor) (University of
Minnesota, Minneapolis, MN).
We thank the reviewers for their many constructive comments and suggestions. The reviewers were not asked to endorse this position or the
supporting paper.
July 2014 Volume 114 Number 7 JOURNAL OF THE ACADEMY OF NUTRITION AND DIETETICS 1103
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22 EXHIBIT C
23
24
25
26
27
28
Page 7 of 9
nutrients
Article
Impact of Second Trimester Maternal Dietary Intake
on Gestational Weight Gain and Neonatal
Birth Weight
Malshani L. Pathirathna 1,2, *, Kayoko Sekijima 1 , Mieko Sadakata 1 , Naoshi Fujiwara 3 ,
Yoshiyuki Muramatsu 1 and Kuruppu M.S. Wimalasiri 4
1 Department of Nursing, Graduate School of Health Sciences, Niigata University, 2-746 Asahimachi-dori,
Chuo-ku, Niigata 951-8518, Japan; sekijima@clg.niigata-u.ac.jp (K.S.); atom@clg.niigata-u.ac.jp (M.S.);
murayosi@clg.niigata-u.ac.jp (Y.M.)
2 Department of Nursing, Faculty of Allied Health Sciences, University of Peradeniya,
Peradeniya 20400, Sri Lanka
3 Department of Medical Technology, Graduate School of Health Sciences, Niigata University,
2-746 Asahimachi-dori, Chuo-ku, Niigata 951-8518, Japan; fujiwaranaoshi@gmail.com
4 Department of Food Science and Technology, Faculty of Agriculture, University of Peradeniya,
Peradeniya 20400, Sri Lanka; swarnaw@pdn.ac.lk
* Correspondence: malshanilakshika@gmail.com or malshanilakshika@pdn.ac.lk; Tel.: +81-70-3604-4661
Received: 26 May 2017; Accepted: 14 June 2017; Published: 17 June 2017
Abstract: Poor maternal nutrition is a major contributor to the high incidence of low birth weight
deliveries in developing countries. This study aimed to assess the impact of second trimester
maternal dietary intake on gestational weight gain and neonatal birth weight. A longitudinal study
was conducted in a tertiary care hospital in Sri Lanka. Participants were 141 pregnant women at
18–24 weeks gestation who were followed up until delivery. Maternal dietary intake was assessed
using a validated Food Frequency Questionnaire at 21.1 ± 1.8 gestational weeks. Gestational weight
gain was examined at the end of 28 weeks gestation and at the end of pregnancy. Energy and nutrient
intakes were calculated using NutriSurvey 2007 (EBISpro, Willstaett, Germany) nutrient analysis
software, modified for Sri Lankan foods. The mean total gestational weight gain of women with low
carbohydrate intake (229–429 g/day) was 2.2 kg less than that of women with moderate carbohydrate
intake (430–629 g/day) (95% confidence interval (CI) 0.428–4.083 kg; p = 0.016). Similarly, babies
of women with low carbohydrate intake were 312 g lighter compared with those of women with a
moderate carbohydrate intake (95% CI 91–534 g; p = 0.006). Our results suggest that second trimester
maternal carbohydrate intake has significant impacts on total gestational weight gain and neonatal
birth weight.
Keywords: low birth weight; maternal diet; macronutrient; gestational weight gain; nutrition;
Sri Lanka
1. Introduction
Despite a consistent decline in maternal and infant mortality, Sri Lanka continues to experience
crucial health problems among pregnant women, infants, and children. A significant problem is the
high percentage of low birth weight (LBW) deliveries. LBW is defined as weight at birth less than
2500 g regardless of gestational age. According to 2012 national health statistics, there were 55,557 LBW
deliveries in Sri Lanka, accounting for 16.3% of total live births [1].
Perinatal complications associated with LBW are mostly attributed to fetal prematurity, but may
also result from intrauterine growth restriction [2]. LBW increases the risk of infant mortality [3],
infectious diseases, inhibited growth, and inhibited cognitive development. Children with LBW
Nutrients 2017, 9, 627; doi:10.3390/nu9060627 www.mdpi.com/journal/nutrients

Nutrients 2017, 9, 627 2 of 12
are also more likely to suffer from chronic illnesses in later life [4]. When overall LBW rates are
examined, Sri Lanka performs better than other countries in the region that have a similar income level.
However, multiple causative factors are associated with Sri Lanka’s low mean birth weight, including
preterm deliveries, pregnancy complications, and increased interest in elective cesarean sections before
40 weeks gestation.
Maternal malnutrition is the main contributing factor underlying high LBW percentages in many
developing countries [5], and plays a major role in both maternal and child health. Poor maternal
nutrition has been related to adverse pregnancy outcomes. However, this association is complex,
influenced by many intrinsic and extrinsic factors, and often results from socio-cultural and behavioral
factors. Body mass index (BMI) and gestational weight gain (GWG) are indicators of maternal nutrition.
Understanding the association between maternal nutrition and neonatal birth weight may inform
development of nutritional interventions that improve neonatal birth weight to within normal limits,
improve long-term quality of life, and reduce the healthcare burden. Early fetal nutrition is considered
the most important factor in child health, even before birth. If a woman is undernourished in early
pregnancy, the fetal metabolism will be altered as fetal adaption to the undernourished utero-placental
environment takes place. Ultimately, this slows the overall fetal growth rate and increases the risk
for LBW. However, there are lack of human studies that clarify the best timing for optimum nutrition
during early pregnancy. In addition, during the first trimester of pregnancy, almost all women
experience loss of appetite, nausea and vomiting, making it difficult to obtain sufficient nutrition.
Therefore, second trimester maternal nutrition can be considered as an important turning point for
both the mother and fetus.
Sri Lankans have a unique dietary pattern. The staple meal is large serving of rice accompanied
by different vegetable, egg, meat or fish side dishes cooked with spices and (most often) coconut milk.
This rice and curry meal is commonly consumed as lunch, although it may also form breakfast and
dinner, depending on personal preferences and factors such as economic status. Traditional morning
and evening meals usually comprise a starchy staple such as string hoppers, hoppers, pittu or bread with
one or two curries. To date, no studies on the relationships among maternal dietary intake, gestational
weight gain, and neonatal birth weight have been conducted in Sri Lanka. This study aimed to assess
the influence of second trimester maternal dietary intake on GWG and neonatal birth weight, and to
explore the relationships among these three factors.
2. Materials and Methods
2.1. Design, Setting and Participants

This study used a longitudinal design, and was conducted in antenatal clinics at the General
Hospital Kurunegala, Sri Lanka from October 2015 to June 2016. A convenience sample of pregnant
women at 18–24 weeks gestation was included and they were followed up until delivery. Initial
recruitment consisted of 150 pregnant women.
2.2. Procedure
Women were invited to participate in the study through a recruitment poster displayed in
the hospital’s outpatient department. Pregnant women who were interested in participating were
requested to inform the researcher directly, in person or via telephone. Written information sheets
were provided to all interested women, including an explanation of the study purpose, data collection
methods, time taken for data collection and confidentiality of personal information.
Exclusion criteria were risk factors according to obstetric history (e.g., miscarriages/abortions,
multiple fetuses, pregnancy-induced hypertension, and gestational diabetes mellitus) and medical
history (e.g., psychiatric disorder or long term cardiac, renal, lung or gastrointestinal disease).
Impending neonates with a 5-minute Apgar score less than 5, and women who expected to deliver
their baby at another hospital were also excluded. Based on these criteria, there were seven exclusions
Nutrients 2017, 9, 627 3 of 12
by the researcher (spontaneous abortion n = 2, multiple fetuses identified at the 20-week ultra sound
scan n = 2, and maternal desire to deliver at another hospital n = 3) and two personal withdrawals,
resulting in nine exclusions between the recruitment process and the end of pregnancy.
Neonatal data were collected from hospital records after participating women had delivered
their babies. The researcher failed to find neonatal data for 14 neonates because of confusion with
hospital delivery registry records, as many similar names made it difficult to accurately locate study
participants. One neonate with a 5-minute Apgar score less than 5 was excluded at the end of data
collection. Data were collected for 126 neonates (Figure S1).
Socioeconomic and demographic data were collected by the researcher through interviews using
a pre-designed and pre-tested questionnaire. Maternal body weight and height were measured
using standard scales. Pre-pregnancy BMI was calculated by pre-pregnancy weight in kg (weight
at the first antenatal clinic visit, usually 6–8 weeks gestation) divided by height in m2 . Women’s
weight at their first antenatal clinic visit was obtained from individual pregnancy cards. Height was
measured at the time of study recruitment. World Health Organization (WHO) international BMI
cut-off points were used for BMI group categorization [6]. The Institute of Medicine (IOM) 2009
Re-examined Guidelines [7] were used to define total GWG categories: 12.5–18 kg for underweight
women, 11.5–16 kg for women of normal weight, 7–11 kg for overweight women, and 5–9 kg for obese
women. Maternal GWG was measured at two points: at the end of 28 weeks gestation and at the end
of pregnancy. GWG up to 28 weeks gestation was calculated by taking the positive difference between
the pre-pregnancy weight and that measured at 28 weeks gestation. Total GWG was calculated by
subtracting pre-pregnancy weight from delivery weight (38.8 ± 1.5 gestational weeks). Delivery
weight was obtained from each woman’s hospital records.
Participants’ dietary intake was assessed using a Food Frequency Questionnaire (FFQ) that was
developed and validated for Sri Lankan adults, including women [8,9]. Thriposha, a supplement of
pulses provided to pregnant women, was included in the FFQ under the pulses group. This is a nutrient
supplement pack that is given to all pregnant women through community maternity clinics to combat
protein, energy, and micronutrient deficiencies. It is a pre-cooked and ready-to-eat cereal legume-based
food. In total, 100 g of Thriposha contains 401.8 kcal of energy, 61.9 g of carbohydrate, 20.0 g of protein,
7.8 g of fat, 1700 IU of vitamin A, and 18 mg of iron. It is recommended to consume 50 g of Thriposha
each day during pregnancy. Participants were asked to complete the FFQ once in the second trimester
(21.1 ± 1.8 gestational weeks). The FFQ was self-administered, with support from the researcher
provided where necessary. Questions focused on the women’s usual dietary intake in the second
trimester. Food photographs were included in the questionnaire to facilitate understanding of portion
sizes. Dietary assessment aids were day-to-day standard household measurements (e.g., plate, cup,
glass and spoons of different sizes). Energy and nutrient intakes were calculated using NutriSurvey
2007 (EBISpro, Willstaett, Germany) nutrient analysis software, after modification for individual Sri
Lankan food items and recipes.
2.2.1. Nutrient Analysis Software Modification for Sri Lankan Food

Modification of the nutrient analysis software involved adding single food items to the original
software using information from the food composition tables of Sri Lanka [10], and the United States
Department of Agriculture (USDA) nutrient database [11]. It should be noted there is no updated Sri
Lankan nutritional database up to date. Macronutrient and micronutrient values for single food items
in the original software and those from the USDA database were changed to reflect local food using the
food composition tables of Sri Lanka. Available nutrient data from food packaging were used for local
food items such as processed foods, cookies and snacks. A standard local recipe book was used for
curry/mixed dishes [12]. All recipes were checked for validity by consulting a convenience sample of
10 participants from the study group, and the recipes were modified accordingly. To estimate portion
sizes for curry/mixed dishes, a procedure was adapted from a previous study [13] (Figure S2):
Nutrients 2017, 9, 627 4 of 12
• For curry/mixed dishes, ingredients were weighed to the nearest 1 g of edible portion using a
standard kitchen weighing scale (Tanita, no. 1155).
• Dishes were cooked according to the validated recipes.
• The final products were measured using standard household measurement utensils.
Recipes (curry/mixed dishes) were entered in to the modified software by entering single food
items according to the validated recipes. For each recipe, a cooking method was applied from the
available options in the software to approximate weight loss due to water evaporation from different
food preparation methods. The software then automatically calculated and standardized the nutrient
composition for 100 g of final product. One cup was considered 150 mL and one glass 200 mL.
The portion size of each food item in the FFQ was also measured and the software was modified
as necessary. After the final software modification, the FFQ was entered in the software with the
option of changing frequency of consumption to daily, weekly or monthly. Finally, participants’ FFQ
data were entered into the software and estimated energy, carbohydrate, and protein intakes per day
were calculated.
2.2.2. Calculation of Estimated Energy Requirement (EER)

The EER for the second trimester of pregnancy for each woman was calculated based on IOM
guidelines [14]:
EER (second trimester) = non-pregnant EER + 340 kcal (1)
EER (non-pregnant) = 354 (6.91 ⇥ age (years)) + physical activity coefficient ⇥

(2)
(9.36 ⇥ weight (kg) + 726 ⇥ height (m))
The value of active physical level (1.27) for women aged 19 years and older [14] was used as the
physical activity coefficient. All the pregnant women reported that they were engaging in normal
day-to-day activities and there were no women with prescribed activity limitations.
2.2.3. Recommended Dietary Allowance (RDA) of Protein

The RDA of protein during the second trimester for each woman was calculated as 1.1 g/kg/day,
using the women’s pre-pregnancy weight [15].
2.3. Data Analysis

All data were entered and rechecked in Microsoft Excel 2007. Data for energy and macronutrient
intakes were transferred from NutriSurvey 2007 to Minitab version 17 for statistical analysis.
Descriptive statistics were expressed as mean ± standard deviation (SD). Correlations between neonatal
birth weight, gestational weight gain, dietary intake and other continuous variables were evaluated
with Pearson’s correlation analysis. To test the effects of different levels of energy and macronutrient
intakes on gestational weight gain and neonatal birth weight, participating women were divided
into two groups by energy intake (1 = daily energy intake less than the EER; 2 = daily energy intake
greater than or equal to EER) and two protein intake groups (1 = less than the RDA of protein intake;
2 = greater than or equal to RDA of protein intake) based on estimated dietary intakes. As all women
were above the RDA for carbohydrate, they were grouped in three categories with same class interval:
1 = 229–429 g/day; 2 = 430–629 g/day; and 3 = 630–829 g/day. One-way analysis of variance (ANOVA)
was used to examine the differences in means of gestational weight gain, energy and nutrient intakes
and neonatal birth weight. Two general linear models were constructed to define independent factors
associated with total gestational weight gain and neonatal birth weight, controlling for confounders.
Total energy intake was not included in the general linear models to avoid multicollinearity, as the total
energy intake represents energy from carbohydrate, protein, and fat. For each analysis, 95% confidence
intervals were calculated, and p < 0.05 was considered statistically significant.
Nutrients 2017, 9, 627 5 of 12
2.4. Ethics
Informed written consent was obtained from all the subjects before they participated in the
study. The study was conducted in compliance with the principles of the Declaration of Helsinki, and
the protocol was approved by the Ethical Review Committee, Graduate School of Health Sciences,
Niigata University, Japan (No: 125); the Ethical Review Committee, Faculty of Allied Health Sciences,
University of Peradeniya, Sri Lanka; and the Institutional Ethical Review Committee, Teaching Hospital
Kurunegala, Sri Lanka (No: ERC/2015/06).
3. Results
3.1. Participants’Characteristics
The final sample included 126 healthy newborns. Of these, 22 (17.4%) were LBW babies. In total,
20.6% women were underweight when they became pregnant and 56.7% had normal BMI. Participants’
characteristics are summarized in Table 1.
Table 1. Participants’ characteristics (n = 141).
Variable Mean (SD) n (%)

Age (years) 28.8 (6.2) -
Gestational age (weeks) a 38.8 (1.5) -
Pre-pregnancy weight (kg) b 51.9 (10.2) -
Pre-pregnancy BMI (kg/m2 ) 22.1 (4.3) -
Parity
Primiparous - 47 (33.3)
Multiparous - 94 (66.7)
Presence of hyperemesis gravidarum c - 23 (16.3)
History of miscarriage or abortion - 38 (26.9)
History of LBW delivery - 28 (19.9)
Total energy intake (kcal/day) 2921.5 (687.7) -
Carbohydrate intake (g/day) 532.7 (133.8) -
Total protein intake (g/day) 71.2 (16.8) -
Fat intake (g/day) 45.8 (16.9) -
an = 126. b Measured at the first antenatal clinic visit, usually around 6–8 weeks gestation. c During first trimester.
BMI: body mass index; LBW: low birth weight. SD: standard deviation.
Of the 141 pregnant women, 138 returned a completed FFQ. Two incomplete questionnaires were
excluded from the final analysis. The mean EER was 2224.4 ± 138.6 kcal/day; 26 (19.1%) participants
were below the EER, whereas almost all women (100%) had a carbohydrate intake above the RDA
(equal to 175 g/day). The mean RDA of protein was 57.3 ± 11.3 g/day, and 81.6% women had a protein
intake above the RDA (calculated on an individual basis). Of the total energy intake, 72.8% was from
carbohydrate and 9.8% was from protein. The protein intake mainly comprised plant protein (75.7%
of total protein intake), with only 12.8% of the total protein intake being animal protein. The mean
animal protein intake was 9.1 g/day and the mean plant protein was intake 53.9 g/day. Women with
the lowest level of monthly household income showed the lowest total energy, carbohydrate, and
protein intakes, but this did not reach statistical significance. Women living in rural areas showed the
highest carbohydrate intake, but this did not significantly differ from women in urban and sub-urban
areas (Table S1).
3.2. Correlation among Maternal Parameters, GWG and Neonatal Birth Weight
The correlation between total GWG and neonatal birth weight was 0.194 (p = 0.046). When
the analysis was repeated with pre-pregnancy BMI and gestational age fixed, there was a moderate
positive correlation between total GWG and neonatal birth weight (r = 0.302, p = 0.002). Maternal height
(r = 0.283), pre-pregnancy weight (r = 0.247), and pre-pregnancy BMI (r = 0.340) were significantly
Nutrients 2017, 9, 627 6 of 12
correlated with total GWG (p < 0.05). Gestational age (r = 0.315), pre-pregnancy weight (r = 0.234),
and pre-pregnancy BMI (r = 0.187) were significantly correlated with neonatal birth weight (p < 0.05).
A strong positive correlation was observed between GWG up to 28 weeks gestation and total GWG
(r = 0.812, p < 0.001). None of the studied dietary factors (total energy, carbohydrate, protein, and fat
intakes) showed a significant correlation with total GWG or neonatal birth weight.
3.3. Effects of Maternal and Neonatal Characteristics on GWG

A univariate analysis was performed to test the effects of maternal and neonatal characteristics on
GWG. Of the factors assessed, only the pre-pregnancy BMI category showed a significant relationship
with GWG up to 28 weeks gestation (p < 0.001), whereas both the pre-pregnancy BMI category
(p < 0.001) and dietary protein intake category (p < 0.05) were significantly associated with total GWG
(Table 2).
The fitted general linear model (Table 3) showed significant effects of education level (p < 0.05),
pre-pregnancy BMI category (p < 0.001), and daily carbohydrate intake category (p < 0.05) on total GWG.
The respective effects of fat intake, parity, and dietary protein intake category were not significant
(p > 0.05). Women with an underweight pre-pregnancy BMI showed a higher mean total GWG
(3.8 kg) compared with women with normal pre-pregnancy BMI (p < 0.001). Similarly, women with
underweight pre-pregnancy BMI showed a 4.8 kg higher mean total GWG compared with women
who were overweight (p < 0.001) and a 5.4 kg higher gain than those who were obese (p < 0.001).
The mean total GWG for women with a carbohydrate intake of 229–429 kcal/day was 2.2 kg below
that of women with a carbohydrate intake of 430–629 kcal/day (p = 0.016).
3.4. Effects of Maternal and Neonatal Characteristics on Neonatal Birth Weight

The univariate analysis revealed that five of 12 factors (area of residence, history of LBW delivery,
total GWG category, total energy intake category, and carbohydrate intake category) were significantly
associated with neonatal birth weight (p < 0.05) (Table 4).
The general linear model (R2 (adjusted) = 13.3%) showed that pre-pregnancy BMI (p = 0.025),
gestational age (p < 0.001), parity (p < 0.05), and carbohydrate intake category (p < 0.05) had significant
effects on neonatal birth weight. The babies of urban mothers were 258 g lighter than those of rural
mothers; However, the difference was not significant (p = 0.062). The mean birth weight of babies of
primiparous mothers was 187.4 g below that of babies of multiparous mothers (p = 0.046). For the
carbohydrate intake category, the mean neonatal birth weight for category 1 (229–429 kcal/day) was
312 g below the mean birth weight of those in category 2 (430–629 kcal/day) (p = 0.006). Income level
had no significant effect on neonatal birth weight (p > 0.05) (Table 5).
3.5. Effects of Supplemental Foods on GWG and Neonatal Birth Weight

The mean Thriposha intake was 34.8 ± 29.3 g/day (range 0–200 g/day); 11.8% (16/136) of women
reported they were not consuming the supplement at all. There was no significant difference in total
GWG between women who consumed 0–49 g/day of Thriposha (9.0 ± 3.7 kg) and those who consumed
50 g/day (9.8 ± 3.9 kg) (p = 0.271). The mean neonatal birth weight was 2859.1 ± 488.4 g in mothers
who consumed 0–49 g/day of Thriposha, and 2906.3 ± 523.1 g in those who consumed 50 g/day.
There was no significant difference in mean neonatal birth weight between the two Thriposha groups
(F = 0.26, p = 0.608).
Nutrients 2017, 9, 627 7 of 12
Table 2. Gestational weight gain (GWG) by maternal and neonatal characteristics (ANOVA).
GWG up to 28 Weeks (n = 105) Total GWG (n = 119)
Variable Sub-category n 95% CI p-Value 95% CI p-Value
Mean (SD) Mean (SD)
All - 6.33 (2.85) 5.78–6.89 - 9.30 (3.72) 8.63–9.98 -
None/primary 16 5.72 (2.80) 4.20–7.24 7.34 (3.72) 5.51–9.17
Education level Secondary 99 6.46 (2.92) 5.85–7.07 0.519 9.64 (3.71) 8.90–10.37 0.072
Higher 2 5.13 (0.11) 1.84–8.42 8.60 (0.57) 3.42–13.78
<9000 5 6.67 (2.57) 3.33–10.00 9.08 (4.91) 5.75–12.41
9000–13,999 20 6.09 (2.62) 4.73–7.46 8.30 (3.44) 6.64–9.96
Income level
14,000–19,999 32 6.27 (3.35) 5.26–7.27 0.978 9.42 (3.23) 8.11–10.74 0.474
(LKR (Sri Lankan rupee))
20,000–31,999 45 6.52 (2.77) 5.60–7.45 9.23 (4.21) 8.12–10.34
32,000 16 6.04 (2.44) 4.30–7.79 10.70 (3.31) 8.77–12.61
Urban 9 4.53 (3.49) 2.40–6.66 7.44 (3.93) 4.97–9.92
Area of residence Sub-urban 52 6.68 (3.33) 5.86–7.50 0.167 9.28 (3.99) 8.25–10.31 0.306
Rural 55 6.18 (2.15) 5.37–7.00 9.52 (3.46) 8.52–10.52
Yes 22 5.33 (3.03) 4.14–6.52 8.04 (3.36) 6.48–9.59
History of LBW deliveries 0.062 0.077
No 97 6.60 (2.76) 5.99–7.21 9.60 (3.76) 8.85–10.33
History of miscarriage Yes 32 6.47 (2.75) 5.43–7.51 8.87 (3.75) 7.56–10.17
0.759 0.442
or abortion No 87 6.28 (2.91) 5.62–6.94 9.46 (3.77) 8.67–10.25
Yes 20 6.34 (3.21) 5.03–7.65 9.04 (4.20) 7.37–10.70
Hyperemesis gravidarum a 0.978 0.730
No 97 6.32 (2.80) 5.70–6.93 9.36 (3.67) 8.60–10.12
Primiparous 40 6.89 (2.39) 5.86–7.92 10.00(4.00) 8.84–11.16
Parity 0.208 0.146
Multiparous 79 6.11 (3.00) 5.46–6.76 8.95 (3.54) 8.12–9.77
Underweight 23 8.43 (2.94) 1 7.24–9.62 12.00(4.07) 1 10.58–13.43
Pre-pregnancy BMI Normal 69 6.18 (2.26) 2 5.51–6.85 9.10 (3.12) 2 8.27–9.92
category b <0.001 * <0.001 *
Overweight 20 5.73 (3.38) 2,3 4.60–6.86 7.89 (3.81) 2 6.37–9.42
Obese 7 2.76 (0.97) 3 0.44–5.08 6.50 (3.30) 2 3.92–9.08
<EER 21 5.57 (3.04) 4.32–6.82 8.05 (3.44) 6.44–9.67
Energy intake c 0.188 0.083
EER 93 6.51 (2.85) 5.87–7.16 9.62 (3.80) 8.86–10.39
229–429 25 5.97 (3.08) 4.78–7.15 8.03 (3.60) 6.56–9.50
Carbohydrate intake
430–629 58 6.41 (2.99) 5.60–7.23 0.798 10.00 (3.68) 9.03–10.96 0.089
(g/day)
630–829 31 6.45 (2.63) 5.29–7.61 9.16 (3.88) 7.84–10.49
<RDA 21 5.49 (3.21) 4.24–6.74 7.81 (3.56) 6.20–9.41
0.141
9.68 (3.74) 8.92–10.44
0.039 *
Male 53 6.44 (2.78) 5.63–7.25 9.23 (3.31) 8.25–10.21
Sex of the newborn 0.552 0.976
Female 53 6.10 (2.75) 5.28–6.91 9.21 (3.86) 8.23–10.19
a During first trimester. b Based on WHO international BMI cut-off values [6]. c EER for the pregnancy second trimester based on IOM 2009 guidelines [14]. d RDA based on IOM Food and
Nutrition Board 2002/2005 guidelines [15]. 1,2,3 Values with the same superscript Arabic numeral do not represent a significance difference. Compared using one-way ANOVA followed by
Tukey’s post-hoc test. “n” column represents the subject numbers corresponding to total GWG. * p < 0.05. RDA: recommended dietary allowance; WHO: World Health Organization; EER:
estimated energy requirement; CI: confidence interval; IOM: Institute of Medicine.
Nutrients 2017, 9, 627 8 of 12
Table 3. General linear model for total gestational weight gain (kg).

Constant 11.35 8.30–14.39 7.4 <0.001 *
Fat intake 0.04 0.08–0.00 1.77 0.08
Education (none/primary)—reference level
Education (secondary) 2.19 0.40–3.97 2.43 0.017 *
Education (higher) 1.29 6.24–3.65 0.52 0.605
Pre-pregnancy BMI category
(underweight)—reference level
Pre-pregnancy BMI category (normal) 3.84 5.49– 2.19 4.62 <0.001 *
Pre-pregnancy BMI category (overweight) 4.80 6.96– 2.64 4.41 <0.001 *
Pre-pregnancy BMI category (obese) 5.44 8.49– 2.40 3.55 0.001 *
Parity (multiparous) 0.93 2.27–0.41 1.38 0.171
Category of carbohydrate intake (430–629 g/day) 2.26 0.43–4.08 2.45 0.016 *
Category of carbohydrate intake (630–829 g/day) 1.60 0.49–3.7 1.52 0.132
Category of protein intake ( RDA) 0.42 1.68–2.52 0.4 0.691
R2 (adjusted) = 16.5%. n = 112. * p < 0.05.
Table 4. Neonatal birth weight by maternal and neonatal characteristics, ANOVA.
Birth Weight (g)

Mean (SD)
All - 126 2874.6 (497) 2787.0–2962.2 -
None/primary 20 2831 (428) 2612.5–3049.5
Education level Secondary 100 2867.6 (508) 2769.9–2965.3 0.629
Higher 4 3091 (391) 2603–3580
<9000 5 3130 (432) 2702–3558
9000–13,999 22 2702 (499) 2498–2906
Income level (LKR) 14,000–19,999 31 2965 (573) 2793–3136 0.093
20,000–31,999 50 2803.3 (454) 2667.9–2938.7
32,000 16 3036.9 (352.8) 2797.6–3276.2
Urban 13 2754 (519) 1,2 2492–3016
Area of residence Sub-urban 54 2747.1 (461.9) 1 2618.0–2875.7 0.011 *
Rural 57 3011.4 (481.9) 2 2886.3–3136.5
History of LBW Yes 24 2684.2 (418.2) 2486.1–2882.2
0.036 *
deliveries No 102 2919.4 (505.2) 2823.3–3015.4
History of miscarriage or Yes 33 2898.9 (507.6) 2727.1–3070.8
0.744
abortion No 93 2865.9 (495.7) 2763.5–2968.3
Hyperemesis Yes 20 2800.5 (366.5) 2582.4–3018.6
gravidarum a 0.499
No 104 2882.1 (512.5) 2786.4–2977.7
Primiparous 45 2770.1 (430.5) 2624.7–2915.5
Parity 0.079
Multiparous 81 2932.6 (523.9) 2824.2–3041.0
Underweight 27 2771.5 (459.8) 2523.2–2899.8
Pre-pregnancy BMI Normal 70 2912.5 (442.5) 2795.5–3029.5
category b 0.231
Overweight 22 2895 (673) 2687–3104
Obese 7 3059 (464) 2689–3428
Within recommended 32 2912.8 (539.7) 1,2 2740.8–3084.8
Total GWG category Below recommended 71 2863.9 (458.5) 1 2748.4–2979.3 0.042 *
Over recommended 3 3600 (721) 2 3038–4162
Energy intake c <EER 24 2692 (533) 2491–2892
0.039 *
EER 97 2927.8 (487) 2828.0–2037.6
Nutrients 2017, 9, 627 9 of 12
Table 4. Cont.
Birth Weight (g)

Mean (SD)
229–429 30 2686.7 (498.1) 1 2508.5–2864.9
Carbohydrate intake
430–629 64 2957.7 (486) 2 2853.7–3097.7 0.033
(g/day)
630–829 27 2872.2 (503.6) 1,2 2684.4–3060.1
<RDA 24 2913 (601) 2708–3117
0.733
Sex of the newborn Male 61 2852.9 (477.3) 2726.5–2979.2
0.637
Female 65 2894.9 (517.7) 2772.5–3017.3
a b c
During first trimester. Based on WHO international BMI cut-off values. EER for the pregnancy second trimester
based on IOM 2009 guidelines [14]. d RDA based on IOM Food and Nutrition Board 2002/2005 guidelines [15].
1,2,3 Values with the same superscript Arabic numeral do not represent a significance difference. Sample sizes vary
slightly because of missing data. Compared using one-way ANOVA followed by Tukey’s post-hoc test. * p < 0.05.
Table 5. General linear model for neonatal birth weight (g).

Constant 2011 4349–327 1.71 0.091
Pre-pregnancy BMI 23.7 3.0–44.4 2.27 0.025 *
Gestational age 102.1 46.7–157.5 3.65 <0.001 *
Fat intake 2.83 8.21–2.54 1.05 0.298
Income (<9000 LKR)—reference level
Income (9000–13,999 LKR) 209 679–262 0.88 0.381
Income (14,000–19,999 LKR) 43 495–410 0.19 0.852
Income (20,000–31,999 LKR) 74 523–374 0.33 0.743
Income ( 32,000 LKR) 156 322–634 0.65 0.518
Area of residence (urban)—reference level
Area of residence (sub-urban) 36 235–308 0.27 0.791
Area of residence (rural) 258 14–529 1.88 0.062
History of LBW deliveries (yes)—reference level
History of LBW deliveries (no) 209 6.0–424.0 1.93 0.057
Parity (multiparous) 187.4 3.2–371.6 2.02 0.046 *
Category of carbohydrate intake (430–629 g/day) 312 91–534 2.8 0.006 *
Category of carbohydrate intake (630–829 g/day) 237 44–517 1.67 0.097
Category of protein intake ( RDA) 66 326–194 0.51 0.615
R2 (adjusted) = 13.3%. n = 118. * p < 0.05.
4. Discussion
The present study showed that 17.4% of deliveries were LBW babies, which was slightly above
the national LBW rates for 2014 [16]. This might be because data for the present study were collected
in a large tertiary care hospital that included more complicated pregnancies. In the present study,
the average total GWG was 9.3 ± 3.7 kg, which was slightly below that observed for Sri Lankan
women in a previous study [17]. However, this mean total GWG was below that recommended for
underweight and normal BMI women. We found a moderate positive correlation between total GWG
and neonatal birth weight. Therefore, total GWG may be considered a good predictor of neonatal
birth weight, this emphasizes the importance of appropriate GWG when it is sought to prevent LBW
deliveries. As there was a strong positive correlation between total GWG and GWG up to 28 weeks
gestation, GWG up to 28 weeks can be used as a meaningful predictor to optimize individualized
Nutrients 2017, 9, 627 10 of 12
antenatal care for women who show low weight gain. Pre-pregnancy BMI also showed a significant
association with neonatal birth weight, indicating the importance of maternal nutrition at the time a
woman becomes pregnant.
In the present study, all participating women had a carbohydrate intake during pregnancy above
the RDA. The mean energy intake was significantly higher than previous studies [18,19]. However,
consuming a large serving of rice and/or other starchy staple in all three main meals and the daily
consumption of supplemental food with a higher energy value are central to this high energy intake.
In addition, the concept of “eating for two” during pregnancy still prevails in Sri Lanka, even though
it has no scientific basis. In the present study, 18.4% of women had a protein intake during pregnancy
below the RDA, with main protein supply being plant protein. Although protein from animal sources
is of greater nutritional value because it contains all essential amino acids, the animal protein intake
of women in our sample was low compared with Western countries [20]. This may be explained
by the vegetarian trend of the younger generation in Sri Lanka because of religious and cultural
influences and the poor economic status. There is no separate RDA for fat intake during pregnancy,
meaning that fat is recommended to be 25–35% of the total calorie intake, as for the general population.
Therefore, in an average 2200 kcal diet, approximately 550 kcal should be fat (approximately 60 g).
The present study showed a mean fat intake of 45.8 ± 16.9 g per day, which was moderately below the
general recommendation.
Our study revealed a significant relationship between second trimester carbohydrate intake and
neonatal birth weight. Godfrey et al. found that birth weight was inversely related to carbohydrate
intake in early pregnancy [20], which is inconsistent with our results. However, gestational age at
the time of nutrient assessment in Godfrey et al.’s study [20] was around 15 weeks gestation, which
differed from the present study, making it difficult to conduct a fair comparison. We found that
moderate carbohydrate intake was associated with both the total GWG and neonatal birth weight.
Godfrey et al. [20] suggested that high carbohydrate intake in early pregnancy suppressed placental
growth (and thus fetal growth), as did low dairy protein intake in late pregnancy. However, it was not
possible to compare our present results with those of Godfrey et al. because we focused on dietary
intake in the second trimester. No significant difference in any background characteristic was apparent
between the high and moderate carbohydrate-intake groups. Although statistical significance was
not attained, the moderate-intake group contained a higher proportion (54.7%) of wealthier women
(monthly household income 2000 LKR); this may partially explain the higher GWG and birth weight
in this group. The present study showed that babies of rural mothers were heavier than those of
urban mothers, although this did not reach the level of significance. This might be attributable to the
higher carbohydrate consumption of rural women compared with urban women (Table S1). No direct
relationship was observed between second trimester protein intake and neonatal birth weight, which is
consistent with results shown by Godfrey et al. [20]. Our univariate analysis showed that women with
a total energy intake below the EER delivered neonates with significantly lower mean birth weight
compared with women who were above the EER. Although the majority of women in our study were
above the EER and RDA for selected macronutrients, individual dietary analysis showed an imbalance
in meal habits, for example, a high proportion of carbohydrate and low amount of other important
nutrients. In particular, mean fat intake was relatively low. Findings of the current study can be
explained by the importance of carbohydrate intake for total GWG, and thereby neonatal birth weight.
Provision of the Thriposha supplement for pregnant women aims to maintain satisfactory GWG
and reduce the LBW percentage. Although this supplement is distributed free of charge, the present
study indicated that the mean Thriposha intake was below the recommended intake of 50 g/day.
In addition, 11.8% of women reported that they were not consuming the Thriposha supplement
even though they received it from the community clinics. This may be because the supplement
was consumed by the entire family rather than the pregnant woman. However, in the present study,
supplement consumption showed no significant effect on either GWG or neonatal birth weight. Further
Nutrients 2017, 9, 627 11 of 12
large scale, nationwide studies are recommended to evaluate the effects of supplements distributed
free of charge by community clinics.
4.1. Limitations
There was a notable data loss between recruitment and the end of data collection, resulting in
a smaller sample size than expected. In addition, using a FFQ to collect dietary intake data might
have resulted in over/under estimation of actual intake. Despite these limitations, the present study
provides the first local estimates of energy and macronutrient consumption among pregnant women
in Sri Lanka.
4.2. Implications
The Sri Lanka Ministry of Health recommends that the total GWG should be based on the IOM
2009 revised guidelines. Over 15% of women with inadequate GWG delivered LBW infants, indicating
that although the IOM guidelines were developed for Americans with larger body frames, they
were useful in our setting for identifying women at risk of delivering LBW infants. Culturally and
economically competent health care is required to allow Sri Lankan women to achieve a desirable GWG;
the emphasis should be on a healthy diet and regular weight monitoring. Nutritional education should
be integrated into first-trimester antenatal education sessions, with a focus on meals meeting the dietary
requirements of pregnancy and featuring a variety of foods from all food groups. Individual nutritional
counseling should be provided to women at risk, especially those with an underweight pre-pregnancy
BMI, those exhibiting low GWG, and those on poor diets. We also found that almost all women
consumed more than the RDA of carbohydrates during pregnancy. Thus, culturally appropriate
dietary recommendations relevant in the Sri Lankan context should be essential components of
strategies seeking to prevent low birth weight.
5. Conclusions
Second trimester maternal carbohydrate intake shows significant relationships with total GWG
and neonatal birth weight. Maintaining a moderate level of carbohydrate intake (430–629 g/day)
during the second trimester of pregnancy may promote favorable total GWG and neonatal birth weight
in the Sri Lankan context. Our results highlight the need for primary healthcare providers to be vigilant
in assessing maternal dietary intake during the second trimester. Individualized education should be
provided about good sources of carbohydrates in meals for pregnant women who show weight gain
below the recommended levels.
Supplementary Materials: The following are available online at www.mdpi.com/2072-6643/9/6/627/s1,

Figure S1: Process of data collection, Figure S2: Example estimation of portion sizes for curry dishes (bean
curry), Table S1: Second trimester maternal energy and macro-nutrient intake by maternal and neonatal
characteristics; ANOVA.
Acknowledgments: The authors thank all women who participated in this study, obstetricians, special grade
nursing officers, and the staff nurses of the antenatal clinics for their support of this study. This study received no
specific grant from any public, commercial, or not-for-profit sector.
Author Contributions: M.L.P., K.S., M.S., N.F. and K.M.S.W. contributed to the conception and design of this
study. M.L.P. conducted the survey. M.L.P. and K.S. performed the statistical analysis. M.L.P. wrote the paper.
K.S., M.S., N.F. Y.M. and K.M.S.W. critically reviewed the manuscript and supervised the whole study process.
All authors read and approved the final manuscript.
Conflicts of Interest: The authors declare no conflict of interest.
References
1. Ministry of Health. Annual Health Bulletin; Ministry of Health, Nutrition and Indigenous Medicine: Colombo,
Sri Lanka, 2012.
Nutrients 2017, 9, 627 12 of 12
2. Patterson, R.M.; Prihoda, T.J.; Gibbs, C.E.; Wood, R.C. Analysis of birth weight percentile as a predictor of
perinatal outcome. Obstet. Gynecol. 1986, 68, 459–463. [PubMed]
3. Joseph, K.S.; Kramer, M.S. Recent trends in infant mortality rates and proportions of low-birth-weight live
births in Canada. CMAJ 1997, 15, 535–541.
4. Barker, D.J.P. Mothers, Babies and Health in Later Life, 2nd ed.; Churchill Livingstone: Edinburgh, UK, 1998.
5. Ramakrishnan, U. Nutrition and low birth weight: From research to practice. Am. J. Clin. Nutr. 2004, 79,
17–21. [PubMed]
6. World Health Organization. Global Database on Body Mass Index. Available online: http://apps.who.int/bmi/
index.jsp?introPage=intro-3.html (accessed on 10 April 2016).
7. Institute of Medicine. Weight Gain during Pregnancy: Reexamining the Guidelines (2009). Available
online: http://www.nationalacademies.org/hmd/~/media/Files/Report%20Files/2009/Weight-Gain-
During-Pregnancy-Reexamining-the-Guidelines/Report%20Brief%20-%20Weight%20Gain%20During%
20Pregnancy.pdf (accessed on 10 March 2017).
8. Jayawardena, R.; Swaminathan, S.; Byrne, N.M.; Soares, M.J.; Katulanda, P.; Hills, A.P. Development of a
food frequency questionnaire for Sri Lankan adults. Nutr. J. 2012, 11, 63. [CrossRef] [PubMed]
9. Jayawardena, R.; Byrne, N.M.; Soares, M.J.; Katulanda, P.; Hills, A.P. Validity of a food frequency
questionnaire to assess nutritional intake among Sri Lankan adults. SpringerPlus 2016, 5, 162. [CrossRef]
[PubMed]
10. Wimalasiri, W.R. Composition of Sri Lankan foods; Faculty of Dental Sciences, University of Peradeniya:
Peradeniya, Sri Lanka, 1999.
11. United States Department of Agriculture. USDA Food Composition Databases. Available online:
https://ndb.nal.usda.gov/ndb/ (accessed on 5 May 2016).
12. Dissanayake, C. Ceylon Cookery, 6th ed.; Stamford Lake (Pvt) Ltd.: Pannipitiya, Sri Lanka, 2005.
13. Jayawardena, R.; Thennakoon, S.; Byrne, N.; Soares, M.; Katulanda, P.; Hills, A. Energy and nutrient intakes
among Sri Lankan adults. Int. Arch. Med. 2014, 7, 34. [CrossRef] [PubMed]
14. Institute of Medicine. Weight Gain during Pregnancy: Reexamining the Guidelines (2009). Available online:
https://www.ncbi.nlm.nih.gov/books/NBK32812/ (accessed on 13 June 2016).
15. Institute of Medicine. Dietary Reference Intakes for Energy, Carbohydrate, Fiber, Fat, Fatty Acids, Cholesterol,
Protein and Amino Acids. Available online: http://www.nationalacademies.org/hmd/Home/Global/
News%20Announcements/%20DRI (accessed on 3 June 2016).
16. Ministry of Health, Nutrition and Indigenous Medicine. Annual Health Bulletin; Ministry of Health, Nutrition
and Indigenous Medicine: Colombo, Sri Lanka, 2014.
17. Abeysena, C.; Jayawardana, P. Body mass index and gestational weight gain in two selected Medical Officer
of Health areas in the Gampaha District. J. Coll. Community Physicians Sri Lanka 2011, 16, 30–34. [CrossRef]
18. Mathews, F.; Yudkin, P.; Neil, A. Influence of maternal nutrition on outcome of pregnancy: Prospective
cohort study. BMJ 1999, 319, 339–343. [CrossRef] [PubMed]
19. Kubota, K.; Itoh, H.; Tasaka, M.; Naito, H.; Fukuoka, Y.; Muramatsu, K.; Kobayashi, K.; Sugihara, K.;
Kanayama, N.; Hamamatsu Birth Cohort Study Team. Changes of maternal dietary intake, body weight
and fetal growth throughout pregnancy in pregnant Japanese women. J. Obstet. Gynaecol. Res. 2013, 39,
1383–1390. [CrossRef] [PubMed]
20. Godfrey, K.; Robinson, S.; Barker, D.J.P.; Osmond, C.; Cox, V. Maternal nutrition in early and late pregnancy
in relation to placental and fetal growth. BMJ 1996, 312, 410–414. [CrossRef] [PubMed]
© 2017 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access
article distributed under the terms and conditions of the Creative Commons Attribution
(CC BY) license (http://creativecommons.org/licenses/by/4.0/).
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
EXHIBIT D
22
23
24
25
26
27
28
Page 8 of 9
HHS Public Access
Author manuscript
Birth Defects Res A Clin Mol Teratol. Author manuscript; available in PMC 2017 November
Author Manuscript
01.
Published in final edited form as:
Birth Defects Res A Clin Mol Teratol. 2016 November ; 106(11): 940–949. doi:10.1002/bdra.23570.
Association between Antibiotic Use among Pregnant Women

with Urinary Tract Infections in the First Trimester and Birth
Defects, National Birth Defects Prevention Study 1997 to 2011
Elizabeth C. Ailes1,*, Suzanne M. Gilboa1, Simerpal K. Gill2, Cheryl S. Broussard1, Krista S.
Crider1, Robert J. Berry1, Tonia C. Carter3, Charlotte A. Hobbs4, Julia D. Interrante1,5,
Jennita Reefhuis1, and The National Birth Defects Prevention Study
Author Manuscript
1NationalCenter on Birth Defects and Developmental Disabilities, Centers for Disease Control
and Prevention, Atlanta, Georgia
2Duchesnay Inc, Blainville, Quebec, Canada
3Marshfield Clinic, Marshfield, Wisconsin
4College of Medicine, University of Arkansas for Medical Sciences, Little Rock, Arkansas
5Oak Ridge Institute for Science and Education, Oak Ridge, Tennessee
Abstract
Background—Previous studies noted associations between birth defects and some antibiotics
(e.g., nitrofurantoin, sulfonamides) but not others (e.g., penicillins). It is unclear if previous
Author Manuscript
findings were due to antibiotic use, infections, or chance. To control for potential confounding by
indication, we examined associations between antibiotic use and birth defects, among women
reporting urinary tract infections (UTIs).
Methods—The National Birth Defects Prevention Study is a multi-site, population-based case-

control study. Case infants/fetuses have any of over 30 major birth defects and controls are live-
born infants without major birth defects. We analyzed pregnancies from 1997 to 2011 to estimate
the association between maternally reported periconceptional (month before conception through
the third month of pregnancy) use of nitrofurantoin, trimethoprim-sulfamethoxazole, or
cephalosporins and specific birth defects, among women with periconceptional UTIs. Women with
periconceptional UTIs who reported penicillin use served as the comparator.
Results—Periconceptional UTIs were reported by 7.8% (2029/26,068) of case and 6.7%

Author Manuscript
(686/10,198) of control mothers. Most (68.2% of case, 66.6% of control mothers) also reported
antibiotic use. Among 608 case and 231 control mothers reporting at least one periconceptional
UTI and certain antibiotic use, compared with penicillin, nitrofurantoin use was associated with
oral clefts in the offspring (adjusted odds ratio, 1.97 [95% confidence interval, 1.10–3.53]),
trimethoprim-sulfamethoxazole use with esophageal atresia (5.31 [1.39–20.24]) and diaphragmatic
*
Correspondence to: Elizabeth Ailes, National Center on Birth Defects and Developmental Disabilities, Centers for Disease Control
and Prevention, 4770 Buford Hwy NE, MS E-86, Atlanta, GA 30341. eailes@cdc.gov.
None of the authors have any conflicts of interest to disclose.
Ailes et al. Page 2
hernia (5.09 [1.20–21.69]), and cephalosporin use with anorectal atresia/stenosis (5.01 [1.34–
Author Manuscript
18.76]).
Conclusion—Periconceptional exposure to some antibiotics might increase the risk for certain
birth defects. However, because individual birth defects are rare, absolute risks should drive
treatment decisions.
Keywords
birth defects; antibiotic; cephalosporin; nitrofurantoin; penicillin; trimethoprim-sulfamethoxazole;
urinary tract infection
Introduction
Urinary tract infections (UTIs) are common bacterial infections in pregnancy; 1 to 4% of
Author Manuscript
pregnant women experience acute cystitis and 4 to 10% have asymptomatic bacteriuria
(Foxman, 2002). If left untreated, UTIs can progress to pyelonephritis, which has been
associated with preterm labor and low birth weight (Delzell and Lefevre, 2000). Even
uncomplicated UTIs have been associated with intrauterine growth restriction and preterm
labor, and maternal UTIs have been associated with early onset sepsis and UTIs in the
neonate (Schieve et al., 1994; Bhutta and Yusuf, 1997; Delzell and Lefevre, 2000;
Emamghorashi et al., 2012). To prevent these adverse pregnancy outcomes, pregnant women
are screened for UTIs during early pregnancy while in prenatal care (U.S. Preventive
Services Task Force, 2008).
According to Infectious Diseases Society of America guidelines, nonpregnant women with

uncomplicated UTIs should be treated with nitrofurantoin or trimethoprim-sulfamethoxazole
Author Manuscript
(Gupta et al., 2011). However, Crider and colleagues’ analysis of 1997 to 2003 National
Birth Defects Prevention Study (NBDPS) data raised concern that these commonly
prescribed antibiotic treatments for UTIs might be associated with specific birth defects
(Crider et al., 2009). In 2011, the American College of Obstetricians and Gynecologists
(ACOG) issued a Committee Opinion stating that sulfonamides and nitrofurantoin can be
prescribed in the first trimester of pregnancy if other antimicrobial therapies are deemed
inappropriate (American College of Obstetricians and Gynecologists Committee on
Obstetric Practice, 2011).
More recent epidemiologic studies have found trimethoprim to be associated with both
miscarriage and birth defects, but trimethoprim-sulfonamides not to be associated with birth
defects (Andersen et al., 2013a, 2013b; Hansen et al., 2016). Subsequent studies of
nitrofurantoin use in pregnancy have not found it to be associated with birth defects,
Author Manuscript
although these studies may have been underpowered to detect associations with specific
birth defect types (Goldberg et al., 2013; Nordeng et al., 2013). Despite rising rates of
antibiotic resistance in some geographic areas (Mazzulli, 2012), penicillins are still
frequently used to treat UTIs in pregnant women (Car, 2006; Petersen et al., 2010), likely
because of their purported relative safety (Crider et al., 2009).
Birth Defects Res A Clin Mol Teratol. Author manuscript; available in PMC 2017 November 01.
Ailes et al. Page 3
Previous studies have typically examined the association of antibiotic classes with specific
Author Manuscript
birth defects regardless of indication (Crider et al., 2009; Andersen et al., 2013b; Goldberg
et al., 2013; Nordeng et al., 2013). However, a potential confounding factor in these studies
is the indication for antibiotic use, as maternal urinary tract, genital, and sexually transmitted
infections have been associated with birth defects in previous studies (Cleves et al., 2008;
Feldkamp et al., 2008). To minimize confounding by indication, we examined the
association between maternally reported periconceptional antibiotic use and birth defects,
among women with UTIs, using 1997 to 2011 NBDPS data, by comparing the odds of birth
defects among women reporting periconceptional nitrofurantoin, trimethoprim-
sulfamethoxazole, or cephalosporin use to women reporting periconceptional penicillin use.
Materials and Methods

NBDPS
Author Manuscript
The NBDPS is a population-based, multi-site, case–control study of risk factors for select
major birth defects (Reefhuis et al., 2015). Our analysis included case and control infants or
fetuses born on or after October 1, 1997, with estimated dates of delivery (EDD) on or
before December 31, 2011. Cases included infants/fetuses with 1 of more than 30 major
birth defects identified through birth defects surveillance systems in the states of Arkansas
(1998–2011), Iowa, New Jersey (1998–2002), and Utah (2003–2011), or select counties in
California, Georgia, Massachusetts, North Carolina (2003–2011), New York, and Texas.
Cases were live-born infants (all sites), stillbirths of ≥20 weeks gestation (nine sites), and
elective terminations (eight sites). Infants/fetuses with major chromosomal abnormalities,
single-gene disorders, and birth defects with known etiology were excluded. Controls were
live-born infants without major birth defects selected annually from vital or hospital
discharge records from the same catchment areas as case infants/fetuses.
Author Manuscript
Case and control mothers were interviewed by telephone in English or Spanish using a
computer-based questionnaire 6 weeks to 24 months after the EDD. Information was
obtained on maternal demographics; nutritional, behavioral, and occupational exposures;
maternal disease including infections; and over-the-counter and prescription medications,
vitamins, and supplements. Clinical data regarding the birth defects were abstracted from
medical records, and birth defects were classified by clinicians and clinical geneticists using
previously described NBDPS classification criteria (Rasmussen et al., 2003; Botto et al.,
2007). During 1997 to 2011, the participation rate was 67% for case mothers and 64% for
control mothers. All subjects consented to participation and the NBDPS was approved by
the Institutional Review Boards at the Centers for Disease Control and Prevention and all
participating institutions.
Author Manuscript
EXPOSURE DEFINITION
During the structured interview, women were asked whether they experienced any of the
following: a kidney, bladder, or UTI from 3 months before conception through the end of
pregnancy. Subsequent questions ascertained whether the infection was diagnosed by a
doctor, when the infection occurred (pregnancy month[s] or trimester[s]), any over-the-
counter and prescription medication(s) taken for the infection, and the name, dates taken,
Ailes et al. Page 4
duration, and frequency of medication(s) used. Later in the questionnaire, women were
Author Manuscript
asked if they took any medications not already discussed and to provide the name, dates
taken, duration, and frequency of use, although not the specific indication for use.
Exposure was defined as maternal report of any antibiotic use in the periconceptional period
(defined as the month before conception through the third month of pregnancy). Medications
were coded using the Slone Drug Dictionary (Boston University), collapsing specific
antibiotics into their respective antibiotic classes (Werler et al., 2011). We created mutually
exclusive categories of antibiotics: penicillin, nitrofurantoin, trimethoprim-
sulfamethoxazole, or cephalosporins, as well as other, unknown or use of multiple
antibiotics. Women reporting a periconceptional doctor-diagnosed UTI but no
periconceptional use of oral antibiotics, including those using only topical antibiotics or
another type of oral anti-infective medication (e.g., antivirals, antifungals), were excluded.
Author Manuscript
ANALYSES
We excluded women with type 1 or type 2 diabetes, multiple or missing number of
gestations, pelvic inflammatory disease, fever during the month before conception through
the third month of pregnancy, missing UTI information, UTIs not diagnosed by a doctor, and
missing antibiotic exposure dates (Fig. 1). We restricted our analysis of the association
between specific antibiotics and birth defects to defect types with at least 100 total or at least
four exposed and four unexposed cases to the four specific antibiotic classes under
investigation (penicillins, nitrofurantoin, trimethoprim-sulfamethoxazole, and
cephalosporins). Due to limited sample sizes or interpretability, we did not assess other
antibiotics, multiple antibiotics or unknown types of antibiotics and birth defects
associations. We selected potential covariates because of their association with birth defects
in previous studies. These included: maternal age, race/ethnicity (non-Hispanic white vs.
Author Manuscript
other), education (<high school vs. ≥high school), prepregnancy body mass index (body
mass index ≥ 30 kg/m2 vs. < 30 kg/m2), use of any folic acid supplements in the month
before through the first month of pregnancy, periconceptional smoking, periconceptional
alcohol use, EDD year (1997–2003, 2004–2011), time from EDD to interview (≤ 6 months,
7–12 months, 13–24 months), and NBDPS study site.
We constructed multivariable conditional logistic regression models to estimate adjusted

odds ratios and their corresponding 95% confidence intervals. We matched on NBDPS study
site, and among the covariates of interest noted above, adjusted for maternal race/ethnicity
and body mass index because these factors varied across antibiotic exposures among control
mothers (Table 1). Given the potentially serious consequences of untreated UTIs in
pregnancy, a comparison population of women with UTIs unexposed to antibiotics would be
Author Manuscript
inappropriate. Thus, to have a disease-matched comparison group, our comparator was

women with penicillin-treated periconceptional UTIs, given the relative lack of associations
between penicillin and birth defects (Crider et al., 2009). This approach has been used in
other studies (Berkovitch et al., 2000; Hansen et al., 2016). We performed one post hoc
sensitivity analysis: we removed any cases or controls with a first degree family member
(i.e., mother, father, full sibling, or previous pregnancy) with a type of birth defect
ascertained in NBDPS.
Ailes et al. Page 5
Results
Author Manuscript
Periconceptional UTIs were common: 7.8% (2029/26,068) of case and 6.7% (686/10,198) of
control mothers reported having at least one doctor-diagnosed, fever-free UTI in the month
before conception through the third month of pregnancy. The majority (68.2% of case,
66.6% of control mothers) also reported taking an antibiotic (Fig. 1). The most commonly
reported antibiotics used periconceptionally were penicillins and nitrofurantoin; however,
more than one-quarter of women could not recall the specific antibiotic taken (Fig. 2).
Comparing the early (1997–2003) and later (2004–2011) eras of the NBDPS, “other”
antibiotics accounted for a greater proportion of use in the latter period of the study among
both case and control mothers.
To determine which potential confounders were associated with exposure, we compared

potential confounders between control mothers with and without the relevant exposure (i.e.,
Author Manuscript
compared women with UTIs and use of nitrofurantoin, trimethoprim-sulfamethoxazole, or

cephalosporin antibiotics periconceptionally to women with UTIs and penicillin antibiotic
use periconceptionally). Factors significantly associated with periconceptional
nitrofurantoin, trimethoprim-sulfamethoxazole, or cephalosporin use rather than penicillin
use among control mothers with UTIs included maternal race/ethnicity and body mass index
(Table 1).
After restricting to phenotypes with at least 100 total, or 4 exposed and 4 “unexposed” (i.e.,
penicillin-exposed) cases, 608 cases and 231 control mothers with periconceptional UTIs
and penicillin, nitrofurantoin, trimethoprim-sulfamethoxazole, or cephalosporin antibiotic
use were eligible for analysis (Fig. 1). Compared with women with UTIs and penicillin use
periconceptionally, women with periconceptional UTIs and nitrofurantoin use had
Author Manuscript
significantly greater odds of having offspring with oral clefts (adjusted odds ratio, 1.97 [95%
confidence interval, 1.10– 3.53]) and borderline significant for the subtype of cleft lip with
or without cleft palate (1.90 [1.00–3.63]), in particular (Table 2). Those reporting UTIs and
trimethoprim-sulfamethoxazole use periconceptionally had significantly greater odds of
having offspring with esophageal atresia (5.31 [1.39–20.24]) and diaphragmatic hernia (5.09
[1.20–21.69]; and those reporting cephalosporin use had significantly greater odds of having
offspring with anorectal atresia/stenosis (5.01 [1.34–18.76]). Results from the posthoc
analyses were similar to those of the primary analysis. Exclusion of the 23 cases and 7
controls with a first degree family member with a birth defect type ascertained in NBDPS
tended to strengthen the associations observed in the primary analysis and the association
between cephalosporins and cleft lip with or without cleft palate became significant (5.67
[1.67–18.76]; other data not shown).
Author Manuscript
Discussion
Reports of UTIs in early pregnancy were common; 8% of case and 7% of control mothers
reported at least one UTI from the month before conception through the third month of
pregnancy. To minimize potential confounding by indication, we restricted our analysis to
women reporting fever-free, doctor-diagnosed UTIs, and antibiotic use periconceptionally.
Overall, compared with women reporting penicillin use, we found significant associations
Ailes et al. Page 6
between periconceptional use of nitrofurantoin and cleft lip with or without cleft palate;
Author Manuscript
trimethoprim-sulfamethoxazole and esophageal atresia and diaphragmatic hernia; and

cephalosporins and anorectal atresia. These associations persisted in a sensitivity analysis
excluding cases and controls with a family history of birth defects.
Among women with UTIs, periconceptional nitrofurantoin exposure was more commonly
reported than penicillin use by mothers of infants/fetuses with cleft lip with or without cleft
palate. Previous studies have noted associations for nitrofurantoin with cleft palate, rectal/
anal atresia/stenosis, NTDs, nonchromosomal heart defects, and a borderline significant
association with hypospadias (Czeizel et al., 2001a; Kallen and Otterblad Olausson, 2003).
Two previous retrospective cohort studies did not find associations between nitrofurantoin
and major malformations, cardiac defects, or cleft lip or palate, although their samples sizes
may have limited their ability to detect significant associations (Goldberg et al., 2013;
Nordeng et al., 2013).
Author Manuscript
Among women with a UTI, periconceptional trimethoprim-sulfamethoxazole exposure was

more commonly reported than penicillin use by mothers of infants/ fetuses with esophageal
atresia and with diaphragmatic hernia. Trimethoprim is a dihydrofolate reductase inhibitor
that interferes with DNA synthesis (Schweitzer et al., 1990). A randomized clinical trial
found significantly lower levels of serum folate among a sample of healthy men taking a
seven day course of trimethoprim compared with those taking placebo (Meidahl Petersen et
al., 2016). Dihydrofolate reductase inhibitor use in early pregnancy has also been associated
with a variety of reproductive outcomes, including miscarriage and selected birth defects
(Andersen et al., 2013a, 2013b; Hernandez-Diaz et al., 2000, 2001). Using data from 1976 to
1998, Hernandez-Diaz et al. (2001) showed that the association between dihydrofolate
reductase inhibitors and NTDs diminished with increasing folic acid supplement use
Author Manuscript
(Hernandez-Diaz et al., 2001).
In our study, when compared with women reporting periconceptional penicillin use, women
reporting cephalosporin use periconceptionally had significantly elevated odds for anorectal
atresia/stenosis. Few published studies describe the associations between cephalosporins and
birth defects. An unpublished analysis of Michigan Medicaid data from 1985 to 1992
showed higher than expected numbers of congenital defects, including heart defects, among
approximately 4000 women taking cephalosporins during their first trimester (Briggs et al.,
2011). A larger Hungarian case–control study again found no association with major
malformations overall but did find a significantly elevated association with cardiac defects
(Czeizel et al., 2001b).
It is worth noting that we saw several additional elevated but nonsignificant associations. Of
Author Manuscript
the 18 birth defects with calculable odds ratios for nitrofurantoin, there were an additional
four elevated (adjusted odds ratio > 1.50) associations: hypoplastic left heart syndrome, cleft
palate, esophageal atresia, and craniosynostosis. For trimethoprim-sulfamethoxazole, of the
16 birth defects with calculable odds ratios, there were an additional eight nonsignificantly
elevated odds ratios for coarctation of the aorta, neural tube defects, oral clefts overall as
well as the specific subtypes of cleft palate and cleft lip with and without cleft palate,
anorectal atresia/stenosis, and limb deficiency, including the subtype of transverse limb
Ailes et al. Page 7
deficiency. For cephalosporins, there were nine calculable odds ratio; nonsignificantly
Author Manuscript
elevated odds ratios were observed for pulmonary valve stenosis, perimembranous
ventricular septal defects, neural tube defects overall, and the subtype of spina bifida, and
oral clefts overall, as well as the subtype of cleft lip with or without cleft palate. Only three
“protective” odds ratios (adjusted odds ratio < 0.66) were observed: two for nitrofurantoin
(with coarctation of the aorta and gastroschisis) and one for trimethoprim-sulfamethoxazole
(with gastroschisis).
A major limitation of our analysis is the potential for exposure misclassification because of
maternal self-report of both UTIs and antibiotic exposure, ascertained 6 weeks to 2 years
after the EDD, with no verification from other sources (e.g., medical or pharmacy records).
To minimize misclassification of UTI status, we restricted our analysis to women reporting
doctor-diagnosed infections. Antibiotic information was more problematic; close to one-
quarter of women could not recall the name of the antibiotic taken. We included medication
Author Manuscript
exposures reported in the month before conception in addition to those in the first trimester
of pregnancy to allow for the potential misclassification of timing of antibiotic use. Recall of
medication timing is often better for short-term use, the circumstance for most of our
antibiotic exposures, and is also improved when women are prompted with specific
medication names, a feature of our study as well (Mitchell et al., 1986; Radin et al., 2013).
Finally, while indication for antibiotic use was not available for all women in our study, 96%
of case and 97% of control mothers with periconceptional UTIs indicated that their UTI was
the indication for their antibiotic use.
There are several considerations in the interpretation of our findings. First, because some
birth defects had elevated odds ratios for multiple antibiotic classes, it remains possible that
confounding by indication still impacted our results due to lack of additional information on
Author Manuscript
UTI severity, the specific type of infection (as women could have had either a kidney,
bladder, or UTI), and the bacterial etiology of the UTI. Second, given the large number (n =
43) of associations examined, multiple comparisons might have impacted our results
because approximately two statistically significant associations would be expected by
chance alone; however, we do not know which of the four associations we observed are
more likely to be due to chance alone. Third, although using conditional logistic regression
allowed us to account for differences by NBDPS study site, this modeling method has been
shown to potentially lead to biased estimates (Greenland et al., 2000). Fourth, by only
including information from women for whom information was available, we assumed the
missing information was missing completely at random and, therefore, the impact of
informative missingness on our results is unknown.
Strengths of this study include the use of multi-site, population-based data with stringent
Author Manuscript
birth defect classification criteria. Clinical geneticists or pediatric cardiology specialists

familiar with the diagnosis of birth defects reviewed all potential NBDPS cases and
confirmed that they met study eligibility requirements (Rasmussen et al., 2003; Botto et al.,
2007). The study’s large size also allowed for the assessment of associations between
specific antibiotic exposures and more refined categories of birth defects. Furthermore, our
study size and design allowed us to minimize potential confounding by indication, by
Ailes et al. Page 8
restricting our primary analysis to case and control mothers reporting UTIs
Author Manuscript
periconceptionally.
Our findings support continued caution with periconceptional use of trimethoprim-

sulfamethoxazole, cephalosporins, and nitrofurantoin. However, the body of literature on the
topic of maternal antibiotic use and birth defects is limited and has not produced consistent
findings. Furthermore, the birth defects found to be associated with antibiotics in this study
(cleft lip with or without cleft palate, esophageal atresia, diaphragmatic hernia, and anorectal
atresia/stenosis) are rare, with a birth prevalence of approximately 0.99%, 0.22%, 0.27%,
and 0.41%, respectively, reported by active birth defects surveillance programs during the
period 2008 to 2012 (Mai et al., 2015). By comparison, nearly 1 in 10 women experience a
UTI during early pregnancy and as the maternal and fetal complications associated with
untreated UTIs are common and serious, women should seek treatment if they suspect they
might have a UTI. Therefore, although some of the associations observed in this study were
Author Manuscript
relatively strong, it is important to note that individual birth defects are rare, and potential
absolute risks of birth defects associated with antibiotic use remain small (~ 1%) compared
with the risks of complications due to untreated UTIs in pregnancy.
Acknowledgments
Supported by an appointment to the Research Participation Program at the National Center on Birth Defects and
Developmental Disabilities, Centers for Disease Control and Prevention (CDC), administered by the Oak Ridge
Institute for Science and Education through an interagency agreement between the U.S. Department of Energy and
CDC.
The findings and conclusions in this report are those of the authors and do not necessarily represent the official
position of the Centers for Disease Control and Prevention.
Author Manuscript
References
American College of Obstetricians and Gynecologists Committee on Obstetric Practice. Committee
Opinion No. 494: sulfonamides, nitrofurantoin, and risk of birth defects. Obstet Gynecol. 2011;
117:1484–1485. [PubMed: 21606771]
Andersen JT, Petersen M, Jimenez-Solem E, et al. Trimethoprim use in early pregnancy and the risk of
miscarriage: a register-based nationwide cohort study. Epidemiol Infect. 2013a; 141:1749–1755.
[PubMed: 23010291]
Andersen JT, Petersen M, Jimenez-Solem E, et al. Trimethoprim use prior to pregnancy and the risk of
congenital malformation: a register-based nationwide cohort study. Obstet Gynecol Int. 2013b;
2013:364526. [PubMed: 23476656]
Berkovitch M, Segal-Socher I, Greenberg R, et al. First trimester exposure to cefuroxime: a
prospective cohort study. Br J Clin Pharmacol. 2000; 50:161–165. [PubMed: 10930968]
Bhutta ZA, Yusuf K. Early-onset neonatal sepsis in Pakistan: a case control study of risk factors in a
birth cohort. Am J Perinatol. 1997; 14:577–581. [PubMed: 9394171]
Author Manuscript
Botto LD, Lin AE, Riehle-Colarusso T, et al. Seeking causes: classifying and evaluating congenital
heart defects in etiologic studies. Birth Defects Res A Clin Mol Teratol. 2007; 79:714–727.
[PubMed: 17729292]
Briggs, G.; Freeman, R.; Yaffe, S. Drugs in pregnancy and lactation: a reference guide to fetal and
neonatal risk. 9th. Philadelphia, PA: Lippincott, Williams, and Wilkins; 2011.
Car J. Urinary tract infections in women: diagnosis and management in primary care. BMJ. 2006;
332:94–97. [PubMed: 16410583]
Cleves MA, Malik S, Yang S, et al. Maternal urinary tract infections and selected cardiovascular
malformations. Birth Defects Res A Clin Mol Teratol. 2008; 82:464–473. [PubMed: 18452156]
Ailes et al. Page 9
Crider KS, Cleves MA, Reefhuis J, et al. Antibacterial medication use during pregnancy and risk of
birth defects: National Birth Defects Prevention Study. Arch Pediatr Adolesc Med. 2009; 163:978–
Author Manuscript
985. [PubMed: 19884587]

Czeizel AE, Rockenbauer M, Sorensen HT, Olsen J. Nitrofurantoin and congenital abnormalities. Eur J
Obstet Gynecol Reprod Biol. 2001a; 95:119–126. [PubMed: 11267733]
Czeizel AE, Rockenbauer M, Sorensen HT, Olsen J. Use of cephalosporins during pregnancy and in
the presence of congenital abnormalities: a population-based, case-control study. Am J Obstet
Gynecol. 2001b; 184:1289–1296. [PubMed: 11349204]
Delzell JE Jr, Lefevre ML. Urinary tract infections during pregnancy. Am Fam Physician. 2000;
61:713–721. [PubMed: 10695584]
Emamghorashi F, Mahmoodi N, Tagarod Z, Heydari ST. Maternal urinary tract infection as a risk
factor for neonatal urinary tract infection. Iran J Kidney Dis. 2012; 6:178–180. [PubMed:
22555480]
Feldkamp ML, Reefhuis J, Kucik J, et al. Case-control study of self reported genitourinary infections
and risk of gastroschisis: findings from the National Birth Defects Prevention Study, 1997–2003.
BMJ. 2008; 336:1420–1423. [PubMed: 18558640]
Author Manuscript
Foxman B. Epidemiology of urinary tract infections: incidence, morbidity, and economic costs. Am J
Med. 2002; 113(Suppl 1A):5S–13S.
Goldberg O, Koren G, Landau D, et al. Exposure to nitrofurantoin during the first trimester of
pregnancy and the risk for major malformations. J Clin Pharmacol. 2013; 53:991–995. [PubMed:
23873250]
Greenland S, Schwartzbaum JA, Finkle WD. Problems due to small samples and sparse data in
conditional logistic regression analysis. Am J Epidemiol. 2000; 151:531–539. [PubMed:
10707923]
Gupta K, Hooton TM, Naber KG, et al. International clinical practice guidelines for the treatment of
acute uncomplicated cystitis and pyelonephritis in women: a 2010 update by the Infectious
Diseases Society of America and the European Society for Microbiology and Infectious Diseases.
Clin Infect Dis. 2011; 52:e103–e120. [PubMed: 21292654]
Hansen C, Andrade SE, Freiman H, et al. Trimethoprim-sulfonamide use during the first trimester of
pregnancy and the risk of congenital anomalies. Pharmacoepidemiol Drug Saf. 2016; 25:170–178.
[PubMed: 26599424]
Author Manuscript
Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Folic acid antagonists during pregnancy
and the risk of birth defects. N Engl J Med. 2000; 343:1608–1614. [PubMed: 11096168]
Hernandez-Diaz S, Werler MM, Walker AM, Mitchell AA. Neural tube defects in relation to use of
folic acid antagonists during pregnancy. Am J Epidemiol. 2001; 153:961–968. [PubMed:
11384952]
Kallen BA, Otterblad Olausson P. Maternal drug use in early pregnancy and infant cardiovascular
defect. Reprod Toxicol. 2003; 17:255–261. [PubMed: 12759093]
Mai CT, Isenburg J, Langlois PH, et al. Population-based birth defects data in the United States, 2008
to 2012: presentation of state-specific data and descriptive brief on variability of prevalence. Birth
Defects Res A Clin Mol Teratol. 2015; 103:972–993. [PubMed: 26611917]
Mazzulli T. Diagnosis and management of simple and complicated urinary tract infections (UTIs). Can
J Urol. 2012; 19(Suppl 1):42–48. [PubMed: 23089347]
Meidahl Petersen K, Eplov K, Kjaer Nielsen T, et al. The effect of trimethoprim on serum folate levels
in humans: a randomized, double-blind, placebo-controlled trial. Am J Ther. 2016; 23:e382–e387.
Author Manuscript
[PubMed: 26566149]
Mitchell AA, Cottler LB, Shapiro S. Effect of questionnaire design on recall of drug exposure in
pregnancy. Am J Epidemiol. 1986; 123:670–676. [PubMed: 3953545]
Nordeng H, Lupattelli A, Romoren M, Koren G. Neonatal outcomes after gestational exposure to
nitrofurantoin. Obstet Gynecol. 2013; 121(Pt 1):306–313. [PubMed: 23344280]
Petersen I, Gilbert R, Evans S, et al. Oral antibiotic prescribing during pregnancy in primary care: UK
population-based study. J Antimicrob Chemother. 2010; 65:2238–2246. [PubMed: 20716554]
Radin RG, Mitchell AA, Werler MM. Predictors of recall certainty of dates of analgesic medication
use in pregnancy. Pharmacoepidemiol Drug Saf. 2013; 22:25–32. [PubMed: 22674801]
Ailes et al. Page 10
Rasmussen SA, Olney RS, Holmes LB, et al. Guidelines for case classification for the National Birth
Defects Prevention Study. Birth Defects Res A Clin Mol Teratol. 2003; 67:193–201. [PubMed:
Author Manuscript
12797461]
Reefhuis J, Gilboa SM, Anderka M, et al. The National Birth Defects Prevention Study: a review of the
methods. Birth Defects Res A Clin Mol Teratol. 2015; 103:656–669. [PubMed: 26033852]
Schieve LA, Handler A, Hershow R, et al. Urinary tract infection during pregnancy: its association
with maternal morbidity and perinatal outcome. Am J Public Health. 1994; 84:405–410. [PubMed:
8129056]
Schweitzer BI, Dicker AP, Bertino JR. Dihydrofolate reductase as a therapeutic target. FASEB J. 1990;
4:2441–2452. [PubMed: 2185970]
U. S. Preventive Services Task Force. Screening for asymptomatic bacteriuria in adults: U.S.
Preventive Services Task Force reaffirmation recommendation statement. Ann Intern Med. 2008;
149:43–47. [PubMed: 18591636]
Werler MM, Louik C, Mitchell AA. Case-control studies for identifying novel teratogens. Am J Med
Genet C Semin Med Genet. 2011; 157C:201–208. [PubMed: 21766434]
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
FIGURE 1.
Selection of subjects with UTIs and antibiotic use, National Birth Defects Prevention Study,
1997 to 2011. B1, month before conception; B3, 3 months before conception; EDD,
estimated date of delivery; P3, third month of pregnancy; P9, ninth month of
pregnancy. aAfter restricting to defects with 100 total or at least 4 cases exposed to and
unexposed to a specific antibiotic group (see the Materials and Methods section). bOnly
included in descriptive analysis.
Author Manuscript
Author Manuscript
Author Manuscript
Author Manuscript
FIGURE 2.
Proportion of women reporting each type of antibiotic, among women with UTIs and
antibiotic use in the month before conception through the third month of pregnancy, by Era,
Among Mothers of Cases (A), and Controls (B), National Birth Defects Prevention Study,
1997 to 2011.
Author Manuscript
TABLE 1
Maternal Characteristics of Controls Mothers Reporting Urinary Tract Infections and Antibiotic Use
Author Manuscript
Periconceptionallya National Birth Defects Prevention Study, 1997 to 2011
Exposed
(UTIs and nitrofurantoin,
trimethoprim-sulfamethoxazole Comparator
or cephalosporin use, (UTIs and penicillin use,
periconceptionally) periconceptionally)
(N = 111) (N = 120) p-Value
Maternal age (years)
<20 13 (12%) 17 (14%) 0.632
20–24 36 (32%) 37 (31%)
25–29 27 (24%) 36 (30%)
30–34 21 (19%) 21 (18%)
> = 35 14 (13%) 9 (8%)

Author Manuscript
Maternal non-Hispanic white race/ethnicity
Yes 76 (68%) 57 (48%) 0.001
No 35 (32%) 63 (53%)
Maternal pre-pregnancy body mass index > = 30 (kg/m2)
Yes 18 (16%) 33 (28%) 0.027
No 92 (83%) 82 (68%)
Missing 1 (1%) 5 (4%)
Maternal education >high school

Yes 67 (60%) 66 (55%) 0.440
No 42 (38%) 51 (43%)
Unknown 2 (2%) 3 (3%)

Author Manuscript
Any maternal smokinga
Yes 28 (25%) 35 (29%) 0.453
No 82 (74%) 82 (68%)
Unknown 1 (1%) 3 (3%)
Any maternal alcohola
Yes 46 (41%) 49 (41%) 0.992
No 64 (58%) 68 (57%)
Unknown 1 (1%) 3 (3%)
Any maternal folic acid useb
Yes 65 (59%) 58 (48%) 0.136
No 46 (41%) 61 (51%)
Author Manuscript
Unknown 0 (0%) 1 (1%)
Time from estimated date of delivery to interview (months)
≤6 55 (50%) 57 (48%) 0.836
7–12 38 (34%) 37 (31%)
Exposed
(UTIs and nitrofurantoin,
trimethoprim-sulfamethoxazole Comparator
Author Manuscript
or cephalosporin use, (UTIs and penicillin use,

periconceptionally) periconceptionally)
(N = 111) (N = 120) p-Value
13–24 17 (15%) 21 (18%)
Missing 1 (1%) 5 (4%)
Estimated date of delivery year
1997–2003 54 (49%) 57 (48%) 0.861
2004–2011 57 (51%) 63 (53%)
NBDPS study site
Arkansas 27 (24%) 30 (25%) 0.388
California 15 (14%) 12 (10%)
Georgia 11 (10%) 12 (10%)
Iowa 11 (10%) 11 (9%)

Author Manuscript
Massachusetts 9 (8%) 9 (8%)
New Jersey 2 (2%) 4 (3%)
New York 5 (5%) 17 (14%)
North Carolina 8 (7%) 9 (8%)
Texas 14 (13%) 12 (10%)
Utah 9 (8%) 4 (3%)
a
From the month before conception through the third month of pregnancy.
b
From the month before conception through the first month of pregnancy.
B1, month before conception; UTI, urinary tract infection; P3, third month of pregnancy.
Author Manuscript
Author Manuscript
Author Manuscript Author Manuscript Author Manuscript Author Manuscript
TABLE 2
Associations between Antibiotic Use and Birth Defectsa, among Women Reporting Urinary Tract Infections and Antibiotic Use Periconceptionallyb:
National Birth Defects Prevention Study, 1997 to 2011
Ailes et al.
Women with periconceptionalb UTIs and … use
Nitrofurantoin Trimethoprim-sulfamethoxazole Cephalosporins

Penicillin
N
Defecta (N) N cOR (95% CI) aOR(95% CI) N cOR (95% CI) aOR(95% CI) N cOR (95% CI) aOR(95% CI)
Controlsc 120 60 30 21
Heart Defects
Conotruncal defects
Tetralogy of Fallot (21) 10 6 1.20 (0.42,3.46) 1.12 (0.36,3.51) 3 2
Left ventricular outflow tract obstructions
Hypoplastic left heart syndrome (17) 6 6 2.00 (0.62,6.46) 1.95 (0.55,6.86) 2 3
Coarctation of the aorta (23) 12 4 0.67 (0.21,2.15) 0.48 (0.13,1.78) 5 1.67 (0.55,5.09) 1.82 (0.53,6.27) 2
Right ventricular outflow tract obstructions
Pulmonary valve stenosis (31) 15 7 0.91 (0.35,2.35) 0.99 (0.34,2.87) 4 1.10 (0.34,3.56) 0.83 (0.20,3.46) 5 1.92 (0.63,5.86) 1.66 (0.46,6.00)
Septal defects
Ventricular septal defect 18 6 0.67 (0.25,1.77) 0.67 (0.24,1.85) 4 0.89 (0.28,2.82) 0.78 (0.23,2.62) 5 1.59 (0.53,4.74) 1.63 (0.49,5.43)
(perimembranous) (33)
Atrial septal defect 40 21 1.05 (0.57,1.94) 0.84 (0.43,1.62) 13 1.30 (0.62,2.73) 1.34 (0.60,2.99) 10 1.43 (0.62,3.29) 1.15 (0.47,2.83)
(secundum or NOS) (84)
Non-heart defects
Neural tube defects (49) 23 13 1.13 (0.54,2.39) 1.23 (0.54,2.84) 7 1.22 (0.48,3.10) 1.78 (0.61,5.22) 6 1.49 (0.54,4.10) 2.16 (0.71,6.61)
Anencephaly and 6 4 1.33 (0.36,4.91) 1.30 (0.29,5.75) 3 2

craniorachischisis (15)
Page 15
Author Manuscript Author Manuscript Author Manuscript Author Manuscript
Women with periconceptionalb UTIs and … use
Nitrofurantoin Trimethoprim-sulfamethoxazole Cephalosporins

Penicillin
N
Ailes et al.
Defecta (N) N cOR (95% CI) aOR(95% CI) N cOR (95% CI) aOR(95% CI) N cOR (95% CI) aOR(95% CI)
Spina bifida (31) 15 9 1.20 (0.50,2.90) 1.26 (0.46,3.42) 3 4 1.52 (0.46,5.04) 2.84 (0.73,11.14)
Oral clefts (100) 37 37 2.03 (1.17,3.53) 1.97 (1.10,3.53) 14 1.57 (0.75,3.27) 1.71 (0.74,3.94) 12 1.85 (0.83,4.12) 1.63 (0.70,3.81)
Cleft palate (27) 11 10 1.85 (0.74,4.60) 1.93 (0.73,5.10) 4 1.50 (0.45,5.07) 1.59 (0.40,6.30) 2
Cleft lip +/− cleft palate (73) 26 27 2.11 (1.13,3.93) 1.90 (1.00,3.63) 10 1.59 (0.69,3.67) 1.61 (0.64,4.02) 10 2.20 (0.93,5.22) 1.72 (0.70,4.23)
Esophageal atresia (22) 6 7 2.33 (0.75,7.25) 2.57 (0.71,9.22) 7 4.67 (1.46,14.91) 5.31 (1.39,20.24) 2
Anorectal atresia/stenosis (19) 6 2 5 3.33 (0.95,11.66) 3.33 (0.95,11.66) 6 5.71 (1.68,19.41) 5.01 (1.34,18.76)
Hypospadias 2nd/3rd degree (50) 25 16 1.07 (0.51,2.24) 1.29 (0.52,3.20) 7 1.14 (0.42,3.09) 1.24 (0.40,3.87) 2
Limb deficiency (27) 12 9 1.50 (0.60,3.76) 1.11 (0.39,3.15) 4 1.33 (0.40,4.43) 1.64 (0.45,5.96) 2
Transverse limb deficiency (21) 10 6 1.20 (0.42,3.46) 0.78 (0.23,2.60) 4 1.60 (0.47,5.46) 1.93 (0.51,7.31) 1
Craniosynostosis (37) 15 13 1.73 (0.78,3.88) 1.72 (0.70,4.22) 5 1.33 (0.45,3.96) 1.18 (0.36,3.84) 4 1.52 (0.46,5.04) 0.83 (0.21,3.29)
Diaphragmatic hernia (16) 6 3 6 4.00 (1.20,13.28) 5.09 (1.20,21.69) 1
Gastroschisis (45) 28 10 0.71 (0.33,1.57) 0.53 (0.22,1.25) 4 0.57 (0.19,1.75) 0.43 (0.13,1.44) 3
Data in boldface type are significant at p < 0.05.

a
Restricted to birth defect types with at least 4 exposed cases and 4 unexposed cases or 100 total cases.
b
In the month before conception through the third month of pregnancy.
c
As pulmonary valve stenosis, and oral cleft cases were only ascertained by a subset of study sites in certain years, and hypospadias cases were only male fetus/infants, controls for these analyses were
similarly restricted. For pulmonary valve stenosis, there were a total of 222 controls, of which 115 reported penicillin exposure in the month before conception through the third month of pregnancy, 59
nitrofurantoin, 28 trimethoprim-sulfamethoxazole, and 20 cephalosporins. For oral clefts, there were a total of 229 controls, of which 120 reported penicillin exposure in the month before conception
through the third month of pregnancy, 59 nitrofurantoin, 29 trimethoprim-sulfamethoxazole, and 21 cephalosporins. For hypospadias, there were a total of 131 controls, of which 65 reported penicillin
exposure in the month before conception through the third month of pregnancy, 39 nitrofurantoin, 16 trimethoprim-sulfamethoxazole, and 11 cephalosporins.
aOR, odds ratio after conditioning on NBDPS study site and adjusting for maternal body mass index and race/ethnicity; cOR, crude (unadjusted) odds ratio; NOS, not otherwise specified; UTI, urinary tract
infection.
Page 16
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21 EXHIBIT E
22
23
24
25
26
27
28
Page 9 of 9
First Trimester Bleeding

MARK DEUTCHMAN, MD, University of Colorado Denver School of Medicine, Aurora, Colorado
AMY TANNER TUBAY, MD, and DAVID K. TUROK, MD, MPH, University of Utah School of Medicine, Salt Lake City, Utah
Vaginal bleeding in the first trimester occurs in about one fourth of pregnancies. About one half of those who bleed
will miscarry. Guarded reassurance and watchful waiting are appropriate if fetal heart sounds are detected, if the
patient is medically stable, and if there is no adnexal mass or clinical sign of intraperitoneal bleeding. Discriminatory
criteria using transvaginal ultrasonography and beta subunit of human chorionic gonadotropin testing aid in dis-
tinguishing among the many conditions of first trimester bleeding. Possible causes of bleeding include subchorionic
hemorrhage, embryonic demise, anembryonic pregnancy, incomplete abortion, ectopic pregnancy, and gestational
trophoblastic disease. When beta subunit of human chorionic gonadotropin reaches levels of 1,500 to 2,000 mIU per
mL (1,500 to 2,000 IU per L), a normal pregnancy should exhibit a gestational sac by transvaginal ultrasonography.
When the gestational sac is greater than 10 mm in diameter, a yolk sac must be present. A live embryo must exhibit
cardiac activity when the crown-rump length is greater than 5 mm. In a normal pregnancy, beta subunit of human
chorionic gonadotropin levels increase by 80 percent every 48 hours. The absence of any normal discriminatory
findings is consistent with early pregnancy failure, but does not distinguish between ectopic pregnancy and failed
intrauterine pregnancy. The presence of an adnexal mass or free pelvic fluid represents ectopic pregnancy until
proven otherwise. Medical management with misoprostol is highly effective for early intrauterine pregnancy failure
with the exception of gestational trophoblastic disease, which must be surgically evacuated. Expectant treatment is
effective for many patients with incomplete abortion. Medical management with methotrexate is highly effective for
properly selected patients with ectopic pregnancy. Follow-up after early pregnancy loss should include attention to
future pregnancy planning, contraception, and psychological aspects of care. (Am Fam Physician. 2009;79(11):985-
992, 993-994. Copyright © 2009 American Academy of Family Physicians.)
A
Patient information: bout one fourth of all pregnant 10 to 11 weeks since last normal menses),
A handout on first trimes- women experience spotting or and bimanual examination for masses and
ter bleeding, written by bleeding in the first several weeks tenderness. Guarded reassurance and watch-
the authors of this article,
is provided on page 993. of pregnancy, and one half of ful waiting are appropriate if fetal heart
those who bleed will miscarry.1 Family phy- sounds are detected with Doppler, if the
sicians who are familiar with the normal patient is stable, and if there is no adnexal
This clinical content progression of early pregnancy anatomy, mass, significant tenderness, or clinical sign
conforms to AAFP criteria sonographic findings, and beta subunit of of intraperitoneal bleeding.
for evidence-based con- human chorionic gonadotropin ( -hCG) Adnexal tenderness and the presence of
tinuing medical education
(EB CME). values can make a definitive diagnosis and a mass may indicate ectopic pregnancy.
proceed with appropriate treatment. Hypotension with other symptoms of hemo-
peritoneum (e.g., shoulder pain, absent
Managing First Trimester Bleeding bowel sounds, distended doughy abdomen)
By definition, bleeding before 20 weeks may point to ruptured ectopic pregnancy
of gestation constitutes threatened abor- and prompt immediate hospitalization for
tion (Table 12,3), but the majority of such evaluation.
pregnancies progress normally. The pace Examination with a vaginal speculum
of evaluation depends on the patient’s his- may reveal nonobstetric causes of bleeding,
tory, signs, and symptoms. If known, the such as cervicitis, vaginitis, cystitis, trauma,
time since the patient’s last normal menses cervical cancer, or polyps; or nonvaginal
may be used to estimate the gestational age. causes of bleeding, such as hemorrhoids.
In stable patients, the physical examination Significant cervical dilation or visible prod-
includes documentation of the size and posi- ucts of conception are indicative of an inevi-
tion of the uterus, auscultation of fetal heart table abortion. Tissue may be removed by
sounds by Doppler (if it has been at least gentle traction with ring forceps, and may be
June 1, 2009 Volume 79, Number 11 www.aafp.org/afp American Family Physician 985
Downloaded from the American Family Physician Web site at www.aafp.org/afp. Copyright © 2009 American Academy of Family Physicians. For the private, noncommercial
use of one individual user of the Web site. All other rights reserved. Contact copyrights@aafp.org for copyright questions and/or permission requests.
SORT: KEY RECOMMENDATIONS FOR PRACTICE
Evidence
Clinical recommendation rating References
Evidence does not support the routine use of antibiotics in all women with incomplete abortion. A 5
A normal pregnancy should exhibit a gestational sac when beta subunit of human chorionic C 1
gonadotropin levels reach 1,500 to 2,000 mIU per mL (1,500 to 2,000 IU per L), a yolk sac when
the gestational sac is greater than 10 mm in diameter, and cardiac activity when the embryonic
crown-rump length is greater than 5 mm.
Because expectant and surgical management of miscarriage are equally effective, the patient’s C 16
preference should play a dominant role in choosing a treatment.
When the patient has an incomplete abortion, nonsurgical treatments have a high likelihood of A 13-15
success; when the patient has an embryonic demise or anembryonic pregnancy, misoprostol
(Cytotec) or surgical treatment is more effective than expectant treatment.
Vaginal misoprostol is safer and more effective than oral misoprostol, with fewer gastrointestinal A 13, 15
side effects.
After a first trimester pregnancy loss, patients who are Rh negative should receive 50 mcg of anti D C 27
immune globulin.
Acknowledgment of grief, sympathy, and reassurance are useful techniques in counseling patients C 31
after miscarriage.
A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evidence; C = consensus, disease-
oriented evidence, usual practice, expert opinion, or case series. For information about the SORT evidence rating system, go to http://www.aafp.
org/afpsort.xml.
Table 1. Definition of Terms Applied to Early Pregnancy Loss
Term Definition
Anembryonic pregnancy Presence of a gestational sac larger than 18 mm without evidence of embryonic tissues (yolk sac or
embryo); this term is preferable to the older and less accurate term “blighted ovum”
Ectopic pregnancy Pregnancy outside the uterine cavity (most commonly in the fallopian tube) but may occur in the
broad ligament, ovary, cervix, or elsewhere in the abdomen
Embryonic demise An embryo larger than 5 mm without cardiac activity; this replaces the term “missed abortion”
Gestational trophoblastic Complete mole: placental proliferation in the absence of a fetus; most have a 46,XX chromosomal
disease or hydatidiform composition; all derived from paternal source
mole Partial mole: molar placenta occurring with a fetus; most are genetically triploid (69,XXY)
Heterotopic pregnancy Simultaneous intrauterine and ectopic pregnancy; risk factors include ovulation induction, in vitro
fertilization, and gamete intrafallopian transfer
Recurrent pregnancy loss More than two consecutive pregnancy losses; “habitual aborter” has also been used but is no
longer appropriate
Spontaneous abortion Spontaneous loss of a pregnancy before 20 weeks’ gestation
Complete abortion Complete passage of all products of conception
Incomplete abortion Occurs when some, but not all, of the products of conception have passed
Inevitable abortion Bleeding in the presence of a dilated cervix; indicates that passage of the conceptus is unavoidable
Septic abortion Incomplete abortion associated with ascending infection of the endometrium, parametrium,
adnexa, or peritoneum
Subchorionic Sonographic finding of blood between the chorion and uterine wall, usually in the setting of
hemorrhage vaginal bleeding
Threatened abortion Bleeding before 20 weeks’ gestation in the presence of an embryo with cardiac activity and
closed cervix
Vanishing twin A multifetal pregnancy is identified and one or more fetuses later disappear (occurs more often
now that early ultrasonography is common); if early in pregnancy, the embryo is
often reabsorbed; if later in pregnancy, it leads to a compressed or mummified fetus or
amorphous material
Information from references 2 and 3.
986 American Family Physician www.aafp.org/afp Volume 79, Number 11 June 1, 2009
examined for the presence of chorionic villi (Figure 1)

or sent for pathologic examination. Chorionic villi are
indicative of spontaneous abortion. Table 2 shows risk
factors of spontaneous abortion.2-4
Cervical testing for gonorrhea and chlamydia may
be performed. Fever and significant adnexal or perito-
neal symptoms are found in septic abortion. Treatment
of septic abortion is urgent, including prompt antibi-
otic administration and uterine evacuation. Infection,
retained products of conception, and uterine perfora-
tion are more common if the history includes attempted
abortion by someone who is untrained. Evidence does
not support the routine use of antibiotics in all women
with incomplete abortion.5 Figure 1. Passed tissue can be examined for chorionic villi.
If history and physical examination do not yield a If chorionic villi are present, the pregnancy was intrauter-
diagnosis, ultrasonography with or without -hCG test- ine, except in the rare heterotopic pregnancy.
ing is required. Application and interpretation of this
testing will help determine the differential diagnosis of
early pregnancy failure. Table 2. Risk Factors for Spontaneous Abortion
and Ectopic Pregnancy
Normal First Trimester Pregnancy Markers
HUMAN CHORIONIC GONADOTROPIN Spontaneous abortion
The first measurable finding in pregnancy is an elevated Endocrine (e.g., progesterone deficiency, thyroid disease,
uncontrolled diabetes)
-hCG level, which is produced by the placenta after
Genetic aneuploidy (accounts for about one half of
implantation of the blastocyst. This occurs at approxi-
spontaneous abortions)
mately 23 menstrual days’ gestation, or as early as eight
Immunologic (e.g., antiphospholipid syndrome, lupus)
days after conception. Readily available home urine
Infection (e.g., chlamydia, gonorrhea, herpes, listeria,
pregnancy tests detect -hCG levels as low as 25 mIU per mycoplasma, syphilis, toxoplasmosis, ureaplasma)
mL (25 IU per L) International Reference Preparation. Occupational chemical exposure
Therefore, it is possible to diagnose pregnancy before a Radiation exposure
missed period.6 Uterine (e.g., congenital anomalies)
Quantitative serum -hCG levels rise in a predictable
Ectopic pregnancy
fashion during the first four to eight weeks of normal
Current intrauterine device
pregnancy, increasing by 80 percent every 48 hours.
History of ectopic pregnancy
This rate of increase in -hCG levels is reassuring, but
History of in utero exposure to diethylstilbestrol
not indicative of normal pregnancy. Inadequately ris-
History of genital infection, including pelvic inflammatory
ing -hCG levels do not distinguish between ectopic disease, chlamydia, or gonorrhea
and failing intrauterine pregnancy. Unexpectedly high History of tubal surgery, including tubal ligation or
-hCG levels require consideration of gestational tro- reanastomosis of the tubes after tubal ligation
phoblastic disease. In vitro fertilization
Infertility
ULTRASONOGRAPHY
Smoking
Early detection of pregnancy depends on transvaginal
ultrasonography using transducer frequency of 5 MHz Information from references 2 through 4.
or greater.1 A 5-mm sonolucent gestational sac should be
visible in the endometrium at the fundus by five men-
strual weeks.1 The normal sac of an intrauterine preg- The yolk sac is visible using transvaginal scanning by
nancy consists of a central blastocyst surrounded by a six menstrual weeks. This confirms an intrauterine preg-
double ring of echogenic chorionic villi and decidua. nancy (Figure 2). By the end of the sixth week, a 2- to
This distinguishes it from a pseudogestational sac asso- 5-mm embryo or fetal pole becomes visible (Figure 3).
ciated with ectopic pregnancy. Measurement of the embryonic crown-rump length is
the most accurate way to date pregnancy (Figure 4). Car-

diac activity should be present when the embryo exceeds
5 mm in length.1 Transabdominal scanning is less sensi-
tive, and will show these landmarks about one week after
they are visible transvaginally.
DISCRIMINATORY CRITERIA
The predictable, linked progression of laboratory and

sonographic findings constitutes discriminatory criteria,
as shown in Table 3.1 A normal pregnancy should exhibit
Figure 2. Yolk sac (YS) within the gestational sac at five to a gestational sac when -hCG levels reach 1,500 to 2,000
six menstrual weeks. This is the first sonographic finding mIU per mL (1,500 to 2,000 IU per L), a yolk sac when
that positively confirms intrauterine pregnancy.
the gestational sac is greater than 10 mm in diameter, and
cardiac activity when the crown-rump length is greater
than 5 mm.1 The absence of an expected discrimina-
tory finding is consistent with pregnancy failure; how-
ever, because of the emotional impact of pregnancy loss,
imaging may be repeated one week later to confirm the
diagnosis. After an embryo with a heartbeat is confirmed
using these criteria, continued follow-up by ultrasonog-
raphy provides more specific information about the state
of the pregnancy than serial measurements of -hCG.
Making a Diagnosis
If the diagnosis is not clear from clinical examination alone,
knowledge of discriminatory criteria based on transvaginal
ultrasonography and -hCG findings facilitates the differ-
ential diagnosis of first trimester bleeding (Table 3 1).
ECTOPIC PREGNANCY
Ectopic pregnancy is responsible for 6 percent of all U.S.
Figure 3. The embryo is first visible as a fetal pole adjacent to maternal deaths.4 Intrauterine pregnancy with yolk sac
the yolk sac (YS). Cardiac activity is often visible at this time. or embryo rules out ectopic pregnancy, with the excep-
tion of a one in 4,000 chance of heterotopic pregnancy.
Transvaginal ultrasonography demonstrates an intra-
uterine gestational sac with nearly 100 percent sensitiv-
ity at -hCG levels of 1,500 to 2,000 mIU per mL.1 If the
-hCG level is above this discriminatory cutoff and a
gestational sac is not visible, then there is a high likeli-
hood of ectopic pregnancy.
An embryo with cardiac activity outside the uterus
proves ectopic pregnancy. Conditions that impede the
tubal transport of a fertilized ovum are associated with
ectopic pregnancy7 (Table 2 2-4), although many ectopic
pregnancies occur in women without risk factors. Early
diagnosis is the key to preventing morbidity and mortal-
ity, and preserving fertility.
An adnexal mass or free pelvic fluid (Figure 5) signifies
Figure 4. Measurement of the embryonic crown-rump a high probability of ectopic pregnancy, even if the gesta-
length is the most accurate way to date pregnancy. This tional sac or embryo is not visible. The presence of a cor-
10.5-week pregnancy measures 38 mm. pus luteum cyst of pregnancy may confuse the picture.
Table 3. Discriminatory Findings in Early Pregnancy
Menstrual Embryologic Laboratory and transvaginal sonographic

age event discriminatory findings
SPONTANEOUS ABORTION
AND EMBRYONIC DEMISE Three to Implantation Decidual thickening
four weeks site
Products of conception in the cervix or an Four weeks Trophoblast Peritrophoblastic flow on color flow Doppler
intrauterine embryo without cardiac activ- Four to five Gestational Present when beta subunit of human
ity proves incomplete abortion, inevitable weeks sac chorionic gonadotropin level is greater
abortion, or embryonic demise. Their pres- than 1,500 to 2,000 mIU per mL (1,500
ence rules out ectopic pregnancy, although to 2,000 IU per L; varies with sonographer
there is a one in 4,000 chance of hetero- experience and quality of ultrasonography)
topic pregnancy. Anembryonic pregnancy Five to six Yolk sac Present when diameter of gestational sac
weeks is greater than 10 mm
is often suspected when the patient reports
Five to six Embryo Present when diameter of gestational sac
regression of pregnancy symptoms or when weeks is greater than 18 mm
Doppler fails to detect fetal heart sounds Five to six Cardiac Present when embryonic crown-rump
by 10 to 11 weeks after the last normal weeks activity length is greater than 5 mm
menses.
Adapted with permission from Paspulati RM, Bhatt S, Nour SG. Sonographic evalua-
GESTATIONAL TROPHOBLASTIC DISEASE tion of first-trimester bleeding [published correction appears in Radiol Clin North Am.
2008;46(2):437]. Radiol Clin North Am. 2004;42(2):299. http://www.sciencedirect.
Gestational trophoblastic disease is char- com/science/journal/00338389.
acterized by a “snowstorm” of amorphous
material filling the uterus in patients pre-
senting in the first trimester.8,9 Prompt surgical evacua-
tion is required with close follow-up because of the risk
of metastatic disease.9
SUBCHORIONIC HEMORRHAGE
Subchorionic hemorrhage (Figure 6) is a common find-
ing with first trimester bleeding and may also be an
incidental finding in uncomplicated pregnancies. It is
important to note whether embryonic cardiac activity is
present. Subchorionic hemorrhage may be mistaken for
a twin gestational sac.
THE DIFFICULT DIAGNOSIS

It is common for -hCG levels to be less than 1,500 mIU Figure 5. The presence of free pelvic fluid in the cul-de-sac
is highly suggestive of ectopic pregnancy.
per mL with sonographic findings that are nondiagnos-
tic. It is reasonable to perform repeat ultrasonography
after one week in a stable patient. This interval allows
significant growth of the gestational sac or embryo, both
of which should grow at the rate of 1 mm per day. In
this situation, serial quantitative -hCG levels in com-
bination with follow-up imaging are also useful. Based
on a prospective study, the minimum -hCG increase
necessary for a living pregnancy over a 48-hour interval
is 53 percent.10 When -hCG rises more slowly than this,
plateaus, or falls, the differential diagnosis is limited to
failing intrauterine pregnancy or ectopic pregnancy.
When -hCG levels are not rising normally and ultra-
sonography cannot confirm pregnancy location, a dilata-
tion and curettage or manual vacuum aspiration may be
helpful. Manual vacuum aspiration requires a specially- Figure 6. Subchorionic hemorrhage (SCH) appears as a
designed 60-mL syringe with attached cannula to apply sonolucent area adjacent to the gestational sac, which
suction to the uterine cavity. If evacuation of the uterus contains an embryo (E) and yolk sac (YS).
yields chorionic villi, then a failed intrauterine pregnancy treatment.14,16 Misoprostol has fewer gastrointestinal side
is diagnosed and treatment for an ectopic pregnancy effects when given vaginally than when given orally.13,15
may be avoided. When ectopic pregnancy is suspected Based on these findings, increased use of medical man-
but cannot be confirmed with noninvasive testing, con- agement has benefits for patients, although misoprostol
sultation for diagnostic laparoscopy or treatment with is not approved by the U.S. Food and Drug Administra-
methotrexate is appropriate. tion for use in treating miscarriage.17 The doses used
in published studies are 800 mcg vaginally or 600 mcg
Management orally.13-15 If a single vaginal dose of 800 mcg does not
THREATENED ABORTION result in complete expulsion by day 3, the dose should
The presence of an intrauterine embryo with cardiac activ- be repeated. If complete expulsion has not occurred by
ity on ultrasonography should be reassuring because it day 8, manual vacuum aspiration should be offered.18
essentially rules out ectopic pregnancy. It is also associ- Additional published protocols are available.19
ated with a pregnancy loss rate of only 2 percent in women Confirming a negative urine -hCG four to six weeks
35 years and younger and 16 percent in women older than after early pregnancy loss excludes the presence of
35 years.11 If subchorionic hemorrhage (Figure 6) is present persistent gestational trophoblastic disease.20 Although
in an intrauterine pregnancy with fetal heart
sounds, the likelihood of spontaneous abor-
tion is 9 percent and may be even higher if the Table 4. Criteria for Managing Ectopic Pregnancy
patient is older than 35 years or if the hema-
toma is large.12 In this situation, the physician Expectant management
should caution patients to expect continued No evidence of tubal rupture
bleeding and possible impending miscarriage. Minimal pain or bleeding
Patient reliable for follow-up
SPONTANEOUS ABORTION, EMBRYONIC Starting -hCG level less than 1,000 mIU per mL (1,000 IU per L) and falling
DEMISE, AND ANEMBRYONIC PREGNANCY
Ectopic or adnexal mass less than 3 cm or not detected
Most first trimester miscarriages occur com- No embryonic heartbeat
pletely and spontaneously without interven- Medical management with methotrexate
tion. Although dilatation and curettage has Stable vital signs and few symptoms
historically been the treatment of choice, No medical contraindication for methotrexate therapy (e.g., normal liver
several recent trials confirm that expectant enzymes, complete blood count and platelet count)
management or medical management with Unruptured ectopic pregnancy
misoprostol (Cytotec) can be as effective and Absence of embryonic cardiac activity
safer while offering the patient more control Ectopic mass of 3.5 cm or less
over her care.13-16 Starting -hCG levels less than 5,000 mIU per mL (5,000 IU per L)
Clinical trials comparing expectant, medi- Dosage: single intramuscular dose of 1 mg per kg, or 50 mg per m2
cal, and surgical management reach several Follow-up: -hCG on the fourth and seventh posttreatment days, then
weekly until undetectable, which usually takes several weeks
conclusions. In incomplete abortion, high
Expected -hCG changes: initial slight increase, then 15 percent decrease
success rates have been demonstrated for
between days 4 and 7; if not, repeat dosage or move to surgery
expectant management (86 percent) and med-
Special consideration: prompt availability of surgery if patient does not
ical management (100 percent).13 However, respond to treatment
expectant management is more likely to fail Surgical management
in embryonic demise or anembryonic preg- Unstable vital signs or signs of hemoperitoneum
nancy; in these patients, the success of expect- Uncertain diagnosis
ant management by day 7 drops to 29 percent, Advanced ectopic pregnancy (e.g., high -hCG levels, large mass, cardiac
compared with 87 percent for medical man- activity)
agement.13 Women treated expectantly have Patient unreliable for follow-up
more outpatient visits than those treated med- Contraindications to observation or methotrexate
ically.13 Women treated medically have more
bleeding but less pain than those treated surgi- -hCG = beta subunit of human chorionic gonadotropin.
cally.14 Surgery is associated with more trauma Information from references 23 through 26.
and infectious complications than medical
substantially reduces the risk of neural tube defects.29

Table 5. Approaches to Grief Counseling After The psychological impact of pregnancy loss can be dev-
Miscarriage astating to the patient and her partner.30 Approaches to
managing patient grief are shown in Table 5.31
Acknowledge and attempt to dispel guilt
This article is one in a series on “Advanced Life Support in Obstetrics
Acknowledge and legitimize grief
(ALSO),” initially established by Mark Deutchman, MD, Denver, Colo. The
Provide comfort, sympathy, and ongoing support series is now coordinated by Patricia Fontaine, MD, MS, ALSO Managing
Reassure the patient about the future Editor, Minneapolis, Minn., and Larry Leeman, MD, MPH, ALSO Associate
Counsel the patient on how to tell family and friends about Editor, Albuquerque, N.M.
the miscarriage: Figure 5 provided by Matthew F. Reeves, MD, MPH, Magee-Womens
Speak simply and honestly Hospital of the University of Pittsburgh (Pa.) Medical Center.
Avoid medical details
Recognize that others may react emotionally The Authors
Explain how others can help, if known
MARK DEUTCHMAN, MD, is a professor in the Department of Family Medi-
Warn patients of the anniversary phenomenon cine at the University of Colorado Denver School of Medicine in Aurora,
Include the patient’s partner in your psychological care Colorado. He received his medical degree from The Ohio State University
Assess level of grief and adjust counseling accordingly in Columbus, and completed a family medicine residency at Sacred Heart
Medical Center in Spokane, Wash.
Information from reference 31. AMY TANNER TUBAY, MD, is an adjunct instructor in the Department of
Family and Preventive Medicine at the University of Utah School of Medi-
cine in Salt Lake City. She received her medical degree from Emory Univer-
sity School of Medicine in Atlanta, Ga., and completed a family medicine
there is solid evidence to support the use of prophylac- residency and an obstetrics fellowship at the University of Utah.
tic antibiotics for induced abortion, there is no evidence DAVID K. TUROK, MD, MPH, is an assistant professor in the Department
supporting this practice in early pregnancy failure.21 of Obstetrics and Gynecology and the Department of Family and Preven-
tive Medicine at the University of Utah School of Medicine. He received
his medical and master of public health degrees from Tufts University in
ECTOPIC PREGNANCY
Boston, Mass., and completed a family medicine residency at Brown Uni-
Early diagnosis of ectopic pregnancy brings management versity in Pawtucket, R.I., and a residency in obstetrics and gynecology at
into the outpatient setting. Current treatment options the University of Utah School of Medicine.
favor medical and laparoscopic management,22 with Address correspondence to Mark Deutchman, MD, University of Colorado
expectant management reserved for patients with a declin- Denver School of Medicine, Mail Stop F-496, Academic Office 1, 12631
East 17th Ave., Room 3617, Aurora, CO 80045 (e-mail: mark.deutchman@
ing quantitative -hCG of less than 1,000 mIU per mL ucdenver.edu). Reprints are not available from the authors.
(1,000 IU per L) and open surgical management reserved
Author disclosure: Nothing to disclose.
for patients with tubal rupture and hemoperitoneum.
Medical management with methotrexate is appropriate for
REFERENCES
properly selected patients. Randomized trials have shown
medical management to be safer, more effective, and less 1. Paspulati RM, Bhatt S, Nour SG. Sonographic evaluation of first-
trimester bleeding [published correction appears in Radiol Clin North
expensive than conservative surgical treatment, and to Am. 2008;46(2):437]. Radiol Clin North Am. 2004;42(2):297-314.
result in equal or better subsequent fertility.23-26 Ectopic http://www.sciencedirect.com/science/journal/00338389.
pregnancy management options, including methotrexate 2. Creinin MD, Schwartz JL, Guido RS, Pymar HC. Early pregnancy failure–cur-
rent management concepts. Obstet Gynecol Surv. 2001;56(2):105-113.
dosages and follow-up criteria, are shown in Table 4.23-26
3. Chen BA, Creinin MD. Contemporary management of early pregnancy
failure. Clin Obstet Gynecol. 2007;50(1):67-88.
Care After Pregnancy Loss 4. Berg CJ, Chang J, Callaghan WM, Whitehead SJ. Pregnancy-related
Several follow-up issues must be addressed after any type mortality in the United States, 1991-1997. Obstet Gynecol. 2003;101(2):
of pregnancy loss. Women who are Rh negative and mis- 289-296.
5. May W, Gülmezoglu AM, Ba-Thike K. Antibiotics for incomplete abor-
carry during the first trimester should receive 50 mcg of tion. Cochrane Database Syst Rev. 2007;(4):CD001779.
anti D immune globulin.27 Contraception should be dis- 6. Committee on Gynecologic Practice, The American College of Obste-
cussed and started immediately. All methods are equally tricians and Gynecologists. ACOG. Committee opinion: number 278,
safe immediately following spontaneous abortion or November 2002. Avoiding inappropriate clinical decisions based on false-
positive human chorionic gonadotropin test results. Obstet Gynecol.
ectopic pregnancy. There is no good evidence suggest- 2002;100(5 pt 1):1057-1059.
ing an ideal interpregnancy interval,28 but folic acid 7. Tay JI, Moore J, Walker JJ. Ectopic pregnancy [published correction
supplementation before attempts at future conception appears in BMJ. 2000;321(7258):424]. BMJ. 2000;320(7239):916-919.
8. Dogra V, Paspulati RM, Bhatt S. First trimester bleeding evaluation. 20. Seckl MJ, Gillmore R, Foskett M, Sebire NJ, Rees H, Newlands ES. Rou-
Ultrasound Q. 2005;21(2):69-85. tine terminations of pregnancy–should we screen for gestational tro-
9. Committee on Practice Bulletins–Gynecology, American College of phoblastic neoplasia? Lancet. 2004;364(9435):705-707.
Obstetricians and Gynecologists. ACOG Practice Bulletin number 53. 21. Prieto JA, Eriksen NL, Blanco JD. A randomized trial of prophylactic dox-
Diagnosis and treatment of gestational trophoblastic disease. Obstet ycycline for curettage in incomplete abortion. Obstet Gynecol. 1995;
Gynecol. 2004;103(6):1365-1377. 85(5 pt 1):692-696.
10. Barnhart KT, Sammel MD, Rinaudo PF, Zhou L, Hummel AC, Guo W. 22. Hajenius PJ, Mol F, Mol BW, Bossuyt PM, Ankum WM, van der Veen F.
Symptomatic patients with an early viable intrauterine pregnancy; HCG Interventions for tubal ectopic pregnancy. Cochrane Database Syst Rev.
curves redefined. Obstet Gynecol. 2004;104(1):50-55. 2007;(1):CD000324.
11. Smith KE, Buyalos RP. The profound impact of patient age on preg- 23. Cohen MA, Sauer MV. Expectant management of ectopic pregnancy.
nancy outcome after early detection of fetal cardiac activity. Fertil Steril. Clin Obstet Gynecol. 1999;42(1):48-54.
1996;65(1):35-40. 24. ACOG practice bulletin. Medical management of tubal pregnancy.
12. Bennett GL, Bromley B, Lieberman E, Benacerraf BR. Subchorionic hem- Number 3, December 1998. Clinical management guidelines for obste-
orrhage in first-trimester pregnancies: prediction of pregnancy outcome trician-gynecologists. American College of Obstetricians and Gynecolo-
with sonography. Radiology. 1996;200(3):803-806. gists. Int J Gynaecol Obstet. 1999;65(1):97-103.
13. Bagratee JS, Khullar V, Regan L, Moodley J, Kagoro H. A randomized 25. Stovall TG, Ling FW. Single-dose methotrexate: an expanded clinical
controlled trial comparing medical and expectant management of first trial. Am J Obstet Gynecol. 1993;168(6 pt 1):1759-1762.
trimester miscarriage. Hum Reprod. 2004;19(2):266-271. 26. Lipscomb GH, McCord ML, Stovall TG, Huff G, Portera SG, Ling FW.
14. Weeks A, Alia G, Blum J, et al. A randomized trial of misoprostol com- Predictors of success of methotrexate treatment in women with tubal
pared with manual vacuum aspiration for incomplete abortion. Obstet ectopic pregnancies. N Engl J Med. 1999;341(26):1974-1978.
Gynecol. 2005;106(3):540-547. 27. ACOG practice bulletin. Prevention of Rh D alloimmunization. Number
15. Zhang J, Giles JM, Barnhart K, Creinin MD, Westhoff C, Frederick MM, 4, May 1999 (replaces educational bulletin number 147, October 1990).
for the National Institute of Child Health and Human Development Clinical management guidelines for obstetrician-gynecologists. Ameri-
(NICHD) Management of Early Pregnancy Failure Trial. A comparison of can College of Obstetricians and Gynecologists. Int J Gynaecol Obstet.
medical management with misoprostol and surgical management for 1999;66(1):63-70.
early pregnancy failure. N Engl J Med. 2005;353(8):761-769. 28. Vlaanderen W, Fabriek LM, van Tuyll van Serooskerken C. Abortion risk and
16. Nanda K, Peloggia A, Grimes D, Lopez L, Nanda G. Expectant care ver- pregnancy interval. Acta Obstet Gynecol Scand. 1988;67(2):139-140.
sus surgical treatment for miscarriage. Cochrane Database Syst Rev. 29. Lumley J, Watson L, Watson M, Bower C. Periconceptional supple-
2006;(2):CD003518. mentation with folate and/or multivitamins for preventing neural tube
17. Winikoff B. Pregnancy failure and misoprostol–time for a change. N Engl defects. Cochrane Database Syst Rev. 2001;(3):CD001056.
J Med. 2005;353(8):834-836. 30. Deutchman M, Eisinger S, Kelber M. First trimester pregnancy com-
18. Robledo C, Zhang J, Troendle J, et al. Clinical indicators for success of plications. In: ALSO: Advanced Life Support in Obstetrics course syl-
misoprostol treatment after early pregnancy failure. Int J Gynaecol Obstet. labus. 4th ed. Leawood, Kan.: American Academy of Family Physicians;
2007;99(1):46-51. 2000:1-27.
19. Reproductive Health Access Project. Protocol of medical treatment 31. DeFrain J, MilIspaugh E, Xie X. The psychosocial effects of miscar-
of missed or incomplete abortion with misoprostol. http://www. riage: implications for health professionals. Fam Syst Health. 1996;
reproductiveaccess.org/m_m/protocol.htm. Accessed January 23, 2009. 14(3):331-347.

Declaration of Savannah O&#39;Neill

Diunggah oleh

Informasi Dokumen

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Declaration of Savannah O&#39;Neill

Diunggah oleh

Hak Cipta:

Format Tersedia

Case 3:18-cv-00050-JD Document 17-12 Filed 01/31/18 Page 1 of 62

LAW OFFICES OF YOLANDA HUANG

Received: 26 May 2017; Accepted: 14 June 2017; Published: 17 June 2017

Nutrients 2017, 9, 627; doi:10.3390/nu9060627 www.mdpi.com/journal/nutrients

Nutrients 2017, 9, 627 2 of 12

2. Materials and Methods

2.1. Design, Setting and Participants

Nutrients 2017, 9, 627 3 of 12

2.2.1. Nutrient Analysis Software Modification for Sri Lankan Food

Nutrients 2017, 9, 627 4 of 12

2.2.2. Calculation of Estimated Energy Requirement (EER)

EER (non-pregnant) = 354 (6.91 ⇥ age (years)) + physical activity coefficient ⇥

2.2.3. Recommended Dietary Allowance (RDA) of Protein

2.3. Data Analysis

Nutrients 2017, 9, 627 5 of 12

Table 1. Participants’ characteristics (n = 141).

Variable Mean (SD) n (%)

Nutrients 2017, 9, 627 6 of 12

3.3. Effects of Maternal and Neonatal Characteristics on GWG

3.4. Effects of Maternal and Neonatal Characteristics on Neonatal Birth Weight

3.5. Effects of Supplemental Foods on GWG and Neonatal Birth Weight

Nutrients 2017, 9, 627 8 of 12

Variable in Model Coefficient 95% CI t p-Value

Table 4. Neonatal birth weight by maternal and neonatal characteristics, ANOVA.

Birth Weight (g)

Nutrients 2017, 9, 627 9 of 12

Birth Weight (g)

Table 5. General linear model for neonatal birth weight (g).

Variable in Model Coefficient 95% CI t p-Value

Nutrients 2017, 9, 627 10 of 12

Nutrients 2017, 9, 627 11 of 12

Supplementary Materials: The following are available online at www.mdpi.com/2072-6643/9/6/627/s1,

Nutrients 2017, 9, 627 12 of 12

FROM THE ACADEMY

Position of the Academy of Nutrition and Dietetics:

ABSTRACT POSITION STATEMENT

Received: 26 May 2017; Accepted: 14 June 2017; Published: 17 June 2017

Nutrients 2017, 9, 627; doi:10.3390/nu9060627 www.mdpi.com/journal/nutrients

Nutrients 2017, 9, 627 2 of 12

2. Materials and Methods

2.1. Design, Setting and Participants

Nutrients 2017, 9, 627 3 of 12

2.2.1. Nutrient Analysis Software Modification for Sri Lankan Food

Nutrients 2017, 9, 627 4 of 12

2.2.2. Calculation of Estimated Energy Requirement (EER)

EER (non-pregnant) = 354 (6.91 ⇥ age (years)) + physical activity coefficient ⇥

2.2.3. Recommended Dietary Allowance (RDA) of Protein

2.3. Data Analysis

Nutrients 2017, 9, 627 5 of 12

Table 1. Participants’ characteristics (n = 141).

Variable Mean (SD) n (%)

Nutrients 2017, 9, 627 6 of 12

3.3. Effects of Maternal and Neonatal Characteristics on GWG

3.4. Effects of Maternal and Neonatal Characteristics on Neonatal Birth Weight

3.5. Effects of Supplemental Foods on GWG and Neonatal Birth Weight

Nutrients 2017, 9, 627 8 of 12

Variable in Model Coefficient 95% CI t p-Value

Table 4. Neonatal birth weight by maternal and neonatal characteristics, ANOVA.

Birth Weight (g)

Nutrients 2017, 9, 627 9 of 12

Birth Weight (g)

Table 5. General linear model for neonatal birth weight (g).

Variable in Model Coefficient 95% CI t p-Value

Nutrients 2017, 9, 627 10 of 12

Declaration of Savannah O'Neill

Declaration of Savannah O'Neill