icronutrient Deficiencies in Early Pregnancy Are Co
- © 2005 The American Society for Nutritional Sciences
mmon, Concurrent, and Vary by Season among Rural Nepali Pregnant Women
1
- Tianan Jiang,
- Parul Christian2,
- Subarna K. Khatry*,
- Lee Wu, and
- Keith P. West Jr
-
Center for Human Nutrition, Department of International Health, Johns Hopkins Bloomberg School of Public Health, Baltimore,
MD 21205, and
-
*Nepal Nutrition Intervention Project-Sarlahi (NNIPS), Nepal Netra Jyoti Sangh, Tripureswor, Kathmandu, Nepal
- ↵2To whom correspondence should be addressed. E-mail: pchristi{at}jhsph.edu.
Abstract
Pregnant women in developing countries are
vulnerable to multiple micronutrient deficiencies. We investigated
their prevalence
and seasonal variation as part of a baseline
assessment in a population-based, maternal micronutrient supplementation
trial
conducted in the rural Southeastern plains of
Nepal. Serum concentrations of 11 micronutrients were assessed in 1165
pregnant
women in the 1st trimester before supplementation.
Using defined cutoff values, the prevalence of deficiencies of vitamins
A, E, and D were 7, 25, and 14%, respectively.
Nearly 33% of the women were deficient in riboflavin, and 40 and 28% had
serum
vitamin B-6 and B-12 deficiencies, respectively.
Only 12% of the women were folate deficient, but 61% were zinc
deficient.
The prevalence of low serum iron concentration was
40%, and 33% were anemic (hemoglobin < 110 g/L). Multiple
micronutrient
deficiencies were common among pregnant women. Over
10% of the pregnant women were both anemic and deficient in B-complex
vitamins, whereas 22% of women were both anemic and
zinc deficient. Only 4% of women had no deficiency, whereas ∼20% of the
women had 2, 3, or 4 deficiencies. Almost 18% of
women had ≥5 deficiencies. Micronutrient status varied by season; it was
generally best during the winter months, except for
serum vitamin D concentration, which peaked during the hot summer and
monsoon months. Women in rural South Asia are
likely to begin a pregnancy with multiple micronutrient deficiencies
that may
vary with seasonality in micronutrient-rich food
availability.
Micronutrient deficiency in women of
reproductive age is recognized as a major public health problem in many
developing countries
(
1–
4).
Pregnant women are particularly vulnerable to nutritional deficiencies
because of the increased metabolic demands imposed
by pregnancy involving a growing placenta, fetus, and
maternal tissues, coupled with associated dietary risks (
5,
6). In turn, maternal undernutrition may predispose a mother to poor health, including infection, pre-eclampsia/eclampsia,
and adverse pregnancy outcomes such as premature birth and intrauterine growth retardation (
1,
2). Micronutrient deficiencies tend to coexist in impoverished settings in part because of uniformly low consumption of foods
rich in multiple micronutrients.
Maternal iron deficiency and consequent
anemia comprise a major problem in developing countries, affecting
>50% of women during
pregnancy (
1–
3). Other micronutrient deficiencies are likely to be widely prevalent, especially those of iodine, zinc, vitamin A, and the
vitamin B-complex (
1–
3,
7). However, little information is available on the extent and severity of multiple micronutrient deficiencies during pregnancy
in community-based studies (
8,
9).
Most data on vitamin status are from select populations, hospital-based
settings, or are from cross-sectional studies in
which nutrient status was assessed at varying single
time points during gestation. Because marginal micronutrient deficiency
in the 1st trimester could lead to more severe
deficiency later on due to stresses imposed by pregnancy and parturition
(
10), nutritional status in early pregnancy may be an important predictor of nutritional risk in late pregnancy (
8). In addition, seasonality appears to markedly influence the prevalence of deficiency (
7,
11,
12), which may sometimes result in unexpected patterns and interpretation of micronutrient deficiencies.
In previous studies, we documented a high prevalence of vitamin A and iron deficiency anemia among Nepali pregnant women (
13–
15)
at various stages of pregnancy. In the present study, we investigated
the prevalence of deficiency for vitamins A, E, D,
riboflavin, B-6, B-12, folate, zinc, iron, and copper
during early pregnancy (<12 wk) in a rural population of Nepalese
pregnant
women using published serological cutoff values. The
coexistence of multiple deficiencies and seasonal variations in
micronutrient
status were also examined.
SUBJECTS AND METHODS
Study design and population.
The present study utilized baseline data collected from a double-masked, cluster-randomized, controlled trial conducted in
the Southeastern plains District of Sarlahi, Nepal, from December 1998 to April 2001 (
15,
16).
The main objectives of the trial were to ascertain the effect of daily
maternal supplementation with folic acid, folic
acid + iron, folic acid + iron + zinc, and a
multiple micronutrient formulation, all with vitamin A, compared with an
active
placebo (containing vitamin A alone), on
reducing low birth weight, fetal loss, and infant mortality and
morbidity. The study
was approved by the ethical review committees
of the Ministry of Health in Nepal and the Johns Hopkins Bloomberg
School of
Public Health in Baltimore, MD.
To identify pregnancies in early
gestation, all eligible women of reproductive age (married women, 15–45 y
of age who were
not menopausal, sterilized, or not already
breast-feeding an infant < 12 mo of age) in the study area were
visited every 5
wk and monitored for pregnancy. Pregnancy was
ascertained with a urine test (human chorionic gonadotrophin antigen
test; Clue,
Orchid Biomedical Systems) among women who
had reportedly not menstruated in the past 30 d. Women who tested
positive were
enrolled after obtaining consent. At
enrollment, newly identified pregnant women were administrated a
baseline interview to
obtain data on 1-wk frequencies of symptoms
of morbidity, intake of food, alcohol, and tobacco use, and information
on household
socioeconomic status. Anthropometric
measurements included weight, height and mid-upper arm circumference
(MUAC)
3 measurements.
Of 30 village development
communities, comprising approximately one third of the total in the
study area, 9 were selected
to represent different geographic and ethnic
communities; all had reasonable access to the main roads and relative
proximity
to the project laboratory. Women from these
communities were invited to participate in a substudy involving venous
blood collection
for subsequent serum analysis of
micronutrient status twice during pregnancy, before supplementation and
in the third trimester.
The samples were drawn via venipuncture and
collected into 7-mL trace metal-free vacuum test tubes (Vacutainer,
Becton Dickinson).
The vacutainers containing blood were kept on
ice and brought to the clinic where they were centrifuged at 750 × g
for 20 min to separate the serum. Serum aliquots were placed into trace
element–free cryotubes (Nalgene Company, Sybron International),
stored in liquid nitrogen tanks, and shipped
to the Johns Hopkins Bloomberg School of Public Health in Baltimore, MD,
where
they were stored at −80°C before analyses.
Laboratory analysis.
Hemoglobin (Hb) assessment was done
on the spot using a homeglobinometer (HemoCue). Serum ferritin
concentration was analyzed
with ELISA procedures using a commercial
fluoroimmunometric assay (DELFIA® Ferritin, Perkin Elmer Wallac). The
interassay
CV for ferritin was ∼5%, using pooled serum
samples. Serum iron, zinc, and copper concentrations were analyzed by
atomic absorption
spectrometry (AAnalyst 600, Perkin Elmer).
Serum folate was measured by a microbiological assay in 96-well
microplates using
a chloramphenicol-resistant strain of
Lactobacillus rhamnosus (NCIMB 10463) (
17). Serum vitamin B-12 was determined by a microbiological assay in 96-well microplates using a colistin-sulfate-resistant
strain of
Lactobacillus lactis (NCIMB 12519) (
18,
19).
Both microorganisms were cryopreserved and the cultures were stable for
many months. The interassay CVs for folate and
vitamin B-12 were 7.6 and 8.4%, respectively.
Serum 25-hydroxyvitamin D concentration was used to evaluate the
vitamin D status,
determined by an immunoassay method with kits
from the Nichols Institute. The inter- and intra-assay CVs were 16.4
and 5.6%,
respectively.
Serum retinol, β-carotene, and α-tocopherol were determined simultaneously by a reverse-phase HPLC (Beckman, System Gold)
attached to an autosampler (717 Plus AS, Waters) using a procedure described by Yamini et al. (
20) with modifications. The internal standard used was all-
trans-ethyl-β-apo-8′-carotenoate
(Fluka Chem). The column (Allsphere ODS-2, 3 μm, 150 × 4.6 mm, Alltech
Associate) was eluted isocratically
with a mobile phase consisting of 84%
acetonitrile, 14% tetrahydrofuran, 6% methanol (added 0.2% ammonium
acetate), and 0.1%
triethylamine. Quality control was maintained
by repeated analyses of standard reference material (SRM, 968c, the
National
Institute of Standards and Technology, NIST,
Gaithersburg, MD) and pooled reference standards. The precision and
accuracy
of the method were also assessed through
participation in the Micronutrients Measurement Quality Assurance
Program of Round
Robin Proficiency Testing from the NIST, in
which 12 “unknown” samples are analyzed and submitted yearly for the
entire study
period.
Serum riboflavin concentration was
determined as a surrogate for riboflavin using reverse-phase HPLC (model
1100, Agilent
Technologies) with a fluorescence detector
(Model FP-1520, Jasco). Before HPLC, 50 μL of serum was deproteinized
using trichloroacetic
acid, and the supernatant was heated for 15
min. HPLC was performed using C18-coated silica column
(Alltech) with a mobile phase consisted of 65% (v:v) 5 mmol/L ammonium
acetate and 35% (v:v) methanol.
Fluorescence excitation and emission
wavelengths were 460 and 525 nm, respectively. The minimum detectable
concentration was
0.35 nmol riboflavin/L. Intra- and interassay
CVs were 1.5 and 4.8%, respectively, using pooled human serums. The
serum concentration
of pyridoxal 5′-phosphate, the active form of
vitamin B-6, was measured using HPLC. The serum (100 μL) was
deproteinized by
the addition of perchloric acid. Precolumn
derivatization was performed with potassium cyanide. The fluorescent
cyanide derivatives
were detected by fluorometry (Model FP-1520,
Jasco) with wavelengths for excitation at 318 nm and for emission at 418
nm.
The analytical HPLC column was an Alltech 3
μm ODS (C18), with a guard column packed with 40 μm C18
material (Alltech). The mobile phase was 50 mmol/L potassium phosphate
buffer (pH 3.2) containing 50 mmol/L sodium perchlorate
and mmol/L heptane sulfonic acid. The minimum
detectable concentration was 2 nmol/L. Published cutoff values for
defining
deficient concentrations of micronutrients
were used.
Statistical analysis.
Descriptive statistical tests were applied to maternal biological, demographic and socioeconomic, and biochemical variables.
Maternal age, gestational age at sample collection, BMI (calculated as kg/m2), and biochemical variables were treated as continuous variables, and diet and seasons were treated as categorical variables.
The proportion of women with
deficient status was calculated for single and multiple micronutrients.
We defined the 4 seasons
as follows: Spring (March–May), Summer
(June–August), Fall (September–November), and Winter
(December–February). Univariate
analyses were done to examine the association
between season and micronutrient status. We performed step-wise
multiple linear
regression analyses to assess seasonal
variation in micronutrient concentration using Spring as the reference
season. Maternal
age, gestational age, BMI, parity, tobacco
and alcohol use, socioeconomic status, and ethnicity were adjusted in
the model.
The prevalence of micronutrient deficiency
was determined for each season, and significant differences among the
prevalence
were assessed using χ2 analyses. The results were expressed as means ± SD, and a P-value
of <0.05 was considered significant. Statistical analyses were
performed using SAS software version 8.2 (SAS Institute).
RESULTS
Over a period of 2 y, 1316 (26.3%) of a
total of 4996 pregnant women were eligible for enrollment into the
substudy of the
trial. Of these, 1165 (88.5%) agreed to a venous
blood draw at baseline, and a few more agreed to a finger prick (n
= 67) allowing us to do Hb assessments on a total of 1232 (93.6%)
women. Sample size varied across the micronutrient determinations
(n = 1158–1165) because of inadequate quantities of serum in some cases.
Maternal age and gestational age at
enrollment were 23.6 ± 6.0 y and 10.9 ± 4.6 wk, respectively; 26% of the
women were nulliparous.
The women were stunted with a height of 150.5 ± 5.5
cm; 14% were below the cutoff value of 145 cm (
21). Subjects were also thin and wasted with a MUAC of 21.9 cm, and BMI of 19.3 kg/m
2.
Serum micronutrient concentrations were normally distributed during early pregnancy except for riboflavin, vitamin B-12, folate,
and ferritin, which were slightly skewed to the left (
Table 1).
TABLE 1
Mean, median, and percentile distributions of serum concentrations of vitamins, trace minerals, and Hb indices among Nepalese
pregnant women in the 1st trimester
Using defined cutoff values (
7,
13,
22–
29),
the prevalence of deficiencies of vitamins A, E, and D were 7, 25, and
14%, respectively. Nearly one third of the women
were deficient in riboflavin, and 40 and 28% had
serum vitamin B-6 and B-12 deficiencies, respectively. Only 12% of the
women
were folate deficient, but 61% were zinc deficient.
The prevalence of low serum iron concentration was 40%, whereas 33%
were
anemic (hemoglobin < 110 g/L). Multiple
micronutrient deficiencies were common among pregnant women. Over 10% of
the pregnant
women were both anemic and deficient in B-complex
vitamins, whereas 22% of women were both anemic and zinc deficient.
Because
zinc deficiency was the most common micronutrient
deficiency, occurring in ∼60% of women, it was also the most frequent
coexisting
micronutrient deficiency (
Table 2).
Only 4% of women had no deficiency, whereas 14.3, 21.5, 20.6, 21.9, and
17.7 had 1, 2, 3, 4, or 5 or more deficiencies,
respectively (data not shown). Women who were
deficient in certain nutrients were more likely to have other
micronutrient
deficiencies (
Table 3).
For example, among the women with vitamin A deficiency, a higher
proportion was deficient in vitamin E (70%), vitamin B-6
(52%), or riboflavin (44%) or were anemic (49%)
relative to the overall prevalence of these in the population. The
prevalence
of micronutrient deficiencies differed by season (
Table 4).
Overall, most micronutrient deficiencies assessed tended to be less
prevalent during the winter season. Serum retinol concentrations
were significantly lower in the hot and humid
monsoon season (April–September), whereas β-carotene and vitamin B-6
concentrations
were higher in summer and winter than in other
seasons (
Fig. 1).
As expected, vitamin D levels were lowest in the winter months and
highest in summer (June–August) and fall (September–November).
Serum zinc concentrations were higher in fall and
lower in summer. Multiple regression analysis also revealed that with
the
exception of vitamin D and copper, women had better
micronutrient status during the winter months (December–February) than
in the other seasons (data not shown).
TABLE 2
The prevalence of concurrent deficiencies of 2 micronutrients among Nepalese pregnant women in the 1st trimester1
TABLE 3
Prevalence
of micronutrient deficiencies among Nepalese pregnant women in the 1st
trimester by the presence of an index micronutrient
deficiency1
TABLE 4
The prevalence of micronutrient deficiency by season among Nepalese pregnant women in the 1st trimester1, 2
FIGURE 1
Seasonal
variation in mean serum micronutrient concentrations among Nepalese
pregnant women in the 1st trimester over a 2-y
period from December 1998 until April 2001.
As shown on the figure keys, concentrations of some nutrients were
adjusted to
scale: retinol level was adjusted upward by a
factor of 10, β-carotene by a factor of 100, and zinc by a factor of 2,
whereas
vitamin D concentration was adjusted downward
by a factor of 2 and vitamin B-12 by a factor of 10.
DISCUSSION
In the present study we documented that
the prevalence of multiple micronutrient deficiencies was common among
women in the
1st trimester of pregnancy in rural Nepal, likely
reflecting dietary inadequacy as they entered pregnancy. Because
micronutrient
status was assessed at a gestational age of 10.9 ±
4.6 wk, the confounding effect of hemodilution that peaks in the 3rd
trimester
(
30,
31) was likely to be minimal.
The present study has several advantages
over previous studies that examined micronutrient status during
pregnancy. Apart
from having a large sample size, a wide range of
micronutrients was examined simultaneously in a community-based study,
allowing
estimation of the extent of maternal micronutrient
deficiencies in early pregnancy in this rural Nepali setting. In
addition,
we report the extent to which multiple deficiencies
coexist, data that are scarce in rural developing country settings. The
results of the present study also have the
potential to provide valuable reference values for assessing nutritional
status.
However, the assessment of vitamin and mineral
status during pregnancy is complicated because there is a general lack
of pregnancy-specific
laboratory indices for nutritional evaluation (
6), and pregnancy itself may alter “normal” values (
9) independently of the effects of hemodilution. Furthermore, because of the lack of standardization of the assay and different
cutoff values to define deficient status, the prevalence of a nutrient deficiency may vary between studies.
Inadequacy of a single nutrient is most
likely associated with deficiencies of other micronutrients. Our
population study
provides evidence that rural pregnant women in
South Asia are likely to suffer from multiple deficiencies. This was
evident
from our estimate that simultaneous deficiencies
for ≥2 micronutrients affected 82% of women who entered the trial early
in
pregnancy. The likelihood of certain micronutrient
deficiencies was higher in the presence of specific other micronutrient
deficiencies, suggesting potential metabolic
interactions. For example, among women with vitamin A deficiency, nearly
half
or more also had vitamin E, riboflavin, and vitamin
B-6 deficiencies and were anemic, substantially above their respective
univariate rates in the population. Joint,
apparently noninteractive deficiencies were also evident, suggested by
comparable
index and conditional prevalences. For example,
B-complex vitamin (riboflavin, vitamins B-6 and B-12) and iron
deficiencies
were comparable irrespective of concurrent zinc
deficiency, possibly derived in part from a shared dietary deficit of
good
food sources such as meat, little of which is
consumed in this setting (data not shown) and elsewhere in rural South
Asia.
Compounding the adverse effects of infrequent
intake of micronutrient-rich foods is also a diet that is
characteristically
high in inhibitors of mineral absorption. Phytates,
for example, which are abundant in rice and other grains, inhibit zinc
absorption in particular (
32); this could help explain the higher prevalence of this nutrient deficit.
Coexisting nutritional deficiencies could reduce the potential benefit of a single nutrient supplement in improving nutrition
status and morbidity (
33,
34). The role of vitamin deficiencies in the etiology of anemia was described (
33–
36). Specifically, vitamin A, riboflavin, vitamin B-6, vitamin B-12, and folate exert hematopoietic function (
35–
37), suggesting that anemic women should possibly be supplemented not only with iron but also with vitamin A (
33) and other micronutrients (
36,
37).
However, less is known about metabolic interactions of micronutrients.
Zinc may interact with vitamin A to potentiate the
effect of vitamin A in restoring night vision among
night-blind pregnant women with low initial serum zinc concentrations
(
38).
The diet of the majority of rural Nepalese is monotonous and low in vegetables and animal sources (
39); it is highly dependent on a largely local market system and seasonal availability. In Nepal, in the period before the monsoon
season, the availability of fruit and vegetables drops dramatically, causing an increase in their prices (
39).
The rainy monsoon season causes pronounced seasonal shortages, whereas
vegetables tend to be more abundant in the dry,
mid-winter season. Such variation in availability,
and cost, of various foods can affect the status of certain
micronutrients.
In the present study, for example, biochemical
concentrations of most, although not all, assessed micronutrients were
higher
in the dry, mid-winter months. Alternatively, late
dry season concentration peaks in serum β-carotene and vitamin B-6
presumably
correspond to the mango harvest and to increased
banana availability at that time of year in the southeastern plains of
Nepal.
Previously, we found maternal night blindness
prevalence to increase during the hot summer months before the monsoon
season,
but then to decline during the short mango season
in June–July (
40), mimicking the seasonal dynamic in xerophthalmia that has long been reported among South Asian children (
41). A seasonal pattern in Hb was similar to that reported by Bondevik et al. (
12)
in pregnant women in a hospital setting in Nepal, in which the
prevalence of anemia was highest during and after the monsoon
period. No seasonal variation was observed in serum
concentrations of vitamin B-12 and ferritin, similar to reports by
Ronnenberg
et al. (
7) among Chinese women and Backstrand et al. (
42)
among Mexican women. Vitamin D concentrations peaked in the monsoon and
hot summer months, presumably in response to increased
exposure to sunlight. Exposure to UV light during
the monsoon season may be high, despite cloud cover, because of the
planting
and other agricultural work in which women may be
involved during this season (
43).
Knowing the high-risk seasonal periods for maternal micronutrient
deficiencies can aid in developing and targeting interventions
for their control.
In summary, we report here the population
prevalence of low and deficient maternal status with respect to
multiple micronutrients
in the Southern plains of Nepal, based on published
serum concentration cutoff values. We found that >80% of women
exhibited
evidence of ≥2 micronutrient deficiencies, and that
seasonality can markedly and differentially influence maternal status,
likely associated with seasonality of
micronutrient-rich foods as well as other potential epidemiologic risk
factors (e.g.,
infection). Because the findings reflect maternal
status in the 1st trimester of pregnancy, they may be also taken to
represent
the burden of micronutrient deficiency in rural
women of reproductive age in this population, and provide regional
guidance
on approaches and timing to the control of multiple
micronutrient deficiencies early in pregnancy and throughout the
reproductive
years in rural South Asia.
Acknowledgments
Apart from the authors, the following
members of the Nepal study team helped in the successful implementation
of the study:
Steven C. LeClerq, Sharada Ram Shrestha and Jonne
Katz; Field Managers Tirtha Raj Shakya and Rabindra Shrestha; Field
Supervisors
Uma Shankar Sah, Arun Bhetwal, Gokarna Subedi, and
Dhrub Khadka; and Laboratory Scientist Tracey Wagner for conducting
laboratory
analyses. Special thanks to the team of
phlebotomists for their hard work in conducting home-based blood
collection, which
made this analysis possible; Elizabeth K. Pradhan
and Gwendolyn Clemens for computer programming and data management; Ravi
Ram, Seema Rai, and Sunita Pant for data cleaning
and supervision.
Footnotes
-
↵1
This work was carried out by the Center for Human Nutrition, the
Department of International Health of the Johns Hopkins
Bloomberg School of Public Health, Baltimore,
MD, in collaboration with the National Society for the Prevention of
Blindness,
Kathmandu, Nepal, under the Micronutrients
for Health Cooperative Agreement No. HRN-A-00-97-00015-00 and the Global
Research
Activity Cooperative Agreement No.
GHS-A-00-03-00019-00 between the Johns Hopkins University and the Office
of Health, Infectious
Diseases and Nutrition, United States Agency
for International Development, Washington, DC, and grants from the Bill
and Melinda
Gates Foundation, Seattle, WA; UNICEF,
Kathmandu, Nepal; and, the Sight and Life Research Institute, Baltimore,
MD.
-
↵3
Abbreviations used: Hb, hemoglobin; MUAC, mid-upper arm circumference;
NIST, the National Institute of Standards and Technology.