Pregnancy Hypertension 13 (2018) 254–259

Contents lists available at ScienceDirect

Pregnancy Hypertension
journal homepage: www.elsevier.com/locate/preghy

Feeding practices in the first 6 months after delivery: Effects of gestational T

hypertension
⁎
Márcia Rejane Strapassona,b, , Charles Francisco Ferreiraa,c, José Geraldo Lopes Ramosa
a
Post Graduation Program in Health Sciences, Gynecology and Obstetrics (PPGGO), Faculty of Medicine, Hospital de Clínicas de Porto Alegre (HCPA), Universidade
Federal do Rio Grande do Sul (UFRGS), Porto Alegre, RS, Brazil
b
Universidade do Vale do Rio Dos Sinos – UNISINOS, São Leopoldo, RS, Brazil
c
Research Group: Climacteric and Menopause, Faculty of Medicine (FAMED), Hospital de Clínicas de Porto Alegre (HCPA), Universidade Federal do Rio Grande do Sul
(UFRGS), Porto Alegre, RS, Brazil

A R T I C LE I N FO A B S T R A C T

Keywords: Objective: To identify the effects of gestational hypertension on feeding practices in the first 6 months after
Breastfeeding delivery.
Hypertension Study design: A prospective cohort study enrolling 168 mother-newborn pairs (Gestational hypertension group
Lactation n = 42, Normotensive group n = 124). The gestational hypertension diagnosis criteria was established as a
Weaning
systolic pressure of ≥140 mmHg or a diastolic pressure of ≥90 mmHg after 20 weeks of gestation, while its
Preeclampsia
severity was categorized according to blood pressure, proteinuria, clinical and laboratory analysis. Demographic,
clinical and social information were collected from the patient’s medical records. In order to collect information
about the newborn’s feeding practices and possible difficulties in breastfeeding the mothers were interviewed via
telephone 30, 60, 120 and 180 days after delivery.
Main outcome measures: Feeding practices (eg. exclusive breastfeeding, predominant breastfeeding, com-
plementary breastfeeding and bottle-feeding) within the first 6 months after delivery.
Results: The mothers with Gestational hypertension displayed greater difficulties in maintaining exclusive
breastfeeding over time, when compared to normotensive mothers. There was a greater introduction of milk
formulas in the group of women with gestational hypertension, and they presented greater difficulties in
maintaining exclusive breastfeeding over time when compared to the group of normotensive mothers at hospital
admission (p ≤ 0,0001). The group with gestational hypertension reported higher frequencies of predominant
breastfeeding practices and presented shorter durations of breastfeeding after 6 months after delivery.
Conclusions: Women with gestational hypertension are at risk of using complementary breastfeeding and
breastfeeding for shorter durations.

1. Introduction categories as: Exclusive breastfeeding that involves only maternal

human breast milk direct from the breast or expressed from the breast
Breastfeeding is recognized worldwide as the best method of nu- with no other liquids or solids, except drops or syrups containing vi-
trition for infants [1]. It is proven to give nutritional health benefits to tamins, mineral supplements or medicines; Predominant breastfeeding:
the child, as well as in the brain [2], cognite [3], neurological, immune human breast milk predominates, but the child may receive water or
[4], psychological [4] development, increased intellectual quotient water-based drinks, fruit juices, oral rehydration salts solution, drops or
[5,6], the formation of bonds between mother and baby [7], among syrups of vitamins, minerals and medicines; Breastfeeding: the child
others. To this end, exclusive breastfeeding is recommended up to the receives human milk (straight from the breast or pumped);
sixth month of the baby’s life and after this period of time, to continue Complementary breastfeeding: the child receives breast milk and other
complementary breastfeeding until the child is two years old or older foods or liquids, solids and semi-solids including non-human milk and
[8]. formula; Bottle-feeding: any liquid including breast milk or semi-solid
The World Health Organization (WHO) defines breastfeeding in food from a bottle with nipple/teat or any food, liquid including non-

⁎
Corresponding author at: Programa de Pós-Graduação em Ciências da Saúde, Ginecologia e Obstetrícia, Faculdade de Medicina da Universidade Federal do Rio
Grande do Sul, Rua Ramiro Barcelos 2350, Largo Eduardo Faraco, Serviço de Ginecologia e Obstetrícia, CEP 90035-903 Porto Alegre, RS, Brazil.
E-mail addresses: [email protected] (M.R. Strapasson), [email protected] (J.G.L. Ramos).

https://doi.org/10.1016/j.preghy.2018.07.002
Received 13 January 2018; Received in revised form 23 June 2018; Accepted 9 July 2018
Available online 10 July 2018
2210-7789/ © 2018 International Society for the Study of Hypertension in Pregnancy. Published by Elsevier B.V. All rights reserved.
M.R. Strapasson et al. Pregnancy Hypertension 13 (2018) 254–259

human milk and formula [8]. The gestational age was adjusted for the cases of newborns who, due
Despite the benefits, only 35% of infants worldwide are exclusively to medical indication, received nothing orally or used other feeding
breastfed until the fourth month of life [7]. Early weaning occurs due to practices without sucking the breast. When the mother was allowed to
multiple factors, such as low socio-economic conditions [9] low levels breastfeed their newborn again, the age was adjusted.
of schooling, psychosocial problems [10,11], postpartum depression The sample calculation was performed with the use of the WINPEPI
[12,13] and working mothers [14]. However, there are very few studies (PEPI for Windows) program. For a difference of 5% and a power of
that address the theme of breastfeeding in the maternal population with 80%, considering a prevalence of 7.5% of preeclampsia found pre-
gestational hypertension and its complications which can affect 6–8% viously by Gaio et al. (2001) [21] in the Brazilian population; the
of pregnant women [15]. Gestational hypertension is one of the most sample size was calculated to contain 160 participants, 40 recent mo-
serious diseases specific to pregnancy, increasing the maternal and thers with preeclampsia and 120 without preeclampsia.
perinatal morbidity and mortality rate, and represents a serious public The main outcome measure was the conditions of breastfeeding and
health problem [15]. newborn feeding during hospitalization and a 6 month follow-up period
Preeclampsia is characterized by hypertension, proteinuria and after birth, collected through a form which was developed to evaluate
edema, potentially leading to intra-utero growth restriction and pre- the practices of breastfeeding and feeding of the baby.
term birth [16]. Women with preeclampsia have a higher difficulty In Rooming-in, the mothers were informed about the research, and
starting breastfeeding than normotensive women who have recently were asked for their authorization. The selection of the mothers and
given birth [17]. It is possible that this condition is associated with newborn medical records was done by convenience sample, in ac-
prematurity, premature separation of the newborn and disease co- cordance with the criteria, including gestational age paired samples.
morbidities [18,19]. The access to the selected records occurred from the patient care re-
Considering that gestational hypertension needs to be better un- cords through electronic consultation. After hospital discharge, tele-
derstood in relation to its repercussions in exclusive breastfeeding, this phone contact was made with the mother during the baby’s first
study aims to identify the associations of gestational hypertension with 30 days, 60 days, 120 days and 180 days, with the objective of following
the interruption of exclusive breastfeeding in the first 6 months of life. the child’s feeding habits and possible breastfeeding difficulties. The
collected information was recorded in the data collection form and
2. Methods inserted in an Excel program database. The data were collected by the
researchers after training in a pilot project.
A prospective cohort study conducted in a large teaching hospital In data analysis, continuous variables were expressed as
located in Canoas, in the metropolitan region of Porto Alegre, in the averages ± standard deviation of the mean (SD) or medians (md) and
south of Brazil. The sample consisted of new mothers diagnosed with 95% confidence intervals (CI 95%, upper and lower limits), defined by
gestational hypertension, the controls (normotensive new mothers), Shapiro-Wilk’s normality test. Categorical variables were described by
and their newborns, admitted to the maternity ward in the study absolute frequencies (n) and (n%). The comparative analysis between
period, from January 4th 2015 to January 10th 2017. For this, women continuous variables were performed by the Student t test for in-
who had recently given birth and who had a diagnosis of gestational dependent variables or by the Mann-Whitney test, where applicable. To
hypertension, preeclampsia, superimposed, preeclampsia or chronic evaluate the relation between categorical variables, the Pearson's Chi-
hypertension, eclampsia and HELLP, over 20 weeks of gestation and the square test was applied. In case of statistical significance, the standar-
control group (normotensive with gestational age paired) and their dized residuals test was adopted. The Pearson Chi-square test with re-
newborns (admitted in the Rooming-in and Neonatal Intensive Care sidual adjustment was used to evaluate the association between cate-
Unit) were included. New mothers under 18 years age, with a diagnosis gorical variables and Generalized Estimates Equation (GEE) analysis
of fetal malformation, dead fetus, diagnosed with psychiatric illness and considering time (30, 60, 120 and 180 days) and groups (hypertensive
human immunodeficiency virus positive women or who had another and normotensive) in relation to practices of breastfeeding (exclusive,
reason for the absolute contraindication to breastfeeding were excluded predominant, complementary and bottle-feeding). The Poisson regres-
from the study population. Due to problems with prenatal records and sion models, for each time measurement (30, 60, 120 and 180 days after
lack of information, such as gestational age due to menstruation or delivery), included the following variables: mode of delivery (vaginal or
ultrasound, it was decided to maintain gestational age according to the cesarean), parity, gestational age, and maternal hypertension categories
Capurro evaluation. (normotensive or hypertensive).
The diagnosis of gestational hypertension was considered with a To assist the analysis, we used the SPSS software, version 18.0, and
systolic blood pressure level of 140 mmHg and/or a diastolic blood the statistical significance was set at 5% (p ≤ 0.05) for all the analyses.
pressure level of 90 mmHg or higher, after 20 weeks of gestation, the Maternal parameters were characterized in Tables 1 and 2 and
diagnosis and the severity of the disease was based on the blood pres- perinatal variables, in Tables 3 and 4.
sure levels, proteinuria, clinical and laboratory findings, according to For this study, the ethical issues set out in Brazilian Ministerial
the criteria established by the International Society for the Study of Resolution No. 466/2012 were considered, which deals with research
Hypertension in Pregnancy (ISSHP), classified as preeclampsia cate- on humans [22]. The survey was conducted after the approval of the
gorized by changes in blood pressure and the presence of proteinuria; ethics and Research Committee (CEP) of the Hospital Mãe de Deus
when severe preeclampsia when diastolic blood pressure mmHg systolic under the opinion paragraph 756160.
≥110 mmHg or systolic ≥160 mmHg; proteinuria equal/higher than
3,0 g in 24 h or 3 + in urinary tape; oliguria; serum creatinine levels 3. Results
higher than 1,2 mg/dL; signs of hypertensive encephalopathy; epigas-
tric or right hypochondrium pain; clinical and/or laboratory coagulo- Women (N = 286) were submitted to an initial assessment. 118 out
pathy evidence; thrombocytopenia (< 150,000/dL); elevated liver en- of the 286 did not meet the inclusion criteria or ceased participation in
zymes and bilirubin. Preeclampsia together with chronic hypertension: the study because of the impossibility of contact. Therefore, the sample
characterized by the emergence of preeclampsia in women with chronic was composed of 168 women, namely, 126 normotensive women and
hypertension. Eclampsia: with the presence of tonic-clonic seizures; 42 hypertensive women (Supplemental Fig. 1).
HELLP, characterized by hemolysis, elevated liver enzymes and In the studied population, there was a predominance of white
thrombocytopenia and White coat hypertension when blood pressure in women (74.4%), with completed high school education (38.1%), mul-
the clinic or office is recorded by a nurse rather than by a physician, tiparous, mostly belonging to the normotensive group (62.7%),
preferably using repeated blood pressure reading [20]. (p ≤ 0.009), with vaginal birth being the most frequently performed in

255
M.R. Strapasson et al. Pregnancy Hypertension 13 (2018) 254–259

Table 1 24.0[24.4–28.5] in hypertensive women. The medians for blood pres-

Distribution of maternal hospitalization variables between the two groups. sure levels, protein/creatinine ratio, 24- urine protein test, were sig-
Variable Normotensive Hypertensive p value* nificantly higher in the hypertensive group (p ≤ 0.0001, p ≤ 0.0001,
n(n%) n(n%) p ≤ 0.000, respectively).
The average newborn weight was 3060.8 ± 557.7 in the normo-
Type of birth Vaginal 86(68.3) 12(28.6) ≤0.0001
tensive group, decreasing to 2614.8 ± 671.3 in the hypertensive group
Cesarean 40(31.7) 30(71.4)
Breastfed in the Yes 68(54.0) 12(28.6) 0.007
(p ≤ 0.0001). The median gestational age according to Capurro was
delivery room No 58(46.0) 30(71.4) 265.0[262.5–266.6] for the normotensive group and
Presence of colostrum Yes 126(100.0) 42(100.0) 1.000 264.0[255.4–265.5] for the hypertensive group, while the adjusted age
No 0(0.0) 0(0.0) was statistically significant in the hypertensive group (p ≤ 0.0001) and
Other children Yes 79(62.7) 16(38.1) 0.009
the first minute Apgar and 5 min Apgar in the normotensive group
No 47(37.3) 26(61.9)
Breastfed previously Yes 71(56.3) 14(33.3) 0.016 (p ≤ 0.0001, p ≤ 0.001, respectively), as shown in Table 3.
No 55(43.7) 28(66.7) There was no statistical difference with regard to the sex of the
Actual breastfeeding Yes 125(99.2) 39(92.9) 0.080 newborn between the groups. The newborn hospitalization was in
No 1(0.8) 3(7.1)
rooming-in which proved statistically significant (p ≤ 0.001), with a
Bleeding Yes 0(0.0) 4(9.5) 0.003
No 126(100.0) 38(90.5)
difference of (96.0%) in the normotensive group (78.6%) for the hy-
Type of nipple Altered 57(45.2) 20(47.6) 0.929 pertensive group, while the use of milk formula proved to be statisti-
Not altered 69(54.8) 22(52.4) cally significant (p ≤ 0.0001), with a greater occurrence in the hy-
Problems with breast Yes 56(44.4) 18(42.9) 1.000 pertensive group (52.4%), followed by the normotensive group
No 70(55.6) 24(57.1)
(16.7%). However, all the newborns admitted to the neonatal intensive
Legend: n: absolute frequency, n%: relative frequency. care unit were breastfed (p ≤ 0.001). Only (4.8%) of the newborns from
* Represent different distribution in total sample by Chi-Square test. the hypertensive group were bottle-fed, being statistically significant
Significance set as p ≤ 0.05 for all analysis. (p ≤ 0.014), as described in Table 4.
GEE analyses were conducted considering the practices of breast-
Table 2 feeding (exclusive, predominant, complementary breastfeeding, bottle-
Continuous variables – maternal data. feeding) the maternal groups (Hypertensive and Normotensive) and the
* moments in which the information was collected in the postpartum
Normotensive Hypertensive p Value
(n = 126) (n = 42) period (30, 60, 120 and 180 days) (Table 5). When compared together
(Table 5A), it was observed that the hypertensive mothers practiced
Maternal age (in years) 27.0[26.2–28.3] 24.0[24.4–28.5] 0.302 exclusive breastfeeding less than the normotensive mothers. In addi-
– md[95%CI]
tion, the predominant breastfeeding practices increased in both ma-
Parity – md[95%CI] 2.0[2.1–2.7] 2.0[1.7–2.7] 0.149
Abortion – md[95%CI] 0.0[0.2–0.4] 0.0[0.1–0.6] 0.552
ternal groups 30 days after delivery, being higher in each time point
Antenatal care visits – 7.5[6.7–7.8] 7.0[6.2–8.1] 0.791 examined. Considering the practice of complementary breastfeeding,
md[95%CI] both maternal groups behaved similarly. When analyzing the practice
Previous breastfeeding 30.0[187.0–329.7] 0.0[87.7–373.2] 0.055 of the groups who did breastfeed, it was observed that hypertensive
length (in days) –
mothers have smaller indexes in the first periods examined (30 and
md[95%CI]
60 days), but this index becomes similar to the normotensive group
Legend: n: absolute frequency, n%: relative frequency, 95%CI: 95% Confidence after 120 days. When we consider the population as a whole, for the
Interval [Lower Bound–Upper Bound]. md: median. p = Index of statistical verification of breastfeeding practices (exclusive, predominant, com-
significance. plementary and bottle-feeding) and postpartum periods examined (30,
* Represent difference between groups by Mann-Whitney test for in- 60, 120 and 180 days) (Table 5B), the decrease of exclusive breast-
dependent test comparisons. Significance set as p ≤ 0.05 for all analysis. feeding 60 and 180 days after delivery was observed, predominant
breastfeeding from the 120 days after delivery increased, as well as the
the normotensive group (68.3%) and cesarean section in the hy- stability of complementary breastfeeding throughout the analysis
pertensive group (71.4%), (p ≤ 0.0001). Regarding breastfeeding in the period. Finally, the rates from the group of mothers who did no
delivery room, the normotensive group showed a frequency of (54.0%) breastfeed increased from 120 days postpartum. Additionally, it was
while the hypertensive group showed a frequency of (28.6%). In ad- evidenced that breastfeeding/lactation practices (exclusively maternal
dition, breastfeeding in previous gestations was higher in the normo- and complementary) and bottle-feeding, equalized 120 days after de-
tensive group (56.3%) compared with the hypertensive group (33.3%), livery and before this time there is a greater frequency of breastfeeding/
even though breastfeeding during the hospitalization period did not lactation practices (exclusive and complementary), and after this mo-
present significant differences among both groups (p ≤ 0080), as seen ment the indexes in both practices of formula feeding and com-
in Table 1. plementary lactation have increased indexes.
The median age was 27.0[26.2–28.3] for normotensive women and Table 6 presents the Poisson regression model for Breastfeeding

Table 3
Continuous variables – perinatal data.
*
Normotensive (n = 126) Hypertensive (n = 42) p Value

Newborn weight – mean ± SD 3060.8 ± 557.7 2614.8 ± 671.3 ≤0.0001

Capurro Gestational Age – md[95%CI] 265.0[262.5–266.6] 264.0[255.4–265.5] 0.272
Ballard Gestational Age – mean ± SD 202.0 ± 0.00 264.0 ± 22.9 0.143
Adjusted for age– md[95%CI] 0.0[-2.0–6.1] 0.0[8.3–65.1] ≤0.0001
Apgar 1st minute – md[95%CI] 8.0[8.1–8.4] 7.5[6.9–7.7] ≤0.0001
Apgar 5th minute – md[95%CI] 9.0[8.8–9.0] 9.0[8.2–8.7] 0.001

LEGEND: n = absolute frequency, p = Index of statistical significance, md = median, P25-75 = percent is 25–75, SD = standard deviation.
* p value = Mann-Whitney test or Student's t-test.

256
M.R. Strapasson et al. Pregnancy Hypertension 13 (2018) 254–259

Table 4
Distribution of perinatal variables between the two groups.
Variable Normotensive Hypertensive p value*
n(n%) n(n%)

Sex of newborn Female 60(47.6) 21(50.0) 0.929

Male 66(52.4) 21(50.0)
Newborn hospitalization Rooming-in 121(96.0) 33(78.6) 0.001
Neonatal intensive care unit 5(4.0) 9(21.4)
Formula in hospitalization Yes 21(16.7) 22(52.4) ≤0.0001
No 105(83.3) 20(47.6)
Breast feeding in the neonatal intensive care unit Newborn hospitalized in neonatal intensive care unit breastfed 5(4.0) 9(21.4) 0.001
Newborn interned in neonatal intensive care unit, breastfeeding not released 0(0.0) 0(0.0)
Hospitalized in neonatal intensive care unit, breastfeeding released but is not 0(0.0) 0(0.0)
sucking.
Newborn not hospitalized in neonatal intensive care unit 121(96.0) 33(78.6)
Pacifier Yes 8(6.3) 2(4.8) 0.233
No 118(93.7) 39(92.9)
Not informed 0(0.0) 1(2.4)
Artificial formula Yes 0(0.0) 2(4.8) 0.014
No 126(100.0) 39(92.9)
Not informed 0(0.0) 1(2.4)

Legend. n: absolute frequency, n%: relative frequency, p = Index of statistical significance.

* Represent statistically different distribution in total sample by Chi-Square test. Significance set as p ≤ 0.05 for all analysis.

practices (for each measurement: 30, 60, 120 and 180 days after de- observed in this study are similar to national rates and current litera-
livery) and Maternal Hypertension categories (Normotensive and Hy- ture.
pertensive), taking into account other factors that might affect this re- This variation in the fourth month may be possibly influenced by
lationship (e.g. mode of delivery: vaginal or cesarean; parity and the mother's return to work. A study performed by Mascarenhas et al.
gestational age). It showed association with exclusive breastfeeding at (2006) [23] held in Pelotas, showed significant statistical association
30 and 60 days (RP = 1.73, 95% CI 1.124–2.670, p = 0.013 and between working mothers at the three month mark and the absence of
RP = 1.64, 95% CI 1,164–2,311p = 0.005 respectively), com- exclusive breastfeeding. Corroborating with these findings, a pro-
plementary feeding after 120 days (RP = 1.26, 95% CI 1.011–1.578, spective controlled study conducted in Chile with working mothers
p = 0.040) and predominant breastfeeding at 180 days (RP = 1.052, after maternity leave, found that only 53% of women who expressed
95% CI 1.052–1.618, p = 0.015). breast milk were able to continue exclusive breastfeeding until the sixth
month of the child's life, 6% more, compared to those that did not [14].
A study performed by Queluz et al. (2012) [24] found that early
4. Discussion
weaning is three times more likely in women who work and who do not
have maternity leave. Therefore, the importance of extending the ma-
For the population of the study pattern, exclusive breastfeeding
ternity leave from four to six months is important as it can lead to the
proved to increase within the first 30 and 60 days, suffering significant
increase of exclusive breastfeeding in Brazil.
reduction from 120 days and a minimum at 180 days. The rates

Table 5
Breastfeeding practice modification among groups, measurements and pairwise comparisons.
Breastfeeding practice 30 days 60 days 120 days 180 days

A. Considering breastfeeding practices, maternal groups (hypertensive or not) and measurements (days after delivery)
Exclusive breastfeeding
Normatensive mothers 0.72 ± 0.04aA 0.60 ± 0.04aB 0.30 ± 0.04aAB 0.09 ± 0.03aA
Hypertensive mothers 0.45 ± 0.08aA 0.33 ± 0.07aB 0.26 ± 0.07aAB 0.07 ± 0.04bAB

Predominant breastfeeding
Normatensive mothers 0.05 ± 0.02aA 0.08 ± 0.02aB 0.10 ± 0.03aC 0.33 ± 0.04aD
Hypertensive mothers 0.12 ± 0.05bA 0.14 ± 0.05aB 0.26 ± 0.07aAB 0.17 ± 0.06aA

Complementary feeding
Normatensive mothers 0.17 ± 0.03aA 0.21 ± 0.04aB 0.33 ± 0.04aAB 0.25 ± 0.04aA
Hypertensive mothers 0.38 ± 0.07aA 0.45 ± 0.08aB 0.19 ± 0.06aAB 0.36 ± 0.07aAB

Bottle-feeding
Normatensive mothers 0.06 ± 0.02aA 0.11 ± 0.03aB 0.27 ± 0.04aC 0.33 ± 0.04aC
Hypertensive mothers 0.05 ± 0.03bA 0.07 ± 0.04bB 0.29 ± 0.07aAB 0.40 ± 0.08aAB

B. Considering breastfeeding practices and measurements (days after delivery)

Exclusive breastfeeding 0.59 ± 0.04 0.47 ± 0.04 0.28 ± 0.04 0.08 ± 0.02
Predominant breastfeeding 0.08 ± 0.03 0.11 ± 0.03 0.18 ± 0.04 0.25 ± 0.04
Complementary feeding 0.27 ± 0.04 0.33 ± 0.04 0.26 ± 0.04 0.30 ± 0.03
Bottle-feeding 0.06 ± 0.02 0.09 ± 0.02 0.28 ± 0.04 0.37 ± 0.04

An interaction between groups and time effect was observed in maternal breastfeeding practices, hypertensive disorder and measurements (days after delivery)
pairwise comparisons by Generalized Estimating Equations (p ≤ 0.001). Data expressed as mean ± standard error of mean. A. Considering breastfeeding practices,
maternal groups (hypertensive or not) and measurements (days after delivery). abDifferent lowercase letters indicate difference proportion among the studied groups
between the same breastfeeding practice. ABDifferent uppercase letters show the evolution of a certain group over time. Significance set as p ≤ 0.05 for all analysis. B.
Considering breastfeeding practices and measurements (days after delivery).

257
M.R. Strapasson et al. Pregnancy Hypertension 13 (2018) 254–259

Table 6 while in the normotensive group despite the indexes being higher at 30
Poisson regression for breastfeeding practices and maternal hypertensive ca- than at 60 days, it was only different at 180 days of the child's life,
tegories. which makes it possible to affirm that normotensive women breast-fed
PR 95%CI p-value more than the hypertensive women over time.
A study performed by Leeners et al. (2005) [17], in Germany, with
30 days women with gestational hypertension (case group, number: 877) and
Exclusive breastfeeding
normotensive women, (control group, number: 623), showed that new
Hypertensive mothers 1.732 1.124–2.670 0.013
Normotensive mother 1 mothers with the HELLP syndrome began breastfeeding less often than
women with preeclampsia and gestational hypertension, although after
Predominant breastfeeding
Hypertensive mothers 0.895 0.780–1.026 0.111 the third month post-partum there was no significant difference in the
Normotensive mother 1 breastfeeding rate, diverging from the present study. Prematurity,
Complementary feeding
prolonged recovery of mothers, the newborn hospitalization and the
Hypertensive mothers 0.859 0.675–1.093 0.215 insufficient support of health professionals were referred to as factors
Normotensive mother 1 affecting initial reduction of breastfeeding [17].
Bottle-feeding However, other factors associated with the beginning of breast-
Hypertensive mothers 0.987 0.916–1.062 0.722 feeding in women with severe preeclampsia who had late preterm and
Normotensive mother 1 term deliveries included African-American race, younger age, low
educational level, multiple births, smoking and obesity, but the most
60 days
Exclusive breastfeeding significant statistical predictor was the intention to breastfeed [19]. A
Hypertensive mothers 1.640 1.164–2.311 0.005 study reiterates that the rates of breastfeeding in premature and low
Normotensive mother 1 birth weight were lower than those at term and can range from 48% to
Predominant breastfeeding 73% [25]. Prematurity is a condition for initial breastfeeding, where
Hypertensive mothers 0.897 0.765–1.051 0.180 mothers of premature babies breastfed less in relation to those with
Normotensive mother 1 term pregnancy [26]. For Haoa et al. (2017) [27], premature infants
Complementary feeding generally have more health problems and difficulty sucking, staying in
Hypertensive mothers 0.774 0.579–1.034 0.083 the neonatal intensive care unit for longer periods separated from their
Normotensive mother 1
mothers. This trend was observed in the present study; however, it was
Bottle-feeding not possible to associate late-preterm births observed in hypertensive
Hypertensive mothers 1.004 0.911–1.107 0.932 group as cause and effect in reducing rates of exclusive breastfeeding
Normotensive mother 1
over time. Another study by Caminha et al. (2014) [28] in the State of
120 days Pernambuco, found no statistical association between birth weight and
Exclusive breastfeeding breastfeeding rates in children over four months of age. Statistical as-
Hypertensive mothers 1.081 0.863–1.352 0.498 sociation was only present with weights less than 2500 g when analyzed
Normotensive mother 1
with maternal education and type of delivery. However, it is under-
Predominant breastfeeding stood that gestational hypertensive comorbidities can affect the initia-
Hypertensive mothers 0.840 0.689–1.025 0.086
tion and establishment of breastfeeding, especially when associated
Normotensive mother 1
with maternal and perinatal severe complications such as eclampsia,
Complementary feeding cerebral hemorrhage, placenta abruption, kidney failure, maternal
Hypertensive mothers 1.263 1.011–1.578 0.040
Normotensive mother 1
death [29,30], fetal growth restriction, prematurity, perinatal asphyxia
and fetal death, requiring larger studies to define these relationships
Bottle-feeding
[31].
Hypertensive mothers 0.890 0.694–1.141 0.358
Normotensive mother 1 The study showed that in addition to breastfeeding more than hy-
pertensive mothers, normotensive mothers took more time to start ex-
180 days clusive breastfeeding. The practice of predominant breastfeeding
Exclusive breastfeeding
showed a difference at 60 days and remained over time, while the
Hypertensive mothers 0.991 0.874–1.124 0.888
Normotensive mother 1
complementation with other type of milks was not different between
the groups. Weaning began earlier at 60 days in the hypertensive group,
Predominant breastfeeding
Hypertensive mothers 1.305 1.052–1.618 0.015
while in the normotensive this event occurred from 120 days.
Normotensive mother 1 A study by Vianna et al. (2007) [32] in Northeastern Brazil, with
11,076 children converged with the present study showing a gradual
Complementary feeding
Hypertensive mothers 0.926 0.712–1.205 0.567 decrease of exclusive breastfeeding and the introduction of other
Normotensive mother 1 feeding practices. Only 22.4% of all children under four months of age
Bottle-feeding were in exclusive breastfeeding and 19.4% were predominantly
Hypertensive mothers 0.813 0.600–1.103 0.184 breastfed. At six months of age, the prevalence for exclusive breast-
Normotensive mother 1 feeding was 16.6% and 15.9% for predominant breastfeeding. In ad-
dition, a study in northeastern Brazil on breastfeeding practices,
Legend: PR: Prevalence ratio. CI: Confidence interval. Poisson regression.
showed that 72% of newborns were given water or tea on the day of
Variables included in the analysis: mode of delivery, parity, gestational age,
birth and 80% in the first week of life. Other milk was introduced in the
maternal hypertension categories. Significance set as p ≤ 0.05 for all analysis.
first 30 days by 58% of mothers. In this population, the median for
breastfeeding was 65 days for mothers who introduced another type of
Despite breastfeeding while still in the hospital was not statistically
food other than breast milk in the first month, and 165 days for those
significant between the groups, it was noted that new mothers with
that did not [33]. In the present study, the use of bottles for hy-
hypertensive disorders of pregnancy, used milk formulas more fre-
pertensive mothers still in hospital proved to be statistically significant.
quently. This fact can justify the reduced practice of exclusive breast-
This practice occurred most probably due to the separation of the bi-
feeding in this population over time when compared to the normo-
nomial mother/baby or the prolonged recovery of the mother.
tensive group, falling at 60 days, keeping stable at 120 and 180 days,
New support and encouragement strategies for exclusive

258
M.R. Strapasson et al. Pregnancy Hypertension 13 (2018) 254–259

breastfeeding need to be developed, especially with the mothers with [4] W.H. Oddy, F. Rosales, A systematic review of the importance of milk TGF-β on
gestational hypertension. The design and implementation of a support immunological outcomes in the infant and young child, Pediatr. Allergy Immunol.
21 (1-Part-I) (2010) 47–59.
program on prenatal and breastfeeding in the hospital with a focus on [5] Organization H. World, Collaborative Study Team on the Role of Breastfeeding on
pregnancy can minimize situations that prevent exclusive breast- the Prevention of Infant Mortality, Effect of breastfeeding on infant and child
feeding. mortality due to infectious diseases in less developed countries: a pooled analysis,
Lancet 355 (2000) 451–455.
[6] C.G. Victora, B.L. Horta, C. Loret de Mola, L. Quevedo, R.T. Pinheiro, D.P. Gigante,
5. Conclusions et al., Association between breastfeeding and intelligence, educational attainment,
and income at 30 years of age: a prospective birth cohort study from Brazil, Lancet
Glob. Health 3 (4) (2015) e199–e205.
The study made it possible to identify associations between [7] W.H. Organization, UNICEF, Global Strategy for Infant and Young Child Feeding,
breastfeeding and women with gestational hypertension. Geneva, 2003.
Severe preeclampsia and the HELLP syndrome were associated with [8] W.H. Organization, Indicators for Assessing Infant and Young Child Feeding
Practices: Conclusions of a Consensus Meeting Held 6–8 November 2007 in
impeding exclusive breastfeeding, and this relationship may be attrib-
Washington, DC, USA. Part 1: Definitions, Geneva, 2008.
uted to the seriousness of the disease and its consequences such as [9] E. Nishioka, M. Haruna, E. Ota, M. Matsuzaki, R. Murayama, K. Yoshimura, et al., A
prematurity, prolonged recovery of mothers and newborn hospitaliza- prospective study of the relationship between breastfeeding and postpartum de-
tion in the neonatal intensive care unit. pressive symptoms appearing at 1–5months after delivery, J. Affect. Disord. 133 (3)
(2011) 553–559.
Mothers diagnosed with hypertension during pregnancy had greater [10] L.A. Nommsen-Rivers, C.J. Chantry, R.J. Cohen, K.G. Dewey, Comfort with the idea
difficulties in maintaining exclusive breastfeeding over time in relation of formula feeding helps explain ethnic disparity in breastfeeding intentions among
to normotensive mothers. On admission, the hypertensive group did not expectant first-time mothers, Breastfeeding Med. 5 (1) (2009) 25–33.
[11] M. Inoue, C.W. Binns, K. Otsuka, M. Jimba, M. Matsubara, Infant feeding practices
present statistical difference regarding breastfeeding but showed a and breastfeeding duration in Japan: a review, Int. Breastfeeding J. 7 (1) (2012) 15.
greater introduction of milk formulas. [12] C.-L. Dennis, K. McQueen, The relationship between infant-feeding outcomes and
Over time and considering breastfeeding practices, the hypertensive postpartum depression: a qualitative systematic review, Pediatrics 123 (4) (2009)
e736.
group recorded more predominant breastfeeding than their controls [13] M.C.M. Machado, K.F. Assis, Fd.C.C. Oliveira, A.Q. Ribeiro, R.M.A. Araújo,
and started weaning at 60 days, while the normotensive group began at A.F. Cury, et al., Determinants of the exclusive breastfeeding abandonment: psy-
120 days, very probably due to the end of their maternity leave and chosocial factors, Rev. Saúde Públ. 48 (6) (2014) 985–994.
[14] V. Valdés, E. Pugin, J. Schooley, S. Catalán, R. Aravena, Clinical support can make
returning to work, among other conditions. the difference in exclusive breastfeeding success among working women, J. Trop.
The study did not establish a definite cause and effect relationship Pediatr. 46 (3) (2000) 149–154.
between the possible associations and the practice of breastfeeding in [15] R.H. Diaz, F.L. Garcia, Prenatal breastfeeding intentions in a group of women with
high risk pregnancies, Boletin de la Asoc. Méd. Puerto Rico 102 (1) (2010) 21–23.
the groups.
[16] N. Vitoratos, D. Hassiakos, C. Iavazzo, Molecular mechanisms of preeclampsia, J.
As this is a cohort study where the action happened through tele- Pregnancy 2012 (2012) 5.
phone contact, a limitation exists regarding the veracity of the in- [17] B. Leeners, W. Rath, S. Kuse, P. Neumaier-Wagner, Breast-feeding in women with
formation and the registered losses. hypertensive disorders in pregnancy, J. Perinat. Med. (2005) 553.
[18] H.C. Lee, J.B. Gould, Factors influencing breast milk versus formula feeding at
New studies are needed to elucidate the relationship between discharge for very low birth weight infants in California, J. Pediatr. 155 (5) (2009)
breastfeeding and hypertension. 657-62.e2.
[19] L. Cordero, C.J. Valentine, P. Samuels, P.J. Giannone, C.A. Nankervis, Breastfeeding
in women with severe preeclampsia, Breastfeeding Med. 7 (6) (2012).
6. Conflicts of interest and source of funding [20] A.L. Tranquilli, G. Dekker, L. Magee, J. Roberts, B.M. Sibai, W. Steyn, G.G. Zeeman,
M.A. Brown, The classification, diagnosis andmanagement of the hypertensive
All authors listed above participated in the study to a significant disorders of pregnancy: a revised statement from the ISSHP, Preg. Hypertens. Int. J.
Women’s Cardiovasc. Health 4 (2014) 97–104.
extent. Márcia Strapasson, Charles Ferreira, and José Ramos worked on [21] D.S. Gaio, M.I. Schmidt, B.B. Duncan, L.B. Nucci, M.C. Matos, L. Branchtein,
analysis and interpretation of data, critically reviewed and revised the Hypertensive disorders in pregnancy: frequency and associated factors in a cohort
manuscript. Márcia Strapasson and José Ramos worked on the con- of Brazilian women, Hypertens. Preg. 20 (3) (2001) 269–281.
[22] BRASIL, Resolução n° 466 de 12 de dezembro de 2012, in: Md. Saúde, (Ed.),
ception, design and data collection. All authors read and approved the
Brasilia, 2012.
submitted manuscript. None of the authors has any conflicts of interest [23] M.L.W. Mascarenhas, E.P. Albernaz, M.B.D. Silva, R.B.D. Silveira, Prevalence of
related to this study, whether financial or of any other nature. exclusive breastfeeding and its determiners in the first 3 months of life in the South
of Brazil, J. Pediatria 82 (4) (2006) 289–294.
[24] M.C. Queluz, M.J.B. Pereira, C.B.D. Santos, A.M. Leite, R.G. Ricco, Prevalência e
Acknowledgments determinantes do aleitamento materno exclusivo no município de Serrana, São
Paulo, Brasil, Rev. Escola de Enfermagem da USP 46 (3) (2012) 537–543.
The authors thank all the researchers from the Research Group [25] L. Furman, N. Minich, M. Hack, Correlates of lactation in mothers birth weight
infants, Pediatrics 109 (4) (2002) e57.
Assistência à Saúde da Mulher – UFRGS, Porto Alegre, Brazil. [26] P. Leahy-Warren, H. Mulcahy, A. Phelan, P. Corcoran, Factors influencing initiation
Additionally, we would like to express our deepest gratitude to the and duration of breast feeding in Ireland, Midwifery 30 (2014) 345–352.
participants for their time and patience throughout this study. [27] N. Haoa, H. Jianga, M. Wua, T. Panb, B. Yana, J. Liuc, et al., Breastfeeding initia-
tion, duration and exclusivity in mothers with epilepsy from South West China,
Epilepsy Res. 135 (2017) 168–175.
Appendix A. Supplementary data [28] Md.F.C. Caminha, P.T.A.C.Cd. Azevedo, B.B. Sampaio, V.M.C. Acioly, M.P.M. Belo,
P.I.C. Lira, et al., Aleitamento materno em crianças de 0 a 59 meses no Estado de
Pernambuco, Brasil, segundo o peso ao nascer, Rev. Ciência e Saúde Coletiva 19 (7)
Supplementary data associated with this article can be found, in the (2014) 2021–2032.
online version, at https://doi.org/10.1016/j.preghy.2018.07.002. [29] E.A.P. Steegers, P. von Dadelszen, J.J. Duvekot, R. Pijnenborg, Preeclampsia, Lancet
376 (9741) (2010) 631–644.
[30] B. Sibai, G. Dekker, M. Kupferminc, Preeclampsia, Lancet 365 (9461) (2005)
References
785–799.
[31] J.R. Barton, B.M. Sibai, Prediction and prevention of recurrent preeclampsia,
[1] B. Al-Sahab, H. Tamim, G. Mumtaz, M. Khawaja, M. Khogali, R. Afifi, et al., Obstetr. Gynecol. 112 (2, Part 1) (2008) 359–372.
Predictors of breast-feeding in a developing country: results of a prospective cohort [32] R.Pd.T. Vianna, M.F. Rea, S.I. Venancio, M.M. Escuder, A prática de amamentar
study, Public Health Nutr. 11 (12) (2008) 1350–1356. entre mulheres que exercem trabalho remunerado na Paraíba, Brasil: um estudo
[2] C.M. Herba, S. Roza, P. Govaert, A. Hofman, V. Jaddoe, F.C. Verhulst, et al., transversal, Cadernos de Saúde Públ. 23 (10) (2007) 2403–2409.
Breastfeeding and early brain development: the Generation R study, Maternal Child [33] N.M. Marques, P.I.C. Lira, M.C. Lima, N.Ld. Silva, Filho M Batista, et al.,
Nutr. 9 (3) (2013) 332–349. Breastfeeding and early weaning practices in northeast Brazil: a longitudinal study,
[3] C. McCrory, A. Murray, The effect of breastfeeding on neuro-development in in- Pediatrics 108 (4) (2001).
fancy, Matern. Child Health J. 17 (9) (2013) 1680–1688.

259