Assisted reproductive technology and breastfeeding outcomes: a case-control study Antonella Cromi, Ph.D.,a Maurizio Serati, M.D.,a Ilario Candeloro, M.D.,a Stefano Uccella, Ph.D.,a Sara Scandroglio, M.D.,a Massimo Agosti, M.D.,b and Fabio Ghezzi, M.D.a a Department of Obstetrics and Gynecology, University of Insubria; and Intensive Care Unit, Del Ponte Hospital, Varese, Italy
b
Department of Neonatology and Neonatal
Objective: To determine if breastfeeding outcomes differ between mothers who conceived spontaneously compared with those who conceived with assisted reproductive technology (ART). Design: Matched case-control study. Setting: Teaching hospital. Patient(s): Ninety-four women having a singleton baby conceived with ART, matched by maternal age, parity, mode of delivery, and gestational age to controls who conceived spontaneously. Intervention(s): Cases and controls were interviewed using a standardized, structured questionnaire, to obtain information on lactation. Exposure to maternity care practices contributing to breastfeeding success was investigated. Main Outcome Measure(s): Initiation, exclusivity, and continuation of breastfeeding. Result(s): Cases were as likely as controls to initiate breastfeeding (89.4% vs. 90.4%), but by 6 weeks postpartum, a greater proportion of mothers who conceived through ART has ceased breastfeeding (20.2% vs. 5.3%). The percentage of mothers who exclusively breastfed their child for 6 months was similar among the 2 groups. On univariate conditional logistic regression, a history of using ART was the only predictor of early breastfeeding cessation (odds ratio ¼ 65.3 [95% confidence interval: 1.5–2889.3]). Conclusion(s): Women who have conceived with ART should be regarded as being at higher risk for early breastfeeding cessation. This study serves as a first step in the investigation of potential modifiable factors that contribute to breastfeeding failure among women who give birth after using ART, and may help in efforts to customize breastfeeding support strategies. Use your smartphone (Fertil SterilÒ 2015;103:89–94. Ó2015 by American Society for Reproductive Medicine.) to scan this QR code Key Words: Assisted reproductive technology, breastfeeding, assisted conception, lactation, and connect to the parenting Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/cromia-art-breastfeeding-outcomes/
B
reastfeeding is recognized as the best source of nutrition for most infants, and exclusive breastfeeding is recommended for the first 6 months of life by many authorities, including the World Health Organization (WHO; 1, 2). Few health behaviors have such a broad-spectrum and longlasting impact on health status, with the potential to improve life chances, health, and well-being (3, 4).
With births occurring increasingly as a result of assisted reproductive technology (ART), the impact of infertility treatments on initiation and duration of breastfeeding is an important issue to explore. Whereas obstetric outcomes among women who conceive as a result of ART have been extensively studied, only a moderate body of research addresses maternal postnatal adjustment, and psychological aspects of early
Received August 7, 2014; revised October 6, 2014; accepted October 7, 2014; published online October 29, 2014. A.C. has nothing to disclose. M.S. has nothing to disclose. I.C. has nothing to disclose. S.U. has nothing to disclose. S.S. has nothing to disclose. M.A. has nothing to disclose. F.G. has nothing to disclose. Reprint requests: Antonella Cromi, Ph.D., Department of Obstetrics and Gynecology, University of Insubria, Del Ponte Hospital, Piazza Biroldi 1, 21100 Varese, Italy (E-mail: antonella.comi@ uninsubria.it). Fertility and Sterility® Vol. 103, No. 1, January 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.10.009 VOL. 103 NO. 1 / JANUARY 2015
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
parenting, after conception with ART (5). The limited evidence about the potential impact of an ART treatment history on breastfeeding capacity is inconclusive and contradictory, mainly because of significant methodological limitations in the studies to date, including recruitment strategies, inadequate sample sizes, lack of appropriate comparison groups, and failure to control for known risk factors for lactation difficulties (6–9). Women who conceive with ART are more likely to be first-time mothers; they are, on average, older and more likely to have pregnancy complications, multiple births, operative delivery, labor induction, premature birth, and low–birth weight babies than 89
ORIGINAL ARTICLE: ASSISTED REPRODUCTION women who conceive spontaneously. Thus, comparisons made with data drawn from the general population of childbearing women need to be interpreted with extreme caution. Moreover, understanding breastfeeding outcomes requires an exploration beyond maternal and newborn factors that could affect lactation success. Maternity care policy and practices, along with provider and hospital factors, may have an impact on successful breastfeeding (10, 11). Therefore, studies recruiting participants from infertility clinics may fail to control for confounders related to the setting in which a woman gives birth. We designed a case-control study to investigate initiation, duration, and exclusivity of breastfeeding among every eligible case of women during the study period who had conceived with the help of ART, compared with an appropriately matched control group of patients who conceived spontaneously and gave birth at the same hospital.
MATERIALS AND METHODS Setting The study was conducted in the Obstetrics Department of Del Ponte Hospital at the University of Insubria, in Varese, Italy, a teaching and tertiary referral center for high- and low-risk obstetrics. Although the ‘‘Baby-Friendly Hospital Initiative’’ designation (12) was not taken into consideration, the changes required for successful implementation of the ‘‘ten steps to successful breastfeeding,’’ developed by WHO and UNICEF (13), had been undertaken at our institution long before the study was conducted. Maternity care practices related to breastfeeding did not change over the study period.
Study Sample Cases were defined as all women conceiving through ART who gave birth at our institution between January 1, 2010 and October 31, 2013. We included live singleton births, either late-preterm (34 weeks to 36 weeks, 6 days of gestation) or at term (37 weeks to 41 weeks, 6 days of gestation). We created a matched control cohort of women who gave birth with no history of ART. Control births were matched oneto-one by maternal age (within 1 year either way); parity (nulliparous vs. multiparous); type of delivery (planned or unplanned cesarean delivery vs. vaginal birth); and gestational age (within 1 week). Matching on these factors was performed by searching our perinatal database. For each case, we selected as the control the first available patient with the closest delivery date who met all matching criteria. Mothers were excluded if they had infants with significant congenital defects, or had conditions in which breastfeeding is contraindicated or not encouraged (eg, human immunodeficiency virus infection, psychiatric illness requiring sedating psychotherapeutic drugs, or active substance use that results in cognitive impairment).
insemination) that did not involve oocyte retrieval or ET were excluded. Pregnancies obtained by gamete and embryo donation were also excluded. The terms used to measure the breastfeeding outcomes for this study are derived from the WHO-recommended definitions (14). Exclusive breastfeeding was defined as infants being fed breast milk only (allowable exceptions were expressed breast milk, oral rehydration solutions, drops or syrups of vitamins, minerals, and medicines).
Data Collection Mother and infant demographic variables, the mother's medical and obstetric history, and information about labor and delivery were extracted from our computerized, researchquality, perinatal database. Pregnancy and delivery information for all patients who deliver at our center is entered into the database. The database is compiled prospectively by dedicated abstractors to ensure accuracy and minimize missing data; it is reviewed periodically to ensure its accuracy by direct comparison with the Registry of Births and the Hospital Discharge Data System. Data on breastfeeding outcomes were collected via structured telephone interviews that included study-specific questions. Cases and controls were invited to participate in the study via a telephone call; a telephone interview was scheduled with those who provided express oral consent. Study-specific fixed-choice questions were used to assess initiation, exclusivity, and continuation of breastfeeding. Data on initiation of breastfeeding, and whether mothers were breastfeeding at 6 weeks, were derived from retrospective questions: ‘‘Have you ever breastfed your baby?’’ (yes/ no) and ‘‘If ‘yes,’ for how long?’’ The response alternatives were as follows: 6 months. Data on how the infant was being fed at 3 months and 6 months postpartum were collected (recoded as exclusive breastfeeding, partial breastfeeding, and no breast milk). Timing of introduction of formula was also recorded. Women who had initiated breastfeeding were asked about their perception of exposure to 6 of the ‘‘ten steps to successful breastfeeding’’ during the postnatal hospital stay. Particularly, we assessed compliance with step 3 (information in pregnancy); step 4 (breastfeed within 1 hour after birth); step 6 (no human milk substitutes); step 7 (room-in), step 9 (pacifiers); and step 10 (postdischarge support). Table 1 presents the questions used to assess exposure to the steps. The exposure variables included: [1] lacking any 1 specific step, and [2] lacking any 2 steps. The construct of interest was ‘‘not receiving the care necessary for compliance with the steps.’’ At our institution, research involving analysis of existing data, or the use of surveys when the information is recorded in such a way that individuals cannot be identified, is exempt from formal institutional review board approval requirements.
Definitions and Terms Used We adopted the definition of ART as involving homologous in vitro fertilization and embryo transfer (ET). Women undergoing a fertility treatment (ovulation induction or artificial 90
Data Analysis We used paired t tests for measured outcomes with a symmetrical distribution, and Wilcoxon's signed rank test for VOL. 103 NO. 1 / JANUARY 2015
Fertility and Sterility®
TABLE 1 Questions used to assess lack of exposure of mothers who conceived using ART, compared with mothers who conceived spontaneously, to the ‘‘ten steps to successful breastfeeding.’’ Step 3 4 6 7 9 10
Question Did you receive information about the benefits of breastfeeding during pregnancy? Did you initiate breastfeeding more than 1 hour after birth? Was the infant fed water, sugar water, or formula? Did the infant stay in the room with you day and night? Did the hospital staff provide the infant with a pacifier during the hospital stay? Did the hospital provide you with information about local breastfeeding support groups before discharge?
Note: The ‘‘ten steps to successful breastfeeding’’ are those defined by WHO and UNICEF (13). ‘‘Lack of’’ was defined as ‘‘not receiving the care necessary for compliance with any step.’’
cases and controls in the proportion of mothers who received care that was lacking any 2 of the steps [29 of 75 (38.7%) vs. 34 of 75 (45.3%), P¼ .70]. Univariate conditional logistic regression was done to test the association of explanatory variables with breastfeeding cessation within 6 weeks. Only matched pairs who initiated breastfeeding were considered (n ¼ 76 pairs). A total of 18 variables were examined, including ART history, maternal age, obesity, parity, previous breastfeeding experience, highest level of education, employment outside the home, maternal smoking during pregnancy, mode of delivery, birth weight, admission to the neonatal intensive care unit, gestational age at birth, and lacking exposure to any step to successful breastfeeding (steps 3, 4, 6, 7, 9, and 10). Conception through ART was the only predictor of a short time period (less than 6 weeks) for breastfeeding (odds ratio ¼ 65.3 [95% confidence interval: 1.5–2889.3], P¼ .03).
Cromi. Breastfeeding after assisted conception. Fertil Steril 2015.
measured outcomes as a nonparametric alternative. Dichotomous outcomes were analyzed using McNemar's test. Univariable conditional logistic regression analysis was used to test the association of one explanatory variable at a time with the outcome of interest. A P value of .05 (two-tailed) was considered statistically significant. All statistical analysis was conducted using SPSS Release 20.0.0 (SPSS, Inc.).
RESULTS During the study period, 173 singletons were born after ART procedures at our institution. Of these, 40 were achieved through intrauterine insemination; 23 were achieved with heterologous techniques; and 8 were born before 34 gestational weeks, leaving 102 mother-baby pairs eligible for the study. Ninety-four mothers agreed to participate in the research and made up the case group. In this group, 50 (53.2%) pregnancies were obtained with intracytoplasmic sperm injection (ICSI), and 44 (46.8%) were achieved after homologous in vitro fertilization and ET. Each case was individually matched with one control. The characteristics of the mother-infant pairs, and labor and delivery outcomes in the case and control groups, are displayed in Table 2. Women who conceived after ART were more likely to be married than those who had conceived naturally. The 2 groups did not differ in any other sociodemographic or clinical variables. Breastfeeding outcomes in the study groups are displayed in Table 3. No statistically significant differences were found between the study groups in the proportion of women who attempted breastfeeding. Mothers who conceived through ART were at greater risk of stopping breastfeeding within the first 6 weeks postpartum. At 6-months postpartum, no differences were found in exclusive breastfeeding between women who used ART and women who conceived spontaneously. Data on maternal perception of compliance with 6 of the WHO and UNICEF ‘‘ten steps to successful breastfeeding’’ are displayed in Table 4. Nonadherence to individual steps was similar between groups. No difference was found between VOL. 103 NO. 1 / JANUARY 2015
DISCUSSION Our results suggest that women who give birth after experiencing infertility and undergoing ART are at higher risk for early breastfeeding cessation. At 1 month after giving birth, almost one quarter of women who used ART were not providing any breast milk. This finding is consistent with an earlier observation in a prospective cohort study by Hammarberg et al. (6). By 6 weeks after delivery, the proportion of ART women providing any breast milk had declined to 77%, and at 3 months, a smaller proportion of these mothers were breastfeeding exclusively, compared with other Australian mothers (46% vs. 57.3%). However, in that study (6), the ART group was compared with the general population of childbearing Australian women, and the ART and spontaneous conception groups differed significantly in characteristics such as age, education, parity, gestational age, twins rate, and birth weight, that could have contributed to the findings. Moreover, recruitment from infertility clinics rather than birth settings did not allow for the study to be controlled for policies and practices during the intrapartum and very early postnatal period that may have affected breastfeeding. As implementation of supportive interventions that improve breastfeeding outcomes is not consistent across health care services, the setting in which a woman gives birth should be considered a potential confounder that needs to be controlled for in comparative studies. O'Quinn et al. (7) suggest that in a Canadian context, the ART population is not different from mothers who conceive spontaneously, with regard to breastfeeding practices or experiences. The participants for that study were recruited from the community; thus, control subjects came from the same source population as the cases. However, no adjustment was made for factors that positively or negatively influence breastfeeding initiation and duration, including mode of delivery, the hospital in which the infants were born, whether the infant was admitted to the neonatal intensive care unit, and gestational age at delivery. Moreover, in that study, 1 in 3 pregnancies resulted from ovulation-inducing drugs only, which might have mitigated the actual effects that 91
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 2 Maternal and infant characteristics and obstetric outcome, for women who conceived using ART, compared with women who conceived spontaneously. Characteristic/Outcome Age (y) Parity Primiparous Multiparous Past breastfeeding experience Pre-pregnancy body mass index (BMI; kg/m2) Obesity (pre-pregnancy BMI >30 kg/m2) Gestational weight gain (kg) Maternal smoking Early pregnancy Throughout pregnancy Partner who smokes Marital status Married Common-law Single, separated, divorced, widowed Italian Foreign-born Maternal education Primary/secondary school High school graduate College graduate Occupation Employed Unemployed/student/housewife Induction of labor Mode of delivery Spontaneous vaginal Vacuum Cesarean Scheduled Urgent Pain medication during delivery General anesthesia Spinal/epidural No pain medications Infant birth weight (g) Birth weight
b
Department of Neonatology and Neonatal
Objective: To determine if breastfeeding outcomes differ between mothers who conceived spontaneously compared with those who conceived with assisted reproductive technology (ART). Design: Matched case-control study. Setting: Teaching hospital. Patient(s): Ninety-four women having a singleton baby conceived with ART, matched by maternal age, parity, mode of delivery, and gestational age to controls who conceived spontaneously. Intervention(s): Cases and controls were interviewed using a standardized, structured questionnaire, to obtain information on lactation. Exposure to maternity care practices contributing to breastfeeding success was investigated. Main Outcome Measure(s): Initiation, exclusivity, and continuation of breastfeeding. Result(s): Cases were as likely as controls to initiate breastfeeding (89.4% vs. 90.4%), but by 6 weeks postpartum, a greater proportion of mothers who conceived through ART has ceased breastfeeding (20.2% vs. 5.3%). The percentage of mothers who exclusively breastfed their child for 6 months was similar among the 2 groups. On univariate conditional logistic regression, a history of using ART was the only predictor of early breastfeeding cessation (odds ratio ¼ 65.3 [95% confidence interval: 1.5–2889.3]). Conclusion(s): Women who have conceived with ART should be regarded as being at higher risk for early breastfeeding cessation. This study serves as a first step in the investigation of potential modifiable factors that contribute to breastfeeding failure among women who give birth after using ART, and may help in efforts to customize breastfeeding support strategies. Use your smartphone (Fertil SterilÒ 2015;103:89–94. Ó2015 by American Society for Reproductive Medicine.) to scan this QR code Key Words: Assisted reproductive technology, breastfeeding, assisted conception, lactation, and connect to the parenting Discuss: You can discuss this article with its authors and with other ASRM members at http:// fertstertforum.com/cromia-art-breastfeeding-outcomes/
B
reastfeeding is recognized as the best source of nutrition for most infants, and exclusive breastfeeding is recommended for the first 6 months of life by many authorities, including the World Health Organization (WHO; 1, 2). Few health behaviors have such a broad-spectrum and longlasting impact on health status, with the potential to improve life chances, health, and well-being (3, 4).
With births occurring increasingly as a result of assisted reproductive technology (ART), the impact of infertility treatments on initiation and duration of breastfeeding is an important issue to explore. Whereas obstetric outcomes among women who conceive as a result of ART have been extensively studied, only a moderate body of research addresses maternal postnatal adjustment, and psychological aspects of early
Received August 7, 2014; revised October 6, 2014; accepted October 7, 2014; published online October 29, 2014. A.C. has nothing to disclose. M.S. has nothing to disclose. I.C. has nothing to disclose. S.U. has nothing to disclose. S.S. has nothing to disclose. M.A. has nothing to disclose. F.G. has nothing to disclose. Reprint requests: Antonella Cromi, Ph.D., Department of Obstetrics and Gynecology, University of Insubria, Del Ponte Hospital, Piazza Biroldi 1, 21100 Varese, Italy (E-mail: antonella.comi@ uninsubria.it). Fertility and Sterility® Vol. 103, No. 1, January 2015 0015-0282/$36.00 Copyright ©2015 American Society for Reproductive Medicine, Published by Elsevier Inc. http://dx.doi.org/10.1016/j.fertnstert.2014.10.009 VOL. 103 NO. 1 / JANUARY 2015
discussion forum for this article now.*
* Download a free QR code scanner by searching for “QR scanner” in your smartphone’s app store or app marketplace.
parenting, after conception with ART (5). The limited evidence about the potential impact of an ART treatment history on breastfeeding capacity is inconclusive and contradictory, mainly because of significant methodological limitations in the studies to date, including recruitment strategies, inadequate sample sizes, lack of appropriate comparison groups, and failure to control for known risk factors for lactation difficulties (6–9). Women who conceive with ART are more likely to be first-time mothers; they are, on average, older and more likely to have pregnancy complications, multiple births, operative delivery, labor induction, premature birth, and low–birth weight babies than 89
ORIGINAL ARTICLE: ASSISTED REPRODUCTION women who conceive spontaneously. Thus, comparisons made with data drawn from the general population of childbearing women need to be interpreted with extreme caution. Moreover, understanding breastfeeding outcomes requires an exploration beyond maternal and newborn factors that could affect lactation success. Maternity care policy and practices, along with provider and hospital factors, may have an impact on successful breastfeeding (10, 11). Therefore, studies recruiting participants from infertility clinics may fail to control for confounders related to the setting in which a woman gives birth. We designed a case-control study to investigate initiation, duration, and exclusivity of breastfeeding among every eligible case of women during the study period who had conceived with the help of ART, compared with an appropriately matched control group of patients who conceived spontaneously and gave birth at the same hospital.
MATERIALS AND METHODS Setting The study was conducted in the Obstetrics Department of Del Ponte Hospital at the University of Insubria, in Varese, Italy, a teaching and tertiary referral center for high- and low-risk obstetrics. Although the ‘‘Baby-Friendly Hospital Initiative’’ designation (12) was not taken into consideration, the changes required for successful implementation of the ‘‘ten steps to successful breastfeeding,’’ developed by WHO and UNICEF (13), had been undertaken at our institution long before the study was conducted. Maternity care practices related to breastfeeding did not change over the study period.
Study Sample Cases were defined as all women conceiving through ART who gave birth at our institution between January 1, 2010 and October 31, 2013. We included live singleton births, either late-preterm (34 weeks to 36 weeks, 6 days of gestation) or at term (37 weeks to 41 weeks, 6 days of gestation). We created a matched control cohort of women who gave birth with no history of ART. Control births were matched oneto-one by maternal age (within 1 year either way); parity (nulliparous vs. multiparous); type of delivery (planned or unplanned cesarean delivery vs. vaginal birth); and gestational age (within 1 week). Matching on these factors was performed by searching our perinatal database. For each case, we selected as the control the first available patient with the closest delivery date who met all matching criteria. Mothers were excluded if they had infants with significant congenital defects, or had conditions in which breastfeeding is contraindicated or not encouraged (eg, human immunodeficiency virus infection, psychiatric illness requiring sedating psychotherapeutic drugs, or active substance use that results in cognitive impairment).
insemination) that did not involve oocyte retrieval or ET were excluded. Pregnancies obtained by gamete and embryo donation were also excluded. The terms used to measure the breastfeeding outcomes for this study are derived from the WHO-recommended definitions (14). Exclusive breastfeeding was defined as infants being fed breast milk only (allowable exceptions were expressed breast milk, oral rehydration solutions, drops or syrups of vitamins, minerals, and medicines).
Data Collection Mother and infant demographic variables, the mother's medical and obstetric history, and information about labor and delivery were extracted from our computerized, researchquality, perinatal database. Pregnancy and delivery information for all patients who deliver at our center is entered into the database. The database is compiled prospectively by dedicated abstractors to ensure accuracy and minimize missing data; it is reviewed periodically to ensure its accuracy by direct comparison with the Registry of Births and the Hospital Discharge Data System. Data on breastfeeding outcomes were collected via structured telephone interviews that included study-specific questions. Cases and controls were invited to participate in the study via a telephone call; a telephone interview was scheduled with those who provided express oral consent. Study-specific fixed-choice questions were used to assess initiation, exclusivity, and continuation of breastfeeding. Data on initiation of breastfeeding, and whether mothers were breastfeeding at 6 weeks, were derived from retrospective questions: ‘‘Have you ever breastfed your baby?’’ (yes/ no) and ‘‘If ‘yes,’ for how long?’’ The response alternatives were as follows: 6 months. Data on how the infant was being fed at 3 months and 6 months postpartum were collected (recoded as exclusive breastfeeding, partial breastfeeding, and no breast milk). Timing of introduction of formula was also recorded. Women who had initiated breastfeeding were asked about their perception of exposure to 6 of the ‘‘ten steps to successful breastfeeding’’ during the postnatal hospital stay. Particularly, we assessed compliance with step 3 (information in pregnancy); step 4 (breastfeed within 1 hour after birth); step 6 (no human milk substitutes); step 7 (room-in), step 9 (pacifiers); and step 10 (postdischarge support). Table 1 presents the questions used to assess exposure to the steps. The exposure variables included: [1] lacking any 1 specific step, and [2] lacking any 2 steps. The construct of interest was ‘‘not receiving the care necessary for compliance with the steps.’’ At our institution, research involving analysis of existing data, or the use of surveys when the information is recorded in such a way that individuals cannot be identified, is exempt from formal institutional review board approval requirements.
Definitions and Terms Used We adopted the definition of ART as involving homologous in vitro fertilization and embryo transfer (ET). Women undergoing a fertility treatment (ovulation induction or artificial 90
Data Analysis We used paired t tests for measured outcomes with a symmetrical distribution, and Wilcoxon's signed rank test for VOL. 103 NO. 1 / JANUARY 2015
Fertility and Sterility®
TABLE 1 Questions used to assess lack of exposure of mothers who conceived using ART, compared with mothers who conceived spontaneously, to the ‘‘ten steps to successful breastfeeding.’’ Step 3 4 6 7 9 10
Question Did you receive information about the benefits of breastfeeding during pregnancy? Did you initiate breastfeeding more than 1 hour after birth? Was the infant fed water, sugar water, or formula? Did the infant stay in the room with you day and night? Did the hospital staff provide the infant with a pacifier during the hospital stay? Did the hospital provide you with information about local breastfeeding support groups before discharge?
Note: The ‘‘ten steps to successful breastfeeding’’ are those defined by WHO and UNICEF (13). ‘‘Lack of’’ was defined as ‘‘not receiving the care necessary for compliance with any step.’’
cases and controls in the proportion of mothers who received care that was lacking any 2 of the steps [29 of 75 (38.7%) vs. 34 of 75 (45.3%), P¼ .70]. Univariate conditional logistic regression was done to test the association of explanatory variables with breastfeeding cessation within 6 weeks. Only matched pairs who initiated breastfeeding were considered (n ¼ 76 pairs). A total of 18 variables were examined, including ART history, maternal age, obesity, parity, previous breastfeeding experience, highest level of education, employment outside the home, maternal smoking during pregnancy, mode of delivery, birth weight, admission to the neonatal intensive care unit, gestational age at birth, and lacking exposure to any step to successful breastfeeding (steps 3, 4, 6, 7, 9, and 10). Conception through ART was the only predictor of a short time period (less than 6 weeks) for breastfeeding (odds ratio ¼ 65.3 [95% confidence interval: 1.5–2889.3], P¼ .03).
Cromi. Breastfeeding after assisted conception. Fertil Steril 2015.
measured outcomes as a nonparametric alternative. Dichotomous outcomes were analyzed using McNemar's test. Univariable conditional logistic regression analysis was used to test the association of one explanatory variable at a time with the outcome of interest. A P value of .05 (two-tailed) was considered statistically significant. All statistical analysis was conducted using SPSS Release 20.0.0 (SPSS, Inc.).
RESULTS During the study period, 173 singletons were born after ART procedures at our institution. Of these, 40 were achieved through intrauterine insemination; 23 were achieved with heterologous techniques; and 8 were born before 34 gestational weeks, leaving 102 mother-baby pairs eligible for the study. Ninety-four mothers agreed to participate in the research and made up the case group. In this group, 50 (53.2%) pregnancies were obtained with intracytoplasmic sperm injection (ICSI), and 44 (46.8%) were achieved after homologous in vitro fertilization and ET. Each case was individually matched with one control. The characteristics of the mother-infant pairs, and labor and delivery outcomes in the case and control groups, are displayed in Table 2. Women who conceived after ART were more likely to be married than those who had conceived naturally. The 2 groups did not differ in any other sociodemographic or clinical variables. Breastfeeding outcomes in the study groups are displayed in Table 3. No statistically significant differences were found between the study groups in the proportion of women who attempted breastfeeding. Mothers who conceived through ART were at greater risk of stopping breastfeeding within the first 6 weeks postpartum. At 6-months postpartum, no differences were found in exclusive breastfeeding between women who used ART and women who conceived spontaneously. Data on maternal perception of compliance with 6 of the WHO and UNICEF ‘‘ten steps to successful breastfeeding’’ are displayed in Table 4. Nonadherence to individual steps was similar between groups. No difference was found between VOL. 103 NO. 1 / JANUARY 2015
DISCUSSION Our results suggest that women who give birth after experiencing infertility and undergoing ART are at higher risk for early breastfeeding cessation. At 1 month after giving birth, almost one quarter of women who used ART were not providing any breast milk. This finding is consistent with an earlier observation in a prospective cohort study by Hammarberg et al. (6). By 6 weeks after delivery, the proportion of ART women providing any breast milk had declined to 77%, and at 3 months, a smaller proportion of these mothers were breastfeeding exclusively, compared with other Australian mothers (46% vs. 57.3%). However, in that study (6), the ART group was compared with the general population of childbearing Australian women, and the ART and spontaneous conception groups differed significantly in characteristics such as age, education, parity, gestational age, twins rate, and birth weight, that could have contributed to the findings. Moreover, recruitment from infertility clinics rather than birth settings did not allow for the study to be controlled for policies and practices during the intrapartum and very early postnatal period that may have affected breastfeeding. As implementation of supportive interventions that improve breastfeeding outcomes is not consistent across health care services, the setting in which a woman gives birth should be considered a potential confounder that needs to be controlled for in comparative studies. O'Quinn et al. (7) suggest that in a Canadian context, the ART population is not different from mothers who conceive spontaneously, with regard to breastfeeding practices or experiences. The participants for that study were recruited from the community; thus, control subjects came from the same source population as the cases. However, no adjustment was made for factors that positively or negatively influence breastfeeding initiation and duration, including mode of delivery, the hospital in which the infants were born, whether the infant was admitted to the neonatal intensive care unit, and gestational age at delivery. Moreover, in that study, 1 in 3 pregnancies resulted from ovulation-inducing drugs only, which might have mitigated the actual effects that 91
ORIGINAL ARTICLE: ASSISTED REPRODUCTION
TABLE 2 Maternal and infant characteristics and obstetric outcome, for women who conceived using ART, compared with women who conceived spontaneously. Characteristic/Outcome Age (y) Parity Primiparous Multiparous Past breastfeeding experience Pre-pregnancy body mass index (BMI; kg/m2) Obesity (pre-pregnancy BMI >30 kg/m2) Gestational weight gain (kg) Maternal smoking Early pregnancy Throughout pregnancy Partner who smokes Marital status Married Common-law Single, separated, divorced, widowed Italian Foreign-born Maternal education Primary/secondary school High school graduate College graduate Occupation Employed Unemployed/student/housewife Induction of labor Mode of delivery Spontaneous vaginal Vacuum Cesarean Scheduled Urgent Pain medication during delivery General anesthesia Spinal/epidural No pain medications Infant birth weight (g) Birth weight
