Advertisement
JOGC

Perinatal Factors Associated With Breastfeeding Trends After Preterm Birth <29 Weeks Gestation: Can We Predict Early Discontinuation?

Open AccessPublished:November 24, 2022DOI:https://doi.org/10.1016/j.jogc.2022.11.002

      Abstract

      Objective

      To determine the rates and perinatal factors associated with initiation and early discontinuation of breastfeeding among very preterm neonates.

      Methods

      This was a retrospective cohort study of very preterm infants (<29 weeks gestation) admitted to 2 regional Level III neonatal intensive care units (NICUs) from January 1, 2015, to December 31, 2019. A national neonatal database was used to evaluate initiation and continuation rates of breastfeeding and associated perinatal factors. Stored nutrition profiles and delivery record books were used to determine feeding volumes associated with continuation of breastfeeding to hospital discharge for a subgroup of infants at a single site. Descriptive and inferential statistics were used to present the results between groups, and logistic regression modeling was used to calculate crude and adjusted odds ratios (OR) and 95% CI.

      Results

      Of 391 eligible neonates, 84% initiated breastfeeding but only 38% continued to discharge. Interestingly, frequency of breastfeeding initiation (P < 0.001) and continuation (P < 0.001) declined over the study period. After adjustment for confounders, younger maternal age, earlier gestational age, cigarette smoking, and multiparity were significantly associated with early discontinuation of breastfeeding prior to hospital discharge. Early discontinuation of breastfeeding was also related to lower volumes of breastmilk by day 7 of life (P = 0.004).

      Conclusion

      Very preterm neonates are at high risk for non-initiation and early discontinuation of breastfeeding. The early postnatal period represents a critical time to establish breastmilk volumes, and the identification of key perinatal risk factors allows for early and targeted breastfeeding support.

      Résumé

      Objectif

      Déterminer les taux d’instauration et d’arrêt précoce de l’allaitement chez les grands prématurés et définir les facteurs périnataux associés.

      Méthodologie

      Il s’agit d’une étude de cohorte rétrospective de grands prématurés (< 29 SA) admis dans deux unités régionales de soins intensifs néonataux de niveau III entre le 1er janvier 2015 et le 31 décembre 2019. Une base de données nationale sur les nouveau-nés a été utilisée pour évaluer les taux d’instauration et de continuation de l’allaitement et définir les facteurs périnataux associés. Les profils d’alimentation conservés et les registres d’accouchement ont été utilisés pour déterminer les volumes d’alimentation associés à la continuation de l’allaitement jusqu’à la sortie de l’hôpital pour un sous-groupe de nourrissons d’un seul site. Des statistiques descriptives et inférentielles ont été utilisées pour présenter les résultats entre les groupes, et la modélisation de régression logistique a été utilisée pour calculer les rapports de cotes bruts et ajustés et l’intervalle de confiance à 95 %.

      Résultats

      L’allaitement a été instauré chez 84 % des 391 nouveau-nés admissibles, mais il a été maintenu jusqu’au congé pour seulement 38 %. Il est à remarquer que la fréquence d’instauration (p < 0,001) et de continuation (p < 0,001) de l’allaitement a diminué pendant la période à l’étude. Après ajustement en fonction des facteurs de confusion, on observe que le jeune âge maternel, le jeune âge gestationnel, le tabagisme et la multiparité sont significativement associés à l’arrêt précoce de l’allaitement avant le congé de l’hôpital. L’arrêt précoce de l’allaitement est aussi lié au faible volume de lait maternel au jour 7 de vie (p = 0,004).

      Conclusion

      Les grands prématurés présentent un risque élevé de non-instauration et d’arrêt précoce de l’allaitement. Le début de la période postnatale est un moment critique pour établir le volume de lait maternel, et la reconnaissance des principaux facteurs de risque périnataux permet de donner un soutien à l’allaitement précoce et ciblé.

      Keywords

      Introduction

      Exclusive breastfeeding for the first 6 months of life with continued breastfeeding for at least 2 years is recommended by the World Health Organization and the Canadian Paediatric Society.
      World Health Organization
      Health topics: breastfeeding.
      ,
      • Boland M.C.
      Exclusive breastfeeding should continue to six months.
      Breastfeeding is especially important in the preterm population; benefits include lower rates of necrotizing enterocolitis, re-hospitalization, sepsis, and overall mortality, as well as improved growth, neurodevelopment, and infant-parent attachment.
      • Callen J.
      • Pinelli J.
      A review of the literature examining the benefits and challenges, incidence and duration and barriers to breastfeeding in preterm infants.
      • Corpeleijn W.E.
      • Kouwenhoven S.M.P.
      • Paap M.C.
      • et al.
      Intake of own mother’s milk during the first days of life is associated with decreased morbidity and mortality in very low birth weight infants during the first 60 days of life.
      • Vohr B.R.
      • Poindexter B.B.
      • Dusick A.M.
      • et al.
      Persistent beneficial effects of breast milk ingested in the neonatal intensive care unit on outcomes of extremely low birth weight infants at 30 months of age.
      There are also long-term benefits of breastfeeding for offspring and mothers.
      • Halipchuk J.
      • Temple B.
      • Dart A.
      • et al.
      Prenatal, obstetric and perinatal factors associated with the development of childhood-onset type 2 diabetes.
      • Ma J.
      • Qiao Y.
      • Zhao P.
      • et al.
      Breastfeeding and childhood obesity: a 12-country study.
      • Sattari M.
      • Serwint J.
      • Levine D.M.
      Maternal implications of breastfeeding: a review for the internist.
      However, premature infants are at risk for not breastfeeding because of immature physiologic feeding mechanisms, barriers to establishing milk supply, and related comorbidities of prematurity.
      • Callen J.
      • Pinelli J.
      A review of the literature examining the benefits and challenges, incidence and duration and barriers to breastfeeding in preterm infants.
      In addition, recommended strategies to improve breastmilk supply, such as skin-to-skin contact, rooming-in, and early breastmilk expression, become even more difficult to establish for those admitted to the neonatal intensive care unit (NICU).
      • Spatz D.
      Beyond BFHI: the Spatz 10-step and breastfeeding resource nurse model to improve human milk and breastfeeding outcomes.
      ,
      • Ikonen R.
      • Paavilainen E.
      • Kaunonen M.
      Preterm infants’ mothers’ experiences with milk expression and breastfeeding: an integrative review.
      With improved awareness of the benefits of breastfeeding, initiation rates in Canada have increased from <25% to 90% over the past 50 years.
      • Green M.
      • Chalmers B.
      • Hanvey L.
      • et al.
      Chapter 6: breastfeeding. Family-centered maternity and newborn care: national guidelines.
      However, early discontinuation rates remain high, with close to 25% of all parents stopping breastfeeding before their infant is 1 month old.
      • Green M.
      • Chalmers B.
      • Hanvey L.
      • et al.
      Chapter 6: breastfeeding. Family-centered maternity and newborn care: national guidelines.
      Breastfeeding rates are lower among mothers with lower educational status and household income, younger age, those who smoke, and those with comorbid medical conditions.
      • Cohen S.S.
      • Alexander D.D.
      • Krebs N.F.
      • et al.
      Factors associated with breastfeeding initiation and continuation: a meta-analysis.
      • Heaman M.I.
      • Martens P.J.
      • Brownell M.D.
      • et al.
      The association of inadequate and intensive prenatal care with maternal, fetal and infant outcomes: a population-based study in Manitoba, Canada.
      • Ramji N.
      • Quinlan J.
      • Murphy P.
      • et al.
      The impact of maternal obesity on breastfeeding.
      Cesarean delivery—increased among preterm births—is associated with delayed milk production and breastfeeding difficulties.
      • Cohen S.S.
      • Alexander D.D.
      • Krebs N.F.
      • et al.
      Factors associated with breastfeeding initiation and continuation: a meta-analysis.
      However, these risk factors are less well-studied in the preterm population, and interpreting existing studies is challenging as many lack differentiation between parental breastmilk and donor milk.
      • Callen J.
      • Pinelli J.
      A review of the literature examining the benefits and challenges, incidence and duration and barriers to breastfeeding in preterm infants.
      ,
      • Ericson J.
      • Eriksson M.
      • Hoddinott P.
      • et al.
      Breastfeeding and risk for ceasing in mothers of preterm infants – long-term follow-up.
      ,
      • Wang Y.
      • Briere C.E.
      • Xu W.
      • et al.
      Factors affecting breastfeeding outcomes at six months in preterm infants.
      Additionally, interventions for improving breastmilk supply for the general population need specific study among preterm infants.
      • Soltani H.
      • Scott A.M.S.
      Antenatal breast expression in women with diabetes: outcomes from a retrospective cohort study.
      ,
      • Foudil-Bey I.
      • Murphy M.S.Q.
      • Dunn S.
      • et al.
      Evaluating antenatal breastmilk expression outcomes: a scoping review.
      Because breastfeeding is associated with improved short- and long-term outcomes, interventions to improve initiation and duration among those neonates deliveringed preterm are urgently needed.
      • Pylypjuk C.
      • Bokhanchuk A.
      • Day C.
      • et al.
      Antenatal breastfeeding promotion amongst pregnancies at high-risk for newborn admission to the NICU: a cross-sectional study.
      The aim of this study was to evaluate breastfeeding patterns and the perinatal factors associated with breastfeeding initiation and continuation among very preterm neonates admitted to the NICU. The results will allow for a more targeted approach to breastfeeding support for those at highest risk of early discontinuation.

      Methods

      This was a retrospective cohort study of preterm neonates delivered <29 weeks gestation and admitted to 2 regional level III NICUs. The Canadian Neonatal Network electronic database was used to identify eligible cases (January 1, 2015, to December 31, 2019). The database is maintained by a multi-disciplinary network of trained abstractors using standardized definitions and is audited regularly.
      • Shah P.S.
      • Siedlitz W.
      • Chan P.
      • et al.
      Internal audit of the Canadian Neonatal Network Data Collection System.
      From the database, information about breastfeeding, maternal demographics, pregnancy complications, birth events, and newborn characteristics were collected. The Score for Neonatal Acute Physiology (SNAP-II)—which is an illness severity score that correlates with risk of death in very preterm infants—was also collected.
      • Dammann O.
      • Shah B.
      • Naples M.
      • et al.
      Interinstitutional variation in prediction of death by SNAP-II and SNAPPE-II among extremely preterm infants.
      Neonates who were moribund on admission were ineligible. Breastfeeding was defined as feeding of the biologic parent’s own breastmilk (either direct feeding at the breast or feeding of expressed breastmilk). A subgroup of neonates admitted to a single site with stored nutrition profiles were additionally evaluated to determine the exact daily feeding volumes and sources associated with duration of breastfeeding. The primary outcome for our study was the rate of breastfeeding initiation and continuation to discharge. The secondary outcomes included the perinatal factors and feeding volumes associated with breastfeeding initiation and duration.
      Descriptive and inferential statistics (i.e., t tests, χ2 tests, Wilcoxon rank sum tests) were used to present the results and compare outcomes between groups (breastfeeding initiation vs. non-initiation and continuation vs. early discontinuation). Multivariate logistic regression modelling adjusting for potential confounders (those factors found to be significant in univariate analysis) was used to generate adjusted OR (aOR) with 95% CIs. A P value of < 0.05 was used to assume statistical significance for the primary outcome and < 0.01 for the secondary outcomes. The main statistical analyses were performed using SAS software version 9.4 (SAS Institute, Cary, NC). Stata 12.0 (Stata Corp., LLC, College Station, TX) was used for the subgroup analyses of feeding volumes. Ethics approval for this project was obtained from our university human research ethics board (H2021:128 [HS24762]) and the Executive Committee of the Canadian Neonatal Network.

      Results

      From the database, there were 391 eligible neonates admitted to the NICU during the study period (Figure 1). Thirty-five percent of neonates were delivered to multiparous patients with a mean maternal age of 30 years and at a mean gestational age of 26 weeks (Table 1). Over half (55%) were delivered by cesarean (Table 1). Neonatal survival to hospital discharge was 81%. Breastfeeding was initiated in 84% (n = 329) of very preterm neonates but was continued to hospital discharge in only 38% (n = 125). Interestingly, there was a significant decline in breastfeeding initiation (P < 0.001) and continuation to hospital discharge (P < 0.001) over the study period (Figure 2).
      Table 1Perinatal factors associated with BF during NICU admission following preterm birth <29 weeks gestation (January 1, 2015–December 31, 2019)
      BF: breastfeeding; NICU: neonatal intensive care unit; SNAP: Score for Neonatal Acute Physiology.
      VariableNo. (%)
      Unless otherwise specified.
      P valueNo. (%)
      Unless otherwise specified.
      P value
      Non-initiationof BF n = 62Initiationof BF n = 329Early discontinuationof BF n = 204Continuationof BF n = 125
      Maternal history and obstetric complications
       Maternal age, mean (SD), y30.1 (5.6)30.4 (6.4)0.7229.5 (7.0)31.9 (5.2)< 0.01
       Cigarette smoking19 (31%)43 (13%)< 0.0140 (20%)3 (2%)< 0.01
       Multiparity (>1)39 (64%)99 (30%)< 0.0177 (38%)22 (18%)< 0.01
       Maternal hypertension10 (18%)48 (15%)0.5925 (13%)23 (19%)0.13
       Maternal diabetes8 (14%)27 (8%)0.1820 (10%)7 (6%)0.18
       Rupture of membranes (>24 h)24 (40%)91 (29%)0.0865 (33%)26 (21%)0.02
       Multiple gestation11 (18%)86 (26%)0.1653 (26%)33 (26%)0.93
       Chorioamnionitis9 (19%)27 (10%)0.0617 (10%)10 (9%)0.89
       Abruption/antepartum hemorrhage4 (6%)14 (4%)0.4510 (5%)4 (3%)0.46
      Neonatal and delivery events
       Small for gestational age4 (6%)18 (5%)0.7614 (7%)4 (3%)0.21
       Gestational age, mean (SD), w25.7 (1.7)26.2 (1.6)0.0625.9 (1.7)26.6 (1.3)< 0.01
       Cesarean delivery38 (61%)176 (54%)0.26106 (52%)70 (56%)0.48
       Antenatal corticosteroids51 (82%)284 (87%)0.31175 (87%)109 (88%)0.74
       Birth weight, mean (SD), g887 (255)963 (239)0.02938 (258)1002 (198)0.01
       Male sex40 (65%)183 (56%)0.19114 (56%)69 (55%)0.90
       5-min Apgar <733 (53%)138 (42%)0.1190 (45%)48 (38%)0.27
       Outborn5 (8%)41 (12%)0.3233 (16%)8 (6%)0.01
       SNAP score, median (IQR)14 (5, 29)9 (0, 21)0.0714 (0, 24)9 (0, 14)< 0.01
       Survival to discharge33 (53%)285 (87%)< 0.01161 (79%)124 (99%)< 0.01
      a Unless otherwise specified.
      Figure thumbnail gr2
      Figure 2Cross-sectional trends in initiation and continuation of BF and usage of donor milk by year of study.
      BF: breastfeeding; DM: donor milk.
      Although there was no difference in maternal age between groups (P = 0.72), there were significantly more multiparas (64% vs. 30%; P < 0.01) and cigarette smokers (31% vs. 13%; P < 0.01) among those that did not initiate breastfeeding. There was also a trend towards lower birth weight among those that did not initiate breastfeeding (887 g [SD 255] vs. 963 g [SD 239]; P = 0.02). Otherwise, there were no significant differences in perinatal factors between those initiating breastfeeding and those that did not (Table 1). However, compared to those who continued breastfeeding to hospital discharge, the early discontinuation group was comprised of younger mothers (29.5 years [SD 7.0] vs. 31.9 years [SD 5.2]; P < 0.01), more multiparas (38% vs. 18%; P < 0.01), and more smokers (20% vs. 2%; P < 0.01) (Table 1). Neonates in the early discontinuation group were born at an earlier mean gestational age (25.9 weeks [SD 1.7] vs. 26.6 weeks [SD 1.3]; P < 0.01) with higher SNAP-II scores (14 [IQR 0, 24] vs. 9 [IQR 0,14]; P < 0.01), were more often outborn (16% vs. 6%; P = 0.01), and were more often born after prolonged rupture of membranes >24 hours (33% vs. 21%; P = 0.02). Neonatal survival to hospital discharge was more common in the breastfeeding initiation and continuation groups (Table 1).
      After adjustment for potential confounders, cigarette smoking (aOR 8.02; 95% CI 2.27–28.28) and multiparity (aOR 3.14; 95% CI 1.66–5.98) remained strongly associated with a higher odds of early breastfeeding discontinuation (Table 2). Conversely, older gestational age was associated with 32% lower odds of early discontinuation (aOR 0.68; 95% CI 0.57–0.81), and older maternal age with a 7% reduction in odds (aOR 0.93; 95% CI 0.89–0.97). Rupture of membranes >24 hours and outborn status were not found to be independently associated with early discontinuation after adjustment (Table 2).
      Table 2Perinatal factors associated with early discontinuation of BF prior to NICU discharge using multiple logistic regression analysis
      aOR: adjusted odds ratio; BF: breastfeeding; NICU: neonatal intensive care unit.
      VariableUnadjusted OR (95% CI)
      The unadjusted ORs (95% CIs) of early discontinuation of BF were estimated based on the univariate regression analysis (with continuing BF to hospital discharge as the reference group).
      P valueaOR (95% CI)
      The aORs (95% CIs) were determined after adjusting for maternal age, cigarette smoking, multiparity, rupture of membranes (>24 h), gestational age, and outborn status.
      P value
      Maternal age, mean (SD), y0.94 (0.91–0.98)< 0.010.93 (0.89–0.97)< 0.01
      Cigarette smoking9.90 (2.99–32.75)< 0.018.02 (2.27–28.28)< 0.01
      Multiparity (>1)2.83 (1.65–4.87)< 0.013.14 (1.66–5.98)< 0.01
      Rupture of membranes (>24 h)1.85 (1.10–3.13)0.021.59 (0.88–2.88)0.13
      Gestational age, mean (SD), w0.75 (0.65–0.88)< 0.010.68 (0.57–0.81)< 0.01
      Outborn2.82 (1.26–6.33)0.011.49 (0.55–4.03)0.43
      a The unadjusted ORs (95% CIs) of early discontinuation of BF were estimated based on the univariate regression analysis (with continuing BF to hospital discharge as the reference group).
      b The aORs (95% CIs) were determined after adjusting for maternal age, cigarette smoking, multiparity, rupture of membranes (>24 h), gestational age, and outborn status.
      Two hundred and seven neonates born <29 weeks had stored nutrition profiles available for analysis and 187 initiated breastfeeding. The first day of breastfeeding initiation was later in the early discontinuation group (day 5 of life [IQR 3, 7.5]) compared to those continuing breastfeeding to discharge (day 4 of life [IQR 2, 6]; P < 0.01) (Table 3). Both groups experienced an increase in daily volumes of breastmilk between days 3 and 7, although there were higher total volumes achieved in the continuation group (Figure 3). By day 7, those in the continuation group had a median proportion of total daily feeds by breastmilk of 100% compared to only 50% in the early discontinuation group (P = 0.004). After day 7, there was significant divergence in breastmilk volumes between groups (Figure 3). For those in the early discontinuation group, there was a plateau in the absolute volume of breastmilk and a sharp decline in the relative proportion of daily feeds from parent’s own breastmilk after day 7, and a steady decrease in both volume measures after day 14 (Figure 3, Table 3). Conversely, those with continued breastfeeding to hospital discharge exhibited a sharp increase in absolute daily volumes of breastmilk, and a sustained proportion of total daily feeds from breastmilk close to 100% from day 14 onwards (Figure 3). The percentage of total daily feeds of breastmilk in the continued breastfeeding group was as high as 95% (311 mL) on the day of discharge (Table 3). Interestingly, there was no difference in formula supplementation between those who continued breastfeeding to hospital discharge and those who did not in the first 28 days of life (Table 3). There was also no significant difference in donor milk usage between groups (P = 0.188).
      Table 3Trends in daily feeding volumes between continuation of BF versus early discontinuation groups for neonates born <29 weeks and admitted to NICU with available nutritional profiles from a single site (n = 187)
      BF: breastfeeding; NICU: neonatal intensive care unit.
      Variable of interestContinuation of BF n = 110Discontinuation of BF n = 77P value
      Day of BF initiation, median (IQR)4 (2, 6)5 (3, 7.5)< 0.001
      BF on day 3 of life
       BF volume in mL, median (IQR)12 (10.8, 64)0 (0, 6)0.021
       BF volume as a percentage of total daily feeds, median (IQR)25 (0, 50)0 (0, 50)0.026
       Receipt of formula, % of neonates that received formula3.6%2.6%0.702
      BF on day 7 of life
       BF volume in mL, median (IQR)40.5 (9.6, 103.8)12 (0, 69)0.004
       BF volume as a percentage of total daily feeds, median (IQR)100 (50, 100)50 (0, 100)0.004
       Receipt of formula, % of neonates that received formula0%1.3%0.232
      BF on day 14 of life
       BF volume in mL, median (IQR)120 (46, 174)21.8 (0, 120)< 0.001
       BF volume as a percentage of total daily feeds, median (IQR)100 (62.5, 100)25 (0, 100)< 0.001
       Receipt of formula, % of neonates that received formula0.9%2.6%0.367
      BF on day 21 of life
       BF volume in mL, median (IQR)156 (66, 216)0 (0, 126)< 0.001
       BF volume as a percentage of total daily feeds, median (IQR)100 (75, 100)0 (0, 100)< 0.001
       Receipt of formula, % of neonates that received formula6.4%7.8%0.712
      BF on day 28 of life
       BF volume in mL, median (IQR)174 (91.8, 243)6 (0, 125)< 0.001
       BF volume as a percentage of total daily feeds, median (IQR)100 (50, 100)12.5 (0, 100)< 0.001
       Receipt of formula, % of neonates that received formula6.4%13%0.124
      BF on day of discharge
       BF volume in mL, median (IQR)
      n of 64 because of unreported volumes from ad lib BF at time of discharge.
      311.2 (168, 436.1)N/A-
       BF volume as a percentage of total daily feeds, median (IQR)
      n of 64 because of unreported volumes from ad lib BF at time of discharge.
      95 (50, 97)N/A-
       Receipt of formula, % of neonates that received formula30%100%< 0.001
      Receipt of donor milk, % of neonates that received formula82.7%89.6%0.188
      a n of 64 because of unreported volumes from ad lib BF at time of discharge.
      Figure thumbnail gr3
      Figure 3Breastfeeding volumes of preterm neonates born <29 weeks and admitted to NICU in millilitres per day (A) and as a percentage of total daily feeds (B), compared between groups (continuation of breastfeeding vs. early discontinuation).
      BF: breastfeeding; NICU: neonatal intensive care unit.

      Discussion

      Even with the known challenges of breastfeeding for very preterm neonates, the overall initiation rate in our region was high at 84%. Unfortunately, more than half (62%) of patients discontinued breastfeeding prior to discharge from NICU, further highlighting how barriers to breastfeeding may disproportionately impact families with a prolonged neonatal hospital stay. Despite increasing awareness about the benefits of breastfeeding, there was a trend towards declining initiation and continuation rates over the study period. It is unclear if general practice changes towards earlier postpartum discharges for mothers may hinder the early breastfeeding supports necessary to establish an adequate milk supply in this high-risk population. Local policy and staffing changes resulting in the reduction of funded lactation consultants may have also contributed to this negative trend in breastfeeding. Higher discontinuation rates in multiparous patients suggest that additional barriers exist specifically for families with other children, and reflects the conflict between the time needed for pumping/breastfeeding and bedside presence in the NICU and the demands of daily activities within the larger family unit. The unique challenges experienced by non-urban families warrant consideration given the geographic distribution of Canadian NICUs in predominantly large urban centres.
      The rates of breastfeeding at hospital discharge for all neonates born preterm vary dramatically from 49% to 91%.
      • Callen J.
      • Pinelli J.
      A review of the literature examining the benefits and challenges, incidence and duration and barriers to breastfeeding in preterm infants.
      Another recent Canadian national cohort study of infants born at <33 weeks reported a 70% overall rate of receipt of breastmilk at discharge from hospital.
      • Dharel D.
      • Singhai N.
      • Wood C.
      • et al.
      Rates and determinants of mother’s own milk feeding in infants born very preterm.
      Our rate of breastmilk receipt at discharge was lower by comparison, but can be at least partially explained by our focus on the subgroup of neonates born <29 weeks. Whereas prior studies have focused on factors for non-initiation of breastfeeding, our study identified risk factors specifically for early discontinuation of breastfeeding, namely cigarette smoking, young maternal age, multiparity, earlier gestational age at birth, and higher SNAP-II scores. Smoking has been reported as one of the strongest risk factors against breastfeeding, and may be a marker for other social or health behaviours that are negatively associated with breastfeeding.
      • Cohen S.S.
      • Alexander D.D.
      • Krebs N.F.
      • et al.
      Factors associated with breastfeeding initiation and continuation: a meta-analysis.
      ,
      • Ericson J.
      • Eriksson M.
      • Hoddinott P.
      • et al.
      Breastfeeding and risk for ceasing in mothers of preterm infants – long-term follow-up.
      ,
      • Killersreiter B.
      • Grimmer I.
      • Bührer C.
      • et al.
      Early cessation of breast milk feeding in very low birthweight infants.
      • Napierala M.
      • Mazela J.
      • Merritt A.
      • et al.
      Tobacco smoking and breastfeeding: effect on the lactation process, breast milk composition and infant development. A critical review.
      • Fleurant E.
      • Schoeny M.
      • Hoban R.
      • et al.
      Barriers to human milk feeding at discharge of very-low-birth-weight infants: maternal goal setting as a key social factor.
      Reports about the influence of maternal age in breastfeeding are conflicting.
      • Wang Y.
      • Briere C.E.
      • Xu W.
      • et al.
      Factors affecting breastfeeding outcomes at six months in preterm infants.
      ,
      • Ericson J.
      • Lampa E.
      • Flacking R.
      Breastfeeding satisfaction post hospital discharge and associated factors – a longitudinal cohort study of mothers of preterm infants.
      In our cohort, multiparity—frequently associated with older maternal age—was a risk factor for both non-initiation of breastfeeding and early discontinuation. Some studies have identified a positive relationship between multiparity and continuation of breastfeeding, others negative.
      • Cohen S.S.
      • Alexander D.D.
      • Krebs N.F.
      • et al.
      Factors associated with breastfeeding initiation and continuation: a meta-analysis.
      ,
      • Wang Y.
      • Briere C.E.
      • Xu W.
      • et al.
      Factors affecting breastfeeding outcomes at six months in preterm infants.
      ,
      • Dharel D.
      • Singhai N.
      • Wood C.
      • et al.
      Rates and determinants of mother’s own milk feeding in infants born very preterm.
      By better understanding the risk factors associated with non-initiation and early discontinuation, we can identify specific barriers to breastfeeding and offer better supports to families most in need.
      Our study was unique in quantifying the specific breastmilk volumes associated with continued breastfeeding to hospital discharge. Failure to achieve these volume thresholds—specifically, 40 mL/day and approximately 100% of total daily feeds from breastmilk by day 7 and approximately 120 mL/day of breastmilk by day 14—could be used to identify those at risk of early discontinuation. As a trend towards lower breastmilk volumes can already be seen by day 3 of life in those with early breastfeeding discontinuation, our study highlights the urgent need for early interventions. These findings also dispel the myths that it takes weeks to establish an adequate milk supply, or that it is acceptable to wait until the 6-week postpartum visit to initiate prescriptions for galactagogues. Postnatal day 7 marks a critical timepoint where proportion of feeds from breastmilk begin to decline in the early discontinuation group, whereas volumes in the continuation group continue to increase.
      There is evidence that the provision of early breastfeeding education and specific instruction about breastmilk expression increases breastfeeding initiation rates.
      • Cohen S.S.
      • Alexander D.D.
      • Krebs N.F.
      • et al.
      Factors associated with breastfeeding initiation and continuation: a meta-analysis.
      ,
      • Theurich M.A.
      • McCool-Myers M.
      • Koletzko B.
      Supporting breastfeeding of small, sick and preterm neonates.
      The Baby-Friendly Hospital Initiative and Spatz 10-Step Model outline the importance of pumping within 1 hour of delivery in the setting of preterm delivery to establish and maintain a reliable milk supply.
      • Spatz D.
      Beyond BFHI: the Spatz 10-step and breastfeeding resource nurse model to improve human milk and breastfeeding outcomes.
      ,
      • Spatz D.L.
      Ten steps for promoting and protecting breastfeeding for vulnerable infants.
      ,
      • Parker L.A.
      • Sullivan S.
      • Krueger C.
      • et al.
      Effect of early breast milk expression on milk volume and timing of lactogenesis stage II among mothers of very low birth weight infants: a pilot study.
      The promotion of skin-to-skin care improves breastmilk volumes and results in higher rates of exclusive breastfeeding at 6 months corrected age among preterm infants.
      • Wang Y.
      • Zhao T.
      • Zhang Y.
      • et al.
      Positive effects of kangaroo mother care on long-term breastfeeding rates, growth and neurodevelopment in preterm infants.
      Poorer breastfeeding among more unwell infants with higher SNAP scores illustrates the need for direct support by skilled care providers in the preterm NICU population.
      • Theurich M.A.
      • McCool-Myers M.
      • Koletzko B.
      Supporting breastfeeding of small, sick and preterm neonates.
      A major strength of our study was the unique access to nutritional profiles, allowing for the identification of breastmilk target volumes as a "marker" of early breastfeeding discontinuation. Another strength was the use of an electronic dataset with rigorous prospective data collection by trained abstractors using standardized definitions.
      • Shah P.S.
      • Siedlitz W.
      • Chan P.
      • et al.
      Internal audit of the Canadian Neonatal Network Data Collection System.
      We were limited by missing socioeconomic, education, and other maternal demographic data. In addition, past maternal experience with breastfeeding was not captured by the database. Large prospective studies that account for other sociodemographic factors are still needed to identify specific barriers to breastfeeding in the NICU and to evaluate the impact of early breastfeeding interventions targeted to those experiencing a preterm birth.

      Conclusion

      Over 80% of very preterm infants can breastfeed, yet more than half will discontinue breastfeeding prior to discharge from hospital. With a heightened awareness of the short timeline to achieve target breastfeeding volumes and key perinatal predictors of breastfeeding challenges, resources can be better targeted to improve breastfeeding success and outcomes for these highest-risk newborns.

      Acknowledgements

      The authors would like to thank Prakesh Shah for supporting the data analysis for the database portion of this project.

      References

        • World Health Organization
        Health topics: breastfeeding.
        (Available at:) (Accessed on February 27, 2021)
        • Boland M.C.
        Exclusive breastfeeding should continue to six months.
        Paediatr Child Health. 2005; 10: 148
        • Callen J.
        • Pinelli J.
        A review of the literature examining the benefits and challenges, incidence and duration and barriers to breastfeeding in preterm infants.
        Adv Neonatal Care. 2005; 5: 72-88
        • Corpeleijn W.E.
        • Kouwenhoven S.M.P.
        • Paap M.C.
        • et al.
        Intake of own mother’s milk during the first days of life is associated with decreased morbidity and mortality in very low birth weight infants during the first 60 days of life.
        Neonatology. 2012; 102: 276-281
        • Vohr B.R.
        • Poindexter B.B.
        • Dusick A.M.
        • et al.
        Persistent beneficial effects of breast milk ingested in the neonatal intensive care unit on outcomes of extremely low birth weight infants at 30 months of age.
        Pediatrics. 2007; 120: e953-e959
        • Halipchuk J.
        • Temple B.
        • Dart A.
        • et al.
        Prenatal, obstetric and perinatal factors associated with the development of childhood-onset type 2 diabetes.
        Can J Diabetes. 2018; 42: 71-77
        • Ma J.
        • Qiao Y.
        • Zhao P.
        • et al.
        Breastfeeding and childhood obesity: a 12-country study.
        Matern Child Nutr. 2020; 16e12984
        • Sattari M.
        • Serwint J.
        • Levine D.M.
        Maternal implications of breastfeeding: a review for the internist.
        Am J Med. 2019; 132: 912-920
        • Spatz D.
        Beyond BFHI: the Spatz 10-step and breastfeeding resource nurse model to improve human milk and breastfeeding outcomes.
        J Perinat Neonat Nurs. 2018; 32: 167-174
        • Ikonen R.
        • Paavilainen E.
        • Kaunonen M.
        Preterm infants’ mothers’ experiences with milk expression and breastfeeding: an integrative review.
        Adv Neonatal Care. 2015; 15: 394-406
        • Green M.
        • Chalmers B.
        • Hanvey L.
        • et al.
        Chapter 6: breastfeeding. Family-centered maternity and newborn care: national guidelines.
        Public Health Agency of Canada. 2019; 6: 1-54
        • Cohen S.S.
        • Alexander D.D.
        • Krebs N.F.
        • et al.
        Factors associated with breastfeeding initiation and continuation: a meta-analysis.
        J Pediatr. 2018; 203: 190-196
        • Heaman M.I.
        • Martens P.J.
        • Brownell M.D.
        • et al.
        The association of inadequate and intensive prenatal care with maternal, fetal and infant outcomes: a population-based study in Manitoba, Canada.
        J Obstet Gynaecol Can. 2019; 41: 947-959
        • Ramji N.
        • Quinlan J.
        • Murphy P.
        • et al.
        The impact of maternal obesity on breastfeeding.
        J Obstet Gynaecol Can. 2016; 38: 703-711
        • Ericson J.
        • Eriksson M.
        • Hoddinott P.
        • et al.
        Breastfeeding and risk for ceasing in mothers of preterm infants – long-term follow-up.
        Matern Child Nutr. 2018; 14e12618
        • Wang Y.
        • Briere C.E.
        • Xu W.
        • et al.
        Factors affecting breastfeeding outcomes at six months in preterm infants.
        J Hum Lact. 2019; 35: 80-89
        • Soltani H.
        • Scott A.M.S.
        Antenatal breast expression in women with diabetes: outcomes from a retrospective cohort study.
        Int Breastfeed J. 2012; 7: 18
        • Foudil-Bey I.
        • Murphy M.S.Q.
        • Dunn S.
        • et al.
        Evaluating antenatal breastmilk expression outcomes: a scoping review.
        Int Breastfeed J. 2021; 16: 25
        • Pylypjuk C.
        • Bokhanchuk A.
        • Day C.
        • et al.
        Antenatal breastfeeding promotion amongst pregnancies at high-risk for newborn admission to the NICU: a cross-sectional study.
        Eur J Obstet Gynecol Reprod Biol. 2022; 7100160
        • Shah P.S.
        • Siedlitz W.
        • Chan P.
        • et al.
        Internal audit of the Canadian Neonatal Network Data Collection System.
        Am J Perinatol. 2017; 34: 1241-1249
        • Dammann O.
        • Shah B.
        • Naples M.
        • et al.
        Interinstitutional variation in prediction of death by SNAP-II and SNAPPE-II among extremely preterm infants.
        Pediatrics. 2009; 124: e1001-e1006
        • Dharel D.
        • Singhai N.
        • Wood C.
        • et al.
        Rates and determinants of mother’s own milk feeding in infants born very preterm.
        J Pediatr. 2021; 236: 21-27
        • Killersreiter B.
        • Grimmer I.
        • Bührer C.
        • et al.
        Early cessation of breast milk feeding in very low birthweight infants.
        Early Hum Dev. 2001; 60: 193-205
        • Napierala M.
        • Mazela J.
        • Merritt A.
        • et al.
        Tobacco smoking and breastfeeding: effect on the lactation process, breast milk composition and infant development. A critical review.
        Environ Res. 2016; 151: 321-338
        • Fleurant E.
        • Schoeny M.
        • Hoban R.
        • et al.
        Barriers to human milk feeding at discharge of very-low-birth-weight infants: maternal goal setting as a key social factor.
        Breastfeed Med. 2017; 12: 20-27
        • Ericson J.
        • Lampa E.
        • Flacking R.
        Breastfeeding satisfaction post hospital discharge and associated factors – a longitudinal cohort study of mothers of preterm infants.
        Int Breastfeed J. 2021; 16: 1-9
        • Theurich M.A.
        • McCool-Myers M.
        • Koletzko B.
        Supporting breastfeeding of small, sick and preterm neonates.
        Sem Perinatol. 2021; 45: 1-6
        • Spatz D.L.
        Ten steps for promoting and protecting breastfeeding for vulnerable infants.
        J Perinat Neonat Nurs. 2004; 18: 385-396
        • Parker L.A.
        • Sullivan S.
        • Krueger C.
        • et al.
        Effect of early breast milk expression on milk volume and timing of lactogenesis stage II among mothers of very low birth weight infants: a pilot study.
        J Perinatol. 2012; 32: 205-209
        • Wang Y.
        • Zhao T.
        • Zhang Y.
        • et al.
        Positive effects of kangaroo mother care on long-term breastfeeding rates, growth and neurodevelopment in preterm infants.
        Breastfeed Med. 2021; 16: 282-291