Factors Associated with Mother’s Own Milk Feeding and Direct Breastfeeding at Discharge in Preterm Infants with Feeding Difficulties: Clinical and Research Implications

Abstract

Background:

Prematurity presents numerous barriers to mother’s own milk (MOM) feeding and direct breastfeeding (DBF).

Aim:

This study aimed to determine factors associated with MOM feeding and DBF at discharge from the neonatal intensive care unit (NICU) in preterm-born infants presenting with feeding difficulties.

Methods:

A retrospective study of data from 237 preterm-born infants referred for evaluation of feeding difficulties and discharged home on full oral feeds was examined. Maternal and infant characteristics and oral feeding milestones were examined for their association with MOM intake and DBF at discharge using bivariate and multivariate regression analyses.

Results:

MOM feeding at discharge occurred in 35.4% (n = 84) infants. The odds of any MOM feeding at discharge were higher with higher maternal age, absence of maternal substance use, and fewer days between full per oral (PO) and discharge (all, p < 0.05). Among the 84 MOM-fed infants, 4.76% (n = 4) were exclusively breastfed, whereas 39.3% (n = 33) were partially DBF at discharge. The DBF infants had higher birthweight, no incidence of being small for gestational age, lower incidence of respiratory support at birth and intraventricular hemorrhage, lower postmenstrual age (PMA) at full PO, shorter duration from first PO to full PO, and lower PMA at discharge (all, p < 0.05).

Conclusion:

We found reduced use of MOM and DBF among preterm-born infants with feeding difficulties at NICU discharge. Clinical management and research advocacy must focus on targeted interventions in this setting by recognizing significant modifiable factors applicable to prepregnancy, pregnancy, NICU, and postdischarge care.

Introduction

Mother’s own milk (MOM) is widely acknowledged as the most preferred form of nutrition for all infants, including preterm infants, owing to its nutritional, immunological, and health-promoting qualities.^1,2 In addition to containing macronutrients and micronutrients, MOM uniquely contains non-nutritive bioactive elements, including immunoglobulins, cytokines, chemokines, antimicrobial agents, hormones, mucins, oligosaccharides, glycans, cells, and growth factors.^3,4 These bioactive elements defend against infection and inflammation while also contributing to immunological maturation, regeneration and repair, organ development, and healthy microbial colonization.³

Thus, MOM feeding is superior to infant formula feeding and is linked to numerous short- and long-term advantages for the preterm infant. These include a reduced risk of prematurity-related morbidities, such as necrotizing enterocolitis (NEC), retinopathy of prematurity, bronchopulmonary dysplasia, intraventricular hemorrhage (IVH), and severe infections.^5,6 By reducing the risk of these prematurity-related morbidities, MOM feeding is associated with a decreased length of hospitalization and, most importantly, mortality.^5,6 In addition, MOM feeding is associated with improved neurodevelopmental outcomes.⁷

In preterm infants, particularly those born before 32 weeks gestational age (GA), MOM feedings typically begin with the use of either nasogastric or orogastric tubes.⁸ This is because these fragile, medically compromised infants often cannot initially feed orally due to poor coordination of sucking, swallowing, and breathing.⁹ As a result, mothers must continuously rely on expression with their hands or pumps to establish and maintain milk supply for several weeks, or even months, until their infants can safely eat orally.¹⁰

Once the infants become physiologically and developmentally ready for oral feeding, they may receive MOM directly from the breast or through a bottle.¹¹ Bottle feeding is assumed to be easier and leads to a speedier transition to complete oral feeding and earlier discharge than direct breastfeeding (DBF).^12,13 However, one study reported no differences in oral feeding characteristics between infants who received ≥50% direct breastfeeds versus <50% direct breastfeeds during hospitalization.¹⁴ Also, the length of hospitalization was not different in both groups.¹⁴

DBF has additional benefits over bottle feeding of expressed MOM. It is associated with improved neurodevelopmental outcomes,¹⁵ reduced risk of dental malocclusion,¹⁶ containment, pain management,¹⁷ and improved maternal and infant bonding, as well as providing sensory, verbal, and nonverbal (emotional) communication,^18,19 and longer durations of MOM feeding.¹¹ Compared to DBF, expressing, and feeding MOM requires considerable effort and is resource-intensive.²⁰ Furthermore, the expression of MOM may result in contamination, and the freezing, thawing, heating, and storage of expressed MOM degrade its nutritional and immunological properties.^21,22

In the preterm infant population admitted to the neonatal intensive care unit (NICU), neither the rate of MOM feeding nor DBF is optimal due to numerous challenges. Delayed attainment of oral feeding and continuous reliance on MOM expression increase the risk of stopping MOM feeds before discharge.¹⁰ Fortification of MOM is often a necessary practice in preterm infants of <32 weeks gestation; however, this practice may interfere with DBF as the milk will need to be expressed before being mixed with fortifiers. The physical separation of the mother from the infant during hospitalization and the fragility of the preterm infant limit the opportunity for frequent skin-to-skin contact, non-nutritive breastfeeding, and bonding which are all essential elements required for the establishment of successful DBF.²³

The presence of maternal morbidities such as gestational diabetes mellitus, mental disorders, substance abuse, hypertensive disorders, and stress arising from pregnancy and preterm delivery can further make MOM feeding and DBF initiation and progression difficult.^24–26 In addition, inadequate support in the NICU, preconceived notions on the DBF-related competence of preterm infants, and a lack of clear guidelines on when to establish DBF are impediments to DBF initiation and continuation.^13,14

Understanding factors during prepregnancy, pregnancy, and neonatal period influencing MOM feeding and DBF is important so that MOM feeding and DBF rates improve at NICU discharge. Therefore, this study aimed to determine infant, maternal, and oral feeding milestone characteristics associated with MOM feeding and DBF at discharge from the NICU in preterm infants.

Methods

Study design and setting

This study incorporates a retrospective design examining the data from preterm infants admitted to the all-referral level IV NICU at Nationwide Children’s Hospital, Columbus, Ohio, who were referred to our innovative neonatal and infant feeding disorders program for the evaluation and management of severe feeding difficulties. The clinical characteristics of the feeding difficulties included an inability to consume adequate and prescribed oral feeds and not meeting oral feeding milestones, gavage tube dependence, gastroesophageal reflux disease, feeding or postprandial related-cardiorespiratory spells, frequent emesis, coughing, choking, gagging, arching, irritability, stridor during feeding, refusal to feed, and/or poor sucking ability. Institutional Review Board approval was obtained, and Health Insurance Portability and Accountability Act guidelines were followed.

Study sample

The study sample consisted of 237 preterm-born infants. Infants were included if they were born between January 1, 2019, and December 31, 2023, and born <37 weeks gestation. Infants were excluded if they were not on full oral feeding at discharge and had incomplete birth and discharge history. During the selected study period, 614 infants were referred to our innovative feeding disorders program, of these, 505 infants were born preterm. At discharge, 307 infants were on full oral feeding. Seventy infants were excluded as there were inadequate complete data characteristics for the purpose of this investigation, leaving a final sample of 237 infants.

Measurement and data collection

The primary outcomes were any MOM feeding at discharge and any DBF at discharge in any MOM subgroup. Any MOM feeding at discharge was defined as when an infant received MOM irrespective of the volume/ratio. Any DBF at discharge was defined as when a MOM-fed infant suckled directly at the breast for nutritive purposes, irrespective of the duration or the number of daily sessions. Several infant, maternal, and oral feeding milestone characteristics were examined for their relationship with any MOM feeding and any DBF at discharge.

The infant characteristics included sex, race, GA at birth, birth weight, birth weight category (small for gestational age [SGA], appropriate for gestational age [AGA], and large for gestational age [LGA]) using the Fenton growth charts, APGAR scores at 5 minutes, respiratory support at birth, 36 weeks and at discharge, length of hospital stay, postmenstrual age (PMA) at discharge, and infant morbidities, including patent ductus arteriosis (PDA), IVH (grades I–IV), hypoxic-ischemic encephalopathy (HIE), NEC (stages I–III) and sepsis. The prepregnancy and pregnancy maternal characteristics included age in completed years, mental disorder, diabetes mellitus, preeclampsia, substance use, hypertension disorder of pregnancy, delivery type (vaginal versus cesarean section), and delivery complication.

Infant oral feeding milestone characteristics were defined as follows: “first per oral (PO)” as the first day of oral feeding irrespective of the feeding volume, and “full PO” as the day when the infant was able to take all feedings by PO without a feeding tube for 2 consecutive days. The following feeding-related outcomes were calculated: PMA at first and full PO, days from first PO to full PO, days from first PO to discharge, and days from full PO to discharge. These feeding-related outcomes were selected because delayed initiation and progression of oral feeding and continuous reliance on MOM expression may increase the risk of discontinuing MOM feeds before discharge.¹⁰ Also, delayed attainment of oral feeding also means delayed initiation of direct breastfeeds. Given the previously reported perceived difficulty of DBF in preterm infants,^13,27 we hypothesized that DBF at discharge would be more likely in infants who had achieved oral feeding competencies earlier.

Statistical analysis

To compare demographic and clinical characteristics (a) between any MOM-fed and formula-fed infants at discharge and (b) between any DBF and non-DBF infants at discharge, the chi-square test or Fisher exact test was used as appropriate for categorical variables, whereas linear mixed models and generalized estimating equations methods were used for continuous variables.

To identify the optimal factors associated with the likelihood of discharge (a) with any MOM versus infant formula and (b) with any DBF versus non-DBF from a pool of demographic and clinical variables, two stepwise logistic regression analyses were performed with the bidirectional elimination method (combined forward and backward). The bidirectional stepwise is an iterative procedure, where testing variables to be included or excluded happen in each step based on a prechosen significant level. To allow a variable into the model, a significance level of 0.3 is required, and a significance level of 0.15 is required for a variable to stay in the model. We chose a higher significant level to ensure that important variables were not missed. Hosmer and Lemeshow test was used to assess the goodness of fit of the optimal models.

Data are presented as LSmean (least square mean) ± SEM (standard error of the mean), %, n (%) or odds ratios, and 95% confidence intervals (CIs). Statistical analysis was carried out using IBM SPSS Statistics, version 28, SAS9.4. All p-values were two-sided and were considered significant at p < 0.05.

Results

Infant and maternal characteristics of the total sample stratified by MOM feeding at discharge are displayed in Table 1. The mean GA of infants in this study was 29.89 ± 0.22 weeks. The median (IQR) Apgar score at 5 minutes was 8 (6–8). A majority (95.4%) of the infants required respiratory support at birth. The average length of hospital stay was 55.55 ± 2.56 days and the average PMA at discharge was 42.71 ± 0.23 weeks. The average maternal age was 28.79 ± 0.38 years and most mothers (75.1%) delivered via cesarean sections. Of the 237 infants in the sample, 84 (35.4%) received some MOM at discharge, and 153 (64.6%) received only infant formula. In the bivariate analysis, mothers of infants who received MOM were significantly older (p = 0.008). The incidence of maternal substance use was significantly lower (p < 0.0010) in mothers whose infants received MOM at discharge. Importantly, no differences were found in any of the examined infant characteristics among MOM-fed and formula-fed infant groups.

Table 1.

Maternal and Infant Characteristics of the Total Sample

	Total (n = 237)	MOM (n = 84)	FM only (n = 153)	p
Infant characteristics
Sex (male:female)	130:107 (54.9:45.1)	47:37 (56:44)	83:70 (54.2:45.8)	0.890
Race (White:Black:Other)	145:56:36 (61.2:23.6:15.2)	54:16:14 (64.3:19:16.7)	91:40:22 (59.5:26.1:14.4)	0.461
GA, weeks	29.89 ± 0.22	30.04 ± 0.35	29.81 ± 0.28	0.602
Birth weight	1457.56 ± 46.51	1455.13 ± 68.94	1458.90 ± 6125	0.967
SGA:AGA:LGA	28:184:25 (11.8:77.6:10.5)	12:60:12 (14.3:71.4:14.3)	16:124:13 (10.5:81:8.5)	0.219
Respiratory support at birth	226 (95.4)	80 (95.2)	146 (95.4)	1.000
Respiratory support at 36 weeks	105 (44.3)	41 (48.8)	64 (41.8)	0.339
Respiratory support at discharge	81 (34.2)	27 (32.1)	54 (35.3)	0.669
Length of hospital stay	55.55 ± 2.56	53.24 ± 4.65	56.82 ± 3.02	0.516
PMA at discharge	42.71 ± 0.23	42.37 ± 0.38	42.90 ± 0.28	0.253
Discharge weight	3676.40 ± 56.12	3585.65 ± 95.30	3726.23 ± 68.86	0.23
Weight gain velocity (g/kg/day)	9.62 ± 0.37	9.55 ± 0.32	9.66 ± 0.17	0.760
Infant morbidity
PDA	57 (24.7)	17 (20.2)	40 (26.1)	0.343
IVH	49 (20.7)	14 (16.7)	35 (22.9)	0.315
HIE	4 (1.7)	3 (3.6)	1 (0.7)	0.129
NEC	26 (11)	9 (10.7)	17 (11.1)	1.000
Sepsis	44 (18.6)	14 (16.7)	30 (19.6)	0.606
Maternal characteristics
Age, years	28.79 ± 0.38	30.07 ± 0.55	28.09 ± 0.50	0.008
Mental disorder	24 (10.1)	5 (6)	19 (12.4)	0.175
Diabetes	49 (20.7)	22 (26.2)	27 (17.6)	0.133
Preeclampsia	63 (26.6)	23 (27.4)	40 (26.1)	0.878
Substance use	51 (21.5)	3 (3.6)	48 (31.4)	<0.001
Hypertension	44 (18.6)	17 (20.2)	27 (17.6)	0.727
Delivery (vaginal:cesarean)	59:178 (24.9:75.1)	16:68 (19:81)	43:110 (28.1:71.9)	0.082
Delivery complications	205 (86.5)	71 (84.5)	134 (87.6)	0.553

AGA, appropriate for gestational age; FM, formula; GA, gestational age; HIE, hypoxic-ischemic encephalopathy; IVH, intraventricular hemorrhage; LGA, large for gestational age; MOM, mother’s own milk; NEC, necrotizing enterocolitis; PDA, patent ductus arteriosis; PMA, postmenstrual age; SGA, small for gestational age.

Of the 84 infants who received some MOM at discharge, only 4 (4.76%) were on an exclusive MOM diet, whereas 33 (39.2%) were directly breastfeeding and supplementing with bottles. Infant and maternal characteristics of this subsample of infants stratified by DBF at discharge are displayed in Table 2. Directly breastfed infants had significantly higher (p = 0.030) birth weights, and none was categorized as SGA at birth. Interestingly, the proportion of infants who required respiratory support at birth was significantly lower (p = 0.020) for infants who were later directly breastfed at discharge. Importantly, directly breastfed infants (which is a later functional activity) had a significantly lower (p = 0.040) incidence of IVH (recognized in the first week after birth) and were discharged at a significantly lower PMA (p = 0.010).

Table 2.

Maternal and Infant Characteristics of the Subsample of MOM-Fed Infants at Discharge

	Total (n = 84)	DBF (n = 33)	NDBF (n = 51)	p
Infant characteristics
Sex (male:female)	47:37 (56:44)	18:15 (54.5:45.5)	29:22 (56.9:43.1)	1.000
Race (White:Black:Other)	54:16:14 (64.3:19:16.7)	25:5:3 (75.8:15.2:9.1)	29:11:10 (56.9:21.6:19.6)	0.230
GA, weeks	30.04 ± 0.35	30.39 ± 0.53	29.81 ± 0.46	0.410
Birth weight	1455.13 ± 68.94	1646.39 ± 121.06	1331.37 ± 76.42	0.030
SGA:AGA:LGA	12:60:12 (14.3:71.4:14.3)	0:25:8 (0:75.8:24.2)	12:35:4 (23.5:68.6:7.8)	0.003
Respiratory support at birth	80 (95.2)	29 (87.9)	51 (100)	0.020
Respiratory support at 36 weeks	41 (48.8)	18 (54.5)	23 (45.1)	0.503
Respiratory support at discharge	27 (32.1)	8 (24.1)	19 (37.3)	0.240
Length of hospital stay	53.24 ± 4.65	45.58 ± 5.92	58.20 ± 6.49	0.150
PMA at discharge	42.37 ± 0.38	41.26 ± 0.53	43.08 ± 0.49	0.010
Discharge weight	3585.65 ± 95.30	3462.12	3665.59	0.30
Weight gain velocity (g/kg/day)	9.55 ± 0.32	8.83 ± 0.71	10.01 ± 0.24	0.120
Infant morbidity
PDA	17 (20.2)	6 (18.2)	11 (21.6)	0.790
IVH	14 (16.7)	2 (6.1)	12 (23.5)	0.040
HIE	3 (3.6)	2 (2.4)	1 (1.2)	0.560
NEC	9 (10.7)	1 (3)	8 (15.69)	0.080
Sepsis	14 (16.7)	4 (4.8)	10 (11.9)	0.550
Maternal characteristics
Age, years	30.07 ± 0.55	30.42 ± 0.85	29.84 ± 0.72	0.600
Mental disorder	5 (6)	1 (3)	4 (7.8)	0.644
Diabetes	22 (26.2)	7 (21.2)	15 (29.4)	0.456
Preeclampsia	23 (27.4)	7 (21.2)	16 (31.4)	0.331
Substance use	3 (3.6)	1 (3)	2 (3.9)	1.000
Hypertension	17 (20.2)	7 (21.2)	10 (19.6)	1.000
Delivery (vaginal:cesarean)	16:68 (19:81)	8:25 (24.2:75.8)	8:43 (15.7:84.3)	0.398
Delivery complications	71 (84.5)	27 (81.8)	42 (86.3)	0.760

AGA, appropriate for gestational age; DBF, direct breastfeeding; GA, gestational age; HIE, hypoxic-ischemic encephalopathy; IVH, intraventricular hemorrhage; LGA, large for gestational age; MOM, mother’s own milk; NDBF, no direct breastfeeding; NEC, necrotizing enterocolitis; PDA, patent ductus arteriosis; PMA, postmenstrual age; SGA, small for gestational age.

Oral feeding milestone characteristics are presented in Figure 1. In the total sample, the number of days from full PO to discharge was significantly shorter (p = 0.033) for the MOM-fed infants. In the subgroup of MOM-fed infants, partially directly breastfed infants achieved full oral feeding at a significantly lower (p = 0.010) PMA and transitioned to full PO in a significantly shorter time (p = 0.010).

FIG. 1.

Oral feeding milestone characteristics. The oral feeding milestones are compared between MOM-fed infants and exclusively formula-fed infants (left column). Days from full PO to discharge are significantly shorter in the MOM group. The oral feeding milestones are compared between directly breastfeeding and nondirectly breastfeeding human milk-fed infants (right column). Directly breastfeeding infants have lower PMA at full PO and fewer days from the first PO to full PO. MOM, mother’s own milk; PMA, postmenstrual age; PO, per oral.

In performing multivariable logistic regression while assessing the factors associated with any MOM feeding at discharge, the odds of any MOM feeding at discharge were significantly higher with increasing maternal age (odds ratio = 1.06; 95% CI, 1.008–1.115; p = 0.0219). The odds of any MOM feeding at discharge were also significantly higher in mothers who did not abuse substances (odds ratio = 12.802; 95% CI, 3.805–43.072; p < 0.001). The odds of any MOM feeding at discharge were significantly decreased with increasing days from the attainment of full PO to discharge (odds ratio = 0.981; 95% CI, 0.964–0.998; p = 0.031) (Table 3).

Table 3.

Factors Associated with MOM Feeding at NICU Discharge

	OR (95% CI)	p
Maternal age, years	1.06 (1.008–1.115)	0.022
Maternal substance use	12.802 (3.805–43.072)	<0.0001
Full PO to discharge, days	0.981 (0.964–0.998)	0.031

CI, confidence interval; MOM, mother’s own milk; NICU, neonatal intensive care unit; OR, odds ratio; PO, per oral.

In performing multivariable logistic regression while assessing the factors associated with DBF at discharge, the odds of any DBF at discharge were significantly higher in infants without an IVH diagnosis (odds ratio = 8.311; 95% CI, 0.99–69.799; p = 0.0219). The odds of any DBF at discharge were significantly decreased with increasing PMA at full PO (odds ratio = 0.877; 95% CI, 0.772–0.996; p = 0.0439) (Table 4).

Table 4.

Factors Associated with Direct Breast Feeding at NICU Discharge

	OR (95% CI)	p
IVH	8.311 (0.99–69.799)	0.047
HIE	0.094 (0.005–1.973)	0.128
PMA at full PO	0.877 (0.772–0.996)	0.044

CI, confidence interval; IVH, intraventricular hemorrhage; HIE, hypoxic-ischemic encephalopathy; NICU, neonatal intensive care unit; OR, odds ratio; PMA, postmenstrual age; PO, per oral.

Discussion

Successful intake of MOM and/or providing DBF to preterm-born infants before and after discharge from the NICU remains a major challenge in many clinical settings worldwide, be it in industrialized or resource-limited nations. While it is abundantly well recognized that the benefits of these approaches are invaluable, their prevalence is alarmingly low. Earlier studies assessed factors associated with MOM and/or DBFs, and this is the first time the current innovative study is examining the role of MOM and DBFs in preterm infants with oral feeding difficulties. Particularly among the preterm infants with feeding difficulties, interestingly, MOM and/or DBF in preference to formula feeding achieved superior results. The antecedent factors are dependent not only on the mother–infant dyad but also on the social determinants and healthcare systems. Healthy preterm-born infants and healthy maternal–infant dyads have undoubtedly higher success rates of MOM intake and/or DBF. However, the nutritional, immunological, neurotrophic, and emotional components of such approaches are much more needed for preterm-born infants with multiple morbidities to improve long-term outcomes. Hence, this study was undertaken to unravel the factors associated with MOM feeding and DBF at discharge in prematurely born infants referred for the evaluation of feeding difficulties who were requiring chronic tube feeding. We anticipate that this work will have immediate clinical and research implications, as well as unravel opportunities for focused educational and quality improvement strategies globally.

In this retrospective study of 237 preterm infants, the rate of any MOM feeding at discharge was 35.4% and the rate of any DBF in the MOM subsample was 39.2%. According to national data collected from 802 U.S. hospitals participating in the Vermont Oxford Network, the provision of any human milk at discharge was 52% in 2017.²⁸ Considerable variations in rates of DBF at NICU discharge have been reported, with rates ranging from 51% to 64%.^29,30 These rates, an important perinatal health statistic, have been used as a measurable target for assessing the efficiency of health care systems; however, these values depend on whether it is in-born or delivery hospitals, or tertiary care settings. Thus, such MOM and DBF metrics cannot compare ‘the apples to apples’ in all hospital NICU systems in an unambiguous way, because preterm infants encompass a select group with highly variable prenatal, perinatal, and NICU characteristics, as well as variability with feeding milestones.

These low rates reflect how challenging it is to provide MOM and directly breastfeed hospitalized preterm infants with feeding difficulties and the need for more interventions and implementation during NICU settings and after discharge. While MOM feeding irrespective of the feeding mode is important and often the focus of most interventions, supporting DBF cannot be overlooked. DBF leads to a greater likelihood of continued MOM feeding upon discharge and many months after discharge.^11,31 Thus, DBF should be emphasized as critical for MOM-feeding success. The amount of work needed to ensure the success of these methods is undoubtedly greater and involves several personnel supporting the mother–infant dyad, thus requiring a greater allocation of resources.

In this current study, we identified a lack of differences in infant characteristics between infants who received MOM at discharge and those who did not. Other studies have, however, identified GA, birthweight, incidence of morbidities, and length of hospitalization to be associated with MOM feeding at discharge.^32,33 This discrepancy may be related to our distinct patient population, which included infants with significant feeding challenges during their NICU course. Also, almost all the MOM-fed infants were receiving infant formula at varying doses, which may have made the distinction between the two groups less straightforward and the dose–response effect of MOM feeding on clinical outcomes less visible.

In the subsample of MOM-fed infants, DBF infants had higher birthweight, no incidence of SGA, lower incidence of respiratory support at birth, lower incidence of IVH diagnosis, and a younger PMA at NICU discharge. An IVH diagnosis and PMA at discharge were the only infant characteristics significantly associated with DBF at discharge. Interestingly, although directly breastfeeding is a later functional mother–infant dyad activity, the proportion of DBF infants who required earlier respiratory support at birth was lower, had lower incidence of IVH recognized in the first week after birth, and were discharged at lower PMA. Thus, it appears plausible that there is a protective effect of the choice of DBF later in the course (and use of MOM) even among the preterm infants presenting with feeding difficulties.

Although there were no statistically significant differences in the length of hospital stay between DBF and non-DBF infants, the former were discharged on average 12 days shorter than the latter. An earlier discharge from the NICU undoubtedly costs less in resources and health care expenditures regardless of statistical significance, improves parent/family satisfaction, and enhances mother–infant bonding, all with added benefits.

The reluctance to offer DBF to sicker and more high-risk infants may be related to the reported perception that DBF is more difficult than bottle feeding, and it is thought to delay the transition to full oral feeding and thus delay hospital discharge.^13,14 Despite current evidence demonstrating that the method of oral feeds at hospital discharge does not influence the length of hospitalization, it is likely that health professionals still perceive direct breastfeeds to affect the length of hospitalization.¹⁴

A lower birth weight and a higher incidence of SGA may result in favoring the use of supplemental feeds to improve weight, thus interfering with MOM provision and DBF. The need to know the exact volumes of milk consumed and to carefully quantify intake in growth-restricted and sick preterm infants may also interfere with DBF. The hospital atmosphere has been regarded as a stressful one with little privacy, and extended hospital stays may lead mothers to forgo DBF, and this may also explain why the nondirectly breastfed infants had higher PMA at discharge and longer permanence in the NICU.

Findings from this study show that older mothers were more likely to provide MOM at discharge. Other studies have reported similar findings.^33,34 Reasons for this may be that older mothers are more likely to have more breastfeeding experience, better health behaviors, and self-efficacy and are more aware of the benefits of breastfeeding. However, a few studies found younger age of mothers to be associated with MOM at discharge,^35,36 whereas some other studies found no differences at all.^32,37 Older maternal age has also been found to be associated with DBF^29,38; however, we found no such association in this current study.

Mothers who abused substances were less likely to provide MOM at discharge in this current study. This is consistent with other studies that have reported that women who abuse substances are at an increased risk of not initiating and maintaining MOM feeding.³⁹ The reasons for this may include a high co-occurrence of medical and psychiatric conditions that can impact lactation.^40,41 Also, infants who are exposed to substances may not gain weight properly, leading to supplementation with infant formula. These infants may also have trouble with latching and may be hospitalized for longer periods.⁴²

All infants in this study achieved full PO at discharge. Among the MOM-fed infants, the time lapse between the attainment of full PO and discharge was shorter; however, no other differences in oral feeding milestones were found. Lapse in hospital discharge once oral feeds are achieved may be related to other medical discharge criteria the infants have not achieved, such as physiological stability and temperature regulation. Discharge from the hospital is also dependent on not only infant characteristics but also provider and caregiver characteristics.⁹ It is possible that caregivers of the MOM-fed group had better discharge readiness due to more frequent interactions and greater opportunities for education with the healthcare team when they come in to pump and breastfeed. As noted earlier, substance use was significantly higher in the infant formula group, and this may likely complicate discharge as the infant’s environment after discharge may be compromised.

In this current study, the directly breastfed infants transitioned from first PO to full PO in a shorter time frame than nondirectly breastfed infants. The directly breastfed infants also achieved full PO at a lower PMA. This may be because DBF facilitates more skin-to-skin contact, which is known to accelerate oral feeding skills.⁴³ Skin-to-skin contact may facilitate early oral feeding milestones through its effect on physiological stability, which is important for better coordination of sucking, swallowing, and breathing. DBF also contributes to the development of stomatognathic structures and functions and helps facilitate oral skills maturation.⁴⁴ Importantly, advanced neurophysiological characteristics manifested as oral sensory-motor processing and regulation, superior adaptational skills while off-tube feeding, sustained growth rates, and the ability to achieve all these skills at a lower PMA support accelerated maturation when there is a commitment to provide MOM and/or DBF.

Although the long-term benefits of MOM and DBF are known, our very early findings are novel observations gleaned during the NICU period, which will likely provide more motivation to the propagation of MOM and/or DBF during NICU stay. Understanding factors that influence MOM feeding and DBF has important clinical implications. The NICU can be strengthened to support the use of MOM and DBF by targeting those mothers and infants at lower odds. Implementing feeding policies such as first oral feeding being MOM and directly at the breast, delaying bottle feeding, increasing visitation hours for parents, involving parents in feeding their infants, engaging more lactation consultants, rooming in, and modeling the baby-friendly hospital initiative are all important interventions.^29,45

Educational and support programs targeted at parents and NICU staff also enhance the use of MOM and DBF and may help change perceptions and attitudes toward these practices. These programs may include breastfeeding education and counseling, peer support programs, provision of breast pumping equipment, and the training of NICU staff.⁴⁵ Breastfeeding education interventions before conception and during the early periods of pregnancy are particularly beneficial to help mothers make informed decisions about feeding their infants. This way, mothers would not be burdened to make unplanned feeding decisions in the case of preterm birth, which often happens unexpectedly.

Kangaroo mother care (KMC) characterized by early and continuous skin-to-skin contact between mother and infant is strongly associated with exclusive breastfeeding at discharge and should be encouraged where possible.⁴⁵ Non-nutritive sucking and oral stimulation may benefit preterm infants who are unable to nurse directly and should be encouraged until breastfeeding is established.⁴⁶ Until the infant is capable of oral feeding, emphasis on early initiation and continuous MOM expression should be made to ensure that maternal milk production does not wane off.

Limitations

The retrospective nature of the study design limited data collection to information available in the medical records. The results from the study are from an all-referral NICU setting, thus limiting the generalization of the findings. Potential variables that may affect MOM feeding and DBF, such as the use of donor human milk, the start of MOM expression, type of feeding at first oral feeding, maternal education, maternal birth order, and other sociodemographic characteristics, were not available.

Conclusions and Future Directions

The current study found limited use of MOM and DBF among preterm infants referred for feeding difficulties at NICU discharge. The odds of MOM feeding at discharge were higher with higher maternal age, absence of maternal substance use, and fewer days between full PO and discharge. Infants who were directly breastfeeding at discharge had a higher birthweight, no incidence of being small for GA, lower incidence of respiratory support at birth and IVH, lower PMA at full PO, shorter duration from first PO to full PO, and lower PMA at discharge. Targeted breastfeeding interventions based on infant and maternal characteristics are paramount in this setting and should be considered a priority, as there are potential clinical, economic, and neurodevelopmental benefits in vulnerable infants. Effective interventions include supportive policies within NICUs, educational and breastfeeding support programs, and KMC. Such interventions should be targeted toward populations at the highest risk of not utilizing MOM, such as younger mothers and mothers who abuse alcohol. Health professionals tending to very sick preterm infants should also be trained to encourage more MOM use and DBF in these infants. Further studies should consider studying the safety of DBF as opposed to bottle feeding in infants with different morbidities.

Footnotes

Acknowledgments

We wish to acknowledge Kailyn Marcano, RN, and Zakia Sultana, BA, for their help with data validation.

Authorship Contribution

F.B.: Writing—original draft, writing—reviewing and editing, and formal data analyses and interpretation. E.A.: Formal statistical analyses and validation. S.R.J.: Conceptualization, writing—data analysis and interpretation, reviewing and editing, resources, and supervision.

Disclosure Statement

The authors have no competing interests to declare that are relevant to the content of this article.

Funding Information

This study is supported in part by The Nationwide Foundation Endowed Chair in Neonatal Research award (to S.R.J.).

References

Meek

, Noble

, Section on B. Policy statement: Breastfeeding and the use of human milk. Pediatrics, 2022; 150(1):e2022057988; doi: 10.1542/peds.2022-057988

Parker

, Stellwagen

, Noble

, et al. SECTION ON BREASTFEEDING, COMMITTEE ON NUTRITION, COMMITTEE ON FETUS AND NEWBORN. Promoting human milk and breastfeeding for the very low birth weight infant. Pediatrics, 2021; 148(5); doi: 10.1542/peds.2021-054272

Ballard

, Morrow

. Human milk composition: Nutrients and bioactive factors. Pediatr Clin North Am, 2013; 60(1):49–74; doi: 10.1016/j.pcl.2012.10.002

Andreas

, Kampmann

, Mehring Le-Doare

. Human breast milk: A review on its composition and bioactivity. Early Hum Dev, 2015; 91(11):629–635; doi: 10.1016/j.earlhumdev.2015.08.013

Miller

, Tonkin

, Damarell

, et al. A systematic review and meta-analysis of human milk feeding and morbidity in very low birth weight infants. Nutrients, 2018; 10(6); doi: 10.3390/nu10060707

Pados

. State of the science on the benefits of human milk for hospitalized, vulnerable neonates. Nurs Womens Health, 2023; 27(2):121–140; doi: 10.1016/j.nwh.2023.01.008

Lechner

, Vohr

. Neurodevelopmental outcomes of preterm infants fed human milk: A systematic review. Clin Perinatol //2017; 44(1):69–83; doi: 10.1016/j.clp.2016.11.004

Jadcherla

, Dail

, Malkar

, et al. Impact of process optimization and quality improvement measures on neonatal feeding outcomes at an All-Referral neonatal intensive care unit. JPEN J Parenter Enteral Nutr, 2016; 40(5):646–655; doi: 10.1177/0148607115571667

Hasenstab

, Jadcherla

. Evidence-based approaches to successful oral feeding in infants with feeding difficulties. Clin Perinatol, 2022; 49(2):503–520; doi: 10.1016/j.clp.2022.02.004

10.

Pineda

. Predictors of breastfeeding and breastmilk feeding among very low birth weight infants. Breastfeed Med, 2011; 6(1):15–19; doi: 10.1089/bfm.2010.0010

11.

Briere

, McGrath

, Cong

, et al. Direct-breastfeeding in the neonatal intensive care unit and breastfeeding duration for premature infants. Appl Nurs Res, 2016; 32:47–51; doi: 10.1016/j.apnr.2016.04.004

12.

Briere

. Breastfed or bottle-fed: Who goes home sooner? Adv Neonatal Care, 2015; 15(1):65–69; doi: 10.1097/anc.0000000000000159

13.

Niela-Vilén

, Axelin

, Melender

, et al. Aiming to be a breastfeeding mother in a neonatal intensive care unit and at home: A thematic analysis of peer-support group discussion in social media. Matern Child Nutr, 2015; 11(4):712–726; doi: 10.1111/mcn.12108

14.

Wener

, Dow

, Fucile

. Evaluation of methods of breast or bottle feeding on length of hospitalization of preterm infants. Breastfeed Med, 2021; 16(11):899–903; doi: 10.1089/bfm.2021.0066

15.

Keim

, Sullivan

, Sheppard

, et al. Feeding infants at the breast or feeding expressed human milk: Long-Term cognitive, executive function, and eating behavior outcomes at age 6 years. J Pediatr, 2021; 233:66–73.e1; doi: 10.1016/j.jpeds.2021.02.025

16.

Abate

, Cavagnetto

, Fama

, et al. Relationship between breastfeeding and malocclusion: A systematic review of the literature. Nutrients, 2020; 12(12); doi: 10.3390/nu12123688

17.

Koukou

, Theodoridou

, Taousani

, et al. Effectiveness of Non-Pharmacological Methods, Such as Breastfeeding, to Mitigate Pain in NICU Infants. Children (Basel), 2022; 9(10); doi: 10.3390/children9101568

18.

Smith

, Forrester

. Maternal time use and nurturing: Analysis of the association between breastfeeding practice and time spent interacting with baby. Breastfeed Med, 2017; 12:269–278; doi: 10.1089/bfm.2016.0118

19.

Modak

, Ronghe

, Gomase

. The psychological benefits of breastfeeding: Fostering maternal well-being and child development. Cureus, 2023; 15(10):e46730; doi: 10.7759/cureus.46730

20.

Fan

HSL

, Fong

DYT

, Lok

KYW

, et al. Experiences of expressed human milk feeding: A qualitative study. Sex Reprod Healthc, 2023; 38:100903; doi: 10.1016/j.srhc.2023.100903

21.

Paulaviciene

, Liubsys

, Eidukaite

, et al. The effect of prolonged freezing and holder pasteurization on the macronutrient and bioactive protein compositions of human milk. Breastfeed Med, 2020; 15(9):583–588; doi: 10.1089/bfm.2020.0219

22.

Boo

, Nordiah

, Alfizah

, et al. Contamination of breast milk obtained by manual expression and breast pumps in mothers of very low birthweight infants. J Hosp Infect, 2001; 49(4):274–281; doi: 10.1053/jhin.2001.1117

23.

Lau

. Breastfeeding challenges and the preterm mother-infant dyad: A conceptual model. Breastfeed Med, 2018; 13(1):8–17; doi: 10.1089/bfm.2016.0206

24.

Pinheiro

, Goldani

, IVAPSA group. Maternal pre-pregnancy overweight/obesity and gestational diabetes interaction on delayed breastfeeding initiation. PLoS One, 2018; 13(6):e0194879; doi: 10.1371/journal.pone.0194879

25.

Fallon

, Bennett

, Harrold

. Prenatal anxiety and infant feeding outcomes: A systematic review. J Hum Lact, 2016; 32(1):53–66; doi: 10.1177/0890334415604129

26.

Horsley

, Chaput

, Da Costa

, et al. Hypertensive disorders of pregnancy and breastfeeding practices: A secondary analysis of data from the All Our Families Cohort. Acta Obstet Gynecol Scand, 2022; 101(8):871–879; doi: 10.1111/aogs.14378

27.

Nyqvist

. Lack of knowledge persists about early breastfeeding competence in preterm infants. J Hum Lact, 2013; 29(3):296–299; doi: 10.1177/0890334413489774

28.

Parker

, Greenberg

, Edwards

, et al. National trends in the provision of human milk at hospital discharge among very low-birth-weight infants. JAMA Pediatr, 2019; 173(10):961–968; doi: 10.1001/jamapediatrics.2019.2645

29.

Briere

, McGrath

, Cong

, et al. Direct-Breastfeeding premature infants in the neonatal intensive care unit. J Hum Lact, 2015; 31(3):386–392; doi: 10.1177/0890334415581798

30.

Fucile

, Wener

, Dow

. Enhancing breastfeeding establishment in preterm infants: A randomized clinical trial of two non-nutritive sucking approaches. Early Hum Dev, 2021; 156:105347; doi: 10.1016/j.earlhumdev.2021.105347

31.

Smith

, Sweigart

, Falatic

, et al. Direct breastfeeding frequency of late preterm and term infants in the neonatal intensive care unit and availability of mother’s own milk at six months of age. a retrospective cohort study. J Perinatol, 2024; 44(9):1307–1311; doi: 10.1038/s41372-024-01972-w

32.

Oliveira

, Valle Volkmer

. Factors associated with breastfeeding very low birth weight infants at neonatal intensive care unit discharge: A single-center Brazilian experience. J Hum Lact, 2021; 37(4):775–783; doi: 10.1177/0890334420981929

33.

Lussier

, Tosi

, Brownell

. Predictors of mother’s own milk feeding at discharge in preterm infants. Adv Neonatal Care, 2019; 19(6):468–473; doi: 10.1097/anc.0000000000000678

34.

Romaine

, Clark

, Davis

, et al. Predictors of prolonged breast milk provision to very low birth weight infants. J Pediatr, 2018; 202:23–30.e1; doi: 10.1016/j.jpeds.2018.07.001

35.

Mamemoto

, Kubota

, Nagai

, et al. Factors associated with exclusive breastfeeding in low birth weight infants at NICU discharge and the start of complementary feeding. Asia Pac J Clin Nutr, 2013; 22(2):270–275; doi: 10.6133/apjcn.2013.22.2.11

36.

Riley

, Schoeny

, Rogers

, et al. Barriers to human milk feeding at discharge of very low-birthweight infants: Evaluation of neighborhood structural factors. Breastfeed Med, 2016; 11(7):335–342; doi: 10.1089/bfm.2015.0185

37.

Chauhan

, Kumar

, Tripathi

, et al. Breastfeeding rates at discharge for very low birthweight neonates and their determinants: An observational study from a tertiary care neonatal intensive care unit in India. J Paediatr Child Health, 2022; 58(9):1653–1660; doi: 10.1111/jpc.16074

38.

Casey

, Fucile

, Dow

. Determinants of successful direct breastfeeding at hospital discharge in high-risk premature infants. Breastfeed Med, 2018; 13(5):346–351; doi: 10.1089/bfm.2017.0209

39.

Nidey

, Groh

, Agnoli

, et al. Breastfeeding initiation and continuation among women with substance and tobacco use during pregnancy: Findings from the pregnancy risk assessment monitoring system 2016-2018. Breastfeed Med, 2022; 17(6):544–549; doi: 10.1089/bfm.2021.0337

40.

Mennella

. Regulation of milk intake after exposure to alcohol in mothers’ milk. Alcohol Clin Exp Res, 2001; 25(4):590–593.

41.

Arnaudo

, Andraka-Christou

, Allgood

. Psychiatric Co-Morbidities in Pregnant women with opioid use disorders: Prevalence, impact, and implications for treatment. Curr Addict Rep, 2017; 4(1):1–13; doi: 10.1007/s40429-017-0132-4

42.

Cook

, Larson

. Breastworks: Breastfeeding practices among women with substance use disorder. Appl Nurs Res, 2019; 47:41–45; doi: 10.1016/j.apnr.2019.04.006

43.

Ciochetto

, Bolzan

, Gonçalves

DDS

, et al. Effects of kangaroo Care on the development of oral skills and achievement of exclusive oral feeding in preterm infants. Codas, 2023; 35(5):e20220070; doi: 10.1590/2317-1782/20232022070

44.

Buckley

, Charles

. Benefits and challenges of transitioning preterm infants to at-breast feedings. Int Breastfeed J, 2006; 1:13; doi: 10.1186/1746-4358-1-13

45.

Song

, Kinshella

, Kawaza

, et al. Neonatal intensive care unit interventions to improve breastfeeding rates at discharge among preterm and low birth weight infants: A systematic review and meta-analysis. Breastfeed Med, 2023; 18(2):97–106; doi: 10.1089/bfm.2022.0151

46.

WHO. Guideline: Protecting, promoting and supporting breastfeeding in facilities providing maternity and newborn services. World Health Organization; 2017.