Effect of Breast Milk on the Frequency of Bronchopulmonary Dysplasia in Very Low Birth Weight Premature Infants: A Meta-analysis

Abstract

Purpose:

To analyze the effect of different feeding types on bronchopulmonary dysplasia (BPD) in very low birth weight preterm infants.

Methods:

The Cochrane Library, PubMed, Embase, China National Knowledge Infrastructure (CNKI), Wanfang Data Knowledge Service Platform, China Biomedical Literature Database (CBM) were searched for literature related to breastfeeding and BPD, with a search period from their inception to January 2023. Two researchers independently screened the literature, extracted data, and assessed the quality of included studies before analyzing the data using Stata16 and RevMan5.4.1 software.

Results:

A total of 17 studies were included. The results showed that there was no significant difference in the frequency of BPD between human milk (HM) and donor human milk (DHM) (OR = 0.54, 95% CI: 0.29–1.03, p = 0.07). However, DHM had a significant effect in reducing the frequency of BPD compared to preterm formula (PF) (OR = 0.62, 95% CI: 0.41–0.94, p = 0.02). Exclusive HM also had a significant effect in reducing the frequency of BPD compared to exclusive PF (OR = 0.51, 95% CI: 0.34–0.78, p = 0.002), as well as compared to any PF (OR = 0.57, 95% CI: 0.37–0.88, p = 0.01). Furthermore, mainly (>50%) HM had a significant effect in reducing the frequency of BPD compared to mainly PF (OR = 0.72, 95% CI: 0.55–0.93, p = 0.01). However, there was no statistically significant difference between any HM and exclusive PF (OR = 0.88, 95% CI: 0.62–1.23, p = 0.46).

Conclusions:

Our study findings suggest that both HM and DHM have a significant protective effect in reducing the frequency of BPD occurrence compared to PF. Furthermore, even when the amount of HM is insufficient, feeding more than 50% of the HM volume still provides a protective effect against the frequency of BPD. Therefore, we recommend feeding infants with more than 50% of HM to harness the protective effect of HM against BPD occurrence.

Introduction

Very low birth weight infants (VLBWIs) are defined as newborns with a birth weight less than 1500 g. The incidence of VLBWI in China is ∼6.6%, while in the United States it is about 1.5%.¹ VLBWIs have immature development in various organs and tissues and often require admission to neonatal intensive care units (NICUs) for treatment after birth. Most of these infants require mechanical ventilation and continuous oxygen therapy in the early stages, not only leading to high rates of complications and mortality but also increasing the risk of severe disabilities among survivors. One of the common severe complications in preterm infants is bronchopulmonary dysplasia (BPD),² which significantly affects the quality of life of affected infants.

BPD,³ also known as chronic lung disease of the newborn, is a common respiratory system disorder among premature infants, especially those with VLBWI or extremely low birth weight (ELBW). The main causes of BPD are damage to the immature lungs caused by various adverse factors such as oxygen toxicity, barotrauma or volutrauma, infection or inflammation, as well as abnormal repair of lung tissue after injury. The smaller the gestational age (GA) and the lower the birth weight the higher the incidence of BPD. BPD not only prolongs the hospital stay of premature infants and increases hospitalization costs but also increases the risk of long-term complications, rehospitalization, and neurocognitive impairment.⁴ BPD is not just a lung disease, and its adverse effects on the respiratory, nervous, and other systems can persist into adulthood, affecting adult health and quality of life.⁵ With the development of assisted reproductive technology, perinatal medicine, and neonatology, the success rate of rescuing VLBWI and ELBWI in China has been increasing. However, this has led to an increasing incidence of BPD.⁶

Breastfeeding has been widely recognized for its role in improving the survival rate and health of preterm infants. Clinical studies both domestically and internationally have shown that breastfeeding can promote lung development in preterm infants, regulate immune function, and reduce the incidence of BPD by improving nutritional status and providing antioxidant and immune active factors.⁷ Currently, the exclusive breastfeeding rate for preterm infants in NICUs in China is less than 15%,⁸ with most being fed a mixture of breast milk and formula at varying ratios. In recent years, numerous studies both domestically and internationally have pointed out the protective effect of breastfeeding on BPD in preterm infants. Therefore, in light of the current low exclusive breastfeeding rate in China, this study is based on the current research status of the impact of breast milk on the risk of BPD, aiming to summarize the existing evidence on the relationship between breast milk and BPD.

Materials and Methods

Database and retrieval strategy

We conducted a systematic search of English databases, including Cochrane Library, PubMed, and Embase, as well as Chinese databases, including China National Knowledge Infrastructure (CNKI), Wan Fang Data, and China Biology Medicine (CBM), from their inception to January 2023. Both subject headings and free-text terms were used for the search. The search terms included “Milk, Human or Breast Feeding or Breast Milk Expression or Milk Banks AND Premature Birth or Infant, Premature or Intensive Care Units, Neonatal or Infant, Very Low Birth Weight or Infant, Extremely Low Birth Weight AND Bronchopulmonary Dysplasia or Outcome Assessment, Health Care”. In addition, we performed a secondary search of the reference lists of included studies to identify relevant research that met our criteria.

Qualification criteria and research selection

The Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram (Fig. 1) and PRISMA checklist (Supplementary Table S1) were used.

FIG. 1.

Flow-chart of search strategy.

Inclusion criteria: (1) randomized controlled trials (RCTs) or cohort studies; (2) study participants were extremely preterm infants (<32 weeks) and/or VLBWIs (weight <1500 g); (3) outcome measures included BPD.

Exclusion criteria: (1) absence of a control group; (2) mixed feeding without specific feeding proportions; (3) erroneous data or study quality lower than 6 points; (4) case reports, meta-analyses, reviews, and other studies.

The definition of BPD was oxygen dependency at 28 days postnatal or at a corrected GA of 36 weeks.

Literature screening and data extraction

Two researchers (L.X. and G.Y.) independently searched the database using the same inclusion criteria to identify potentially eligible studies. Subsequently, the two members compared the initially included studies and resolved any discrepancies through discussion and consultation with a third member when necessary, ultimately selecting the literature that met the study's requirements and could be extracted. Using a predesigned data extraction table, data on study name, publication year, country, grouping status, birth weight, GA, gender, BPD definition, duration of mechanical ventilation, and duration of oxygen therapy were extracted independently by the two researchers, with any disputes resolved through joint discussion and judgment. The registration number for this article is CRD42023380749.

Literature quality evaluation

Two researchers (L.X. and G.Y.) used RevMan 5.4.1 to assess the risk of bias in the included RCT studies and evaluated the quality of the included observational studies using the Newcastle–Ottawa Scale (NOS). The NOS items include selection of studies (0–4 points), comparability (0–2 points), and outcome/exposure (0–3 points), with a total score of 9 points. For RCT studies, the Cochrane risk of bias assessment tool was used, which includes selection bias (random sequence generation, allocation concealment), performance bias (blinding of participants and personnel), detection bias (blinding of outcome assessors), attrition bias (incomplete outcome data), reporting bias (selective reporting of study results), and other potential biases. After independent evaluation by the two researchers, the first and second authors cross-checked and discussed the literature with different opinions. If unresolved, the third researcher arbitrated.

Statistical methods

All data were imported into Stata 16 software and RevMan 5.4.1 software for Meta-analysis. The effect size was measured using the odds ratio (OR), 95% confidence interval (CI), and p-value. Results with a p-value <0.05 were considered statistically significant. Considering the heterogeneity among studies, we used a random-effects model for analysis, which accounts for between-study variability. Subgroup analysis and sensitivity analysis were also performed to further explore the sources of heterogeneity and assess the stability of the results. For publication bias, we evaluated RCT studies using the Cochrane risk of bias assessment tool and assessed all included studies using Egger's Test and Begg's and Mazumdar's Tests. Subsequently, meta-regression analyses were conducted on publication year, sex, birth weight, and GA using a random regression model.

Result

General situation of inclusion in the study

After screening titles, abstracts, and full texts, 138 potentially relevant studies were identified. Ultimately, 17 studies^9–25 met the inclusion criteria, with 3 being RCTs^17,20,23 and 14 being observational studies.^{9–16,18,21,22,24,25} The literature inclusion process is shown in Figure 1, and the general characteristics of the included studies are presented in Table 1. To facilitate data synthesis, multiple intervention groups in some studies were merged for analysis according to the inclusion criteria. Feeding types were categorized into six groups: human milk (HM) and donor human milk (DHM); DHM and preterm formula (PF); Exclusive HM and Exclusive PF; Exclusive HM and Any PF; Mainly HM and Mainly PF; Any HM and Exclusive PF.

Table 1.

Characteristics of Included Studies

Author	Year	Country	Study design	Study population	Groups	Sample(s)	Respiratory outcome(s)	NOS
Assad et al.⁹	2016	America	Retrospective cohort	<1500 g and/or <28 w	HM Bovine, HM M: Bovine, HM, PF PF	293	BPD36	7
Spiegler et al.¹⁰	2016	Germany	Prospective cohort	22–32 w	HM PF M	462	BPD36	9
Xu et al.¹¹	2020	China	Retrospective cohort	<1500 g and <34 w	PF 1–24 mL HM 25–49 mL HM >50 mL HM	964	BPD28	6
Serrao et al.¹²	2016	Italy	Prospective cohort	<31 w	>50 mL HM <50 mL HM	52	BPD36	7
Sun et al.¹³	2019	China	Prospective cohort	<30 w	HM DHM	207	BPD36	7
Furman et al.¹⁴	2003	America	Prospective cohort	<30 w	PF 1–24 mL HM 25–49 mL HM >50 mL HM	119	BPD36	7
Montjaux-Régis et al.¹⁵	2011	France	Prospective cohort	<32 w	<20% HM 20–80% HM > ≥ 80%HM	48	BPD36	7
Fang et al.¹⁶	2021	China	Prospective cohort	<1500 g or <32 w	DHM PF	304	BPD28	7
Petrova et al.¹⁸	2020	America	Prospective cohort	≤32 w	<35% HM 35–84.9% HM ≥85%HM	84	Unknown	9
Kreissl et al.¹⁹	2017	Austria	Prospective cohort	<1500 g and <32 w	DHM PF	283	Unknown	7
Sisk et al.²¹	2007	America	Prospective cohort	≤30 w	<50% HM ≥50%HM	202	Unknown	6
Madore et al.²²	2017	America	Prospective cohort	<1000 g	HM PF DHM	81	BPD36	7
Lund et al.²⁴	2022	Sweden	Prospective cohort	<32 w	<20% HM 20–80% HM >80%HM	440	BPD36	7
Vohr et al.²⁵	2006	America	Prospective cohort	<1500 g	HM PF	1035	BPD36	9
Cossey et al.¹⁷	2013	Belgium	RCT	<1500 g and/or <32 w	HM DHM	303	BPD36	—
Costa et al.²⁰	2018	Italy	RCT	<32 w	DHM PF	70	Unknown	—
Schanler et al.²³	2005	America	RCT	≤29 w	HM PF DHM	243	Unknown	—

HM, human milk; PF, preterm formula; DHM, donor human milk; M, mixed feeding with HM and PF; NOS, Newcastle–Ottawa Scale; RCT, Randomized Controlled Trial; BPD36, need for supplemental oxygen or any respiratory support, including moderate and severe BPD, at 36 weeks postmenstrual age; BPD28, moderate to severe BPD defined as requiring >21% supplemental oxygen for at least 28 days of oxygen therapy.

Quality assessment and bias risk assessment

Three RCTs^17,20,23 scored high in terms of random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, and selective reporting. One study²³ was assessed as having an unclear risk of bias due to incomplete outcome data, and two studies^17,20 were assessed as having an unclear risk of bias for other potential biases (Supplementary Fig. S10). Fourteen observational studies^{9–16,18,21,22,24,25} scored between 6 and 9 points on the NOS, with the highest score being 9 points. The average score for the included studies was 7.3 points (Table 2).

Table 2.

Newcastle–Ottawa Scale Scoring Results

Study	Selection (0–4) points)	Comparability (0–2) points)	Outcome (0–3) points))	Total (0–9)points))
Assad et al.⁹	4	0	3	7
Spiegler et al.¹⁰	4	2	3	9
Xu et al.¹¹	4	0	2	6
Serrao et al.¹²	4	0	3	7
Sun et al.¹³	4	0	3	7
Furman et al.¹⁴	4	0	3	7
Montjaux-Régis et al.¹⁵	4	0	3	7
Fang et al.¹⁶	4	0	3	7
Petrova et al.¹⁸	4	2	3	9
Kreissl et al.¹⁹	4	0	3	7
Sisk et al.²¹	4	0	2	6
Madore et al.²²	4	0	3	7
Lund et al.²⁴	4	0	3	7
Vohr et al.²⁵	4	2	3	9

Selection, Maximum score of 4 points; Comparability, Maximum score of 2 points; Outcome, Maximum score of 3 points.

Meta-analysis and meta-regression results

HM versus DHM

Four studies^13,17,22,23 (n = 717) compared infants fed exclusively HM to those fed DHM. The meta-analysis results showed that, compared to DHM, exclusive HM had no significant effect on BPD (OR = 0.54, 95% CI: 0.29–1.03, p = 0.07) (Table 3 and Supplementary Fig. S1). After excluding the study by Schanler et al.,²³ who did not provide a clear definition of BPD in their article, exclusive HM still had no significant effect on BPD (OR = 0.45, 95% CI: 0.19–1.09, p = 0.08). After excluding the study by Sun et al.,¹³ whose control group was frozen HM without pasteurization, the effect of exclusive HM on BPD remained statistically nonsignificant (OR = 0.70, 95% CI: 0.45–1.09, p = 0.11).

Table 3.

Meta-Analysis Results

	Pooled analysis		Heterogeneity		Publication bias
	OR (95% CI)	p-Value	I ²	p-Value	Egger's test		Begg's and Mazumdar's tests
	OR (95% CI)	p-Value	I ²	p-Value	T	p-Value	Z	p-Value
HM vs. DHM k = 4 ^13,17,22,23	0.54 (0.29–1.03)	0.03	48	0.12	1.17	0.362	0	1.000
DHM vs. PF k = 5 ^{16,19,20,22,23}	0.62 (0.41–0.94)	0.02	0	0.9	−0.24	0.822	−0.24	1.000
Exclusive HM vs. Exclusive PF k = 4 ^9,10,22,23	0.51 (0.34–0.78)	0.002	6.5	0.36	−1.03	0.410	1.02	0.308
Exclusive HM vs. Any PF k = 5 ^{9,10,22,23,25}	0.57 (0.37–0.88)	0.01	63.3	0.028	2.6	0.081	0.24	0.806
Mainly HM vs. Mainly PF k = 12 ^{9–15,18,21–25}	0.72 (0.55–0.93)	0.01	38.1	0.087	−0.55	0.594	1.56	0.119
Any HM vs. Exclusive PF k = 7 ^{9–11,14,22,23,25}	0.88 (0.62–1.23)	0.46	65.7	0.008	−0.58	0.589	0.30	0.764

HM; human milk; PF, preterm formula; DHM, donor human milk; OR, odds ratio; 95% CI, 95% confidence interval.

When analyzing the between-group differences in GA, the meta-analysis found no significant differences (OR = −0.05, 95% CI: −0.20 to 0.10, p = 0.59) (Supplementary Fig. S2). Two of the studies^17,23 were RCTs, and a subgroup analysis showed that in RCTs, exclusive HM had no significant effect on BPD (OR = 0.69, 95% CI: 0.42–1.12, p = 0.13) (Supplementary Fig. S3). Sensitivity analysis revealed that after removing any single study, exclusive HM had no significant effect on BPD (Supplementary Fig. S4).

Three studies^13,17,23 reported data on the number of days on mechanical ventilation. The meta-analysis showed that exclusive HM had a significant effect on reducing the number of days on mechanical ventilation (MD −2.66, 95% CI: −4.14 to −1.18, p = 0.0004) (Supplementary Fig. S5).

The meta-regression analysis results showed that year, sex, GA, and birth weight had no statistical significance for BPD, indicating that these factors do not affect the incidence of BPD (Table 4).

Table 4.

Meta-Regression Results

	No. of primary studies	Total sample size	Coef.	Std.err	R-squared	p-Value
HM vs. DHM k = 4
Year	4	717	−1.500	1.123	0.21	0.313
Male	4	717	22.059	64.504	−0.42	0.765
GA	4	717	−7.693	3.954	0.48	0.191
Birth weight	3	510	−0.061	0.034	0.52	0.327
DHM vs. PFk = 5
Year	5	882	0.952	0.3544	0.61	0.074
Male	5	882	−20.862	39.592	−0.22	0.635
GA	5	882	2.220	2.183	0.01	0.384
Birth weight	5	882	0.014	0.017	−0.10	0.487
HM vs. PFk = 4
Year	4	795	−1.314	1.585	−0.12	0.494
Male	4	795	−16.624	167.158	−0.19	0.930
GA	4	795	−12.170	6.973	0.41	0.223
Birth weight	3	678	−0.156	0.064	0.71	0.247
Exclusive HM vs. Any PFk = 5
Year	5	2850	17.450	23.370	−0.12	0.509
Male	5	2850	−505.974	2057.500	−0.31	0.822
GA	5	2850	245.657	65.358	0.77	0.033
Birth weight	4	2684	1.732	0.985	0.41	0.221
Mainly HM vs. Mainly PFk = 12
Year	12	3617	2.402	2.310	0.57	0.346
Male	11	3415	−0.319	0.134	0.57	0.063
GA	12	3617	−25.565	24.151	0.57	0.338
Birth weight	11	3500	−0.177	0.222	0.57	0.460
Any HM vs. Exclusive PFk = 7
Year	7	4060	−15.519	13.045	0.06	0.288
Male	7	4060	506.808	1785.032	−0.18	0.788
GA	7	4060	−124.109	75.043	0.23	0.158
Birth weight	6	3767	−0.631	0.800	−0.09	0.485

HM, human milk; PF, preterm formula; DHM, donor human milk; GA, gestational age; k, numbers of primary studies.

DHM versus PF

Five studies^{16,19,20,22,23} (n = 882) compared infants fed DHM to those fed exclusively PF. The meta-analysis found a statistically significant effect of DHM in reducing the frequency of BPD compared to PF (OR = 0.62, 95% CI: 0.41–0.94, p = 0.02) (Table 3, and Supplementary Fig. S1). When analyzing the between-group differences in GA, the meta-analysis found no significant differences (OR = 0.04, 95% CI: −0.09 to 0.17, p = 0.55) (Supplementary Fig. S2); sensitivity analysis indicated that the results remained significant after excluding any single study.

The meta-regression analysis results showed that year, sex, GA, and birth weight had no statistical significance for BPD (Table 4).

Exclusive HM versus exclusive PF

Four studies^9,10,22,23 (n = 795) compared infants fed exclusively HM to those fed exclusively PF. The meta-analysis found a statistically significant effect of exclusive HM in reducing the frequency of BPD compared to exclusive PF (OR = 0.51, 95% CI: 0.34–0.78, p = 0.002) (Table 3 and Supplementary Fig. S1). One study²³ did not provide a clear definition of BPD, and after excluding it, the meta-analysis showed no statistically significant effect of exclusive HM on BPD (OR = 0.59, 95% CI: 0.32–1.07, p = 0.08). When we analyzed the between-group differences in GA, the meta-analysis found no significant differences (OR = −0.03, 95% CI: −0.17 to 0.12, p = 0.33) (Supplementary Fig. S2). However, one study had a significant between-group difference in GA of ≥0.5 weeks. After excluding the study by Assad et al.,⁹ the results still showed a significant effect of exclusive HM on BPD (OR = 0.46, 95% CI: 0.30–0.69, p = 0.0002).

Two studies^10,23 provided data on the number of days of mechanical ventilation. The meta-analysis showed that receiving exclusive HM had no significant effect on the duration of mechanical ventilation (MD = −2.84, 95% CI: −9.57 to 3.90, p = 0.41) (Supplementary Fig. S6).

The meta-regression analysis results showed that year, sex, GA, and birth weight had no statistical significance for BPD (Table 4).

Exclusive HM versus any PF

Five studies^{9,10,22,23,25} (n = 2850) compared infants fed exclusively HM to those fed any amount of PF (including exclusive PF). The meta-analysis found a significant effect of feeding exclusive HM in reducing the frequency of BPD compared to any PF (OR = 0.57, 95% CI: 0.37–0.88, p = 0.01) (Table 3 and Supplementary Fig. S1). After excluding the study by Schanler et al.,²³ which did not provide a clear definition of BPD, the results showed a significant effect of exclusive HM on BPD (OR = 0.61, 95% CI: 0.38–0.97, p = 0.04). Upon analyzing the between-group differences in GA, we found no significant overall differences (MD = −0.01, 95% CI: −0.10 to 0.08, p = 0.55) (Supplementary Fig. S2). However, one study⁹ had a between-group GA difference of ≥0.5 weeks. After excluding this study, the meta-analysis of the remaining four studies still showed a significant effect of exclusive HM on BPD (OR = 0.61, 95% CI: 0.38–0.98, p = 0.04).

The meta-regression analyses showed that there were no significant results regarding year, birth weight, and sex, but they showed a statistically significant result with regard to GA (Table 4).

Mainly HM versus mainly PF

Twelve studies^{9–12,14,15,18,21–25} (n = 3617) compared infants fed mainly HM to those fed mainly PF (including exclusive HM and exclusive PF). The meta-analysis found a significant effect of mainly HM in reducing the frequency of BPD compared to mainly PF (OR = 0.72, 95% CI: 0.55–0.93, p = 0.01) (Table 3 and Supplementary Fig. S1). However, three studies^18,21,22 did not provide a clear definition of BPD. After excluding these studies, the meta-analysis still showed a significant effect of mainly HM on reducing BPD (OR = 0.71, 95% CI: 0.53–0.95, p = 0.02).

In addition, when analyzing the between-group differences in GA, we found significant overall differences in GA (OR = −0.14, 95% CI: −0.21 to −0.07, p < 0.00001) (Supplementary Fig. S2). However, it is worth noting that four studies had substantial differences in GA between groups. To address this, we conducted a subgroup analysis by excluding these four studies^9,14,18,21 with between-group GA differences of ≥0.5 weeks, leaving eight studies,^{10–12,15,22–25} which showed a significant effect of mainly HM on BPD (OR = 0.63, 95% CI: 0.46–0.86, p = 0.004) (Supplementary Fig. S7).

Four studies^10,12,21,23 provided data on the number of days on mechanical ventilation. The meta-analysis showed that receiving mainly HM had no significant effect on the duration of mechanical ventilation (MD = 0.23, 95% CI : −1.77 to 2.23, p = 0.82) (Supplementary Fig. S8). Three studies^10,12,21 provided data on the number of days on oxygen supplementation. The meta-analysis showed that receiving mainly HM had no statistically significant effect on this outcome (MD = 3.21, 95% CI: −13.40 to 19.83, p = 0.70) (Supplementary Fig. S9).

The meta-regression analysis results showed that year, sex, GA, and birth weight had no statistical significance for BPD (Table 4).

Any HM versus exclusive PF

Seven studies^{9–11,14,22,23,25} (n = 4060) compared infants fed any HM to those fed exclusively PF (including exclusive HM). The meta-analysis found no significant effect of any HM on BPD (OR = 0.88, 95% CI: 0.62–1.23, p = 0.46) (Table 3 and Supplementary Fig. S1). After excluding the study by Xu et al.,¹¹ which used a different BPD definition, there was still no significant effect of any HM on BPD (OR = 0.96, 95% CI: 0.62–1.48, p = 0.84). Heterogeneity was considerable (p = 0.008, I² = 65.7%). After analyzing the between-group differences in GA, we found significant overall differences (OR = −0.15, 95% CI: −0.23 to −0.06, p = 0.004) (Supplementary Fig. S2). GA might be a confounding factor. In the subgroup analysis, we excluded three studies^9,11,14 with between-group GA differences ≥0.5 weeks, leaving four studies.^10,22,23,25 The meta-analysis of these studies showed no significant effect of any HM on BPD (OR = 0.78, 95% CI: 0.57–1.06, p = 0.11), with moderate heterogeneity (p = 0.18, I² = 39%).

The meta-regression analysis results showed that year, sex, GA, and birth weight had no statistical significance for BPD (Table 4).

Discussion

Apart from when HM and DHM are used as the exclusive sources of nutrition, a predominantly HM-based feeding approach similarly demonstrates significant protective effects against BPD.

This study systematically analyzed the evidence of the effect of HM on BPD compared to DHM or PF. The results of this meta-analysis showed that although feeding exclusive HM had no significant effect on BPD compared to DHM (p = 0.07), it significantly reduced the number of days on mechanical ventilation (p = 0.0004); infants fed with DHM had a lower incidence of BPD compared to PF (p = 0.02). Furthermore, in the analysis of HM and PF, both exclusive HM and mainly HM significantly reduced the risk of BPD compared to PF (p = 0.002, p = 0.01), while any HM showed no significant protective effect on BPD compared to exclusive PF (p = 0.46).

This meta-analysis showed that HM had no protective effect on BPD compared to DHM, which is similar to the results obtained by Eduardo Villamor-Martínez et al.²⁶ This may be due to the limited number of included studies, where two RCT studies received partial or mainly HM feeding. However, the study by Eduardo Villamor-Martínez et al. demonstrated that raw mother's own milk significantly reduced the risk of BPD compared to pasteurized mother's own milk. When comparing exclusive HM, mainly HM, and PF, there were significant effects on the risk of BPD, which is consistent with other studies. For example, Fonseca et al.²⁷ found that HM intake of 7 mL/(kg·d) within 6 weeks after birth could prevent the occurrence of BPD, and HM had a significant protective effect on moderate and severe BPD. Similarly, in a 5-year single-center prospective observational study in the United States in 2017,²⁸ univariate and multivariate analyses showed that for every 10% increase in HM feeding by 36 weeks of corrected GA, the incidence of BPD decreased by 9.5%, confirming a dose-effect relationship between HM and BPD.

HM has significant protective effects against BPD. However, current breastfeeding rates are low. It is necessary to implement more measures to increase breastfeeding rates.

HM feeding is the main preventive and therapeutic measure for BPD and has become the “gold standard” for enteral nutrition support in premature infants due to its nutritional richness, low osmolarity, fast gastric emptying, and abundant immune-active substances.²⁹ Clinical studies both domestically and internationally have shown that HM feeding can promote lung development, regulate immune function, reduce the incidence of BPD by improving nutritional status, and providing antioxidant and immunomodulatory factors.^30,31 A 2018 Chinese study³² found that exclusive HM feeding was superior to liquid preterm formula feeding in reducing BPD incidence by 10.4% in VLBWIs. Our recent study³³ in 2020 showed that oral colostrum administration for mechanically ventilated ELBW infants within 24 hours after birth, along with a 5-day continuous intervention, could reduce the incidence of BPD by 11.25%.

Current research indicates that HM also plays a protective role in preventing retinopathy of prematurity and necrotizing enterocolitis in preterm infants. However, the low rates of breastfeeding in NICUs, as evidenced by China's NICU-exclusive breastfeeding rate of only 9.3%,³⁴ highlight the need for concerted efforts to improve breastfeeding practices. Comprehensive strategies encompassing education, support, and policy interventions are necessary to increase breastfeeding rates and optimize the benefits of HM feeding for preterm infants.

The results of the meta-regression analysis indicate that GA may be a potential factor influencing the risk of BPD.

Our meta-regression analysis results only found a potential impact of GA on the risk of BPD and could not explain the relationships between publication year, sex, birth weight, and BPD. However, in the study by Meizhu et al.,³⁵ it was shown that birth weight, male sex, mechanical ventilation, mechanical ventilation >7 days, respiratory distress syndrome in preterm infants, patent ductus arteriosus in preterm infants, sepsis, and NEC in preterm infants were all risk factors for BPD. Therefore, caution should be exercised in interpreting our results, which may be attributed to the limited number of studies included in our analysis. Further research, incorporating more and higher quality studies, is required to gain a better understanding of the relationship between these factors and BPD.

Limitation

We acknowledge the limitations of our study, which include the relatively small number of studies included and the significant differences in sample sizes analyzed among them. In addition, the majority of studies included in our analysis were observational in nature, resulting in significant heterogeneity in the demographic characteristics of the populations studied. It is important to note that our study did not specifically focus on exploring the protective effect of breast milk on the severity of BPD. To address these limitations, future research should aim to include larger and more diverse samples, as well as utilize more rigorous study designs such as RCTs. Furthermore, future studies should aim to explore the impact of breast milk on both the frequency and severity of BPD to provide a more comprehensive understanding of its protective effects.

Conclusion

In conclusion, our study findings suggest that both HM and DHM have a significant protective effect in reducing the frequency of BPD occurrence compared to PF. If the amount of HM is insufficient, feeding a proportion of more than 50% HM still provides a protective effect on BPD. GA may potentially play a role in the risk of BPD. The dose-effect relationship between HM and BPD still requires further research to explore the changes in BPD incidence when feeding more specific amounts of HM.

Footnotes

Acknowledgment

We have not received funds in support of research work or for covering the costs to public in open access.

Authors' Contributions

X.C.: Guarantor of the article, study concept and design, literature search, data analysis, and article writing. X.L.: literature search, data abstraction, participant article writing. Y.G.: contributed to interpretation of results and reviewed and revised the article. C.L.: contributed to the statistical analyses and interpretation of the results, and reviewed and revised the article. M.P.: literature search. All authors approved the final article as submitted.

Disclosure Statement

The authors declare no conflict of interest.

Funding Information

No funding was received for this article.

Supplementary Material

References

, Wang

, Wu

, et al. A survey of neonatal birth conditions in urban China in 2005. Chin J Contemp Pediatr, 2012; 14(1):7–10.

Jiang

, Yan

, Li

, et al. Mortality and Morbidity in Infants <34 Weeks' Gestation in 25 NICUs in China: A Prospective Cohort Study. Front Pediatr, 2020; 8:33; doi: 10.3389/fped.2020.00033

Shao

, Ye

, Qiu

. Practical Neonatology 5th edition[M]. People's Medical Publishing House: Beijing; 2019; p. 1150.

Principi

, Di Pietro

, Esposito

. Bronchopulmonary dysplasia: clinical aspects and preventive and therapeutic strategies. J Transl Med, 2018; 16(1):36; doi: 10.1186/s12967-018-1417-7

Davidson

, Berkelhamer

. Bronchopulmonary dysplasia: Chronic lung disease of infancy and long-term pulmonary outcomes. J Clin Med, 2017; 6(1):4; doi: 10.3390/jcm6010004

, Ma

Importance of respiratory support for severe bronchopulmonary dysplasia. Chin J Pediatr, 2020:355-356–357.

Gila-Diaz

, Arribas

, Algara

, et al. A review of bioactive factors in human breastmilk: A focus on prematurity. Nutrients, 2019; 11(6):1307; doi: 10.3390/nu11061307

Zhang

, Li

, Yu

. Investigation and analysis of human milk feeding status during hospitalization of extremely low and very low birth weight infants. Chin J Pediatr, 2020; 58(5):387–390.

Assad

, Elliott M

, Abraham J

. Decreased cost and improved feeding tolerance in VLBW infants fed an exclusive human milk diet. J Perinatol, 2016; 36(3):216–220.

10.

Spiegler

, Preuß

, Gebauer

, et al. Does breastmilk influence the development of bronchopulmonary dysplasia?. J Pediatr, 2016; 169:76–80.e4.

11.

, Yu

, Li

, et al. Dose-dependent effect of human milk on Bronchopulmonary dysplasia in very low birth weight infants. BMC Pediatr, 2020; 20(1):522.

12.

Serrao

, Papacci

, Costa

, et al. Effect of early expressed human milk on insulin-like growth factor 1 and short-term outcomes in preterm infants. PLoS One, 2016; 11(12):e0168139.

13.

Sun

, Han

, Cheng

, et al. Testing the feasibility and safety of feeding preterm infants fresh mother's own milk in the NICU: A pilot study. Sci Reports, 2019; 9(1):941.

14.

Furman

, Taylor

, Minich

, et al. The effect of maternal milk on neonatal morbidity of very low-birth-weight infants. Arch Pediatr Adolesc Med, 2003; 157(1):66–71.

15.

Montjaux-Régis

, Cristini

, Arnaud

, et al. Improved growth of preterm infants receiving mother's own raw milk compared with pasteurized donor milk. Acta Paediatr Int J Paediatr, 2011; 100(12):1548–1554.

16.

Fang

, Zhang

, Wu

, et al. Is preterm donor milk better than preterm formula for very-low-birth-weight infants?. Food Nutr Res, 2021; 65; doi: 10.29219/fnr.v65.5346

17.

Cossey

, Vanhole

, Eerdekens

, et al. Pasteurization of mother's own milk for preterm infants does not reduce the incidence of late-onset sepsis. Neonatology, 2013; 103(3):170–176.

18.

Petrova

, Eccles

, Mehta

. Role of the proportional intake of fortified mother's own milk in the weight gain pattern of their very-preterm-born infants. Nutrients, 2020; 12(6):1571.

19.

Kreissl

, Sauerzapf

, Repa

, et al. Starting enteral nutrition with preterm single donor milk instead of formula affects time to full enteral feeding in very low birthweight infants. Acta Paediatr Int J Paediatr, 2017; 106(9):1460–1467.

20.

Costa

, Maggio

, Alighieri

, et al. Tolerance of preterm formula versus pasteurized donor human milk in very preterm infants: A randomized non-inferiority trial. Italian J Pediatr, 2018; 44(1):96.

21.

Sisk

, Lovelady

, Dillard

, et al. Early human milk feeding is associated with a lower risk of necrotizing enterocolitis in very low birth weight infants. J Perinatol, 2007; 27(7):428–433.

22.

Madore

, Bora

, Erdei

, et al. Effects of donor breastmilk feeding on growth and early neurodevelopmental outcomes in preterm infants: An observational study. Clin Ther, 2017; 39(6):1210–1220.

23.

Schanler

, Lau

, Hurst

, et al. Randomized trial of donor human milk versus preterm formula as substitutes for mothers' own milk in the feeding of extremely premature infants. Pediatrics, 2005; 116(2):400–406; doi: 10.1542/peds.2004-1974

24.

Lund

A-M

, Domellöf

, Pivodic

, et al. Mother's own milk and its relationship to growth and morbidity in a population-based cohort of extremely preterm infants. J Pediatr Gastroenterol Nutr, 2022; 74(2):292–300.

25.

Vohr

, Poindexter

, Dusick

, et al.; NICHD Neonatal Research Network. Beneficial effects of breast milk in the neonatal intensive care unit on the developmental outcome of extremely low birth weight infants at 18 months of age. Pediatrics, 2006; 118(1):e115–e123; doi: 10.1542/peds.2005-2382

26.

Villamor-Martínez

, Pierro

, Cavallaro

, et al. Donor human milk protects against bronchopulmonary dysplasia: A systematic review and meta-analysis. Nutrients, 2018; 10(2):238; doi: 10.3390/nu10020238

27.

Fonseca

, Senna

, Silveira

, et al. Association between breast milk and bronchopulmonary dysplasia: A single center observational study. Amer J Perinatol, 2016; 34(03):264–269; doi: 10.1055/s-0036-1586503

28.

Patel

, Johnson

, Robin

, et al. Influence of own mother's milk on bronchopulmonary dysplasia and costs. Arch Dis Child Fetal Neonatal Ed, 2017; 102(3):F256–F261; doi: 10.1136/archdischild-2016-310898

29.

Wang

DH.

Improving nutritional management by standardizing feeding of premature infants. Chin J Pediatr, 2020; 58(2):155.

30.

Huang

, Tong

. Protective effect of breastfeeding on bronchopulmonary dysplasia in preterm infants. Chin J Neonatol, 2022; 37(5):470–473.

31.

Huang

, Zhang

, Tang

, et al. Human milk as a protective factor for bronchopulmonary dysplasia: A systematic review and meta-analysis. Arch Dis Child Fetal Neonatal Ed, 2019; 104(2):F128–F136.

32.

Chen

, Zheng

, Chen

, et al. Value of exclusive human milk feeding for extremely low birth weight preterm infants. J Nurses Adv Pract, 2018:33(9):783–785.

33.

, Pan

, Chen

, et al. Effects of early oral colostrum administration on mechanical ventilation of extremely low birth weight preterm infants. Chin J Nurs, 2020; 55(6):884–888.

34.

Wang

, Wei

, Luo

, et al. Analysis of influencing factors of exclusive breastfeeding for 445 premature infants within 6 months. Chin J Gen Pract, 2021; 19(10):1686.

35.

, Liu

, Xiao

, et al. Meta-analysis of risk factors for bronchopulmonary dysplasia in premature infants. Chin J Gen Pract, 2023; 26(03):356–366.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.26 MB

0.27 MB

0.25 MB

0.23 MB

0.21 MB

0.12 MB

0.11 MB

0.16 MB

0.20 MB

0.08 MB