Abstract
Introduction:
Restricted fetal and neonatal growth is a known risk factor for bronchopulmonary dysplasia (BPD) in premature infants. However, the impact of nutrition and infant growth specifically on lung growth in BPD in unknown. Moreover, whether all lung growth in BPD is beneficial is unclear. We hypothesized that lung growth and development and severity of BPD directly relate to caloric and protein intake, weight gain, and linear growth of premature neonates.
Methods:
In this retrospective study, caloric and protein intake for the first four weeks of life, growth parameters along with lung volume, mass, density, and BPD severity obtained by ultrashort echo time (UTE) MRI, were analyzed.
Results:
The cohort included 95 neonates with mean GA 26.1 weeks and BW 790 g. Infants with Grade 2 and 3 BPD had less caloric and protein intake during first 4 weeks of life vs Grade 1 BPD (96/98 vs 106 kcal/kg/d; 3.79/3.75 vs 3.99 g protein/kg/d; p< 0.05). UTE MRI showed that lung mass per body surface area increased with increasing BPD severity (237, 311, 384 g/m2 for Grade 1, 2, and 3, respectively, p< 0.05). Increased caloric intake was associated with decreased lung mass (p=0.02) and improved BPD score on MRI (p=0.04)
Discussion/Conclusion:
Decreased nutritional intake during the first four weeks of life appears to be associated with more severe BPD, increased lung mass and more severe lung disease on MRI.
Keywords: outcomes prediction, nutrition, pulmonary imaging, neonatal lung disease, prematurity
Introduction:
Bronchopulmonary dysplasia (BPD) is one of the most common diseases of prematurity and significantly contributes to mortality and morbidity of preterm infants beyond infancy through childhood and adulthood [1]. Numerous perinatal and postnatal insults affect lung growth and development.
Although restricted fetal and neonatal growth is a known risk factor for BPD, there is a limited understanding of how nutritional intake, weight gain, and linear growth are directly related to lung growth and development in preterm infants with BPD [2-4]. Infants with BPD frequently have poor growth despite meeting recommended caloric goals for a given weight and age. There is some evidence that infants with BPD have increased energy expenditure and failure to grow because of increased work of breathing, early use of diuretics and corticosteroids, a history of intrauterine growth restriction, fluid restriction, and decreased feeding tolerance [5-8]. Studies show that both under provision and over provision of macronutrients can have a negative impact on the health and development of preterm lungs [3, 6, 8-11]. As a result, the optimal nutritional balance for infants at risk of BPD is unknown [3, 5-9, 12].
Previous research has focused on the association of nutrition with clinical outcomes of BPD rather than a more direct analysis of lung growth. Recent advancements in lung magnetic resonance imaging (MRI), particularly with ultrashort echo time (UTE) sequences, allow for high-resolution imaging of lung tissues. We previously used UTE MRI in a pilot study to evaluate lung mass growth in premature infants with BPD and demonstrated an association between lung mass and BPD severity such that infants with BPD who survived had a higher lung mass than control infants without BPD and infants with BPD that did not survive [13]. Therefore, it is unclear if all increases in lung size correspond to lung growth or if increases in lung mass are due to other factors such as inflammation, scarring and fibrosis. Lastly, it is unclear if lung growth improves or worsens outcomes in infants with BPD. In our pilot study, we also found an association between lung mass growth and infant growth; however, the specific factors that influence lung mass growth in premature infants are unknown [13]. In the current study, we focused on the potential link between nutritional and lung growth in BPD and hypothesize that increases in early infant nutrition will associate with improved lung growth and decreased severity of BPD.
Materials and methods:
This study was a secondary analysis of a cohort of 10 control infants and 95 infants with BPD born 2014-2021. Institutional review board approval was obtained prior to research activities (IRB 2013-6101 and IRB 2018-0958). Consent was obtained from parents/legal guardian prior to MRI.
2.1. Population and variables.
Infants born before 32 weeks of gestational age (GA) with weight <1500 gram and diagnosed with BPD of any severity who underwent lung UTE MRI were included. The severity of BPD (Grade 1, 2, and 3) was determined based on the amount of respiratory support at 36 weeks corrected GA based on Jensen criteria, 2019 [14]. Infants (n=5) with major congenital abnormalities or genetic disorders that may affect lung development, with missing data of the early nutritional intake and significant bulk motion artifact on MRI, or respiratory infection at the time of MRI were excluded. Ten late preterm and term neonates with no history of lung disease were included as a control group [15].
Z-scores for anthropometric data were obtained from the Fenton growth chart according to sex and corrected GA. Weight gain velocity was calculated using the Patel Exponential Method [16]. Body surface area (BSA) was calculated using the Mosteller equation (the square root of [height (in centimeters) x weight (in kilograms) / 3600]).
Nutritional data included daily fluid, protein, and caloric intake for the first 28 days of life. Protein, lipid, carbohydrate, and total caloric intake from TPN or IVF were calculated for each day for the first 28 days of life based on the content of IVF or TPN (extracted from the actual TPN or IVF orders) and actual volume delivered as per nursing documented flowsheets of “Intake and output”. Protein and caloric intake were calculated for enteral feeds based on consumed (not ordered) volume and assumption of protein and caloric content of unfortified breastmilk as 1.1 g/100 ml and 20 kcal/oz or fortified breastmilk. The manufacturer product information sheet was used for enteral formula and parenteral nutrition macronutrient calculations.
2.2. MRI Equipment and Acquisition.
Lung imaging was obtained on a small-footprint 1.5T orthopedic scanner (ONI Medical Systems; currently operating with HDx software, GE Healthcare, Waukesha, WI), modified for neonatal imaging [17]. Infants were prepared for MRI examination using the “feed and swaddle” method. All subjects were imaged during tidal breathing (no paralysis, sedation or breath holds) on their clinically indicated respiratory support with no intravenous contrast agents. Sedation/anesthesia were not administered for the purposes of the research MRI.
2.3. Image analysis.
Total lung volume was calculated via segmentation of whole-lung parenchyma on 3D UTE images using Amira (FEI Visualization Sciences Group) [13]. Lung density was calculated for the entire lung by normalizing lung tissue intensities between air (noise) and muscle intensity values on UTE MRI, yielding approximately volumetric density in grams per cubic centimeter (g/cm3), as previously described [15]. Whole-lung mass was calculated as a product of whole-lung volume and mean whole-lung density. MRI score as a measure of BPD lung disease severity was determined as previously described; the range of total score was 0 (no BPD) through 14 (severe) [17].
2.4. Statistical analysis.
Comparison of quantitative metrics between groups of patients was made using ANOVA. Least significant difference post-hoc tests were performed as applicable. Spearman correlations were used for pairwise relationships between the continuous covariates/outcomes of interest. Chi-squared test or exact chi-squared (the exact chi-squared was used when more than 20% of data cells had a lower than expected count (<5)) was used for two categorical variables. For adjusted linear regression analysis, the variables significantly associated with both an outcome and either BPD variable at the 0.1 level were included in the model for testing the adjusted impact of the intake variable on the corresponding outcome. Comparisons were considered statistically significant for p-values < 0.05; given the small sample size, we also indicated those that were marginally significant for p-values between 0.05 and 0.10. All p-values were two-tailed.
Results:
Ninety-eight potentially eligible preterm infants had lung MRI. Two infants were excluded due to bulk motion artifact and missing clinical data, and one due to a genetic disorder with congenital tracheal stenosis. The final cohort included ninety-five neonates with nineteen (20%) infants diagnosed with Grade 1, 32 (34%) – Grade 2, and 44 (46%) – Grade 3 BPD (Table 1).
Table 1.
(A) Characteristics of a cohort of premature infants with BPD and Controls and (B) Postnatal clinical characteristics of infants with different severity of BPD
| A | ||
|---|---|---|
| Patient characteristics | BPD n=95 | Controls n=10 |
| Gestational age, weeks, mean (range) | 26.1 (23.0 - 30.4) | 37.3 (34.7-39.2) |
| Birth weight, g, mean (range) | 790 (430 - 1720) | 2759 (1940-3255) |
| Sex, male, n (%) | 41 (43) | 2 (20) |
| SGA, n (%) | 15 (16) | 0 (0) |
| Race, n (%) | ||
| White | 65 (68) | 8 (80) |
| Black | 29 (31) | 2 (20) |
| Asian | 1 (1) | 0 |
| Ethnicity, n (%) | ||
| Hispanic | 3 (3) | 0 |
| Not Hispanic | 92 (97) | 10 (100) |
| BPD severity, n (%) | ||
| Grade 1 | 19 (20) | |
| Grade 2 | 32 (34) | |
| Grade 3 | 44 (46) | |
| Singleton, n (%) | 76 (80) | 10 (100) |
| Maternal age, mean (SD) | 28 (6) | 27 (17-35) |
| Cesarian delivery, n (%) | 63 (66) | 4 (40%) |
| Antenatal steroids, 2 doses or >, n (%) | 80 (84) | 0 (0) |
| Maternal comorbidities, n (%) | ||
| Diabetes | 9 (10) | 1 (10) |
| Hypertension | 28 (30) | 1 (10) |
| Preeclampsia | 25 (26) | 1 (10) |
| Chorioamnionitis | 32 (34) | 1 (10) |
| Placental abruption | 16 (17) | 1 (10) |
| Postnatal steroids, n (%) | 73 (77) | 0 |
| Surfactant, 1 or more doses, n (%) | 83 (87) | 0 |
| B | ||||
|---|---|---|---|---|
| Clinical course | Grade 1 BPD n=19 |
Grade 2 BPD n=32 |
Grade 3 BPD n=44 |
p value |
| Full enteral feeds, day of life, mean (SD)* | 25 (15) | 30 (26) | 27 (22) | 0.7 |
| Wean to RA, day of life, mean (range) | 102 (32 – 500) | 287 (97 – 500) | 315 (57 – 630) | <0.001 |
| Extubation, day of life, mean (range) | 13 (0 – 50) | 59 (7 – 171) | 89 (50 - 126) | <0.001 |
| Medical NEC, n (%) | 0 (0) | 5 (16) | 6 (14) | 0.7 |
| Surgical NEC/SIP, n (%) | 5 (26) | 5 (16) | 10 (23) | 0.6 |
| IVH grade 3-4, n (%) | 5 (26) | 9 (28) | 10 (23) | 0.9 |
| Blood culture negative infections, episodes per patient, mean (range) | 0.7 (0-2) | 1.2 (0-3) | 2.1 (0-5) | <0.001 |
| Blood culture positive infections, episodes per patient, mean (range) | 0.2 (0-1) | 0.5 (0-2) | 0.8 (0-2) | 0.004 |
Full enteral feeds were achieved upon stopping parenteral nutrition and dextrose-containing IVF excluding running at low rate (≤1 ml/hr) to maintain IV access.
SGA – small for gestational age defined as a birth weight of less than 10th percentile for gestational age, BPD – bronchopulmonary dysplasia
Infants with Grade 2 and 3 BPD were born at significantly earlier GA and had smaller BW and weight, length, and HC z-scores at birth than infants with Grade 1 BPD. Infants with Grade 2 and 3 BPD lost less weight during the first week of life (−0.43 and −0.49 for Grades 2 and 3 vs −0.79 z-score change for Grade 1, p=0.03) and had a higher velocity of weight gain during the first 4 weeks of life (13.4 and 14.1 for Grades 2 and 3 vs 11.2 g/kg/d for Grade 1, p=0.04). Despite the lower decline in weight z-scores and greater rate of weight gain over the first 4 weeks of life, infants with Grade 2 or 3 BPD consistently remained smaller than infants in the Grade 1 group due to the markedly smaller size at birth. In addition, they had consistently lower weight and HC z-scores up to term PMA when MRI was performed and were significantly shorter for PMA at 4 weeks of life, 36 weeks PMA, and time of MRI comparing with infants with Grade 1 BPD (Table 2).
Table 2.
Anthropometric characteristics of infants by different severity of BPD
| Characteristics | Grade 1 (n=19) |
Grade 2 (n=32) |
Grade 3 (n=44) |
P values |
|---|---|---|---|---|
| GA, weeks, mean (ranges) | 27.1(24.0 – 30.0) | 26.3 (23.7-30.4) | 25.5 (23.0-29.6) | 0.001 |
| Birth weight, g, mean (SD) | 1000 (270) | 770 (160) | 710 (250) | <0.0011 |
| Birth weight, z-score, mean (SD) | 0.22 (0.92) | −0.38 (0.84) | −0.40 (1.09) | 0.061 |
| Weight change, birth to 1 week, z-score, mean (SD) | −0.79 (0.39) | −0.43 (0.40) | −0.49 (0.57) | 0.031 |
| Weight at 4-week age, z-score, mean (SD) | −0.60 (0.67) | −0.88 (0.75) | −0.80 (0.73) | 0.42 |
| Weight change over 4 weeks, z-score, mean (SD) | −0.82 (0.42) | −0.50 (0.47) | −0.35 (0.73) | 0.021 |
| Weight at 36 weeks PMA, z-score, mean (SD) | −0.87 (1.07) | −1.40 (1.05) | −1.37 (0.96) | 0.16 |
| Weight at time MRI, z-score, mean (SD) | −0.65 (1.14) | −1.10 (1.17) | −1.29 (1.11) | 0.102 |
| Weight gain by 4 weeks of life, g/kg/day, mean (SD) | 11 (3) | 13 (4) | 14 (5) | 0.041 |
| Regain birth weight, day of life, mean (SD) | 11 (4) | 9 (5) | 9 (6) | 0.29 |
| Nadir weight loss, % from BW, mean (SD) | 10 (5) | 8 (5) | 9 (7) | 0.45 |
| Length at birth, z-score, mean (SD) | 0.28 (0.79) | −0.64 (1.08) | −0.70 (0.93) | 0.0021 |
| Length at 4 weeks age, z-score, mean (SD) | −0.94 (0.80) | −1.57 (0.86) | −1.60 (0.83) | 0.011 |
| Length at 36 weeks PMA, z-score, mean (SD) | −1.42 (1.04) | −2.07 (0.94) | −2.60 (1.12) | <0.001 |
| Length at time of MRI, z-score, mean (SD) | −1.20 (0.92) | −2.28 (1.22) | −2.80 (1.38) | <0.001 |
| Linear growth by 4 weeks of life, cm/week, mean (SD) | 0.6 (0.2) | 0.8 (0.4) | 0.7 (0.3) | 0.40 |
| HC at birth, z-score, mean (SD) | 0.16 (1.12) | −0.68 (0.89) | −0.76 (0.98) | 0.0071 |
| HC at 4 weeks of age, z-score, mean (SD) | −1.09 (0.87) | −1.40 (0.93) | −1.52 (0.85) | 0.21 |
| HC at 36 weeks PMA, z-score, mean (SD) | −1.16 (1.23) | −1.76 (1.03) | −1.69 (0.98) | 0.12 |
| HC at time MRI, z-score, mean (SD) | −1.00 (1.22) | −1.31 (1.06) | −1.64 (1.02) | 0.05 |
GA – gestational age, PMA - postmenstrual age, MRI - magnetic resonance imaging, HC – head circumference
one-way ANOVA test is significant; post hoc test showed significant difference between mild and moderate, and mild and severe disease but not significant between moderate and severe.
one-way ANOVA test is marginally significant; post hoc test showed significant difference between mild and severe disease but not significant between mild and moderate, and moderate and severe.
There was no significant difference in the daily fluid intake overall for the first 4 weeks of life between groups, but infants who developed Grade 2 and 3 BPD had less daily caloric and protein intake over the first 4 weeks of life than infants with Grade 1 BPD. When analyzed weekly, the lower caloric intake in neonates with Grade 2 and 3 BPD persisted through all 4 weeks reaching significance at third week and marginal significance at fourth week of life. There was no difference in protein intake during the first and fourth weeks of life, but neonates with more severe lung disease consistently received less protein in the second and third weeks of life. There was no significant difference in fluid, caloric or protein intake at any time when directly comparing infants with Grade 2 versus Grade 3 BPD (Table 3A).
Table 3.
(A) Nutritional intake during the first 4 weeks of life and (B) Spearman correlation for caloric and protein intake for first 4 weeks of life versus z-score for length at 4 weeks of life, 36 weeks PMA and at the time of MRI.
| A | ||||
|---|---|---|---|---|
| Characteristics* | Grade 1 BPD n=19 |
Grade 2 BPD n=32 |
Grade 3 BPD n=44 |
P values |
| Fluid, week 1, ml/kg/day | 145 (16) | 145 (19) | 146 (20) | 0.72 |
| Fluid, week 2, ml/kg/day | 162 (16) | 153 (12) | 154 (18) | 0.021 |
| Fluid, week 3, ml/kg/day | 157 (8) | 155 (15) | 152 (17) | 0.11 |
| Fluid, week 4, ml/kg/day | 152 (8) | 149 (11) | 152 (15) | 0.54 |
| Fluid, over 4 weeks of life, ml/kg/day | 154 (7) | 150 (10) | 151 (12) | 0.12 |
| Kcal/kg/d, week 1 | 74 (10) | 72 (10) | 71 (11) | 0.66 |
| Kcal/kg/d, week 2 | 109 (20) | 101 (17) | 101 (16) | 0.20 |
| Kcal/kg/d, week 3 | 121 (14) | 105 (15) | 104 (19) | 0.0041 |
| Kcal/kg/d, week 4 | 120 (13) | 113 (12) | 109 (19) | 0.072 |
| Kcal/kg/d, over 4 weeks | 106 (11) | 98 (10) | 96 (12) | 0.031 |
| Protein/kg/day, week 1 | 3.43 (0.41) | 3.45 (0.35) | 3.42 (0.42) | 0.98 |
| Protein/kg/day, week 2 | 4.23 (0.40) | 3.92 (0.48) | 3.98 (0.50) | 0.062 |
| Protein/kg/day, week 3 | 4.22 (0.59) | 3.72 (0.51) | 3.78 (0.57) | 0.021 |
| Protein/kg/day, week 4 | 4.06 (0.60) | 3.92 (0.43) | 3.97 (0.62) | 0.72 |
| Protein/kg/day, over 4 weeks | 3.99 (0.32) | 3.75 (0.28) | 3.79 (0.38) | 0.051 |
| B | z-score for length | ||
|---|---|---|---|
| 4 weeks of life | 36 weeks PMA | Time of MRI | |
| Caloric intake, first 4 weeks of life, kcal/kg/day | R=0.21, p=0.04 | R=0.25, p=0.02 | R=0.41, p<0.001 |
| Protein intake, first 4 weeks of life, g/kg/day | R=0.24, p=0.02 | R=0.12, p=0.25 | R=0.21, p=0.04 |
all data presented as mean (SD); week 1 = day of life 1-7, week 2 = day 8-14, week 3 = d 15-21, week 4 = day 22-28
one-way ANOVA test is significant; post hoc test showed significant difference between grades 1 and 2, and grades 1 and 3 disease but not significant between grades 2 and 3.
one-way ANOVA test is marginally significant; post hoc test showed significant difference between grades 1 and 3 disease but not significant between grades 1 and 2, and grades 2 and 3.
Lower early caloric intake was positively correlated with poorer linear growth at 4 weeks of life, 36 weeks PMA, and at term corrected age, and the correlation became stronger with time. Similar but less sustained correlation was observed between early protein intake and linear growth that reached significance at 4 weeks PMA and at term corrected age (Table 3B). Infants with lower caloric intake during the first 4 weeks of life required significantly longer mechanical ventilation (r=−0.28, p<0.05), oxygen supplementation (r=−0.33, p<0.05) and had more episodes of blood culture negative (r=−0.42, p<0.05) and blood culture positive infections (r=−0.24, p<0.05). Lower protein intake during the first 4 weeks of life significantly correlated only with the longer need for oxygen supplementation (r=−0.38, p<0.05).
Lung mass and lung volume per BSA increased from Grade 1 to Grade 2 and to Grade 3 BPD (237, 311, 384 g/m2 and 440, 535, 741 ml/m2 respectively, p<0.001)(Figure 1). Total lung density increased from control term/near term to Grade 1 and Grade 2 BPD but significantly dropped in Grade 3 BPD when compared with Grade 1 and Grade 2 groups (p=0.01) (Table 4). The percent volume of lung tissue with normal density significantly declined as the level of respiratory support required at the time of MRI progressed (p<0.001)(Figure 1C). There was significant positive correlation between lung mass per BSA and duration of mechanical ventilation (r=0.33, p=0.001), need for oxygen supplementation (r=0.35; p<0.001), and number of blood culture negative infections (r=0.43, p<0.001);between lung volume per BSA and duration of mechanical ventilation (r=0.3, p=0.003) and number of blood culture negative infections (r=0.44, p<0.001)(data not shown).
Figure 1.
Lung mass (A) and volume (B) calculated per body surface area for infants with and without BPD. Lung density distribution (C) in controls and infants with BPD categorized by level of respiratory support required at the time of MRI. Middle column represents normal density (i.e. within one standard deviation of the median density for normal infants) while left and right column are greater than 1 standard deviation less dense (left column) or more dense (right column). Percent normal density lung volume across all groups compared by ANOVA one way analysis and found to be significant between all groups (p<0.001)
Table 4.
MRI related patient’s characteristics
| MRI data | Control n=10 |
Grade 1 BPD n=19 |
Grade 2 BPD n=32 |
Grade 3 BPD n=44 |
P values |
|---|---|---|---|---|---|
| Respiratory support at time of MRI, n (%)* | |||||
| Room air | 10 (100) | 13 (68) | 0 | 0 | |
| Nasal canula | 0 | 6 (32) | 8 (25) | 3 (7) | |
| HFNC | 0 | 0 | 9 (28) | 2 (5) | - |
| CPAP | 0 | 0 | 7 (22) | 2 (5) | |
| NIPPV | 0 | 0 | 3 (9) | 3 (5) | |
| Mechanical ventilation | 0 | 0 | 5 (16) | 34 (77) | |
| PMA at time of MRI, weeks, mean (range) | 40.0 (37.0-42.6) | 38.5 (35.3-42.7) | 40.9 (35.6-49.3) | 41.1 (36.1-47.0) | 0.011 |
| BPD score, mean (SD)** | - | 2.9 (2.5) | 6.4 (3.7) | 8.9 (3.2) | <0.0012 |
| Total lung volume, ml, median (SD) | 102 (23) | 85 (18) | 107 (34) | 145 (53) | <0.0013 |
| Lung density, whole, g/ml, mean (SD) | 0.492 (0.133) | 0.542 (0.153) | 0.584 (0.151) | 0.525 (0.184) | 0.014 |
| Total lung mass, whole, g, mean (SD) | 50 (12) | 46 (10) | 62 (18) | 75 (27) | <0.0015 |
| Total lung volume/BSA, mL/m2, mean (SD) | 495 (111) | 440 (63) | 535 (137) | 741 (258) | <0.0013 |
| Total lung mass/BSA, g/m2, mean (SD) | 243 (52) | 237 (33) | 311 (78) | 384 (127) | <0.0015 |
MRI - magnetic resonance imaging, HFNC – high flow nasal canula, CPAP – continuous positive airway pressure, NIPPV – non-invasive positive pressure ventilation, PMA - postmenstrual age, BSA – body surface area
All infants were diagnosed with BPD at 36 weeks PMA based on 2019 Jensen criteria. Some infants with Grade 3 BPD were extubated while some infants with Grade 2 BPD were intubated between 36 week PMA and the time of MRI.
BPD score is a measure of BPD lung disease severity based on MRI characteristics as previously described17,26; the range of total score is 0 (no BPD) through 14 (Grade 3 BPD).
one-way ANOVA test is significant; post hoc test was significant between all groups except not significant between control and Grade 2, control and Grade 3, and Grade 2 and Grade 3 BPD.
one-way ANOVA test is significant; post hoc test was significant between all groups.
one-way ANOVA test is significant; post hoc test was significant between all groups except not significant between control and Grade 1 and control and Grade 2 BPD.
one-way ANOVA test is significant; post hoc test was significant between all groups except not significant between Grade 1 and Grade 3, and control and Grade 3 BPD.
one-way ANOVA test is significant; post hoc test was significant between all groups except not significant between control and Grade 1 BPD.
There was a significant correlation between reduced early caloric intake and greater MRI-based total lung mass per BSA and a worse (i.e. higher) MRI-based BPD score (r=−0.23, p=0.02 and r=−0.21, p=0.04, respectively) (Figure 2).
Figure 2.
Correlation of caloric intake for the first 4 weeks of life with lung mass per body surface area (A, R=−0.23, p=0.02) and MRI based BPD severity score (B, R=−0.21, p=0.04). Data were compared using Spearman correlations.
As per unadjusted linear regression analysis, caloric intake during the first four weeks of life negatively affected the lung mass (95% CI −2.17 [−4.16, −0.18]). Week 4 caloric intake had a significant unadjusted impact on lung volume (−3.05 [−5.82, −0.28]). The caloric intake during first four weeks of life significantly impacted all growth outcomes at term PMA: weight (0.03[0.01, 0.05]), length (0.05 [0.03,0.07]), HC (0.03 [0.01, 0.05]). Early protein intake had no significant impact on total lung mass or BPD score (r=−0.11 and r=−0.11, p=0.28)., weight gain or HC but only on z-score for length at term PMA (0.85 [0.03, 1.67]). However, after adjusting for the selected covariates, these significant associations did not persist.
Discussion:
In this retrospective observational clinical study of the association of early nutrition with the severity of BPD and lung development in preterm neonates using UTE MRI, we found that lower caloric was significantly associated with clinically more severe BPD, larger lung mass and volume, and a decreased proportion of normal density lung parenchyma.
Compared with infants with Grade 1 BPD, infants with Grade 2 and Grade 3 BPD were born at earlier GA with significantly smaller BW, length, and HC. Our cohort was consistent with other studies that show the strong association of more severe lung disease with prematurity and fetal growth restriction reflecting chronic impaired prenatal nutrition that likely negatively impacts distal lung and airway growth and development as seen in animal models [4, 6, 8, 12, 18-23]. Similar to other studies, in our cohort, 20 out of 21 neonates less than 26 weeks GA with BW below the 25th percentile developed Grade 2 or Grade 3 BPD [4]. In addition, infants with Grade 2 and Grade 3 BPD, despite significantly higher average daily weight gain than infants with Grade 1 BPD during the first four weeks of life, stay behind in their weight and head circumference percentiles with strikingly delayed linear growth that progressively worsens up to term age [4, 8].
Although infants experienced variations in caloric and protein intake, we observed no significant difference in fluid intake for the cumulative first four weeks of life between infants with different severity of BPD. In our cohort, the average daily fluid intake varied between groups in tight ranges of 150 and 154 ml/kg with a standard deviation of 7-12 ml/kg. Despite no significant difference in fluid intake, infants with Grade 2 and 3 BPD received significantly lower calories and protein during the cumulative first four weeks of life. The difference in average daily caloric and protein intake during this period between infants with Grade 1 and Grade 3 BPD in our cohort comprised 10 kcal/kg and 0.2 g/kg, respectively. This is comparable to a previous study that found every additional 10 kcal/kg/day of energy and 0.5 g/kg/day of protein was associated with a decreased risk of BPD by 9% and 25%, correspondently [24].
Although other studies also found data linking decreased early nutrition and poor growth with worse respiratory outcome in preterm infants, our study also analyzed the effect of early nutrition on lung development using UTE MRI to measure lung mass, volume, and density, and linking these findings with the severity of BPD, the nutritional data, and clinical course from birth to term PMA [4, 6, 8, 18-21]. Consistent with our initial evaluation of lung volume and mass in BPD, we found by UTE MRI that absolute and normalized lung volume and mass at or around term PMA increase as the severity of BPD progresses [13]. Initial hypotheses thought that larger lung volumes were due to hyperinflation caused by positive pressure ventilation and distorted lung architecture with cysts, emphysema, and air trapping; however, this mechanism should lead to hyperinflation with lower lung density and mass. In our study, lung density increased from control to Grade 1 and Grade 2 BPD. This increase in density along with increased lung volume resulted in increased lung mass. Interestingly, in infants with Grade 3 BPD lung density declined but lung mass was still substantially higher in these infants when compared with control and Grade 1 and Grade 2 BPD.
We observed that increased lung mass with reduced proportion of normal density parenchyma in infants with Grade 2 and 3 BPD was associated with lower early caloric intake and less favorable clinical course including longer duration of mechanical ventilation and oxygen supplementation and higher rate of blood culture negative infections, all of which may inhibit infant growth. We acknowledge the potential contradiction that our data suggests when infants with more severe BPD had less caloric and protein intake, but a higher rate of weight gain while still remaining smaller. As shown in animal models, malnutrition during the prenatal and early postnatal period leads to abnormal distal lung and airway formation causing disrupted alveolarization, altered pulmonary vasculature growth, surfactant deficiency, excessive extracellular matrix, abnormal epithelial cell division, reduced amount of cartilage with abundant mucosal folding in trachea and larger bronchi [3, 12, 25].
Our study is not without limitations. The retrospective design precludes finding causative relationships but rather associations and has a potential for errors in charting and inaccurate data collection. We had a high portion of patients with Grade 3 BPD in our study population (46%). Lung growth in BPD is a complex process and our measurements of lung volume, mass and density certainly do not represent all the components of lung growth and measurements such as volume and density may change in a given patient. That limitation notwithstanding, our study provides initial insight on lung growth in infants with BPD that was previously unavailable. Repeated measures of lung structural components and less variable timing for MRI study would provide more insight and should be considered for future studies. We gathered our nutrition data over a four-week period, which may limit the validity of week-to-week comparisons, however the week to week data was included for completeness. Lastly, BPD is a multifactorial disease with many offending factors that may have confounded the nutrition, growth and MRI results of our study (unknown prenatal risk factors, postnatal infections, proportion of donor vs maternal breastmilk atelectasis, emphysema, edema, etc). Indeed, we observed associations with important clinical factors such as number of infections therefore, it is difficult to discern between an association versus a causative relationship. We attempted an adjusted analysis, but none of our significant findings persisted likely due to the small sample size and the lack of statistical power to test the adjusted impact of the caloric or protein intake on BPD and lung growth.
Conclusion:
Optimization of early postnatal nutrition might be a powerful tool to improve lung growth and function, ameliorate the BPD severity, and achieve full growth potential with a less complicated clinical course. Pulmonary MRI allows us to directly assess components of lung growth and development and link these data with early nutrition.
Funding Sources:
This work was supported by National Institutes of Health R01 HL146689 and a grant from the Academic Research Committee at Cincinnati Children’s Hospital. Nara Higano was supported by National Institutes of Health T32HL007752.
Abbreviations:
- BPD
bronchopulmonary dysplasia
- UTE
ultrashort echo time
- MRI
magnetic resonance imaging
- PMA
postmenstrual age
- GA
gestational age
- BW
birth weight
- SD
standard deviation
- HC
head circumference
- BSA
body surface area
Footnotes
Statement of Ethics: This study was reviewed and approved by the Institutional Review Board of Cincinnati Children’s Hospital Medical Center, Cincinnati Ohio, approval IRB 2013-6101 and IRB 2018-0958. Written informed consent was obtained from the parent/legal guardian of participants prior to the study MRI. All work complied with accepted standards for research practice and reporting.
Conflict of Interest Disclosures: All authors have no conflicts of interest to disclose.
Data Availability Statement:
All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.
References
- 1.Jobe AH, Bancalari E. An All-Inclusive Perspective on Bronchopulmonary Dysplasia. J Pediatr. 2021. Jul;234:257–59. [DOI] [PubMed] [Google Scholar]
- 2.Ofek Shlomai N, Reichman B, Zaslavsky-Paltiel I, Lerner-Geva L, Eventov-Friedman S, Israel Neonatal N. Neonatal morbidities and postnatal growth failure in very low birth weight, very preterm infants. Acta Paediatr. 2022. Aug;111(8):1536–45. [DOI] [PubMed] [Google Scholar]
- 3.Arigliani M, Spinelli AM, Liguoro I, Cogo P. Nutrition and Lung Growth. Nutrients. 2018. Jul 18;10(7). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Nyp MF, Taylor JB, Norberg M, Truog WE. Impaired growth at birth and bronchopulmonary dysplasia classification: beyond small for gestational age. Am J Perinatol. 2015. Jan;32(1):75–82. [DOI] [PubMed] [Google Scholar]
- 5.Lima PAT, Meio M, Moreira MEL, de Abranches AD, Milanesi BG, Gomes Junior SCS. Energy expenditure and body composition in infants with bronchopulmonary dysplasia at term age. Eur J Pediatr. 2022. Aug;181(8):3039–47. [DOI] [PubMed] [Google Scholar]
- 6.Malikiwi AI, Lee YM, Davies-Tuck M, Wong FY. Postnatal nutritional deficit is an independent predictor of bronchopulmonary dysplasia among extremely premature infants born at or less than 28weeks gestation. Early Hum Dev. 2019. Apr;131:29–35. [DOI] [PubMed] [Google Scholar]
- 7.Huysman WA, de Ridder M, de Bruin NC, van Helmond G, Terpstra N, Van Goudoever JB, et al. Growth and body composition in preterm infants with bronchopulmonary dysplasia. Arch Dis Child Fetal Neonatal Ed. 2003. Jan;88(1):F46–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Milanesi BG, Lima PA, Villela LD, Martins AS, Gomes-Junior SCS, Moreira MEL, et al. Assessment of early nutritional intake in preterm infants with bronchopulmonary dysplasia: a cohort study. Eur J Pediatr. 2021. May;180(5):1423–30. [DOI] [PubMed] [Google Scholar]
- 9.Mayor RS, Finch KE, Zehr J, Morselli E, Neinast MD, Frank AP, et al. Maternal high-fat diet is associated with impaired fetal lung development. Am J Physiol Lung Cell Mol Physiol. 2015. Aug 15;309(4):L360–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Lock MC, McGillick EV, Orgeig S, McMillen IC, Muhlhausler BS, Zhang S, et al. Differential effects of late gestation maternal overnutrition on the regulation of surfactant maturation in fetal and postnatal life. J Physiol. 2017. Nov 1;595(21):6635–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kuiper-Makris C, Selle J, Nusken E, Dotsch J, Alejandre Alcazar MA. Perinatal Nutritional and Metabolic Pathways: Early Origins of Chronic Lung Diseases. Front Med (Lausanne). 2021;8:667315. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Wignarajah D, Cock ML, Pinkerton KE, Harding R. Influence of intrauterine growth restriction on airway development in fetal and postnatal sheep. Pediatr Res. 2002. Jun;51(6):681–8. [DOI] [PubMed] [Google Scholar]
- 13.Giltmier AJ, Higano NS, Woods JC, Kingma PS. Evaluation of regional lung mass and growth in neonates with bronchopulmonary dysplasia using ultrashort echo time magnetic resonance imaging. Pediatr Pulmonol. 2023. Oct 3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Jensen EA, Dysart K, Gantz MG, McDonald S, Bamat NA, Keszler M, et al. The Diagnosis of Bronchopulmonary Dysplasia in Very Preterm Infants. An Evidence-based Approach. Am J Respir Crit Care Med. 2019. Sep 15;200(6):751–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Hahn AD, Higano NS, Walkup LL, Thomen RP, Cao X, Merhar SL, et al. Pulmonary MRI of neonates in the intensive care unit using 3D ultrashort echo time and a small footprint MRI system. J Magn Reson Imaging. 2017. Feb;45(2):463–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Patel AL, Engstrom JL, Meier PP, Kimura RE. Accuracy of methods for calculating postnatal growth velocity for extremely low birth weight infants. Pediatrics. 2005. Dec;116(6):1466–73. [DOI] [PubMed] [Google Scholar]
- 17.Higano NS, Spielberg DR, Fleck RJ, Schapiro AH, Walkup LL, Hahn AD, et al. Neonatal Pulmonary Magnetic Resonance Imaging of Bronchopulmonary Dysplasia Predicts Short-Term Clinical Outcomes. Am J Respir Crit Care Med. 2018. Nov 15;198(10):1302–11. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Bose C, Van Marter LJ, Laughon M, O'Shea TM, Allred EN, Karna P, et al. Fetal growth restriction and chronic lung disease among infants born before the 28th week of gestation. Pediatrics. 2009. Sep;124(3):e450–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Uberos J, Jimenez-Montilla S, Molina-Oya M, Garcia-Serrano JL. Early energy restriction in premature infants and bronchopulmonary dysplasia: a cohort study. Br J Nutr. 2020. May 14;123(9):1024–31. [DOI] [PubMed] [Google Scholar]
- 20.Al-Jebawi Y, Agarwal N, Groh Wargo S, Shekhawat P, Mhanna MJ. Low caloric intake and high fluid intake during the first week of life are associated with the severity of bronchopulmonary dysplasia in extremely low birth weight infants. J Neonatal Perinatal Med. 2020;13(2):207–14. [DOI] [PubMed] [Google Scholar]
- 21.Rocha G, Guimaraes H, Pereira-da-Silva L. The Role of Nutrition in the Prevention and Management of Bronchopulmonary Dysplasia: A Literature Review and Clinical Approach. Int J Environ Res Public Health. 2021. Jun 9;18(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Maritz GS, Cock ML, Louey S, Suzuki K, Harding R. Fetal growth restriction has long-term effects on postnatal lung structure in sheep. Pediatr Res. 2004. Feb;55(2):287–95. [DOI] [PubMed] [Google Scholar]
- 23.Orgeig S, Crittenden TA, Marchant C, McMillen IC, Morrison JL. Intrauterine growth restriction delays surfactant protein maturation in the sheep fetus. Am J Physiol Lung Cell Mol Physiol. 2010. Apr;298(4):L575–83. [DOI] [PubMed] [Google Scholar]
- 24.Klevebro S, Westin V, Stoltz Sjostrom E, Norman M, Domellof M, Edstedt Bonamy AK, et al. Early energy and protein intakes and associations with growth, BPD, and ROP in extremely preterm infants. Clin Nutr. 2019. Jun;38(3):1289–95. [DOI] [PubMed] [Google Scholar]
- 25.Gaultier C. Malnutrition and lung growth. Pediatr Pulmonol. 1991;10(4):278–86. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data generated or analyzed during this study are included in this article. Further enquiries can be directed to the corresponding author.