The present study demonstrates an independent association between severe BPD and lower caloric intake and higher total fluid intake during the first two weeks of life. Cumulative total fluid intake was higher and mean total caloric intake lower in infants with severe BPD through week 12 compared to those without severe BPD. Therefore, it appears that early fluid and nutritional management in the first two weeks of life may predispose to the subsequent development of severe BPD. In contrast, nutritional intake through week 12 is more likely to reflect the effects of the development and presence of established chronic lung injury including efforts by clinicians to increase nutritional intakes, and the potential for increased energy expenditure resulting from increased work of breathing (11).
Many prior studies have focused on the potential for higher total fluid intake predisposing to the development of BPD. Oh et al. (12) demonstrated that less postnatal weight loss and higher fluid intake during the first ten days of life were significantly associated with a higher risk of developing BPD in a cohort of ELBW infants. A Cochrane review analyzing 5 RCTs of varying water intake in premature infants indicated restricted fluid intake was associated with increased postnatal weight loss, as well as reduced risk of patent ductus arteriosus and necrotizing enterocolitis. However, restricted fluid intake was only associated with a trend towards a reduced risk of BPD but this was not statistically significant (13). The present findings can be interpreted to suggest that early restriction of fluid intake is critical to attenuate the development of severe BPD because higher cumulative and mean total fluid intakes during the first two weeks of life were significantly associated with an increased risk of developing severe BPD.
More recent reports have addressed the influence of nutritional intakes on the development of BPD (6, 14, 15). Al-Jebawi, et al. recently reported that low caloric intakes in the first week of life were associated with an increased risk of developing moderate to severe BPD (OR 0.94, 0.91–0.97) (14). However, ELBW infants in this study were exclusively parenterally fed during the first week of life (14). In contrast, although our study is consistent with these findings, the infants in our study received a combination of parenteral and enteral nutrition. Furthermore, our findings examine the role of caloric deficit in the early management through week 2 as well as through 12 weeks PMA, demonstrating that both are associated with the development of severe BPD.
Malikiwi, et al. demonstrated that the odds of developing BPD was lower after receiving a higher four-week average daily caloric intake (OR = 0.89) in a retrospective case-control study consisting of 66 infants (6). Likewise, another cohort of 296 infants born less than 27 weeks of gestation exhibited reduced risk of developing BPD after receiving higher energy intakes from postnatal days seven to 27 (15). However, the definition of BPD utilized in the aforementioned studies described infants requiring supplemental oxygen at 36 weeks postmenstrual age, which would be consistent with a definition of moderate to severe BPD as defined by National Institutes of Health (NIH) consensus guidelines of 2001 (8).
The nutritional evaluation in our study examined a subgroup of infants with complex medical needs who developed severe BPD, in contrast to the previous reports (6, 14, 15). In addition, the current study represents a retrospective analysis of prospectively collected nutritional and growth profiles. There is a paucity of previous studies that have examined the contribution of nutritional profiles specifically to the development of severe BPD. The present study emphasizes the importance of early higher caloric intake in infants who are more likely to develop severe BPD. Moreover, higher caloric intake through week 12 reduced the odds of developing severe BPD by approximately four percent for each one kcal/kg. Nonetheless, the current study along with the previous findings (6, 14, 15) represent retrospective analysis of low-birth-weight infants that were destined to develop BPD at 36 weeks of age. In addition, we cannot rule out the possibility that factors known to contribute to the development of BPD including gestational age, birth weight, severity of lung disease, and the level of respiratory support, did not also influence the nutritional intake of the infants who were destined to develop severe BPD.
The infants with severe BPD also received a higher cumulative total protein intake through week 12. This is most likely because of the longer duration of exposure to parenteral nutrition in the infants who developed severe BPD compared with the control infants. The control infants also received more breast milk, which has a lower protein content than parenteral alimentation.
A Cochrane meta-analysis (16) from 2006 attempted to explore randomized trials that assessed the effects of increased energy intake on outcomes of preterm infants with or destined to develop chronic lung disease (CLD)/BPD, but eligible trials were not identified, thus emphasizing the need for future randomized controlled trials to prospectively examine the benefits of more optimal nutritional intakes in reducing the severity of BPD.
The present study also demonstrated an association between the development of BPD and reduced weight, length, and head circumference growth trajectories. It has long been recognized infants with BPD demonstrate poor postnatal growth (17). Vohr et al. (17) originally showed that infants with BPD were more likely to have poor weight gain with weight less than the third percentile at 4 and 12 months postnatal age compared with infants without BPD. The lower postnatal weight gain was suggested to be associated with poor nutritional intake and greater expenditure of energy (17). Subsequently, Ehrenkranz et al. (18) examined a cohort of ELBW infants divided into quartiles of in-hospital growth velocity rates. Fifty six percent of infants in the lowest growth quartile (mean weight gain of 12.0 g/kg/day) developed BPD, compared to 31% of infants in the highest growth quartile (mean weight gain of 21.2 g/kg/day), p < 0.001.
The results of the current study agree with the results in the prior studies that demonstrate poor postnatal weight gain in infants with severe BPD, but also emphasizes growth failure across anthropometric measures in these infants, including length and head circumference. The present study also accentuates the importance of the association between linear growth and the development of severe BPD. It is well documented that height is an important predictor of lung function both in children and adults (19). Moreover, the importance of length as it relates to lung growth in preterm infants is becoming increasingly recognized (20). Sanchez-Solis et al. (21) showed in a multivariate analysis that lung function in preterm infants with and without BPD was associated with gain in height velocity, but not specifically with increases in weight gain. These findings are consistent with the results of our logistic regression analysis, in which the anthropometric measure of strongest significance was the length trajectory. These findings suggest that future efforts to prevent BPD should include an emphasis on the evaluation of linear growth, as well as weight gain velocity.
We also observed that head circumference Z-score changes between birth and 36 weeks PMA (or discharge if sooner) were lower in infants with severe BPD. Head circumference is often utilized as an indirect measure of postnatal brain growth, and the literature has shown an association between suboptimal brain growth and subsequent neurodevelopmental impairment (22–24). Nesterenko, et al. demonstrated a correlation between oxygen exposure and delays in head growth at six months corrected age in infants with moderate to severe BPD with birth head circumference less than the 50th percentile (25). However, it remains to be determined whether the oxygen exposure is directly related to the pathogenesis of delayed head growth or is simply a marker of clinical illness severity. Additional studies are required to elucidate the significance of head growth as it relates to BPD and neurodevelopmental outcomes.
There are several limitations to the present study. Although the nutritional and fluid intakes and growth parameters were prospectively collected, a major limitation is the retrospective nature of analysis of the study. Furthermore, given the patient population in this study included only ELBW infants, it was difficult to identify infants without BPD. Consequently, the infants who did not develop severe BPD included some infants with an oxygen requirement at discharge, but who did not develop severe BPD as this was the population of interest for our study. Further, despite our best efforts to match the populations with and without severe BPD, the group with severe BPD had significantly lower weights and gestational ages at birth compared with those without severe BPD, though within our defined measures of closeness. A larger cohort may allow for more precise matching. Moreover, as this is a single center study, our findings may not be generalizable to other centers because the incidence of BPD varies widely among centers (26).
In conclusion, higher fluid and lower total caloric intakes and reductions in linear growth were independently associated with an increased risk of developing severe BPD in ELBW infants.