Virtual elimination of necrotising enterocolitis for 5 years – reasons?

Abstract

A standardised feeding regimen was adopted in 1997 for guiding enteral feeding of neonates <32 weeks’ gestation during clinical trials (18 months each) involving erythromycin (n=73) as a prokinetic and carboxymethylcellulose (n=70) as a laxative as well as for during 2 years (n=155) without any trials. Most aspects of the feeding regimen (e.g., milk increments-total volume/day, use of breast milk by choice, etc) were not significantly different from current practices.
Results. 298 neonates <32 weeks’ gestation (<28 weeks; n=78) were enterally fed during the 5 years. Their demographic characteristics and median (interquartile) age in days at starting (AST) and days to reach full enteral feeds (FFT) of 150 ml/kg/day were not significantly different during these 5 years: [AST: 5 (3–7.5)], [FFT: 4 (3–7)] Only one case of definite NEC (⩾Stage II) occurred during the 5 years. The time to reach full feeds was also reduced by over 54% (including for neonates <28 weeks gestation) compared with a historical cohort.
Conclusion. Sustained reduction in the time to reach full feeds with virtual elimination of ⩾Stage II NEC for 5 years indicates continued benefits of a standardised feeding regimen as a simple preventive strategy to prevent NEC. Whether our specific policy of no enteral feeds in presence of hemodynamic instability associated with PDA requiring indomethacin, and/or sepsis played a role in achieving the significant results needs controlled trials.

Introduction

A standardised feeding regimen was adopted in 1997 for guiding enteral feeding of neonates <32 weeks’ gestation during clinical trials (18 months each) involving erythromycin as a prokinetic and carboxymethylcellulose as a laxative as well as for a period of 2 years without any trials [1], [2], [3]. The feeding regimen prospectively defined readiness for commencing, guidelines for upgrading, and indications for stopping/restarting enteral feeds (3). Most aspects of this regimen (e.g., daily maximum milk increments and total volume) were not significantly different from current practices except for the specific policy of not commencing/ceasing enteral feeds in presence of hemodynamic instability associated with patent ductus arteriosus (PDA) requiring indomethacin, and/or sepsis (3). A total of 298 neonates <32 weeks’ gestation (including 78 <28 week) were enterally fed during the 5 year period. Their demographic characteristics, median (interquartile) age at starting feeds and days to reach full enteral feeds of 150 ml/kg/day were not significantly different between the study and control groups during both the trials and overall remained comparable during the entire period of 5 years (Table 1, Table 2). Except for a single case definite NEC (⩾Stage II) was virtually eliminated from the nursery during these 5 years. This was in contrast to six deaths from NEC per year for 5 years prior to adopting the standardised feeding regimen (4). The time to reach full feeds was also reduced by over 54% (including for neonates <28 weeks gestation) during these 5 years compared with a historical cohort with comparable demographic characteristics (3). The significant results most probably reflect the continued benefits of a standardised feeding regimen (increased awareness leading to early detection and management of signs of “feed intolerance”) [3], [5]. However contribution by our specific policy in relation to PDA and/or sepsis cannot be ruled out given the currently accepted risk factors and pathogenesis of NEC. We review the complex interplay between immaturity, enteral feeds, intestinal bacterial colonisation and a hypoxia-compromised gut mucosa that may occur specifically in preterm neonates with hemodynamic instability due to PDA and/or sepsis. We also emphasise the need for clinical trials in this area in the light of current variations in feeding practices (6), our results over 5 years, and our feeding practices in presence of these high-risk factors.