Fatores associados ao escore-z de peso em prematuros de muito baixo peso e extremo baixo peso ao nascer durante a internação hospitalar

Abstract

Introduction: Prematurity is a global public health problem, accounting for 9.9% of births in 20201. In Brazil, the rate is similar, with prematurity being the leading cause of mortality in children under five. Preterm newborns (PTNB) with very low birth weight (VLBW) and extremely low birth weight (ELBW) account for 15% of this rate and require more attention regarding their growth and related factors, especially in early postnatal life. Objective: To investigate the behaviour of the z-score (z-s) of weight and associated factors of PTNBs with VLBW and ELBW, comparing them with those with low birth wheigt (LBW), during a four-week hospitalisation in a neonatal intensive care unit (NICU). Methods: A longitudinal study using data from the NICU PTNB records of a university hospital in southern Brazil. Data were obtained from patients admitted within the first 48 hours of life, from January 2017 to December 2020. PTNB of both sexes, non-twin, with gestational age ≥24 and <37 weeks, with birth weight (BW) ≥500g, receiving parenteral/enteral nutrients were included. In total, records were obtained for 558 PTNB, of which 297 were excluded, 29 deaths were observed, and 261 were included in this study. The data of eligible preterm infants were categorised into five phases according to the number of days of hospitalisation and met the following criteria Week 1) 5 to 7 days; Week 2) 12 to 14 days; Week 3) 19 to 21 days and Week 4) 26 to 28 days. Of the 213 PTNB included, 125, 82 and 54 remained in the NICU at weeks 1, 2, 3 and 4 respectively. The outcome was z-s weight, the exposure was the BW of the PTNB, categorised as 1) VLBW and ELBW: < 1500 g and 2) LBW: ≥1500 g to 2500 g. Energy (kcal/kg/day) and protein (g/kg/day) intakes were assessed. Analyses were performed in JAMOVI version 2.5. Two-factor ANOVA and multiple linear regression were used to evaluate the association between demographic, clinical and nutritional factors and weight z-s behaviour. Results: Among PTNB, 60% were male, 65% weighed ≥1500g, 80% had respiratory distress syndrome, 41% and 27% had sepsis and jaundice respectively.There was a significant interaction between BW category and length of stay on z-s (F=4.0; p=0.003). In the VLBW and ELBW groups, weight z-s was significantly lower in the first week of hospitalisation compared to the LBW group [-1.05 (-1.34; -0.75) vs -0.34 (-0.49; -0.18), z-s]. Associated factors were male gender, sepsis, start of enteral nutrition (EN), protein intake (F=9.19; p<0.001). Conclusion: The downward behaviour of the weight z-s curve was associated with the interaction between BW and length of hospital stay in PTNBs with VLBW and ELBW. In addition, factors such as male sex and the presence of sepsis contributed to the observed decline. On the other hand, starting EN before 24 hours of hospitalisation and providing more than 2.0 g/kg/day of protein in the first days of life attenuated the decrease in z-s. However, these factors were not sufficient to limit the nutritional deficit in PTNB with VLBW and ELBW compared to those with LBW.