Patient population
This center-based study was carried out at the NICU of Careggi University Hospital in Florence, Italy. Infants were included in the study if they were born at 23+0–31+6 weeks of gestation. Exclusion criteria were major congenital malformations, chromosomal disorders, inherited metabolic diseases, and death before reaching full enteral feeding (150 mL/kg).
Study design
The new local protocol for GRs monitoring was introduced in October 2020 and routine evaluations of GRs before every feeding were discontinued. Patients undergoing selective GRs monitoring were prospectively studied from December 2020 to June 2021, while patients in the routine GRs monitoring group were retrospectively studied from December 2019 to June 2020.
Evaluation of GRs in the selective group was performed in case of established (positive blood culture) or suspected sepsis (pathological values of C-reactive protein (CRP) and consistent clinical signs and symptoms), NEC at stage 1 or higher [15], established or suspected gastrointestinal occlusion, ascites, and established feeding intolerance. The latter was determined by abdomen pathological physical examination, regurgitations/vomits, GRs with volume ≥ 100% of previous feed and/or altered (i.e.: hematic, fecaloid), mucous or bloody stools, and onset of apnea/bradycardia [16]. The severity of these conditions was graded as minor and major criteria of feeding intolerance and were used for deciding the interruption of enteral feeding: the occurrence of one to two minor criteria suggested no change in the volume of feed, while the occurrence of three minor criteria or one major criterion suggested discontinuation of enteral feeding [16]. (Table S1, supplemental materials) Evaluation of GRs in the routine group was performed before every feeding and the decisions to increase, reduce, or halt feeding were made based on the same criteria 17.
All infants followed the same enteral nutrition protocol: trophic feeding was initiated within 24 h after birth and continued at 20–40 mL/kg/d as tolerated for up to five days. Subsequently, the amount was increased by 20 ml/kg per day if enteral nutrition was tolerated. Enteral nutrition goals were 150 mL/kg/d and 120 kcal/kg/d. All patients received their mother’s or donor’s human milk administered through an oro-gastric tube as bolus or continuous milk feeding. Human milk was enriched with a fortifier (Prenidina FM85®, Nestlè, La Tour-de-Peilz, France; 1 g/25 mL of milk) when enteral feeding of 100 ml/kg/d was reached. Preterm formula was administered when human milk was not available. Parenteral nutrition was continued until feeds of 100 mL/kg/d were reached.
Study outcomes
The primary outcome of this study was age at full enteral feeding (150 mL/kg/d). Prespecified secondary outcomes included age at full oral (breast and bottle) enteral feeding and at birth weight recovery, duration of parenteral nutrition, stay in NICU and in hospital, and evaluation of Z-scores of weight, length, and head circumference at discharge.
Data collection
Clinical and demographic data were collected by reviewing patients’ electronic medical records. The following data were recorded for each studied infant: gestational age, birth weight, birth weight < 10th percentile, antenatal steroids, mode of delivery, need and duration of non-invasive and invasive ventilation, surfactant and postnatal steroid treatment, occurrence of patent ductus arteriosus (PDA) requiring treatment, bronchopulmonary dysplasia (BPD), necrotizing enterocolitis (NEC), sepsis, intraventricular hemorrhage (IVH), retinopathy of prematurity (ROP), mortality, duration of stay in NICU and in hospital, and occurrence of breastmilk (exclusive or mixed) feeding at discharge. BPD was defined as oxygen requirement at 36 weeks of post-menstrual age. PDA was diagnosed by echocardiography; NEC was defined as Bell’s stage ≥ 2 [15] sepsis was defined as positive blood culture. IVH was classified according to the Papile classification scheme [17]; ROP was graded according to the international classification of retinopathy of prematurity [18]. The examined maternal variables included clinical chorioamnionitis, maternal diabetes, hypertensive disorders of pregnancy, and prolonged premature rupture of membranes (pPROM) > 18 h. Z-scores of weight, length, and head circumference at discharge were calculated using Anthro version 3.2.2 (World Health Organization, Geneva, Switzerland) [19]. Growth Z-scores are normed for age and sex to provide a more meaningful description of postnatal growth than raw values.
Statistical analysis
Clinical characteristics of infants in the selective and routine GRs monitoring groups were reported as mean and standard deviation, rate and percentage, or median and range. Statistical analysis was performed using the Student “t” test for parametric continuous variables, the two sample Wilcoxon rank-sum test for non-parametric continuous variables, and the Fisher test for categorical variables. A p < 0.05 was considered statistically significant.
We planned to perform logistic regression analyses to assess the possible correlation between the primary outcome and variables that at univariate analysis were different between the groups (p < 0.100) excluding those which were found to be collinear by calculating variance inflation factors (VIF). However, there were not significant differences between the no residual and residual groups and logistic regression analysis was not required. Post-hoc analysis demonstrated that our study had 73% power to detect as statistically significant the observed differences of age at full enteral feeding between the groups.