Does Human Milk Fortifier Affect Intestinal Inflammation in Preterm Infants?

Introduction

Intestinal inflammation in preterm infants is concerning, as it may be an early marker of disease, such as necrotizing enterocolitis (NEC). ¹ The site of inflammation characteristically is marked by neutrophil migration and sequestration.^1,2 Intestinal neutrophil action often is correlated with the presence of calprotectin in the feces. ³ A common neonatal intensive care practice for preterm infants is the feeding of human milk and to optimize nutritional intake with the addition of human milk fortifier. The feeding of human milk containing intact bovine protein as the bovine-based human milk fortifier has been associated with greater fecal calprotectin levels, a marker of intestinal inflammation, compared with feeding of unfortified human milk. ⁴ Newer human milk fortifiers containing protein hydrolysates instead of intact protein now are available. We investigated fecal calprotectin levels in human milk-fed preterm infants before and after the addition of human milk fortifier containing protein hydrolysates in place of intact protein.

Methods

Preterm infants born ≤32 weeks gestation were enrolled in their first week if their birth weight was ≤1,500 g, were free of major congenital anomalies, and were receiving enteral feedings of human milk (mother's own milk or donor human milk). The feeding protocol was the same for all infants, which included trophic feeding, slow advancement, and the addition of human milk fortifier (Similac Human Milk Fortifier, Abbott Laboratories, Columbus, OH) when infants reached a milk intake of 80 mL/kg per day. Subsequently, feedings were advanced as tolerated to ∼160 mL/kg per day. Fecal samples were collected serially from day of enrollment to day 60. These samples were stored at 2–8°C for 1–5 days and subsequently transferred to −80°C storage until analyzed in duplicate for calprotectin (Calprotectin ELISA Assay Kit; Eagle Bioscience, Nashua, NH). The sensitivity of assay was 0.9 μg/g. The following data were recorded: gestational age, birth weight, type of feeding, postnatal age when human milk fortifier was added, postnatal morbidities such as sepsis, patent ductus arteriosus (PDA), PDA ligation, NEC, antibiotic and other medication use, and prenatal maternal characteristics, including the presence of preterm labor and/or chorioamnionitis, and the receipt of antenatal antibiotics and/or steroids.

SPSS (Statistical Package for the Social Sciences) version 21.0 was used for data analysis. Calprotectin levels 2–5 days before and 2–5 days after human milk fortification were analyzed by paired t test. Pearson correlation and regression analyses were used to compare associations between calprotectin and study variables. Statistical significance was set at p < 0.05. The study was approved by the Institutional Review Board of Northwell Health. Written informed consent was obtained from parents.

Results

A total of 401 stool samples were collected from 24 human milk-fed infants (Table 1) who had mean (±standard deviation) birth weight of 993 ± 277 g (range 440–1,400 g), gestational age of 27.5 ± 2.8 weeks (range 23–31 weeks), and initiation of human milk fortification at 14 ± 9 days postnatally. Median fecal calprotectin levels before and after fortification were similar, 7 μg/g (interquartile range [IQR] 4; 18 μg/g) and 8.5 μg/g (IQR 2; 20 μg/g), respectively, p = 0.36.

Calprotectin levels were not correlated with birth weight, gestational age, neonatal morbidities, or maternal factors of preterm labor, chorioamnionitis, and antenatal steroids, but declined with postnatal age (p = 0.005), and receipt of human milk fortifier (p < 0.001) and antibiotic, given either antenatally (p = 0.007) or postnatally (p = 0.008) (Table 1). When adjusted for postnatal age, calprotectin levels declined with the receipt of human milk fortifier, p < 0.001 (Fig. 1). Independently, when adjusted for postnatal age and receipt of human milk fortifier, calprotectin levels declined with the receipt of antibiotics, p < 0.001.

OPEN IN VIEWER

FIG. 1.

Relationships among fecal calprotectin, human milk fortification, and postnatal age in preterm infants. Significantly lower calprotectin in preterm infants receiving human milk fortifier, when adjusted for postnatal age, p < 0.001.

Discussion

We found no differences in fecal calprotectin levels before and after the addition of human milk fortifier suggesting that the fortifier is not associated with intestinal inflammation in preterm infants. Importantly, these observations occur in a setting where the human milk fortifier contained a hydrolyzed bovine-based protein source compared with previous formulations using intact bovine-based protein. This association has not been reported previously. The protein hydrolysate-containing human milk fortifier most likely did not affect intestinal inflammation as has been shown with intact protein-containing human milk fortifier. ⁴

Calprotectin has been shown to decline in feces with dietary changes favoring the use of protein hydrolysate-containing bovine-based infant formula. ⁴ Similarities in calprotectin levels have been reported in preterm infants fed either protein hydrolysate-containing formulas or unfortified human milk. ⁵ Moreover, it is reported that elevated calprotectin levels decline in infants with cow milk allergic colitis after treatment with hydrolyzed protein-containing formula. ⁶ Thus, the observations in this study are pertinent and highlight the beneficial nature of a protein hydrolysate-containing human milk fortifier for human milk-fed preterm infants.

Calprotectin is resistant to degradation by fecal bacterial and is not specifically temperature sensitive, so it is a useful marker. ⁷ Since calprotectin levels correlate with neutrophil migration to sites of inflammation in the intestine it has been a used as a marker of disease in children with inflammatory bowel diseases.^8,9

We observed that fecal calprotectin levels independently were low in the presence of antibiotics, but both fortifier and antibiotics had independent associations when adjusted for postnatal age. Antibiotics presumably delay bacterial colonization and its associated localized inflammation. ¹⁰ Fecal calprotectin levels decline with postnatal age. ¹¹ Greater calprotectin levels in early stool samples may be explained by the presence of meconium, which is associated with higher levels of calprotectin than later stools.^12,13 Furthermore, elevated early fecal calprotectin levels may be a marker of neutrophil migration that results from intestinal immaturity and increased intestinal permeability in early postnatal life.^12,13

This small observational study reported comparatively low but consistent levels of calprotectin in the study preterm infants. The low levels may be due to the study conditions of an exclusive human milk diet with fortifier containing protein hydrolysates as bovine-based protein source.^14,15 Thus, for preterm infants fortifiers containing hydrolyzed protein appear to benefit in contrast to those fortifiers with intact bovine-based intact protein.