Somatostatin in preterm infants: postnatal changes and response to stress

Hypermagnesemia and feeding intolerance in preterm infants: a cohort study

BACKGROUND

Feeding intolerance (FI) is a common clinical problem in preterm infants often caused by some neonatal disorders and drugs, including antenatal exposure to magnesium sulfate (MgSO₄ ).

OBJECTIVE

To evaluate the association between hypermagnesemia at birth and FI in preterm infants during the first 72 hours of life.

METHOD

This was a cohort study conducted with preterm infants < 34 weeks of gestation. Infants presenting at least two of the following signs were considered as having FI: vomiting, abdominal distension, the need for continuous intermittent feeding, and delayed meconium passage. Hypermagnesemia was characterized by umbilical serum magnesium levels >2.5 mEq/l.

RESULTS

251 infants were evaluated. The median birth weight and gestational age were 1390 (IQR 1020 -1070) g and 31 (IQR 28-32) weeks, respectively. The FI rate was 17.5%. The exposure rate to MgSO₄ was similar in the tolerant and intolerant groups (53.1% x 63.6%, p=0.204), but hypermagnesemia was more frequent in the FI group (40.9% x 24.2%, p= 0.024). The univariate analysis showed that infants with hypermagnesemia were two-fold more likely to present FI (OR 2.16 95% CI 1.09-4.26). In the multiple logistic regression analysis, we found that hypermagnesemia was independently associated with FI (OR 2.51 - 95% CI 1.06-5.91), as well as maternal diabetes mellitus (OR 2.56 95% CI 1.07 - 6.14), SNAPPE-II (OR 1.051 95% CI 1.025-1.078), and brain hemorrhage (OR 3.61 95% CI 1.31 - 9.91).

CONCLUSION

In addition to other factors, hypermagnesemia at birth was independently associated with early feeding intolerance in preterm infants. This article is protected by copyright. All rights reserved.

Gastric emptying in premature newborns with acute respiratory distress

OBJECTIVES

The authors hypothesized that acute respiratory distress (ARD) delays gastric emptying. The objective was to test this hypothesis by assessing gastric emptying on the second and seventh days of life in premature infants with ARD resulting from pulmonary disease.

METHODS

Thirty-nine newborns with ARD starting on the first day of life were selected and paired with 39 healthy control newborns matched by weight (within 250 g). Gestational age was <or =35 weeks and birth weight was < or =1750 g for all subjects. Gastric emptying was assessed at 48.0 +/- 24.0 hours and at 168.0 +/- 24.0 hours of life. A test meal consisting of 3 mL/kg of 5% glucose in water labeled with phenol red was administered by gastric tube over 1 minute and gastric retention was determined as percent test meal remaining in the stomach 30 minutes after administration.

RESULTS

Gastric retention at 30 minutes varied considerably in both groups and was significantly higher (P < 0.01) in newborns with ARD (61.4%) than controls (51.8%) at 48.0 +/- 24.0 hours, decreasing significantly after partial or full remission of ARD at 168 +/- 24 hours of life. Gastric retention was 60.2% in newborns with feeding intolerance and 36.8% in tolerant newborns (P < 0.001) at 168 hours. ARD and periventricular or intraventricular hemorrhage were predictors of gastric retention at 48 +/- 24 hours of life, whereas feeding intolerance and gestational age were predictors of gastric retention at 168 +/- 24 hours. Gastric retention was inversely correlated with gestational age.

CONCLUSION

Gastric emptying is delayed in premature infants with ARD during the first 72 hours of life and may impair the initiation of enteral feeding.

Plasma somatostatin and cholecystokinin levels in sick preterm infants during their first six weeks of life

The present study reports the levels of plasma somatostatin and cholecystokinin in 19 preterm infants with asphyxia [n = 10, GA (median; range) 26; 23-30 weeks] and respiratory distress syndrome (n = 14, GA 27; 23-29 weeks) compared with preterm infants without any of these conditions (reference group, n = 59, GA 33; 25-36 weeks). In the reference group 37 infants received phototherapy and their peptide levels were compared with those not receiving phototherapy (n = 22). Plasma somatostatin and cholecystokinin were analysed by specific radioimmunoassays on day 1, day 3-4 and at 6 weeks of life. Plasma somatostatin levels, but not cholecystokinin levels, of reference infants were inversely related to gestational age on day 1 and day 3-4. Asphyxiated infants and infants with respiratory distress syndrome had significantly higher somatostatin levels than reference infants on day 1 and day 3-4. These differences disappeared when the levels were adjusted for gestational age. Plasma cholecystokinin levels were not influenced by respiratory distress syndrome and asphyxia. Phototherapy did not affect plasma somatostatin and cholecystokinin levels.

Plasma somatostatin and cholecystokinin levels in preterm infants during their first two years of life

Plasma somatostatin and cholecystokinin are two gut peptides with opposite functions which are regulated by two different parts of the autonomic nervous system. Previously we have shown that plasma somatostatin and cholecystokinin levels are higher in preterm infants during the 1st d of life than in adults or in their mothers, and that plasma somatostatin is negatively correlated to gestational age. We have longitudinally studied these two peptides in 28 preterm infants, 17 boys and 11 girls, up until the age of 2 y. The mean (SD) gestational age was 32.3 (2.8) wk, the mean birth weight was 1877 (515) g, and the mean birth length was 42.8 (3.8) cm. Blood samples were taken on the 1st d of life, at 6 wk, and at 3, 6, 12, and 24 mo of age. Plasma was analyzed by specific somatostatin and cholecystokinin RIAs. The median plasma somatostatin and cholecystokinin levels were lowest at 3 mo (somatostatin = 17.4/cholecystokinin = 10.5 pmol/L) and highest at 6 mo (somatostatin = 37.3/cholecystokinin = 27.1 pmol/L). At 24 mo plasma somatostatin remained at the same level, and cholecystokinin had decreased to half that level. After the 1st d of life plasma somatostatin and cholecystokinin levels were not correlated to gestational age or attained weight or length. The plasma somatostatin level at 3 mo of age was negatively correlated to the increment in knee-heel distance between 3 and 6 mo of age.