GLP-2 levels in infants with intestinal dysfunction.

Glucagon Like Peptide 2 (GLP-2) has been proposed as an important regulatory hormone in nutrient absorption. The present study was conducted in human infants with intestinal dysfunction undergoing surgery, correlating postprandial GLP-2 levels with intestinal length, nutrient absorption, and patient outcome. We hypothesized that GLP-2 levels would be inversely related to nutrient absorption; we further hypothesized that post prandial GLP-2 levels would be predictive of the ability to wean patients from total parenteral nutrition (TPN), and tolerance of enteral feeding. Infants prospectively identified with nutrient malabsorption following intestinal surgery were monitored and after initiation of feeds GLP-2 levels were measured in the fed state. Intestinal length was recorded intraoperatively and nutrient absorption was quantified using both a balance study, and carbohydrate probe method. 12 infants had GLP-2 levels successfully measured; two patients had repeated studies. Average gestational age was 32.7 +/- 3.4 wk, age at testing was 1.7 +/- 1.4 mo and average weight was 3.5 +/- 1.1 kg. Causes of intestinal loss were necrotizing enterocolitis, atresia and volvulus. Five patients had severe short bowel syndrome (<50% of normal small intestinal length), 3 died. GLP-2 levels were best correlated with residual small intestinal length (r2 = 0.75). Correlations with total intestinal length including colon were less significant; residual colon appeared to not contribute to measurable GLP-2 production. GLP-2 levels were well correlated with tolerance of enteral feeds. Contradicting the initial hypothesis, GLP-2 levels were directly correlated with nutrient absorptive capacity (correlation with fat absorption: r2 = 0.72, carbohydrate = 0.50 and protein = 0.54 respectively). There were no apparent changes in GLP-2 levels with gestational or postnatal age. As a corollary to the correlation with bowel length, a postprandial level of 15 pmol/L appeared to be discriminatory; infants with postprandial GLP-2 levels of > 15 pmol/L were able to be weaned from total parenteral nutrition, while 3 of 4 infants who had GLP-2 levels less than 15 could not be weaned by one year. These results show that in infants with intestinal dysfunction, GLP-2 levels are correlated with residual small bowel length and nutrient absorption, and may be predictive of outcome. In contrast to adults with intact colon and SBS, infants with SBS and intact colon do not appear able to produce GLP-2 in response to feeding stimulation. Further studies are suggested to examine the ontogeny of the GLP-2 axis and the possible therapeutic role of GLP-2 supplementation.