Shulman, Robert J.

Gastrointestinal (GI) function in the preterm infant is underdeveloped compared with that in the term infant. Consequently, preterm infants experience great difficulty in tolerating enteral feedings as evidenced by symptoms such as gastric residuals after feeding and abdominal distention. This feeding intolerance is associated with significant morbidity (complications) and in the extreme case of necrotizing enterocolitis, mortality (death). These clinical morbidities delay the progression of feedings and consequently, the time to full enteral feedings. This relationship is important, as we have observed a relationship between the time required to reach full enteral feedings and the duration of hospitalization. Hence, strategies that reduce feeding intolerance will hasten the progression to full enteral feedings and ultimately will reduce the duration of hospitalization. One of the key components to developing strategies to reduce feeding intolerance is to be able to identify which infants are most at risk. Presumably the risk is greater with increasing degrees of prematurity. However, at the present time, it is not even possible to predict among infants of the same gestational age which are most at risk. Studies are needed to define the clinical characteristics that reflect feeding intolerance and to develop clinical tests that can predict feeding intolerance. Such tests would allow early intervention and potentially prevent or ameliorate the most severe manifestation of feeding intolerance, necrotizing enterocolitis

We hypothesize that in preterm infants 25 to 32 weeks gestational age the following tests:

Measurement of:

a. Gastric emptying using the 13C-octanoic acid breath test; b. Gastrointestinal mucosal integrity with the sucrose/lactulose/mannitol/sucralose urinary excretion (permeability) test (PT); c. Lactase activity using the lactase activity test (part of the PT); and d. Fecal excretion of calprotectin and alpha-1-antitrypsin

1. Can detect the presence of feeding intolerance (defined by increased gastric residuals, abnormal clinical abdominal examination, presence of green gastric residuals) before it becomes clinically apparent; and 2. Can be used to predict the time required to reach full enteral feedings.

Starting at 7 days postnatally the infants will undergo simple, noninvasive tests to assess gastrointestinal (GI) function. The assessments will include:

1. Gastric emptying measured every two weeks using the 13C-octanoic acid breath test; 2. Measurement of GI mucosal integrity every 2 weeks with the sucrose/lactulose/mannitol/sucralose urinary excretion (permeability) test (PT); 3. Measurement of lactase activity every 2 weeks using the lactulose/lactose urinary excretion test (part of the PT); 4. Measurement of GI inflammation weekly using fecal excretion of calprotectin and alpha-1-antitrypsin.

This is a preliminary investigation designed to assess potential early markers of feeding intolerance in preterm infants. The results of this preliminary investigation will be used to define further studies leading to improvement in the clinical care of these patients. The sample size of 60 neonates is intended to provide preliminary estimates of prognostic associations between various measures of feeding intolerance and potential predictors of such intolerance. For example, dichotomous cut points for potential prognostic indicators (e.g., fecal alpha-1-antitrypsin; normal/abnormal) will be related to dichotomized feeding intolerance outcomes (e.g., green gastric residual; yes/no) using sensitivity, specificity, and positive/negative predictive values. Since it is not possible to know the ultimate denominators for sensitivity, specificity, etc. measures, the sample size will provide preliminary estimates of these measures with corresponding confidence intervals. The sample size of 60 will also provide 15 subjects (general rule of thumb) for each of the four tests to be included in the ordinary multiple regression analysis to predict continuous outcome measures (e.g., gastric residual volume and time to full enteral feeding) and for multiple logistic regression for prediction of dichotomized feeding tolerance outcomes. This sample size is also sufficient to detect bivariate correlations as small as r = 0.30 between continuous outcomes such as time to full enteral feeding and continuous markers. We consider a correlation of 0.30 as indicating the smallest relationship of clinical interest.

Infants born 25-32 wk gestational age without cyanotic heart disease, congenital GI anomalies including gastroschisis and omphalocele, chromosomal abnormalities, or inherited deficiencies, whose attending physicians and parents agree to study participation.

We are actively recruiting for the study now.