Short bowel syndrome (SBS) occurs when there is insufficient intestinal mass to support normal growth and development. Approximately 30 out of every 100,000 babies are affected by SBS in North America, and these infants remain dependent on intravenous, parenteral nutrition (PN) for prolonged periods of time. Children with SBS frequently fail to achieve sufficient linear growth and weight gain despite receiving calories in excess of that required by age-matched healthy children. Poor intestinal absorption, motility and increased inflammation all contribute to poor growth in these patients. In addition, children with SBS are known to have significant disturbances to their normal commensal gut bacteria. They may experience a depletion of specific groups of beneficial gut bacteria, and their metabolic by-products, specifically short-chain fatty acids (SCFAs), which can lead to intestinal inflammation, malabsorption, and a less efficient use of consumed calories.

In the proposed study, I hypothesize that children with SBS who are given supplements of targeted probiotics will have an increase in beneficial anti-inflammatory bacteria in their gut that more closely resembles the microbiota profile of healthy children. In addition, the children receiving probiotic supplementation will have increased concentrations of fecal SCFAs and improved growth compared to children with SBS who are not receiving supplementation. The central hypothesis will be tested by 1) prospectively characterizing the intestinal bacterial populations (by using next-gen sequencing methods), and measuring SCFA concentrations in the stool of children with SBS receiving probiotic treatment compared to those receiving no supplementation and 2) determining differences in the growth trajectory of the children in both groups by measuring sequential anthropometrics.

Enrolled patients will be randomized to either continue with standard of care, or to receive a daily probiotic for 3 months. A total of 3 stool samples will be collected from each patient (at the beginning, midpoint and end of the study) and fecal 16S rDNA microbial sequencing and SCFA concentrations will be compared between groups, as will the groups growth trajectory.

The long-term objective of the study is to determine how to effectively change the gut microbiota in children with SBS to restore a healthy balance and maximize growth and development. Although children with SBS have known disturbances to their intestinal microbiota, it is unclear whether providing an oral probiotic is an effective approach to correct these disturbances.

1. Introduction and Purpose: The goal of this project is to improve linear growth and weight gain in children with short bowel syndrome by changing their intestinal microbiota and ultimately intestinal bacterial metabolism.

2. Background: Short bowel syndrome (SBS) occurs when there is insufficient intestinal mass to support normal growth and development. Approximately 35 babies out of every 100,000 live birth are affected by SBS in North America, and these infants remain dependent on intravenous, parenteral nutrition (PN) for a prolonged period of time. One of the major challenges these children face is achieving sufficient linear growth and weight gain despite receiving calories in excess of that required by age matched healthy children. Problems with poor intestinal absorption and motility and increased inflammation all contribute to poor growth in these patients. An important factor in the healthy function of the remaining small bowel is the intestinal microbial community. It is known that depletion of certain groups of beneficial gut bacteria, and their metabolic by-products, specifically short-chain fatty acids (SCFAs) can lead to intestinal inflammation, malabsorption, and less efficient use of consumed calories.

   The long-term objective of the study is to determine how to effectively change the gut microbiota in children with SBS to restore a healthy balance and maximize growth and development. Although children with SBS have known disturbances to their intestinal microbiota, it is unclear whether providing an oral probiotic is an effective approach to correct these disturbances.

   Aim 1: To determine if probiotic supplementation can significantly modulate gut microbiota and improve growth in SBS patients.

   Aim 2: To determine levels of fecal SCFAs in children with SBS, including those receiving probiotics.

3. Study description: All eligible children will be consented at the time of an outpatient clinic visit or during an inpatient stay. Enrolled participants will then be randomized to "placebo" where they will receive the standard of care medical and nutritional treatment for short bowel syndrome in addition to 1mL of pure MCT (medium chain triglyceride) oil daily for 3 months, or "probiotic therapy" where they will receive a combination of Lactobacillus rhamnosus (5 billion colony-forming units per day) and Lactobacillus johnsonii,(200 million colony-forming units per day) daily for 3 months in addition to the standard of care. Randomization will not be blinded. The probiotics were chosen because they have been well studied in infants and children and both of these species have been found to be deficient in children with SBS and poor growth. The placebo was chosen because it is the only inactive ingredient included in the suspension of the probiotic, it is odorless and tasteless, and has no known side effects. The probiotic and MCT oil will be given daily either orally or through an existing gastrostomy tube. Patients who drop out of the study, or who are withdrawn will not be replaced, and the data from these subjects will be included as part of the analysis in an intention-to-treat manner.

Stool (approximately 200mg per sample) will be collected at 3 separate time points for each patient (at the beginning, middle and end of the study). Stool will be collected either during an inpatient stay or at an outpatient clinic visit. If a patient is unable to provide a stool sample, the family can collect one at home and it can be frozen and brought in at the time of the next clinic visit. A frozen sample can be used for 16S rRNA sequencing and metabolomics analysis, but not for SCFA concentration. Microbial gDNA can be recovered from stool that has been frozen for up to 6 months, with similar sequencing results compared to freshly processed stool. If a fresh sample is provided, a portion will be used to determine the concentration of SCFAs. Stool used for 16S rRNA sequencing and metabolomics will be kept frozen until analysis is performed. A one-time stool collection will also be obtained from 10 control patients with short bowel syndrome who are growing well and will be used for 16S rRNA and metabolomics analysis.

Stool used for 16S rRNA sequencing will be kept frozen until the time of analysis. Bacterial gDNA will then be extracted from fecal contents. Amplicons of the bacterial 16S rRNA genes will be generated and sequenced using the Illumina Hi-Seq 2000 (UTSW Genomics Core). Sequencing reads will be filtered and quality trimmed and analyzed using the QIIME software package.59 Microbial diversity will be estimated by calculating the Shannon diversity index and Bray-Curtis dissimilarity index.

Bacterial abundance will be determined using standard curves constructed with reference to cloned bacterial DNA corresponding to a short segment of the 16S rRNA gene that was amplified from bacterial reference strains.63 Dr. Koh has also validated qPCR protocols for 6 specific bacterial groups: 1) Eubacteria (all bacteria); 2)Lactobacillus (a subset of firmicutes); 3) Bacteroides (a subset of Bacteroidetes); 4) Enterobacteriaceae (ENTERO; pro-inflammatory gram-negatives such as E.coli, Klebsiella spp, etc); 5) CLEPT (a subset of Firmicutes; butyrate producing anti-inflammatory Clostridia); 6)EREC (a subset of Firmicutes; butyrate producing anti-inflammatory Clostridia). Bacterial group specific qPCR provides quantification of overall abundance of specific groups of bacteria

Concentrations of SCFAs in the stool will be measured at all time points that a fresh sample can be provided. Briefly, fresh fecal samples will be weighed, suspended in water with 0.5% phosphoric acid and then flash frozen immediately with liquid nitrogen. Given the volatility of SCFAs, flash freezing will provide more accurate measurements. Fecal suspensions will then be thawed, homogenized, and extracted with ethyl acetate. Organic extracts will be spiked with carbon-labeled SCFA standards (in order to accurately quantify SCFA levels) and run on the GC-MS.

Randomization: Enrolled patients will be randomized using a centralized process based on a random table. Randomization will be determined by calling a central project phone number answered by the research nurse and requesting a subject ID and an allocation to either standard treatment or conditional surgery. This is an open-label trial and patients will be randomized in a 1:1 manner to one of two treatment arms.

Treatment Phase:

Placebo patients will receive the standard of care medical and nutritional treatment for short bowel syndrome in addition to daily MCT oil. All enrolled patients will have detailed growth assessment during the study. The primary anthropometric measures of growth will include weight, height, head circumference (for children <2 years) and mid-arm circumference, which will be documented upon enrollment, and monthly during the study. Body mass index and z-scores for growth variables will be calculated at each time point. Expected and median values for anthropometric parameters will be obtained from the World Health Organization growth chart for patients <2 years and the Centers for Disease Control growth charts for patients ≥2 years.

Clinical information will also be collected from the medical records for all enrolled patients including information about age, gender, ethnicity, etiology of SBS, quality and frequency of daily bowel movements, antibiotic usage in the 4 weeks prior to enrollment and during the study, infections during the study, surgical history, including any bowel-lengthening procedures, length and anatomy of small bowel and colon remaining, the presence of a stoma and/or gastrostomy tube, medical history, medications, detailed nutrition information (caloric intake from parenteral and enteral nutrition), and routine nutritional blood work. Stepwise logistical regression analysis of these clinical characteristics will be performed to model the independent predictors of poor growth, using a significance level of p<.05. A Fischer's exact test will be performed to assess the overall distribution of discrete clinical variables among patient analyses.

Probiotic Group - where they will receive a combination of Lactobacillus rhamnosus ( 5 billion colony-forming units per day) and Lactobacillus johnsonii (200 million colony-forming units per day) daily for 3 months in addition to the standard of care.

Stool samples will be collected by both groups at 3 separate time points for each patient, upon initial enrollment, the midpoint and end of the study). The samples will be collected at a standard of care clinic visit, during an inpatient stay, or at home by the caregiver and frozen. Caregivers will be provided with instructions and stool collection kits.

Inclusion Criteria:

• have a diagnosis of short bowel syndrome (<25% of their expected small bowel length based on age or > 6 weeks of parenteral nutrition after small bowel resection
• cared for by the Intestinal Rehabilitation Team at Children's Health
• are receiving at least a portion of calories from enteral nutrition

Exclusion Criteria:

• have used probiotics within 2 weeks of enrollment
• are not able to come to regularly scheduled appointments during the study period
• are unwilling to collect stool samples