Microbiome Survey of Preterm Infants

Overview

The skin and intestinal tracts of humans are covered with microbes, and the gene products of these microbes are collectively known as the human microbiome. Many studies in recent years have linked the microbiome to many aspects of human health and disease. Premature infants are vulnerable to invasive infections, some of which may originate from the microbes that colonize the skin and in the intestinal tract. Therefore, determining the patterns of early life colonization and the sources of colonization in these infants may be critical to determining infectious risks. This prospective study is proposed to identify the sources of microbes that colonize extremely low birth weight (ELBW; \<1000 g birth weight) infants immediately following birth through the first month of life. The study team plans to non-invasively survey the skin and stool of ELBW preterm infants in combination with sampling of their mothers' skin and stool and the intensive care nursery (ICN) environment. As controls, the study team will perform a similar survey of full term mother-infant pairs. Samples will be analyzed using high throughput sequencing techniques to determine microbial content. Topical barrier sprays or emollients are commonly used early in life in the ELBW infant to reduce insensible water loss and enhance the integrity of the skin barrier. The effect of these treatments on the development of the skin microbiome remains unknown. As part of this study, the study team also proposes the application of two different skin barrier therapies to small areas on infant's legs to determine the effects of the different treatments on skin colonization. While all parts of the study are described in this summary and the detailed description, the remainder of the record focuses on this interventional portion of the study (Group 2). The goals of this study are outlined as follows: 1. To elucidate the relationship between microbes in the health care environment and colonization patterns of the intestinal tracts in preterm and term infants. 2. To understand the relationship between mode of delivery, the health care environment, and the colonization of the skin in preterm and term infants. 3. To determine the effects of topical skin barrier therapies on the colonization patterns of the skin in preterm infants. 4. To ascertain the relationship between the composition of the microbes colonizing preterm and term infants and the development of mucosal and systemic immunity to those microbes.

The long term goals of this project are to understand what constitutes protective and beneficial gut and skin microbial communities and determine approaches to foster and preserve these communities. By defining the gut and skin-associated microbiological communities and how these communities are affected by the environment and medical interventions including barrier therapies, the study team will make essential first-steps towards achieving these long term goals. Design \& Procedures Study Groups: Potential subjects will be identified among Duke Hospital-admitted women with expectant preterm (≤30 weeks gestation) or term delivery. A total of 45 infants will be enrolled into one of three groups. Preterm microbiome Group 1 (n=10, ELBW preterm infants) will be part of an intensive survey of infant colonization with bacteria and fungi in addition to the intensive care environment. This will be a strictly observational group. Skin microbiome Group 2 (n=30, ELBW preterm infants) will be prospectively evaluated for the development of the skin microbiome in the context of a topical skin therapy typically used in the study nursery. Each infant will serve as his/her own control with a well-circumscribed, small area of the anterior thigh left untreated as the control area. Full term microbiome Group 3 (n=5, full term infants) will be surveyed for skin and intestinal colonization and serve as a control group. This will be a strictly observational group. Preterm and Full Term One-time Group 4 (n=100 preterm infants born at \< 2000 grams and 100 full term). This will strictly be an observational group. Comprehensive Microbiome Survey of Preterm Infants: Biological and environmental samples will be collected from enrolled mother and infant subjects (Study Group 1) from the time of delivery through the first two weeks of life. For stool samples, a minimum of 1 gram will be collected. For blood samples, 4 ml will be collected. Infant stool samples will be collected from diapers. Maternal stool and vaginal samples will be collected at the time of caesarian section using a sterile nylon swab of the distal rectum and vagina. Skin samples will be collected by gentling rubbing the target area with a sterile nylon swab moistened with sterile saline. Breast milk samples will be scavenged from left over material after the feeding of each infant. Since breast milk undergoes a single thaw from frozen storage per feeding, excess milk may not be stored for subsequent feedings and would be put to waste. A portion of stool, blood, and breast milk samples will be placed into sterile, nucleic acid-free tubes. An additional portion of stool, blood, and breast milk samples will be placed into sterile, nucleic acid-free tubes containing the preservative RNA later. All stool, breast milk, and blood samples will be frozen at -20°C within 1 hour of collection. Remaining samples will be placed at 4°C within 18 hr of collection. Samples will be transferred to -80°C storage within 24 hours of collection. A label carrying a unique sample number will be placed on each sample as it is collected. A binder with sample labels and a log of sample numbers, time of collection, sample type key, and special notes regarding collection, will be placed at the bedside. It is standard of care for ELBW infants in the Duke Intensive Care Nursery that non-sterile gloves are worn by all providers and family members for contact with the infant for the first two weeks of life. Providers and relatives will be asked to place used gloves in a zip-lock bag at the bedside after contact with the infant. The gloves will be grouped into zip-lock bags by provider type (medical provider, nurse, or ancillary staff) or relative. The bags will be collected each shift or day. Glove samples will not be labeled or linked with individual providers or relatives. When available, scavenged serum will be used for assessment of antibody specificity in relationship to colonization patterns of each infant. In some cases, stool samples will be analyzed for the presence, quantity and specificity of mucosal immunoglobulins. A video recorder will be positioned above the incubator on a tripod attached to the incubator vertical rail to capture changes in the infant's environment, including medical or nursing interventions. The camera will be positioned to capture the infant and the inside incubator surroundings, while avoiding the faces of those interacting with the infants to preserve anonymity. When hands or instruments are placed inside the incubator, they will be recorded. Video will be downloaded onto an external hard drive every 24 hours, and then onto a secure computer server for analyses. Video recording will be continuous for two weeks except for the period of time it takes to change the memory card. The video will be coded for the start and stop times of all clinical care events in order to explore the influence of environmental changes on the diversity of the infant's microbiome using Mangold Interact software. Inter-rater reliability will be checked by having video coders score a 30-minute period of each video, then computing Cohen's kappa. Microbiome Survey of Full Term Infants: The study team will perform a more limited survey of five full term infants as a control group (Study Group 3). Samples will be collected from mothers and infants. Samples will be collected daily while the mother and infant are in the hospital, which is generally expected to be approximately three days. At hospital discharge, the parents will be sent home supplies in a prepaid return mailing envelope to be used to collect skin and stool samples when the infant reaches two weeks of life. Skin Microbiome Substudy Plan: A controlled study of a commonly used skin treatments (No Sting protective spray and no treatment) will be performed in 30 ELBW infants (Study Group 2) to determine if No Sting promotes different skin microbial colonization. The current standard of skin care for ELBW infants in the Duke ICN is application of No Sting to all external exposed, non-mucosal areas shortly after birth and again at one week of age. A quarter 1.5 x 3 inch silicon patch with two holes will be applied to each of the infant's thighs. One hole in each patch will be left untreated (control) while the other is treated with No Sting. The patches will be used to define the treatment and no treatment areas. The remainder of the infant's skin will be treated with No Sting therapy at birth and seven days of life, as per the current standard of care. Daily sampling of each of the sites (No Sting, and untreated) will be performed by gentle application of a nylon swab for the first two weeks of life, and then twice weekly until one month of age. Samples will be analyzed for microbial content. Additional sampling will include infant stool as well as maternal skin, vaginal, perirectal, and stool samples as outlined in Table 1. The study team will use non-invasive skin probes to perform daily assessments of transepidermal water loss (TEWL) and skin pH. TEWL will be measured using the DermaLab® TEWL probe (Cortex Technology, Hadsund, Dermark), which consists of an open probe with paired sensors placed at different distances from the skin. Humidity and temperature are measured in each sensor to calculate vapor pressure gradients. The difference between two vapor pressure gradient measures is representative of TEWL at that point on the skin \[19, 20\]. Environmental humidity levels will also be measured using the DermaLab® before obtaining the TEWL measures. The average humidity readings from the two sensors will be recorded as environmental humidity. The probe will be thoroughly cleaned with 70% ethanol prior to and between uses to minimize cross-contamination or microbial transmission. The ethanol will be allowed to air dry prior to use of the probe. The Extech PH100 meter will be used to measure skin pH in the range of 0.00 to 14.00. The small flat surface electrode provides a non-intrusive accurate measure of pH. No Sting will be reapplied on day seven. Preterm and Full Term One-time Microbiome Survey (Study Group 4): One hundred preterm infants born at \< 2000 grams and 100 full term infants will be included for a one-time data collection of infant only microbiome samples. For premature infants environmental samples will also be collected. The same procedures described for Study Group 1 and 2 above will be used for this single data collection.