Oral Entrainment on Suck Development of Preterm Infants Born 23 to 34 Weeks Gestational Age (GA)

Study Design Primary Hypothesis The use of a patterned oral somatosensory entrainment stimulation program (NTrainer©) will safely accelerate the development of oromotor skills (suck) in preterm infants and is expected to significantly enhance the development of oral feeding skills. Non-nutritive suck development will be assessed on several dimensions of oromotor output including suck burst structure, minute-rates for suck cycle production, and a novel quantitative measure of suck pattern formation known as the non-nutritive suck spatiotemporal index (NNS STI) (Poore, Barlow, Wang, Estep, 2007, revision for Dev Med Child Neurology; Poore, Barlow, Wang, Gu, 2007, in prep for Dev Med Child Neurol). Measures of feeding skill include measurement of the transition time (days) to 100% oral feed, feed rate (ml/min), pulse oximetry, and documentation of reflux or spit up events. Secondary measures will include documentation of infant growth rate compared to no intervention. The growth rate will be assessed by examining the change in weight, head circumference and length during the 28 days following initiation of oral feedings. Safety will be assessed by monitoring mortality, and the occurrence of feeding problems (numbers of days when feedings are withheld; ie NPO days), necrotizing enterocolitis, and aspiration events.

Enrollment We will conduct a randomized control trial to evaluate the efficacy of the NTrainer on suck development and oral feed performance.

We estimate we will need to study 100 preterm infants to test our primary hypothesis. (see sample size calculations in Statistical Analysis section). Enrollment will include 50 treatment / 50 controls. Expected ethnic proportion for Kansas [US Federal Census]African American 5.9%, Asian American 1.8%, Hispanic American 5.6%,Native American 0.9%, White 86.3%, All others --

Study Population Inclusion Criteria

• Documentation of informed consent. Informed consent will be obtained by the principal investigator(s) and/or designated research staff for each subject before being randomized into the study. We have a PACIRB approved Spanish version of the consent form available for interested participants.
• Gestational age between 23 weeks and 0/7 days and 34 weeks as per the best estimate by the neonatologist.
• If subject is transferred to another hospital, the ability to obtain follow-up data on outcomes.
• Ready for oral feedings. Neonates who meet eligibility criteria (A-E) will be assessed by the health care team (neonatology and occupational/developmental therapist) with the NTrainer beginning at 32 weeks post menstrual age (e.g., this will be 9 weeks of age for a neonate born at 23 weeks estimated gestational age). If they are considered ready for feedings they will be enrolled. Feeding readiness criteria include:

A. Initiate when the health care team feels the baby is ready to try oral feeds.

B. Hemodynamically stable. C. Stable O2 saturation levels during suck (infant will maintain SpO2 levels ≥ 90 during suck).

D. Alert and active: neurologic exam appropriate for post-menstrual age E. No signs of sepsis or intolerance of gastric feedings

Exclusion criteria

• Gestational age < 23 weeks or > 34 weeks.
• Any major congenital anomalies (chromosomal abnormalities, nervous system anomalies, cyanotic congenital heart disease, gastroschisis, omphalocele, diaphragmatic hernia and/or other major gastrointestinal anomalies).
• Major neurological injury including intraventricular hemorrhage (IVH grades III or IV), or periventricular leukomalacia (PVL).
• Not ready for oral feedings as determined by the health care team.

Outcome Measures

Primary Outcome Measures (Primary Study Endpoints)

• The Non-Nutritive Suck SpatioTemporal Index (NNS STI) will be used to characterize the emergence and integrity of the suck central pattern generator (sCPG) through quantitative and statistical analyses of suck pattern stability. This procedure involves the calculation of nipple compression variability across multiple suck bursts. By calculating the cumulative sum of the standard deviations of an amplitude- and time- normalized set of NNS pressure trajectories, suck development can be represented by a single numerical value known as the NNS STI. The mathematics underlying STI are well suited to quantitatively track the emergence of ororhythmic stereotypy during NNS development in preterm infants (Poore, Barlow, Wang, Estep, Lee, 2007). The STI indicates the degree to which the set of motor trajectories converges on a single underlying template, or the stability of the neuromotor sequences exhibited by the preterm infant. This approach is conceptually unique from all previous studies on suck development which have attempted to characterize ororhythmic activity through parametric analyses such as mean sucks per burst, sucks per second per burst, suck width and intersuck width. The NNS STI will be calculated at 7, 14, 21 and 28 days after randomization to assess maturation of suck coordination.

Details of the NNS STI calculation: Two minutes of the digitized ororhythmic pressure waveform with the greatest number of pressure peaks greater than 1 cm H2O, reflecting each infant's most active period of oromotor output, will be selected for NNS STI analysis. Because analyses of NNS spatiotemporal stability depend on comparing suck bursts with a fixed peak (suck cycle) number, the first five peaks from five successive bursts will be identified from the two-minute ororhythmic pressure waveform sample. For infants with degraded NNS pattern structure, the first five most burst-like mouthing movements are identified, based on period, amplitude, and duration.

The selection of NNS bursts and calculation of the non-nutritive suck STI will be completed with a specialized LabVIEW© software program developed in Dr. Barlow's laboratory known as NNS STI. NNS STI is programmed to initially perform pressure peak detection for each burst to identify the time location for each peak. After the desired pressure peak locations are identified, the start- and end-points for the selected NNS burst are calculated by extending the analysis window 300 samples prior to the first peak and 300 samples following the fifth peak, in order to ensure accurate pressure peak waveform discrimination. Amplitude and time normalization of these five bursts is completed next. Time normalization is based on linear interpolation, which projects the five-peak-bursts ensemble to an analysis window based on a preset abscissa scale of 10,000 data samples. Amplitude normalization is accomplished by deducting the mean and then dividing by the standard deviation for each trajectory. The NNS STI represents the cumulative sum of the standard deviations, indexed at 100 ms intervals, on the normalized NNS burst waveforms.

• Transition-to-Oral Feed (TOF): Time from initiation of oral feedings to full independent feedings (i.e., no gavage supplementation and taking at least 120-180 ml/kg/day.)
• Length-of-Stay (LOS) in the NICU.
• Occurrence of any of the following during the 28 day study period:

• necrotizing enterocolitis
• clinical suspicion of aspiration during feeding event (i.e., cough, spit-ups, O2 desaturation).
• number of days feeding withheld for more than 12 hours (NPO)
• death

Secondary Outcome Measures

Growth velocity during the 28 days after initiation of oral feedings:

• Weight Gain (gms/kg/day) = [wt (gms) at 28 days - bw (gms)] / bw / 28 days
• Head Growth (cm/wk) = [hc (cm) at 28 days - hc (cm) at birth] / 3 wks

• Occipitofrontal circumference = place measuring tape around the front of the head, above the brow and the occipital area. The measuring tape should be above the ears.
• Length Growth (cm/wk) = [len (cm) at 28 days - len (cm) at birth] / 3 wks Abbreviations:birth weight (bw); weight (wt); grams (gms); head circumference (hc); centimeters (cm),length (len) and weeks (wks).

During the course of the study, adverse events will be monitored to ensure that the rate of reported events does not exceed the expected rate.

Randomization and Treatment Assignment Randomization will take place using an electronic randomization schedule generated by Minitab v. 15 statistical software program. Neonates who are started on a trial of oral feedings will be randomly assigned to one of two groups - initiation of oral feedings with the NTrainer or to the Control group.

Inclusion criteria

• Estational age between 23 weeks and 0/7 days and 34 weeks as per the best estimate by the neonatologist.
• No major anomalies (chromosomal abnormalities, cyanotic congenital heart disease, gastroschisis, omphalocele, diaphragmatic hernia or other major gastrointestinal anomalies, major neurological injury or anomaly, and multiple congenital anomalies)
• Ready for oral feedings. Neonates who meet eligibility criteria (A-E) will be assessed by the health care team (neonatology and occupational therapist) with the NTrainer beginning at 32 weeks post menstrual age A. Hemodynamically stable and off all vasopressors B. Tolerating 120 ml/kg/day of enteral feedings C. Respiratory rate <80 breaths per minute D. Alert and active neurologic exam appropriate for post-menstrual age E. No signs of sepsis or intolerance of gastric feedings

Exclusion criteria

• Gestational age < 23 weeks or > 34 weeks
• Any major congenital anomalies
• Not ready for oral feedings