Comparison of Esophageal Clearance Times of Oral Budesonide Preparations

This study will be a blinded randomized cross over trial study, comparing one of two alternative slurries (xanthan gum or honey) against the current standard sucralose slurry in each subject and determining esophageal clearance for each.

A total of 24, up to 30 healthy adult subjects will be enrolled and randomized to their treatment group by sequentially numbered, opaque, sealed envelopes (SNOSE) prepared by an uninvolved third party. These envelopes will be labeled numerically "Subject 1" through "Subject 24", and will randomly contain which treatment will be administered, and in what order. If additional subjects are recruited, 6 additional SNOSE will be prepared in a similar manner. There will be 4 types envelopes prepared- Sucralose-Xanthan Gum, Xanthan Gum-sucralose, Sucralose-Honey and Honey-Sucralose. One of these regimens will be randomly inserted into an opaque envelope as described previously and sealed. The envelope will be given to the subject after enrollment and consent, and will be opened by the nuclear medicine pharmacist to determine the assigned preparation that will be compounded and administered. Subjects will be randomized in a 1:1:1:1 ratio using a computer program based on random number generation.

The sucralose slurry will consist of 10 grams of sucralose mixed with 4ml of 0.5mg/2ml nebulized budesonide solution (for a total budesonide dose of 1mg) and 1mCi (1ml) of Tc99 sulfur colloid for a total volume of approximately 10ml.

The honey slurry will consist of 5ml of honey mixed with 4ml of 0.5mg/2ml nebulized budesonide solution (for a total budesonide dose of 1mg) and 1mCi (1ml) of Tc99 sulfur colloid for a total volume of approximately 10ml.

The xanthan gum slurry will consist of 50mg xanthan gum powder mixed with 4ml of a 0.5mg/2ml nebulized budesonide solution (for a total budesonide dose of 1mg), along with a flavoring packet of cold pressed crystallized orange flavor and 1mCi (1ml) of Tc99 sulfur colloid and 5ml of distilled water for a total volume of approximately 10ml.

The study technician, interpreting radiologist and study coordinators will be blinded to the type of slurry ingested. Each subject will be asked to guess which of the three slurries he/she received following consumption of each slurry. In addition, the subject will complete the hedonic general labeled magnitude scale (h-gLMS) following consumption of each slurry.

Data will be collected at initial entry (age, sex, race, height, complete blood cell count ) and recorded on a data spreadsheet. The first swallow study will be recorded as a time interval (1 second intervals from initial swallow until 10 minutes have elapsed) as a count of tracer in the esophagus, from which an area under the curve of total esophageal contact time will be calculated. After a 1 hour "washout" period and two 8 oz glasses of water are ingested the second slurry will be administered and the data from the same time points will be collected. This data will be obtained from the nuclear medicine department. The same individual will denote the areas of interest on the nuclear medicine study (upper to lower esophagus) as well as obtain, record, and calculate clearances to eliminate inter-operator variability. This technique is validated and is currently utilized to determine esophageal clearance for the diagnosis of achalasia in adults and will utilize the same camera and protocol (NM 250- Esophageal Transit Study). Data results will be plotted as quantity of tracer remaining (y axis) vs time (x axis) to calculate a total Area Under the Curve (amount remaining at each time point summed until all tracer has cleared).

In a previous study comparing nebulized and viscous topical corticosteroid treatments (Dellon et al, Gastroenterology 2012;143:321-324), the mucosal medication contact time measured by scintigraphy and represented by the area under the curve (AUC) for the overall esophagus. The AUC was higher for the oral viscous slurry group (median=48,900) compared to the nebulized/swallowing group (median=19,200), p=0.005.

The measurement units, range and standard deviation for these contact time AUCs were not provided. There is also no data provided on within subject changes for any outcomes.

Attempting to extrapolate the expected variability in the AUCs in this study, standard deviations of ~29000 for the oral viscous slurry group and ~11400 for the nebulized/swallowing group would result in a similar P value. Both of these theoretical standard deviations are ~60% of the medians.

1. Assumptions for sample size estimation for the comparisons of the standard oral viscous slurry to an experimental slurry):
2. The mean AUC for the oral viscous slurry treatment (the 'standard') will be 48,900, with a standard deviation of 29,000.
3. For each group of subjects (xanthan gum and honey), the probability of a Type I error is controlled at α=0.05.
4. There will be no crossover or carryover effect.

If the study design was not a crossover and two separate groups were studied (one receiving the standard slurry and the other receiving the experimental slurry), a sample of 13 subjects per group would have 80% power to detect a difference of 24450 in AUC (i.e. a 50% change from the standard oral viscous slurry).

Given that this study design has each subject measured for both the standard and experimental slurries, we assume that there will be a correlation between studies (r=0.70, within subject standard deviation of 22,460) and we anticipate that a sample of 12 subjects will have 80% power to detect a change of 20000 between oral viscous slurry and the experimental slurry (either honey or gum).

To allow for dropouts, up to 30 subjects will be recruited for this within-subject comparison in the study.

Data Analysis Plan:

1. Demographic characteristics of the study groups will be presented using means with standard deviations and counts with percentages.
2. A patient flow diagram will be presented to describe the recruitment process (i.e. number of subjects consented, number of withdrawals), randomization and completion rates.
3. The primary outcome is the the mucosal medication contact time measured by scintigraphy and represented by the area under the curve (AUC) for the overall esophagus. A secondary outcome will be the number of total seconds until 100% clearance.
4. Two separate experiments will be analyzed: Comparison of xanthan gum to budesonide and comparison of honey to budesonide. With the AUC as the dependent variable, repeated measures analysis of variance for the crossover design will be used to compare slurries and the order of testing.
5. Secondary outcomes will be the number of seconds until clearance and the taste of the substance as measured by gLMS. Repeated measures analysis of variance for the crossover design will be used to compare slurries and the order of testing.
6. Differences in AUC and taste between honey and xanthan gum will be explored using a two sample t-test or Wilcoxon rank sum test as appropriate