14781-koppen

157 Water-perfused colonic manometry 7 Frequency spectra were analyzed using a Bayesian estimation method based on statistical modeling using the t-distribution. We utilized the Markov chain Monte Carlo (MCMC) technique using software from the Stan Development Team (PyStan: the Python interface to Stan, Version 2.4). Analysis of t-distributions was chosen because it is a robust approach to handle outliers. We have used the MCMC technique in previous publications, where the technique is described in detail. 13,17 Here, the mean RMS distribution for each frequency and patient type is computed with MCMC. Statistical differences between the grouped means between preprandial and postprandial data within pediatric subjects were then calculated. This was achieved by subtracting the preprandial means from the postprandial means. Where the 95% highest density interval of the differences between the means being compared did not contain a 0 (i.e. the value was greater than 0), this was considered to be a statistically significant difference. The greater the value from 0, the greater the effect size. RESULTS Colonic manometry was performed in 19 children (median age 15 years; range, 4–18 years). In one of the subjects, the catheter tip was placed in the cecum, which resulted in manometric recordings from the ascending colon, the transverse colon, and proximal the descending colon only. The data of this patient have been excluded from all analyses, leaving 18 patients (five males). In one child (no. 12, Table 1), all of the sensors were located in the sigmoid colon. Thus, data for descending colon are reported from 17 children. Of the nine children with measured colonic transit, six had proven slow-transit constipation and the remaining three had normal colonic transit (Table 1). However, in the children diagnosed with ‘normal transit’, laxative medication was taken. Of 13 children who had anorectal manometry, 11 had demonstrable evidence of a RAIR and normal or slightly elevated anal sphincter resting pressure (Table 1). The remaining two children did not have manometric evidence of RAIR. However, both children have since had Hirschsprung’s disease excluded from their pathology. The adult data came from14 patients with scintigraphy-diagnosed slow-transit constipation (two men; median age, 52 years; range, 24–76 years) and 12 healthy adults (five men; median age, 51 years; range, 27–69 years). 13,17 The patients with slow-transit constipation all reported a long history of constipation with 10 of 14 patients reporting constipation from childhood and the remaining four patients reporting constipation worsening from puberty into adulthood.