14781-koppen

205 Colonic manometry and contrast enemas 9 There are uncertainties to take into account when interpreting the results of our study. Due to the retrospective study design, this study is at risk of selection and information bias. In addition, the fact that this study was performed at a tertiary care setting may have resulted in selection bias. At the same time, children with severe, intractable FC are most commonly seen in a tertiary care setting and colonic manometry and contrast enemas are usually performed in specialized centers, which makes it likely that our sample is representative for the population. Another bias may have resulted from the fact that a significant number of patients had to be excluded based on the quality of their contrast enema or colonic manometry studies. This high rate of exclusions is expected, since contrast enemas and colonic manometry had been performed for clinical purposes and were not purposefully performed to fulfil our inclusion criteria. In this study, we only investigated the colon distally from the splenic flexure (the left colon). We chose to do so because in most patients with FC this segment is also evaluated by colonic manometry and contrast enemas. Colonic manometry of the entire colon would be preferable. However, with a colonic manometry catheter that usually is only 70 cm in length this is not always possible, especially in older children, and physicians therefore choose to evaluate the left colon where dysmotility is most often observed. We also evaluated studies performed with different types of catheters (solid state vs water-perfused) and with different sensor interspacings (3, 10, and 15 cm). A previous study has shown that identification of HAPCs with a solid-state catheter correlates overall well with a water-perfused catheter. 35 The high-resolution manometry catheter (36 sensors, 3 cm apart) provides a more detailed characterization of colonic motility and the point of HAPC termination than the low-resolution motility catheters that were used. High-resolution colonic manometry is a relatively new diagnostic tool and has only been implemented in our center since 2013. Therefore, most patients who were included in this study had received low-resolution colonic manometry. Since the contrast enemas and colonic manometries were performed within a 12-month time period, a change in the anatomical or functional situation could have occurred in the meantime, which may have affected our results. However, the median duration between contrast enema and colonic manometry was approximately two months and it is doubtful that in these patients with such longstanding symptoms major changes would have occurred in that time frame. In addition, colonic manometry recordings were performed after a bowel cleanout, while contrast enemas were performed without a preceding bowel cleanout. Finally, we compared our data with a normative population that was younger than our sample (<6 years of age). It is uncertain if these normative data can appropriately be used in older children. Unfortunately, there are no other normative data available for colonic size characteristics in children. The same is true for the definition that was used for megacolon/