265 Treatment of jejunoileal atresia by primary anastomosis or enterostomy Chapter 12 both treatment groups in our cohort which suggests they don’t explain the difference in complication rates between both treatments. Thirdly, in our cohort a significant difference in the location of the atresia between both treatment groups exists, showing that a jejunum atresia is more likely to be treated by primary anastomosis. Since it is known that jejunostomies are more at risk of becoming a high-output enterostomy, this might influence the surgeons choice of treatment in jejunal atresias in favour of primary anastomosis (22). Although in most (78%) of the enterostomy patients an ileostomy was made, the incidence of a high-output enterostomy in our cohort is high (48%), but lower than the 64% reported previously (7). This could be an important argument for the surgeon not to perform primary enterostomy, including ileostomies, in future cases. Lastly, the relatively high amount of patients born with a gastroschisis within the enterostomy group (seven versus one in the primary anastomosis group) could have been of influence of a higher complication rate. Previous studies report higher chances of mortality as well as morbidity, such as adhesive obstruction and incisional hernias, in patients suffering from both a jejunoileal atresia and gastroschisis (23, 24). Two out of three of the patients that died in the group treated with an enterostomy had an associated gastroschisis, which underlines these results. All surviving six patients experienced a Clavien-Dindo ≥III complication. These results suggest that for gastroschisis patients other considerations might influence choice of treatment. Still, it is unclear what the influence of surgical treatment is to the occurrence of complications in this important sub cohort. A sensitivity analysis, excluding all gastroschisis patients, resulted in non-significant differences between the two treatment groups for Clavien-Dindo ≥III complications, adhesive bowel obstruction and short bowel syndrome, whilst surgical site infection and wound dehiscence stayed significant. It could be that due to the exposure to amniotic fluid intrauterine the bowel of gastroschisis patients is already in a worse shape making it more prone for complications. However, the change to non-significance could also be related to power issues due to a reduction of more than 20% (7 of 31) in the enterostomy group. These small sample sizes might have influenced our results although our cohort is of similar size as previous studies. Also, our results are based on historical data which might have resulted in forms of bias such as reporter bias. Another sensitivity analysis, which split the cohort in half, based on the year of the first operation (before or after 2010), showed that primary anastomosis were not performed significantly more often after 2010 and that there were no significant differences in complication rates between the two decades included in this study.
RkJQdWJsaXNoZXIy MTk4NDMw