Laurens Schattenkerk

139 Incisional hernia after surgical correction of abdominal congenital anomalies in infants Chapter 5 Since the incidence in incisional hernia in the pediatric population is low, most cohort studies only describe limited numbers of patients with an incisional hernia. Still, in most of these studies thorough statistical analysis is done, often by chi-square or regression analysis. However, both analyses demand a sufficient amount of hernia cases ensuring the least expected count to be five for chi-squared statistics and a minimum of ten hernias per one controlling variable (events per variable (EPV)) for any form of regression analysis [368, 369]. Not abiding these statistical rules can lead to inaccurate results causing conclusions to be, to some extent, unsupported or deceptive. Future research should ensure statistical power with a sufficiently large cohort. The incidence provided by this review can aid in power size calculations. This review was limited because none of the included studies used a specific classification for incisional hernias or described the way they were diagnosed. Moreover, many studies did not report on duration of follow-up or had a follow up of less than a year, possibly resulting in missed incisional hernias. Also, different surgical approaches and disease severities were included. These differences dissolve when combining the study results into proportional meta-analysis, which could have influenced our results. The broad inclusion criteria, resulting in differences between compared studies, can partly explain the broad confidence intervals of gastroschisis and omphalocele. In adults, consensus has been reached that an incisional hernia should be classified by its location, length and width [370]. Whilst infantile incisional hernias might differ from adults, it seems advisable for prospective studies to include these characteristics as well as to describe the diagnostic modality as to aid in the generalizability of the interpretations. Using any form of diagnostic imaging will increase the total amount of detected IHs. Yet, this increase will be accompanied with disagreement between observers, as is the case in adults, making the IHs less generalizable which is why we suggest physical examination could be sufficient [371]. In our opinion, the clinical significance of IHs in infants undetectable by physical examination is doubtful. Smaller hernias have been reported to close without necessitating surgery, as is also the case in congenital umbilical hernias [354]. Since most IHs develop within a year after surgery it is desirable to plan an extra repeat visit at least a year after surgery as part of the research protocol [318]. Another limitation of our study is that most included studies did not describe the incisional hernia cases in detail. Often, just the total number was noted without describing further case characteristics. This withheld us from thoroughly analyzing indications for redo-surgery or known risk factors such as having a history of stoma or surgical site infection. However, for gastroschisis specifically, though these finding are based on non-randomized retrospective cohort studies, our results suggest that SILO closure and complex cases are more at risk of IH. The high overall heterogeneity seems

RkJQdWJsaXNoZXIy MTk4NDMw