173 Incisional hernia after abdominal surgery in infants: A retrospective analysis of incidence and risk factors Chapter 7 Statistical analysis Descriptive data were reported according to distribution as median with range or mean ± standard deviation (SD). Categorical data were analysed with the Chi-square-test, Fisher’s exact test or One Way ANOVA accordingly. Multivariate logistic regression analysis was performed to identify risk factors for incisional hernia. The assumption of linearity of the logit of ordinal variables was assessed using the Box-Tidwell test. Only if the assumption was not met, a variable was categorized. Forward Wald selection was used for selection of variables and assessment of confounding (increase in B-coefficient of >10%) and effect modification (significant interaction term). Significant risk factors were reported in odds ratio (OR) with 95% confidence intervals (95%-CI). Significant risk factors were defined by a 95%-CI not including the one. Additionally, the adjusted R-squared will be reported to show the proportion of the variance in IH explained by the model. Hosmer-Lemeshow test will be performed to assess the model’s goodness of fit. the Subgroup analysis was performed within two subgroups of the overall cohort. The first subgroup separated infants who developed an IH following a laparoscopically or a laparotomy. In this subgroup we evaluated if the distribution of the risk factors identified in the overall cohort differed between the IHs following the different surgical approaches. The second subgroup evaluated all infants operated by laparotomy to evaluate if technical factors (e.g. suture resorption speed) differed significantly between those that did and did not develop IH following laparotomy. The significance level was set at P < 0.05 and corrected for Bonferroni in case of the Chi-square and Fisher’s test. Because we executed six tests the corrected significant p-value was set at P < 0.008 for the subgroup analysis. All analyses were performed with IBM SPSS statistics, version 23 (IBM Corp., Armonk, NY, United States). Results Patient and surgical characteristics Between 1998 and 2018, 2066 infants underwent abdominal surgery, 11 declined participation leaving 2055 infants accessible for inclusion. From a total of 43 different included diseases (Appendix 1), the three most reported reasons for the main surgery were infantile hypertrophic pyloric stenosis (PS) (420/2055), necrotizing enterocolitis (NEC) (189/2055) and anorectal malformations(240/2055). Patients characteristics are displayed in Table 1. Out of 2055 patients, 66% were male. Twenty-eight percent were born preterm with a median gestational age of 33 weeks (range: 23-73) and a median birthweight of 1805 grams (range: 523 – 4315). Age at main surgery was a median of 47 days (range: 0 – 1095 days). The main surgery was deemed as urgent in 51%. Laparoscopic surgery was performed in 20% of the infants. Out of these laparoscopic procedures, 75%
RkJQdWJsaXNoZXIy MTk4NDMw