Page 307 - Prevention and Treatment of Incisional Hernia- New Techniques and Materials

Measurements
In case of severe infectious complications animals were euthanized before the intended endpoint. These animals were analysed together with the surviving animals sacri ced at the intended endpoint. Half of the surviving animals were sacri ced after 90 days and half after 180 days. During sacri ce the abdomen was shaved, disinfected and opened through a U-shaped incision extending lateral and caudal to the mesh. Mesh incorporation was de ned as percentage of the mesh edge incorporated into the abdominal wall, taking into account any shrinkage (example in Figure 1). In all mesh groups mesh surface and coverage of the mesh surface with adhesions was scored using a grid placed over the mesh, dividing it into 30 equal squares and facilitating accurate estimation of adhesion formation. Tenacity of adhesions was scored using the Zühlke-score, a 4-degree classi cation of adhesions based on histological and morphological criteria(14). Pictures of abdominal wall with mesh and the present adhesions were taken (5.0 megapixels digital camera; Sony Cybershot). Macroscopically mesh infection was de ned as the presence of abscesses of the mesh. Shrinkage was de ned as the relative loss of surface compared with the original size of the implanted mesh measured with a calliper. All measurements were performed by 2 independent observers and disagreements reconciled after discussion. The animals were euthanized by cardiac cut.
Statistical analysis
Incorporation, mesh infection, tenacity and percentage of adhesions, abscess
formation, survival, weight and shrinkage were compared using non-parametric
tests (Kruskal Wallis, Mann Witney, Chi-square, Fisher’s exact and Spearman’s
rho) since the data did not show a normal distribution. Therefore all results
are presented using the median and the interquartile range (IQR). In case the
overall test showed di erences, the pairwise tests were done to determine the 15 groups causing the overall signi cance. Exact methods for signi cance were
used when computational limits allowed these. All reported p-values are two- sided and considered signi cant if less than 0.05. In view of the small sizes of the groups, it was not possible to adjust the p-values using Bonferroni’s correction. Statistical analysis was performed using PSAW statistical software package, version 17 (IBM SPSS statistics).
Incorporation of biological meshes
305