Page 164 - Prevention and Treatment of Incisional Hernia- New Techniques and Materials

Chapter 8
value of yi is μi=xi exp (_ati), or log μi=logxi_ati. To estimate a, the GLM is used, assuming for yi a Poisson distribution with expectation μi and a logarithmic link function. The linear predictor is hi=logxi+ati, with an o set log xi, and no intercept. Using the GLM function in the R system, a is estimated(29). A larger a means a smaller probability per month follow-up of getting a recurrence. The hazard (a) is transformed from a monthly risk to a yearly risk of getting a recurrence during long-term follow-up.
Results
The systematic database search identi ed 1749 records and 80 additional records were identi ed through additional cross-referencing. After removal of all duplicates 1467 unique records remained. All abstracts were screened for eligibility and for 410 records the full-text article was assessed. Fifty- ve articles containing 3945 patients met the inclusion criteria and were selected for review and included in GLM analysis. The PRISMA ow-chart of the selection of relevant studies can be found in gure 1(30). Three techniques of open reconstruction without mesh were described comprising 460 patients. Six di erent techniques of open reconstruction with mesh were described comprising 3002 patients and 483 patients were repaired by laparoscopic approach. An overview of the di erent LIH repair techniques and results are shown in table 2.
1. Open reconstruction without mesh
a) Components Separation Technique (CST/Ramirez)
CST was rstly described by Albanese in 1951(82) and named as such after the publication of Ramirez in 1990(8). During CST the abdominal wall is augmented by creating relaxing incisions in the external oblique aponeurosis and separating the external oblique from the internal oblique muscles and elevating the overlying rectus muscle from the posterior rectus sheath ( gure 2). In 7 studies, including one RCT, a total of 219 LIH were repaired using CST(9, 11, 31-35). In, approximately 40% of cases, patients had a complex LIH.
Postoperative mortality was 1.3%, and postoperative complications occurred in almost 50%. Infection or necrosis of the wound occurred in 20%, hematoma in 8%, seroma in 9%, and pulmonary complications in 7%. In one patient a rupture of the abdominal wall at site of the relaxing incisions occurred
162