The rapid development of endoscopic techniques over the last three decades has enabled the safe repair of complex ventral hernias with a smoother recovery then a�er open ventral hernia repair and with less persis�ng postopera�ve pain. Endoscopic ventral hernia repair followed the introduc�on of the laparoscopic technique in the early 1990's, as with most other laparoscopic procedures. The procedure involved covering the hernia defect intraperitoneally with a mesh i.e., intra-abdominal Onlay Mesh (IPOM) repair, and was shown to have several advantages compared to established open techniques at that �me. Early follow-up studies on IPOM repair showed low risk for postopera�ve complica�on, low recurrence rates, and high pa�ent sa�sfac�on [1]. However, in contrast to open repairs that were mainly based on the onlay or sublay approach, IPOM repair involved placing the mesh within the abdominal cavity. Direct contact with the intes�nes, however, introduced the risks of adhesions, infec�on, or fistulae involving the mesh, and new types of mesh had to be developed to reduce those risks.A�er two decades' development, refinement, and numerous atempts to improve the technique, the unavoidable disadvantages of intra-abdominal mesh placement have become apparent [2]. The tension caused by the mesh may lead to long-term pain, and even if new compound meshes have improved the situa�on, intraperitoneal mesh-related complica�ons cannot be en�rely avoided. Furthermore, compound meshes required for IPOM repair are more expensive.In recent years, our understanding of anatomy derived from open hernia surgery has come to use when carrying out endoscopic hernia repair. Technological advances and increased experience have made other approaches to ventral hernia repair possible, entering anatomical areas that were previously only considered accessible with an open technique. The new endoscopic techniques for placing the mesh outside the abdominal cavity include endoscopic Mini-or Less-Open Sublay repair (eMILOS), endoscopic Totally Extraperitoneal Approach (TEA), TransAbdominal PrePeritoneal repair (TAPP), and enhanced-view totally extraperitoneal repair (eTEP). In comparison with onlay repairs, these techniques lead to greater reten�on strengths [3].With the TAPP technique, the preperitoneal space is entered through the abdominal cavity, a�er which the peritoneum is closed over the mesh. Laparoscopic TransAbdominal Retromuscular (TARM) repair is performed through a longitudinal incision in the peritoneum and posterior rectus sheath, providing access to the retromuscular space which enables placement of the mesh in a sublay posi�on.The development of the new techniques has to a great extent been fueled by the possibili�es provided by the new technologies. Nevertheless, the purpose of these innova�ons should be on an improvement of the long-term outcome and biomechanical proper�es of the repairs and not on the benefits perceived intraopera�vely [4]. As for many other cases when a progress in medicine is seen, the goals tend to be defined by the healthcare provider handling the new tools rather than quality measures defined by pa�ent perspec�ves.The new approaches are technically demanding, but the development of robot-assisted laparoscopy has overcome many of obstacles that could not be managed with conven�onal laparoscopy. Robo�c assistance facilitates dissec�on and provides access to all corners of the surgical field. In a study from Germany, eTEP, eMILOS and TAPP were introduced as rou�ne procedures for ventral hernia repair parallel to training in robot surgery at the clinic [5]. That study suggested that it is possible to go from open preperitoneal and retromuscular techniques to robot-assisted transabdominal ventral hernia repair placing the mesh in the same anatomical posi�ons. In this context, IPOM repair represents only a transitory era in the development of ventral hernia surgery.In a study from Sweden, a pragma�c approach was applied to robot-assisted ventral hernia repair where the primary aim was to carry out TAPP or TARM repair while being prepared to convert to IPOM if this was found to be too difficult. The study showed that this strategy was safe and resulted in rapid recovery and a low rate of long-term postopera�ve pain [6].Repair of a parastomal hernia is widely considered a challenging procedure as it requires restora�on of fascial strength while not interrup�ng con�nuity between the intra-abdominal intes�ne and the external stoma. However, a case report showed that totally extraperitoneal placement of a polypropylene mesh around the stoma using conven�onal laparoscopy was possible [7]. This approach avoids the risks associated with intra-abdominal mesh such as with the Sugerbaker technique.Concurrent with the development of endoscopic techniques, surgical approaches were developed, such as transverse abdominal release, to reduce the tension caused when closing the abdominal wall in very large incisional hernia repair. Recent studies have shown, however, that botulinum toxin injec�ons in the abdominal muscle prior to repair may lead to relaxa�on of the abdominal wall and reduce the need for surgical component separa�on. In a case series of pa�ents undergoing IPOM repair or laparoscopic-open-laparoscopic ventral hernia repair, preopera�ve administra�on of botulinum toxin was found to be safe, and that it increased abdominal volume and lengthened laterally retracted abdominal muscles [8].In many respects, the development of these new endoscopic repair techniques for ventral hernia repair has led to the rediscovery of the same anatomical spaces that were used for open ventral repair prior to the introduc�on of the IPOM technique. Since these procedures retain func�onal anatomy using mesh to reinforce the abdominal wall rather than to overbridge the defect, they have more favorable outcomes than laparoscopic IPOM repair, with smooth and rapid postopera�ve recovery as well as reduced risk for long-term postopera�ve pain.The greatest poten�al benefit from the transfer from IPOM and onlay repairs is probably the improved biomechanical proper�es from using mesh with sufficient size and placing them in an appropriate posi�on. By providing access to anatomical spaces other than the abdominal cavity or the prefascial space they enable placing the mesh with sizes adapted to the defect size and fixing them in a way that the �ssue tension is adjusted to the func�onal anatomy of the abdominal wall, thereby reducing the risk of persis�ng pain and recurrences [8]. Future studies should focus on evalua�ng ventral hernia surgery in terms of biomechanical strength and resistance to the forces caused by cyclic loading of the abdominal wall.
No takes yet. Share an insight, caveat, or question.
Gabriel Sandblom (2023) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: