A. C. de Beaux

Clinical prognostic scoring system to aid decision-making in gastro-oesophageal cancer

Accurate prediction of prognosis in gastro‐oesophageal cancer remains challenging. The aim of this study was to develop a robust model for outcome prediction.

Laparoscopic ventral hernia repair.

Laparoscopic repair of ventral hernias (primary anterior abdominal wall and incisional hernias not including the groin) has become increasingly popular since its description in 19931. The steps of the procedure are easy to describe: insert the camera port well away from the site of the hernia, clear the anterior abdominal wall of adhesions, lay on an intraperitoneal mesh with generous overlap in all directions, then fix it in place with sutures, tacks or glue. The apparent simplicity of the operation and the intuitive appeal of a ‘keyhole’ approach have been the main drivers behind an explosion in the use of this technique. Despite this rapid and widespread uptake, fewer than 1000 patients have been reported in randomized trials comparing laparoscopic with open techniques2. Surgeons new to the technique have been encouraged by these studies, which have consistently reported a lower incidence of wound infection, less postoperative pain, shorter hospital stay and faster return to full activity with laparoscopic repair. Early recurrences are uncommon and generally reflect technical error, but with longer follow-up there is mounting evidence of late recurrences3 resulting from the slow process of mesh migration and/or mesh contracture4,5. Thus, the overall recurrence rate seems no better than with open surgery, and may well be worse. Although postoperative complications are fewer after laparoscopic repair, those that do occur tend to be more serious. A recent study comparing laparoscopic and open methods reported that 23 per cent of complications in the laparoscopic group required surgical correction (compared with 7 per cent in the open group) and that 5 per cent were lifethreatening (compared with none in the open group)6. In addition, a number of case reports have highlighted new complications unique to the laparoscopic approach, including bowel obstruction, visceral erosion and fistula related to the intraperitoneal mesh, and rectus sheath haematoma and cardiac tamponade from tack penetration7. Dense adhesions to the mesh can also compromise the success of subsequent laparoscopic surgery for other conditions. As operative strategies evolve, there remain unanswered questions. Who is suitable? Which mesh should be used, and what size? How best to fix it? Should the defect be closed? Are there medicolegal consequences? Despite a lack of experimental evidence to indicate that patients fare better with laparoscopic repair, there is general agreement that all but the largest and smallest ventral hernias are potentially operable this way. The presence of multiple defects and obesity tips the balance in favour of the laparoscopic approach. For defects less than 2 cm in diameter, some would argue that an open suture repair is adequate, especially if it is achievable under local anaesthetic. For very large defects it is difficult to stop the mesh from bulging into the defect, and there may be inadequate tissue peripherally to which the mesh can be securely fixed8. When intra-abdominal forces exceed fixation strength, the mesh slides progressively into the defect (mesh migration)4,5. Emergency laparoscopic repair is feasible, but current opinion advises against this in the presence of strangulated bowel. Even during an elective repair, although minor small bowel perforations with minimal contamination may be repaired laparoscopically before proceeding as planned, intraoperative faecal soiling mandates abandoning the operation and/or converting to an open procedure9. Cosmesis is often important. From the patient’s point of view, abdominal wall distortion and deformity may be the drivers for surgical repair. Several small laparoscopic incisions may be intuitively more cosmetic than a single large one, but patients with a significant bulge and redundant skin may be better served by an open repair with excision of redundant skin, in some cases extending to a formal abdominoplasty. Most surgeons accept that ‘naked’ polypropylene mesh in the abdominal cavity is no longer acceptable owing to the long-term risk of adhesion formation. These adhesions can be dense and are associated with an increased risk of small bowel resection at any subsequent abdominal surgery, because of difficulty in gaining access to the abdominal cavity10. Newer ‘antiadhesion’ meshes, designed specifically for intraperitoneal use, do result in fewer adhesions11 but cannot be described as ‘non-stick’. The optimal size of mesh remains contentious. A minimum overlap of 5 cm all around the hernia defect has been recommended12, but this is an

Authors' reply: Laparoscopic ventral hernia repair (Br J Surg 2012; 99: 1319-1321).

Sir We thank Dr Aellen for his comments to our editorial. We note his concerns regarding the risks of laparoscopic ventral hernia repair (LVHR) and feel they are similar to our own. We also support his view that a retromuscular sublay open repair is a good operation, but we must acknowledge that there is little evidence to support the view that it is the preferred operation in the majority of patients. The majority of, although not all, studies comparing the open approach with LVHR report fewer wound complications and mesh infections with the laparoscopic approach, along with a shorter hospital stay and faster return to full activity. These advantages must be weighed up against the shortand long-term risks associated with placing mesh within the peritoneal cavity, well described by Dr Aellen. The medicolegal side is interesting. Although the majority of cases referred to the authors for medicolegal opinions have no case to answer with respect to negligence, what is clear is that the number of cases referred following laparoscopic surgery exceeds that following open surgery, despite open repairs being performed more often than laparoscopic repairs in the UK. Although we are advocates of LVHR in selected patients, with the right mesh and state-of-the-art technique, we make a plea in our editorial for more information, as there remain too many unanswered questions regarding the laparoscopic approach. Information can be gathered by careful audit of our results and combining the results from many centres. We believe the online platform developed by a European working party has much to recommend it in facilitating this end (the platform can be viewed at http://www.eurahs.eu). A. de Beaux and B. Tulloh Department of General Surgery, Royal Infirmary of Edinburgh, Edinburgh EH16 4SA, UK (e-mail: adebeaux@doctors.org.uk) DOI: 10.1002/bjs.9080

Simulation-based training and learning curves in laparoscopic Roux-en-Y gastric bypass (Br J Surg 2012; 99: 887–895)

Background: Ex vivo simulation-based technical skills training has been shown to improve operating room performance and shorten learning curves for basic laparoscopic procedures. The application of such training for laparoscopic Roux-en-Y gastric bypass (LRYGBP) has not been reviewed. Methods: Relevant studies were identified by one author from a search of MEDLINE and Embase databases from 1 January 1994 to 30 November 2010. Studies examining the learning curves and ex vivo training methods for LRYGBP were included; all other types of bariatric operations were excluded. A manual search of the references was also performed to identify additional potentially relevant papers. Results: Twelve studies (5 prospective and 7 retrospective case series) were selected for review. The learning curve for LRYGBP was reported to be 50–100 procedures. Bench-top laparoscopic jejunojejunostomy, anaesthetized animals and Thiel human cadavers made up the bulk of the reported models for ex vivo training. Most studies were of relatively poor quality. An evidence-based ex vivo training curriculum for LRYGBP is currently lacking. Conclusion: Better quality studies are needed to define the learning curve for LRYGBP. Future studies should focus on the design and validation of training models, and a comprehensive curriculum for training and assessment of cognitive, technical and non-technical components of competency for laparoscopic bariatric surgery.

Morbidity and mortality before and after bariatric surgery for morbid obesity compared with the general population (Br J Surg 2011; 98: 811–816)

1Upper Gastrointestinal Research, Department of Molecular Medicine and Surgery, and 2Department of Public Health Sciences, Karolinska Institutet, and 3Department of Clinical Sciences, Danderyd Hospital, Karolinska Institute, Stockholm, Sweden, and 4Division of Cancer Studies, King’s College of Medicine, London, UK Correspondence to: Dr M. Plecka Östlund, Upper Gastrointestinal Research, Norra Stationsgatan 67, SE-171 76 Stockholm, Sweden (e-mail: magdalena.plecka-ostlund@sodersjukhuset.se)