Validation of laparoscopic surgical skills training outside the operating room: a long road.

BACKGROUND: Surgical skills training outside the operating room is beneficial. The best methods have yet to be identified. The authors aimed to document the predictive validity of simulation training in three different studies. METHODS: Study 1 was a prospective, randomized, multicenter trial comparing performance in the operating room after training on a laparoscopic simulator and after no training. The Global Operative Assessment of Laparoscopic Skills (GOALS) was used to evaluate operative performance. Study 2 retrospectively reviewed the operative performance of junior residents before and after implementation of a laparoscopic skills training curriculum. Operative time was the variable used to determine resident improvement. Study 3 was a prospective, randomized trial evaluating intern operative performance of laparoscopic cholecystectomy in a porcine model before and after training on a simulator. Operative performance was assessed using GOALS. RESULTS: All three studies failed to demonstrate predictive validity. With GOALS used as the assessment tool, no difference was found between trained and untrained residents in studies 1 and 3. In study 2, the trained group took significantly longer to complete a laparoscopic cholecystectomy than the untrained group. CONCLUSIONS: No correlation was found between the three types of training outside the operating room, and no improved operative performance was observed. Possible explanations include too few subjects, training introduced too late in the learning curve, and training criteria that were too easy. Additionally, simulator training focuses on precision, which may actually increase task time. Awareness of these issues can improve the design of future studies.

Tissue ingrowth and bowel adhesion formation in an animal comparative study: polypropylene versus Proceed versus Parietex Composite.

BACKGROUND: The optimal prosthesis for laparoscopic ventral hernia repair would combine excellent parietal surface tissue ingrowth with minimal visceral surface adhesiveness. Currently, few data are available from randomized trials comparing the commercially available prostheses. METHODS: In a pig model designed to incite adhesions, three 10 x 15-cm pieces of mesh (Proceed, Parietex Composite [PCO], and polypropylene [PPM]) were randomly positioned intraperitoneally in each of 10 animals using sutures and tack fixation. After a 28-day survival, the amount of shrinkage, the area and peel strength of visceral adhesions, the peak peel strength, the work required to separate mesh from the abdominal wall, and a coefficient representing the adhesiveness of tissue ingrowth were averaged for each type of mesh and then compared with the averages for the other prostheses. The histologic appearance of each prosthesis was documented. RESULTS: Proceed had more shrinkage (99.6 cm2) than PCO (105.8 cm2) or PPM (112 cm2), although the difference was not statistically significant. The mean area of adhesions to PCO (11%) was significantly less than for Proceed (48%; p < 0.008) or PPM (46%; p < 0.008). Adhesion peel strength was significantly less for PCO (5.9 N) than for Proceed (12.1 N; p < 0.02) or PPM (12.9 N; p < 0.02). According to a filmy-to-dense scale of 1 to 5, adhesions were more filmy with PCO (1.7) than with PPM (2.9) or Proceed (3.7) (p < 0.007). Peak peel strength from the abdominal wall was significantly higher for PCO (17.2 N) than for Proceed (10.7 N) or PPM (10 N; p < 0.002). The histology of each prosthesis showed a neoperitoneum only with PCO. CONCLUSIONS: With less shrinkage, fewer and less dense adhesions to the viscera, and significantly stronger abdominal wall adherence and tissue ingrowth at 28 days in this animal study, PCO was superior to both Proceed and PPM in all categories. Furthermore, PCO demonstrated all the favorable qualities needed in an optimal prosthesis for laparoscopic ventral hernia repair, including the rapid development of a neoperitoneum.

A comparative study of adhesion formation and abdominal wall ingrowth after laparoscopic ventral hernia repair in a porcine model using multiple types of mesh.

BACKGROUND: The ideal mesh for laparoscopic ventral hernia repair is not yet identified. METHODS: We laparoscopically placed polypropylene (PPM), expanded polytetrafluoroethylene (ePTFE), and polyester with antiadhesive collagen layer (PCO) in eight pigs using sutures and tacks for fixation. After 28-day survival, we compared adhesion formation, fibrous ingrowth, and shrinkage among the types of mesh. RESULTS: Mean area of adhesions to PCO (8.25%) was less than that to ePTFE (57.14%, p < 0.001) or PPM (79.38%, p < 0.001). Adhesions peel strength was less for PCO (2.3 N) than for PPM (16.1 N, p < 0.001) or ePTFE (8.8 N, p = 0.02). Peel strength of mesh from the abdominal wall was less for ePTFE (1.3 N/cm of mesh width) than for PCO (2.8 N/cm, p = 0.001) or PPM (2.1 N/cm, p = 0.05). ePTFE area (94.4 cm(2)) was less than that for PCO (118.6 cm(2), p < 0.001) or PPM (140.7 cm(2), p < 0.02). CONCLUSION: PCO had fewer and less severe adhesions than ePTFE or PPM while facilitating excellent ingrowth of the adjacent parietal tissue.

Construct validity for the LAPSIM laparoscopic surgical simulator.

BACKGROUND: The skills required for laparoscopic surgery are amenable to simulator-based training. Several computerized devices are now available. We hypothesized that the LAPSIM simulator can be shown to distinguish novice from experienced laparoscopic surgeons, thus establishing construct validity. METHODS: We tested residents of all levels and attending laparoscopic surgeons. The subjects were tested on eight software modules. Pass/fail (P/F), time (T), maximum level achieved (MLA), tissue damage (TD), motion, and error scores were compared using the t-test and analysis of variance. RESULTS: A total of 54 subjects were tested. The most significant difference was found when we compared the most (seven attending surgeons) and least experienced (10 interns) subjects. Grasping showed significance at P/F and MLA (p < 0.03). Clip applying was significant for P/F, MLA, motion, and errors (p < 0.02). Laparoscopic suturing was significant for P/F, MLA, T, TD, as was knot error (p < 0.05). This finding held for novice, intermediate, and expert subjects (p < 0.05) and for suturing time between attending surgeons and residents (postgraduate year [PGY] 1-4) (p < 0.05). CONCLUSIONS: LAPSIM has construct validity to distinguish between expert and novice laparoscopists. Suture simulation can be used to discriminate between individuals at different levels of residency and expert surgeons.

Comparison of two composite meshes using two fixation devices in a porcine laparoscopic ventral hernia repair model.

INTRODUCTION: Laparoscopic ventral hernia repair uses tacks to secure mesh. The mesh is designed to maximize tissue ingrowth while minimizing adhesions. We hypothesized: (1) a collagen-coated polyester mesh (PCO) will form fewer adhesions than an ePTFE-polypropylene composite (BC) and (2) absorbable tacks are equivalent to metal tacks. METHODS: In a porcine model of adhesion formation, three pieces of 10x15-cm mesh were placed on the anterior abdominal wall. PCO was secured with absorbable (PLA) or metal tacks (PT), BC with PT. At 28 days, adhesion formation, abdominal-wall adherence, and tissue ingrowth were analyzed. RESULTS: PCO induced fewer adhesions (14.5% vs 53.4%, P = 0.007). On an adhesion scale (0 5), BC scored 3.6 vs 1.75 for PCO (P < 0.03). There was no difference in adhesion strength, tack adhesions, or abdominal-wall peel force. Histology showed equal ingrowth. CONCLUSIONS: PCO induces fewer adhesions than BC. There is no difference in the ingrowth of the two mesh types. The PLA achieves equivalent mesh incorporation to the PT.

Hand-assisted laparoscopic splenectomy for giant spleens.

BACKGROUND: Laparoscopic splenectomy for massive splenomegaly is technically difficult, and the morcellated splenic tissue may be inadequate for histologic study. A hand-assisted technique may provide a technical advantage and allow removal of larger pieces of spleen. METHODS: Patients who underwent hand-assisted laparoscopic splenectomy for massive splenomegaly were reviewed. Demographic information, operative data, and outcomes data were tabulated. RESULTS: Sixteen patients met these criteria. Mean age was 56 years (range, 35-78 years). Operating time averaged 240 min (range, 165-360 min), and median blood loss was 425 cc (range, 100-1800 cc). There were no conversions to an open procedure. Mean weight of extracted spleens was 2008 g (range, 543-4090 g). Postoperative length of stay averaged 3.3 days (range, 2-7 days). There was one postoperative complication (6.25%) and no mortality. CONCLUSIONS: Hand-assisted laparoscopic splenectomy for massive splenomegaly is feasible and safe while preserving the recovery benefits of minimal access surgery. It provides an adequate specimen for histologic study.

The use of a personal digital assistant for wireless entry of data into a database via the Internet.

BACKGROUND: Researchers typically record data on a worksheet and at some later time enter it into the database. Wireless data entry and retrieval using a personal digital assistant (PDA) at the site of patient contact can simplify this process and improve efficiency. METHODS: A surgeon and a nurse coordinator provided the content for the database. The computer programmer created the database, placed the pages of the database on the PDA screen, and researched and installed security measures. RESULTS: Designing the database took 6 months. Meeting Health Insurance Portability and Accountability Act of 1996 (HIPAA) requirements for patient confidentiality, satisfying institutional Information Services requirements, and ensuring connectivity required an additional 8 months before the functional system was complete. CONCLUSIONS: It is now possible to achieve wireless entry and retrieval of data using a PDA. Potential advantages include collection and entry of data at the same time, easy entry of data from multiple sites, and retrieval of data at the patient's bedside.