JOURNAL OF LAPAROENDOSCOPIC & ADVANCED SURGICAL TECHNIQUES Volume 24, Number 4, 2014 ª Mary Ann Liebert, Inc. DOI: 10.1089/lap.2013.0358
Evaluation of a Bochdalek Diaphragmatic Hernia Rabbit Model for Pediatric Thoracoscopic Training Jesu´s Uso´n-Casau´s, DVM, PhD,1 Eva Marı´a Pe´rez-Merino, DVM, PhD,1 Ramo´n Rivera-Barreno, DVM, PhD,2 Carlos A. Rodrı´guez-Alarco´n, DVM, PhD,2 and Francisco M. Sa´nchez-Margallo, DVM, PhD 3
Abstract
Background/Aim: This study evaluated the usefulness of a Bochdalek hernia rabbit model as a tool for advanced thoracoscopic training, teaching the specific skills required for thoracoscopic repair of congenital diaphragmatic hernia. Materials and Methods: An incision was made in the Bochdalek triangle of 25 New Zealand rabbits (weighing 3–3.5 kg) to induce an experimental diaphragmatic hernia. At 72 hours later, a thoracoscopic repair of the hernia as described for newborns was performed by 25 pediatric surgeons divided into two groups: expert and novice. The tasks assessed were organ relocation and diaphragm suture. A visual analog scale was used to evaluate technical performance. The objective performance measure was completion time. Complications were recorded, and suture quality was scored. The surgeons evaluated the model by completing a questionnaire, grading items on a 5-point scale. Results: All 25 animals developed a diaphragmatic hernia with protrusion of the intestine into the thoracic cavity. Expert trainees had significantly shorter completion times and better performance scores than novices. Experts also received higher scores for suture quality. Five novices caused perforations or bleeding, but no experts did. The surgeons rated the model positively, highlighting the similarities between the model and newborn hernias and its usefulness for pediatric training programs. Conclusions: The Bochdalek hernia rabbit model can be used to detect different levels of experience in pediatric thoracoscopy. This realistic and easily reproducible model can help to perfect thoracoscopic skills in a realistic recreation of a pediatric Bochdalek hernia repair. Introduction
T
horacoscopy is the third most frequent approach for repairing neonatal congenital diaphragmatic hernia (CDH), behind laparotomy and thoracotomy.1 To embrace this technique and ensure it is safe for pediatric patients, adequate training is essential. However, individual centers and surgeons may have limited experience with congenital diaphragmatic hernias and even less experience with the minimally invasive surgical (MIS) approach.2,3 There are many levels of laparoscopic training that surgeons must master. Practice in both the simulation laboratory and the animal operating room should, ideally, precede the human operating room. Low-fidelity models (box trainers) are well suited for novice learners. These inanimate models provide a reliable and reproducible method for learning instrument-handling maneuvers and mastering basic laparo1 2 3
scopic skills such as suturing and knot tying.4 However, the clinical use of minimally invasive surgery should be preceded by a training period with a model that closely resembles human conditions.5 Even for learning isolated mechanical tasks it has been shown that training can be provided using a pelvic trainer or animal model with similar effects, but in vivo performance of the same task requires secondary surgical skills, which are conveyed only during live training with greater success. Consequently, stepwise teaching with both modules seems reasonable before these procedures are approached in neonates or small children.6 The development of specific surgical training models that mimic the human situation can enhance advanced laparoscopic skills and facilitate the development of surgical skills for rare diseases. Several rabbit models for different endosurgical pediatric procedures have been described,7 but there is no specific animal model for thoracoscopic training
Department of Animal Medicine and Surgery, Faculty of Veterinary Science, University of Extremadura, Ca´ceres, Spain. Department of Veterinary Sciences, Institute of Biomedical Sciences, Autonomous University of Juarez, Ciudad Juarez, Mexico. Department of Laparoscopic Surgery, Jesu´s Uso´n Minimally Invasive Surgery Centre, Ca´ceres, Spain.
280
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
281
on diaphragmatic hernias. Our aim is to determine the value of this experimental Bochdalek hernia rabbit model (BHRM) for pediatric thoracoscopic training. Materials and Methods
Prior to the in vivo procedures, rabbit cadavers were used to identify the Bochdalek triangle and determine the best place to create the diaphragmatic defect (Fig. 1) and the optimum positions for trocars. Diaphragmatic hernia induction
Twenty-five New Zealand white rabbits weighing 3–3.5 kg were used. The protocol was approved by the Ethical Committee for Animal Research of the Jesu´s Uso´n Minimally Invasive Surgery Centre ( JUMISC), Ca´ceres, Spain. Animal care complied with the Guide for the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, Commission on Life Sciences, U.S. National Research Council. The rabbits were premedicated with subcutaneous buprenorphine (0.050 mg/kg) and meloxicam (0.20 mg/kg). Anesthesia was induced with dexmedetomidine (0.015 mg/kg) and ketamine (8 mg/kg) through an intravenous catheter in the marginal ear vein and maintained with propofol (1–2 mg/kg). A midline minilaparotomy (4 cm) was performed. To induce the diaphragmatic defect, an incision (3 cm) was made in the dorsal costal area of the left diaphragm, at the small triangle between the pars lumbalis and pars costalis and the retractor costae muscle. An oval-shaped defect was created. A chest tube was inserted through the defect and brought out through a laparotomy, with a purse-string stitch around it. The laparotomy was routinely closed. Before the animal regained consciousness, the pneumothorax was aspirated, the tube was removed, and the wound was stitched closed. Thoracoscopic repair
At 72 hours after the hernia was induced, the animals were premedicated with intravenous atropine (0.05 mg/kg) and
FIG. 2.
Correct positioning of the rabbit.
midazolam (0.25 mg/kg). Anesthesia was induced using intravenous propofol (10 mg/kg). Endoscopically guided intubation was performed, and anesthesia was maintained using mechanical ventilation with 1.5% isoflurane and fentanyl (0.01 mg/kg every 30 minutes). The rabbits were positioned in the reverse Trendelenburg position and right lateral decubitus position on a specially designed operating table made of expanded polystyrene (length · width · height, 50 cm · 12 cm · 15 cm) so that the rabbit’s head could be raised to an angle of 20 (Fig. 2). The 5-mm port for the optic was positioned in the fourth left intercostal space, under the caudal border of the scapula. The left upper limb was stretched up toward the head and secured with adhesive strips to free up as much chest surface as possible. Two 3.5-mm ports for the instruments were positioned in the sixth left intercostal space, one 1.5 cm from the spine and the other 2 cm from the sternum (Fig. 3). The lung was collapsed by an intrapleural carbon dioxide insufflation (2–4 mm Hg). During the thoracoscopy, the herniated abdominal organs were relocated to the peritoneal cavity by pushing with two endoscopic graspers, and the diaphragmatic defect was repaired with five stitches using the intracorporeal technique with interrupted 5-0 glyconate suture and a laparoscopic needle. Model assessment
After Institutional Review Board approval was obtained, thoracoscopic repair of experimental hernias was performed
FIG. 1. Determination of the optimal location of the diaphragmatic defect in the cadaver.
FIG. 3. Placement of trocars and instrument during thoracoscopy. The black arrow indicates the sternum.
282
by 25 pediatric surgeons during an intensive 3-day hands-on pediatric thoracoscopy course at the JUMISC. The novice group was made up of 20 trainees (pediatric residents, postgraduate year [PGY]-1–4) divided into 10 two-person teams, with each person alternating between the role of surgeon and assistant for each surgery. Before operating on the animals, all novices were trained in basic laparoscopic skills using a box trainer (SIMULAP-IC05; CCMIJU, Ca´ceres, Spain). Five pediatric surgeons with advanced minimally invasive surgery training and previous clinical thoracoscopy experience were recruited as an experienced group. After the trocars were correctly positioned, two tasks were assessed: repositioning of abdominal organs and diaphragmatic suture. The assessment tools included a visual analog scale (VAS) to measure overall technical competence at each task. The scale used is a 10-cm line with descriptive anchors at each end. The anchors were ‘‘Unable to complete the task with maximum guidance’’ and ‘‘Could perform the task safely and independently (fully competent).’’ The evaluator wrote an ‘‘X’’ on the line to indicate the score he or she assigned for that particular item. The scores were determined numerically by measuring where along the 10-cm line that the mark was placed. The participants were evaluated by two trained observers: an attending pediatric surgeon and a trainer in minimally invasive surgery at the JUMISC. The average score from the two reviewers was calculated and used in the final data analysis. Completion time was the objective performance measure for both tasks. After the procedure, the animals were euthanized, and their abdominal and thoracic organs were carefully examined to detect any severe hemorrhage, visceral injuries, or intestinal perforation. The diaphragms were also removed to evaluate final suture quality. Correct suture was considered to be five stitches spaced 5 – 1 mm, each with a square knot tied by four alternating throws, and without tears. Each correct stitch was given 3 points (suture maximum score, 15 points). Tears, incorrect tension, or incorrect distance was penalized by deducting 1 point each. At the end of the course, the surgeons were given a questionnaire consisting of six statements about how realistic the model was and its usefulness for training. These were presented on a 5-point ordinal scale (1 = not at all, 2 = a little, 3 = enough, 4 = a lot, and 5 = completely). The statistical software package SPSS version 15.0 (SPSS, Inc., Chicago, IL) was used for the statistical analysis. Because of the smaller sample size and the non-normality of the data, a two-tailed Mann–Whitney U test was used to evaluate differences between the two groups for the different parameters (time, VAS performance, and final suture quality). P < .05 was considered statistically significant. Intraclass correlation coefficients (ICCs) were used to assess inter-rater reliability of the VAS for competence.
´ N-CASAU ´ S ET AL. USO
FIG. 4. Operative view of the diaphragmatic hernia in the rabbit model. operating table were optimal, allowing an excellent diaphragmatic approach, and repair was achieved successfully in all animals (Fig. 5). Although carbon dioxide insufflation facilitated the reduction of the abdominal viscera, this technique was avoided as much as possible to force the trainees to use their tissue-handling skills. Table 1 provides a summary of the results of the different procedure assessments. The VAS scores for competence show that the experienced group performed better in both tasks. The total ICC for VAS-hernia reduction was 0.79 (95% confidence interval, 0.61–0.91), and the total ICC for VASsuture was 0.81 (0.95% confidence interval, 0.63–0.94). The overall time for both the tasks combined ranged from 28 minutes for the most expert surgeon to 81 minutes for a novice, and the expert group completed both tasks in a significantly shorter time. The expert participants continued to achieve higher scores for suture quality. Five of the novices caused perforations (three PGY-3 and one PGY-4) or hemorrhage (one PGY-4), but no experts made these mistakes.
Results
The original 3-cm diaphragmatic incision was sufficient to cause herniation with protrusion of several intestinal loops and the stomach into the thoracic cavity. The original incision increased in size over the 72-hour period because of progressive herniation of the peritoneal organs (Fig. 4). The positioning of the trocars and of the animals on the special
FIG. 5. Appearance of the diaphragm defect with partial diaphragmatic suture after relocation of the organs.
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
Table 1. Results of Different Assessment Tools for the Procedures by Experience Group Assessment
Novice (n = 20)
Experts (n = 5)
Completion time (minutes) Hernia reduction 24.4 – 5.7 14.1 – 0.6 Diaphragmatic suture 27.8 – 5.9 16.1 – 0.4 VAS Hernia reduction 4.3 – 2.7 9.1 – 2.1 Diaphragmatic suture 3.7 – 2.1 8.9 – 2.4 Final suture quality (out of 15) 6.3 – 2.2 10.4 – 1.0 Perforations/hemorrhage (n) 4/1 0/0
P .0008 .0006 .0002 .0002 .0002
Data are mean – standard deviation values unless indicated otherwise. VAS, visual analog scale.
Table 2 shows the mean ratings for the trainees’ first impression of the BH-RM. For both groups, the highest mean score was for the question about the usefulness of this model for pediatric thoracoscopic training. The second highest rating was for the capacity of the model to improve thoracoscopic skills, and the third for the similarities between the model and a newborn CDH. The lowest mean score was for the capacity of the model to facilitate management of complications in CDH surgery. The statistical analysis revealed no significant difference between the scores of the two groups (P > .3), indicating a general agreement on all questions. Discussion
Thoracoscopy has proved to be a feasible and safe surgical approach under two conditions: for carefully selected patients and for pediatric surgeons with sufficient training to perform the MIS procedure.8,9 Several diaphragmatic models have been used to measure different aspects of this congenital defect. However, we were unable to find a live diaphragmatic hernia model for thoracoscopic training reported in the literature. Many years of experience at the JUMISC have shown that both simulators and animal models should be used in laparoscopic training. The JUMISC has a fully validated training method based on a pyramid structure with four levels: (1) basic and advanced skills with box trainers, (2) anatomical protocols and advanced skills with animal models, (3) advanced procedural skills with telesurgical applications, and (4) practice in the operating room. This curriculum has driven the development of the different animal models, conceived as
283
a training method for the second level in the pyramid.5,10 This program has been used in pediatric and neonatal training, obtaining very high ratings over the last few years. In order to choose the most suitable animal model for each specialty, aspects relating to comparative anatomy and size have been taken into account, to ensure that the training models are as similar to humans as possible. For example, pigs are used as a model for digestive and urological laparoscopy, and rabbits are used for neonatal laparoscopy. Surgical techniques included in the training program are tested by pediatric surgery specialists and selected based on their feasibility, usefulness, and realism.11,12 Recently, attempts have been made to replace live animals in pediatric laparoscopic training with high-fidelity models that recreate procedures such as thoracoscopic esophageal atresia/tracheoesophageal fistula13 or CDH14 repair. The cost of one BH-RM (rabbit + surgeries + 72 hours of care = e200) is similar to that of those in vitro simulators. However, the latter are obviously reusable and do not require facilities and veterinary staff for housing and care of the animals. There are also still legal and ethical concerns about the use of animals for single-use training labs. However, at present, the weakest physical attributes of the diaphragmatic hernia simulator are the realism of the tissues, including the skin, diaphragm, and intestine, and the poor haptic feedback of the synthetic materials.14 In contrast, our questionnaire scores highlight the positive response of the surgeons with regard to the similarity of the BH-RM to real-life scenarios and its usefulness. For training in handling tissues, in vitro models are poor substitutes for in vivo models because pulsating vessels, bleeding, and control of bleeding are essential to surgery. The BH-RM exactly replicates the classic Bochdalek hernia, and laparoscopic surgery can be performed in a real environment with similar instrument size, port placement, and rabbit position to those described for thoracoscopy in newborns.15–17 Furthermore, the model not only allows tissue handling of varying degrees of difficulty, but also makes it possible to practice several other laparoscopic tasks such as cutting, coagulation, stitching, and intracorporeal knot tying. Once the laparoscope was inserted, the surgeons pointed out how similar the model was to what they see thorascopically in classic cases of newborn CDH. The large cecum size in the rabbit and the huge amount of herniated intestine mean that the required reduction maneuvers are very similar in both cases. Various authors have suggested that piglets may provide the optimum animal model for neonatal thoracoscopy because the fragility of the rabbit coupled with a limited intrathoracic working space limits its use for this
Table 2. Observed Results for Bochdalek Hernia Rabbit Model Assessment Score Item Similarities between BH-RM and newborn CDH Similarities between rabbit thoracic dimension and newborn thorax Degree of technical difficulty compared with real scenario Training capacities of model to improve thoracoscopic skills Training capacities of model to facilitate management of CDH surgical complications Degree of usefulness of this model in a training program in pediatric thoracoscopy surgery
Novice
Expert
P
4.10 – 0.68 3.84 – 0.62 4.00 – 0.36 4.20 – 0.50 3.44 – 0.94 4.29 – 0.81
3.90 – 0.31 3.27 – 0.52 3.76 – 0.42 4.50 – 0.63 3.05 – 0.83 4.55 – 0.85
.716 .317 .628 .815 .496 .856
All items scored on a scale range of 1–5 points, where 1 = not at all, 2 = a little, 3 = enough, 4 = a lot, and 5 = completely. Data are mean – standard deviation values. BH-RM, Bochdalek hernia rabbit model; CDH, congenital diaphragmatic hernia.
´ N-CASAU ´ S ET AL. USO
284
technique.18,19 However, the similar proportions of rabbits and newborns and the rabbit model’s value for laparoscopic pediatric training are well established.6,7,20 Although most of the surgeons noticed that the rabbit thorax is flatter whereas the newborn thorax is more rounded, the intrathoracic working space is similar. We agree that rabbit tissue is very fragile, but it ensures that surgeons use precise and careful maneuvers, acquiring an extremely refined technique. Handling intestinal loops in a small thoracic space and pulling them through a diaphragmatic defect is an excellent exercise that requires precise, careful movements. Diaphragmatic suturing was chosen because it is a fundamental MIS skill and because traditionally the optimal neonatal candidate for a CDH thoracoscopic repair would be one in whom the diaphragm could be repaired primarily.8 Both are well-defined procedures that allow objective evaluation (time, presence/absence of perforation/hemorrhage, suture quality) and an overall competence score (VAS). In the future, our aim is to develop a valid procedure-specific tool for this surgical procedure, similar to the modified GOALS tool used in the assessment of laparoscopic incisional or inguinal hernia repair.21,22 Thoracoscopy for recurrent herniation and the use of a diaphragmatic patch are other real scenarios23,24 that we plan to include in the model in the future. The BH-RM provided a clear distinction between expert and novice groups. Novices with some experience handling laparoscopic tools carried out both techniques (relocation of organs and suturing) better and faster than the novice subjects with very little laparoscopic experience. Perforation and hemorrhage occurred most frequently among the novices, especially in senior trainees, probably because of a more aggressive technique. This demonstrates the need for training with real tissues before operating on patients. Although the surgeons believed that thoracoscopic skills could be substantially improved, the model’s usefulness in the management of surgical complications received the lowest score. The expert surgeons reasoned that most major surgical complications (such as intestinal perforation) involve moving from thoracoscopy to open surgery, and, moreover, conditions such as lung hypoplasia or pulmonary hypertension are not considered in this model. In conclusion, we have developed an inexpensive, realistic, and easily reproducible, in the short term, diaphragmatic hernia model, to be used to hone thoracoscopic skills and develop specific skills required by surgeons to repair Bochdalek CDH in newborns using the thoracoscopic approach. This model could be included in pediatric MIS training courses and pediatric surgery fellowship programs as a further step toward excellence in MIS skills.
2.
3. 4. 5.
6. 7.
8. 9. 10. 11. 12.
13.
14.
15.
Acknowledgments
This study was supported by Junta de Extremadura (PRI 08A105). Disclosure Statement
No competing financial interests exist.
16.
17. 18.
References
1. Tsao K, Lally PA, Lally KP; Congenital Diaphragmatic Hernia Study Group. Minimally invasive repair of con-
19.
genital diaphragmatic hernia. J Pediatr Surg 2011;46:1158– 1164. Cho SD, et al. Analysis of 29 consecutive thoracoscopic repairs of congenital diaphragmatic hernia in neonates compared to historical controls. J Pediatr Surg 2009;44: 80–86. Nguyen TL, Le AD. Thoracoscopic repair for congenital diaphragmatic hernia: Lessons from 45 cases. J Pediatr Surg 2006;41:1713–1715. Molinas CR, Binda MM, Mailova K, Koninckx PR. The rabbit nephrectomy model for training in laparoscopic surgery. Hum Reprod 2004;19:185–190. Uso´n J, Sa´nchez FM, Sa´nchez MA, Pe´rez FJ, Hashizume M. Principios ba´sicos. In: Uso´n J, Sa´nchez FM, Pascual S, Climent S (eds.). Formacio´n en Cirugı´a Laparosco´pica Paso a Paso, 3rd ed. Ca´ceres, Spain: Centro de Cirugı´a de Mı´nima Invasio´n, 2007:21–89. Heinrich M, Tillo N, Kirlum HJ, Till H. Comparison of different training models for laparoscopic surgery in neonates and small infants. Surg Endosc 2006;20:641–644. Kirlum HJ, Heinrich M, Till H. Rabbit model serves as a valuable operative experience and helps to establish new techniques for abdominal and thoracic endosurgery. Pediatr Surg Int 2005;21:91–93. Yang EY, et al. Neonatal thoracoscopic repair of congenital diaphragmatic hernia: Selection criteria for successful outcome. J Pediatr Surg 2005;40:1369–1375. Okazaki T, et al. Indications for thoracoscopic repair of congenital diaphragmatic hernia in neonates. Pediatr Surg Int 2011;27:35–38. Uso´n-Gargallo J, et al. Modelo de formacio´n piramidal para la ensen˜anza de cirugı´a laparosco´pica. Cir Cir 2013;81: 420–430. Pe´rez-Duarte FJ, et al. Design and validation of a training model on paediatric and neonatal surgery. Cir Pediatr 2012;25:121–125. Pe´rez-Duarte, FJ, et al. Validation of the lagomorph animal model for training and skills acquisition on neonatal laparoscopic surgery. Presented at the 24th Conference of the Society for Medical Innovation and Technology, Barcelona, 2012. www.smit.de/typo3/index.php?id = 90 (accessed November 30, 2013). Barsness KA, Rooney DM, Davis LM. Collaboration in simulation: The development and initial validation of a novel thoracoscopic neonatal simulator. J Pediatr Surg 2013;48:1232–1238. Barsness KA, Rooney DM, Davis LM. The development and evaluation of a novel diaphragmatic hernia repair simulator. J Laparoendosc Adv Surg Tech A 2013;23:714– 718. Arca MJ, et al. Early experience with minimally invasive repair of congenital diaphragmatic hernias: Results and lessons learned. J Pediatr Surg 2003;38:1563–1568. Gomes Ferreira C, et al. Neonatal minimally invasive surgery for congenital diaphragmatic hernias: A multicenter study using thoracoscopy or laparoscopy. Surg Endosc 2009;23:1650–1659. Keijzer R, Puri P. Congenital diaphragmatic hernia. Semin Pediatr Surg 2010;19:180–185. Jones VS, et al. An optimum animal model for neonatal thoracoscopy. J Laparoendosc Adv Surg Tech A 2008; 18:759–762. Bidarkar SS, et al. Porcine models for pediatric minimally invasive surgical training: A template for the
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
20.
21. 22. 23.
future. J Laparoendosc Adv Surg Tech A 2012;22:117– 122. Simforoosh N, et al. Laparoscopic animal surgery for training without sacrificing animals; Introducing the rabbit as a model for infantile laparoscopy. J Laparoendosc Adv Surg Tech A 2011;21:929–933. Ghaderi I, et al. Evaluation of surgical performance during laparoscopic incisional hernia repair: A multicenter study. Surg Endosc 2011;25:2555–2563. Vaillancourt M, et al. GOALS-incisional hernia: A valid assessment of simulated laparoscopic incisional hernia repair. Surg Innov 2011;18:48–54. Keijzer R, et al. Thoracoscopic repair in congenital diaphragmatic hernia: Patching is safe and reduces the recurrence rate. J Pediatr Surg 2010:45:953–957.
285
24. Kunisaki SM, et al. Thoracoscopic repair of recurrent Bochdalek diaphragmatic hernias in children. J Laparoendosc Adv Surg Tech A 2012;22:1004–1009.
Address correspondence to: Eva Marı´a Pe´rez-Merino, DVM, PhD Department of Animal Medicine and Surgery Faculty of Veterinary Science University of Extremadura Avenida Universidad s/n 10003 Ca´ceres Spain E-mail: [email protected]
Evaluation of a Bochdalek Diaphragmatic Hernia Rabbit Model for Pediatric Thoracoscopic Training Jesu´s Uso´n-Casau´s, DVM, PhD,1 Eva Marı´a Pe´rez-Merino, DVM, PhD,1 Ramo´n Rivera-Barreno, DVM, PhD,2 Carlos A. Rodrı´guez-Alarco´n, DVM, PhD,2 and Francisco M. Sa´nchez-Margallo, DVM, PhD 3
Abstract
Background/Aim: This study evaluated the usefulness of a Bochdalek hernia rabbit model as a tool for advanced thoracoscopic training, teaching the specific skills required for thoracoscopic repair of congenital diaphragmatic hernia. Materials and Methods: An incision was made in the Bochdalek triangle of 25 New Zealand rabbits (weighing 3–3.5 kg) to induce an experimental diaphragmatic hernia. At 72 hours later, a thoracoscopic repair of the hernia as described for newborns was performed by 25 pediatric surgeons divided into two groups: expert and novice. The tasks assessed were organ relocation and diaphragm suture. A visual analog scale was used to evaluate technical performance. The objective performance measure was completion time. Complications were recorded, and suture quality was scored. The surgeons evaluated the model by completing a questionnaire, grading items on a 5-point scale. Results: All 25 animals developed a diaphragmatic hernia with protrusion of the intestine into the thoracic cavity. Expert trainees had significantly shorter completion times and better performance scores than novices. Experts also received higher scores for suture quality. Five novices caused perforations or bleeding, but no experts did. The surgeons rated the model positively, highlighting the similarities between the model and newborn hernias and its usefulness for pediatric training programs. Conclusions: The Bochdalek hernia rabbit model can be used to detect different levels of experience in pediatric thoracoscopy. This realistic and easily reproducible model can help to perfect thoracoscopic skills in a realistic recreation of a pediatric Bochdalek hernia repair. Introduction
T
horacoscopy is the third most frequent approach for repairing neonatal congenital diaphragmatic hernia (CDH), behind laparotomy and thoracotomy.1 To embrace this technique and ensure it is safe for pediatric patients, adequate training is essential. However, individual centers and surgeons may have limited experience with congenital diaphragmatic hernias and even less experience with the minimally invasive surgical (MIS) approach.2,3 There are many levels of laparoscopic training that surgeons must master. Practice in both the simulation laboratory and the animal operating room should, ideally, precede the human operating room. Low-fidelity models (box trainers) are well suited for novice learners. These inanimate models provide a reliable and reproducible method for learning instrument-handling maneuvers and mastering basic laparo1 2 3
scopic skills such as suturing and knot tying.4 However, the clinical use of minimally invasive surgery should be preceded by a training period with a model that closely resembles human conditions.5 Even for learning isolated mechanical tasks it has been shown that training can be provided using a pelvic trainer or animal model with similar effects, but in vivo performance of the same task requires secondary surgical skills, which are conveyed only during live training with greater success. Consequently, stepwise teaching with both modules seems reasonable before these procedures are approached in neonates or small children.6 The development of specific surgical training models that mimic the human situation can enhance advanced laparoscopic skills and facilitate the development of surgical skills for rare diseases. Several rabbit models for different endosurgical pediatric procedures have been described,7 but there is no specific animal model for thoracoscopic training
Department of Animal Medicine and Surgery, Faculty of Veterinary Science, University of Extremadura, Ca´ceres, Spain. Department of Veterinary Sciences, Institute of Biomedical Sciences, Autonomous University of Juarez, Ciudad Juarez, Mexico. Department of Laparoscopic Surgery, Jesu´s Uso´n Minimally Invasive Surgery Centre, Ca´ceres, Spain.
280
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
281
on diaphragmatic hernias. Our aim is to determine the value of this experimental Bochdalek hernia rabbit model (BHRM) for pediatric thoracoscopic training. Materials and Methods
Prior to the in vivo procedures, rabbit cadavers were used to identify the Bochdalek triangle and determine the best place to create the diaphragmatic defect (Fig. 1) and the optimum positions for trocars. Diaphragmatic hernia induction
Twenty-five New Zealand white rabbits weighing 3–3.5 kg were used. The protocol was approved by the Ethical Committee for Animal Research of the Jesu´s Uso´n Minimally Invasive Surgery Centre ( JUMISC), Ca´ceres, Spain. Animal care complied with the Guide for the Care and Use of Laboratory Animals of the Institute of Laboratory Animal Resources, Commission on Life Sciences, U.S. National Research Council. The rabbits were premedicated with subcutaneous buprenorphine (0.050 mg/kg) and meloxicam (0.20 mg/kg). Anesthesia was induced with dexmedetomidine (0.015 mg/kg) and ketamine (8 mg/kg) through an intravenous catheter in the marginal ear vein and maintained with propofol (1–2 mg/kg). A midline minilaparotomy (4 cm) was performed. To induce the diaphragmatic defect, an incision (3 cm) was made in the dorsal costal area of the left diaphragm, at the small triangle between the pars lumbalis and pars costalis and the retractor costae muscle. An oval-shaped defect was created. A chest tube was inserted through the defect and brought out through a laparotomy, with a purse-string stitch around it. The laparotomy was routinely closed. Before the animal regained consciousness, the pneumothorax was aspirated, the tube was removed, and the wound was stitched closed. Thoracoscopic repair
At 72 hours after the hernia was induced, the animals were premedicated with intravenous atropine (0.05 mg/kg) and
FIG. 2.
Correct positioning of the rabbit.
midazolam (0.25 mg/kg). Anesthesia was induced using intravenous propofol (10 mg/kg). Endoscopically guided intubation was performed, and anesthesia was maintained using mechanical ventilation with 1.5% isoflurane and fentanyl (0.01 mg/kg every 30 minutes). The rabbits were positioned in the reverse Trendelenburg position and right lateral decubitus position on a specially designed operating table made of expanded polystyrene (length · width · height, 50 cm · 12 cm · 15 cm) so that the rabbit’s head could be raised to an angle of 20 (Fig. 2). The 5-mm port for the optic was positioned in the fourth left intercostal space, under the caudal border of the scapula. The left upper limb was stretched up toward the head and secured with adhesive strips to free up as much chest surface as possible. Two 3.5-mm ports for the instruments were positioned in the sixth left intercostal space, one 1.5 cm from the spine and the other 2 cm from the sternum (Fig. 3). The lung was collapsed by an intrapleural carbon dioxide insufflation (2–4 mm Hg). During the thoracoscopy, the herniated abdominal organs were relocated to the peritoneal cavity by pushing with two endoscopic graspers, and the diaphragmatic defect was repaired with five stitches using the intracorporeal technique with interrupted 5-0 glyconate suture and a laparoscopic needle. Model assessment
After Institutional Review Board approval was obtained, thoracoscopic repair of experimental hernias was performed
FIG. 1. Determination of the optimal location of the diaphragmatic defect in the cadaver.
FIG. 3. Placement of trocars and instrument during thoracoscopy. The black arrow indicates the sternum.
282
by 25 pediatric surgeons during an intensive 3-day hands-on pediatric thoracoscopy course at the JUMISC. The novice group was made up of 20 trainees (pediatric residents, postgraduate year [PGY]-1–4) divided into 10 two-person teams, with each person alternating between the role of surgeon and assistant for each surgery. Before operating on the animals, all novices were trained in basic laparoscopic skills using a box trainer (SIMULAP-IC05; CCMIJU, Ca´ceres, Spain). Five pediatric surgeons with advanced minimally invasive surgery training and previous clinical thoracoscopy experience were recruited as an experienced group. After the trocars were correctly positioned, two tasks were assessed: repositioning of abdominal organs and diaphragmatic suture. The assessment tools included a visual analog scale (VAS) to measure overall technical competence at each task. The scale used is a 10-cm line with descriptive anchors at each end. The anchors were ‘‘Unable to complete the task with maximum guidance’’ and ‘‘Could perform the task safely and independently (fully competent).’’ The evaluator wrote an ‘‘X’’ on the line to indicate the score he or she assigned for that particular item. The scores were determined numerically by measuring where along the 10-cm line that the mark was placed. The participants were evaluated by two trained observers: an attending pediatric surgeon and a trainer in minimally invasive surgery at the JUMISC. The average score from the two reviewers was calculated and used in the final data analysis. Completion time was the objective performance measure for both tasks. After the procedure, the animals were euthanized, and their abdominal and thoracic organs were carefully examined to detect any severe hemorrhage, visceral injuries, or intestinal perforation. The diaphragms were also removed to evaluate final suture quality. Correct suture was considered to be five stitches spaced 5 – 1 mm, each with a square knot tied by four alternating throws, and without tears. Each correct stitch was given 3 points (suture maximum score, 15 points). Tears, incorrect tension, or incorrect distance was penalized by deducting 1 point each. At the end of the course, the surgeons were given a questionnaire consisting of six statements about how realistic the model was and its usefulness for training. These were presented on a 5-point ordinal scale (1 = not at all, 2 = a little, 3 = enough, 4 = a lot, and 5 = completely). The statistical software package SPSS version 15.0 (SPSS, Inc., Chicago, IL) was used for the statistical analysis. Because of the smaller sample size and the non-normality of the data, a two-tailed Mann–Whitney U test was used to evaluate differences between the two groups for the different parameters (time, VAS performance, and final suture quality). P < .05 was considered statistically significant. Intraclass correlation coefficients (ICCs) were used to assess inter-rater reliability of the VAS for competence.
´ N-CASAU ´ S ET AL. USO
FIG. 4. Operative view of the diaphragmatic hernia in the rabbit model. operating table were optimal, allowing an excellent diaphragmatic approach, and repair was achieved successfully in all animals (Fig. 5). Although carbon dioxide insufflation facilitated the reduction of the abdominal viscera, this technique was avoided as much as possible to force the trainees to use their tissue-handling skills. Table 1 provides a summary of the results of the different procedure assessments. The VAS scores for competence show that the experienced group performed better in both tasks. The total ICC for VAS-hernia reduction was 0.79 (95% confidence interval, 0.61–0.91), and the total ICC for VASsuture was 0.81 (0.95% confidence interval, 0.63–0.94). The overall time for both the tasks combined ranged from 28 minutes for the most expert surgeon to 81 minutes for a novice, and the expert group completed both tasks in a significantly shorter time. The expert participants continued to achieve higher scores for suture quality. Five of the novices caused perforations (three PGY-3 and one PGY-4) or hemorrhage (one PGY-4), but no experts made these mistakes.
Results
The original 3-cm diaphragmatic incision was sufficient to cause herniation with protrusion of several intestinal loops and the stomach into the thoracic cavity. The original incision increased in size over the 72-hour period because of progressive herniation of the peritoneal organs (Fig. 4). The positioning of the trocars and of the animals on the special
FIG. 5. Appearance of the diaphragm defect with partial diaphragmatic suture after relocation of the organs.
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
Table 1. Results of Different Assessment Tools for the Procedures by Experience Group Assessment
Novice (n = 20)
Experts (n = 5)
Completion time (minutes) Hernia reduction 24.4 – 5.7 14.1 – 0.6 Diaphragmatic suture 27.8 – 5.9 16.1 – 0.4 VAS Hernia reduction 4.3 – 2.7 9.1 – 2.1 Diaphragmatic suture 3.7 – 2.1 8.9 – 2.4 Final suture quality (out of 15) 6.3 – 2.2 10.4 – 1.0 Perforations/hemorrhage (n) 4/1 0/0
P .0008 .0006 .0002 .0002 .0002
Data are mean – standard deviation values unless indicated otherwise. VAS, visual analog scale.
Table 2 shows the mean ratings for the trainees’ first impression of the BH-RM. For both groups, the highest mean score was for the question about the usefulness of this model for pediatric thoracoscopic training. The second highest rating was for the capacity of the model to improve thoracoscopic skills, and the third for the similarities between the model and a newborn CDH. The lowest mean score was for the capacity of the model to facilitate management of complications in CDH surgery. The statistical analysis revealed no significant difference between the scores of the two groups (P > .3), indicating a general agreement on all questions. Discussion
Thoracoscopy has proved to be a feasible and safe surgical approach under two conditions: for carefully selected patients and for pediatric surgeons with sufficient training to perform the MIS procedure.8,9 Several diaphragmatic models have been used to measure different aspects of this congenital defect. However, we were unable to find a live diaphragmatic hernia model for thoracoscopic training reported in the literature. Many years of experience at the JUMISC have shown that both simulators and animal models should be used in laparoscopic training. The JUMISC has a fully validated training method based on a pyramid structure with four levels: (1) basic and advanced skills with box trainers, (2) anatomical protocols and advanced skills with animal models, (3) advanced procedural skills with telesurgical applications, and (4) practice in the operating room. This curriculum has driven the development of the different animal models, conceived as
283
a training method for the second level in the pyramid.5,10 This program has been used in pediatric and neonatal training, obtaining very high ratings over the last few years. In order to choose the most suitable animal model for each specialty, aspects relating to comparative anatomy and size have been taken into account, to ensure that the training models are as similar to humans as possible. For example, pigs are used as a model for digestive and urological laparoscopy, and rabbits are used for neonatal laparoscopy. Surgical techniques included in the training program are tested by pediatric surgery specialists and selected based on their feasibility, usefulness, and realism.11,12 Recently, attempts have been made to replace live animals in pediatric laparoscopic training with high-fidelity models that recreate procedures such as thoracoscopic esophageal atresia/tracheoesophageal fistula13 or CDH14 repair. The cost of one BH-RM (rabbit + surgeries + 72 hours of care = e200) is similar to that of those in vitro simulators. However, the latter are obviously reusable and do not require facilities and veterinary staff for housing and care of the animals. There are also still legal and ethical concerns about the use of animals for single-use training labs. However, at present, the weakest physical attributes of the diaphragmatic hernia simulator are the realism of the tissues, including the skin, diaphragm, and intestine, and the poor haptic feedback of the synthetic materials.14 In contrast, our questionnaire scores highlight the positive response of the surgeons with regard to the similarity of the BH-RM to real-life scenarios and its usefulness. For training in handling tissues, in vitro models are poor substitutes for in vivo models because pulsating vessels, bleeding, and control of bleeding are essential to surgery. The BH-RM exactly replicates the classic Bochdalek hernia, and laparoscopic surgery can be performed in a real environment with similar instrument size, port placement, and rabbit position to those described for thoracoscopy in newborns.15–17 Furthermore, the model not only allows tissue handling of varying degrees of difficulty, but also makes it possible to practice several other laparoscopic tasks such as cutting, coagulation, stitching, and intracorporeal knot tying. Once the laparoscope was inserted, the surgeons pointed out how similar the model was to what they see thorascopically in classic cases of newborn CDH. The large cecum size in the rabbit and the huge amount of herniated intestine mean that the required reduction maneuvers are very similar in both cases. Various authors have suggested that piglets may provide the optimum animal model for neonatal thoracoscopy because the fragility of the rabbit coupled with a limited intrathoracic working space limits its use for this
Table 2. Observed Results for Bochdalek Hernia Rabbit Model Assessment Score Item Similarities between BH-RM and newborn CDH Similarities between rabbit thoracic dimension and newborn thorax Degree of technical difficulty compared with real scenario Training capacities of model to improve thoracoscopic skills Training capacities of model to facilitate management of CDH surgical complications Degree of usefulness of this model in a training program in pediatric thoracoscopy surgery
Novice
Expert
P
4.10 – 0.68 3.84 – 0.62 4.00 – 0.36 4.20 – 0.50 3.44 – 0.94 4.29 – 0.81
3.90 – 0.31 3.27 – 0.52 3.76 – 0.42 4.50 – 0.63 3.05 – 0.83 4.55 – 0.85
.716 .317 .628 .815 .496 .856
All items scored on a scale range of 1–5 points, where 1 = not at all, 2 = a little, 3 = enough, 4 = a lot, and 5 = completely. Data are mean – standard deviation values. BH-RM, Bochdalek hernia rabbit model; CDH, congenital diaphragmatic hernia.
´ N-CASAU ´ S ET AL. USO
284
technique.18,19 However, the similar proportions of rabbits and newborns and the rabbit model’s value for laparoscopic pediatric training are well established.6,7,20 Although most of the surgeons noticed that the rabbit thorax is flatter whereas the newborn thorax is more rounded, the intrathoracic working space is similar. We agree that rabbit tissue is very fragile, but it ensures that surgeons use precise and careful maneuvers, acquiring an extremely refined technique. Handling intestinal loops in a small thoracic space and pulling them through a diaphragmatic defect is an excellent exercise that requires precise, careful movements. Diaphragmatic suturing was chosen because it is a fundamental MIS skill and because traditionally the optimal neonatal candidate for a CDH thoracoscopic repair would be one in whom the diaphragm could be repaired primarily.8 Both are well-defined procedures that allow objective evaluation (time, presence/absence of perforation/hemorrhage, suture quality) and an overall competence score (VAS). In the future, our aim is to develop a valid procedure-specific tool for this surgical procedure, similar to the modified GOALS tool used in the assessment of laparoscopic incisional or inguinal hernia repair.21,22 Thoracoscopy for recurrent herniation and the use of a diaphragmatic patch are other real scenarios23,24 that we plan to include in the model in the future. The BH-RM provided a clear distinction between expert and novice groups. Novices with some experience handling laparoscopic tools carried out both techniques (relocation of organs and suturing) better and faster than the novice subjects with very little laparoscopic experience. Perforation and hemorrhage occurred most frequently among the novices, especially in senior trainees, probably because of a more aggressive technique. This demonstrates the need for training with real tissues before operating on patients. Although the surgeons believed that thoracoscopic skills could be substantially improved, the model’s usefulness in the management of surgical complications received the lowest score. The expert surgeons reasoned that most major surgical complications (such as intestinal perforation) involve moving from thoracoscopy to open surgery, and, moreover, conditions such as lung hypoplasia or pulmonary hypertension are not considered in this model. In conclusion, we have developed an inexpensive, realistic, and easily reproducible, in the short term, diaphragmatic hernia model, to be used to hone thoracoscopic skills and develop specific skills required by surgeons to repair Bochdalek CDH in newborns using the thoracoscopic approach. This model could be included in pediatric MIS training courses and pediatric surgery fellowship programs as a further step toward excellence in MIS skills.
2.
3. 4. 5.
6. 7.
8. 9. 10. 11. 12.
13.
14.
15.
Acknowledgments
This study was supported by Junta de Extremadura (PRI 08A105). Disclosure Statement
No competing financial interests exist.
16.
17. 18.
References
1. Tsao K, Lally PA, Lally KP; Congenital Diaphragmatic Hernia Study Group. Minimally invasive repair of con-
19.
genital diaphragmatic hernia. J Pediatr Surg 2011;46:1158– 1164. Cho SD, et al. Analysis of 29 consecutive thoracoscopic repairs of congenital diaphragmatic hernia in neonates compared to historical controls. J Pediatr Surg 2009;44: 80–86. Nguyen TL, Le AD. Thoracoscopic repair for congenital diaphragmatic hernia: Lessons from 45 cases. J Pediatr Surg 2006;41:1713–1715. Molinas CR, Binda MM, Mailova K, Koninckx PR. The rabbit nephrectomy model for training in laparoscopic surgery. Hum Reprod 2004;19:185–190. Uso´n J, Sa´nchez FM, Sa´nchez MA, Pe´rez FJ, Hashizume M. Principios ba´sicos. In: Uso´n J, Sa´nchez FM, Pascual S, Climent S (eds.). Formacio´n en Cirugı´a Laparosco´pica Paso a Paso, 3rd ed. Ca´ceres, Spain: Centro de Cirugı´a de Mı´nima Invasio´n, 2007:21–89. Heinrich M, Tillo N, Kirlum HJ, Till H. Comparison of different training models for laparoscopic surgery in neonates and small infants. Surg Endosc 2006;20:641–644. Kirlum HJ, Heinrich M, Till H. Rabbit model serves as a valuable operative experience and helps to establish new techniques for abdominal and thoracic endosurgery. Pediatr Surg Int 2005;21:91–93. Yang EY, et al. Neonatal thoracoscopic repair of congenital diaphragmatic hernia: Selection criteria for successful outcome. J Pediatr Surg 2005;40:1369–1375. Okazaki T, et al. Indications for thoracoscopic repair of congenital diaphragmatic hernia in neonates. Pediatr Surg Int 2011;27:35–38. Uso´n-Gargallo J, et al. Modelo de formacio´n piramidal para la ensen˜anza de cirugı´a laparosco´pica. Cir Cir 2013;81: 420–430. Pe´rez-Duarte FJ, et al. Design and validation of a training model on paediatric and neonatal surgery. Cir Pediatr 2012;25:121–125. Pe´rez-Duarte, FJ, et al. Validation of the lagomorph animal model for training and skills acquisition on neonatal laparoscopic surgery. Presented at the 24th Conference of the Society for Medical Innovation and Technology, Barcelona, 2012. www.smit.de/typo3/index.php?id = 90 (accessed November 30, 2013). Barsness KA, Rooney DM, Davis LM. Collaboration in simulation: The development and initial validation of a novel thoracoscopic neonatal simulator. J Pediatr Surg 2013;48:1232–1238. Barsness KA, Rooney DM, Davis LM. The development and evaluation of a novel diaphragmatic hernia repair simulator. J Laparoendosc Adv Surg Tech A 2013;23:714– 718. Arca MJ, et al. Early experience with minimally invasive repair of congenital diaphragmatic hernias: Results and lessons learned. J Pediatr Surg 2003;38:1563–1568. Gomes Ferreira C, et al. Neonatal minimally invasive surgery for congenital diaphragmatic hernias: A multicenter study using thoracoscopy or laparoscopy. Surg Endosc 2009;23:1650–1659. Keijzer R, Puri P. Congenital diaphragmatic hernia. Semin Pediatr Surg 2010;19:180–185. Jones VS, et al. An optimum animal model for neonatal thoracoscopy. J Laparoendosc Adv Surg Tech A 2008; 18:759–762. Bidarkar SS, et al. Porcine models for pediatric minimally invasive surgical training: A template for the
BOCHDALEK HERNIA MODEL FOR THORACOSCOPIC TRAINING
20.
21. 22. 23.
future. J Laparoendosc Adv Surg Tech A 2012;22:117– 122. Simforoosh N, et al. Laparoscopic animal surgery for training without sacrificing animals; Introducing the rabbit as a model for infantile laparoscopy. J Laparoendosc Adv Surg Tech A 2011;21:929–933. Ghaderi I, et al. Evaluation of surgical performance during laparoscopic incisional hernia repair: A multicenter study. Surg Endosc 2011;25:2555–2563. Vaillancourt M, et al. GOALS-incisional hernia: A valid assessment of simulated laparoscopic incisional hernia repair. Surg Innov 2011;18:48–54. Keijzer R, et al. Thoracoscopic repair in congenital diaphragmatic hernia: Patching is safe and reduces the recurrence rate. J Pediatr Surg 2010:45:953–957.
285
24. Kunisaki SM, et al. Thoracoscopic repair of recurrent Bochdalek diaphragmatic hernias in children. J Laparoendosc Adv Surg Tech A 2012;22:1004–1009.
Address correspondence to: Eva Marı´a Pe´rez-Merino, DVM, PhD Department of Animal Medicine and Surgery Faculty of Veterinary Science University of Extremadura Avenida Universidad s/n 10003 Ca´ceres Spain E-mail: [email protected]
