Volume 33, Issue 2 , Pages 208-213, February 2007
A New Vascular Anastomosis Model: Relation between Outcome and Experience
Article Outline
Background
Vascular anastomosis is a complex task that requires multiple skills. Existing training methods lack the ability to objectively quantify surgical skill. In this study we tested a new vascular anastomosis model for bench training.
Materials and methods
Surgical performance was assessed based on the new vascular anastomosis training model. Thirty- eight subjects were asked to (1) close the end of a 6-mm polytetrafluoroethylene (PTFE) graft, using a continuous suturing technique with 6-0 polypropylene; (2) perform end-to-end and (3) end-to-side anastomosis using the same materials and techniques.
Results
The mean age (sd) of all participants was 28.3 (2.1) years. More surgically experienced trainees did better in all measures of technical skill. Although there was a tendency for those with previous experience with the training model to do better in terms of the technical outcomes, these differences were not statistically significant. Multivariable analysis revealed that level of surgical training and type of anastomosis were the only significant factors related to completion time.
Conclusions
Our study confirmed the impact of increasing surgical experience on the technical skills of surgical trainees. Trainees with higher levels of training made fewer errors and completed the procedures faster than those with lower levels of training.
Keywords: Surgical education, Anastomosis leakage, PTFE graft
Introduction
The traditional surgical apprenticeship has been used for training surgeons for centuries. The surgeon-trainee is systematically exposed to graduated clinical experience in the operating room for several years of residency training. Surgical residents acquire skill by observing and assisting experienced surgeons in action. As their skill levels increase they receive more responsibility and perform more surgical procedures by themselves.1
Technical ability is one of the most important components of surgical competency. Operative procedures require 75% decision making and 25% dexterity.2 Decision making can be learned and assessed via written and oral examinations. The assessment of surgical skill has long lagged behind that of decision making.3, 4 Many methods have been used to evaluate technical performance of the surgeon. Measuring the time taken to complete procedures is one of the earliest methods, with the faster surgeon presumably being better than the slower one. Another way of quantifying skill is the documentation of the number and range of procedures completed. However, this is only number counting, as no relationship between the experience of a trainee and postoperative complication rates can be assessed.5
In the current system of surgical training, most of the technical skill is acquired during supervised training on a limited number of patients available at teaching centers.6 The training process on live patients has an inherent risk to the patient and increases the time required for the operations. The Institute of Medicine reported that 44,000 to 98,000 Americans die in hospitals each year as a result of medical error.7 This is more than the number of deaths from highway accidents or from breast cancer. The public increasingly rejects a training model that uses the operating room and real humans as the format for training.
Surgical skill training can be augmented by the use of live animals, cadavers and inert materials. Each of these alternative methods has some limitations. Existing training methods also suffer from an inability to objectively quantify surgical skill. It is difficult to assess a trainee's technical progress other than through subjective rating.
The advantages of a bench-training model as a teaching modality over the operating room environment include the possibility to repeat practice and reuse materials, lower costs due to elimination of live patients or animals, and less stressful environment.8 Trainees can learn step by step from the instructors who can provide individualized feedback to them.
In this study, we tested a new vascular anastomosis model for bench training. Satisfaction and outcome after practicing with the model were evaluated in the context of previous experience with the model (prior participation in previous workshops) and level of surgical training (year of residency training).
Materials and Methods
Model for vascular anastomosis
The vascular anastomosis training model consisted of three metal tubes for holding the vascular conduit for anatomosis. Vascular anastomosis can be done in an end-to-end or end-to-side fashion. Each tube was designed to be adjustable for the distance and angle of anastomosis. The metal rods at the edge of the model were used to hold the vascular conduit during anastomosis. A submersible pump, SP-500 (Resun, Bangkok, Thailand), with adapter and control system was designed and applied to test leakage of vascular anastomoses.
Subjects
Thirty eight surgeons-in-training participated in this study. There were fourteen first-year surgical residents from all subspecialties of surgery, including orthopaedics, cardiac surgery, urology and neurosurgery. All were trained under the same first-year surgical residency program at the Department of Surgery, Ramathibodi Hospital, Mahidol University. Other subjects included one vascular surgery fellow, nine second-year, six third-year, and eight fourth–year general surgery residents.
Seventeen subjects practiced on the model prior to assessment, including three first-year, four second-year, three third-year, six fourth-year residents, and one vascular surgery fellow.
Methods
Surgical performance was assessed based on the new vascular anastomosis training model. Subjects were asked to (1) close the end of a 6-mm polytetrafluoroethylene (PTFE) graft, using a continuous suturing technique with 6-0 polypropylene; (2) perform end-to-end (Fig. 1) and (3) end-to-side (Fig. 2) anastomoses using the same materials and techniques. A questionnaire was used to evaluate the subject's satisfaction of the training model after completion of the procedures. Levels of satisfaction were graded on a 5-point scale, with 5 being the most satisfactory and 1 unsatisfactory.
Fig. 1
Assessment of surgical performance: end-to-end anastomosis.
Fig. 2
Assessment of surgical performance: end-to-side anastomosis.
Assessment of surgical technique
Three outcome measures were used to evaluate adequacy of surgical technique: time to completion of each procedure, the grading of leakage across the anastomosis, and the diameter of the anastomosis at the narrowest point.
Time to completion. Subjects were given a standardized starting position – sitting down facing the model, with both hands placed palm down on the side of the model. Recordings were started from this position and ended when the suture was cut after the final knot. Each trainee completed one anastomosis of each type.
Grading of leakage. A submersible pump, SP-500 (Resun, Bangkok, Thailand), with adapter and control system was used to test the leakage of the end-to-side vascular anastomosis. One end of the PTFE graft was connected to the pump system with other ends clamped off. The pressure used was 160
mmHg. We used the same pump and setting for all experiments. The grading of leakage was defined as: 0 – no leakage; 1 – water dripping from the anastomosis (Fig. 3); 2 – water ejecting from the anastomosis (Fig. 4); and 3 – water bursting from the anastomosis (Fig. 5).
Fig. 3
Grade 1 of leakage.
Fig. 4
Grade 2 of leakage.
Fig. 5
Grade 3 of leakage.
Anastomosis diameter. The diameter of the end-to-end anastomosis at the narrowest point was measured using a Vernier caliper in millimeters.
Statistical analysis
Age and time to completion of each procedure were summarized as mean (standard deviation). Gender, levels of satisfaction, grading of leakage, previous experience with the model and levels of surgical training were summarized as counts and percentages. Tests for the effects of age, gender, levels of training and previous experience (with the vascular anastomosis training model) on the time to procedure completion were done within a linear statistical model framework with random effects for the individual trainee. The effects of the same variables on grades of leakage were assessed using an ordinal logistic regression model. Finally, effects on the diameter of the vascular anastomosis were tested using an ordinary linear regression model. Statistical analyses were performed using Stata v. 7 (Stata Corp, College Station, TX, USA). Statistical significance was defined as a p-value of 0.05 or less.
Results
Participant demographics
The mean age (sd) of all participants was 28.3 (2.1) years. Thirty-five resident trainees were men and three were women. Details for the various levels of surgical training are presented in Table 1. The mean age (sd) of participants who have practiced with the model in a hands-on workshop and received verbal feedback from expert vascular surgeons was 28.5 (1.7) while the mean age (sd) of subjects who have not practiced was 28 (2.4). More surgically experienced trainees have received more vascular anastomosis training.
Table 1. Demographics, prior surgical and vascular anastomosis training experience
| Year of residency training (n=38) | |||||
|---|---|---|---|---|---|
| 1 (n=14) | 2 (n=9) | 3 (n=6) | 4 (n=8) | Fellow (n=1) | |
| Age (years): Mean (sd) | 26.9 (1.3) | 27.6 (1.1) | 29.8 (3.3) | 30.1 (0.83) | 29 |
| Male gender: Number (%) | 13 (93) | 8 (89) | 6 (100) | 7 (88) | 1 |
| Previous workshop experience: Number (%) | 3 (21) | 4 (44) | 3 (50) | 6 (75) | 1 |
Relation between years of training and technical outcome
The mean (sd) time consumed during the anastomoses, the summary of the grades of leakage and the mean (sd) of the diameter of vascular anastomoses for each level of surgical training are shown in Table 2. More surgically experienced trainees did better in all measures of technical skill.
Table 2. Measures of technical skill and level of surgical training
| Year of residency training | Completion time (minutes) | Grades of leakage∗ | Diameter of anastomosis∗∗ (mm) Mean (sd) | ||||
|---|---|---|---|---|---|---|---|
| Mean (sd) | Number (%) | ||||||
| End | End-to-end | End-to-side | 1 | 2 | 3 | ||
| 1 (n=14) | 11.6 (4.7) | 21.8 (9.2) | 23.9 (8.7) | 8 (58) | 3 (21) | 3 (21) | 5.70 (0.09) |
| 2 (n=9) | 10.8 (5.1) | 16.1 (4.7) | 20.6 (3.0) | 7 (78) | 1 (11) | 1 (11) | 5.80 (0.08) |
| 3 (n=6) | 7.5 (2.1) | 11.0 (2.3) | 15.8 (4.1) | 5 (83) | 0 | 1 (17) | 5.82 (0.04) |
| 4 (n=8) | 7.5 (3.0) | 13.1 (4.4) | 16.5 (4.7) | 8 (100) | 0 | 0 | 5.90 (0.05) |
| Fellow (n=1) | 3.0 | 7.0 | 10.0 | 1 | 0 | 0 | 5.90 |
∗Maximum grade of leakage for end-to-side anastomosis. |
∗∗Diameter of end-to-end anastomosis at the narrowest point. |
Relation between previous experience with anastomosis training model and technical outcomes
A summary of time for completion of various anastomoses, grades of leakage and diameter of anastomoses for students with or without prior experience with the training model are shown in Table 3. Although there was a tendency for those with previous experience to do better in terms of the technical outcomes, these differences were not statistically significant (see Table 4). Subgroup analyses comparing residents with prior experience and those with no prior experience for each level of surgical training (year of residency training) did not change this conclusion (analysis not shown).
Table 3. Previous experience with anastomosis training model and technical outcomes
| Previous experience with model | Completion time (minutes) | Grades of leakage∗ | Diameter of anastomosis∗∗ (mm) mean (sd) | ||||
|---|---|---|---|---|---|---|---|
| Mean (sd) | Number (%) | ||||||
| End | End-to-end | End-to-side | 1 | 2 | 3 | ||
| Yes (n=17) | 8.4 (3.8) | 14.5 (6.5) | 18.8 (7.5) | 15 (88) | 1 (6) | 1 (6) | 5.80 (0.08) |
| No (n=21) | 10.4 (4.8) | 18.1 (8.3) | 20.8 (6.6) | 14 (67) | 3 (14) | 4 (19) | 5.70 (0.08) |
∗Maximum grade of leakage for end-to-side anastomosis. |
∗∗Diameter of end-to-end anastomosis at the narrowest point. |
Table 4. Factors related to completion time (multivariable analysis)
| Factor | Average change in completion time∗ | 95% CI | p-value |
|---|---|---|---|
| Previous experience | 0.19 minutes increase | −3.0 to 3.4 mins | 0.908 |
| Level of training | −2.6 minutes (decrease) per 1 higher level | −3.9 to −1.3 mins | <0.001 |
| End-to-end anastomosis | 7.1 minutes increase | 5.1 to 9.0 mins | <0.001 |
| End-to-side anastomosis | 10.4 minutes increase | 8.4 to 12.4 mins | <0.001 |
∗Base line completion time is for end closure, which was 15.3 minutes on average for 1st year residents with no prior training. |
Adjusted multivariable analysis
Multivariable analysis revealed that level of surgical training and type of anastomosis were the only significant factors related to completion time (Table 4). Previous experience with the model, once adjusted for these significant factors, was not an important factor.
The only significant factor related to grades of leakage was the diameter of the anastomosis. A tenth of a millimeter increase in the anastomosis diameter was dramatically associated with lower grades of leakage. Although it might appear that higher levels of surgical training was associated with higher grades of leakage, this difference was not statistically significant (p=0.211). When adjusted for anastomosis diameter, previous experience with the vascular anastomosis training model was not significant.
The only significant factor related to diameter of the end-to-end anastomosis was level of surgical training. Adjusted for this factor, previous experience with the model was only of borderline significance (p=0.059).
Satisfaction with the model
All participants answered the questionnaire. A summary of the satisfaction rates is presented in Table 5. Most participants rated the vascular anastomosis training model as acceptable or better in all items.
Table 5. Satisfaction with the vascular anastomosis training model
| Satisfaction item | Satisfaction scale (N=38) | ||||
|---|---|---|---|---|---|
| Number (%) | |||||
| 5 | 4 | 3 | 2 | 1 | |
| Improve understanding | 4 (11) | 16 (42) | 13 (34) | 5 (13) | 0 |
| Help practice | 2 (5) | 12 (33) | 23 (61) | 1 (3) | 0 |
| Realistic | 0 | 11 (29) | 21 (55) | 6 (16) | 0 |
| Appropriate for learning | 7 (18) | 21 (55) | 9 (24) | 1 (3) | 0 |
| Applicability | 6 (16) | 25 (66) | 7 (18) | 0 | 0 |
Discussion
It is well known that in procedure-based specialties such as surgery, better skills result in better outcome.9, 10, 11 Our study has confirmed the impact of increasing surgical experience on the technical skills of surgical trainees. Trainees with higher levels of training made fewer errors and completed the procedures faster than those with lower levels of training.
Although there seemed to be a trend for trainees with previous experience using the new vascular anastomosis training model to do better technically than those without previous experience (Table 3), this was not confirmed on multivariable analyses (Table 4). Once adjusted for level of surgical training, previous exposure to vascular anastomosis training was no longer an important factor. This is explained by the confounding effect of level of surgical training: more higher level trainees had previous exposure to the anastomosis training model (Table 1).
More experienced trainees tended to construct wider vascular anastomoses, the single most important factor in reducing anastomotic leakage. For this technical outcome, previous experience with the vascular anastomosis training model was of borderline significance (p=0.059).
Most participants were satisfied with the anastomosis training model (Table 5), but there is room for improvement, especially concerning the realism of the materials used. Some authors have used rubber gloves or silicone as training materials for vascular anastomosis. It is important for the trainees to familiarize themselves with the instruments and materials that resemble those in actual practice.
The results of this study emphasized the importance of level of surgical training. This training is a sum of all surgical experience a trainee receives up to the time of testing. Practice on the vascular anastomosis training model is also a part of this training. The reason why prior practice or experience with the training model did not clearly and independently improve technical skills was likely due to limited exposure (only one workshop experience) at some distant past (six months prior to the study) without repeated exposure. If the evaluation of vascular surgical skills in this study is valid, then more practice with the training model should improve technical competence.
Vascular anastomosis is a complex task that has multiple requisite skills which include manual dexterity and visual-spatial ability.8 The process of learning requires practice and multiple attempts in order to integrate the multiple skills required. The operator needs to 1) plan the sequence of events required to complete the task, 2) prepare the blood vessels and the operative field, 3) move their body and arms as necessary to facilitate suture placement, 4) grasp the needle at the appropriate place and angle for each placement of suture, 5) adjust the patient's body position or room lighting to make the anastomosis technically possible, and 6) place the sutures at appropriate intervals and distances from the edges of the blood vessels with minimal tissue damage. Expert surgeons are able to simultaneously integrate all these skills to complete the task while trainees often focus only on 1 to 2 of these skills during the learning process.
There are many methods available for training of vascular skill. Jackson, et al. reviewed the literature on surgical skills training and assessment. They concluded that training for vascular surgeon can be improved by various methods such as workshop, laboratory, and seminar room based methods.12 Pandey, et al. demonstrated the efficacy of workshop training on simulators.13 Other studies also showed the assessment of trainee competency.14, 15, 16
These methods of bench training will not replace the standard approach to surgical training but will complement and extend the learning experience. The first step towards acceptance of the model is to perform more validation studies. The second step is to incorporate the model into the surgical curriculum, where it is necessary to perform objective analysis of technical skills. The final step is the evaluation of technical competency.
A major challenge facing surgical educators is to develop sophisticated education centers that bring together the expertise of instructors and investigators in surgical skills. Individualized verbal feedback from an expert surgeon is important in teaching surgical skills.
A recent study has shown no significant difference in the technical skills outcome in a checklist and global rating for trainees who received repeated video feedback compared to a control group who received expert surgical feedback alone.8 The impact of instructor feedback on outcome of vascular anastomosis was not assessed in this study.
In summary, numerous challenges lie ahead in the effective delivery of surgical skills education. It is important to assess surgical skills by both expert reviewers and a technology driven process. However, more work is needed to develop and refine the assessment methods. Instructors must do their best to help the trainees in their efforts to learn and acquire the necessary surgical skills.
References
- . Surgical handicraft: teaching and learning surgical skills. Am J Surg. 1987;153:422–427
- . Relationship between skill and outcome in the laboratory-based model. Surgery. 2002;131:318–323
- . Evaluation of procedural skills of internal medicine residents. Acad Med. 1989;64:213–216
- . The next generation of vascular surgeons: how should they be trained and credentialed? Who should pay for it?. Semin Vasc Surg. 1997;10:72–80
- . Assessment of surgical residents' competence based on postoperative complications. Int Surg. 1987;72:230–232
- Measuring and developing suturing technique with a virtual reality surgical simulator. J Am Coll Surg. 1999;189:114–127
- . Health care quality and how to achieve it. Acad Med. 2002;77:91–99
- . Effectiveness of repeated video feedback in the acquisition of a surgical technical skill. Can J Surg. 2005;48:195–200
- Virtual reality training improves operating room performance: results of a randomized, double-blinded study. Ann Surg. 2002;236:458–464
- . Teaching surgical techniques and procedures using advanced educational tools and concepts. Asian J Surg. 2005;28:159–162
- . The relationship between motion analysis and surgical technical assessments. Am J Surg. 2002;184:70–73
- . Training for infrainguinal bypass surgery. Eur J Vasc Endovasc Surg. 2003;26(5):457–466
- European Vascular Workshop, Pontresina. Do workshops improved the technical skill of vascular surgical trainees?. Eur J Vasc Endovasc Surg. 2005;30(4):441–447
- . Education and Training Committee of the Vascular Society of Great Britain and Ireland. Setting standards for the assessment of operative competence. Eur J Vasc Endovasc Surg. 2005;30(2):215–218
- Robotically assisted aorto-femoral bypass grafting: lessons learned from our initial experience. Eur J Vasc Endovasc Surg. 2004;27(5):507–511
- . The efficacy of robot-assisted versus conventional laparoscopic vascular anastomoses in an experimental model. Eur J Vasc Endovasc Surg. 2004 Mar;27(3):283–286
PII: S1078-5884(06)00538-7
doi:10.1016/j.ejvs.2006.09.026
© 2006 Published by Elsevier Inc.
Volume 33, Issue 2 , Pages 208-213, February 2007
