Article ID: CJ-19-0386
Background: Since endovascular aneurysm repair has become predominant, the issue of training young vascular surgeons in open abdominal aortic aneurysm (AAA) surgery has received significant attention. Through learning curve analysis, we aimed to determine the number of cases needed for young surgeons to achieve satisfactory open surgical skills.
Methods and Results: A total of 562 consecutive patients who underwent open repair either by an attending surgeon (group A) or 6 young vascular surgeons (group Y) were included and assessed with regards to the preparation, clamp, and total operation times. Although some of the patients’ characteristics were different, the surgical procedures were comparable between the 2 groups. There was a clear trend towards a decrease in each 10 successive cases in group Y. The operation times in group A were constant at 72±30 (preparation), 48±10 (clamp), and 231±59 min (total), which were achieved by young vascular surgeons in 10, 30, and 10 cases, respectively. In the cumulative sum analysis, 25–27 cases were necessary for young vascular surgeons to enhance their surgical skills. The complication rate in group Y was no higher than that in group A.
Conclusions: Young vascular surgeons can safely learn open AAA repair without increasing operation time or complications. Approximately 30 cases would be necessary to gain satisfactory surgical skills.
With the worldwide use of endovascular aneurysm repair (EVAR), the number of open surgeries (OS) for abdominal aortic aneurysms (AAA) has decreased.1,2 In most countries, more than 50% of all AAA cases are now treated by EVAR.1–3 This development deprives young vascular surgeons of opportunities to learn OS for AAA, which is a fundamental and indispensable skill for vascular surgeons, even in the endovascular era. In fact, current young vascular surgeons perform a fewer procedures during their surgical residency than before.4–6 Recent studies suggest that more complicated cases are more likely to be treated by OS.3 Moreover, increasing morbidity and mortality rates after OS, partly attributed to fewer opportunities to learn open repair, have been reported.3,7 The question of how to train young vascular surgeons in OS has now become a significant issue.
Vascular surgeons need to learn several skills to safely perform OS for AAA. A previous study revealed that simulation-based training works well to improve open AAA repair skills.8 Another study reported that a surgeon’s case volume is associated with in-hospital deaths after OS for AAA.9,10 However, few studies concerning the relationship between clinical experience and surgical skill improvement have been reported, especially during the learning period. In the endovascular era, it is becoming increasingly difficult for young vascular surgeons to gain experience in OS. Therefore, to efficiently learn open surgical skills for AAA, setting a clear target would be useful.
The objective of the present study was to analyze the learning curve and determine the number of cases needed to establish satisfactory surgical skills in OS for intact AAA.
This was a retrospective study of prospectively accumulated data. The study followed the principles outlined in the Declaration of Helsinki. It was conducted according to the guidelines of the research ethics committee of Asahi General Hospital and the institutional review board approved this study (registration no. 2016071908). The discrete data elements were entered into a database prospectively, and informed consent for non-invasive clinical studies was obtained by opt out method from all the patients at the time of the operation. Consecutive patients who underwent OS for AAA or iliac artery aneurysm operated on either by an attending surgeon or 6 young vascular surgeons between 2003 and 2017 were enrolled. Patients with an AAA diameter ≥5.0 cm on computed tomography were candidates for repair. Patients with smaller AAAs that grew rapidly by 5 mm in 6 months or those that formed a saccular aneurysm and enlarged during the follow-up period were also recommended for surgical intervention. Solitary iliac artery aneurysms treated with aneurysmorrhaphy without arterial reconstruction and ruptured aneurysms were excluded.
The patients’ characteristics were evaluated and included age, sex, aneurysmal diameter, hypertension, diabetes mellitus, chronic heart disease, respiratory disease, chronic kidney disease, cerebrovascular disease, history of smoking, and history of abdominal surgery (hostile abdomen). In the present study, chronic heart disease was defined as the presence of either coronary artery disease, moderate or severe regurgitation in the aortic or mitral valve, severe arrhythmia, history of cardiac surgery, or history of percutaneous coronary intervention. Respiratory disease was defined as a vital capacity <80% of the predicted value, forced expiratory volume in 1 second to forced vital capacity ratio <70%, asthma, emphysema, history of lung resection, or use of home oxygen therapy. Cerebrovascular disease was defined as a history of stroke, transient ischemic attack, or cerebral hemorrhage. Chronic kidney disease was defined as creatinine clearance ≤70 mL/min, serum creatinine ≥1.5 mg/dL, or the use of hemodialysis.
Preoperative EvaluationAs the first step, we confirmed smoking cessation before elective surgery. Next, we assessed the patient’s general condition and organ function. If the patient had no active cardiac disease and had physical activity levels of ≥4 metabolic equivalents, surgery could proceed.11,12 If the patient had poor exercise capacity (<4 metabolic equivalents), a cardiologist or anesthesiologist evaluated the patient’s ability to tolerate general anesthesia. We also routinely performed blood tests, including hemoglobin A1c and creatinine clearance, and spirometry. High-risk patients whose rupture risk was estimated to be higher than their operative risk were candidates for OS despite advanced age or organ dysfunction. EVAR was introduced in the hospital in 2015, so all patients with AAA who were eligible for repair before 2015 underwent OS. Since 2015, 22 frail patients with AAA have undergone EVAR.
Operative Strategy and Postoperative ManagementThe patients were admitted to hospital on the day before the operation. Generally, we used a transperitoneal approach with a midline incision and heparin was reserved for patients with advanced peripheral arterial disease. Preoperative autologous blood transfusion was not used. Intraoperative autotransfusion systems13 were also not routinely used because the estimated blood loss during the operations was minimal. We usually clamped the infrarenal aorta, sometimes using left renal vein mobilization or simple division without reconstruction. In some cases of an aneurysm with a hostile neck, such as a shaggy or porcelain aorta or suprarenal aortic aneurysm, the suprarenal or supraceliac aorta was the site of proximal clamping. The treatment approach for concomitant internal iliac artery aneurysms (IIAAs) and the indication for branch reconstruction (inferior mesenteric and renal arteries) were left to the surgeon’s discretion. However, as a rule, concomitant IIAAs were simply excluded, treated with aneurysmorrhaphy, or resected with reconstruction if the size exceeded 2.0 cm. The inferior mesenteric artery was reconstructed when the bilateral internal iliac arteries were excluded. When the main renal artery was involved in the AAA, it was reconstructed. The accessory renal arteries were reconstructed when their sizes exceeded 3 mm in diameter. In the analysis, branch reconstruction was considered to include reconstruction of the inferior mesenteric and inferior renal polar arteries. The general criterion for transfusion was hemorrhagic shock status with hemoglobin level <8.0 g/dL. Epidural anesthesia was administered until postoperative days 4–5, if not contraindicated. Patients were extubated in the operating room,14 and the gastric tubes were removed. Patients usually returned to the recovery room in the general ward and entered the intensive care unit only if they had not been extubated or had unstable vital signs. Postoperatively, we encouraged patients to walk as much as possible on day 1. Water intake was also resumed on day 1. Our criterion for resuming normal diet was confirmation of bowel movement without ileus.15 Patients were discharged after recovery of daily activities without wound complications.
Analysis of Surgical ResultsAll of the young vascular surgeons had completed general surgery training. Some had a small amount of open repair training prior to this study. The operation time, intraoperative blood loss and postoperative complications were evaluated. Postoperative hospital stay, and total hospitalization costs were also analyzed. The operation time was divided into preparation time, clamp time, and total operation time. Preparation time was defined as the time from the incision to the dissection of the proximal and distal clamping sites. Clamp time was defined as the time from clamping of the proximal aorta to removal of the clamp from either iliac artery. Postoperative complications were evaluated according to the criteria of Clavien-Dindo.16 Severe complications were defined as those of Clavien-Dindo classification III–IV. Postoperative mortality included 30-day and in-hospital deaths of any cause.
Statistical AnalysisData were analyzed using JMP 9.0 software (SAS Institute Inc., Cary, NC, USA). Continuous data are expressed as mean±standard deviation. Student’s t-tests were used for the analysis of continuous data and the Chi-squared test was used for categorical data. We applied a cutoff value of P<0.05 for significance.
To analyze the shared learning curve among the 6 young vascular surgeons, the mean operation time of each 10 successive cases was plotted, and the following equation was used to estimate the learning curve:17
y=a×x−b
x, experience; y, operation time; a and b, constants
The intraoperative estimated blood loss was also analyzed using this equation. Cumulative sum (CUSUM) analysis was also used to investigate the learning curves.18,19 CUSUM is a sequential analysis technique to determine changes in a certain parameter of the probability distribution. When applied to operation times, CUSUM can detect sequential differences. The 2 inflection points in the CUSUM curve, which divided the 3 phases of the learning curve, were recorded.
xi, individual operation time; µ, mean of overall operation time
Based on these analyses of the learning curve, the number of procedures needed to establish satisfactory surgical skills was determined.
A total of 562 consecutive patients who underwent OS for intact AAAs or iliac artery aneurysms at Asahi General Hospital, Japan between 2003 and 2017 were eligible for analysis. Patients were divided into 2 groups according to the physician who had operated on them: the attending surgeon group (group A) and the young surgeons’ group (group Y). A total of 330 (58.7%) were treated by an attending surgeon and 232 (41.3%) by 6 young vascular surgeons. The 6 young vascular surgeons had experienced 50, 42, 40, 36, 33, and 31 cases, respectively, over 2–3 years. They had sporadically experienced 0–5 cases of OS for AAA before the training at this institution. The baseline characteristics of the 2 groups are shown in Table 1. The mean age (76.2±8.8 vs. 72.5±8.4 years, P<0.0001); aneurysm diameter (5.9±1.1 vs. 5.6±1.0 cm, P=0.001); and prevalence of hostile abdomen (83 [25%] vs. 42 [18%], P=0.05), chronic heart disease (164 [50%] vs. 87 [38%], P=0.004), respiratory disease (159 [48%] vs. 92 [40%], P=0.05), and chronic kidney disease (199 [60%] vs. 100 [43%], P<0.0001) were significantly different between the 2 groups. However, the operative procedure, such as the use of bifurcated reconstruction or additional procedures, was comparable between the 2 groups (Table 2). Moreover, time-consuming procedures such as suprarenal aortic clamp, branch reconstruction, and other additional procedures were also comparable between the 2 groups (Table 2). In the overall cohort, the mean operation time was 226±57 min, with an estimated blood loss of 465±292 mL. Preoperative autologous blood transfusion was never used (0%), and the intraoperative autotransfusion system was utilized in limited cases (69 patients, 12%); 45 patients (8.0%) needed transfusion in the perioperative period.
| Attending surgeons | Young surgeons | P value | |
|---|---|---|---|
| n | 330 | 232 | |
| Age, years | 76.2±8.8 | 72.5±8.4 | <0.0001 |
| Sex (male) | 272 (82%) | 201 (87%) | 0.18 |
| Aneurysm diameter, cm | 5.9±1.1 | 5.6±1.0 | 0.001 |
| Hostile abdomen | 83 (25%) | 42 (18%) | 0.05 |
| Chronic heart disease | 164 (50%) | 87 (38%) | 0.004 |
| Respiratory disease | 159 (48%) | 92 (40%) | 0.05 |
| Cerebrovascular disease | 49 (15%) | 25 (11%) | 0.16 |
| Chronic kidney disease | 199 (60%) | 100 (43%) | <0.0001 |
| Hypertension | 178 (54%) | 108 (47%) | 0.08 |
| History of smoking | 123 (37%) | 102 (44%) | 0.11 |
| Diabetes mellitus | 85 (26%) | 63 (27%) | 0.71 |
| Attending surgeons (n=330) |
Young surgeons (n=232) |
P value | |
|---|---|---|---|
| Bifurcated reconstruction | 243 (74%) | 178 (77%) | 0.41 |
| Suprarenal clamp | 15 (4.5%) | 4 (1.7%) | 0.07 |
| Additional procedures | 67 (20%) | 37 (16%) | 0.19 |
| Planned | 28 (8.4%) | 21 (9.1%) | 0.81 |
| Tri- or quadro-furcated grafting | 6 (1.8%) | 8 (3.4%) | 0.22 |
| Renal artery bypass | 3 (0.9%) | 1 (0.4%) | 0.51 |
| Branch reconstruction | 14 (4.2%) | 10 (4.3%) | 0.97 |
| Abdominal wall hernia repair | 4 (1.2%) | 0 (0%) | 0.09 |
| Other procedures | 12 (3.6%) | 8 (3.4%) | 0.91 |
| Unplanned | 33 (10%) | 11 (4.7%) | 0.02 |
| Additional bypass | 23 (7.0%) | 9 (3.9%) | 0.12 |
| Thrombectomy | 10 (3.0%) | 2 (0.9%) | 0.08 |
Operation time, intraoperative blood loss, and postoperative complications were analyzed for each successive 10 cases (Figures 1–3). Figure 1A shows the total operation time of the 6 young vascular surgeons. All of them exhibited a similar learning curve trend with decreasing operation time as the number of operations performed increased. When combined, the learning curves of the 6 young vascular surgeons followed a power approximation (y=237.74×−0.105, R2=0.9728, Figure 1B). In contrast, the total operation time of the attending surgeon was almost constant, with a mean of 231±59 min (Figure 1C). The preparation and clamp times in group Y also decreased with increasing experience (Figure 1D). Preparation and clamp times in group A were 72±30 min and 48±10 min, respectively. When applying the mean times of the attending surgeon into the estimation curves of Figure 1B and 1D, the young surgeons were able to achieve the attending surgeon’s preparation, clamp, and total operation times after 10, 30, and 10 cases, respectively. The intraoperative blood loss in group Y was almost constant (416±219 mL, Figure 2) and did not exceed that of group A (499±330 mL). The rate of postoperative complications in group Y did not change during the learning period and was no higher than that observed in group A (overall complications, 20–28.3% vs. 34.2%; severe complications, 0–6.7% vs. 9.1%; Figure 3). A breakdown of the complications is given in Table 3. Although the incidence of each complication was not high, embolic complications, such as cholesterol crystal embolism and bowel ischemia, occurred more frequently in group Y, whereas the incidence of pneumonia and acute renal failure were significantly higher in group A. All-cause mortality rates were not statistically different between the 2 groups (8 [2.4%] vs. 5 [2.2%], P=0.83).
Operation times of the young and attending surgeons. (A) Total operation times of the 6 young vascular surgeons. Analysis was conducted for successive 10 cases. (B) Mean of total operation times of the 6 young vascular surgeons. Analysis was conducted for successive 10 cases. The learning curve of the 6 young vascular surgeons followed a power approximation (y=237.74×−0.105, R2=0.9728). (C) Total operation time of an attending vascular surgeon. Analysis was conducted for successive 10 cases. (D) Mean of preparation and clamp times of 6 young vascular surgeons. Analysis was conducted for successive 10 cases.
Mean of estimated blood loss for 6 young vascular surgeons. Analysis was conducted for successive 10 cases.
Rates of postoperative complications according to Clavien-Dindo classification. Analysis was conducted for successive 10 cases.
| Attending surgeons (n=330) |
Young surgeons (n=232) |
P value | |
|---|---|---|---|
| Postoperative hospital stay, days | 9.1±11.4 | 8.8±9.9 | 0.79 |
| Hospitalization costs, USD | 14,000±6,400 | 13,500±4,100 | 0.33 |
| Any morbidity | 114 (34.5%) | 60 (25.9%) | 0.03 |
| Morbidity with CDC 2–5 | 76 (23.0%) | 38 (16.4%) | 0.05 |
| Severe morbidity (CDC 3–5) | 30 (9.1%) | 13 (5.6%) | 0.13 |
| Acute limb ischemia | 7 (2.1%) | 3 (1.3%) | 0.46 |
| Cholesterol crystal embolism | 1 (0.3%) | 5 (2.2%) | 0.04 |
| Bowel ischemia | 1 (0.3%) | 3 (1.3%) | 0.17 |
| Buttock claudication | 1 (0.3%) | 0 (0%) | 0.40 |
| Neural dysfunction (bladder, rectal and sexual) | 5 (1.5%) | 2 (0.9%) | 0.49 |
| Lymphorrhea | 2 (0.6%) | 0 (0%) | 0.23 |
| Wound dehiscence | 8 (2.4%) | 1 (0.4%) | 0.07 |
| Delirium | 28 (8.4%) | 20 (8.6%) | 0.95 |
| Postoperative bleeding | 5 (1.5%) | 0 (0%) | 0.06 |
| Postoperative transfusion | 10 (3.0%) | 2 (0.9%) | 0.08 |
| Pulmonary embolism/deep venous thrombosis | 2 (0.6%) | 1 (0.4%) | 0.78 |
| Reoperation | 11 (3.3%) | 3 (1.3%) | 0.13 |
| Readmission | 9 (2.7%) | 5 (2.2%) | 0.69 |
| Acute myocardial ischemia | 2 (0.6%) | 0 (0%) | 0.23 |
| Heart failure | 4 (1.2%) | 2 (0.9%) | 0.69 |
| Cerebral infarction | 1 (0.3%) | 1 (0.4%) | 0.80 |
| Pneumonia | 20 (6.1%) | 5 (2.2%) | 0.03 |
| Re-intubation | 2 (0.6%) | 1 (0.4%) | 0.78 |
| Other respiratory disorder | 3 (0.9%) | 2 (0.9%) | 0.95 |
| Acute renal failure | 14 (4.2%) | 2 (0.9%) | 0.02 |
| Ileus/gastroparesis | 12 (3.6%) | 7 (3.0%) | 0.69 |
| Upper GI bleeding | 5 (1.5%) | 3 (1.3%) | 0.83 |
| Other GI tract disorder | 4 (1.2%) | 4 (1.7%) | 0.61 |
| Other | 4 (1.2%) | 1 (0.4%) | 0.33 |
| Any cause of death | 8 (2.4%) | 5 (2.2%) | 0.83 |
CDC, Clavien-Dindo Classification; GI, gastrointestinal; USD, United States dollar (1 USD was calculated as equal to 100 Japanese Yen).
CUSUM analyses of the individual young vascular surgeons’ preparation, clamp, and total operation times were conducted. Figure 4 is a representative CUSUM curve of the total operation time of a young vascular surgeon. Two inflection points divide the CUSUM curve into 3 phases: a learning phase, a consolidation phase, and a maturing phase. As shown in Table 4, in terms of preparation time, young vascular surgeons remained in the learning period for up to 10 cases, and 27 cases were necessary for maturing. Similarly, 27 and 25 cases were necessary for maturing in terms of clamp and total operation times, respectively.
Representative CUSUM curve. The 1 st and 2nd inflection points divide the CUSUM curve into 3 phases: learning phase, consolidation phase, and maturing phase. CUSUM, cumulative sum.
| Preparation time | Clamp time | Total operation time | ||||
|---|---|---|---|---|---|---|
| 1 st IP | 2nd IP | 1 st IP | 2nd IP | 1 st IP | 2nd IP | |
| Y1 (n=50) | 15 | 26 | 15 | 33 | 15 | 33 |
| Y2 (n=42) | 7 | 28 | 10 | 34 | 8 | 29 |
| Y3 (n=40) | 8 | 32 | 8 | 23 | 8 | 25 |
| Y4 (n=36) | 16 | 30 | 5 | 27 | 5 | 18 |
| Y5 (n=33) | 10 | 25 | 10 | 27 | 12 | 27 |
| Y6 (n=31) | 6 | 18 | 6 | 20 | 6 | 18 |
| Mean | 10 | 27 | 9 | 27 | 9 | 25 |
CUSUM, cumulative sum; IP, inflection points; Y1–Y6, 6 different young vascular surgeons. Each data was shown in number.
This study showed that 30 cases were necessary for young vascular surgeons to reach the attending surgeons’ technical level and to reach a level of maturity during OS for intact AAAs. Importantly, during the learning period, young vascular surgeons improved their surgical skills in terms of operation time without negatively affecting patient outcomes.
We believe that this study will contribute to a more efficient approach to the teaching of OS for AAAs in the endovascular era, in which the number of open procedures for AAA has been decreasing. There were some valuable findings in the present study. First, there were shared trends in the learning curve among the 6 young vascular surgeons, which indicated that young surgeons can improve their surgical skills with increasing experience. Assuming that young vascular surgeons have already completed general surgical training, a clear target of 30 AAA cases is useful to achieve technical competency. Second, in contrast to previous studies in which the learning curve of a single surgeon was analyzed, in the present study, we analyzed the surgical results of 6 young vascular surgeons individually as well as in combination. Both analyses revealed that experience with 30 cases was necessary to establish satisfactory surgical skills. By the 30-case threshold, young vascular surgeons had learned to perform the operations with preparation, clamp, and total operation times equivalent to those of an attending surgeon.
The concept of a learning curve for open AAA surgery may also be useful for mentors. It was interesting that young surgeons required the greatest number of cases to achieve clamp times (i.e., hemostasis and anastomosis times) equivalent to those of the attending surgeons. This suggests that hemostasis and anastomosis are technically demanding in the open AAA procedure. Although operations performed in the learning period did not lead to increased blood loss or incidence of complications, stepwise training, such as the attending surgeon partially taking over to perform the anastomosis, may be indicated in some cases.
Recent studies have reported worse outcomes after OS for AAA. In 11 European countries and Australia, the clear trend is towards increasing use of EVAR, where it comprised 44.3% and 60.6% of repairs in 2005–2009 and 2010–2013, respectively.3 Although overall perioperative mortality has fallen over time (3.0% vs. 2.4%, P<0.0001), mortality after open repair has increased (3.9% vs. 4.4%, P=0.008), especially in low-volume centers. In contrast, the perioperative mortality after EVAR has decreased in high-volume centers and remained stable in low-volume centers. In addition, another study from the USA revealed that early reinterventions were 1.6-fold more likely after open repair compared with EVAR and led to a 3.9-fold higher risk of perioperative death.7 The findings from those 2 studies indicate that a lower level of experience with OS is an important contributor to worse outcomes. Because the hospital volume-outcome relationship has been well-characterized,20 centralization of open AAA repair has been suggested as a solution.
The requirement for 30 cases over 2–3 years in our study corresponds to the findings from another report.21 Zettervall et al have shown that higher surgeon and hospital volumes are associated with lower mortality rates, especially after OS for AAA. According to their study, surgeons in the top 2 quintiles had a volume of 9–13 and 14–62 cases per year, respectively, and achieved lower mortality rates. Combining their findings with ours, 10–15 cases per year for a total of 30 cases would contribute to the acquisition of surgical skills for young vascular surgeons. In the endovascular era, because more complex surgeries, such as pararenal abdominal aneurysms, are inclined to be performed by open repair, securing the opportunities for young vascular surgeons’ surgical training is very important.
However, a future prediction model estimated that a vascular resident in 2020 would complete 1–3 cases of OS during their residency.22 If this estimate is true, it will be very difficult for vascular residents to complete 30 cases. Moreover, the time taken to experience 30 cases would have some effect on their progress. How to train young vascular surgeons in such a situation remains a matter of debate. Some believe that surgical simulation is a useful option, while others argue that centralization of OS is important. Regarding the training method, 1 study revealed that standardization (where 1 instructor taught the same procedure repeatedly) was more effective than traditional training (where different surgeons supervise simulation training).23 Another study demonstrated that AAA-specific simulation training was effective for improving surgical skills.8 However, these studies were limited by the relationship between simulation training and the outcomes of practical OS, which comprises complex procedures, not being clarified. Both training approaches may work to some extent; however, vascular surgeons must take the concerns regarding the lack of OS training more seriously.
Despite this adverse training environment, the present study included some encouraging findings regarding OS training. During the learning period of the young vascular surgeons, OS was safely performed without increasing intraoperative blood loss, morbidity, or mortality. As was shown by the CUSUM curve analysis, young vascular surgeons were in the learning phase for up to 10 cases, during which the operation time had room for improvement. From 10 to 30 cases, their operation times did not further increase, which indicated that their surgical skills had stabilized. After 30 cases, the operation time became even shorter, implying maturation of skills. During these learning, consolidation, and maturing phases, the estimated blood loss (Figure 2) and complication rate (Figure 3) were not influenced by the cumulative operative experience. Regarding individual morbidities, the incidences of overall morbidity, pneumonia, and acute renal failure were higher in the attending surgeon group. We believe that the differences are explained by the patients’ backgrounds, whereby patients in the attending surgeon’s group were older, and more frequently had chronic heart disease, respiratory disease, and chronic kidney disease. Conversely, cholesterol crystal embolism occurred more frequently in the young surgeons’ group. However, most complications were mild, and the incidences corresponded to those in former reports.24 Although it may be true that 30 cases are necessary to reach maturity, OS training for young vascular surgeons can be performed with acceptable patient safety.
Study LimitationsSome limitations to the interpretation of this study should be noted. First, this was a retrospective study at a single institution over a relatively long term. Changes in surgical and anesthetic techniques might have some effect on surgical skill acquisition. Moreover, cases were not randomly assigned to attending or young vascular surgeons. Second, although operation time, which was the primary outcome measure in this study, is an important surrogate marker, no single outcome measure can be used to determine surgical skills. Third, because this institution had a small amount of EVAR experience in the study period, even simple infrarenal AAAs were operated on by open repair. The number of cases required for young vascular surgeons to experience might be different in the modern era.
Young vascular surgeons can safely learn open AAA repair without increasing operation time or complications. To gain satisfactory surgical skills, young surgeons need to experience 30 cases. Future studies will be necessary to verify the present results; for example, to compare the surgical outcomes of complex surgeries before and after the surgeon has experienced 30 cases.
None.
