Do Early Outcomes of Sleeve Gastrectomy Improve the First Year After Completing a Bariatric Surgery Fellowship?

Introduction

Bariatric surgery has been shown to be the most effective, long-term tool for the treatment of morbid obesity.^1–4 As a result, the number of bariatric procedures performed has risen drastically over the past few years. With the increase in demand comes the increasing need for bariatric surgeons that are capable and competent to provide this complex treatment. Bariatric surgery is a fast-growing subspecialty involving technically challenging procedures with long learning curves. It has been suggested that the learning curve for laparoscopic Roux-en-Y gastric bypass (RYGB) is between 75 and 100 cases.^5,6 This learning curve can be shortened or even eliminated by fellowship training. ⁷ While laparoscopic sleeve gastrectomy (SG) is a less complicated procedure, it can impart serious morbidity with devastating complications. It has been suggested that the learning curve of SG is probably in the range of 50–70 cases. ⁸

Bariatric surgery training either by a mentor or by a formal bariatric surgery fellowship can potentially speed-up the learning curve and improve outcomes. ⁹ Some even claim that a designated fellowship can eliminate the learning curve. ⁷ Having said that, it is also important to remember that different surgeons have different skill sets according to their previous training and associated exposure to advanced laparoscopic procedures, such as bariatric surgery.

There is ongoing debate and varying opinions whether an experienced laparoscopic surgeon should start practicing bariatric surgery without formal fellowship training. ⁹ However, studies that evaluate and compare the outcomes of surgeons before and after fellowship are limited. The aim of this study is to compare the early postoperative outcome of patients who underwent SG and were operated on by a single surgeon before and after completion of a fellowship.

Materials and Methods

After Institutional Review Board approval, we retrospectively identified and analyzed all patients who underwent bariatric surgery by a single surgeon (D.F.). The patients were operated on in a single institution between 2013–2014 and 2016–2017. Patients were divided into two groups: before the fellowship (2013–2014) and after (2016–2017). Only cases in which the surgeon operated with a resident as an assistant were included. Any case that was assisted or partially performed by another senior surgeon was excluded. This was done to reduce any interventions by a more experienced surgeon that could affect the outcomes.

Since the surgeon did not perform any RYGB or single anastomosis gastric bypass/mini gastric bypass (MGB) before the fellowship without a senior surgeon assistant, only SG was included before the fellowship. Hence, the comparison of the early postoperative outcome was performed on patients who underwent SG only. Patient's preparation and education, deep vein thrombosis prophylaxis, and surgical technique for SG did not change after fellowship training.

The technique for performing SG included positioning the patient in modified lithotomy position, applying compression stockings and administration of 2 g of second-generation cephalosporin antibiotics and insertion of one 15 mm, two 12 mm, and one 5 mm trocars and a liver retractor. Disconnecting the gastroepiploic vessels from the grater curvature by using Ligasure sealing vessels device (LigaSure™; Medtronic) starting a 4 cm proximal to the pylorus to the angle of His. Complete mobilization of the stomach fundus and exposure of the left crus. Under a 42F bougie calibration, the stomach was stapled (Tri-Staple™; Medtronic). The resected stomach was extracted through the 15 mm trocar. The upper third of the sleeved stomach was reinforced by oversewing with 2.0 nonabsorbable suture. Hemostatic agent was spread over the stapled line (EVICEL^® Fibrin Sealant; Ethicon).

We then compared preoperative demographic data, intraoperative parameters, and postoperative outcomes. We subcategorized postoperative complications to major and minor complications. Major complications included significant bleeding (requiring blood transfusion with or without return to the operating room), staple line leak, or intra-abdominal abscess demonstrated by imaging. Minor complications included complications that did not required invasive intervention, such as gastro esophageal reflux disease symptoms, wound infection, respiratory or urinary infection, and severe constipation. We compared weight loss parameters, which included body mass index (%BMI) loss and % total body weight loss (%TBWL) at the latest follow-up encounter that was documented. Data are presented as mean ± standard deviation. Comparison between paired and unpaired parameters was performed using Student's t-test for continuous variables and Chi-square test for categorical variables. A p-value of <0.05 was defined as statistically significant for the purpose of this study. A univariate regression analysis was performed for selected parameters. Statistical analysis was facilitated by using Microsoft Excel^® statistical tools (Microsoft 365^®).

Results

We identified 49 patients who underwent bariatric surgery before the fellowship. As previously mentioned, all patients in the comparative analysis evaluating outcomes before and after fellowship had SG, although the surgeons practice now included RYGB and MGB. After fellowship, we identified a total of 146 patients who underwent SG, RYGB, or MGB (Table 1). SG patients composed the majority of this group with 117 (80.1%) patients, followed by 24 (16.4%) RYGB and 5 (3.4%) MGB. None of the groups had mortality and none of the operations had to be converted to open.

The RYGB and MGB patients were excluded from the comparison since there was not an appropriate group to compare them with before the fellowship, however, their postoperative course was unremarkable with no postoperative complications. One RYGB patient had an intraoperative event of positive leak test at the gastrojejunostomy. The leak was closed with suture repair and a gastrostomy tube was placed. Her postoperative course was uneventful.

Basic characteristics of those who underwent SG before and after fellowship are presented in Table 2. The postfellowship group were significantly older (42.3 ± 14.6 vs. 36.4 ± 11.7, p = 0.01). The American Society of Anesthesiology (ASA) score was significantly higher (2.29 ± 0.4 vs. 2.1 ± 0.3, p = 0.02) and the prevalence of hyperlipidemia was also significantly higher (33.3% vs. 18.3%, p = 0.05). No other basic characteristic significant differences between the groups were noted. One patient (2.0%) in the prefellowship group had a revision of adjustable gastric band to SG and 2 (1.7%) similar revisions were performed after fellowship.

There were more sliding hiatal hernias which were identified and repaired in the postfellowship group, although not statistical significant (Table 3). The residency year of the assistant was significantly higher in the postfellowship group (5.0 ± 1.8 vs. 3.89 ± 1.6, p < 0.001). Utilization of univariate analysis did not find this factor to be a significant contributor to major complications (−0.009; −0.02–0.001, p = 0.08). There were no significant differences between the groups in operation time, estimated blood loss, length of stay, return to emergency room, 30-day overall complications, and return to the operating room. One patient (2.0%) in the prefellowship group had an intraoperative bleed due to a deep laceration of the liver by the Nathanson retractor. The patient did not require blood transfusion. Two patients (1.7%) in the postfellowship had intraoperative complications. One had an intraoperative bleed from a short gastric vessel. The patient lost ∼1600 cc of blood and required transfusion of two packed red blood cells. The other patient had the oronasal thermometer cut by the stapler. The incident was recognized only when the anesthesiologist notified the surgical team. An upper endoscopy was performed and a 5 mm gap at the stapled line was identified where the thermometer was cut. This was closed with sutures and a leak test confirmation. Both postoperative courses were unremarkable. They were both discharged on postoperative day 4.

Significant differences were observed between the groups when comparing major postoperative complications (bleeding and leaks) before and after the fellowship, respectively [3 (6.9%) vs. 1 (0.8%), p = 0.04]. The postoperative complications are demonstrated in Table 3. There was one patient in each group who had a postoperative bleeding event that required blood transfusion and laparoscopic exploration. The patient who bled in the prefellowship group returned to the emergency department 1 week after the operation with hypotension, abdominal pain, and hemoglobin drop. The source of bleeding could not be identified at exploration. After blood clot evacuation, hemostatic glue was sprayed along the staple line. The patient who bled in the postfellowship group had blood in their drain and hemoglobin drop. She was taken back to the operating room on postoperative day 2. The source of bleeding was found at the upper staple line. Two patients from the prefellowship group returned to the emergency room with fever, abdominal pain, and dysphagia 10 and 12 days after the operation. A computed tomography scan demonstrated a fluid collection with air bubbles next to the sleeve. Both patients underwent CT-guided drainage. They were treated with intravenous antibiotics and total parenteral nutrition. One of those patients also developed left thorax pleural effusions, which necessitated insertion of a chest drain. Both patients eventually recovered within 3 weeks after the second admission. The other postoperative complications that were recorded were minor complications. In the group before fellowship, one patient had a wound infection and one had gastroesophageal reflux complains. In the postfellowship group, 2 patients suffered from gastroesophageal reflux complaints and one returned to the emergency room with constipation.

We analyzed weight loss parameters at the latest follow-up encounter. Follow-up data were obtained in 75% of the patients before fellowship and in 100% of patients after fellowship. After a mean follow-up of 11.7 ± 1.7 and 8.6 ± 1.3 months for before and after fellowship groups, respectively, no significant differences were found in weight, %BMI loss, and %TWL (Table 3).

Discussion

In this study, we compared SG performed before and after completion of a 1-year clinical bariatric surgery fellowship and found significantly improved surgical outcomes after the fellowship. The 30-day rates of leak and bleeding after the fellowship were significantly lower than before (0.8% vs. 6.9%, p = 0.04). There were no leaks in the postfellowship group and only one case of postoperative bleeding that required laparoscopic exploration and blood transfusion. All other outcomes also trended toward improvement in the postfellowship group but were not statistically significant. Patients operated on after fellowship tended to be older and sicker (higher ASA score). Theoretically, that should have put those patients at a greater risk of complications, but our results in this study demonstrated the opposite. In terms of weight loss outcomes, there were no significant differences between the groups.

Laparoscopic bariatric surgery training has a long learning curve. When compared to other bariatric procedures that include intestinal bypass, SG is associated with lower, but still significant risk of major complications. Mastering of surgical technique is crucial and appropriate attention to the smallest details will narrow the margin of error. Advanced laparoscopic skills such as suturing, stapling, and dissection techniques are essential before laparoscopic bariatric surgery can safely be performed with minimal complications. All these translate into a long learning curve during which patients could be subjected to poorer perioperative outcomes.

Fellowship in bariatric surgery is designed to provide a structured educational and training experience in all aspects of bariatric surgery. The American Society for Metabolic and Bariatric Surgery created a standardized bariatric surgery-training curriculum for fellowship programs. ¹⁰ According to these guidelines the fellow has to perform at least 51 stapling/anastomosis of the gastrointestinal track, 10 restrictive procedures, and 5 revision procedures. The fellow must participate in at least 100 bariatric procedures and be the primary surgeon in at least 51%. In addition to technical skills, the fellow has to participate in the evaluation of patients before surgery (50 patients) and in postoperative encounters and outpatient evaluation (100 patients each). Although the training curriculum evolved over the years, ¹¹ fellows are still evaluated by their skills progression. The minimum number of procedures to attain competence in primary and revisional bariatric surgery remains unclear.

The training level of surgery assistant was not homogenous between the groups. Overall, more advanced and experienced residents assisted operations that were performed after fellowship. This factor is especially crucial once an intraoperative complication occurs, such as bleeding. The presence of an experienced assistant can provide an additional set of experienced eyes to recognize technical errors and more easily assist in the management of intraoperative complications. However, a univariate analysis did not suggest that this factor had significant influence on 30-day major morbidity.

We believe that a 1-year clinical fellowship supervised by a designated bariatric surgery expert achieves optimal competency and proficiency that is paramount before starting a bariatric surgery practice. There are surgeons that cannot accomplish an entire year of bariatric surgery training. This can vary based on geographical region. For instance, the demand may be much higher than the available spots for training or some surgeons may not simply be able to pause their practice for a complete year for logistical reasons. Some universities have developed mini-fellowship programs to address these and other concerns. ¹² However, there is lack of data regarding the outcomes of patients operated on by graduates of mini-fellowship in bariatric surgery.

As intensive as a mini-fellowship can be, it will always lack some key features that the 1-year comprehensive fellowships posses. That includes learning how to traverse the short and intermediate-term complications of the different types of surgeries or taking part in crucial decision-making meetings with a multidisciplinary team. In addition, mini-fellowships may not meet every trainee's needs and probably will not eliminate (or lessen) the learning curve. Therefore, this type of training may be better suited for more experienced laparoscopic surgeons.

There are reports that suggest that bariatric surgery can be performed safely in emerging bariatric centers, by an experienced laparoscopic surgeon under the virtual supervision of experienced bariatric surgeons with comparable complication rates. ¹³ Telemonitoring is relatively a new tool for teaching laparoscopy. It allows the mentor surgeon to communicate interactively with the surgeons who are performing the surgery in the operating room. The system allows the mentor to instruct the surgeon by drawing or marking on the surgical image. ¹⁴

To the best of our knowledge, this study is the first to compare early outcomes of SG performed by a single surgeon before and after their bariatric surgery training. As previously stated, the SG learning curve is shorter than the RYGB curve, and we did not find a similar comparison in the literature for RYGB. We tend to believe that the same study comparing RYGB would have shown much wider differences between the groups. ⁹

This study have some limitations, such as incapability of comparing RYGB patients, its retrospective nature, small cohort of patients, and a single surgeon experience only. Follow-up was <1 year, and although involved 100% of patients in the after fellowship group, involved only 75% of the patients who were operated before fellowship. One can also argue that since SG learning curve is not as long, the surgeon probably would have achieved the ideal proficiency to perform a safe SG even without completing a formal bariatric surgery fellowship. However, studies comparing different surgeons with or without formal fellowship training showed significant differences in favor of the fellowship-trained surgeons.^7,9

Conclusions

Bariatric surgery training has a long learning curve. The SG learning curve is relatively short, but still requires considerable competency and skills. Designated bariatric surgery training is associated with reduced major postoperative complications for SG. Surgeons who wish to practice bariatric surgery should pursue a formal bariatric surgery training program before starting their practice.