Laparoscopic Sleeve Gastrectomy Changes in the Last Decade: Differences in Morbidity and Weight Loss

Introduction

After the introduction of Laparoscopic Sleeve Gastrectomy (LSG) and reporting of excellent early results by Gagner, these were confirmed also by other authors and LSG spread rapidly as a stand-alone laparoscopic bariatric surgical procedure.^1–5 Today, it is the most frequently performed bariatric procedure, both in Europe and in the United States.^6,7

The aim of this study is to report the learning curve and standardization process of the procedure in terms of improvement of intraoperative and postoperative parameters, morbidity, postoperative body mass index (BMI), and %EBMIL (Excess BMI Loss), in the authors' General Surgery unit in which the surgical activity is not only bariatric but also includes upper gastro intestinal (GI), Hepatobiliary, Adrenal, and lower GI procedures, mainly performed with a minimally invasive approach.

Materials and Methods

This study is a retrospective analysis of prospectively collected data. Institutional review board (IRB) approval and informed consent from all individual participants included in the study were obtained. From October 2006 to January 2016, 127 obese patients underwent LSG in the Unit of “Clinica Chirurgica e Tecnologie Avanzate,” Department of General Surgery, and Surgical Specialties “Paride Stefanini,” Policlinico Umberto I, Sapienza University of Rome, Italy.

At admission in the study, the patients' preoperative assessments in each and every case included the following: physical examination, measure of anthropometric parameters (weight, height, and BMI), blood markers' assay (complete blood count, liver and kidney function tests, total cholesterol, low-density lipoprotein [LDL], high-density lipoproteins [HDL], triglycerides, prothrombin time [PT], partial thromboplastin time [PTT], international normalized ratio [INR], fibrinogen, serum iron, ferritin, folic acid, vitamins B12 and D, plasma electrolytes, basal glycemia, HbA_1c, blood assay for thyroid function, thyroid peroxidase antibodies and thyroglobulin antibodies, oral glucose tolerance test, and serology for HIV, hepatitis B virus [HBV], and hepatitis C virus [HCV]), and instrumental examinations (cardiac, thyroid, and abdominal ultrasound, cardiac and pulmonary evaluation for patients' assessments according to the American Society of Anesthesiologists' risk classification, gastroscopy with biopsies for Helicobacter pylori detection, and polysomnography in patients with Epworth sleepiness scale [ESS] index greater than 8). Upon preoperative study completion, the patients were evaluated by a multidisciplinary team, including an endocrinologist, a surgeon, a psychologist and/or a psychiatrist, a dietician, and an anesthesiologist.

Inclusion criteria were as follows: age between 18 and 65 years, BMI ≥40 or ≥35 kg/m² with relevant comorbidities, absence of H. pylori infection or significant reduction of its titer after specific antibiotic eradication therapy, no alcohol or drug addiction, and patients' signature of a detailed informed consent (“Società Italiana di Chirurgia dell'Obesità,” SICOB). Exclusion criteria were as follows: BMI < 35 kg/m², patients at prohibitively high risk for general anesthesia and induction of pneumoperitoneum, sweet eaters, and pregnancy. The final decision for surgery was taken by the multidisciplinary team after discussion with the patients.

For the purpose of this study, patients were divided into three consecutive Groups (Group A, B and C), based on variations in bougie diameter, in technical and instrumental details (Table 1). The decision to modify the surgical technique and patient management was taken by the authors based on the evidence reported in the literature,^7,8 as well as to improve the postoperative results.

Pneumoperitoneum was established with a Veress needle in Groups A and B and with an open technique in Group C. Five trocars and a 30° optic were used. The first 12 mm trocar was inserted along the midline, midway between the umbilicus and xiphoid process. The second and third 12 mm trocars were placed along the left and right pararectal lines, two fingerbreadths below the costal arch. The fourth 12 mm trocar was placed in a subxiphoid position, right of the midline, and a fifth 5 mm trocar was placed in the left hypochondrium, on the anterior axillary line.

According to the protocol, patients with abdominal wall defects at admission underwent LSG alone, and hernia repair was deferred until after weight loss, unless the defect size was very small.

Primary endpoints were as follows: operative time, conversion rate, intraoperative and postoperative complications (according to Clavien-Dindo classification ⁹ ), hospital stay, associated procedures, and mortality. Leaks were classified according to the International Expert Panel Consensus Statement. ⁸ Secondary endpoints were as follows: postoperative BMI and %EBMIL. Weight regain was defined as regain of ≥15% of lost weight.

Results

Patients' characteristics are shown in Table 2 and total mean operative time is shown in Table 3. In Group A, patients underwent LSG between October 2006 and September 2012, in Group B between October 2012 and November 2013, and in Group C between December 2013 and January 2016. Mean operative time for LSG alone in Groups A, B, and C, excluding the time spent to carry out any associated procedure, was 195.6 ± 51.97 minutes (range 95–300 minutes), 142.9 ± 32.51 minutes (range 90–240 minutes), and 142.9 ± 33.02 minutes (range 90–210 minutes), respectively (A versus B P = .0001; A versus C P = .0007; and B versus C P = .9974).

The single conversion to open surgery occurred in Group A, due to the presence of extensive intraperitoneal adhesions in a patient who had previously undergone incisional hernia repair with the positioning of a large abdominal wall mesh. Major morbidity in this group included early (postoperative day [POD] 14 [7]) proximal staple line leakage (2.7%) in one patient, (grade III-a ⁸ ), treated by percutaneous drainage, enteral nutrition through a nasojejunal feeding tube, and antibiotics (Table 3).

Morbidity in Groups B and C is shown in Table 3. In Group B, two early leaks were treated conservatively (III-a) with antibiotics, percutaneous drainage, and enteral nutrition, whereas two acute leaks were treated surgically (III-b) with laparoscopic peritoneal lavage and gastric suture. Two patients with leakage had undergone prior removal of a failed laparoscopic adjustable gastric banding (LAGB) (two-step procedure). Of 4 patients with bleeding, 3 were treated conservatively with transfusions of red blood cells (II) and 1 underwent reoperation (III-b). In the latter case, at reoperation, bleeding was observed from the short gastric vessels. The patient who presented with a spleen hematoma was treated conservatively with antibiotic therapy (II). One acute leak occurred in Group C on POD 3 and was treated by laparoscopic peritoneal lavage and gastric suture. This same patient showed signs of mediastinitis on POD 7 and underwent immediate left thoracotomy, thoracic lavage, drainage, and esophageal metal stent placement (III-b), followed by an uneventful postoperative course. One case of postoperative bleeding in Group C was managed by transfusions (II) (Table 3).

The types and numbers of associated procedures that were performed are shown in Table 3. At admission, abdominal wall defects were observed in 3 Group A, 2 Group B, and 4 Group C patients, respectively. Due to the presence of a small abdominal wall defect, hernia repair during LSG was performed in 3 cases only (1 Group A and 2 Group C patients). Mean hospital stay is shown in Table 3. Overall morbidity and reintervention rates were 8.6% and 3.1%, respectively. Mortality was nil. Time for leakage resolution ranged between 1 day and 4 months.

After leakage healing, patients' BMI losses were the same as for the patients with no leakage (mean postoperative BMI 28.4 kg/m²). Postoperative BMI and %EBMIL are shown in Table 4. Weight regain was observed in 6 Group A (16.6%) patients and in 1 Group B patient (2.1%) (overall weight regain rate 5.5%) and was managed conservatively in all cases, but one, who underwent resleeve gastrectomy elsewhere (Table 5). Strictures were not observed.

Discussion

This is a retrospective analysis of a consecutive series of patients who underwent LSG, divided in three groups based on progressive reduction in bougie size and differences in patient management and technical details, such as type of instrumentation employed, distance of transection from the pylorus, types of stapler load firings, and distance between the staple line and the bougie. These were modified during the experience based on recommendations published in the literature ⁸ and presented at scientific meetings. The results of these changes were fewer postoperative complications and better weight loss outcomes. At present, the procedure in the authors' center is performed by Fellows and elder Residents as part of their surgical training program, under guidance by an expert tutor.

In the literature,^10,11 there are few studies to define the learning curve of LSG. According to Zacharoulis et al., the learning curve is defined as the surgeons' experience in bariatric surgery or in advanced laparoscopic surgery, case volume, adequate training, multidisciplinary team approach, and institutional staff and facilities. ¹⁰ According to Casella et al., the learning curve is defined as the number of consecutive interventions that are necessary to reach proficiency in a particular procedure. ¹¹ This study was conducted with a multidisciplinary approach by a group experienced in advanced laparoscopic surgery in a General Surgery Unit dedicated to upper GI, Hepatobiliary, Adrenal, and lower GI procedures, where the bariatric experience before the beginning of this study was mainly based on LAGB.^12–14 As reported also by other authors, ¹⁰ operative time in the present series plateaued at the end of Group A (36 cases), but a reduction in the morbidity rate, which is a better definition of proficiency, was observed only in Group C.

In Zacharoulis' series, 102 patients were divided into three groups based on case sequence. ¹⁰ In this article, there was no change in surgical technique and operative time was the only parameter that changed during the learning curve, achieving a plateau at the end of Group B (68 patients). ¹⁰ Casella et al. reported a series of 128 patients without relevant changes in the surgical technique, but a reduction in morbidity rate was observed, although not statistically significant. ¹¹ In this study, mean operative time for LSG alone decreased between Groups A and B, but it remained stable between Groups B and C. However, an increase in the total mean operative time was observed between Groups B and C. This is due to an increase in the number of associated procedures performed in the latter group, mainly hiatoplasties (22.2%, with or without buttressing of the diaphragmatic crura), reflecting greater awareness of the potential increase in gastroesophageal reflux disease (GERD) symptoms after LSG. As recommended in the literature,^8,15 the authors now have a more aggressive approach toward the intraoperative identification and management of hiatal hernia. As reported in the literature, ¹⁵ small hiatal hernia defects (<4 cm²) were treated by simple hiatoplasty, while larger defects (> 4 cm²) were treated by hiatoplasty and absorbable mesh buttressing of the diaphragmatic crura.

In the literature, up to 7% leakage rate after LSG is reported.^16–18 In this series, the overall leakage rate is 4.7% with most leaks observed in Group B. This is probably due to the fact that in this group, the suture line was kept close to the bougie with the aim of creating a narrow sleeve, but increasing at the same time the intragastric pressure above the angulus, with subsequent proximal leakage in most cases. For this reason, in Group C, a “floppy” sleeve was created around a 36F bougie, with the consequence that fewer leaks were observed in this group. In fact, a reduction in the postoperative morbidity rate was observed, although the difference was not statistically significant.

In the authors' opinion, beyond meticulous attention to technical details, a high level of suspicion in the postoperative course is mandatory for early detection and aggressive management of suture line leakage, which is the most dreaded, potentially fatal complication of this procedure. This caveat is highlighted in the literature and immediate reoperation is recommended in patients with fever or tachycardia, even when no radiological signs of complications are present. ⁸

In Groups A and B, intraoperative leak test (IOLT) with methylene blue^® (SALF, Bergamo, Italy) solution and postoperative Gastrografin^® (Bayer, Leverkusen, Germany) swallow before starting the liquid diet after surgery were performed routinely, to test for the absence of staple line leakage. In Group C, IOLT was not performed and the postoperative Gastrografin swallow was performed only selectively upon detection of inflammatory response signs, due to the low incidence of leaks detected by routine IOLT and Gastrografin swallow, as reported in the literature.^19–23 With increasing authors' experiences, the use of peritoneal drainages and nasogastric tubes was abolished, with a decrease in mean hospital stay in Group C.

In this series, obese patients with abdominal wall defects underwent LSG alone, except in the case of very small wall defects (< 2 cm), in which the repair was performed during LSG, abdominal wall repair was postponed after weight loss. The drawback of this approach is the potential for hernia incarceration and the need for emergency surgery in the time interval between the two procedures.

At a mean follow-up of 33.3 months, patients in Group B experienced greater weight loss than in Group A. This is probably related to the reduction in bougie size. In the literature, a bougie size between 32F and 36F is recommended as was used in Group C. ⁸ Group C patients need a longer follow-up duration before drawing any definitive conclusion. Weight regain was observed in 7 patients of the entire series (5.5%), 6 of who were in Group A. The weight regain rate reported in the literature ranges from 5.4% to 19.2%.^19,24 No weight regain has yet been observed in Group C. Although this observation is again probably related to bougie size, a longer follow-up duration is required.

The results of this study confirm that with increasing surgeons' experiences, there is a reduction in operative time, postoperative morbidity, and hospital stay. Unlike the article by Casella et al., ¹¹ however, an increase in the rate of associated procedures was observed in this study. In the authors' opinion, this is due to a more aggressive attitude to prevent postoperative GERD, as recommended by other authors. ⁸ Moreover, the use of a smaller bougie size and beginning the staple line closer to the pylorus, but with antrum preservation, improved patients' weight loss and reduced the weight regain rates.