Analysis of Laparoscopic Sleeve Gastrectomy Learning Curve and Its Influence on Procedure Safety and Perioperative Complications

Analysis of Laparoscopic Sleeve Gastrectomy Learning Curve and Its Influence on Procedure Safety... Purpose Laparoscopic sleeve gastrectomy (LSG) has become an attractive bariatric procedure with promising treatment effects yet amount of data regarding institutional learning process is limited. Materials and Methods Retrospective study included patients submitted to LSG at academic teaching hospital. Patients were divided into groups every 100 consecutive patients. LSG introduction was structured along with Enhanced Recovery after Surgery (ERAS) treatment protocol. Primary endpoint was determining the LSG learning curve’s stabilization point, using operative time, intraoperative difficulties, intraoperative adverse events (IAE), and number of stapler firings. Secondary end- points: influence on perioperative complications and reoperations. Five hundred patients were included (330 females, median age of 40 (33–49) years). Results Operative time in G1–G2 differed significantly from G3–G5. Stabilization point was the 200th procedure using operative time. Intraoperative difficulties of G1 differed significantly from G2–G5, with stabilization after the 100th procedure. IAE and number of stapler firings could not be used as predictor. Based on perioperative morbidity, the learning curve was stabilized at the 100th procedure. The morbidity rates in the groups were G1, 13%; G2, 4%; G3, 5%; G4, 5%; and G5, 2%. The reoperation rate in G1 was 3%; G2, 2%; G3, 2%; G4, 1%; and G5, 0%. Conclusion The institutional learning process stabilization point for LSG in a newly established bariatric center is between the 100th and 200th operation. Initially, the morbidity rate is high, which should concern surgeons who are willing to perform bariatric surgery. . . . Keywords Laparoscopic sleeve gastrectomy Learning curve Surgical training Clinical competence * Michał Wysocki Michał Natkaniec michal92wysocki@gmail.com mrnatk@gmail.com Piotr Małczak Piotr Major pmmalczak@gmail.com majorpiotr@gmail.com Michał Nowakowski Jadwiga Dworak mmnowakowski@gmail.com jadwigadworak@gmail.com Andrzej Budzyński Michał Pędziwiatr andrzej.budzynski@uj.edu.pl michal.pedziwiatr@uj.edu.pl Magdalena Pisarska 2nd Department of General Surgery, Jagiellonian University Medical magdalenapisarska@interia.pl College, Kopernika 21 St, 31-501 Krakow, Poland Mateusz Wierdak Centre for Research, Training and Innovation in Surgery (CERTAIN wierdakmateusz@poczta.onet.pl Surgery), Krakow, Poland Anna Zub-Pokrowiecka Department of Medical Education, Jagiellonian University Medical zuban@poczta.onet.pl College, Kopernika 21 St, 31-501 Krakow, Poland OBES SURG (2018) 28:1672–1680 1673 Introduction m ). In our study, we included 18–65-year-old patients, who agreed to voluntarily use their data in retrospective studies and Laparoscopic sleeve gastrectomy (LSG) has become an attrac- were submitted to LSG as primary treatment for morbid obe- tive, one-stage, and bariatric procedure with promising short- sity. Patients meeting the study inclusion criteria were divided and long-term treatment effects in morbidly obese patients [1, into groups every 100 consecutive patients. The flow of pa- 2]. LSG is considered as one of the least technically challeng- tients through the study is shown in Fig. 1. ing bariatric procedures. Along with high safety profile and The bariatric team that performed the surgeries com- good outcomes, it is often a sufficient reason for selecting prised of six operators, including two attending surgeons LSG as the first-line treatment in newly established bariatric and four surgery residents in their third year of training centers for gaining experience in weight loss surgery. The (out of 6 years). In each case, one attending surgeon and perioperative complication rate of LSG is estimated to be be- one resident were scrubbed. Introduction of bariatric pro- tween 7 and 15%. The mortality risk of LSG ranges between cedures was structured. Initially, the most experienced op- 0.1 and 2% [3–5]. However, these rates are considered higher erator in laparoscopic surgery visited high-volume bariat- in low-volume bariatric centers and in newly established cen- ric centers to participate in bariatric internships. ters [6, 7]. In 1885, a German psychologist, Hermann Procedures then started in the department along with the Ebbinghaus, was the first to describe the learning curve pro- introduction of a treatment protocol. Guest surgeons from cess. A learning curve is currently defined as the amount of high-volume bariatric centers proctored the learning pro- repetition required for a particular activity to establish an ex- cess in the bariatric team in the first 15 cases during pert level of performance in this particular action [8]. With 2 weeks. Requirements for surgeons in training were ac- regard to surgical procedures, a learning curve is defined as quirement of an appropriate theoretical background in the number of consecutive procedures required to become an bariatric surgery and prior experience in laparoscopic sur- expert in carrying out a specific operation. Most authors agree gery (including intra-corporal sewing skills and structured that the LSG learning curve is 50–100 procedures per surgeon training on simulators). Prior to training as an operator in [9, 10]. However, the amount of data regarding the LSG learn- the operating room, apart from gaining a theoretical back- ing curve for the whole bariatric center—the institutional ground, surgeons needed to assist in a minimum of 15 learning process and its effect on the bariatric treatment LSGs. After meeting the requirements, the residents course—is scarce. As in most surgical procedures, the safety started operating as the first operator with the attending of the operation depends mainly on the experience of the team surgeon assisting with the patient. We defined the number performing the procedure, not on the brilliant surgeon of 100 operations as a basic unit for further analysis. To performing all of the operations. It can, therefore, be stated minimize bias, the patients’ care was standardized in ac- that the experience of the laparoscopic center and the entire cordance with the principles of the multimodal Enhanced team rather than individual surgeons translates into complica- Recovery after Surgery (ERAS) pathway, as described tion rate and outcomes. Therefore, with attendees and resi- previously [11–13]. During preoperative counseling, pa- dents as main operators, this study aimed to analyze the sta- tients aged over 50 years, with maximal preoperative bilization point of the institutional learning process for bariat- BMI over 45 kg/m , with obesity-related comorbidities ric team of six operators, including two attending surgeons (especially type 2 diabetes mellitus and gastroesophageal and four surgery residents, in a newly established bariatric reflux disease, GERD), hiatal hernia, and previous gastric center. We also attempted to analyze the effect of this learning resections were factors indicating qualification for process on patients’ safety and the incidence of perioperative LRYGB, not for LSG in our facility. Routine endoscopy complications. of the upper gastrointestinal tract was done with assess- ment of the incidence of hiatal hernia and esophageal, gastric or duodenal mucosa pathology. If there were Material and Methods symptoms of GERD and endoscopic findings correspond- ing with GERD, patient was not qualified to LSG. We performed a retrospective analysis of prospectively col- With regard to the surgical technique, it was standardized lected data of patients who were submitted to surgical treat- as described previously [14]. ment of morbid obesity in the Second Department of General The analyzed group was divided into subgroups according Surgery, Jagiellonian University Medical College (academic to the order of the procedure: G1, group number 1 (1st–100th teaching hospital and tertiary referral center for general sur- LSG); G2, group number 2 (101st–200th LSG); G3, group gery). Criteria for surgical treatment were in accordance with number 3 (201st–300th LSG); G4, group number 4 (301st– the guidelines of the Metabolic and Bariatric Surgery 400th LSG); and G5, group number 5 (401st–500th LSG). Section of the Polish Surgical Society (i.e., body mass index The primary endpoint was determining the stabilization [BMI] ≥ 35 kg/m with obesity comorbidities or BMI ≥ 40 kg/ point of the LSG learning curve in each group using operative 1674 OBES SURG (2018) 28:1672–1680 Fig. 1 Study flow chart Enrollment Assessed for eligibility (n=775) Excluded (n=275) Other Procedure (n=258) Lack of necessary data (n=5) Declined to participate (n=12) Allocation Allocated to LSG (n=500) Received allocated intervention (n=500) Analysis Allocated to groups (n=500) st th Group 1 – 1 -100 patient (n=100) st th Group 2 – 101 -200 patient (n=100) st th Group 3 – 201 -300 patient (n=100) st th Group 4 – 301 -400 patient (n=100) st Group 5 – 401 -500 patient (n=100) time, intraoperative difficulties, intraoperative adverse events, range, and odds ratios (ORs) with 95% confidence intervals and the number of stapler firings. As secondary endpoints, the (CIs) when appropriate. To assess statistical significance of effects of the learning curve on perioperative complications qualitative data differences in subgroups, Pearson’schi- and re-operation rate were assessed. square test and discriminant analysis were used. Quantitative Directly after each procedure, every surgeon who per- data were analyzed with the Kruskal–Wallis ANOVA and formed the operation as the main operator was obligated post-hoc testing. Univariate logistic regression was used to to note down the intraoperative difficulties, which were calculate ORs with 95% confidence intervals. Results were defined as surgeon-reported obstacles during the opera- considered statistically significant when the p value was less tion. These obstacles were additional measures that were than 0.05. required to finish the procedure or those that significantly prolonged the procedure. These obstacles also included Patients difficulty in achieving a sufficient working space, difficul- ty in proper setting of the stapler, intra-abdominal anato- From April 2009 to October 2017, 775 patients were my obstructing performance of the surgery, or the need for treated for morbid obesity at the Second Department of assistance from a supervisor. The definition of intraoper- Surgery, Jagiellonian University Medical College, includ- ative difficulties is similar to that reported by other au- ing 500 consecutive patients qualified to LSG. A total of thors [15, 16]. Intraoperative adverse events were defined 500 patients were included in the study [330 females, 170 as any iatrogenic, adverse event during the operation, males median age of 40 (33–49) years, median BMI of which was not derived from the standard LSG technique. 44.84 (min. 34.01; max. 76.44) kg/m ]. Patients’ charac- Perioperative complications were defined as adverse teristics are shown in Table 1. events occurring within 30 days after the procedure. These events were classified according to the Clavien– Dindo classification [17]. Rhabdomyolysis was defined Results as an elevated serum creatine phosphokinase concentra- tion of > 1000 U/L with a concomitant increase in myo- Operative Time and First Operators globin concentrations. The median operative time for LSG was 90 (70–120) minutes. Statistical Analysis The Kruskal–Wallis ANOVA showed that there was a signif- icant difference in the median operative time between the All data were analyzed with Statsoft Statistica version 12.0 PL groups (p <0.001) (Table 2). Multiple comparisons of the me- (StatSoft Inc., Tulsa, OK, USA). The results are presented as dian ranges for all groups showed significant differences be- mean ± standard deviation (SD), median and interquartile tween G1–G2 and G3–G5. Based on multiple comparisons of OBES SURG (2018) 28:1672–1680 1675 the median range tests, the LSG learning curve’sstabilization point was the 200th procedure for the whole newly established bariatric center (Fig. 2). During a study period, we observed a significant decrease in the number of operations performed by the attendees in favor of operations performed by the residents as first operators (Pearson’stest, p <0.001). Conversions All procedures were completed laparoscopically. Intraoperative Difficulties Intraoperative difficulties were reported during 23 (4.6%) pro- cedures (Table 2). Using the group number as a factor in univariate logistic regression, the OR for the occurrence of intraoperative difficulties diminished with every 100 per- formed operations (OR, 0.68; 95% CI, 0.49–0.94; p =0.019) and affected, respectively, in G1, 12 patients (12%); G2, two (2%); G3, three (3%); G4, two (2%); and G5, four (4%). According to Pearson’s chi-square test, there was a significant difference in the occurrence of intraoperative difficulties among the groups (p= 0.003). Based on intraoperative diffi- culties, stabilization point was at the 100th procedure in dis- criminant analysis. Intraoperative Adverse Events Intraoperative adverse events were observed during nine (1.8%) operations (Table 2). In G1, they occurred in three patients, G2—none, G3—three patients, G4—two patients, and G5—two patients. Due to lack of variability, this pa- rameter could not be used for estimating stabilization point (p = 0.607). Number of Stapler Firings The median number of stapler firings in the groups was the following: G1, five (4–5); G2, four (4–5); G3, four (4, 5); G4, four (4, 5); and G5, five (4, 5). The Kruskal–Wallis ANOVA showed a significant difference in the number of stapler firings among the groups (p < 0.001). Multiple comparisons of the median and range of stapler firings for all groups showed that G1 and G5 differed from G2 and G4. Perioperative Morbidity Perioperative complications were diagnosed in 29 (7.6%) pa- tients (Table 3). During the study period, we diagnosed 13 patients with perioperative complications of Clavien–Dindo class I, 4 patients with class II, 10 patients with class III, and 2 patients with class IV. Detailed characteristic of periopera- tive morbidity regarding Clavien–Dindo scale is presented in Table 1 Patients’ characteristics Demography Total G1 G2 G3 G4 G5 P value Females, n (%) 330 (66) 69 63 64 70 64 0.759 Males, n (%) 170 (34) 31 37 36 30 36 Median age, years (IQR) 40 (33–49) 39 (32.5–47) 41 (33–48.5) 40 (33–49) 39 (32–47) 42 (34–50) 0.400 Median BMI at qualification, kg/m (IQR) 46.44 (42.91–51.37) 46.56 (44.10–50.85) 44.62 (41.98–49.22) 46.71 (42.73–51.37) 45.81 (42.76–52.73) 47.11 (43.94–54.65) 0.345 Median BMI on admission, kg/m (IQR) 44.84 (41.33–49.22) 45.48 (42.45–49.80) 43.10 (41.04–47.66) 45.49 (41.14–49.12) 44.01 (40.88–49.51) 45.0 (41.87–50.20) 0.143 Cardiovascular diseases, n(%) 61(12.2) 151612126 0.226 Arterial hypertension, n (%) 320 (64) 59 71 62 55 73 0.034 Respiratory diseases, n(%) 95(19) 2214221522 0.364 Diabetes mellitus, n (%) 119 (24) 22 21 28 24 24 0.808 Dyslipidemia, n (%) 180 (36) 60 50 38 16 16 < 0.001 Fatty liver disease, n (%) 219 (44) 65 60 56 15 23 < 0.001 G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st–300th LSG), G4 group number 4 (301st–400th LSG), G5 group number 5 (401st–500th LSG) 1676 OBES SURG (2018) 28:1672–1680 Table 2 Operative time, operating surgeons, intraoperative difficulties, and intraoperative adverse events Total G1 G2 G3 G4 G5 Median operative time (IQR) 90 (70–120) 130 (100–160) 100 (80–120) 80 (70–100) 90 (70–110) 80 (65–95) First operator Attending 296 (59%) 91/100 87/100 53/100 35/100 30/100 Resident 204 (41%) 9/100 13/100 47/100 65/100 70/100 Intraoperative difficulties Total 23/500 (4.6%) 12/100 (12%) 2/100 (2%) 3/100 (3%) 2/100 (2%) 4/100 (4%) Difficulty to achieve sufficient working space 6 (1.2%) 4 0 1 0 1 Difficulty of proper setting the stapler 7 (1.4%) 5 0 1 0 1 Intra-abdominal adhesions obstructing 4 (0.8%) 2 0 0 1 1 performance of the surgery Fatty liver disease obstructing the procedure 1 (0.2%) 0 1 0 0 0 Irreducible umbilical hernia 1 (0.2%) 0 1 0 0 0 Large umbilical hernia 1 (0.2%) 0 0 0 1 0 Required of the help of the mentor 1 (0.2%) 1 0 0 0 0 Difficulty to achieve sufficient hemostasis 2 (0.4%) 0 0 1 0 1 Intraoperative adverse events Total 9/500 (1.8%) 3/100 (3%) 0/100 (0%) 2/100 (3%) 2/100 (2%) 2/100 (2%) Intraoperatively diagnosed leakage, supplied 3 (0.6%) 3 0 0 0 0 with the additional suturing Intraoperatively diagnosed leakage, supplied 1 (0.2%) 0 0 1 0 0 with setting of stent and drainage Excessive intraoperative blood loss 1 (0.2%) 0 0 1 0 0 Bleeding from liver/spleen 2 (0.4%) 0 0 0 1 1 Bleeding from stapler line 1 (0.2%) 0 0 0 1 0 Gastric content leak from resected part 1 (0.2%) 0 0 0 0 1 G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st–300th LSG), G4 group number 4 (301st– 400th LSG), G5 group number 5 (401st–500th LSG) Table 3. When we used group number as a factor in the uni- Mortality variate logistic regression model, the OR of perioperative morbidity significantly decreased with every 100 performed None of the patients died during the 30-day perioperative procedures (i.e., with every consecutive group) (OR, 0.66; period. 95% CI, 0.50–0.89; p = 0.006). The morbidity rates in the groups were G1, 13%; G2, 4%; G3, 5%; G4, 5%; and G5, 2%, respectively. Pearson’s chi-square test showed that there Discussion was a significant difference in the occurrence of morbidity among the groups (p = 0.011). Based on perioperative morbid- Bariatric surgery in Poland is still a developing discipline. We ity, the learning curve stabilized at the 100th procedure in describe a 9-year experience of LSG in the Second discriminant analysis. Department of General Surgery, JUMC [18]. We aimed to assess our institutional learning process for LSG and the effect of this process on safety and perioperative complications. Reoperations Operative time is frequently regarded as a descriptive pa- rameter for evaluating the learning curve. In our study, the Reoperations were necessary in eight (1.8%) patients. operative time was an average of 90 min. Zacharoulis et al. Pearson’s chi-square test did not reveal significant differences assumed that one of the main determinants of an LSG learning of reoperations in groups (p = 0.508). It was also nonsignifi- curve’s stabilization point is the operative time and estimated cant in univariate logistic regression model (p = 0.344). that 68 consecutive procedures are essential to finish training Basing on reoperations, this parameter could not be used for [19]. Daskalakis et al. observed a significant difference in setting stabilization point. surgical duration between the first and second halves of their OBES SURG (2018) 28:1672–1680 1677 Fig. 2 Graph illustrating operative time changes in institutional learning curve Median IQR Case number Min.-Max. 230 consecutive LSGs [20]. A learning curve’sstabilization We reported intraoperative difficulties during 23 (4.6%) point is usually derived from the literature and tested for indi- procedures. In our study, the OR of the occurrence of intraop- vidual operators to achieve expertise. In our study, we to esti- erative difficulties decreased with every 100 performed LSGs. mate the stabilization point for the whole department by test- Intraoperative difficulties were mainly related to a narrow op- ing the median and range within consecutively operated erative space (six cases) and to proper stapler setting difficulty groups of 100 patients. In our study, the institutional learning (seven cases). Unfortunately, we could not find any previously process’ stabilization point was the 200th procedure. published studies regarding intraoperative difficulties, which Table 3 Perioperative (≤ 30 days) morbidity regarding Clavien–Dindo scale and reoperations C-D Complications No. (%) G1 G2 G3 G4 G5 Total 29/500 (7.6%) 13/100 (13%) 4/100 (4%) 5/100 (5%) 5/100 (5%) 2/100 (2%) 4b Cardiorespiratory failure (ICU stay) 1 (0.2%) 0 1 0 0 0 4a Pulmonary embolism (thrombolysis) 1 (0.2%) 0 0 0 0 1 3b GI leak (relaparoscopy) 4 (0.8%) 3 1 0 0 0 Operation site bleeding (relaparoscopy) 4 (0.8%) 0 1 2 1 0 Wound infection (open drainage) 1 (0.2%) 0 0 0 1 0 3a GI stricture (tube placement) 1 (0.2%) 0 0 1 0 0 2 Varicella infection 1 (0.2%) 0 0 1 0 0 TIA 1 (0.2%) 0 0 0 1 0 Superior mesenteric vein thrombosis (thrombolysis) 1 (0.2%) 0 0 0 1 0 Renal colic 1 (0.2%) 0 0 0 0 1 1 Delayed gastric emptying due to temporary stricture * 5 (1%) 2 1 1 1 0 Dehydration* 1 (0.2%) 1 0 0 0 0 Prolonged drainage 1 (0.2%) 1 0 0 0 0 Rhabdomyolysis 6 (1.2%) 6 0 0 0 0 Reoperations 8 (1.8%) 3 2 2 1 0 *1 patient was diagnosed with both complications C-D Clavien-Dindo classification grade, G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st– 300th LSG), G4 group number 4 (301st–400th LSG), G5 group number 5 (401st–500th LSG), TIA transient ischemic attack 1-100 101-200 Learning curve's stabilization point 201-300 301-400 401-500 Operative time [min] 1678 OBES SURG (2018) 28:1672–1680 are important parameters for assessment of the learning curve reported by Daskalakis et al. (7.4%) [20]. In our study, due to [21, 22]. Intraoperative difficulties were self-reported by sur- low variability of reoperation rate, we could not draw conclu- geons and defined as obstacles requiring additional measures sions from this parameter. during the surgery. We estimated that the intraoperative diffi- None of the patients died during the 30-day perioperative culty rate stabilized after 100 procedures. period. This finding is similar to that by Casella et al. [28]and In our study, intraoperative adverse events, defined as Trastulli et al. [29]. However, Zacharoulis et al. showed that harmful events related to surgery, occurred during nine the mortality rate was 0.98% [19]. Taking into consideration, (1.8%) procedures, and due to lack of variability, they could the mortality of LSG as reported in a systematic review by Shi not be used for estimation of the institutional learning process’ et al. (0–3.3%), the lack of morality in our study on the course stabilization point. There were four cases (0.8%) of intraoper- of the learning curve is satisfactory [26]. atively diagnosed gastrointestinal leak. Similarly, Rubin et al. Usually, a learning curve is evaluated using selected pa- reported that intraoperative adverse events occurred during rameters for particular surgeon. We assessed the institutional 3.33% of 120 LSGs. They found one case of intraoperatively learning process using intraoperative difficulties, periopera- diagnosed gastric leak, two cases of short gastric vessel bleed- tive morbidity, and operative time, while obtaining accurate ing, and one case of probe immobilization in the staple line results. Operative time of the first 200 procedures was signif- [23]. Braghetto et al. reported a rate of 14% for intraoperative icantly longer than that of the next 300 surgeries. Based on adverse events after introduction of LSG in a series of 50 this, the institutional learning process stabilization point was LSGs (four cases of intraoperatively diagnosed gastric leak, the 200th procedure for the whole newly established bariatric one case of short gastric vessel bleeding, one case of liver center. Intraoperative difficulties in the first 100 patients capsule perforation with a retractor, and one case of probe (12%) were significantly more prevalent than that in the next immobilization in the staple line) [24]. 400 procedures (2.75%). Based on intraoperative difficulties, After 100 procedures, the amount of stapler firings de- the institutional learning process’ stabilization point was at the creased by one firing but then in the last 100 patients, it 100th procedure. Finally, the postoperative morbidity rate in increased by one. Kaska et al. found that, to avoid adverse the first 100 patients of 13% was much higher than that in the events and complications, adequately choosing the car- next 400 cases (4%). Based on morbidity, the institutional tridge and over-sewing the staple line are essential but learning process was stabilized at the 100th procedure. the number of stapler firings was not analyzed [16]. We Summarizing all endpoints that enabled to assess stabilization believe that this increase comes just from taking adequate point, we proposed that the institutional learning process sta- measures to operative conditions. Our beliefs are support- bilized between the 100th and 200th case. ed by lack of increase in intraoperative adverse events and This study has several limitations. Our institution is a re- complications. ferral center for general surgery and a teaching hospital. The conversion rate for LSG ranges from 1.05 to 1.85% Therefore, our results probably cannot be extrapolated to all [25]. Despite these rates, in our study, all procedures were hospitals. Additionally, this was a retrospective study. A pro- completed laparoscopically. spective study should be designed to confirm these findings. The LSG learning curve’s stabilization point can also be Finally, the bariatric team consisted of surgeons of varying estimated using perioperative morbidity. In our study, periop- levels of experience, which may have affected the results. A erative morbidity was 7.6%. An almost identical rate of the larger amount of operated patients would enable the use of all LSG learning curve was described by Zacharoulis et al. of the parameters we attempted to test for describing the learn- (7.8%) [19]. Our result is satisfactory compared with the re- ing curve. Further research should be performed using intra- ported a major morbidity rate of 12.1% (0–29%) in a system- operative difficulties as a self-reported assessment for evalua- atic review by Shi et al. [26]. Regardless of implementation of tion of the learning curve process. the ERAS protocol along with introduction of LSG, we ob- served a relatively high perioperative morbidity of 13% in the first 100 patients. Morbidity then decreased to 4% in the rest of the consecutive patients. Discriminant analysis enabled us Conclusion to set the LSG learning curve’s stabilization point based on morbidity at the 100th LSG. Zachariah et al. showed that the The institutional learning process’ stabilization point for LSG perioperative complication rate in the first 50 patients was in a newly established bariatric center is between the 100th significantly higher than that in the last 178 patients (8 vs. and 200th operation. LSG, which is carried out with a learning 1.6%, p =0.02) [27]. Casella et al. showed that perioperative curve, does not affect the safety of the procedure in terms of morbidity was significantly decreased after the 88th LSG [28]. intraoperative adverse events. Initially, with introduction of In our study, eight (1.8%) patients required reoperations. LSG, the morbidity rate is high, which should concern sur- This reoperation rate was significantly smaller than the rate geons who are willing to start performing bariatric surgery. OBES SURG (2018) 28:1672–1680 1679 Funding This research was supported by the Faculty of Medicine, 9. Romero RJ, Kosanovic R, Rabaza JR, et al. Robotic sleeve gastrec- Jagiellonian University Medical College, Leading National Research tomy: experience of 134 cases and comparison with a systematic Centre (KNOW) 2012–2017. review of the laparoscopic approach. Obes Surg. 2013;23(11): 1743–52. https://doi.org/10.1007/s11695-013-1004-1. 10. Buchwald H, Williams SE. Bariatric surgery worldwide 2003. Obes Compliance with Ethical Standards Surg. 2004;14(9):1157–64. https://doi.org/10.1381/ Conflict of Interest The authors declare that they have no conflict of 11. Major P, Wysocki M, Pędziwiatr M, et al. Risk factors for compli- interest. cations of laparoscopic sleeve gastrectomy and laparoscopic Roux- en-Y gastric bypass. Int J Surg. 2017;37:71–8. https://doi.org/10. Ethical Responsibilities of Authors Article follows the COPE 1016/j.ijsu.2016.12.012. guidelines. 12. Major P, Wysocki M, Torbicz G, et al. Risk factors for prolonged length of hospital stay and readmissions after laparoscopic sleeve Ethical Approval All procedures performed in studies involving human gastrectomy and laparoscopic Roux-en-Y gastric bypass. Obes participants were in accordance with the ethical standards of the institu- Surg. 2017; https://doi.org/10.1007/s11695-017-2844-x. tional and/or national research committee and with the 1964 Helsinki 13. Matłok M, Pędziwiatr M, Major P, et al. One hundred seventy-nine declaration and its later amendments or comparable ethical standards. consecutive bariatric operations after introduction of protocol in- The study was approved by the Bioethics Committee of Jagiellonian spired by the principles of enhanced recovery after surgery University (KBET/62/B/2011). (ERAS®) in bariatric surgery. Med Sci Monit. 2015;21:791–7. https://doi.org/10.12659/MSM.893297. 14. Major P, Wysocki M, Pędziwiatr M, et al. More stapler firings Informed Consent Statement Informed consent was obtained from all increase the risk of perioperative morbidity after laparoscopic individual participants included in the study. sleeve gastrectomy. Videosurgery and Other Miniinvasive Techniques/Wideochirurgia i inne techniki małoinwazyjne. 2017; Open Access This article is distributed under the terms of the Creative https://doi.org/10.5114/wiitm.2017.70197. Commons Attribution 4.0 International License (http:// 15. Korenkov M, Gundlach M, Heimbucher J, et al. The impact of creativecommons.org/licenses/by/4.0/), which permits unrestricted use, preoperative classification of expected intraoperative difficulties. distribution, and reproduction in any medium, provided you give appro- A multicenter study. Chirurg. 2014;85(11):999–1004. https://doi. priate credit to the original author(s) and the source, provide a link to the org/10.1007/s00104-013-2702-x. Creative Commons license, and indicate if changes were made. 16. Kaska L, Proczko M, Stefaniak T, et al. Redesigning the process of laparoscopic sleeve gastrectomy based on risk analysis resulted in 100 consecutive procedures without complications. Wideochir Inne Tech Maloinwazyjne. 2013;8(4):289–300. https://doi.org/10.5114/ References wiitm.2011.34797. 17. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 1. Sánchez-Santos R, Masdevall C, Baltasar A, et al. Short- and mid- 6336 patients and results of a survey. Ann Surg. 2004;240(2): term outcomes of sleeve gastrectomy for morbid obesity: the expe- 205–13. rience of the Spanish National Registry. Obes Surg. 2009;19(9): 18. Wylezol MS, Zwirska-Korczala K, Pasnik K. Bariatric surgery in 1203–10. https://doi.org/10.1007/s11695-009-9892-9. Poland from 1993 to 2003. J Physiol Pharmacol. 2005;56(Suppl 6): 2. Arias E, Martínez PR, Ka Ming Li V, et al. Mid-term follow-up after 109–15. sleeve gastrectomy as a final approach for morbid obesity. Obes 19. Zacharoulis D, Sioka E, Papamargaritis D, et al. Influence of the Surg. 2009;19(5):544–8. https://doi.org/10.1007/s11695-009- learning curve on safety and efficiency of laparoscopic sleeve gas- 9818-6. trectomy. Obes Surg. 2012;22(3):411–5. https://doi.org/10.1007/ 3. Sanni A, Perez S, Medbery R, et al. Postoperative complications in s11695-011-0436-8. bariatric surgery using age and BMI stratification: a study using 20. Daskalakis M, Berdan Y, Theodoridou S, et al. Impact of surgeon ACS-NSQIP data. Surg Endosc. 2014;28(12):3302–9. https://doi. experience and buttress material on postoperative complications org/10.1007/s00464-014-3606-7. after laparoscopic sleeve gastrectomy. Surg Endosc. 2011;25(1): 4. Dumon KR, Murayama KM. Bariatric surgery outcomes. Surg Clin 88–97. https://doi.org/10.1007/s00464-010-1136-5. North Am. 2011;91(6):1313–1338, x. https://doi.org/10.1016/j.suc. 21. Mammen JM, Fischer DR, Anderson A, et al. Learning styles vary 2011.08.014. among general surgery residents: analysis of 12 years of data. J Surg 5. Picot J, Jones J, Colquitt JL, et al. The clinical effectiveness and Educ. 2007;64(6):386–9. https://doi.org/10.1016/j.jsurg.2007.08. cost-effectiveness of bariatric (weight loss) surgery for obesity: a systematic review and economic evaluation. Health Technol 22. Quillin RC, Pritts TA, Hanseman DJ, et al. How residents learn Assess. 2009;13(41):1–190, 215–357, iii–iv. https://doi.org/10. predicts success in surgical residency. J Surg Educ. 2013;70(6): 3310/hta13410. 725–30. https://doi.org/10.1016/j.jsurg.2013.09.016. 6. Celio AC, Kasten KR, Brinkley J, et al. Effect of surgeon volume 23. Rubin M, Yehoshua RT, Stein M, et al. Laparoscopic sleeve gas- on sleeve gastrectomy outcomes. Obes Surg. 2016;26(11):2700–4. trectomy with minimal morbidity. Early results in 120 morbidly https://doi.org/10.1007/s11695-016-2190-4. obese patients. Obes Surg. 2008;18(12):1567–70. https://doi.org/ 7. Birkmeyer NJ, Dimick JB, Share D, et al. Hospital complication 10.1007/s11695-008-9652-2. rates with bariatric surgery in Michigan. JAMA. 2010;304(4):435– 24. Braghetto I, Korn O, Valladares H, et al. Laparoscopic sleeve gas- 42. https://doi.org/10.1001/jama.2010.1034. trectomy: surgical technique, indications and clinical results. Obes 8. Sauerland S, Angrisani L, Belachew M, et al. Obesity surgery: Surg. 2007;17(11):1442–50. https://doi.org/10.1007/s11695-008- 9421-2. evidence-based guidelines of the European Association for 25. Rosenthal RJ, Diaz AA, Arvidsson D, et al. International Sleeve Endoscopic Surgery (EAES). Surg Endosc. 2005;19(2):200–21. https://doi.org/10.1007/s00464-004-9194-1. Gastrectomy Expert Panel Consensus Statement: best practice 1680 OBES SURG (2018) 28:1672–1680 guidelines based on experience of >12,000 cases. Surg Obes Relat 28. Casella G, Soricelli E, Giannotti D, et al. Learning curve for lapa- roscopic sleeve gastrectomy: role of training in a high-volume bar- Dis. 2012;8(1):8–19. https://doi.org/10.1016/j.soard.2011.10.019. 26. Shi X, Karmali S, Sharma AM, et al. A review of laparoscopic iatric center. Surg Endosc. 2015;30(9):3741–8. https://doi.org/10. sleeve gastrectomy for morbid obesity. Obes Surg. 2010;20(8): 1007/s00464-015-4670-3. 1171–7. https://doi.org/10.1007/s11695-010-0145-8. 29. Trastulli S, Desiderio J, Guarino S, et al. Laparoscopic sleeve gas- 27. Zachariah SK, Chang PC, Ooi AS, et al. Laparoscopic sleeve gas- trectomy compared with other bariatric surgical procedures: a sys- trectomy for morbid obesity: 5 years experience from an Asian tematic review of randomized trials. Surg Obes Relat Dis. center of excellence. Obes Surg. 2013;23(7):939–46. https://doi. 2013;9(5):816–29. https://doi.org/10.1016/j.soard.2013.05.007. org/10.1007/s11695-013-0887-1. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Obesity Surgery Springer Journals

Analysis of Laparoscopic Sleeve Gastrectomy Learning Curve and Its Influence on Procedure Safety and Perioperative Complications

Free
9 pages
Loading next page...
 
/lp/springer_journal/analysis-of-laparoscopic-sleeve-gastrectomy-learning-curve-and-its-KdfJ7pUmTe
Publisher
Springer US
Copyright
Copyright © 2017 by The Author(s)
Subject
Medicine & Public Health; Surgery
ISSN
0960-8923
eISSN
1708-0428
D.O.I.
10.1007/s11695-017-3075-x
Publisher site
See Article on Publisher Site

Abstract

Purpose Laparoscopic sleeve gastrectomy (LSG) has become an attractive bariatric procedure with promising treatment effects yet amount of data regarding institutional learning process is limited. Materials and Methods Retrospective study included patients submitted to LSG at academic teaching hospital. Patients were divided into groups every 100 consecutive patients. LSG introduction was structured along with Enhanced Recovery after Surgery (ERAS) treatment protocol. Primary endpoint was determining the LSG learning curve’s stabilization point, using operative time, intraoperative difficulties, intraoperative adverse events (IAE), and number of stapler firings. Secondary end- points: influence on perioperative complications and reoperations. Five hundred patients were included (330 females, median age of 40 (33–49) years). Results Operative time in G1–G2 differed significantly from G3–G5. Stabilization point was the 200th procedure using operative time. Intraoperative difficulties of G1 differed significantly from G2–G5, with stabilization after the 100th procedure. IAE and number of stapler firings could not be used as predictor. Based on perioperative morbidity, the learning curve was stabilized at the 100th procedure. The morbidity rates in the groups were G1, 13%; G2, 4%; G3, 5%; G4, 5%; and G5, 2%. The reoperation rate in G1 was 3%; G2, 2%; G3, 2%; G4, 1%; and G5, 0%. Conclusion The institutional learning process stabilization point for LSG in a newly established bariatric center is between the 100th and 200th operation. Initially, the morbidity rate is high, which should concern surgeons who are willing to perform bariatric surgery. . . . Keywords Laparoscopic sleeve gastrectomy Learning curve Surgical training Clinical competence * Michał Wysocki Michał Natkaniec michal92wysocki@gmail.com mrnatk@gmail.com Piotr Małczak Piotr Major pmmalczak@gmail.com majorpiotr@gmail.com Michał Nowakowski Jadwiga Dworak mmnowakowski@gmail.com jadwigadworak@gmail.com Andrzej Budzyński Michał Pędziwiatr andrzej.budzynski@uj.edu.pl michal.pedziwiatr@uj.edu.pl Magdalena Pisarska 2nd Department of General Surgery, Jagiellonian University Medical magdalenapisarska@interia.pl College, Kopernika 21 St, 31-501 Krakow, Poland Mateusz Wierdak Centre for Research, Training and Innovation in Surgery (CERTAIN wierdakmateusz@poczta.onet.pl Surgery), Krakow, Poland Anna Zub-Pokrowiecka Department of Medical Education, Jagiellonian University Medical zuban@poczta.onet.pl College, Kopernika 21 St, 31-501 Krakow, Poland OBES SURG (2018) 28:1672–1680 1673 Introduction m ). In our study, we included 18–65-year-old patients, who agreed to voluntarily use their data in retrospective studies and Laparoscopic sleeve gastrectomy (LSG) has become an attrac- were submitted to LSG as primary treatment for morbid obe- tive, one-stage, and bariatric procedure with promising short- sity. Patients meeting the study inclusion criteria were divided and long-term treatment effects in morbidly obese patients [1, into groups every 100 consecutive patients. The flow of pa- 2]. LSG is considered as one of the least technically challeng- tients through the study is shown in Fig. 1. ing bariatric procedures. Along with high safety profile and The bariatric team that performed the surgeries com- good outcomes, it is often a sufficient reason for selecting prised of six operators, including two attending surgeons LSG as the first-line treatment in newly established bariatric and four surgery residents in their third year of training centers for gaining experience in weight loss surgery. The (out of 6 years). In each case, one attending surgeon and perioperative complication rate of LSG is estimated to be be- one resident were scrubbed. Introduction of bariatric pro- tween 7 and 15%. The mortality risk of LSG ranges between cedures was structured. Initially, the most experienced op- 0.1 and 2% [3–5]. However, these rates are considered higher erator in laparoscopic surgery visited high-volume bariat- in low-volume bariatric centers and in newly established cen- ric centers to participate in bariatric internships. ters [6, 7]. In 1885, a German psychologist, Hermann Procedures then started in the department along with the Ebbinghaus, was the first to describe the learning curve pro- introduction of a treatment protocol. Guest surgeons from cess. A learning curve is currently defined as the amount of high-volume bariatric centers proctored the learning pro- repetition required for a particular activity to establish an ex- cess in the bariatric team in the first 15 cases during pert level of performance in this particular action [8]. With 2 weeks. Requirements for surgeons in training were ac- regard to surgical procedures, a learning curve is defined as quirement of an appropriate theoretical background in the number of consecutive procedures required to become an bariatric surgery and prior experience in laparoscopic sur- expert in carrying out a specific operation. Most authors agree gery (including intra-corporal sewing skills and structured that the LSG learning curve is 50–100 procedures per surgeon training on simulators). Prior to training as an operator in [9, 10]. However, the amount of data regarding the LSG learn- the operating room, apart from gaining a theoretical back- ing curve for the whole bariatric center—the institutional ground, surgeons needed to assist in a minimum of 15 learning process and its effect on the bariatric treatment LSGs. After meeting the requirements, the residents course—is scarce. As in most surgical procedures, the safety started operating as the first operator with the attending of the operation depends mainly on the experience of the team surgeon assisting with the patient. We defined the number performing the procedure, not on the brilliant surgeon of 100 operations as a basic unit for further analysis. To performing all of the operations. It can, therefore, be stated minimize bias, the patients’ care was standardized in ac- that the experience of the laparoscopic center and the entire cordance with the principles of the multimodal Enhanced team rather than individual surgeons translates into complica- Recovery after Surgery (ERAS) pathway, as described tion rate and outcomes. Therefore, with attendees and resi- previously [11–13]. During preoperative counseling, pa- dents as main operators, this study aimed to analyze the sta- tients aged over 50 years, with maximal preoperative bilization point of the institutional learning process for bariat- BMI over 45 kg/m , with obesity-related comorbidities ric team of six operators, including two attending surgeons (especially type 2 diabetes mellitus and gastroesophageal and four surgery residents, in a newly established bariatric reflux disease, GERD), hiatal hernia, and previous gastric center. We also attempted to analyze the effect of this learning resections were factors indicating qualification for process on patients’ safety and the incidence of perioperative LRYGB, not for LSG in our facility. Routine endoscopy complications. of the upper gastrointestinal tract was done with assess- ment of the incidence of hiatal hernia and esophageal, gastric or duodenal mucosa pathology. If there were Material and Methods symptoms of GERD and endoscopic findings correspond- ing with GERD, patient was not qualified to LSG. We performed a retrospective analysis of prospectively col- With regard to the surgical technique, it was standardized lected data of patients who were submitted to surgical treat- as described previously [14]. ment of morbid obesity in the Second Department of General The analyzed group was divided into subgroups according Surgery, Jagiellonian University Medical College (academic to the order of the procedure: G1, group number 1 (1st–100th teaching hospital and tertiary referral center for general sur- LSG); G2, group number 2 (101st–200th LSG); G3, group gery). Criteria for surgical treatment were in accordance with number 3 (201st–300th LSG); G4, group number 4 (301st– the guidelines of the Metabolic and Bariatric Surgery 400th LSG); and G5, group number 5 (401st–500th LSG). Section of the Polish Surgical Society (i.e., body mass index The primary endpoint was determining the stabilization [BMI] ≥ 35 kg/m with obesity comorbidities or BMI ≥ 40 kg/ point of the LSG learning curve in each group using operative 1674 OBES SURG (2018) 28:1672–1680 Fig. 1 Study flow chart Enrollment Assessed for eligibility (n=775) Excluded (n=275) Other Procedure (n=258) Lack of necessary data (n=5) Declined to participate (n=12) Allocation Allocated to LSG (n=500) Received allocated intervention (n=500) Analysis Allocated to groups (n=500) st th Group 1 – 1 -100 patient (n=100) st th Group 2 – 101 -200 patient (n=100) st th Group 3 – 201 -300 patient (n=100) st th Group 4 – 301 -400 patient (n=100) st Group 5 – 401 -500 patient (n=100) time, intraoperative difficulties, intraoperative adverse events, range, and odds ratios (ORs) with 95% confidence intervals and the number of stapler firings. As secondary endpoints, the (CIs) when appropriate. To assess statistical significance of effects of the learning curve on perioperative complications qualitative data differences in subgroups, Pearson’schi- and re-operation rate were assessed. square test and discriminant analysis were used. Quantitative Directly after each procedure, every surgeon who per- data were analyzed with the Kruskal–Wallis ANOVA and formed the operation as the main operator was obligated post-hoc testing. Univariate logistic regression was used to to note down the intraoperative difficulties, which were calculate ORs with 95% confidence intervals. Results were defined as surgeon-reported obstacles during the opera- considered statistically significant when the p value was less tion. These obstacles were additional measures that were than 0.05. required to finish the procedure or those that significantly prolonged the procedure. These obstacles also included Patients difficulty in achieving a sufficient working space, difficul- ty in proper setting of the stapler, intra-abdominal anato- From April 2009 to October 2017, 775 patients were my obstructing performance of the surgery, or the need for treated for morbid obesity at the Second Department of assistance from a supervisor. The definition of intraoper- Surgery, Jagiellonian University Medical College, includ- ative difficulties is similar to that reported by other au- ing 500 consecutive patients qualified to LSG. A total of thors [15, 16]. Intraoperative adverse events were defined 500 patients were included in the study [330 females, 170 as any iatrogenic, adverse event during the operation, males median age of 40 (33–49) years, median BMI of which was not derived from the standard LSG technique. 44.84 (min. 34.01; max. 76.44) kg/m ]. Patients’ charac- Perioperative complications were defined as adverse teristics are shown in Table 1. events occurring within 30 days after the procedure. These events were classified according to the Clavien– Dindo classification [17]. Rhabdomyolysis was defined Results as an elevated serum creatine phosphokinase concentra- tion of > 1000 U/L with a concomitant increase in myo- Operative Time and First Operators globin concentrations. The median operative time for LSG was 90 (70–120) minutes. Statistical Analysis The Kruskal–Wallis ANOVA showed that there was a signif- icant difference in the median operative time between the All data were analyzed with Statsoft Statistica version 12.0 PL groups (p <0.001) (Table 2). Multiple comparisons of the me- (StatSoft Inc., Tulsa, OK, USA). The results are presented as dian ranges for all groups showed significant differences be- mean ± standard deviation (SD), median and interquartile tween G1–G2 and G3–G5. Based on multiple comparisons of OBES SURG (2018) 28:1672–1680 1675 the median range tests, the LSG learning curve’sstabilization point was the 200th procedure for the whole newly established bariatric center (Fig. 2). During a study period, we observed a significant decrease in the number of operations performed by the attendees in favor of operations performed by the residents as first operators (Pearson’stest, p <0.001). Conversions All procedures were completed laparoscopically. Intraoperative Difficulties Intraoperative difficulties were reported during 23 (4.6%) pro- cedures (Table 2). Using the group number as a factor in univariate logistic regression, the OR for the occurrence of intraoperative difficulties diminished with every 100 per- formed operations (OR, 0.68; 95% CI, 0.49–0.94; p =0.019) and affected, respectively, in G1, 12 patients (12%); G2, two (2%); G3, three (3%); G4, two (2%); and G5, four (4%). According to Pearson’s chi-square test, there was a significant difference in the occurrence of intraoperative difficulties among the groups (p= 0.003). Based on intraoperative diffi- culties, stabilization point was at the 100th procedure in dis- criminant analysis. Intraoperative Adverse Events Intraoperative adverse events were observed during nine (1.8%) operations (Table 2). In G1, they occurred in three patients, G2—none, G3—three patients, G4—two patients, and G5—two patients. Due to lack of variability, this pa- rameter could not be used for estimating stabilization point (p = 0.607). Number of Stapler Firings The median number of stapler firings in the groups was the following: G1, five (4–5); G2, four (4–5); G3, four (4, 5); G4, four (4, 5); and G5, five (4, 5). The Kruskal–Wallis ANOVA showed a significant difference in the number of stapler firings among the groups (p < 0.001). Multiple comparisons of the median and range of stapler firings for all groups showed that G1 and G5 differed from G2 and G4. Perioperative Morbidity Perioperative complications were diagnosed in 29 (7.6%) pa- tients (Table 3). During the study period, we diagnosed 13 patients with perioperative complications of Clavien–Dindo class I, 4 patients with class II, 10 patients with class III, and 2 patients with class IV. Detailed characteristic of periopera- tive morbidity regarding Clavien–Dindo scale is presented in Table 1 Patients’ characteristics Demography Total G1 G2 G3 G4 G5 P value Females, n (%) 330 (66) 69 63 64 70 64 0.759 Males, n (%) 170 (34) 31 37 36 30 36 Median age, years (IQR) 40 (33–49) 39 (32.5–47) 41 (33–48.5) 40 (33–49) 39 (32–47) 42 (34–50) 0.400 Median BMI at qualification, kg/m (IQR) 46.44 (42.91–51.37) 46.56 (44.10–50.85) 44.62 (41.98–49.22) 46.71 (42.73–51.37) 45.81 (42.76–52.73) 47.11 (43.94–54.65) 0.345 Median BMI on admission, kg/m (IQR) 44.84 (41.33–49.22) 45.48 (42.45–49.80) 43.10 (41.04–47.66) 45.49 (41.14–49.12) 44.01 (40.88–49.51) 45.0 (41.87–50.20) 0.143 Cardiovascular diseases, n(%) 61(12.2) 151612126 0.226 Arterial hypertension, n (%) 320 (64) 59 71 62 55 73 0.034 Respiratory diseases, n(%) 95(19) 2214221522 0.364 Diabetes mellitus, n (%) 119 (24) 22 21 28 24 24 0.808 Dyslipidemia, n (%) 180 (36) 60 50 38 16 16 < 0.001 Fatty liver disease, n (%) 219 (44) 65 60 56 15 23 < 0.001 G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st–300th LSG), G4 group number 4 (301st–400th LSG), G5 group number 5 (401st–500th LSG) 1676 OBES SURG (2018) 28:1672–1680 Table 2 Operative time, operating surgeons, intraoperative difficulties, and intraoperative adverse events Total G1 G2 G3 G4 G5 Median operative time (IQR) 90 (70–120) 130 (100–160) 100 (80–120) 80 (70–100) 90 (70–110) 80 (65–95) First operator Attending 296 (59%) 91/100 87/100 53/100 35/100 30/100 Resident 204 (41%) 9/100 13/100 47/100 65/100 70/100 Intraoperative difficulties Total 23/500 (4.6%) 12/100 (12%) 2/100 (2%) 3/100 (3%) 2/100 (2%) 4/100 (4%) Difficulty to achieve sufficient working space 6 (1.2%) 4 0 1 0 1 Difficulty of proper setting the stapler 7 (1.4%) 5 0 1 0 1 Intra-abdominal adhesions obstructing 4 (0.8%) 2 0 0 1 1 performance of the surgery Fatty liver disease obstructing the procedure 1 (0.2%) 0 1 0 0 0 Irreducible umbilical hernia 1 (0.2%) 0 1 0 0 0 Large umbilical hernia 1 (0.2%) 0 0 0 1 0 Required of the help of the mentor 1 (0.2%) 1 0 0 0 0 Difficulty to achieve sufficient hemostasis 2 (0.4%) 0 0 1 0 1 Intraoperative adverse events Total 9/500 (1.8%) 3/100 (3%) 0/100 (0%) 2/100 (3%) 2/100 (2%) 2/100 (2%) Intraoperatively diagnosed leakage, supplied 3 (0.6%) 3 0 0 0 0 with the additional suturing Intraoperatively diagnosed leakage, supplied 1 (0.2%) 0 0 1 0 0 with setting of stent and drainage Excessive intraoperative blood loss 1 (0.2%) 0 0 1 0 0 Bleeding from liver/spleen 2 (0.4%) 0 0 0 1 1 Bleeding from stapler line 1 (0.2%) 0 0 0 1 0 Gastric content leak from resected part 1 (0.2%) 0 0 0 0 1 G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st–300th LSG), G4 group number 4 (301st– 400th LSG), G5 group number 5 (401st–500th LSG) Table 3. When we used group number as a factor in the uni- Mortality variate logistic regression model, the OR of perioperative morbidity significantly decreased with every 100 performed None of the patients died during the 30-day perioperative procedures (i.e., with every consecutive group) (OR, 0.66; period. 95% CI, 0.50–0.89; p = 0.006). The morbidity rates in the groups were G1, 13%; G2, 4%; G3, 5%; G4, 5%; and G5, 2%, respectively. Pearson’s chi-square test showed that there Discussion was a significant difference in the occurrence of morbidity among the groups (p = 0.011). Based on perioperative morbid- Bariatric surgery in Poland is still a developing discipline. We ity, the learning curve stabilized at the 100th procedure in describe a 9-year experience of LSG in the Second discriminant analysis. Department of General Surgery, JUMC [18]. We aimed to assess our institutional learning process for LSG and the effect of this process on safety and perioperative complications. Reoperations Operative time is frequently regarded as a descriptive pa- rameter for evaluating the learning curve. In our study, the Reoperations were necessary in eight (1.8%) patients. operative time was an average of 90 min. Zacharoulis et al. Pearson’s chi-square test did not reveal significant differences assumed that one of the main determinants of an LSG learning of reoperations in groups (p = 0.508). It was also nonsignifi- curve’s stabilization point is the operative time and estimated cant in univariate logistic regression model (p = 0.344). that 68 consecutive procedures are essential to finish training Basing on reoperations, this parameter could not be used for [19]. Daskalakis et al. observed a significant difference in setting stabilization point. surgical duration between the first and second halves of their OBES SURG (2018) 28:1672–1680 1677 Fig. 2 Graph illustrating operative time changes in institutional learning curve Median IQR Case number Min.-Max. 230 consecutive LSGs [20]. A learning curve’sstabilization We reported intraoperative difficulties during 23 (4.6%) point is usually derived from the literature and tested for indi- procedures. In our study, the OR of the occurrence of intraop- vidual operators to achieve expertise. In our study, we to esti- erative difficulties decreased with every 100 performed LSGs. mate the stabilization point for the whole department by test- Intraoperative difficulties were mainly related to a narrow op- ing the median and range within consecutively operated erative space (six cases) and to proper stapler setting difficulty groups of 100 patients. In our study, the institutional learning (seven cases). Unfortunately, we could not find any previously process’ stabilization point was the 200th procedure. published studies regarding intraoperative difficulties, which Table 3 Perioperative (≤ 30 days) morbidity regarding Clavien–Dindo scale and reoperations C-D Complications No. (%) G1 G2 G3 G4 G5 Total 29/500 (7.6%) 13/100 (13%) 4/100 (4%) 5/100 (5%) 5/100 (5%) 2/100 (2%) 4b Cardiorespiratory failure (ICU stay) 1 (0.2%) 0 1 0 0 0 4a Pulmonary embolism (thrombolysis) 1 (0.2%) 0 0 0 0 1 3b GI leak (relaparoscopy) 4 (0.8%) 3 1 0 0 0 Operation site bleeding (relaparoscopy) 4 (0.8%) 0 1 2 1 0 Wound infection (open drainage) 1 (0.2%) 0 0 0 1 0 3a GI stricture (tube placement) 1 (0.2%) 0 0 1 0 0 2 Varicella infection 1 (0.2%) 0 0 1 0 0 TIA 1 (0.2%) 0 0 0 1 0 Superior mesenteric vein thrombosis (thrombolysis) 1 (0.2%) 0 0 0 1 0 Renal colic 1 (0.2%) 0 0 0 0 1 1 Delayed gastric emptying due to temporary stricture * 5 (1%) 2 1 1 1 0 Dehydration* 1 (0.2%) 1 0 0 0 0 Prolonged drainage 1 (0.2%) 1 0 0 0 0 Rhabdomyolysis 6 (1.2%) 6 0 0 0 0 Reoperations 8 (1.8%) 3 2 2 1 0 *1 patient was diagnosed with both complications C-D Clavien-Dindo classification grade, G1 group number 1 (1st–100th LSG), G2 group number 2 (101st–200th LSG), G3 group number 3 (201st– 300th LSG), G4 group number 4 (301st–400th LSG), G5 group number 5 (401st–500th LSG), TIA transient ischemic attack 1-100 101-200 Learning curve's stabilization point 201-300 301-400 401-500 Operative time [min] 1678 OBES SURG (2018) 28:1672–1680 are important parameters for assessment of the learning curve reported by Daskalakis et al. (7.4%) [20]. In our study, due to [21, 22]. Intraoperative difficulties were self-reported by sur- low variability of reoperation rate, we could not draw conclu- geons and defined as obstacles requiring additional measures sions from this parameter. during the surgery. We estimated that the intraoperative diffi- None of the patients died during the 30-day perioperative culty rate stabilized after 100 procedures. period. This finding is similar to that by Casella et al. [28]and In our study, intraoperative adverse events, defined as Trastulli et al. [29]. However, Zacharoulis et al. showed that harmful events related to surgery, occurred during nine the mortality rate was 0.98% [19]. Taking into consideration, (1.8%) procedures, and due to lack of variability, they could the mortality of LSG as reported in a systematic review by Shi not be used for estimation of the institutional learning process’ et al. (0–3.3%), the lack of morality in our study on the course stabilization point. There were four cases (0.8%) of intraoper- of the learning curve is satisfactory [26]. atively diagnosed gastrointestinal leak. Similarly, Rubin et al. Usually, a learning curve is evaluated using selected pa- reported that intraoperative adverse events occurred during rameters for particular surgeon. We assessed the institutional 3.33% of 120 LSGs. They found one case of intraoperatively learning process using intraoperative difficulties, periopera- diagnosed gastric leak, two cases of short gastric vessel bleed- tive morbidity, and operative time, while obtaining accurate ing, and one case of probe immobilization in the staple line results. Operative time of the first 200 procedures was signif- [23]. Braghetto et al. reported a rate of 14% for intraoperative icantly longer than that of the next 300 surgeries. Based on adverse events after introduction of LSG in a series of 50 this, the institutional learning process stabilization point was LSGs (four cases of intraoperatively diagnosed gastric leak, the 200th procedure for the whole newly established bariatric one case of short gastric vessel bleeding, one case of liver center. Intraoperative difficulties in the first 100 patients capsule perforation with a retractor, and one case of probe (12%) were significantly more prevalent than that in the next immobilization in the staple line) [24]. 400 procedures (2.75%). Based on intraoperative difficulties, After 100 procedures, the amount of stapler firings de- the institutional learning process’ stabilization point was at the creased by one firing but then in the last 100 patients, it 100th procedure. Finally, the postoperative morbidity rate in increased by one. Kaska et al. found that, to avoid adverse the first 100 patients of 13% was much higher than that in the events and complications, adequately choosing the car- next 400 cases (4%). Based on morbidity, the institutional tridge and over-sewing the staple line are essential but learning process was stabilized at the 100th procedure. the number of stapler firings was not analyzed [16]. We Summarizing all endpoints that enabled to assess stabilization believe that this increase comes just from taking adequate point, we proposed that the institutional learning process sta- measures to operative conditions. Our beliefs are support- bilized between the 100th and 200th case. ed by lack of increase in intraoperative adverse events and This study has several limitations. Our institution is a re- complications. ferral center for general surgery and a teaching hospital. The conversion rate for LSG ranges from 1.05 to 1.85% Therefore, our results probably cannot be extrapolated to all [25]. Despite these rates, in our study, all procedures were hospitals. Additionally, this was a retrospective study. A pro- completed laparoscopically. spective study should be designed to confirm these findings. The LSG learning curve’s stabilization point can also be Finally, the bariatric team consisted of surgeons of varying estimated using perioperative morbidity. In our study, periop- levels of experience, which may have affected the results. A erative morbidity was 7.6%. An almost identical rate of the larger amount of operated patients would enable the use of all LSG learning curve was described by Zacharoulis et al. of the parameters we attempted to test for describing the learn- (7.8%) [19]. Our result is satisfactory compared with the re- ing curve. Further research should be performed using intra- ported a major morbidity rate of 12.1% (0–29%) in a system- operative difficulties as a self-reported assessment for evalua- atic review by Shi et al. [26]. Regardless of implementation of tion of the learning curve process. the ERAS protocol along with introduction of LSG, we ob- served a relatively high perioperative morbidity of 13% in the first 100 patients. Morbidity then decreased to 4% in the rest of the consecutive patients. Discriminant analysis enabled us Conclusion to set the LSG learning curve’s stabilization point based on morbidity at the 100th LSG. Zachariah et al. showed that the The institutional learning process’ stabilization point for LSG perioperative complication rate in the first 50 patients was in a newly established bariatric center is between the 100th significantly higher than that in the last 178 patients (8 vs. and 200th operation. LSG, which is carried out with a learning 1.6%, p =0.02) [27]. Casella et al. showed that perioperative curve, does not affect the safety of the procedure in terms of morbidity was significantly decreased after the 88th LSG [28]. intraoperative adverse events. Initially, with introduction of In our study, eight (1.8%) patients required reoperations. LSG, the morbidity rate is high, which should concern sur- This reoperation rate was significantly smaller than the rate geons who are willing to start performing bariatric surgery. OBES SURG (2018) 28:1672–1680 1679 Funding This research was supported by the Faculty of Medicine, 9. Romero RJ, Kosanovic R, Rabaza JR, et al. Robotic sleeve gastrec- Jagiellonian University Medical College, Leading National Research tomy: experience of 134 cases and comparison with a systematic Centre (KNOW) 2012–2017. review of the laparoscopic approach. Obes Surg. 2013;23(11): 1743–52. https://doi.org/10.1007/s11695-013-1004-1. 10. Buchwald H, Williams SE. Bariatric surgery worldwide 2003. Obes Compliance with Ethical Standards Surg. 2004;14(9):1157–64. https://doi.org/10.1381/ Conflict of Interest The authors declare that they have no conflict of 11. Major P, Wysocki M, Pędziwiatr M, et al. Risk factors for compli- interest. cations of laparoscopic sleeve gastrectomy and laparoscopic Roux- en-Y gastric bypass. Int J Surg. 2017;37:71–8. https://doi.org/10. Ethical Responsibilities of Authors Article follows the COPE 1016/j.ijsu.2016.12.012. guidelines. 12. Major P, Wysocki M, Torbicz G, et al. Risk factors for prolonged length of hospital stay and readmissions after laparoscopic sleeve Ethical Approval All procedures performed in studies involving human gastrectomy and laparoscopic Roux-en-Y gastric bypass. Obes participants were in accordance with the ethical standards of the institu- Surg. 2017; https://doi.org/10.1007/s11695-017-2844-x. tional and/or national research committee and with the 1964 Helsinki 13. Matłok M, Pędziwiatr M, Major P, et al. One hundred seventy-nine declaration and its later amendments or comparable ethical standards. consecutive bariatric operations after introduction of protocol in- The study was approved by the Bioethics Committee of Jagiellonian spired by the principles of enhanced recovery after surgery University (KBET/62/B/2011). (ERAS®) in bariatric surgery. Med Sci Monit. 2015;21:791–7. https://doi.org/10.12659/MSM.893297. 14. Major P, Wysocki M, Pędziwiatr M, et al. More stapler firings Informed Consent Statement Informed consent was obtained from all increase the risk of perioperative morbidity after laparoscopic individual participants included in the study. sleeve gastrectomy. Videosurgery and Other Miniinvasive Techniques/Wideochirurgia i inne techniki małoinwazyjne. 2017; Open Access This article is distributed under the terms of the Creative https://doi.org/10.5114/wiitm.2017.70197. Commons Attribution 4.0 International License (http:// 15. Korenkov M, Gundlach M, Heimbucher J, et al. The impact of creativecommons.org/licenses/by/4.0/), which permits unrestricted use, preoperative classification of expected intraoperative difficulties. distribution, and reproduction in any medium, provided you give appro- A multicenter study. Chirurg. 2014;85(11):999–1004. https://doi. priate credit to the original author(s) and the source, provide a link to the org/10.1007/s00104-013-2702-x. Creative Commons license, and indicate if changes were made. 16. Kaska L, Proczko M, Stefaniak T, et al. Redesigning the process of laparoscopic sleeve gastrectomy based on risk analysis resulted in 100 consecutive procedures without complications. Wideochir Inne Tech Maloinwazyjne. 2013;8(4):289–300. https://doi.org/10.5114/ References wiitm.2011.34797. 17. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 1. Sánchez-Santos R, Masdevall C, Baltasar A, et al. Short- and mid- 6336 patients and results of a survey. Ann Surg. 2004;240(2): term outcomes of sleeve gastrectomy for morbid obesity: the expe- 205–13. rience of the Spanish National Registry. Obes Surg. 2009;19(9): 18. Wylezol MS, Zwirska-Korczala K, Pasnik K. Bariatric surgery in 1203–10. https://doi.org/10.1007/s11695-009-9892-9. Poland from 1993 to 2003. J Physiol Pharmacol. 2005;56(Suppl 6): 2. Arias E, Martínez PR, Ka Ming Li V, et al. Mid-term follow-up after 109–15. sleeve gastrectomy as a final approach for morbid obesity. Obes 19. Zacharoulis D, Sioka E, Papamargaritis D, et al. Influence of the Surg. 2009;19(5):544–8. https://doi.org/10.1007/s11695-009- learning curve on safety and efficiency of laparoscopic sleeve gas- 9818-6. trectomy. Obes Surg. 2012;22(3):411–5. https://doi.org/10.1007/ 3. Sanni A, Perez S, Medbery R, et al. Postoperative complications in s11695-011-0436-8. bariatric surgery using age and BMI stratification: a study using 20. Daskalakis M, Berdan Y, Theodoridou S, et al. Impact of surgeon ACS-NSQIP data. Surg Endosc. 2014;28(12):3302–9. https://doi. experience and buttress material on postoperative complications org/10.1007/s00464-014-3606-7. after laparoscopic sleeve gastrectomy. Surg Endosc. 2011;25(1): 4. Dumon KR, Murayama KM. Bariatric surgery outcomes. Surg Clin 88–97. https://doi.org/10.1007/s00464-010-1136-5. North Am. 2011;91(6):1313–1338, x. https://doi.org/10.1016/j.suc. 21. Mammen JM, Fischer DR, Anderson A, et al. Learning styles vary 2011.08.014. among general surgery residents: analysis of 12 years of data. J Surg 5. Picot J, Jones J, Colquitt JL, et al. The clinical effectiveness and Educ. 2007;64(6):386–9. https://doi.org/10.1016/j.jsurg.2007.08. cost-effectiveness of bariatric (weight loss) surgery for obesity: a systematic review and economic evaluation. Health Technol 22. Quillin RC, Pritts TA, Hanseman DJ, et al. How residents learn Assess. 2009;13(41):1–190, 215–357, iii–iv. https://doi.org/10. predicts success in surgical residency. J Surg Educ. 2013;70(6): 3310/hta13410. 725–30. https://doi.org/10.1016/j.jsurg.2013.09.016. 6. Celio AC, Kasten KR, Brinkley J, et al. Effect of surgeon volume 23. Rubin M, Yehoshua RT, Stein M, et al. Laparoscopic sleeve gas- on sleeve gastrectomy outcomes. Obes Surg. 2016;26(11):2700–4. trectomy with minimal morbidity. Early results in 120 morbidly https://doi.org/10.1007/s11695-016-2190-4. obese patients. Obes Surg. 2008;18(12):1567–70. https://doi.org/ 7. Birkmeyer NJ, Dimick JB, Share D, et al. Hospital complication 10.1007/s11695-008-9652-2. rates with bariatric surgery in Michigan. JAMA. 2010;304(4):435– 24. Braghetto I, Korn O, Valladares H, et al. Laparoscopic sleeve gas- 42. https://doi.org/10.1001/jama.2010.1034. trectomy: surgical technique, indications and clinical results. Obes 8. Sauerland S, Angrisani L, Belachew M, et al. Obesity surgery: Surg. 2007;17(11):1442–50. https://doi.org/10.1007/s11695-008- 9421-2. evidence-based guidelines of the European Association for 25. Rosenthal RJ, Diaz AA, Arvidsson D, et al. International Sleeve Endoscopic Surgery (EAES). Surg Endosc. 2005;19(2):200–21. https://doi.org/10.1007/s00464-004-9194-1. Gastrectomy Expert Panel Consensus Statement: best practice 1680 OBES SURG (2018) 28:1672–1680 guidelines based on experience of >12,000 cases. Surg Obes Relat 28. Casella G, Soricelli E, Giannotti D, et al. Learning curve for lapa- roscopic sleeve gastrectomy: role of training in a high-volume bar- Dis. 2012;8(1):8–19. https://doi.org/10.1016/j.soard.2011.10.019. 26. Shi X, Karmali S, Sharma AM, et al. A review of laparoscopic iatric center. Surg Endosc. 2015;30(9):3741–8. https://doi.org/10. sleeve gastrectomy for morbid obesity. Obes Surg. 2010;20(8): 1007/s00464-015-4670-3. 1171–7. https://doi.org/10.1007/s11695-010-0145-8. 29. Trastulli S, Desiderio J, Guarino S, et al. Laparoscopic sleeve gas- 27. Zachariah SK, Chang PC, Ooi AS, et al. Laparoscopic sleeve gas- trectomy compared with other bariatric surgical procedures: a sys- trectomy for morbid obesity: 5 years experience from an Asian tematic review of randomized trials. Surg Obes Relat Dis. center of excellence. Obes Surg. 2013;23(7):939–46. https://doi. 2013;9(5):816–29. https://doi.org/10.1016/j.soard.2013.05.007. org/10.1007/s11695-013-0887-1.

Journal

Obesity SurgerySpringer Journals

Published: Dec 23, 2017

References

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off