ORIGINAL CONTRIBUTIONS
Early Post-operative Complications: Incidence, Management, and Impact on Length of Hospital Stay. A Retrospective
Comparison Between Laparoscopic Gastric Bypass and Sleeve Gastrectomy
Rudolf A. Weiner&Islam A. El-Sayes&Sophia Theodoridou&
Sylvia R. Weiner&Oliver Scheffel
Published online: 12 July 2013
#Springer Science+Business Media New York 2013
Abstract
Background Laparoscopic Roux-en-Y gastric bypass (LRY GB) and laparoscopic sleeve gastrectomy (LSG) are the most common obesity surgeries. Their early complications may prolong hospital stay (HS).
Methods Data for patients who underwent LRYGB and LSG in our clinic from 2009 through August 2012 were collected.
Early post-operative complications prolonging HS (>5 days) were retrospectively analyzed, highlighting their relative incidence, management, and impact on length of HS.
ResultsSixty-six patients (4.9 %) after 1,345 LRYGB opera- tions vs. 49 patients (7.14 %) after 686 LSG operations devel- oped early complications. This difference is statistically signif- icant (p=0.039). Male gender percentage was significantly higher in complicated LSG group vs. complicated LRYGB group [23 patients (46.9 %) vs. 16 patients (24.2 %)]
(p= 0.042). Mean BMI was significantly higher in the
complicated LSG group (54.2±8.3) vs. complicated LRYGB group (46.8±5.7; p=0.004). Median length of HS was not longer after complicated LSG compared with complicated LRYGB (11 vs. 10 days; p=0.287). Leakage and bleeding were the most common complications after either procedure.
Leakage rate was not higher after LSG (12 patients, 1.7 %) compared with LRYGB (22 patients, 1.6 %;p=0.304). Bleed- ing rate was significantly higher after LSG (19 patients, 2.7 %) than after LRYGB (10 patients, 0.7 %;p=0.004). Prolonged elevation of inflammatory markers was the most common presentation for complications after LSG (18 patients, 36.7 %) and LRYGB (31 patients, 46.9 %).
Conclusions LSG was associated with more early complica- tions. This may be attributed to higher BMI and predomi- nance of males in LSG group.
Keywords Laparoscopic Roux-en-Y gastric bypass . Laparoscopic sleeve gastrectomy . Length of hospital stay . Early post-operative complications . Leakage . Bleeding
Introduction
Surgery proved to be the most effective intervention for mor- bid obesity [1, 2]. Laparoscopic Roux-en-Y gastric bypass (LRYGB) and laparoscopic sleeve gastrectomy (LSG) are currently the most commonly performed techniques [3–6].
These technically demanding procedures may have serious complications which could be either early or late [7]. This study focused on early post-operative complications, which included any complication following either LRYGB or LSG and prolonged the hospital stay (HS) past the fifth post- operative day. The aim of the study was to compare both procedures with regard to the incidence of early post- operative complications and to detect whether the diagnosis The results of this study have been submitted for IFSO 2013 Congress
in Turkey as an oral presentation. However, the result of this submission (acceptance for presentation or rejection) is not yet announced.
R. A. Weiner
:
I. A. El-Sayes (*):
S. Theodoridou:
S. R. Weiner
:
O. ScheffelSurgery Department, Sachsenhausen Hospital, Krankenhaus Sachsenhausen Schulstr. 31, 60594 Frankfurt am Main, Germany e-mail: islam_sayes@yahoo.com R. A. Weiner
e-mail: rweiner@khs-ffm.de S. Theodoridou
e-mail: sophiatheodoridou@t-online.de S. R. Weiner
e-mail: sylvia.weiner@gmx.de O. Scheffel
e-mail: oliverscheffel@mac.com
and management of these complications prolonged HS in our clinic. Our perioperative protocol for prevention and manage- ment of these complications was also highlighted.
Patients and Methods
Patients who underwent either LRYGB or LSG as a primary surgical intervention for morbid obesity from January 2009 through August 2012 in our clinic were included. All pa- tients conformed to the NIH consensus criteria [8] and Ger- man guidelines for bariatric surgery [9]. LRYGB is our standard procedure. However, morbidly obese patients with evidence of inflammatory bowel disease, those who persis- tently consume ulcerogenic medications, those with exten- sive previous abdominal surgeries, and those in need for future endoscopic assessment of the distal stomach or duo- denum seemed to be unsuitable candidates for LRYGB in favor of LSG. On the other hand, those with evidence of gastro-esophageal reflux disease were more liable to benefit from LRYGB. Super obese patients (BMI >50) were more likely to undergo LSG, but the final decision was met intra- operatively. In other words, super obese patients who could benefit almost equally from both procedures and had no contraindications for either LRYGB or LSG were informed for both procedures. Intraoperative diagnostic laparoscopy determined the final decision, based upon liver volume, intra-abdominal fat content, and extent of adhesions. It is also important to mention that the decision was tailored for every patient according to his own presentation and preference.
Early post-operative complications interfering with pa- tient discharge on the fifth post-operative day were retro- spectively analyzed. Their incidence, presentation, manage- ment, and impact on length of HS were reported. Patients with a prolonged HS due to early post-operative complica- tions represented our study cohort. It’s crucial however to outline our recommended perioperative protocol to prevent these complications. Enoxaparin sodium (CLEXANE®, Sanofi-Aventis®, 40 mg once-daily subcutaneous injection) together with elastic stocking usage started the day before the operation until the day of discharge. This went in accordance with German guidelines for obesity surgery [10]. Two grams of a third-generation cephalosporin were infused intraopera- tively. Muscle splitting trocars were used. For LRYGB, blue staple loads of linear endo GIA devices were used for gastrojejunal anastomosis, reinforced by interrupted hand- sewn (0) size Coated VICRYL® Plus Antibacterial (polyglactin 910) sutures (ETHICON, Johnson & Johnson Intl). Enteroenteral anastomosis was performed using white staple loads. Mesenteric defects were closed using nonabsorbable (0) size MERSILK® (ETHICON, LLC) su- ture material. For LSG, a 42-Fr orogastric bougie was used.
Linear green (and blue, if needed) staple loads were used.
Peri-Strips Dry® with Veritas® Collagen Matrix (Synovis®) reinforced the staple line as needed. Hemostatic sutures or ultrasonic coagulation assured hemostasis. Leakage was ex- cluded by methylene blue testing. Drain insertion in the vicinity of the staple line was standardized. Patients were admitted to the post-operative overnight Intensive Care Unit for adequate hemodynamic and hourly urinary catheter out- put monitoring. Patients were encouraged to start oral sips on the first post-operative day, progressing gradually to soft food. Tachycardia (>100 beats/min), fever (≥38.5 °C), ab- dominal pain, tachypnea (>20 breaths/min), or decreased urinary output despite good hydration were handled with a high index of suspicion. Laboratory findings of C-reactive protein (CRP) >150 mg/l and leukocytosis >11,000 cells/μl were suggestive of complications. CRP level and complete blood count were routinely checked on the first and fifth post-operative day. Diagnostic radiological and endoscopic tools were readily available, but not as a routine post- operative practice. Patients undergoing revisional surgery were managed similarly to those primarily operated. All complications were adequately controlled before discharge.
Leakage was defined as trickling of gastric and/or intestinal contents outside the visceral lumen. It was diagnosed radio- logically by detection of contrast outside the lumen in contrast-enhanced computed tomography (CT) scans or gastrografin swallows. This may have been aided by detection of abscess cavity, air foci in the vicinity of operative field, or stranded mesentery on CT scans. Leakage after LRYGB was an indication for reoperation (over sewing and suction drain- age). Endoscopic stenting played a minor role here, in contrast to leakage after LSG, in which a combination of surgical intervention and endoscopic stenting (Niti-S Fully Covered Esophageal Stent®, Taewoong Medical Co., Ltd.) assured better results.
Bleeding was suspected by a post-operative hemoglobin level drop of at least 2 g/dl, aided by CT evidence. It was controlled using surgical intervention in hemodynamically unstable patients, where hematoma evacuation, over sewing, and drainage were done. Otherwise, we resorted to conser- vative methods (fluid resuscitation ±blood transfusion).
Statistical Analysis
Data input was performed using IBM SPSS software version 20.0. Qualitative data were described using numbers (percentage). Comparisons between the groups regarding the categorical variables were evaluated using the Chi- square test. The distributions of quantitative variables were tested for normality. Quantitative data were described using the mean, standard deviation, median, minimum, and maxi- mum. For normally distributed data, the two independent populations were compared using the independent t test.
For abnormally distributed data, the Mann–Whitney test was used to analyze the two independent populations. The sig- nificance test results were reported as two-tailed probabili- ties. Significance was determined at the 5 % level.
Results
Among 2,031 patients who underwent either LRYGB or LSG in our clinic from January 2009 through August 2012, 1,345 LRYGBs (66.2 %) and 686 LSGs (33.8 %) were performed in our clinic (Table1). This difference in the overall tendency to perform LRYGB was statistically significant. Similarly, we performed significantly more LRYGBs than LSGs in 2012. In our clinic, these operations were performed by ten surgeons, with an experience of up to 12 years in both procedures.
Among 1,345 patients who underwent LRYGB, mean BMI was 46.3 kg/m2(±6.4). Males represented 42.3 % (569 patients). Age ranged from 22 to 62 years with a median of 43 years. Among 686 patients who underwent LSG, mean BMI was 57.8 kg/m2 (±2.1). A total of 325 patients were males (47.3 %). Age ranged from 23 to 64 years with a median of 39 years. Mean BMI was significantly higher among pa- tients who underwent LSG.
Regarding complications, 66 patients (4.9 %) developed early complications after LRYGB and 49 patients (7.14 %) developed early complications after LSG resulting in a prolonged HS. This difference was statistically significant (p=0.039). Patients with prolonged HS represented our co- hort. Prolonged HS was subdivided into three time intervals:
7–10, 10–30, and >30 days. Data related to HS are also shown in Tables2and3. Median HS after complicated LSG was not longer than that for LRYGB [11 days (7–92) vs. 10 days (7–
76), respectively] (Mann–Whitney test,p=0.287). Regarding the demographic data of our study cohort, male gender per- centage was significantly higher in the LSG group compared with the LRYGB group [46.9 % (23 patients) vs. 24.2 % (16 patients)] (p=0.042). Similarly, mean BMI was significantly
higher in the LSG group (54.2±8.3) compared with the LRYGB group (46.8±5.7;p=0.004).
To eliminate bias in procedure selection based on gender or BMI differences between both study groups, we divided our study cohort in the LRYGB and LSG groups into four subgroups: males with BMI<50 kg/m2, males with BMI>50 kg/m2, females with BMI<50 kg/m2, and females with BMI>50 kg/m2. Demographic data of the study cohort are shown in Table4.
Regarding early post-operative complications, gastroenteric leakage, bleeding, and elevation of inflammatory markers were the most commonly encountered early complications and causes of prolonged HS. Other less common complica- tions and causes of prolonged HS were also encountered (Tables5and6). Leakage in our cohort presented by elevated inflammatory markers in 19 patients (86.3 %) in the LRYGB group vs. 12 patients (100 %) in the LSG group. Other less common presentations included upper abdominal pain and tachycardia. These data are presented in Table7. Leakage after LRYGB was reported at the site of gastrojejunostomy or proximal gastric pouch (nine patients, 40.9 %), at the enteroenteral anastomosis (nine patients, 40.9 %), at the site of accidental small intestinal injury (two patients, 9.1 %), at the site of the fundal wrap that was released during LRYGB (one patient, 4.5 %), and at the site of adhesiolysis due to a previous open appendectomy (one patient, 4.5 %). Once diagnosed, urgent intervention (operative and/or endoscopic) was manda- tory for adequate control of leakage. Table 8 shows how leakage was managed in our cohort. Open intervention was significantlymore needed for control of leakage after LRYGB than after LSG [ten patients (45 %) vs. two patients (17 %)].
While stenting was significantlymore needed for control of leakage after LSG than after LRYGB [nine patients (75 %) vs.
one patient (4 %)].
Early post-operative bleeding rate wassignificantlyhigher after LSG than after LRYGB. Sites of bleeding are shown in Table 5. Conservative measures (fluid resuscitation±blood transfusion) were successful in three patients (30 %) in the
Table 1 Yearly and total number (percentage) of performed operations (LRYGB and LSG) in our clinic from 2009 through August 2012
LRYGB LSG pvalue
2009 353 (17.3 %) 148 (7.2 %) NS
2010 372 (18.3 %) 207 (10.1 %) NS
2011 333 (16.3 %) 213 (10.4 %) NS
31 August 2012 287 (14.1 %) 118 (5.8 %) p= 0.033*
Total 1,345/2,031 (66.2 %) 686/2,031 (33.7 %) p=0.042*
Percentages are calculated from the total number of LRYGB and LSG performed in our clinic from 2009 through August 2012 (2,031 operations).
Summation of individual percentages may not coincide with total percentage because of rounding NSnon-significant
*p≤0.05, statistically significant using Chi-square test
LRYGB group vs. six patients (31.6 %) in the LSG group.
Laparoscopic intervention (hematoma evacuation, over sew- ing, lavage, and drainage) was necessary in 7 patients (70 %) in the LRYGB group vs. 13 patients (68.4 %) in the LSG group who were hemodynamically unstable. Post-operative pneumonia was managed medically. Adult Respiratory Dis- tress Syndrome necessitated mechanical ventilation. Trocar site herniation was managed by laparoscopic reduction of incarcerated small bowel and excision of gangrenous omen- tum. Small intestinal obstruction was detected in two patients and necessitated open re-do of stenotic enteroenteral anasto- mosis in one patient and laparoscopic herniotomy for irreduc- ible umbilical hernia in the second. Bile leakage at the com- mon bile duct stump from a concomitant laparoscopic chole- cystectomy necessitated laparoscopic revision. Deep vein thrombosis was diagnosed in one patient and was managed medically. Post-operative dysphagia and fluid intolerance were managed by oral corticosteroid preparations for subse- quent follow-up. Repeated dressing was required for infected wounds. One patient with a Douglas pouch abscess was managed by open drainage. A gluteal abscess was drained in one patient. A left shoulder dislocation was reduced under anesthesia in one patient. Other complications were managed medically. Regarding the mortality rate, one patient died after LRYGB (0.07 %) and one patient died after LSG (0.14 %) due to uncontrolled sepsis following leakage.
Discussion
Bariatric surgery is gaining increasing popularity worldwide, with more than a fivefold increase in the number of bariatric surgeries in the USA between 2001 and 2008 [11]. LRYGB is currently the gold standard bariatric procedure. LSG is a relatively recent procedure, with lack of a standardized tech- nique [12]. It attracted however particular interest in the last few years, being primarily indicated in super obese patients [13,14]. We share the same concept. LRYGB is our standard technique. Among 2,031patients in this study, we performed significantly more LRYGB than LSG (1,345 (66.2 %) vs.
686 patients (33.7 %), respectively). LSG was put in consid- eration in patients who were not eligible for LRYGB and in super obese patients. We experienced however patients with BMI of up to 60 in whom LRYGB was safely performed.
That is to say, we balance the risk–benefit ratio of both procedures in every patient. In our opinion, a morbidly obese patient with inflammatory bowel disease benefits more from LSG, while a super obese patient with significant gastro- esophageal reflux symptoms benefits more from LRYGB.
We routinely discharge our patients on the fifth post- operative day as long as the post-operative course is smooth.
Series have reported a shorter post-operative HS of 2 to 3 days in absence of complications [15,16]. Some authors even discharge their patients normally on the first post- Table 2 Number of patients with a prolonged HS due to early complications after either LRYGB or LSG and number of patients operated upon in our clinic from 2009 through August 2012
LRYGB LSG pvaluea
2009 21/353 (5.9 %) 9/148 (6.1 %)
2010 14/372 (3.8 %) 12/207 (5.8 %)
2011 17/333 (5.1 %) 20/213 (9.4 %)
31 August 2012 14/287 (4.9 %) 8/118 (6.9 %)
Total 66/1,345 (4.9 %) 49/686 (7.14 %) 0.039*
Percentages of patients with complications are given in parentheses
*p≤0.05, statistically significant
aChi-square test
Table 3 Differentiation of prolonged HS into three time intervals
Number of patients in each time interval LRYGB LSG pvaluea
Patients with HS of 7–10 daysN(%) 28 (42.4) 17 (34.7)
Patients with HS of 10–30 daysN(%) 33 (50) 26 (53.1)
Patients with HS longer than 30 daysN(%) 5 (7.6) 6 (12.2)
TotalN(%) 66 (100) 49 (100) NS
Number (N) and percentage of patients for each time interval are shown NSnon-significant
aChi-square test
operative day [17]. German health insurance programs sup- port however a longer HS of up to 6 to 8 days after bariatric surgeries [10]. Based on our experience, we suggest that post-operative HS of up to 5 days insures adequate periop- erative patient assessment, unless post-operative complica- tions were encountered.
Post-operative complications remain to be a matter of con- cern in those risky patients in whom laboratory values and imaging studies may also be equivocal. Integrated teamwork is therefore necessary in bariatric practice. Several conditions have been implicated as possible predisposing factors for com- plications following bariatric surgery. High BMI proved in the literature to be a major determinant of post-operative compli- cations [18]. Male gender poses another problem, mostly due to more intra-abdominal fat and associated co-morbidities, with
subsequent increase of post-operative morbidity and mortality [19–21]. Studies even showed that males undergoing LSG had significantly higher BMI and higher perioperative morbidity [21]. Interestingly, our data went in accordance. Male gender percentage was significantly higher in the LSG group com- pared with the LRYGB group [46.9 % (23 patients) vs. 24.2 % (16 patients), respectively]. Similarly, mean BMI was signifi- cantly higher in the LSG group (54.2±8.3) compared with the LRYGB group (46.8±5.7). Additionally, there were more com- plications, especially bleeding (which was significantly more encountered) and a longer HS in the LSG group. Putting in consideration that the difference in male gender percentage between the whole group of patients who underwent LRYGB (1,345) and LSG (686) was statistically insignificant, and changed in the complicated group to become significantly Table 4 Demographic data and incidence of associated co-morbidities in patients with a prolonged HS after either LRYGB or LSG
LRYGB (66) LSG (49)
Males Females Males Females
BMI (kg/m2) <50 >50 <50 >50 <50 >50 <50 >50
Number 7 9 20 30 13 10 10 16
Median age in years (min.–max.) 46 (29–55) 39 (28–50) 48 (32–54) 48 (34–60) 42 (24–47) 50 (36–54) 51 (30–62) 45 (38–60) Mean BMI in kg/m2(±SD) 43.8 (±4.6) 52 (±3.4) 42 (±5.2) 50.4 (±5.7) 49.2 (±5.6) 55.1 (±4.7) 47.2 (±3.4) 60.2 (±4.1)*
Co-morbidities (N, %a)
-Type 2 diabetes 7 (100) 8 (88) 7 (35) 9 (30) 5 (38) 5 (50) 3 (30) 7 (43)
-Hypertension 6 (85) 9 (100) 5 (25) 7 (23) 6 (46) 5 (50) 4 (40) 7 (43)
-Hypertriglyceridemia 2 (28) 4 (44) 3 (15) 7 (23) 3 (23) 3 (30) 3 (30) 4 (25)
-Osteoarthritis 5 (71) 8 (88) 5 (25) 10 (33) 7 (53) 5 (50) 3 (30) 9 (56)
-Sleep apnea 3 (42) 3 (33) 4 (20) 5 (16) 3 (23) 3 (30) 1 (10) 3 (18)
SDstandard deviation
*p≤0.05, statistically significant
aPercentages of co-morbidities are calculated as percentages from the total number of patients in the corresponding column
Table 5 Main causes of prolonged HS after complicated LRYGB and LSG
LRYGB LSG pvalue
Males Females Total number (%) Males Females Total number (%)
BMI (Kg/m2) <50 >50 <50 >50 <50 >50 <50 >50
Leakage 6 4 6 6 22/1,345 (1.6) 4 3 2 3 12/686 (1.7) NS
Bleeding 2 4 2 2 10/1,345 (0.7) 4 4 6 5 19/686 (2.8) <0.05*
Extraluminal 2 2 2 1 4 3 4 4
Intraluminal – 2 – 1 – – – –
Combined – – – – – – 1 –
Abdominal wall – – – – – 1 1 1
Percentages are calculated as percentages from the total number of patients in our study (1,345 in LRYGB and 686 in LSG).Pvalue refers to all patients with either leakage or bleeding after LRYGB and LSG
NSnon-significant
*p≤0.05, statistically significant
higher in the LSG group may propose an association between male gender and post-LSG complications, especially bleeding.
However, the association between male gender, high BMI, and
post-operative complications (especially after LSG) is still in need for further research.
Post-operative complications may be either early or late [7]. Early complications, despite being relatively scarce, prolonged the HS in our clinic. The median length of HS after complicated LSG was not longer than that after complicated LRYGB (11 vs. 10 days).
Incidence of early complications following LSG wassig- nificantlyhigher than that following LRYGB (4.9 vs. 7.14 %).
This rate is comparable to the reported overall complication rate of 5.3 % after nonband procedures [22,23]. Consistent with our results, leakage and bleeding are the most commonly reported early post-operative complications [24].
The incidence of leakage after LRYGB ranges from 0.4 to 5.2 % [25–29]. We report a relatively low leakage rate of 1.63 %. Leakage rate decreased noticeably as our familiarity with the procedure increased. In 2009, 12 early leakages (3.39 %) were encountered after LRYGB vs. 6 cases (1.61 %) in 2010, 3 in 2011 (0.9 %), and only 1 case (0.3 %) in 2012. Authors refer a 40 % reduction in leakage rate to increased familiarity with the technique [30]. Seven sites for leakage after LRYGB are documented in the literature [31]. Only five sites were observed in this study, with gastojejunostomy (±gastric pouch) and enteroenterostomy representing the most common sites (nine patients, 40.9 % for each). Post-LSG leakage is also the most devastating complication of LSG, with a reported incidence of 1–7 % in the literature [32]. We found a leakage rate of 1.75 % after LSG, which is insignificantly higher than that after LRYGB.
Timely diagnosis of leakage is challenging. Tachycardia, fever, abdominal pain, hypotension, and low urinary output should all be handled with a high index of suspicion [33].
Those together with elevated inflammatory markers are strongly suggestive of complications, with serum CRP level Table 6 Causes of prolonged HS after LRYGB and LSG (other than
leakage and bleeding)
Cause of prolonged HS LRYGB LSG
Elevated inflammatory markers
31 18
Respiratory complications 9 5
Trocar site hernia 3 (gangrenous omentum in 1 patient and small bowel loop in 2 patients)
1
Small intestinal obstruction 2 –
Wound infection 2 1
Gastroenteritis 1 1
Arrhythmia 1 1
Intolerance to oral fluids 1 1
Deep vein thrombosis _ 1
Bile leakage at CBD stump from concomitant laparoscopic cholecystectomy
1 1
Erysipelas 1 –
Lower back pain 1 –
Douglas pouch abscess 1 –
Post-operative renal insufficiency
1 –
Post-operative urinary tract infection
– 2
Gluteal abscess – 1
Left shoulder dislocation – 1
Values refer to number of patients having this cause Some patients had more than one cause of prolonged HS
Table 7 Presentations of patients with leakage after LRYGB and LSG
Symptoms/signs/laboratory findings Patients with leakage after
LRYGB (N=22) (100 %)
Patients with leakage after LSG (N=12) (100 %)
Elevated inflammatory markers 19 (86.3 %) 12 (100 %)
Upper abdominal pain 9 (40.9 %) 3 (25 %)
Tachycardia 3 (13.6 %) 1 (8.3 %)
Fever 3 (13.6 %) 1 (8.3 %)
Tachypnea 2 (9 %) 1 (8.3 %)
Intolerance to oral intake/Nausea/Dysphagia 1 (4.5 %) 1 (8.3 %)
Dyspnea 1 (4.5 %) 3 (25 %)
Wound discharge due to wound infection 1 (4.5 %) 0 (0 %)
Oliguria (due to post-operative renal insufficiency) 1 (4.5 %) 0 (0 %)
Lower abdominal pain (due to Douglas pouch abscess diagnosed by CT) 1 (4.5 %) 0 (0 %)
Excessive bloody drain effluent 0 (0 %) 1 (8.3 %)
Values refer to number (percentage) of patients having the corresponding symptom, sign, or laboratory finding. Summation of individual numbers may not coincide with the total number (22 in LRYGB group and 12 in LSG group) due to overlapping clinical presentations
being a more reliable indicator of complications than leuko- cytic count [34]. We routinely depend on elevated serum CRP levels and leukocytosis, aided by contrast-enhanced radiological studies, as indications of leakage. Contrast- enhanced CT series are of a higher sensitivity than upper gastrointestinal series [35, 36]. Accordingly, we routinely adopt the triad of clinical suspicion, altered laboratory pro- file, and CT evidence to confirm a suspected leakage. Once diagnosed, reoperation is a standard for early leakage after LRYGB [33]. All patients in our series (22 patients, 100 %) underwent re-exploration, whether laparoscopic (55.54 %) or open (45.46 %). Repair of the leakage site, lavage, and suction drainage were standard practices. Reoperation com- bined with endoscopic stenting was practiced in only one patient (4.54 %) due to fear of stent migration. Moreover, endoscopic stenting is applicable only in gastric pouch leak- ages. In contrast, endoscopic stenting is well established for post-LSG leakage [37,38], especially when combined with reoperation [39]. In our series, reoperation was required in 11 patients (91.67 %). In eight of them (66.67 %), endoscopic stenting was implemented. Endoscopic stenting as a sole intervention was used effectively in one patient (8.33 %).
Early post-operative hemorrhage is another serious com- plication, with an incidence reaching 4 % after LRYGB [40, 41] and 3.5 % after LSG [42]. Our results suggest that the bleeding rate was significantly higherafter LSG than after LRYGB (2.76 vs. 0.7 %). Features such as excessive bloody drain effluent, tachycardia, drop in the hemoglobin level, hematemesis, and melena warranted a further work-up [27].
Contrast-enhanced CT series have been demonstrated to de- tect all possible bleeding sites, whether intraluminal, extraluminal, or abdominal wall bleeding [43]. In this study, extraluminal bleeding was the most commonly encountered pattern after either LRYGB or LSG (78.9 % of cases with post-LRYGB bleeding vs. 70 % after LSG). Conservative management (fluid resuscitation±blood transfusion) was suc- cessful in three patients (30 %) after LRYGB vs. six patients (31.6 %) after LSG. Reoperation is limited to hemodynamically
unstable patients [41]. This was met with in 7 patients (70 %) after LRYGB vs. 13 patients (68.4 %) after LSG.
Modification of surgical technique has always been prac- ticed, aiming at reducing post-operative complications.
Staple-line reinforcement is an example, with controversy of ideas regarding its efficacy. Some authors attributed a shorter operative time and fewer complications, including leakage, after LRYGB to the use of buttressing materials [44]. Others linked a less bleeding rate along the staple line to the use of Bioabsorbable Seamguard® in LSG [45]. Other studies showed however no clear benefit with the use of buttressing materials [12, 46]. Based on our experience, the use of buttressing materials decreased the rate of our complications after LSG [47]. That is why we recommend staple-line rein- forcement during LSG, where a long stapling distance and cutting through a relatively thick antrum may adversely affect the perioperative safety. Both antral cutting and a long stapling distance are not encountered in LRYGB. Therefore, staple- line reinforcement, in our opinion, is not needed in LRYGB.
Respiratory complications are reported in the literature with an incidence reaching 4.4 % [41]. We encountered respi- ratory complications in only nine patients (0.67 %) and five patients (0.73 %) after LRYGB and LSG, respectively. Med- ical management was satisfactory. Trocar site herniation, ne- cessitating laparoscopic revision, was observed in 0.22 % of patients after LRYGB and 0.14 % of patients after LSG. The average reported incidence for this complication is 1 % [48].
Deep venous thrombosis (DVT) is another serious complica- tion with significant mortality, if complicated by pulmonary embolism, with an incidence of 0–1.3 % in large-size studies [49]. We encountered only one case (0.14 %) of DVT after LSG. This was diagnosed by lower limb Duplex study and managed conservatively.
Interestingly, asignificantlylarger number of patients devel- oped early complications after LSG compared with LRYGB in our data. This is in contrast to data from the USA. The Amer- ican College of Surgeons data showed a higher 30-day mor- bidity rate after LRYGB vs. LSG (5.91 vs. 5.61 %, respective- ly) [50]. The Michigan Bariatric Surgery Collaborative data confirmed these results where 30-day morbidity rate was higher after LRYGB vs. LSG (3.6 vs. 2.2 %) [51]. Analysis of the Bariatric Outcome Longitudinal Database showed that LRYGB was followed by a higher morbidity rate (14.8 %) than that after LSG (10.8 %) [52]. This difference between our series and data from the USA could be explained by a larger cohort in series from the USA as well as a much longer post-operative follow-up period (30 days), compared with only five post- operative days in our study. Monocentric design in our study and ethnic differences could play another role.
Mortality after LRYGB is reported in data of the Ameri- can College of Surgeons, being insignificantly higher than that after LSG (0.14 vs. 0.11 %, respectively) [50]. Data of the Michigan Bariatric Surgery Collaborative revealed as Table 8 Interventions for patients presenting with leakage after either
LRYGB or LSG
LRYGB (N=22)
LSG (N=12) -Reoperation
(over sewing, lavage and suction drainage)
Laparoscopic Plus stenting 1 6 Without stenting 11a 3
Open Plus stenting – 2
Without stenting 10 –
-Only stenting – 1
Values refer to number of patients in whom the corresponding technique was used
aRe-do of enteroenteral anastomosis was necessary in two patients
well a higher post-LRYGB mortality rate when compared with LSG (0.14 vs. 0 %, respectively) [51]. One patient (0.07 %) died after LRYGB in our clinic as a result of uncontrolled sepsis following leakage. For the same reason, one patient (0.14 %) died after LSG.
A limitation of this study was the lack of classification of the sites of post-LSG leakage. Selection bias between both procedures represented another limitation. After LRYGB, there should have been a differentiation between proximal gastric pouch and gastrojejunostomy leaks. The relative inci- dence of different presentations of bleeding was also lacking.
A more accurate assessment of the sensitivity and specificity of different diagnostic modalities was necessary.
Conclusion
LRYGB and LSG are performed in increasing numbers.
Their complications are uncommon but require timely man- agement. Elevation of serum inflammatory markers appears to be highly suggestive of complications, especially when confirmed by radiological signs. LSG was associated with a significantlyhigher incidence of early complications, com- pared with LRYGB, leading to an insignificantly longer median length of HS. Leakage and bleeding were the most common early complications. The incidence of bleeding was significantlyhigher after LSG than after LRYGB. The higher rate of complications after LSG may be related to the pre- dominance of males and the higher BMI in this group.
Conflict of Interest None.
References
1. Abbatini F, Rizzello M, Casella G, et al. Long-term effects of laparoscopic sleeve gastrectomy, gastric bypass, and adjustable gastric banding on type 2 diabetes. Surg Endosc. 2010;24:1005–10.
2. Buchwald H, Estok R, Fahrbach K, et al. Weight and type 2 diabetes after bariatric surgery: systematic review and meta- analysis. Am J Med. 2009;122:248–56.
3. Buchwald H, Avidor Y, Braunwald E, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA. 2004;292:1724–37.
4. Buchwald H, Oien D. Metabolic/bariatric surgery worldwide 2008.
Obes Surg. 2009;19:1605–11.
5. Gagner M, Deitel M, Kalberer T, et al. The Second International Consensus Summit for Sleeve Gastrectomy, March 19–21, 2009.
Surg Obes Relat Dis. 2009;5:476–85.
6. Santry H, Gillen D, Lauderdale D. Trends in bariatric surgical procedures. JAMA. 2005;294:1909–17.
7. Griffith P, Birch D, Sharma A, et al. Managing complications associated with laparoscopic Roux-en-Y gastric bypass for morbid obesity. Can J Surg. 2012;55:329–36.
8. NIH Conference. Gastrointestinal surgery for severe obesity. Consen- sus Development Conference Panel. Ann Intern Med. 1991;115:956–
61.
9. Runkel N, Colombo-Benkmann M, Hüttl T, et al. Evidence-based German guidelines for surgery for obesity. Int J Colorectal Dis.
2011;26:397–404.
10. Stroh C, Luderer D, Weiner R, et al. Actual situation of thromboem- bolic prophylaxis in obesity surgery: data of quality assurance in bariatric surgery in Germany. Thrombosis. 2012. doi:10.1155/2012/
209052.
11. American Society for Metabolic and Bariatric Surgery (2009) www.asmbs.org/Newsite07/media/ASMBS_Metabolic_Bariatric_
Surgery_Overview_FINAL_09.pdf. Accessed 23 March 2011 12. Parikh M, Issa R, McCrillis A, et al. Surgical strategies that may
decrease leak after laparoscopic sleeve gastrectomy: a systematic review and meta-analysis of 9991 cases. Ann Surg. 2013;257:231– 7.
13. Silecchia G, Boru C, Pecchia A, et al. Effectiveness of laparoscopic sleeve gastrectomy (first stage of biliopancreatic diversion with duodenal switch) on co-morbidities in super-obese high-risk pa- tients. Obes Surg. 2006;16:1138–44.
14. Gagner M, Boza C. Laparoscopic duodenal switch for morbid obesity. Exp Rev Med Devices. 2006;3:105–12.
15. Lancaster R, Hutter M. Bands and bypasses: 30-day morbidity and mortality of bariatric surgical procedures as assessed by prospec- tive, multicenter, risk-adjusted ACS-NSQIP data. Surg Endosc.
2008;22:2554–63.
16. Dallal R, Datta T, Braitman L. Medicare and Medicaid status pre- dicts prolonged length of stay after bariatric surgery. Surg Obes Relat Dis. 2007;3:592–6.
17. McCarty T, Arnold D, Lamont J, et al. Optimizing outcomes in bariatric surgery: outpatient laparoscopic gastric bypass. Ann Surg.
2005;242:494–501.
18. Benotti P, Wood G, Winegar D, et al. Risk factors associated with mortality after Roux-en-Y gastric bypass surgery. Ann Surg. 2013.
doi:10.1097/SLA.0b013e31828a0ee4.
19. Buchwald H, Estok R, Fahrbach K, et al. Trends in mortality in bariatric surgery: a systematic review and meta-analysis. Surgery.
2007;142:621–35.
20. Morais A, Faintuch J, Leal A, et al. Inflammation and biochemical features of bariatric candidates: does gender matter? Obes Surg.
2011;21:71–7.
21. Stroh C, Köckerling F, Weiner R, et al. Are there gender-specific aspects of sleeve gastrectomy—data analysis from the quality as- surance study of surgical treatment of obesity in Germany. Obes Surg. 2012;22:1214–9.
22. Inabnet W, Belle S, Bessler M, et al. Comparison of 30-day out- comes after non-LapBand primary and revisional surgical proce- dures from the Longitudinal Assessment of Bariatric Surgery study.
Surg Obes Relat Dis. 2010;6:22–30.
23. LABS Consortium Writing Group. Perioperative safety in the Lon- gitudinal Assessment of Bariatric Surgery. New Engl J Med.
2009;361:445–54.
24. Skroubis G, Karamanakos S, Sakellaropoulos G, et al. Comparison of early and late complications after various bariatric procedures:
incidence and treatment during 15 years at a single institution.
World J Surg. 2011;35:93–101.
25. Fernandez A, DeMaria E, Tichansky D, et al. Experience with over 3,000 open and laparoscopic bariatric procedures: multivariate anal- ysis of factors related to leak and resultant mortality. Surg Endosc.
2004;18:193–7.
26. Marshall J, Srivastava A, Gupta SK, et al. Roux-en-Y gastric bypass leak complications. Arch Surg. 2003;138:520–3.
27. Bellorin O, Abdemur A, Sucandy I, et al. Understanding the sig- nificance, reasons and patterns of abnormal vital signs after gastric bypass for morbid obesity. Obes Surg. 2011;21:707–13.
28. Fullum T, Aluka K, Turner P. Decreasing anastomotic and staple line leaks after laparoscopic Roux-en-Y gastric bypass. Surg Endosc.
2009;23:1403–8.
29. Madan A, Martinez J, Menzo E, et al. Omental reinforcement for intraoperative leak repairs during laparoscopic Roux-en-Y gastric bypass. Am Surg. 2009;75:839–42.
30. Ballesta C, Berindoague R, Cabrera M, et al. Management of anastomotic leaks after laparoscopic Roux-en-Y gastric bypass.
Obes Surg. 2008;18:623–30.
31. Attila Csendes A, Burgos A, Braghetto I. Classification and manage- ment of leaks after gastric bypass for patients with morbid obesity: a prospective study of 60 patients. Obes Surg. 2012;22:855–62.
32. Deitel M, Gagner M, Erickson A, et al. Third international summit:
current status of sleeve gastrectomy. Surg Obes Relat Dis. 2011;7:749– 59.
33. Herron D, Roohipour R. Complications of Roux-en-Y gastric by- pass and sleeve gastrectomy. Abdom Imaging. 2012;37:712–8.
34. Warschkow R, Tarantino I, Folie P. C-Reactive protein 2 days after laparoscopic gastric bypass surgery reliably indicates leaks and moderately predicts morbidity. J Gastrointest Surg. 2012;16:1128– 35.
35. Kolakowski S, Kirkland M, Schuricht A. Routine postopera- tive upper gastrointestinal series after Roux-en-Y gastric by pass: determination of whether it is necessary. Arch Surg.
2007;142:930–4.
36. Yu J, Turner M, Cho S, et al. Normal anatomy and complications after gastric bypass surgery: helical CT findings. Radiology. 2004;231:753–
60.
37. Blackmon S, Santora R, Schwarz P, et al. Utility of removable esophageal covered self-expanding metal stents for leak and fistula management. Ann Thorac Surg. 2010;89:931–6.
38. Tan J, Kariyawasam S, Wijeratne T, et al. Diagnosis and manage- ment of gastric leaks after laparoscopic sleeve gastrectomy for morbid obesity. Obes Surg. 2010;20:403–9.
39. Sakran N, Goitein D, Raziel A, et al. Gastric leaks after sleeve gastrectomy: a multicenter experience with 2,834 patients. Surg Endosc. 2013;27:240–5.
40. Bakhos C, Alkhoury F, Kyriakides T, et al. Early post-operative hemorrhage after open and laparoscopic Roux-en-Y gastric bypass.
Obes Surg. 2009;2:153–577.
41. Schauer P, Ikramuddin S, Gourash W, et al. Outcomes after lapa- roscopic Roux-en-Y gastric bypass for morbid obesity. Ann Surg.
2000;232:515–29.
42. Triantafyllidis G, Lazoura O, Sioka E, et al. Anatomy and compli- cations following laparoscopic sleeve gastrectomy: radiological evaluation and imaging pitfalls. Obes Surg. 2011;21:473–8.
43. Kothari S. Bariatric surgery and postoperative imaging. Surg Clin N Am. 2011;91:155–72.
44. Sajid M, Khatri K, Singh K, et al. Use of staple-line reinforcement in laparoscopic gastric bypass surgery: a meta-analysis. Surg Endosc. 2011;25:2884–91.
45. Consten E, Gagner M, Pomp A, et al. Decreased bleeding after laparoscopic sleeve gastrectomy with or without duodenal switch for morbid obesity using a stapled buttressed absorbable polymer membrane. Obes Surg. 2004;14:1360–6.
46. Giannopoulos G, Tzanakis N, Rallis G, et al. Staple line reinforce- ment in laparoscopic bariatric surgery: does it actually make a difference? A systematic review and meta-analysis. Surg Endosc.
2010;24:2782–8.
47. Daskalakis M, Berdan Y, Theodoridou S, et al. Impact of surgeon experience and buttress material on postoperative complications after laparoscopic sleeve gastrectomy. Surg Endosc. 2011;25:88–97.
48. Chiu C, Lee W, Wang W, et al. Prevention of trocar-wound hernia in laparoscopic bariatric operations. Obes Surg. 2006;16:913–8.
49. Brethauer S, Schauer P. Risk–benefit analysis of laparoscopic bariatric procedures. In: Schauer P, Schirmer B, Brethauer S, editors. Minimally invasive bariatric surgery. New York: Springer; 2007. p. 371.
50. Hutter M, Schirmer B, Jones D, et al. First report from the American College of Surgeons Bariatric Surgery Center Network: laparoscopic sleeve gastrectomy has morbidity and effectiveness positioned be- tween the band and the bypass. Ann Surg. 2011;254:410–22.
51. Birkmeyer N, Dimick J, Share D, et al. Hospital complication rates with bariatric surgery in Michigan. JAMA. 2010;304:435–42.
52. DeMaria E, Pate V, Warthen M, et al. Baseline data from American Society for Metabolic and Bariatric Surgery-designated Bariatric Surgery Centers of Excellence using the Bariatric Outcomes Lon- gitudinal Database. Surg Obes Relat Dis. 2010;6:347–55.
