G Model
ACCPM-6; No. of Pages 8 Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
Review
Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery Philippe Montravers *, Pascal Augustin, Nathalie Zappella, Guillaume Dufour, Konstantinos Arapis, Denis Chosidow, Pierre Fournier, Lara Ribeiro-Parienti, Jean-Pierre Marmuse, Mathieu Desmard AP–HP, CHU Bichat–Claude-Bernard, De´partement d’Anesthe´sie Re´animation, 46, rue Henri-Huchard, 75018 Paris, France
A R T I C L E I N F O
A B S T R A C T
Article history: Available online xxx
Perioperative complications following bariatric surgery (BS) have been poorly analysed and their management is not clearly assessed. The associated frequency of ICU admission is difficult to estimate. Among surgical complications, digestive perforations are the most frequent. The most common postoperative complications of sleeve gastrectomy are fistulas, but bleeding on the stapling line is also commonly reported. Complication rates are higher after Roux-en-Y gastric bypass, mainly due to anastomotic leaks. Medical complications are mainly thromboembolic or respiratory complications. All these surgical and medical complications are not easily detected; clinical signs can be atypical or insidious, often resulting in delayed management. Respiratory signs can be predominant and lead erroneously to pulmonary or thromboembolic diseases. Diagnostic criteria are based on minor clinical signs, tachycardia being probably the most frequent one. Lately, complications are revealed by haemodynamic instability, respiratory failure or renal dysfunction and radiographic findings. Management decision according to these abnormal signs is based on a combined multidisciplanary approach including surgical and/or endoscopic procedures and medical care, depending on the nature and severity of the surgical complication. Medical management is based on supportive ICU care of organ dysfunctions, curative anticoagulation if required, nutritional support, and appropriate anti-infective therapy. Pharmacological data are limited in morbidly obese patients and the appropriate doses are debated, especially for anti-infective agents. Complicated BS cases have a poor outcome, probably largely related to delayed diagnosis and reoperation. ß 2015 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Keywords: Bariatric surgery Morbid obesity Pulmonary embolism Postoperative pneumonia Postoperative peritonitis Sepsis Multidrug-resistant bacteria Antibiotic adequacy
1. Introduction Bariatric surgery (BS) is very commonly used for the treatment of morbid obesity [1]. Most studies have demonstrated these procedures to be safe and well tolerated in the vast majority of cases, with decreased overall mortality and morbidity rates [2,3]. As usual, however, the benefits of a new treatment must be weighed against the possible associated risks. BS has been reported to be associated with a growing number of perioperative complications and postoperative or short-term adverse outcomes over recent years. From the anaesthesiologist’s perspective, these morbidly obese patients represent a new population that was extremely rare thirty
* Corresponding author. Tel.: +33 1 40 25 83 55; fax: +33 1 40 25 63 09. E-mail address: [email protected] (P. Montravers).
years ago. While the intraoperative period has been well described and is uneventful in most cases, challenging issues may occur during the postoperative course that deserve specific attention and require intensive care unit (ICU) admission. This article is designed to review the postoperative surgical and medical complications reported after the most commonly performed BS procedures. We describe the clinical conditions and management of these high-risk patients according to the specificities of the procedure performed. 2. Surgical procedures Many techniques have been proposed for the surgical management of morbid obesity. These operations can be divided into restrictive and malabsorptive procedures [1] that are designed to reduce caloric intake by dramatically modifying the upper mesocolic gastrointestinal tract.
http://dx.doi.org/10.1016/j.accpm.2014.06.002 2352-5568/ß 2015 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 2
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
2.1. Restrictive procedures Several surgical techniques that decrease food intake by creating a small gastric pouch and result in rapid satiety after food intake have been described. Adjustable gastric banding was the first technique to become widely popular. An adjustable silicone ring is placed around the stomach, to create a small gastric pouch with a narrow outlet to induce delayed emptying. The degree of gastric stricture and consequent food restriction is adjusted by means of saline injections into a subcutaneous port. The procedure is usually well tolerated and is now frequently proposed as day-only surgery in patients with no concomitant disease. Vertical restrictive (sleeve) gastrectomy is now frequently proposed as a first-line treatment in replacement of adjustable gastric banding due to its better efficacy in terms of weight loss [4]. The procedure removes two-thirds of the gastric volume and leaves a gastric tube by stapling the entire length of the greater curvature. 2.2. Malabsorptive operations Malabsorptive procedures bypass various portions of the small intestine and reduce jejunal absorption. Proximal Roux-en-Y gastric bypass (RYGB) combines restrictive and malabsorptive procedures and is often considered to be the gold standard because of its excellent efficacy and good safety. RYGB comprises stapling the stomach to create a small upper gastric pouch and division of the small intestine at the mid-jejunum level, while the distal portion (the alimentary, or Roux limb) is anastomosed to the gastric pouch. Intestinal (jejunal) absorption is progressively decreased with increasing length of the Roux limb. Two anastomoses (gastrointestinal and jejuno-intestinal anastomoses) are performed. Other malabsorptive procedures have been proposed, but are often responsible for severe or extreme malabsorption. This is the case of biliopancreatic diversion and duodenal switch (BPD/DS) procedures, sometimes proposed in super-obese patients (BMI > 50 kg/m2) or after failure of previous BS. This procedure comprises sleeve gastrectomy and a short duodeno-ileal circuit with duodeno-ileal and ileo-ileal anastomoses. 3. Epidemiology of surgical complications Descriptions of postoperative surgical complications are based on reports of general or surgical site complications. Very few data are available concerning the incidence and clinical characteristics of complications requiring ICU management [5–7]. Although most surgical complications occur during the procedure, the time to diagnosis usually peaks during the early postoperative period, in the first days or weeks following surgery [8–11]. Surgical complications can be divided into two entities: septic complications, mainly fistula and anastomotic leak, and nonseptic complications, mostly due to haemorrhage. The vast majority of these complications are treated in surgical units and never require ICU management. Anastomotic leaks are a very common complication and are more common than pulmonary embolism in some series [12,13]. In a series of 107 autopsies following BS, anastomotic leak was the leading cause of death among the 97 deaths directly related to surgical complications, observed in 36% of cases, while only 13% of deaths were due to pulmonary embolism [14]. 3.1. Complications of restrictive procedures 3.1.1. Complications of adjustable gastric banding The most recent meta-analysis, comprising more than 850 patients included in randomised control trials (RCT) and
36 000 patients from observational studies (OS), estimated the complication rate of adjustable gastric banding to be 13% in RCTs, while a lower rate was observed in OS (7.8%) [3] (Fig. 1A). Although the procedure appears to be safe, many early complications have been described which can be life-threatening if the diagnosis is delayed. Gastric or oesophageal perforation is a rare complication that is sometimes diagnosed intraoperatively and constitutes a contraindication to foreign body insertion. In an analysis conducted by the French ANAES agency, the incidence of gastric perforation was 0.3% (15/5237 patients) [15]. Early gastric band erosion and perforation of the gastric pouch have been reported during the first postoperative weeks [16]. Erosion is usually related to a technical problem and unrecognized intraoperative gastric perforation. Early postoperative infection or gastric wall ischaemia secondary to a tight band could also account for these erosions. An initially unrecognized perforation can also be revealed by early ring or port suppuration. Ring or port infection, observed in 1.5% of cases, may lead to suppuration at the port site [17]. Interestingly, port infection is not systematically associated with ring infection, and the ring can be left in place in the case of isolated port infection. Ring malposition and band slippage are early but rare postoperative complications [17].
3.1.2. Complications of sleeve gastrectomy The sleeve gastrectomy complication rate has been estimated to be 13% in RCTs and 8.9% in OSs [3] (Fig. 1B). The most common complications during the postoperative course of sleeve gastrectomy are fistulas, reported in 0 to 20% of cases after gastric banding removal [18]. Parikh et al. [19] reported a fistula rate of 2.2% in a meta-analysis of almost 10 000 sleeve gastrectomies. These leaks usually occur in the upper part of the gastric tube, in the proximal third of the stomach, close to the angle of His, in up to 89% of cases [20,21]. Bleeding is the most frequent nonseptic complication of sleeve gastrectomy. The most common source of bleeding is the stapling line, observed in approximately 3.5% of cases [22]. 3.2. Complications of malabsorptive procedures Complication rates are higher after malabsorptive procedures than after restrictive surgery. A meta-analysis by Chang et al. [3], based on more than 600 patients included in RCTs and 71 000 patients from OSs, demonstrated an estimated complication rate of 21% in RCTs and 12% in OSs. Anastomotic leaks are the most common complications of RYGB, with a reported incidence ranging from 0 to 8% [12,13]. A fistula can arise at each segment of the RYGB. The fistula rates reported in the literature are presented in Fig. 1C and Table 1 [8–11,23,24]. The bleeding rate after RYGB is significantly higher than after other BS procedures and is reported in 1 to 4.4% of cases [25,26], requiring a complex diagnostic process to identify the source. The site of bleeding may be situated in the abdominal cavity from one of the staple lines. Another source of bleeding is intraluminal bleeding (gastric or intestinal), which may be difficult to detect [25]. Complications following BPD/DS comprise all of the complications reported following sleeve gastrectomy and gastric bypass. Single-centre studies have reported discordant results following BPD/DS, with either similar complication rates to those observed after other procedures or higher overall complication rates with a trend towards higher bleeding rates [2,27].
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
3
Fig. 1. Sites and frequency of the most frequent surgical complications observed during adjustable gastric banding (A), vertical restrictive sleeve gastrectomy (B) and proximal Roux-en-Y gastric bypass (C).
4. Epidemiology of medical complications The timing of postoperative complications has not been specifically assessed in bariatric surgery. Similarly to other procedures, a majority of medical complications could occur before patients were discharged. In a cohort of 13 871 patients submitted to various bariatric procedures, Morino et al. [28] reported that medical complications caused death within a mean delay of 17 days. Thromboembolic complications accounted for 13 to 38.2% of fatalities [14,28]. In a literature analysis gathering more than 20 publications and focusing on postoperative complications, 58.6% of deaths were related to pulmonary embolism [15]. Non-fatal cases of pulmonary embolism following gastric bypass are reported in various proportions between 0 and up to 4% of cases [15,29]. Recent studies in RYGBP reported incidences ranging between 1.2% and 8%
of cases [5,9,24] and up to 12.5% in a study pooling pulmonary embolism and deep venous thromboses [23]. Respiratory failure is the second respiratory cause of death reported in 11.8% of the fatalities [28]. In non-fatal cases, the incidence of respiratory failure varies largely, depending on the definition, from 10 [24] to 29% of cases [23]. Pleural effusions are noticed in 5 [24] to 23% of patients [23], while atelactasias are reported between 6 [8] and 92% of cases [23]. Pneumonias are reported in various proportions, between 0.1% in laparoscopic bypass (0.33% in open bypass) [29] to 2–3% [9,24] and even 22% for Ballesta et al. [8]. Cardiovascular failure and cerebral ischemia are the two other major medical causes of death reported by Morino et al. [28] in 17.6 and 2.9% of cases, respectively. However, non-fatal cardiovascular complications are rarely reported [23,24].
Table 1 Site of the leak following Roux-en-Y gastric bypass (RYGB).
Ballesta et al. [8], n (%) Carucci et al. [23], n (%) Gonzalez et al. [24], n (%) Hamilton et al. [10], n (%) Marshall et al. [11], n (%) Thodiyil et a. [9], n (%)
Number of procedures
Leaks
Gastric pouch
Gastrojejunal anastomosis
Excluded stomach
Jejuno-jejunal anastomosis
Other sites
1200 904 3018 246 400 2675
59 48 63 9 21 46
6 5 6 1 – 4
40 37 33 3 13 37
2 15 17 2 2 2
3 1 9 2 4 1
4 7 – – 2 2
(5) (5) (3) (4) (5) (2)
(10) (10) (10) (11) (9)
(68) (77) (52) (33) (62) (80)
(3) (31) (27) (22) (10) (4)
(5) (2) (14) (22) (21) (2)
(7) (15)
(10) (4)
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
4
Rhabdomyolysis is reported in more than 20 publications over the last decade and seems to be more frequent in male patients with high BMI (52 kg/m2) who underwent prolonged bariatric surgery [30]. Acute renal failure was observed in 14% of these cases. Renal dysfunction can also be associated with sepsis as reported in postoperative cases in 35% [31] to 58% of cases [32].
or bowel obstruction (n = 2) [31]. These symptoms usually deteriorate over several days leading to organ dysfunction and/ or ICU admission. Interestingly, the frequency of organ failure in BS patients admitted to the ICU for postoperative infection did not differ from those reported in patients with other common forms of postoperative peritonitis [32].
5. Diagnosis of surgical complications
5.2. Radiological assessment
5.1. Clinical presentation
Radiological imaging is strongly recommended in the diagnostic work-up of postoperative complications [34]. However, the diagnosis can be difficult due to the lack of specificity of imaging studies, especially at the early postoperative phase. No consensus has been reached concerning the optimal imaging technique (CTscan or upper gastrointestinal [UGI] studies) to be used in these patients. In a cohort of 300 patients who underwent RYGB, Madan et al. [35] used routine UGI studies on the first postoperative day, and reported positive and negative predictive values for leak identification of 67% and 99%, respectively. Moreover, these authors observed that routine UGI studies were more predictive of early leak diagnosis than clinical signs. Few studies are available assessing the capacity of CT to detect surgical complications after BS, and more specifically anastomotic leaks. Lyass et al. [36] proposed the use of CT rather than UGI studies because of its capacity to detect complications other than anastomotic leaks. Interestingly, no study has ever assessed the specificity and sensitivity of CT in the diagnosis of complicated BS, possibly because of the limitations on the use of imaging procedures in this population: patients exceeding the weight limit of CT-scan systems (above 150 to 200 kg of body weight), limited use of contrast agents in these patients who present a high rate of renal impairment, and/or the radiologist’s limited experience. These technical limitations led some authors to reoperate without radiological proof of complications, as in 2% (1/59) of cases of complicated RYGB for Ballesta et al. [8], 10% (2/21) of cases for Marshall et al. [11], or 18% (27/150) of cases for Lee et al. [33]. Chest radiograph abnormalities are not uncommon in complicated BS patients admitted to the ICU. Kermarrec et al. [31] observed basal pulmonary atelectasis and/or left-sided pleural effusion in 37% and 44% of complicated BS patients, respectively. Upper GI investigations were performed in only 44% of these patients and were reported as normal in one half of them. Similarly, abdomen CT-scan was performed in only one half of cases in a median of 8 days after initial surgery. Imaging studies were reported to be normal in 4 patients with generalized (n = 2) or localized (n = 2) intra-abdominal sepsis. Abnormal CT-scans revealed subphrenic collections (n = 7), supramesocolic collections (n = 2), and small bowel obstruction (n = 1) [31]. In summary, the decision-making process is based on abnormal clinical signs, especially signs of haemodynamic instability,
Obvious surgical complications may be diagnosed intraoperatively. Unfortunately, in the vast majority of cases, surgical complications are not easily detected, consequently often resulting in delayed management. For clinicians not used to the management of BS patients, it is important to remember that these morbidly obese patients constitute an atypical population that differs from the general population of abdominal surgery patients. In a recent publication, we compared the clinical characteristics and management of 134 non-obese patients admitted to our ICU for ‘‘common’’ postoperative peritonitis and a group of 49 patients who developed postoperative peritonitis following BS [32]. The BS patients were younger (45 10 years versus 63 16 years; P < 0.0001), had low cancer rates and lower rates of underlying fatal disease (39% vs. 64%; P = 0.002). However, the time to diagnosis of surgical complications was similar in the two groups of patients [32]. Identification of these complications before the onset of organ dysfunction is frequently based on minor clinical signs, making the diagnosis particularly difficult [10,12] (Table 2). Tachycardia is probably the most frequent sign, largely reported in the literature [8–11,23,24,33]. Polypnea or respiratory distress are other sensitive indicators of gastrointestinal leak [9,10,23]. It is noteworthy that the absence of clinical signs at the time of diagnosis of peritonitis is reported in up to 49% of non-ICU patients [8,9,24]. The reasons for the limited value of clinical signs have not been elucidated, but could be related to the large mass of subcutaneous tissue, the subphrenic site of intraperitoneal sepsis, or impairment of the inflammatory response. A similar low incidence of clinical symptoms was also reported in ICU BS cases [31]. The most common symptoms, fever, dyspnoea and tachycardia, were reported in up to 74, 98 and 100% of cases, respectively [31]. Local and abdominal signs were rarely reported; tenderness, pus, and ileus were observed in only 30%, 33% and 37% of cases, respectively [31]. In many cases, the predominance of respiratory signs may be more suggestive of pleural or pulmonary diseases. In such circumstances, an incorrect initial diagnosis was reported in more than 50% (15/27) of patients. The most frequent erroneous hypotheses proposed by attending teams were pneumonia (n = 7), pulmonary embolism (n = 4), wound abscess (n = 2),
Table 2 Clinical presentation at the time of diagnosis of anastomotic leaks in four cohorts of patients who underwent Roux-en-Y gastric bypass (RYGB).
Number of patients, n Absence of clinical signs, n (%) Presence of abdominal pain, n (%) Presence of left shoulder pain, n (%) Presence of nausea/vomiting, n (%) Abnormal drainage, n (%) Fever > 38 8C, n (%) White blood cell count > 12,000/mm3, n (%) Tachycardia, n (%) Polypnoea, n (%) Hypotension, n (%) Oliguria, n (%) a
Ballesta et al. [8]
Carucci et al. [23]
Gonzalez et al. [24]
Thodiyil et al. [9]
59 29 (49) 30 (51) – 11 (19) 29 (49) 29 (49) – 13 (22) – 7 (12) 8 (13)
48 – 13 (27) 9 (19) 39 (81) 3 (6) 44 (92)a 44 (92)a 44 (92)a 8 (17) 7 (15) –
63 5 (8) 34 (54) 9 (14) 11 (17) 15 (24) 40 (63) 47 (75) 45 (72) – 11 (17) 13 (21)
46 15 (33) 11 (24) 4 (9) 3 (7) 13 (28) 13 (28) 31 (67) 8 (17) 4 (9) 1 (2) 2 (4)
Elevated white blood cell counts, fever and/or tachycardia in 92% of patients.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
5
Abnormal clinical signs
Clinical stability
Clinical instability
Reoperation (laparoscopy or laparotomy / depending on clinical severity)
Upper gastrointestinal imaging
Radiographic signs
CT scan if possible Consider reoperation
No abnormal signs
CT scan cannot be performed
Reoperation Laparoscopy
CT scan available
Radiographic signs Consider reoperation
No signs Survey
Fig. 2. Diagnostic strategy in the case of suspected fistula.
respiratory failure or less frequently renal dysfunction and radiographic findings. The medical team may decide to reoperate in the presence of severe clinical signs, even when there is limited proof of surgical complications. In doubtful cases with a reassuring clinical status, the team may prefer watchful waiting. Based on our experience and a review of the literature, we routinely use the decision tree presented in Fig. 2 as a possible diagnostic approach in patients with suspected fistula. 6. Management of complications Management of complicated BS cases is based on a combined approach including surgical and/or endoscopic procedures and medical care, largely depending on the nature and severity of the complication. The severity of the clinical features at the time of diagnosis is the most important factor guiding the treatment strategy. 6.1. Surgical management General surgical management corresponds to the conventional rules of septic surgery including identification of the infectious source, collection of microbiological samples, peritoneal lavage and possibly drainage depending on the intensity and source of contamination, and elimination/control of the source of infection. In the presence of bleeding, surgery is required to identify the source of bleeding and to control blood loss. The treatment strategy is usually based on second-look laparoscopy possibly converted to laparotomy depending on the surgeon’s experience, technical conditions and the patient’s general status. All devices (gastric ring, tube, port, etc.) must be removed. Based on our experience and a review of the literature, we routinely use the decision tree presented in Fig. 3 for the management of all fistulas observed after bariatric surgery. 6.1.1. Complicated gastric banding Gastric or oesophageal perforation following gastric banding must be treated by abundant lavage and efficient drainage of the
peritoneal cavity. The perforation must be sutured together with drainage in contact with the perforation in every case. Infection of the ring or port of gastric banding requires removal of the device. Placement of a new device is deferred until eradication of the infectious source. Band malposition requires urgent reoperation and removal or repositioning of the gastric band [17,37]. 6.1.2. Complicated sleeve gastrectomy Reoperation must include extensive peritoneal toilet, sometimes combined with suture and drainage in contact with the fistula. The outcome is usually uneventful when the fistula is adequately closed. Endoscopic placement of a prosthesis into the gastric leak has been proposed as either first-line therapy or as second-line therapy after failure to control the infection [18]. Radiological drainage techniques using pigtail drains have also been proposed [18]. In the case of recurrent failure of these strategies or in the presence of necrosis of the stomach, total gastrectomy with primary or delayed reconstruction by oesophagojejunal anastomosis has been successfully proposed [32,38].
Proven fistula
Clinical instability - hemodynamic - respiratory - renal
Reoperation
Clinical stability - hemodynamic - respiratory - renal
No collection
Limited collection
Generalized collection
Conservative management
Consider percutaneous drainage
Reoperation
Fig. 3. Therapeutic strategy in the case of proven fistula.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 6
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
6.1.3. Complicated Roux-en-Y Gastric Bypass The surgical strategy for RYGB leaks is based on revision of the anastomosis [25]. This second-look can be combined with primary or delayed total gastrectomy and reconstruction by oesophagojejunal anastomosis [32]. Some highly experienced teams have successfully tried nonoperative management using closed suction drains in the region of the gastrojejunal anastomosis in patients with stable haemodynamics and in the absence of sepsis [25]. In cases of substantial postoperative haemorrhage, abdominal re-exploration must be performed to clean the peritoneum, and identify and control the source of bleeding. The situation is much more challenging in the absence of an obvious source of bleeding. 6.2. Medical management Medical management is based on supportive ICU care, adequate nutritional support, appropriate anti-infective therapy and management of associated complications. Only limited data are available concerning the therapeutic management of morbidly obese patients in the ICU setting. Overall clinical management does not differ from that of other BS cases or other complicated abdominal surgery cases. Interestingly, most strategies applied as part of supportive ICU care, such as mechanical ventilation, prevention of thromboembolic complications, type and placement of monitoring devices or pharmacokinetic issues, are based on extrapolations from medical management in non-obese cases. Curative anticoagulation using low molecular weight heparin is usually adjusted to patient body weight, but data pertaining to the safety of weight-based doses is limited [39]. A strong relationship also exists between weight and unfractionated heparin dosing requirements. Monitoring of therapeutic effect is required and doses can be adjusted according to response [39]. Nutritional support is poorly investigated in these patients. Most authors report placement of a feeding tube into the gastric remnant or at the jejunum to continue enteral nutrition below the anastomotic leak. Guidelines have been recently published for the nutrition support of adult patients with morbid obesity that can be implemented in the first 48 hours following ICU admission [40]. 6.3. Anti-infective therapy As usually reported in septic complications, adequate antiinfective therapy is the cornerstone of management combined with effective surgical source control. High concentrations of aerobes, anaerobes and fungi have been observed in the saprophytic stomach flora of ‘‘healthy’’ patients undergoing bypass for morbid obesity. Few studies have addressed the microbiological issues raised by septic complicated BS patients. We recently compared two groups of patients undergoing reoperation for septic complicated abdominal surgery: 49 patients previously treated by BS and 134 patients previously treated by conventional surgery. Compared to patients with postoperative peritonitis following non-bariatric surgery, BS patients had a 37% higher rate of Gram-positive cocci, a 33% lower rate of Gram-negative bacilli, and a 50% lower rate of anaerobes and multidrug-resistant strains [32]. As previously reported elsewhere, the extensive use of antibiotics between initial surgery and the diagnosis of septic surgical complication contributes to the emergence of MDR strains and fungi [41,42]. High rates of Candida have also been reported, similar to those observed in conventional surgery [32]. These high rates of Candida were recently confirmed in a cohort of mixed ICU and non-ICU complicated BS cases [21]. Based on our microbiological results, the empirical antibiotic protocols selected for these complicated BS cases comprise broadspectrum agents (usually carbapenems or ureidopenicillins/
betalactamase inhibitors combined with aminoglycosides and glycopeptides) and antifungal agents (echinocandins or triazoles) to ensure a high adequacy rate, targeting all microorganisms in more than 80% of cases [32]. Patients with complicated RYGB are considered to present small bowel perforation rather than upper gastrointestinal perforation [32]. De-escalation is performed in every case, whenever possible, and as soon as microbiological identification and susceptibility testing are available. In our population, step-down therapy was performed in more than two-thirds of cases, achieving good results with low-spectrum agents such as amoxicillin/clavulanic acid [32]. However, this policy can only be achieved with systematic collection of intraoperative microbiological samples. Only limited pharmacological data are available in morbidly obese patients, especially concerning the use of anti-infective agents in the ICU setting. The appropriate antibiotic and antifungal doses in these specific cases have not been clearly defined and are largely based on extrapolations from non-obese patients or plasma assays when available. However, prescribing physicians should always remember that diffusion of antibiotics in the peritoneal space cannot be easily predicted and is impossible to monitor. These patients are at risk of both low concentrations and overdosing, as recently reported in a study of serum beta-lactam concentrations in obese patients, in whom insufficient serum concentrations were observed in 32% of cases, while overdose concentrations were observed in 25% of cases [43].
7. Prognosis of postoperative complications Only limited data are available concerning the incidence and outcome of specific complications. In the most recent metaanalysis, based on nearly 162 000 patients, the estimated postoperative mortality ( day 30) was 0.08% in randomized controlled trials (RCT) and 0.22% in observational studies [3]. The estimated mortality rates in RCTs was 0.11% for adjustable gastric banding, 0.50% for sleeve gastrectomy and 0.08% for gastric bypass, while these estimates in observational studies were 0.07%, 0.29% and 0.38%, respectively [3]. In a meta-analysis of more than 3000 BPD/DS procedures, Buchwald et al. reported the highest early ( 30 postoperative days) mortality rates of all BS procedures (1.1%) compared to 0.5% for gastric bypass and 0.1% for purely restrictive procedures [2]. These authors confirmed this trend in a subsequent publication comparing open and laparoscopic procedures for gastric banding (mortality rate of 0.3% and 0.1%, respectively), gastroplasty (0.3% and 0.2%), gastric bypass (0.5% and 0.2%) and BPD/DS (0.9% and 0.7%) [44]. In a large database of more than 44 000 patients, Khan et al. [45] identified the following independent risk factors for mortality after BS procedures: age > 45 years, male gender, BMI 50 kg/m2, laparotomy, loss of autonomy before surgery, coronary angioplasty, dyspnoea, preoperative intentional weight loss > 10% and bleeding disorders. Underlying diseases play a key role in these complications. In a series of 3073 RYGB, predictive factors for mortality were male gender, age, weight, and obstructive sleep apnoea [46]. However, anastomotic leak was one of the strongest independent risk factors for death, and leak-associated mortality was estimated to be higher than 16%. In this paper, Fernandez et al. [46] highlighted the importance of an experienced surgeon in these difficult cases and advised surgeons against operating on such patients early in their learning curve. From the intensivist’s point of view, the general outcome of morbidly obese patients admitted to the ICU is difficult to assess. In a meta-analysis comprising a total of 62 045 critically ill subjects, obesity was significantly associated with prolonged duration of mechanical ventilation and ICU length of stay, but not with excess
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx Table 3 Mortality rates according to the anatomical site of the leak following Roux-en-Y gastric bypass (RYGB).
Hamilton et al. [10] Lee et al. [33] Marshall et al. [11]
Leaks, n (%)
Gastric pouch gastrojejunal anastomosis, n (%)
Excluded stomach, n (%)
Jejuno-jejunal anastomosis, n (%)
9 150 21
0 9/100 (9) 0
1/2 (50) 0 0
0 6/15 (40) 2/4 (50)
7
medical environment, expertise in the management of these difficult cases is only at the early stages. Collaborative and multidisciplinary approaches are mandatory to further our understanding of the pathophysiological mechanisms driving the clinical and biological responses to aggression, and finally to improve the quality of care of these high-risk patients.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Table 4 Outcome of anastomotic leaks following Roux-en-Y gastric bypass (RYGB) among cases treated conservatively or reoperated. Leaks, n (%)
Ballesta et al. [8] Gonzalez et al. [24] Marshall et al. [11]
59 63 21
Conservative treatment
Reoperation
n (%)
Death rate, n (%)
n (%)
0 0 0
23 40 11
36 23 10
Funding: The author and co-authors have no financial interests to disclose.
References Death rate, n (%) 5 (22) 4 (10) 2 (18)
mortality [47]. In a recent analysis gathering data from more than 119 000 obese patients requiring mechanical ventilation, Kumar et al. [48] observed an increased risk of death among morbidly obese people relative to non-obese people when stratifying by severity of disease. An increased (33%) risk of death has been reported among surgical patients with prolonged ICU stays [49]. Complicated BS cases have a poorer outcome. The mortality rate associated with anastomotic leaks following RYGB range between 6 and 22% [12,50] and up to 50% in some series, depending of the anatomic location of the fistula [10,11,33] (Table 3). Patients undergoing reoperation had high mortality rates of up to 18 to 22% [8,11], and even 40% in some specific subpopulations [33], while no death was reported among cases treated conservatively (Table 4). However, these results must be interpreted cautiously, as conservatively treated cases may have had less severe complications than reoperated patients. In our experience with patients admitted for postoperative peritonitis, despite their young age, these BS patients have a poor prognosis [31]. Complicated BS patients have similar reoperation rates and mortality rates to those of patients with postoperative peritonitis due to other causes [32]. On the other hand, BS did not appear to be a significant risk factor for mortality in the multivariate analysis in this study, suggesting that BS patients had a similar risk to non-BS patients. Delayed surgery has been frequently reported to be associated with an increased risk of organ dysfunction and mortality. The importance of the urgent recognition of complications was recently illustrated by Gonzales et al. [24], who reported a mortality rate as low as 6% for anastomotic leaks after RYGB. In this study, the vast majority of patients presented no signs of organ failure and leaks were detected early (< 48 h) after the index surgery before the onset of any clinical symptoms [24].
8. Conclusion Morbidly obese patients represent a new paradigm for physicians in developed countries. Bariatric surgery has become an increasingly popular approach over recent years. Intensivists are already involved in the management of BS complications, which raises new and challenging issues. While technical and logistic problems (beds, tables, CT-scans and radiology tables, etc) will probably be resolved with progressive improvement of the
[1] DeMaria EJ. Bariatric surgery for morbid obesity. N Engl J Med 2007;356:2176– 83. [2] Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach K, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292: 1724–37. [3] Chang SH, Stoll CR, Song J, Varela JE, Eagon CJ, Colditz GA. The effectiveness and risks of bariatric surgery: an updated systematic review and meta-analysis, 2003–2012. JAMA Surg 2014;149:275–87. [4] Arapis K, Chosidow D, Lehmann M, Bado A, Polanco M, Kamoun-Zana S, et al. Long-term results of adjustable gastric banding in a cohort of 186 super-obese patients with a BMI 50 kg/m2. J Visceral Surg 2012;149:e143–52. [5] Benotti PN, Wood GC, Rodriguez H, Carnevale N, Liriano E. Perioperative outcomes and risk factors in gastric surgery for morbid obesity: a 9-year experience. Surgery 2006;139:340–6. [6] Chen CC, Huang MT, Wei PL, Liang HH, Chen SC, Wu CH, et al. Severe peritonitis due to Streptococcus viridans following adjustable gastric banding. Obes Surg 2010;20:1603–5. [7] van den Broek RJ, Buise MP, van Dielen FM, Bindels AJ, van Zundert AA, Smulders JF. Characteristics and outcome of patients admitted to the ICU following bariatric surgery. Obes Surg 2009;19:560–4. [8] Ballesta C, Berindoague R, Cabrera M, Palau M, Gonzales M. Management of anastomotic leaks after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2008;18:623–30. [9] Thodiyil PA, Yenumula P, Rogula T, Gorecki P, Fahoum B, Gourash W, et al. Selective nonoperative management of leaks after gastric bypass: lessons learned from 2675 consecutive patients. Ann Surg 2008;248:782–92. [10] Hamilton EC, Sims TL, Hamilton TT, Mullican MA, Jones DB, Provost DA. Clinical predictors of leak after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Surg Endosc 2003;17:679–84. [11] Marshall JS, Srivastava A, Gupta SK, Rossi TR, DeBord JR. Roux-en-Y gastric bypass leak complications. Arch Surg 2003;138:520–3 [discussion 3–4]. [12] Fernandez Jr AZ, Demaria EJ, Tichansky DS, Kellum JM, Wolfe LG, Meador J, et al. Multivariate analysis of risk factors for death following gastric bypass for treatment of morbid obesity. Ann Surg 2004;239:698–703. [13] Gonzalez R, Nelson LG, Gallagher SF, Murr MM. Anastomotic leaks after laparoscopic gastric bypass. Obes Surg 2004;14:1299–307. [14] Goldfeder LB, Ren CJ, Gill JR. Fatal complications of bariatric surgery. Obes Surg 2006;16:1050–6. [15] Msika S. Surgery for morbid obesity: 2. Complications. Results of a technologic evaluation by the ANAES. J Chir 2003;140:4–21. [16] Cherian PT, Goussous G, Ashori F, Sigurdsson A. Band erosion after laparoscopic gastric banding: a retrospective analysis of 865 patients over 5 years. Surg Endosc 2010;24:2031–8. [17] Owers C, Ackroyd R. A study examining the complications associated with gastric banding. Obes Surg 2013;23:56–9. [18] Pequignot A, Fuks D, Verhaeghe P, Dhahri A, Brehant O, Bartoli E, et al. Is there a place for pigtail drains in the management of gastric leaks after laparoscopic sleeve gastrectomy? Obes Surg 2012;22:712–20. [19] Parikh M, Issa R, McCrillis A, Saunders JK, Ude-Welcome A, Gagner M. 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. [20] Aurora AR, Khaitan L, Saber AA. Sleeve gastrectomy and the risk of leak: a systematic analysis of 4888 patients. Surg Endosc 2012;26:1509–15. [21] Rebibo L, Dupont H, Levrard M, Cosse C, Dhahri A, Verhaeghe P, et al. Gastric fistula after laparoscopic sleeve gastrectomy: don’t forget to treat for Candida. Obes Surg 2013;23:2106–8. [22] Shi X, Karmali S, Sharma AM, Birch DW. A review of laparoscopic sleeve gastrectomy for morbid obesity. Obes Surg 2010;20:1171–7. [23] Carucci LR, Turner MA, Conklin RC, DeMaria EJ, Kellum JM, Sugerman HJ. Rouxen-Y gastric bypass surgery for morbid obesity: evaluation of postoperative extraluminal leaks with upper gastrointestinal series. Radiology 2006;238: 119–27.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 8
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
[24] Gonzalez R, Sarr MG, Smith CD, Baghai M, Kendrick M, Szomstein S, et al. Diagnosis and contemporary management of anastomotic leaks after gastric bypass for obesity. J Am Coll Surg 2007;204:47–55. [25] Griffith PS, Birch DW, Sharma AM, Karmali S. Managing complications associated with laparoscopic Roux-en-Y gastric bypass for morbid obesity. Can J Surg 2012;55:329–36. [26] Heneghan HM, Meron-Eldar S, Yenumula P, Rogula T, Brethauer SA, Schauer PR. Incidence and management of bleeding complications after gastric bypass surgery in the morbidly obese. Surg Obes Relat Dis 2012;8:729–35. [27] Topart P, Becouarn G, Ritz P. Comparative early outcomes of three laparoscopic bariatric procedures: sleeve gastrectomy, Roux-en-Y gastric bypass, and biliopancreatic diversion with duodenal switch. Surg Obes Relat Dis 2012;8:250–4. [28] Morino M, Toppino M, Forestieri P, Angrisani L, Allaix ME, Scopinaro N. Mortality after bariatric surgery: analysis of 13,871 morbidly obese patients from a national registry. Ann Surg 2007;246:1002–9. [29] Podnos YD, Jimenez JC, Wilson SE, Stevens CM, Nguyen NT. Complications after laparoscopic gastric bypass: a review of 3464 cases. Arch Surg 2003;138:957–61. [30] Chakravartty S, Sarma DR, Patel AG. Rhabdomyolysis in bariatric surgery: a systematic review. Obes Surg 2013;23:1333–40. [31] Kermarrec N, Marmuse JP, Faivre J, Lasocki S, Mognol P, Chosidow D, et al. High mortality rate for patients requiring intensive care after surgical revision following bariatric surgery. Obes Surg 2008;18:171–8. [32] Montravers P, Guglielminotti J, Zappella N, Desmard M, Muller C, Fournier P, et al. Clinical features and outcome of postoperative peritonitis following bariatric surgery. Obes Surg 2013;23:1536–44. [33] Lee S, Carmody B, Wolfe L, Demaria E, Kellum JM, Sugerman H, et al. Effect of location and speed of diagnosis on anastomotic leak outcomes in 3828 gastric bypass cases. J Gastrointest Surg 2007;11:708–13. [34] Merkle EM, Hallowell PT, Crouse C, Nakamoto DA, Stellato TA. Roux-en-Y gastric bypass for clinically severe obesity: normal appearance and spectrum of complications at imaging. Radiology 2005;234:674–83. [35] Madan AK, Stoecklein HH, Ternovits CA, Tichansky DS, Phillips JC. Predictive value of upper gastrointestinal studies versus clinical signs for gastrointestinal leaks after laparoscopic gastric bypass. Surg Endosc 2007;21:194–6. [36] Lyass S, Khalili TM, Cunneen S, Fujita F, Otsuka K, Chopra R, et al. Radiological studies after laparoscopic Roux-en-Y gastric bypass: routine or selective? Am Surg 2004;70:918–21. [37] Chevallier JM, Zinzindohoue F, Douard R, Blanche JP, Berta JL, Altman JJ, et al. Complications after laparoscopic adjustable gastric banding for morbid obesity: experience with 1000 patients over 7 years. Obes Surg 2004;14:407–14.
[38] Moszkowicz D, Arienzo R, Khettab I, Rahmi G, Zinzindohoue F, Berger A, et al. Sleeve gastrectomy severe complications: is it always a reasonable surgical option? Obes Surg 2013;23:676–86. [39] Patel JP, Roberts LN, Arya R. Anticoagulating obese patients in the modern era. Br J Haematol 2011;155:137–49. [40] Choban P, Dickerson R, Malone A, Worthington P, Compher C. American Society for Parenteral and Enteral Nutrition, A.S.P.E.N. Clinical guidelines: nutrition support of hospitalized adult patients with obesity. JPEN J Parenter Enteral Nutr 2013;37:714–44. [41] Augustin P, Kermarrec N, Muller-Serieys C, Lasocki S, Chosidow D, Marmuse JP, et al. Risk factors for multidrug resistant bacteria and optimization of empirical antibiotic therapy in postoperative peritonitis. Crit Care 2010; 14:R20. [42] Seguin P, Fedun Y, Laviolle B, Nesseler N, Donnio PY, Malledant Y. Risk factors for multidrug-resistant bacteria in patients with post-operative peritonitis requiring intensive care. J Antimicrob Chemother 2010;65:342–6. [43] Hites M, Taccone FS, Wolff F, Cotton F, Beumier M, De Backer D, et al. Casecontrol study of drug monitoring of beta-lactams in obese critically ill patients. Antimicrob Agents Chemother 2013;57:708–15. [44] Buchwald H, Estok R, Fahrbach K, Banel D, Sledge I. Trends in mortality in bariatric surgery: a systematic review and meta-analysis. Surgery 2007;142: 621–35. [45] Khan MA, Grinberg R, Johnson S, Afthinos JN, Gibbs KE. Perioperative risk factors for 30-day mortality after bariatric surgery: is functional status important? Surg Endosc 2013;27:1772–7. [46] Fernandez Jr AZ, DeMaria EJ, Tichansky DS, Kellum JM, Wolfe LG, Meador J, et al. Experience with over 3000 open and laparoscopic bariatric procedures: multivariate analysis of factors related to leak and resultant mortality. Surg Endosc 2004;18:193–7. [47] Akinnusi ME, Pineda LA, El Solh AA. Effect of obesity on intensive care morbidity and mortality: a meta-analysis. Crit Care Med 2008;36:151–8. [48] Kumar G, Majumdar T, Jacobs ER, Danesh V, Dagar G, Deshmukh A, et al. Outcomes of morbidly obese patients receiving invasive mechanical ventilation: a nationwide analysis. Chest 2013;144:48–54. [49] Nasraway Jr SA, Albert M, Donnelly AM, Ruthazer R, Shikora SA, Saltzman E. Morbid obesity is an independent determinant of death among surgical critically ill patients. Crit Care Med 2006;34:964–70. [50] Csendes A, Burdiles P, Burgos AM, Maluenda F, Diaz JC. Conservative management of anastomotic leaks after 557 open gastric bypasses. Obes Surg 2005;15: 1252–6.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
ACCPM-6; No. of Pages 8 Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
Review
Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery Philippe Montravers *, Pascal Augustin, Nathalie Zappella, Guillaume Dufour, Konstantinos Arapis, Denis Chosidow, Pierre Fournier, Lara Ribeiro-Parienti, Jean-Pierre Marmuse, Mathieu Desmard AP–HP, CHU Bichat–Claude-Bernard, De´partement d’Anesthe´sie Re´animation, 46, rue Henri-Huchard, 75018 Paris, France
A R T I C L E I N F O
A B S T R A C T
Article history: Available online xxx
Perioperative complications following bariatric surgery (BS) have been poorly analysed and their management is not clearly assessed. The associated frequency of ICU admission is difficult to estimate. Among surgical complications, digestive perforations are the most frequent. The most common postoperative complications of sleeve gastrectomy are fistulas, but bleeding on the stapling line is also commonly reported. Complication rates are higher after Roux-en-Y gastric bypass, mainly due to anastomotic leaks. Medical complications are mainly thromboembolic or respiratory complications. All these surgical and medical complications are not easily detected; clinical signs can be atypical or insidious, often resulting in delayed management. Respiratory signs can be predominant and lead erroneously to pulmonary or thromboembolic diseases. Diagnostic criteria are based on minor clinical signs, tachycardia being probably the most frequent one. Lately, complications are revealed by haemodynamic instability, respiratory failure or renal dysfunction and radiographic findings. Management decision according to these abnormal signs is based on a combined multidisciplanary approach including surgical and/or endoscopic procedures and medical care, depending on the nature and severity of the surgical complication. Medical management is based on supportive ICU care of organ dysfunctions, curative anticoagulation if required, nutritional support, and appropriate anti-infective therapy. Pharmacological data are limited in morbidly obese patients and the appropriate doses are debated, especially for anti-infective agents. Complicated BS cases have a poor outcome, probably largely related to delayed diagnosis and reoperation. ß 2015 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Keywords: Bariatric surgery Morbid obesity Pulmonary embolism Postoperative pneumonia Postoperative peritonitis Sepsis Multidrug-resistant bacteria Antibiotic adequacy
1. Introduction Bariatric surgery (BS) is very commonly used for the treatment of morbid obesity [1]. Most studies have demonstrated these procedures to be safe and well tolerated in the vast majority of cases, with decreased overall mortality and morbidity rates [2,3]. As usual, however, the benefits of a new treatment must be weighed against the possible associated risks. BS has been reported to be associated with a growing number of perioperative complications and postoperative or short-term adverse outcomes over recent years. From the anaesthesiologist’s perspective, these morbidly obese patients represent a new population that was extremely rare thirty
* Corresponding author. Tel.: +33 1 40 25 83 55; fax: +33 1 40 25 63 09. E-mail address: [email protected] (P. Montravers).
years ago. While the intraoperative period has been well described and is uneventful in most cases, challenging issues may occur during the postoperative course that deserve specific attention and require intensive care unit (ICU) admission. This article is designed to review the postoperative surgical and medical complications reported after the most commonly performed BS procedures. We describe the clinical conditions and management of these high-risk patients according to the specificities of the procedure performed. 2. Surgical procedures Many techniques have been proposed for the surgical management of morbid obesity. These operations can be divided into restrictive and malabsorptive procedures [1] that are designed to reduce caloric intake by dramatically modifying the upper mesocolic gastrointestinal tract.
http://dx.doi.org/10.1016/j.accpm.2014.06.002 2352-5568/ß 2015 Socie´te´ franc¸aise d’anesthe´sie et de re´animation (Sfar). Published by Elsevier Masson SAS. All rights reserved.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 2
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
2.1. Restrictive procedures Several surgical techniques that decrease food intake by creating a small gastric pouch and result in rapid satiety after food intake have been described. Adjustable gastric banding was the first technique to become widely popular. An adjustable silicone ring is placed around the stomach, to create a small gastric pouch with a narrow outlet to induce delayed emptying. The degree of gastric stricture and consequent food restriction is adjusted by means of saline injections into a subcutaneous port. The procedure is usually well tolerated and is now frequently proposed as day-only surgery in patients with no concomitant disease. Vertical restrictive (sleeve) gastrectomy is now frequently proposed as a first-line treatment in replacement of adjustable gastric banding due to its better efficacy in terms of weight loss [4]. The procedure removes two-thirds of the gastric volume and leaves a gastric tube by stapling the entire length of the greater curvature. 2.2. Malabsorptive operations Malabsorptive procedures bypass various portions of the small intestine and reduce jejunal absorption. Proximal Roux-en-Y gastric bypass (RYGB) combines restrictive and malabsorptive procedures and is often considered to be the gold standard because of its excellent efficacy and good safety. RYGB comprises stapling the stomach to create a small upper gastric pouch and division of the small intestine at the mid-jejunum level, while the distal portion (the alimentary, or Roux limb) is anastomosed to the gastric pouch. Intestinal (jejunal) absorption is progressively decreased with increasing length of the Roux limb. Two anastomoses (gastrointestinal and jejuno-intestinal anastomoses) are performed. Other malabsorptive procedures have been proposed, but are often responsible for severe or extreme malabsorption. This is the case of biliopancreatic diversion and duodenal switch (BPD/DS) procedures, sometimes proposed in super-obese patients (BMI > 50 kg/m2) or after failure of previous BS. This procedure comprises sleeve gastrectomy and a short duodeno-ileal circuit with duodeno-ileal and ileo-ileal anastomoses. 3. Epidemiology of surgical complications Descriptions of postoperative surgical complications are based on reports of general or surgical site complications. Very few data are available concerning the incidence and clinical characteristics of complications requiring ICU management [5–7]. Although most surgical complications occur during the procedure, the time to diagnosis usually peaks during the early postoperative period, in the first days or weeks following surgery [8–11]. Surgical complications can be divided into two entities: septic complications, mainly fistula and anastomotic leak, and nonseptic complications, mostly due to haemorrhage. The vast majority of these complications are treated in surgical units and never require ICU management. Anastomotic leaks are a very common complication and are more common than pulmonary embolism in some series [12,13]. In a series of 107 autopsies following BS, anastomotic leak was the leading cause of death among the 97 deaths directly related to surgical complications, observed in 36% of cases, while only 13% of deaths were due to pulmonary embolism [14]. 3.1. Complications of restrictive procedures 3.1.1. Complications of adjustable gastric banding The most recent meta-analysis, comprising more than 850 patients included in randomised control trials (RCT) and
36 000 patients from observational studies (OS), estimated the complication rate of adjustable gastric banding to be 13% in RCTs, while a lower rate was observed in OS (7.8%) [3] (Fig. 1A). Although the procedure appears to be safe, many early complications have been described which can be life-threatening if the diagnosis is delayed. Gastric or oesophageal perforation is a rare complication that is sometimes diagnosed intraoperatively and constitutes a contraindication to foreign body insertion. In an analysis conducted by the French ANAES agency, the incidence of gastric perforation was 0.3% (15/5237 patients) [15]. Early gastric band erosion and perforation of the gastric pouch have been reported during the first postoperative weeks [16]. Erosion is usually related to a technical problem and unrecognized intraoperative gastric perforation. Early postoperative infection or gastric wall ischaemia secondary to a tight band could also account for these erosions. An initially unrecognized perforation can also be revealed by early ring or port suppuration. Ring or port infection, observed in 1.5% of cases, may lead to suppuration at the port site [17]. Interestingly, port infection is not systematically associated with ring infection, and the ring can be left in place in the case of isolated port infection. Ring malposition and band slippage are early but rare postoperative complications [17].
3.1.2. Complications of sleeve gastrectomy The sleeve gastrectomy complication rate has been estimated to be 13% in RCTs and 8.9% in OSs [3] (Fig. 1B). The most common complications during the postoperative course of sleeve gastrectomy are fistulas, reported in 0 to 20% of cases after gastric banding removal [18]. Parikh et al. [19] reported a fistula rate of 2.2% in a meta-analysis of almost 10 000 sleeve gastrectomies. These leaks usually occur in the upper part of the gastric tube, in the proximal third of the stomach, close to the angle of His, in up to 89% of cases [20,21]. Bleeding is the most frequent nonseptic complication of sleeve gastrectomy. The most common source of bleeding is the stapling line, observed in approximately 3.5% of cases [22]. 3.2. Complications of malabsorptive procedures Complication rates are higher after malabsorptive procedures than after restrictive surgery. A meta-analysis by Chang et al. [3], based on more than 600 patients included in RCTs and 71 000 patients from OSs, demonstrated an estimated complication rate of 21% in RCTs and 12% in OSs. Anastomotic leaks are the most common complications of RYGB, with a reported incidence ranging from 0 to 8% [12,13]. A fistula can arise at each segment of the RYGB. The fistula rates reported in the literature are presented in Fig. 1C and Table 1 [8–11,23,24]. The bleeding rate after RYGB is significantly higher than after other BS procedures and is reported in 1 to 4.4% of cases [25,26], requiring a complex diagnostic process to identify the source. The site of bleeding may be situated in the abdominal cavity from one of the staple lines. Another source of bleeding is intraluminal bleeding (gastric or intestinal), which may be difficult to detect [25]. Complications following BPD/DS comprise all of the complications reported following sleeve gastrectomy and gastric bypass. Single-centre studies have reported discordant results following BPD/DS, with either similar complication rates to those observed after other procedures or higher overall complication rates with a trend towards higher bleeding rates [2,27].
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
3
Fig. 1. Sites and frequency of the most frequent surgical complications observed during adjustable gastric banding (A), vertical restrictive sleeve gastrectomy (B) and proximal Roux-en-Y gastric bypass (C).
4. Epidemiology of medical complications The timing of postoperative complications has not been specifically assessed in bariatric surgery. Similarly to other procedures, a majority of medical complications could occur before patients were discharged. In a cohort of 13 871 patients submitted to various bariatric procedures, Morino et al. [28] reported that medical complications caused death within a mean delay of 17 days. Thromboembolic complications accounted for 13 to 38.2% of fatalities [14,28]. In a literature analysis gathering more than 20 publications and focusing on postoperative complications, 58.6% of deaths were related to pulmonary embolism [15]. Non-fatal cases of pulmonary embolism following gastric bypass are reported in various proportions between 0 and up to 4% of cases [15,29]. Recent studies in RYGBP reported incidences ranging between 1.2% and 8%
of cases [5,9,24] and up to 12.5% in a study pooling pulmonary embolism and deep venous thromboses [23]. Respiratory failure is the second respiratory cause of death reported in 11.8% of the fatalities [28]. In non-fatal cases, the incidence of respiratory failure varies largely, depending on the definition, from 10 [24] to 29% of cases [23]. Pleural effusions are noticed in 5 [24] to 23% of patients [23], while atelactasias are reported between 6 [8] and 92% of cases [23]. Pneumonias are reported in various proportions, between 0.1% in laparoscopic bypass (0.33% in open bypass) [29] to 2–3% [9,24] and even 22% for Ballesta et al. [8]. Cardiovascular failure and cerebral ischemia are the two other major medical causes of death reported by Morino et al. [28] in 17.6 and 2.9% of cases, respectively. However, non-fatal cardiovascular complications are rarely reported [23,24].
Table 1 Site of the leak following Roux-en-Y gastric bypass (RYGB).
Ballesta et al. [8], n (%) Carucci et al. [23], n (%) Gonzalez et al. [24], n (%) Hamilton et al. [10], n (%) Marshall et al. [11], n (%) Thodiyil et a. [9], n (%)
Number of procedures
Leaks
Gastric pouch
Gastrojejunal anastomosis
Excluded stomach
Jejuno-jejunal anastomosis
Other sites
1200 904 3018 246 400 2675
59 48 63 9 21 46
6 5 6 1 – 4
40 37 33 3 13 37
2 15 17 2 2 2
3 1 9 2 4 1
4 7 – – 2 2
(5) (5) (3) (4) (5) (2)
(10) (10) (10) (11) (9)
(68) (77) (52) (33) (62) (80)
(3) (31) (27) (22) (10) (4)
(5) (2) (14) (22) (21) (2)
(7) (15)
(10) (4)
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
4
Rhabdomyolysis is reported in more than 20 publications over the last decade and seems to be more frequent in male patients with high BMI (52 kg/m2) who underwent prolonged bariatric surgery [30]. Acute renal failure was observed in 14% of these cases. Renal dysfunction can also be associated with sepsis as reported in postoperative cases in 35% [31] to 58% of cases [32].
or bowel obstruction (n = 2) [31]. These symptoms usually deteriorate over several days leading to organ dysfunction and/ or ICU admission. Interestingly, the frequency of organ failure in BS patients admitted to the ICU for postoperative infection did not differ from those reported in patients with other common forms of postoperative peritonitis [32].
5. Diagnosis of surgical complications
5.2. Radiological assessment
5.1. Clinical presentation
Radiological imaging is strongly recommended in the diagnostic work-up of postoperative complications [34]. However, the diagnosis can be difficult due to the lack of specificity of imaging studies, especially at the early postoperative phase. No consensus has been reached concerning the optimal imaging technique (CTscan or upper gastrointestinal [UGI] studies) to be used in these patients. In a cohort of 300 patients who underwent RYGB, Madan et al. [35] used routine UGI studies on the first postoperative day, and reported positive and negative predictive values for leak identification of 67% and 99%, respectively. Moreover, these authors observed that routine UGI studies were more predictive of early leak diagnosis than clinical signs. Few studies are available assessing the capacity of CT to detect surgical complications after BS, and more specifically anastomotic leaks. Lyass et al. [36] proposed the use of CT rather than UGI studies because of its capacity to detect complications other than anastomotic leaks. Interestingly, no study has ever assessed the specificity and sensitivity of CT in the diagnosis of complicated BS, possibly because of the limitations on the use of imaging procedures in this population: patients exceeding the weight limit of CT-scan systems (above 150 to 200 kg of body weight), limited use of contrast agents in these patients who present a high rate of renal impairment, and/or the radiologist’s limited experience. These technical limitations led some authors to reoperate without radiological proof of complications, as in 2% (1/59) of cases of complicated RYGB for Ballesta et al. [8], 10% (2/21) of cases for Marshall et al. [11], or 18% (27/150) of cases for Lee et al. [33]. Chest radiograph abnormalities are not uncommon in complicated BS patients admitted to the ICU. Kermarrec et al. [31] observed basal pulmonary atelectasis and/or left-sided pleural effusion in 37% and 44% of complicated BS patients, respectively. Upper GI investigations were performed in only 44% of these patients and were reported as normal in one half of them. Similarly, abdomen CT-scan was performed in only one half of cases in a median of 8 days after initial surgery. Imaging studies were reported to be normal in 4 patients with generalized (n = 2) or localized (n = 2) intra-abdominal sepsis. Abnormal CT-scans revealed subphrenic collections (n = 7), supramesocolic collections (n = 2), and small bowel obstruction (n = 1) [31]. In summary, the decision-making process is based on abnormal clinical signs, especially signs of haemodynamic instability,
Obvious surgical complications may be diagnosed intraoperatively. Unfortunately, in the vast majority of cases, surgical complications are not easily detected, consequently often resulting in delayed management. For clinicians not used to the management of BS patients, it is important to remember that these morbidly obese patients constitute an atypical population that differs from the general population of abdominal surgery patients. In a recent publication, we compared the clinical characteristics and management of 134 non-obese patients admitted to our ICU for ‘‘common’’ postoperative peritonitis and a group of 49 patients who developed postoperative peritonitis following BS [32]. The BS patients were younger (45 10 years versus 63 16 years; P < 0.0001), had low cancer rates and lower rates of underlying fatal disease (39% vs. 64%; P = 0.002). However, the time to diagnosis of surgical complications was similar in the two groups of patients [32]. Identification of these complications before the onset of organ dysfunction is frequently based on minor clinical signs, making the diagnosis particularly difficult [10,12] (Table 2). Tachycardia is probably the most frequent sign, largely reported in the literature [8–11,23,24,33]. Polypnea or respiratory distress are other sensitive indicators of gastrointestinal leak [9,10,23]. It is noteworthy that the absence of clinical signs at the time of diagnosis of peritonitis is reported in up to 49% of non-ICU patients [8,9,24]. The reasons for the limited value of clinical signs have not been elucidated, but could be related to the large mass of subcutaneous tissue, the subphrenic site of intraperitoneal sepsis, or impairment of the inflammatory response. A similar low incidence of clinical symptoms was also reported in ICU BS cases [31]. The most common symptoms, fever, dyspnoea and tachycardia, were reported in up to 74, 98 and 100% of cases, respectively [31]. Local and abdominal signs were rarely reported; tenderness, pus, and ileus were observed in only 30%, 33% and 37% of cases, respectively [31]. In many cases, the predominance of respiratory signs may be more suggestive of pleural or pulmonary diseases. In such circumstances, an incorrect initial diagnosis was reported in more than 50% (15/27) of patients. The most frequent erroneous hypotheses proposed by attending teams were pneumonia (n = 7), pulmonary embolism (n = 4), wound abscess (n = 2),
Table 2 Clinical presentation at the time of diagnosis of anastomotic leaks in four cohorts of patients who underwent Roux-en-Y gastric bypass (RYGB).
Number of patients, n Absence of clinical signs, n (%) Presence of abdominal pain, n (%) Presence of left shoulder pain, n (%) Presence of nausea/vomiting, n (%) Abnormal drainage, n (%) Fever > 38 8C, n (%) White blood cell count > 12,000/mm3, n (%) Tachycardia, n (%) Polypnoea, n (%) Hypotension, n (%) Oliguria, n (%) a
Ballesta et al. [8]
Carucci et al. [23]
Gonzalez et al. [24]
Thodiyil et al. [9]
59 29 (49) 30 (51) – 11 (19) 29 (49) 29 (49) – 13 (22) – 7 (12) 8 (13)
48 – 13 (27) 9 (19) 39 (81) 3 (6) 44 (92)a 44 (92)a 44 (92)a 8 (17) 7 (15) –
63 5 (8) 34 (54) 9 (14) 11 (17) 15 (24) 40 (63) 47 (75) 45 (72) – 11 (17) 13 (21)
46 15 (33) 11 (24) 4 (9) 3 (7) 13 (28) 13 (28) 31 (67) 8 (17) 4 (9) 1 (2) 2 (4)
Elevated white blood cell counts, fever and/or tachycardia in 92% of patients.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
5
Abnormal clinical signs
Clinical stability
Clinical instability
Reoperation (laparoscopy or laparotomy / depending on clinical severity)
Upper gastrointestinal imaging
Radiographic signs
CT scan if possible Consider reoperation
No abnormal signs
CT scan cannot be performed
Reoperation Laparoscopy
CT scan available
Radiographic signs Consider reoperation
No signs Survey
Fig. 2. Diagnostic strategy in the case of suspected fistula.
respiratory failure or less frequently renal dysfunction and radiographic findings. The medical team may decide to reoperate in the presence of severe clinical signs, even when there is limited proof of surgical complications. In doubtful cases with a reassuring clinical status, the team may prefer watchful waiting. Based on our experience and a review of the literature, we routinely use the decision tree presented in Fig. 2 as a possible diagnostic approach in patients with suspected fistula. 6. Management of complications Management of complicated BS cases is based on a combined approach including surgical and/or endoscopic procedures and medical care, largely depending on the nature and severity of the complication. The severity of the clinical features at the time of diagnosis is the most important factor guiding the treatment strategy. 6.1. Surgical management General surgical management corresponds to the conventional rules of septic surgery including identification of the infectious source, collection of microbiological samples, peritoneal lavage and possibly drainage depending on the intensity and source of contamination, and elimination/control of the source of infection. In the presence of bleeding, surgery is required to identify the source of bleeding and to control blood loss. The treatment strategy is usually based on second-look laparoscopy possibly converted to laparotomy depending on the surgeon’s experience, technical conditions and the patient’s general status. All devices (gastric ring, tube, port, etc.) must be removed. Based on our experience and a review of the literature, we routinely use the decision tree presented in Fig. 3 for the management of all fistulas observed after bariatric surgery. 6.1.1. Complicated gastric banding Gastric or oesophageal perforation following gastric banding must be treated by abundant lavage and efficient drainage of the
peritoneal cavity. The perforation must be sutured together with drainage in contact with the perforation in every case. Infection of the ring or port of gastric banding requires removal of the device. Placement of a new device is deferred until eradication of the infectious source. Band malposition requires urgent reoperation and removal or repositioning of the gastric band [17,37]. 6.1.2. Complicated sleeve gastrectomy Reoperation must include extensive peritoneal toilet, sometimes combined with suture and drainage in contact with the fistula. The outcome is usually uneventful when the fistula is adequately closed. Endoscopic placement of a prosthesis into the gastric leak has been proposed as either first-line therapy or as second-line therapy after failure to control the infection [18]. Radiological drainage techniques using pigtail drains have also been proposed [18]. In the case of recurrent failure of these strategies or in the presence of necrosis of the stomach, total gastrectomy with primary or delayed reconstruction by oesophagojejunal anastomosis has been successfully proposed [32,38].
Proven fistula
Clinical instability - hemodynamic - respiratory - renal
Reoperation
Clinical stability - hemodynamic - respiratory - renal
No collection
Limited collection
Generalized collection
Conservative management
Consider percutaneous drainage
Reoperation
Fig. 3. Therapeutic strategy in the case of proven fistula.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 6
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
6.1.3. Complicated Roux-en-Y Gastric Bypass The surgical strategy for RYGB leaks is based on revision of the anastomosis [25]. This second-look can be combined with primary or delayed total gastrectomy and reconstruction by oesophagojejunal anastomosis [32]. Some highly experienced teams have successfully tried nonoperative management using closed suction drains in the region of the gastrojejunal anastomosis in patients with stable haemodynamics and in the absence of sepsis [25]. In cases of substantial postoperative haemorrhage, abdominal re-exploration must be performed to clean the peritoneum, and identify and control the source of bleeding. The situation is much more challenging in the absence of an obvious source of bleeding. 6.2. Medical management Medical management is based on supportive ICU care, adequate nutritional support, appropriate anti-infective therapy and management of associated complications. Only limited data are available concerning the therapeutic management of morbidly obese patients in the ICU setting. Overall clinical management does not differ from that of other BS cases or other complicated abdominal surgery cases. Interestingly, most strategies applied as part of supportive ICU care, such as mechanical ventilation, prevention of thromboembolic complications, type and placement of monitoring devices or pharmacokinetic issues, are based on extrapolations from medical management in non-obese cases. Curative anticoagulation using low molecular weight heparin is usually adjusted to patient body weight, but data pertaining to the safety of weight-based doses is limited [39]. A strong relationship also exists between weight and unfractionated heparin dosing requirements. Monitoring of therapeutic effect is required and doses can be adjusted according to response [39]. Nutritional support is poorly investigated in these patients. Most authors report placement of a feeding tube into the gastric remnant or at the jejunum to continue enteral nutrition below the anastomotic leak. Guidelines have been recently published for the nutrition support of adult patients with morbid obesity that can be implemented in the first 48 hours following ICU admission [40]. 6.3. Anti-infective therapy As usually reported in septic complications, adequate antiinfective therapy is the cornerstone of management combined with effective surgical source control. High concentrations of aerobes, anaerobes and fungi have been observed in the saprophytic stomach flora of ‘‘healthy’’ patients undergoing bypass for morbid obesity. Few studies have addressed the microbiological issues raised by septic complicated BS patients. We recently compared two groups of patients undergoing reoperation for septic complicated abdominal surgery: 49 patients previously treated by BS and 134 patients previously treated by conventional surgery. Compared to patients with postoperative peritonitis following non-bariatric surgery, BS patients had a 37% higher rate of Gram-positive cocci, a 33% lower rate of Gram-negative bacilli, and a 50% lower rate of anaerobes and multidrug-resistant strains [32]. As previously reported elsewhere, the extensive use of antibiotics between initial surgery and the diagnosis of septic surgical complication contributes to the emergence of MDR strains and fungi [41,42]. High rates of Candida have also been reported, similar to those observed in conventional surgery [32]. These high rates of Candida were recently confirmed in a cohort of mixed ICU and non-ICU complicated BS cases [21]. Based on our microbiological results, the empirical antibiotic protocols selected for these complicated BS cases comprise broadspectrum agents (usually carbapenems or ureidopenicillins/
betalactamase inhibitors combined with aminoglycosides and glycopeptides) and antifungal agents (echinocandins or triazoles) to ensure a high adequacy rate, targeting all microorganisms in more than 80% of cases [32]. Patients with complicated RYGB are considered to present small bowel perforation rather than upper gastrointestinal perforation [32]. De-escalation is performed in every case, whenever possible, and as soon as microbiological identification and susceptibility testing are available. In our population, step-down therapy was performed in more than two-thirds of cases, achieving good results with low-spectrum agents such as amoxicillin/clavulanic acid [32]. However, this policy can only be achieved with systematic collection of intraoperative microbiological samples. Only limited pharmacological data are available in morbidly obese patients, especially concerning the use of anti-infective agents in the ICU setting. The appropriate antibiotic and antifungal doses in these specific cases have not been clearly defined and are largely based on extrapolations from non-obese patients or plasma assays when available. However, prescribing physicians should always remember that diffusion of antibiotics in the peritoneal space cannot be easily predicted and is impossible to monitor. These patients are at risk of both low concentrations and overdosing, as recently reported in a study of serum beta-lactam concentrations in obese patients, in whom insufficient serum concentrations were observed in 32% of cases, while overdose concentrations were observed in 25% of cases [43].
7. Prognosis of postoperative complications Only limited data are available concerning the incidence and outcome of specific complications. In the most recent metaanalysis, based on nearly 162 000 patients, the estimated postoperative mortality ( day 30) was 0.08% in randomized controlled trials (RCT) and 0.22% in observational studies [3]. The estimated mortality rates in RCTs was 0.11% for adjustable gastric banding, 0.50% for sleeve gastrectomy and 0.08% for gastric bypass, while these estimates in observational studies were 0.07%, 0.29% and 0.38%, respectively [3]. In a meta-analysis of more than 3000 BPD/DS procedures, Buchwald et al. reported the highest early ( 30 postoperative days) mortality rates of all BS procedures (1.1%) compared to 0.5% for gastric bypass and 0.1% for purely restrictive procedures [2]. These authors confirmed this trend in a subsequent publication comparing open and laparoscopic procedures for gastric banding (mortality rate of 0.3% and 0.1%, respectively), gastroplasty (0.3% and 0.2%), gastric bypass (0.5% and 0.2%) and BPD/DS (0.9% and 0.7%) [44]. In a large database of more than 44 000 patients, Khan et al. [45] identified the following independent risk factors for mortality after BS procedures: age > 45 years, male gender, BMI 50 kg/m2, laparotomy, loss of autonomy before surgery, coronary angioplasty, dyspnoea, preoperative intentional weight loss > 10% and bleeding disorders. Underlying diseases play a key role in these complications. In a series of 3073 RYGB, predictive factors for mortality were male gender, age, weight, and obstructive sleep apnoea [46]. However, anastomotic leak was one of the strongest independent risk factors for death, and leak-associated mortality was estimated to be higher than 16%. In this paper, Fernandez et al. [46] highlighted the importance of an experienced surgeon in these difficult cases and advised surgeons against operating on such patients early in their learning curve. From the intensivist’s point of view, the general outcome of morbidly obese patients admitted to the ICU is difficult to assess. In a meta-analysis comprising a total of 62 045 critically ill subjects, obesity was significantly associated with prolonged duration of mechanical ventilation and ICU length of stay, but not with excess
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx Table 3 Mortality rates according to the anatomical site of the leak following Roux-en-Y gastric bypass (RYGB).
Hamilton et al. [10] Lee et al. [33] Marshall et al. [11]
Leaks, n (%)
Gastric pouch gastrojejunal anastomosis, n (%)
Excluded stomach, n (%)
Jejuno-jejunal anastomosis, n (%)
9 150 21
0 9/100 (9) 0
1/2 (50) 0 0
0 6/15 (40) 2/4 (50)
7
medical environment, expertise in the management of these difficult cases is only at the early stages. Collaborative and multidisciplinary approaches are mandatory to further our understanding of the pathophysiological mechanisms driving the clinical and biological responses to aggression, and finally to improve the quality of care of these high-risk patients.
Disclosure of interest The authors declare that they have no conflicts of interest concerning this article.
Table 4 Outcome of anastomotic leaks following Roux-en-Y gastric bypass (RYGB) among cases treated conservatively or reoperated. Leaks, n (%)
Ballesta et al. [8] Gonzalez et al. [24] Marshall et al. [11]
59 63 21
Conservative treatment
Reoperation
n (%)
Death rate, n (%)
n (%)
0 0 0
23 40 11
36 23 10
Funding: The author and co-authors have no financial interests to disclose.
References Death rate, n (%) 5 (22) 4 (10) 2 (18)
mortality [47]. In a recent analysis gathering data from more than 119 000 obese patients requiring mechanical ventilation, Kumar et al. [48] observed an increased risk of death among morbidly obese people relative to non-obese people when stratifying by severity of disease. An increased (33%) risk of death has been reported among surgical patients with prolonged ICU stays [49]. Complicated BS cases have a poorer outcome. The mortality rate associated with anastomotic leaks following RYGB range between 6 and 22% [12,50] and up to 50% in some series, depending of the anatomic location of the fistula [10,11,33] (Table 3). Patients undergoing reoperation had high mortality rates of up to 18 to 22% [8,11], and even 40% in some specific subpopulations [33], while no death was reported among cases treated conservatively (Table 4). However, these results must be interpreted cautiously, as conservatively treated cases may have had less severe complications than reoperated patients. In our experience with patients admitted for postoperative peritonitis, despite their young age, these BS patients have a poor prognosis [31]. Complicated BS patients have similar reoperation rates and mortality rates to those of patients with postoperative peritonitis due to other causes [32]. On the other hand, BS did not appear to be a significant risk factor for mortality in the multivariate analysis in this study, suggesting that BS patients had a similar risk to non-BS patients. Delayed surgery has been frequently reported to be associated with an increased risk of organ dysfunction and mortality. The importance of the urgent recognition of complications was recently illustrated by Gonzales et al. [24], who reported a mortality rate as low as 6% for anastomotic leaks after RYGB. In this study, the vast majority of patients presented no signs of organ failure and leaks were detected early (< 48 h) after the index surgery before the onset of any clinical symptoms [24].
8. Conclusion Morbidly obese patients represent a new paradigm for physicians in developed countries. Bariatric surgery has become an increasingly popular approach over recent years. Intensivists are already involved in the management of BS complications, which raises new and challenging issues. While technical and logistic problems (beds, tables, CT-scans and radiology tables, etc) will probably be resolved with progressive improvement of the
[1] DeMaria EJ. Bariatric surgery for morbid obesity. N Engl J Med 2007;356:2176– 83. [2] Buchwald H, Avidor Y, Braunwald E, Jensen MD, Pories W, Fahrbach K, et al. Bariatric surgery: a systematic review and meta-analysis. JAMA 2004;292: 1724–37. [3] Chang SH, Stoll CR, Song J, Varela JE, Eagon CJ, Colditz GA. The effectiveness and risks of bariatric surgery: an updated systematic review and meta-analysis, 2003–2012. JAMA Surg 2014;149:275–87. [4] Arapis K, Chosidow D, Lehmann M, Bado A, Polanco M, Kamoun-Zana S, et al. Long-term results of adjustable gastric banding in a cohort of 186 super-obese patients with a BMI 50 kg/m2. J Visceral Surg 2012;149:e143–52. [5] Benotti PN, Wood GC, Rodriguez H, Carnevale N, Liriano E. Perioperative outcomes and risk factors in gastric surgery for morbid obesity: a 9-year experience. Surgery 2006;139:340–6. [6] Chen CC, Huang MT, Wei PL, Liang HH, Chen SC, Wu CH, et al. Severe peritonitis due to Streptococcus viridans following adjustable gastric banding. Obes Surg 2010;20:1603–5. [7] van den Broek RJ, Buise MP, van Dielen FM, Bindels AJ, van Zundert AA, Smulders JF. Characteristics and outcome of patients admitted to the ICU following bariatric surgery. Obes Surg 2009;19:560–4. [8] Ballesta C, Berindoague R, Cabrera M, Palau M, Gonzales M. Management of anastomotic leaks after laparoscopic Roux-en-Y gastric bypass. Obes Surg 2008;18:623–30. [9] Thodiyil PA, Yenumula P, Rogula T, Gorecki P, Fahoum B, Gourash W, et al. Selective nonoperative management of leaks after gastric bypass: lessons learned from 2675 consecutive patients. Ann Surg 2008;248:782–92. [10] Hamilton EC, Sims TL, Hamilton TT, Mullican MA, Jones DB, Provost DA. Clinical predictors of leak after laparoscopic Roux-en-Y gastric bypass for morbid obesity. Surg Endosc 2003;17:679–84. [11] Marshall JS, Srivastava A, Gupta SK, Rossi TR, DeBord JR. Roux-en-Y gastric bypass leak complications. Arch Surg 2003;138:520–3 [discussion 3–4]. [12] Fernandez Jr AZ, Demaria EJ, Tichansky DS, Kellum JM, Wolfe LG, Meador J, et al. Multivariate analysis of risk factors for death following gastric bypass for treatment of morbid obesity. Ann Surg 2004;239:698–703. [13] Gonzalez R, Nelson LG, Gallagher SF, Murr MM. Anastomotic leaks after laparoscopic gastric bypass. Obes Surg 2004;14:1299–307. [14] Goldfeder LB, Ren CJ, Gill JR. Fatal complications of bariatric surgery. Obes Surg 2006;16:1050–6. [15] Msika S. Surgery for morbid obesity: 2. Complications. Results of a technologic evaluation by the ANAES. J Chir 2003;140:4–21. [16] Cherian PT, Goussous G, Ashori F, Sigurdsson A. Band erosion after laparoscopic gastric banding: a retrospective analysis of 865 patients over 5 years. Surg Endosc 2010;24:2031–8. [17] Owers C, Ackroyd R. A study examining the complications associated with gastric banding. Obes Surg 2013;23:56–9. [18] Pequignot A, Fuks D, Verhaeghe P, Dhahri A, Brehant O, Bartoli E, et al. Is there a place for pigtail drains in the management of gastric leaks after laparoscopic sleeve gastrectomy? Obes Surg 2012;22:712–20. [19] Parikh M, Issa R, McCrillis A, Saunders JK, Ude-Welcome A, Gagner M. 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. [20] Aurora AR, Khaitan L, Saber AA. Sleeve gastrectomy and the risk of leak: a systematic analysis of 4888 patients. Surg Endosc 2012;26:1509–15. [21] Rebibo L, Dupont H, Levrard M, Cosse C, Dhahri A, Verhaeghe P, et al. Gastric fistula after laparoscopic sleeve gastrectomy: don’t forget to treat for Candida. Obes Surg 2013;23:2106–8. [22] Shi X, Karmali S, Sharma AM, Birch DW. A review of laparoscopic sleeve gastrectomy for morbid obesity. Obes Surg 2010;20:1171–7. [23] Carucci LR, Turner MA, Conklin RC, DeMaria EJ, Kellum JM, Sugerman HJ. Rouxen-Y gastric bypass surgery for morbid obesity: evaluation of postoperative extraluminal leaks with upper gastrointestinal series. Radiology 2006;238: 119–27.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
G Model
ACCPM-6; No. of Pages 8 8
P. Montravers et al. / Anaesth Crit Care Pain Med xxx (2015) xxx–xxx
[24] Gonzalez R, Sarr MG, Smith CD, Baghai M, Kendrick M, Szomstein S, et al. Diagnosis and contemporary management of anastomotic leaks after gastric bypass for obesity. J Am Coll Surg 2007;204:47–55. [25] Griffith PS, Birch DW, Sharma AM, Karmali S. Managing complications associated with laparoscopic Roux-en-Y gastric bypass for morbid obesity. Can J Surg 2012;55:329–36. [26] Heneghan HM, Meron-Eldar S, Yenumula P, Rogula T, Brethauer SA, Schauer PR. Incidence and management of bleeding complications after gastric bypass surgery in the morbidly obese. Surg Obes Relat Dis 2012;8:729–35. [27] Topart P, Becouarn G, Ritz P. Comparative early outcomes of three laparoscopic bariatric procedures: sleeve gastrectomy, Roux-en-Y gastric bypass, and biliopancreatic diversion with duodenal switch. Surg Obes Relat Dis 2012;8:250–4. [28] Morino M, Toppino M, Forestieri P, Angrisani L, Allaix ME, Scopinaro N. Mortality after bariatric surgery: analysis of 13,871 morbidly obese patients from a national registry. Ann Surg 2007;246:1002–9. [29] Podnos YD, Jimenez JC, Wilson SE, Stevens CM, Nguyen NT. Complications after laparoscopic gastric bypass: a review of 3464 cases. Arch Surg 2003;138:957–61. [30] Chakravartty S, Sarma DR, Patel AG. Rhabdomyolysis in bariatric surgery: a systematic review. Obes Surg 2013;23:1333–40. [31] Kermarrec N, Marmuse JP, Faivre J, Lasocki S, Mognol P, Chosidow D, et al. High mortality rate for patients requiring intensive care after surgical revision following bariatric surgery. Obes Surg 2008;18:171–8. [32] Montravers P, Guglielminotti J, Zappella N, Desmard M, Muller C, Fournier P, et al. Clinical features and outcome of postoperative peritonitis following bariatric surgery. Obes Surg 2013;23:1536–44. [33] Lee S, Carmody B, Wolfe L, Demaria E, Kellum JM, Sugerman H, et al. Effect of location and speed of diagnosis on anastomotic leak outcomes in 3828 gastric bypass cases. J Gastrointest Surg 2007;11:708–13. [34] Merkle EM, Hallowell PT, Crouse C, Nakamoto DA, Stellato TA. Roux-en-Y gastric bypass for clinically severe obesity: normal appearance and spectrum of complications at imaging. Radiology 2005;234:674–83. [35] Madan AK, Stoecklein HH, Ternovits CA, Tichansky DS, Phillips JC. Predictive value of upper gastrointestinal studies versus clinical signs for gastrointestinal leaks after laparoscopic gastric bypass. Surg Endosc 2007;21:194–6. [36] Lyass S, Khalili TM, Cunneen S, Fujita F, Otsuka K, Chopra R, et al. Radiological studies after laparoscopic Roux-en-Y gastric bypass: routine or selective? Am Surg 2004;70:918–21. [37] Chevallier JM, Zinzindohoue F, Douard R, Blanche JP, Berta JL, Altman JJ, et al. Complications after laparoscopic adjustable gastric banding for morbid obesity: experience with 1000 patients over 7 years. Obes Surg 2004;14:407–14.
[38] Moszkowicz D, Arienzo R, Khettab I, Rahmi G, Zinzindohoue F, Berger A, et al. Sleeve gastrectomy severe complications: is it always a reasonable surgical option? Obes Surg 2013;23:676–86. [39] Patel JP, Roberts LN, Arya R. Anticoagulating obese patients in the modern era. Br J Haematol 2011;155:137–49. [40] Choban P, Dickerson R, Malone A, Worthington P, Compher C. American Society for Parenteral and Enteral Nutrition, A.S.P.E.N. Clinical guidelines: nutrition support of hospitalized adult patients with obesity. JPEN J Parenter Enteral Nutr 2013;37:714–44. [41] Augustin P, Kermarrec N, Muller-Serieys C, Lasocki S, Chosidow D, Marmuse JP, et al. Risk factors for multidrug resistant bacteria and optimization of empirical antibiotic therapy in postoperative peritonitis. Crit Care 2010; 14:R20. [42] Seguin P, Fedun Y, Laviolle B, Nesseler N, Donnio PY, Malledant Y. Risk factors for multidrug-resistant bacteria in patients with post-operative peritonitis requiring intensive care. J Antimicrob Chemother 2010;65:342–6. [43] Hites M, Taccone FS, Wolff F, Cotton F, Beumier M, De Backer D, et al. Casecontrol study of drug monitoring of beta-lactams in obese critically ill patients. Antimicrob Agents Chemother 2013;57:708–15. [44] Buchwald H, Estok R, Fahrbach K, Banel D, Sledge I. Trends in mortality in bariatric surgery: a systematic review and meta-analysis. Surgery 2007;142: 621–35. [45] Khan MA, Grinberg R, Johnson S, Afthinos JN, Gibbs KE. Perioperative risk factors for 30-day mortality after bariatric surgery: is functional status important? Surg Endosc 2013;27:1772–7. [46] Fernandez Jr AZ, DeMaria EJ, Tichansky DS, Kellum JM, Wolfe LG, Meador J, et al. Experience with over 3000 open and laparoscopic bariatric procedures: multivariate analysis of factors related to leak and resultant mortality. Surg Endosc 2004;18:193–7. [47] Akinnusi ME, Pineda LA, El Solh AA. Effect of obesity on intensive care morbidity and mortality: a meta-analysis. Crit Care Med 2008;36:151–8. [48] Kumar G, Majumdar T, Jacobs ER, Danesh V, Dagar G, Deshmukh A, et al. Outcomes of morbidly obese patients receiving invasive mechanical ventilation: a nationwide analysis. Chest 2013;144:48–54. [49] Nasraway Jr SA, Albert M, Donnelly AM, Ruthazer R, Shikora SA, Saltzman E. Morbid obesity is an independent determinant of death among surgical critically ill patients. Crit Care Med 2006;34:964–70. [50] Csendes A, Burdiles P, Burgos AM, Maluenda F, Diaz JC. Conservative management of anastomotic leaks after 557 open gastric bypasses. Obes Surg 2005;15: 1252–6.
Please cite this article in press as: Montravers P, et al. Diagnosis and management of the postoperative surgical and medical complications of bariatric surgery. Anaesth Crit Care Pain Med (2015), http://dx.doi.org/10.1016/j.accpm.2014.06.002
