Enhanced Recovery After Bariatric Surgery (ERABS): Clinical Outcomes from a Tertiary Referral Bariatric Centre

This prospective study collected data on consecutive patients undergoing primary bariatric procedures (LRYGB, sleeve gastrectomy and gastric banding) within an ERABS pathway. Procedures were performed over a 9-month period (15 April 2012 to 15 January 2013) at Imperial Weight Centre (London), a regional high-volume tertiary referral bariatric centre that employed four consultant bariatric surgeons and a bariatric fellow. The study had approval from the hospital regulatory authorities.

Prospectively collected data included demographic details, baseline co-morbidities, operation performed and LOS. Data on occurrence of complications, mortality, re-admissions and re-operations were extracted retrospectively from medical case notes and emergency patient admission lists. The census date for data collection was 1 February 2013 (and included patients operated on until 15 January 2013). Data are presented as mean ± SD or median (interquartile range, IQR) as appropriate.

At the time of the study, regional National Health Service (NHS) commissioning criteria for bariatric surgery were followed. Surgery was indicated for patients with body mass index (BMI) ≥35 kg/m² in the presence of obesity-related medical co-morbidities (defined as difficult to control hypertension, type 2 diabetes mellitus, obstructive sleep apnea or polycystic ovary syndrome). The direct surgical pathway utilized at Imperial Weight Centre included attendance at a patient seminar where information (oral and written) was delivered (by both a metabolic physician and bariatric surgeon) on the various medical and surgical treatments for morbid obesity. Thence followed attendance at a multi-disciplinary one-stop bariatric clinic where within a group setting, patients received presentations delivered by specialist obesity dieticians and clinical nurse specialists describing the perioperative pathway. They were also issued with the appropriate dietary advice sheets and incentive spirometers, ensuring they were familiarized with use of the latter. Pending surgical, medical and anesthetic reviews patients watched a Bariatric Surgery DVD to reiterate indications and expected outcomes from surgery. This ensured sufficient background knowledge was imparted prior to the medical and surgical reviews. Preoperative assessment was subsequently undertaken by a consultant bariatric anesthetist, a metabolic physician and a bariatric surgeon. Appropriate consultations with supporting specialties (renal, cardiac, respiratory and psychiatry physicians) were arranged as indicated. Patients with diabetes mellitus were reviewed by the metabolic physician and their medications adjusted to ensure optimized diabetes control preoperatively. All patients were listed for surgery once any necessary preoperative 'work-up' was completed and reviewed by the multi-disciplinary team. All patients undergoing primary surgery, who lived within a 2-h commute from the hospital (approximately 30 miles distance), and had appropriate home social support (an adult to care for them postoperatively) were included into the ERABS pathway. Otherwise, exclusion criteria included presence of renal (CKD III), cardiac or liver failure that necessitated postoperative admission to a high dependency unit or being wheelchair-dependent. Whilst most ERABS interventions were still utilized in excluded patients, postoperative analgesia, fluid requirements, mobilization and early discharge goals were tailored in a patient-specific manner.

In accordance with the principles of multimodal ERAS pathways utilized in patients undergoing elective colorectal surgery [1], the ERABS pathway utilized at our bariatric centre (Fig. 1) included preoperative, intraoperative and postoperative interventions. Preoperative interventions included: extensive perioperative counselling, shortened fluid fasts (intake of clear liquids permitted up to 2 h pre-operatively) [6], and optimized operating schedules whereby patients were scheduled on operating lists on Mondays, Wednesdays and Fridays, based on their co-morbidities and anticipated postoperative specialist medical reviews and LOS. Patients with potential for day case discharge (e.g., gastric bands) were scheduled first on the operating list, whilst patients with complex co-morbidities were scheduled on Mondays and Wednesdays; anticipating specialist inpatient reviews which would be difficult to arrange during the weekend. As data on preoperative carbohydrate treatment were limited in the bariatric population [7] we did not utilize this intervention in this feasibility study. Patients undertook a preoperative low carbohydrate (800 kcal/day) liver-shrinking diet for a period of 2–4 weeks preoperatively (depending on preoperative BMI and sex). Intraoperative interventions included an optimized bariatric anesthesic protocol that included ramped head-up intubation and extubation, avoidance of long acting opioids (instead remifentanil was utilized intraoperatively), multimodal analgesia (use of intravenous Paracetamol, Diclofenac and Tramadol), volume controlled ventilation, use of high PEEP (6–8 cm/H₂O) and permissive hypercapnea (end tidal CO₂ >6.5 kPa) the latter used for its vasodilatory effects (to allow intra-abdominal bleeding to be detected intraoperatively). Furthermore, no steroids or benzodiazepines were utilized intraoperatively, and the BIS EEG VISTA™ Monitor System (Aspect Medical Systems, Massachusetts, USA) was used to guide and minimize intraoperative anesthetic requirements (BIS to less than 60). Patients typically received 1.5–2 l of crystalloid infusion intraoperatively and local anesthetic was infiltrated at surgical port sites at the start and end of surgery (typically, a total of 40 ml 0.25 % Bupivacaine with 1:200,000 Adrenalin was used). Intraoperative humidification and heating of insufflated CO₂ gas using the HumiGard™ system (Fisher & Paykel Healthcare Ltd, Auckland, New Zealand) was also used as in our experience of this intervention, patients developed less postoperative pain and use of humidified warmed CO₂ reduced laparoscopic lens fogging enhancing speed and clinical efficiency. Alongside the WHO checklist, a dedicated bariatric patient 'time-out' was undertaken identifying patient-specific co-morbidities to the theatre team. Patients were positioned at 60° reverse Trendelenburg to ensure optimal intraabdominal space was available in the supracolic compartment. Laparoscopic RYGB were performed in an antecolic–antegastric fashion (typically 50 cm biliopancreatic limb, 100 cm Roux limb length) using an end-to-side (but functional end-to-end) linear stapled gastrojejunal anastomotic technique. Sleeve gastrectomies were sized using a 34F orogastric tube and utilizing Seamguard® (GORE®) staple line reinforcement. The pars flaccida approach was used to place gastric bands and a gastro-gastric tunnel was formed in all cases to reduce the incidence of band slippage. Nasogastric tubes and urinary catheters were not routinely used and use of surgical drains was reserved for patients whom the surgeon considered at increased postoperative risk of bleeding. Postoperative interventions included protocol-based care using a didactic protocol (Fig. 2) that ensured full mobilization within 4 h of the end of surgery, hourly use of incentive spirometry, a postoperative intravenous fluid balance regimen that avoided fluid overload (previously shown to delay return of gut function [8]) and the administration of regular multimodal (opiate sparing) analgesia (intravenous Paracetamol 1 g QDS, Tramadol melts 50–100 mg QDS), antiemetics (Ondansetron 4 mg TDS), prokinetics (Metoclopromide 10 mg TDS), and laxatives (Lactulose 15 ml BD). Compliance with the aforementioned post-operative interventions was ensured by asking nursing staff to sign opposite each intervention used. Although not formally assessed, lack of compliance with these interventions occurred in only a few patients when certain devices/drugs were out of stock on the surgical ward. A subcutaneous insulin sliding scale regimen was used postoperatively for type 2 diabetic patients with good effect and avoided problems with maintaining 'insulin-exclusive' additional intravenous access postoperatively in these patients. Patients were reviewed postoperatively three times daily by the bariatric fellow at 7:30 am, 1 pm and 4:30 pm to ensure appropriate clinical and discharge concerns were addressed early. Liquid diet was commenced on the first postoperative day without need for routine contrast studies to confirm anastomotic integrity. The latter, however, were selectively used in patients in whom an unsatisfactory intra-operative air-leak test was obtained or when clinically indicated postoperatively. Written discharge information sheets were issued to patients on the morning ward round on the first postoperative day to ensure sufficient time to read and understand information contained therein. Information on the postoperative recovery, diet, alarm symptoms, and a dedicated 24-h emergency bariatric mobile number were also issued to patients. Discharge criteria included the presence of all of: (1) able to drink 1.5 l of fluid per day and tolerating liquid diet; (2) pain adequately controlled with oral analgesia; (3) adequate mobilization; (4) presence of a supervising adult at home; and (5) understand and accept the written information sheets provided. Early postoperative clinical review (proforma-guided) on day 10 was undertaken by the bariatric clinical nurse specialists who had access to an immediate bariatric surgical consult if necessary; with the aim of eliciting early postoperative concerns and to reiterate postoperative dietary instructions to patients.