Advances in surgical and anesthesiological care have allowed major surgery to increasingly be offered to the ageing population as well as in subjects with substantial comorbidity. Thus, in order to reduce risk and to improve clinical outcome in this group of patients, a thorough preoperative preparation and optimization is necessary. This includes a detailed assessment of comorbidity and multidisciplinary involvement in the optimal treatment of hypertension, cardiac and respiratory function [20, 23]. Glucose control should be evaluated using fasting blood glucose or HbA1C levels [20, 23]. Also in nondiabetic individuals, an increased or borderline-increased HcA1C was associated with a threefold increase in postoperative complications after colorectal surgery [24]. In case of anemia, the need of iron supplementation should be considered. Malnourished patients have a high risk of postoperative complications and benefit from preoperative nutritional support, which in most patients is tolerated using the oral route [15]. There is evidence that pharmaconutrition/immunonutrition (supplements containing specific nutrients such as arginine, glutamine, Ω-3 fatty acids, and others) may reduce postoperative infection rates and hospital stay in patients undergoing major abdominal surgery [25]. This intervention may be considered in subjects undergoing procedures associated with a high risk of postoperative infection regardless of preoperative nutritional status.

Prehabilitation comprises preoperative physical conditioning to improve functional and physiological capacity in order to enable patients to recover sooner after surgical stress [20]. A systematic review evaluated the effects of preoperative exercise therapy on postoperative complications and length of stay in surgery of all types [26]. In patients undergoing cardiac, orthopedic, and abdominal surgery, a meta-analysis indicated that prehabilitation led to a reduced length of stay and improved physical fitness. Although the applicability of these studies to patients undergoing specific colorectal or upper GI surgery procedures is unclear, they may be a promising concept.

Tobacco smoking is associated with an increased risk of postoperative morbidity and mortality, attributed mainly to reduced tissue oxygenation (and consequent wound infections), pulmonary complications, and thromboembolism. A recent Cochrane review concluded that cessation of smoking, preferably at least for 4–8 weeks before surgery, was associated with marked reductions in postoperative complications (Intensive care unit intervention, effects on any postoperative complication: RR 0.42; 95 % CI 0.27–0.65) [27]. In addition, hazardous drinking, defined as intake of three alcohol equivalents (12 g ethanol each) or more per day, has long been identified as a risk factor for postoperative complications. Alcohol abstinence for 1 month has been associated with better outcome after colorectal surgery [28]. Available ERAS guidelines for colorectal and upper GI surgery, therefore, recommend cessation of alcohol for abusers and tobacco use in all patients 4 weeks prior to surgery. In bariatric and other benign major abdominal surgery, even longer periods of alcohol abstinence are usually recommended in patients with history of alcohol abuse .

Preoperative information and/or a visit to the surgical ward have been shown to reduce anxiety, and improve compliance with postoperative instructions, postoperative recovery, length of stay, and long-term outcomes after various types of surgery [15]. Although data from studies specifically evaluating the effect in specific procedures such as in upper gastrointestinal surgery are sparse, preoperative counseling is part of currently published ERAS guidelines.

Fasting from midnight before elective surgery is not supported by evidence, and therefore, in most guidelines has been replaced with guidance for fluid intake of clear fluids up to 2 h prior to induction of anesthesia [29]. Solids should, however, be withheld until 6 h before operation to prevent risk of aspiration. A preoperative carbohydrate-rich drink given up to 2 h before anesthesia has been shown to reduce preoperative hunger, thirst, and anxiety [30]. In addition, PONV [31] and surgical stress as measured by postoperative insulin resistance and protein catabolism are improved and length of stay is reduced, with the most pronounced effect after major surgery [6]. Avoiding preoperative fasting using carbohydrate loading is therefore recommended in current ERAS guidelines for colorectal surgery, gastrectomy, pancreaticoduodenectomy [14–16], and bariatric surgery [13].

Prophylactic antibiotics reduce infectious complications. Patients should receive a single dose orally or intravenously at least 30 min before skin incision [15]. Repeated dosing can be administered depending on the half-life of the drug and the duration of surgery. The skin should be prepared with chlorhexidine–alcohol [15].

There is no convincing evidence in the literature of the benefits from long-acting sedatives prior to surgery and their use is therefore not recommended. Short-acting anxiolytics might be used, in particular to facilitate procedures such as insertion of epidural catheters. The data from studies comparing various anesthetic protocols is sparse. However, the use of short-acting induction agents such as propofol and opioids such as sufentanil is usually recommended and included in available ERAS anesthesiological [20], colorectal [15], and upper GI guidelines [20]. In addition, short-acting muscle relaxants are widely used. In particular in laparoscopic surgery, deep neuromuscular block is helpful in order to ensure surgical access. In order to avoid deep sedation, a Bispectral Index (BIS) might be used for titration of anesthetic agents although the evidence for its efficacy is limited.

Near-zero fluid balance, avoiding salt and water overload, has been shown to result in improved outcomes [20, 23]. Vasopressors should be considered as first choice to treat hypotension to avoid unnecessary fluid overload. Goal-directed fluid therapy is recommended to obtain optimal tissue perfusion and in high-risk patients Doppler-guided techniques might be used in order to improve outcome [32], even though the benefits are unclear in patients already managed within an ERAS pathway.

There is convincing documentation of benefits associated with prevention of hypothermia in terms of reducing complications as well as improving postanesthetic recovery [33]. This is usually achieved by the use of active cutaneous airborne heating systems (Bair-hugger) or circulating-water garments. Avoidance of hypothermia is of particular importance in surgical procedures with long operating times such as pelvic procedures or pancreaticoduodenectomy, whereas the effects might be less pronounced in, for example, uncomplicated laparoscopic bariatric surgery [34].

Minimally invasive surgery reduced damage to tissues by changes in surgical access [20]. In open surgery, the length and orientation of incision affect pain and may influence surgical outcomes [20]. The extent of the injury to abdominal wall is further reduced using minimally invasive techniques such as laparoscopy which has been evaluated for the treatment of colorectal cancer in randomized trials [35, 36]. The safety and overall value of robotic surgery remains unclear although present evidence suggests higher costs and at least similar rates of complications [37]. We are awaiting results from a large multicenter (RCT: ROLARR, ClinicalTrials.gov. Identifier: NCT01736072). Other minimally invasive options such as Trans-anal TME, SILS, or NOTES are still under evaluation.

In bariatric surgery, laparoscopy has rapidly superseded open surgery due to improved outcome in terms of reduced complications and improved recovery [38]. For distal gastrectomy, there is evidence supporting the use of laparoscopic-assisted surgery in early gastric cancer, whereas more data on long-term survival after laparoscopic compared to open surgery in advanced disease is still awaited [14]. In total gastrectomy, laparoscopic-assisted approach might be used if expertise is available, since it has been shown to reduce complication rates and improve patient recovery [14]. Although laparoscopic resection of the pancreatic head has been shown to be feasible, too little data is available on oncological outcomes after laparoscopic pancreaticoduodenectomy to recommend its routine use.

A Cochrane meta-analysis concluded that routine nasogastric intubation following open abdominal surgery should be abandoned in favor of selective use [39]. A subgroup analysis of nine RCTs with 1085 patients that underwent gastroduodenal surgery found increased pulmonary complications associated with routine use of postoperative nasogastric tube. In addition, intra-abdominal or pelvic drains have no advantage in colorectal surgery [15] although the evidence in pelvic procedures was based on a small number of patients. However, a large multicenter RCT of prophylactic pelvic drains in low anterior resection (GRECCAR 5, ClinicalTrials.gov Identifier: NCT01269567) was recently completed and preliminary data (Presented by Dr. Denost at ESCP in Dublin, at International Trials Symposium, September 23rd, 2015) show no effect of pelvic drains on the incidence or severity of anastomotic leakage. Peri-anastomotic drains have not been shown to reduce overall complication rates in pancreatic [40] or gastric cancer surgery [41], and are associated with slower recovery [42]. Similarly, no advantages were shown by the use of abdominal drain after gastric bypass for morbid obesity. Thus, no convincing evidence supports the routine use of postoperative drains after upper gastrointestinal surgery. In contrast , the use of a passive subcutaneous drain was associated with a reduction in superficial surgical site infections in a randomized study of 263 patients undergoing open or laparoscopic colorectal surgery [43].

The duration of urinary drainage should be as short as possible, and the catheter can in most cases be removed within 24 h after colorectal surgery without increased incidence of urinary retention [15]. Early removal with intermittent urine drainage as needed has been shown to be safe also in patients with thoracic epidural analgesia [15]. When urinary catheterization of more than 3 days postoperatively is expected (i.e., some pelvic procedures), a suprapubic catheter seems the better choice [15]. The optimal duration of ureteral stents and transurethral neo-bladder catheter after radical cystectomy is still unknown [18].

Comprehensive ERAS guidelines for anesthesia practice in gastrointestinal surgery have recently been published [20]. In open abdominal surgery, epidural analgesia (EDA) has been shown to provide superior postoperative pain control compared with opioids as well as patient-controlled intravenous opioid analgesia (PCA). Moreover, the EDA was reported to be associated with fewer episodes of postoperative ileus, pulmonary complications, and improved insulin sensitivity. A thoracic EDA is recommended in ERAS guidelines for open colorectal and major upper GI surgery such as pancreaticoduodenectomy and gastrectomy. Studies evaluating the use of EDA in open liver resections are sparse. The EDA in laparoscopic colorectal procedures where skin incision and abdominal wall injury is kept minimal has been questioned. In addition, there is no consensus regarding the value of EDA in laparoscopic upper abdominal surgery, such as gastric bypass. In situations where an EDA cannot be used, a PCA is the most commonly used alternative after open abdominal surgery although other alternatives, including various techniques for regional anesthesia and intravenous lidocaine infusion, are recommended in ERAS guidelines [20, 23]. After cessation of EDA or PCA, multimodal systemic analgesia should be used including non-opioid analgesics such as paracetamol and NSAIDs. For opioids, when necessary, the enteral routes should be used as soon as possible.

Although mainly extrapolated from studies in colorectal surgery, available data suggest that patients at risk of PONV should be treated with a multimodal approach with the use of antiemetics according to patient risk factors [20, 23]. This includes the use of propofol for induction of anesthesia and avoidance of volatile anesthetics, opioids, and fluid overload. The recommended antiemetics for PONV prophylaxis vary in their efficacy and include 5-hydroxytryptamine receptor antagonists, corticosteroids, butyrophenones, neurokinin-1 receptor antagonists, antihistamines, and anticholinergics [44].

The risk factors for venous thromboembolism (VTE) include major surgery, malignant disease, and obesity. Therefore, patients undergoing major colorectal and upper GI surgery are at risk. Low molecular weight heparin (LMWH) is effective at preventing VTE and advantageous compared to unfractionated heparin due to its once-daily administration. Mechanical methods such as intermittent pneumatic compression or graduated compression stockings may be used as an adjunct in patients who are at moderate or high risk for VTE. LMWH treatment is usually initiated either the evening before, or within 6 h postoperatively and continued at least until patients are fully mobile. After major open surgical procedures 4 weeks treatment is usually recommended, whereas 7 days is usually considered sufficient after laparoscopic surgery. The risk of spinal or epidural hematoma in patients with EDA should be considered and a 12 h interval between LMWH administration and catheter insertion or removal should be adhered to.

Major open abdominal surgery is associated with long recovery time even in the absence of complications. Prolonged immobilization /bed rest is associated with several adverse effects and should be avoided although scientific data is lacking [20, 23]. Day-to-day targets for mobilization should be defined and progress monitored and documented. Satisfactory pain control is mandatory in order to achieve adequate mobilization. In patients undergoing laparoscopic surgery, early mobilization is normally much easier to achieve, and usually possible within a few hours after surgery [20, 23].

Early oral intake has been shown to be safe and most often feasible after major colorectal as well as upper gastrointestinal surgery [20, 23] and should therefore be encouraged. However, in the presence of impaired gut function, enteral or parenteral nutritional support might be necessary, in particular if complications occur. Return to oral intake should be aimed for as soon as possible. The need for motility-enhancing drugs is usually not required after upper GI surgery compared with after colorectal surgery. Although commonly used after colorectal surgery, only some fast-track programs for pancreatic surgery include the use of laxatives postoperatively, and there is limited documentation of the effectiveness of such regimens after gastrectomy and hepatic surgery.

Patients can generally be discharged when they tolerate adequate oral intake, when they are fully mobilized and when pain can be managed with oral analgesics. Sufficient time should be provided for the patient to independently manage a new stoma . After early discharge, patients should be contacted by a nurse after 2–3 days, to assure that rehabilitation is progressing well. Usually another contact 30 days postoperatively is useful in order to assure a normal postoperative course and to prevent hospital readmissions [45].

A structured audit on perioperative care and clinical outcome is essential for maintaining a successful ERAS program. Using the International ERAS database facilitates this process through a detailed registration on the perioperative care, and the clinical outcome of the patients in combination with a clique view statistical ad on that provides an easy and immediate feedback and analysis of registered data (http://​www.​erassociety.​org/​).