TEA (T6–T11) is one method to provide optimal analgesia and decrease narcotic requirements following gastrointestinal surgery, particularly if done with an open approach. Continuous epidural analgesia (CEA) or patient-controlled epidural analgesia (PCEA) for 48–72 h provides superior static and dynamic analgesia compared to systemic opioids [4]. Combining local anesthetic with lipophilic opioids [5, 6] and/or epidural epinephrine (2 μg/ml) [7, 8] improves the quality of analgesia. Epidural solution containing epidural morphine (0.02 mg/ml) increases segmental analgesia spread and could be recommended for long midline incisions [9]. Epidural catheters should be inserted in the mid-thoracic region (T6–T8) in patients undergoing upper gastrointestinal surgery and in the low-thoracic region (T9–T11) in patients undergoing lower gastrointestinal surgery. Supplement analgesia is required in patients undergoing abdominal perianal resection, in whom perianal pain (S1–S3 dermatomes) is not controlled by TEA.

Besides its analgesic properties TEA also plays a pivotal role in attenuating the stress response induced by surgery and facilitating early surgical recovery. Through the inhibition of hypothalamus-hypophysis-adrenal axis and thoracic sympathetic fibers, TEA decreases insulin resistance and protein breakdown [10, 11], and furthermore decreases the need for anesthetic agents, opioids, and muscle relaxants [10]. Finally, inhibition of thoracic sympathetic fibers and avoidance of systemic opioids facilitate the recovery of bowel function [12]. Despite these favorable effects, TEA is associated with higher risk of hypotension, pruritus, and lower limb motor weakness. PCEA provides similar analgesia but with less side effects than CEA [13]. Arterial hypotension caused by TEA can be particularly dangerous, especially when primary gastrointestinal anastomoses are created [14]. Interestingly, treating hypotension induced by TEA with intravenous fluids does not restore splanchnic blood flow. On the contrary, administration of small doses of vasopressors has been shown to be safe [15] and increase splanchnic circulation [16]. Orthostatic hypotension associated with TEA does not impair the ability to ambulate [17]. Although TEA impairs bladder functions, urinary catheters can be safely removed the day after the surgery, reducing the incidence of urinary tract infections and without increasing the risk of bladder recatheterization [18]. More rare but more serious complications such as post-dural puncture headache, epidural hematoma, and abscess can also occur. Main contraindications include patient refusal, coagulopathy, thrombocytopenia or platelets dysfunction, and systemic infections. In patients receiving antithrombotic or thrombolytic agents, insertion and removal of epidural catheters should be timed according to international guidelines (Table 13.1) [19, 20].

The benefits of TEA have not been observed after laparoscopic gastrointestinal surgery, where alternative techniques have provided satisfactory analgesia. In fact, two recent RCTs found that in colorectal patients undergoing laparoscopic surgery in a context of an enhanced recovery program, TEA delays hospital discharge [21] and prolongs medical recovery [22] compared with patients receiving intrathecal analgesia or systemic opioids. The use of TEA may remain valuable in patients at high risk of postoperative respiratory complications [23], in those with high probability of conversion to laparotomy, and in patients with an 8–10 cm Pfannenstiel-like incision after laparoscopic rectal surgery in whom pain is better controlled with TEA in the first 24 h [24].

When one considers that up to 1/3 of catheters can dislodge, block, leak [25], or not be correctly inserted [26], these modalities are best used as part of a team approach with highly specialized and experienced providers, including specialized nurses and acute pain service, where the success rates can be much higher, and epidurals not providing adequate analgesia quickly troubleshot. One must assess whether such a program exists at a particular institution before deciding on the use of routine thoracic epidurals. Given these largely equivocal results, and potential difficulties in postoperative management, the authors do not use TEA as a routine part of our practice following laparoscopic gastrointestinal surgery, unlike for open abdominal surgery.

Single-shot spinal analgesia with local anesthetic and intrathecal opioids is a valuable analgesic technique in patients undergoing laparoscopic procedures in whom wound pain relief requirements are more modest, as its analgesic effect is limited to the first postoperative 24 h. Although systemic opioid requirements are significantly decreased [27] compared with patients receiving systemic opioids, the risk of pruritus (OR = 3.85, 95 % CI 2.40–6.15) and respiratory depression (although rare) (OR = 2.35, 95 % CI = 1.00–5.51) is higher. Postoperative urinary retention is also more frequent after intrathecal morphine [28]. Similarly, arterial hypotension is higher and persists in the early postoperative period [29]. Contraindications are similar to those of TEA, but the risk of severe complications associated with this technique is significantly lower [30].

Behind providing excellent analgesia [27], spinal analgesia with intrathecal morphine or diamorphine seems an appealing technique to shorten hospital stay in patients undergoing laparoscopic colorectal surgery with an ERP protocol [21, 31].

In view of its antinociceptive and anti-inflammatory properties, systemic administration of IVL as adjuvant to systemic opioids has been shown to improve postoperative analgesia, reduce opioid consumption, accelerate gastrointestinal function [32], and speed surgical recovery [33, 34]. Similar benefits have been observed after laparoscopic abdominal surgeries when compared to systemic opioids [35], but not when compared to TEA [24], and especially in absence of an ERP [24, 36]. A loading dose of 1.5 mg/kg (ideal body weight) should be initiated 30 min before or at the induction of anesthesia and continued until the end of surgery or in the recovery room (2 mg/kg/h-IBW). The exact duration of the infusion providing optimal analgesia and facilitating also recovery remains unknown. Systemic toxicity is rare, and continuous cardiovascular monitoring is required, limiting its use to the operating rooms or to high-dependency intensive care units [34].

CWI of local anesthetic after open abdominal surgery has been shown to improve postoperative analgesia and reduce opioid consumption [37, 38]; however the effect on the recovery of bowel function is unclear [37, 39]. Two recent RCTs have compared the analgesic efficacy of CWI of local anesthetic with TEA but the results are contrasting [40, 41]. Although preperitoneal multihole catheters have consistently provided satisfactory analgesia, and subfascial catheters have provided better results than suprafascial catheters [42], the anatomical location associated with optimal recovery remains undetermined [38, 42]. A recent feasibility study has compared the analgesic efficacy of CWI of local anesthetic with epidural analgesia after laparoscopic abdominal surgery. Pain intensity was similar among patients receiving epidural and CWI of local anesthetic [43]. Continuous infusion of ropivacaine 0.2 % (10 ml/h) for 48–72 h has been used in the majority of the studies. Other amide local anesthetics have also been used. Systemic opioids are still required to control visceral pain. Unfortunately, they do have a tendency to dislodge, so nursing and patient education are key to proper use.

Abdominal trunk blocks such as transversus-abdominis plane (TAP) block and rectus sheath block have been used to control surgical somatic pain originating from the abdominal wall. Significant reduction of pain intensity and opioid consumption after ultrasound-guided single-shot TAP blocks has been observed in the first 24 h after surgery [44–47]. TAP blocks can also be performed by surgeons from the peritoneal cavity before closing the abdominal wall [48, 49] or using laparoscopic guidance [50–52]. Few studies have reported a reduction of some of the opioid side effects such as nausea and vomiting [46] or sedation [48, 53], but these results have not been consistently reproduced [44]. Others have infused local anesthetic through multihole catheters inserted in the transversus-abdominis plane to improve and prolong opioid-based postoperative analgesia up to 48–72 h after abdominal surgery [54–56]. Niraj et al. found that epidural analgesia did not provide better visual analogue scores during coughing than intermittent local anesthetic boluses through bilateral subcostal TAP catheters in the first 72 h after upper abdominal surgery [57]. However the epidural failure rate was high (22 %) and almost half of the TAP catheters had to be replaced in the postoperative period.

Similar benefits have been reported in abdominal laparoscopic procedures [45, 47, 50–52] and in the context of ERPs [50–52, 58]. Despite facilitating hospital discharge [52], bilateral single-shot TAP blocks do not seem to reduce hospital stay after laparoscopic colorectal surgery [58]. A recent RCT has shown that the analgesic efficacy of four-quadrant TAP blocks in adjunct to bilateral posterior continuous TAP blocks was not inferior to TEA after laparoscopic colorectal surgery [59].

A minimal volume of 15 ml of long-acting local anesthetic injected under ultrasound guidance or at the level of the triangle of Petit is required to achieve satisfactory analgesia with single-shot TAP block [60]. A recent meta-analysis showed that preoperative TAP blocks provide greater analgesia than postoperative TAP blocks [50]. Ropivacaine 0.2 % (8–10 ml/h) can be infused for 48–72 h through a multihole catheter. A bilateral infusion (8–10 ml/h each side) is required with a midline incision. A second injection may be performed just beneath the rib cage (subcostal approach). It is unclear what, if any, effect TAP blocks have on length of stay [52]. Early evidence is encouraging regarding the use of these techniques. A meta-analysis of nine studies including 413 patients showed significant reduction in morphine requirements [46] and a potential for reduced length of stay [52].

Rectus sheath blocks have also been used but the evidence is limited in patients undergoing gastrointestinal surgery. Rectus sheath blocks (15–20 ml of long-acting local anesthetic, bilaterally) are particularly useful to control pain originating from midline incisions, as they provide sensory block for the whole midline of the abdomen. Like TAP blocks they can be inserted under ultrasound guidance or without, using a loss of resistance as verification of the correct plane, although surgeons can also insert them under direct vision. Very commonly a catheter is left in situ and local anesthetic can be administered either by bolus dosing or via continuous infusion, as the analgesic effect is shorter than TAP blocks.

IPLA has been shown to improve postoperative analgesia but not reduce opioid consumption after laparoscopic abdominal procedures [61]. The type of procedure seems to influence this as beneficial effects are seen after upper GI procedures [62] but not after colorectal surgery [63]. This effect might be the result of intraperitoneal deafferentation as indicated by low cortisol and cytokine levels after IPLA instillation [64].

NSAIDs and COX-2 inhibitors have been shown to improve postoperative analgesia and reduce opioid consumption and some of their side effects by 30 % [65]. There have been recent concerns about the risk of anastomotic leakage and the use of NSAIDs or COX-2 inhibitors after colorectal surgeries based on experimental, retrospective, and case series studies [66]. Large RCTs are needed to confirm these results. Although not statistically significant, a trend towards higher risk of developing anastomotic leakage after bowel surgery was reported in a recent meta-analysis of six RCTs (480 patients) of patients receiving at least one dose of NSAIDs or COX-2 inhibitors within 48 h of surgery (Peto OR = 2.16 [0.85–5.53]) [67]. Proper use of these and other oral medications for the multimodal treatment of postoperative pain requires routine (rather than PRN) use. This requires education of the entire treatment team to prevent noncompliance. NSAIDs should be stopped or avoided in the setting of renal dysfunction [68]. An additional concern with NSAIDs is the theoretical increased risk of bleeding. The largest published experience comes from the tonsillectomy literature, where large series suggest that the avoidance of these medications is equivocal at best for avoiding postoperative bleeding episodes [69–71].

Acetaminophen improves postoperative analgesia, and has an opioid sparing effect, but does not reduce opioids side effects [72–75]. An IV formulation (propacetamol) is also available and can be used in patients who are unable to tolerate oral medication. These have been shown to significantly decrease PCA-morphine consumption [76]. The maximum dose is 1 g four times per day. There is some evidence to support a 2 g loading dose, with better pain relief with no increase in toxicity. Use of acetaminophen in conjunction with an NSAID has been shown to be superior to either alone [77]. Acetaminophen dose should be reduced (<2 g/day) in patients with pre-existing liver disease [78, 79].

Perioperative intravenous ketamine and gabapentinoids have also shown opioid sparing properties [80, 81], but they have been poorly studied in patients undergoing gastrointestinal surgery and in the context of an ERP. The risk of side effects such as dizziness and sedation potentially limiting early ambulation should be considered. An opioid-free multimodal analgesic strategy based mainly on analgesic adjuvants would be appealing but more studies are warranted to establish the feasibility, efficacy, and safety of such analgesic approaches [82]. The effect of gabapentanoids seems to be most beneficial when given pre-procedurally, and work through modulation of neuropathic pain [83–85]. There is some evidence that narcotic requirements are decreased [85, 86] and progression to chronic pain states seems to be diminished [87, 88].

The use of a peripherally acting μ-opioid receptor antagonist (e.g., alvimopan) can be used to counteract the intestinal (ileus) side effects of opioid medications. It does not cross the blood brain barrier so it does not alter the therapeutic effects of these medications [89]. Use of this receptor blocker has been shown to enhance the return of bowel function and hospital discharge by 11–26 h [90, 91]. The effect appears to be much more profound in open surgical patients than in laparoscopic procedures [92, 93]. One approach is to give a single dose of alvimopan preoperatively to block the effects of opioid administration intraoperatively for all patients, but only continue it postoperatively in open colectomy cases. Some hospitals have a restriction that a preoperative dose must be given, and this allows continued administration postoperatively in the event of conversion from a laparoscopic to an open procedure.