Introduction

Colorectal cancer is the second most common cancer diagnosed in women and third most common in men, and accounts for approximately 10% of all annually diagnosed cancers and cancer-related deaths worldwide.1 The established curative form of treatment is surgical removal preceded by neo-adjuvant combined radio and chemotherapy and/or followed by adjuvant chemotherapy in eligible cases.2 Laparoscopic resection, to different extents according to tumour site, has become the standard technique worldwide given its notable benefits, mainly the shorter length of stay and less postoperative pain.3

Through the PROSPECT (PROcedure-SPECific postoperative pain managemenT) Working Group, anaesthesiologists and surgeons collaborate with the purpose of formulating evidence-based recommendations for postoperative pain management specific to each surgical procedure.4 Overall recommendations assessing the efficacy and adverse effects of each technique (www.postoppain.org) are based on procedure-specific evidence as well as extensive clinical practice information.5 The methodology of the PROSPECT group is unique in that it aims to synthesise clinical evidence while considering risks and benefits of interventions, also taking into account study design. Specifically, the group seeks to determine the relevance of interventions in current peri-operative care, and critically evaluate the baseline pain treatment.5

The aim of this systematic review was to evaluate the current literature on postoperative pain management following laparoscopic colorectal surgery, updating our previous recommendation.6 The primary outcomes were postoperative pain scores and analgesic requirements. We also considered study quality, clinical relevance of trial design and a comprehensive risk–benefit assessment of the analgesic intervention.

Methods Systematic literature search

Five electronic databases (PubMed, Embase, Ovid MEDLINE, Cochrane Database of Systematic Reviews and Cochrane Central Register of Controlled Studies) were comprehensively searched to identify randomised controlled studies (RCT) published before 25 January 2022, taking into account the previously published guidelines,6 and any combination of medical subject heading (MeSH) terms, text words and word variants related to pain interventions for laparoscopic colorectal surgery without language or publication date restrictions. The terms included, but were not limited to, those related to pain, analgesia, pain treatment, systemic analgesics, nerve blocks and adjuvant medication (Supplemental Data File 1, https://links.lww.com/EJA/A900). Reference lists of retrieved reviews, systematic reviews and meta-analyses were hand searched for additional eligible RCTs.

Study inclusion/exclusion criteria

Three independent authors (JB, MH and MS) conducted the literature search, screening and excluding irrelevant articles. Titles and abstracts, and if necessary full texts, were independently screened to identify potentially relevant studies assessing analgesic, anaesthetic or surgical interventions affecting postoperative pain in patients who had undergone laparoscopic colorectal surgery for benign or malignant conditions. Reviewers then compared their initial screening results. Any disagreements were resolved by consensus between reviewers and a senior author (PL), and the potentially relevant RCTs published as full text in English were assessed for the final inclusion criteria. We only included studies that reported pain intensity measured by visual analogue scale (VAS) or numerical rating scale (NRS) in adults undergoing elective laparoscopic colorectal surgery, including studies with a defined laparoscopic colonic resection subgroup in mixed surgical procedures and studies with patient collectives being stated as ‘colorectal’ where only abdominal and no perineal incisions were made.

Analysis of outcome and statistical analysis

The criteria used to assess the quality of eligible studies were summarised in Joshi et al.5 We also recorded whether adequate basic analgesia (defined as round-the-clock paracetamol along with, unless contraindicated, nonsteroidal anti-inflammatory drugs (NSAIDs) or cyclo-oxygenase-2 (COX-2)-inhibitors) was provided to study participants in both intervention and control groups.5 Unless specified otherwise, it was assumed that the pain scores were assessed at rest. We also retrieved comparative opioid consumption when different analgesic regimens were used, particularly when the study design included patient-controlled intravenous opioid (morphine or morphine equivalent) administration. The studies were stratified according to the intervention evaluated: analgesic/anaesthetic or surgical. The effectiveness of each intervention was evaluated qualitatively by assessing the number of studies showing a significant difference. Recommendations are given when at least two congruent studies support an intervention. For this review, we define a change greater than 10 mm on a 100 mm VAS scale or greater than 1 on a 10-point NRS scale as clinically important.5,7

The results of the literature search were presented to the entire Working Group in two subsequent meetings using a modified Delphi approach.5 The lead authors proposed interpretations of the findings and the entire group was engaged in the drafting of the final recommendations in several iterations, collating rounds of individual comments on the evidence and draft recommendations, followed by round-table discussions and further rounds of individual comments, and votes in cases of disagreement. Where applicable, vote results are reported in the Results section. When consensus had been obtained, the lead authors compiled the manuscript, and again, all the Working Group members participated actively in the writing and discussion.

Results

A total of 4461 studies were initially identified in our broad literature search. After exclusion of duplicate and irrelevant studies, based on consensus between the three reviewers (JB, MH and MS) and the senior author (PL) a total of 457 studies were identified for potential inclusion (Fig. 1) and investigated using abstract screening. Of these, 354 manuscripts were excluded because they did not meet inclusion criteria. Of the remaining 103 RCTs selected to undergo full text data extraction, another 31 studies were excluded at closer examination: six studies in a language other than English,8–13 two studies with no access to full-text,14,15 thirteen studies with no pain scores reported16–28 and ten studies that combined different types of surgery without laparoscopic colorectal surgery subgroup analysis.29–38

Fig. 1:

PRISMA Flow Diagram.

A total of 72 RCTs were included. Some studies had more than two relevant study arms. These were separately listed for each relevant comparison, making the sum of studies in the following listing higher than the number of studies included: seven studies investigating adjuvant systemic analgesics,39–45 three studies investigating varying degrees of neuromuscular blockade,46–48 seven studies investigating intravenous lidocaine,43,49–54 six studies investigating epidural analgesia,55–60 four studies investigating spinal opioids,55,61–63 twenty-eight studies investigating truncal nerve blocks,53,58–60,64–87 seven studies investigating wound infiltration,54,57,72–74,88,89 four studies investigating intraperitoneal local anaesthetics (IPLA)90–93 and seventeen studies investigating surgical techniques.94–110 Supplemental Table 1, https://links.lww.com/EJA/A901 lists the summary of key results from studies evaluating systemic analgesics, systemic analgesic adjuncts, regional analgesia and surgical procedures used to support the recommended interventions in patients after laparoscopic colectomy. Supplemental Table 2, https://links.lww.com/EJA/A902 lists the corresponding data from procedures where no consensus could be reached, and Supplemental Table 3, https://links.lww.com/EJA/A903 lists the evidence for procedures not recommended.

Analgesic or anaesthetic interventions Nonopioid analgesics and analgesic adjuncts

Our group had previously recommended the administration of NSAIDs or COX-2-inhibitors.6 This recommendation is strengthened by one study showing an additive analgesic effect of intravenous parecoxib compared to 0.9% saline, when added to an epidural analgesia regimen.40 The group paid particular interest to the issue of NSAID in rectal surgery, and voted 8 : 1 in favour of recommending NSAID/COX-2 specific inhibitors for colonic, but not for rectal surgery, given concern over potential anastomotic leakage in the latter.

The nonopioid analgesic, nefopam, was shown in one recent study to decrease dynamic pain scores after preincisional versus intra-operative administration.39

Pregabalin was found to be opioid-sparing but did not decrease pain scores in a setting of adequate baseline analgesia.41 Dexmedetomidine, in two studies, did not have an analgesic or opioid-sparing effect.42,43 Cho et al.44 found no analgesic or opioid-sparing effect for ketamine when added to an opioid-based regimen. Laparoscopic colorectal surgery was also used as a model to test the analgesic efficacy of a novel analgesic compound (VVZ-149) inhibiting the type 2 glycine and the type 2A serotonin receptor.45

In accordance with PROSPECT methodology, a common core of paracetamol and NSAID/COX-2 specific inhibitors are recommended as baseline analgesia for colonic surgery; paracetamol is recommended for rectal surgery. Pregabalin, dexmedetomidine and ketamine are not recommended because of limited procedure-specific evidence.

Neuromuscular blockade

A study by Kim and colleagues with adequate baseline analgesia compared moderate (1 to 2 train-of-four response) with deep (1 to 2 posttetanic count) muscle relaxation achieved via a continuous infusion of rocuronium in laparoscopic colorectal surgery. Deep neuromuscular blockade showed significantly lower NRS pain scores immediately after surgery, and at 6, 24, 48 h after surgery, and also a reduced incidence of shoulder tip pain and reduction in intravenous opioid and other rescue analgesics requirement.46 In contrast, Cho et al.47 found no analgesic effects of moderate versus deep neuromuscular blockade. More recently, Gu et al.48 demonstrated decreased pain scores up to 24 h after surgery, and an opioid-sparing effect for deep neuromuscular blockade.

In summary, deep neuromuscular blockade is not recommended because of conflicting procedure-specific evidence.

Intravenous lidocaine

Our group had previously recommended the administration of intravenous lidocaine in cases where other analgesics are not ‘adequate or appropriate’,6 based on studies by Kaba111 and Wongyingsinn.112 Since then, Ahn et al.49 compared intra-operative continuous lidocaine infusion with normal saline, and found an analgesic effect during the first 24 h after surgery, in a setting without adequate baseline analgesia, while Tikuisis et al.50 found an analgesic effect at 2, 12 and 24 h, but not at 4 and 8 h postoperatively, and a small decrease in ketorolac rescue analgesia. Elhafz et al.51 had reported analgesic and opioid-sparing effects of lidocaine over placebo in a study with no baseline analgesia. One study by Andjelkovic et al.43 found an opioid-sparing, but not analgesic, effect for lidocaine on the first and second postoperative day. In contrast, Kim et al.52 and Dewinter et al.53 found that under adequate baseline analgesia, lidocaine had no analgesic or opioid-sparing effect. Finally, Beaussier et al.54 failed to detect a significant difference in terms of hyperalgesia, postoperative pain and residual incisional pain at 3 and 6 months, comparing lidocaine to controls.

In summary, based on our previous recommendation, the ambivalent nature of studies published since, and the cautionary findings of a recent meta-analysis113 the recommendation on lidocaine required a vote, with eight members present voting in favour, and six against recommending lidocaine. Because no consensus could be reached, no recommendation could be made for use of intravenous lidocaine as a first-line treatment. However, intravenous lidocaine may be considered when basic analgesia cannot be provided.

Epidural analgesia

Previously, our group did not recommended epidural analgesia for laparoscopic colorectal surgery because of a risk/benefit assessment6 that took six RCTs into account.112,114–118 In another report, Levy et al.55 published a three-arm study with adequate baseline analgesia, comparing spinal morphine, epidural analgesia and intravenous opioid patient-controlled analgesia (PCA). In this study, epidural analgesia, compared with systemic analgesia, had lower pain scores in the recovery room, and through to postoperative day 2, but at the expense of a longer length of hospital stay. In contrast, no beneficial analgesic effect of epidural analgesia over PCA was found by Hubner et al.56 in the presence of baseline analgesia.

Barr et al.57 compared levobupivacaine and fentanyl thoracic epidural anaesthesia (TEA) to continuous local anaesthetic infusion via wound infusion catheter. The authors reported no significant difference in pain scores and mean analgesic requirement; however, adequate baseline analgesia was not used in the study. Bumblyte et al., comparing thoracic epidural analgesia with transversus abdominis plane (TAP) blocks in patients receiving adequate baseline analgesia, found an early opioid-sparing but no analgesic effect (maximal pain score) for epidural analgesia.58 No analgesic benefit was found when comparing TEA with TAP blocks with adequate baseline analgesia,59 or without.60

Considering the invasive nature of epidural analgesia, the doubtful analgesic efficacy in laparoscopic colorectal surgery, combined with a potential for serious complications, and the availability of alternative analgesic methods that are less invasive, epidural analgesia is not recommended.

Intrathecal morphine

The intrathecal administration of morphine was not recommended previously by our group based on limited evidence of benefit and the potential for side-effects such as pruritus, respiratory depression, urinary retention and nausea/vomiting.6 The authors took into account the study by Kong et al.119 and noted a further investigation by Levy et al.55

Two more studies with adequate baseline analgesia compared intrathecal diamorphine to intravenous analgesia, both reporting a significant decrease in pain scores and opioid requirements with spinal morphine.61,62 A study by Koning et al.63 also investigated spinal bupivacaine and morphine versus a control group who received a sham procedure and a bolus of piritramide. The spinal analgesia group needed less intra-operative sufentanil and had significantly lower NRS pain scores on the first day despite receiving less piritramide by PCA. Of note, significantly more patients in the spinal group had pruritis but no difference in nausea, drowsiness, or length of stay was noted. Levy et al.55 also found an early analgesic benefit for spinal morphine over PCA.

To sum up, intrathecal morphine exhibits clear analgesic efficacy, but given that the routine anaesthetic method for laparoscopic colorectal surgery is general anaesthesia, a separate spinal puncture would be required to initiate this treatment, and specific side-effects are a concern. This weighing of risks and benefits prompted a discussion in the group, with seven members voting to recommend, based on the perceived simplicity, cost-effectiveness and widespread availability of the intervention, whereas eight present members voted against a recommendation based on the invasive nature of the procedure and the potential of side effects. Because no consensus could be reached, no recommendation could be made for use of intrathecal morphine.

Truncal blocks

Multiple studies have assessed the effect of truncal blocks in laparoscopic colorectal procedures. The first technique to be investigated was TAP blockade. Here, seven studies compared TAP blockade with controls. Tikuisis et al.64 found analgesic and opioid-sparing effects of TAP blocks in the setting of hand-assisted colectomy, in an experimental design which provided adequate basic analgesia to all patients. Ma and colleagues65 investigated TAP blocks in the setting of conventional laparoscopy, and without basic analgesia, and found analgesic and opioid-sparing effects of TAP blocks. The studies by Zhao et al., Torup et al. and Walter et al. demonstrated an opioid-sparing but not analgesic, effect.66,68,120 Keller et al.69 found analgesic but not opioid-sparing effects. Finally, Oh et al.70 and Smith et al.71 found no beneficial analgesic or opioid-sparing effect of TAP blocks.

Next, TAP blocks were compared with other interventions, such as wound infiltration, intrathecal morphine and epidural analgesia. Compared with wound infiltration, three studies did not find superior analgesia with TAP blocks.72–74 Han et al. demonstrated that TAP blocks did not show additional analgesic effects when superimposed on intrathecal morphine.75 When compared with thoracic epidural analgesia, one study showed less early morphine consumption in the epidural than in the TAP group,58 whereas no difference was detected in two other studies.59,60 Hamada et al.76 showed that the addition of dextran to levobupivacaine could prolong analgesic effects of TAP blocks in a setting without adequate baseline analgesia.

The question whether the TAP block should be performed intra-operatively under ‘direct’ vision, or postoperatively using ultrasound guidance was investigated by three studies. Two studies found no difference in efficacy,77,78 whereas one study tended to sway in favour of surgeon-performed intra-operative blocks.79

Another block discussed is the quadratus lumborum block (QLB), which was shown by Dewinter et al.53 to be equally effective as adequate systemic baseline analgesia, albeit at the cost of a higher incidence of self-reported symptoms of local anaesthetic toxicity. Wang et al.80 found QLB to be more analgesic and opioid-sparing than a sham procedure in a study without basic analgesia. Of note, this study highlighted a relevant safety concern. The QLB group had a significantly higher incidence of subjective symptoms of local anaesthetic systemic toxicity.53 When comparing the QLB to TAP block, Deng et al.81 found opioid-sparing, and Huang et al.82 reported marginal analgesic and opioid-sparing effects in favour of QLB. Aoyama et al.83 found no difference in analgesic effect when QLB was performed as a continuous infusion versus programmed intermittent bolus infusion.

Finally, erector spinae blockade (ESPB) was investigated by two studies. Rao Kadam et al.84 found no analgesic advantages over wound infiltration in a setting of adequate baseline analgesia, and Qi-Hong et al.85 found very small, and probably clinically insignificant, opioid-sparing effects of ESPB over TAP.

Two studies failed to demonstrate analgesic superiority of liposomal bupivacacine over a mixture of bupivacaine–epinephrine–dexamethasone in TAP blocks,86 or standard bupivacaine in QLB.87

In conclusion, truncal blocks (TAP, QLB, ESPB) are not recommended as a first-line analgesic method in laparoscopic colorectal surgery because of inconsistent procedure-specific evidence. TAP blocks may possibly offer clinical benefits if an additional incision is created (hand-assisted surgery).64

Wound infiltration

The previous guidelines did not comment on wound infiltration.6 Moore et al.88 (wound infiltration) and Fustran et al.89 (preperitoneal infiltration) found no analgesic or opioid-sparing effect. Wound infiltration was equally effective as levobupivacaine and fentanyl TEA in a study by Barr et al.57 Three studies found equal acute pain outcomes between TAP blocks and wound infiltration (see above under TAP blocks).72–74 Finally, Beaussier et al.54 found no difference in analgesic efficacy between wound infiltration and systemic lidocaine.

In summary, despite the inconsistent evidence, wound infiltration is recommended because of its simplicity and low cost.

Intraperitoneal local anaesthetic

Our group had previously not recommended IPLA because of questions over optimal dose and timing,6 taking into account the study by Kahokehr et al.121 In more recent studies, Park et al.90 found analgesic and opioid-sparing effects, and Duffield et al.91 found analgesic, but no opioid-sparing effects, both in settings that lacked basic analgesia. Fares et al.92 found that bupivacaine mixed with dexmedetomidine was more effective than plan bupivacaine, which in turn was not more effective than saline. Finally, Stephensen et al.93 investigated the potential efficacy of continuous intraperitoneal local anaesthetic versus saline for 3 days, and found no analgesic or opioid-sparing effects.

Given the inconsistent evidence, and absence of studies where IPLA was studied in the context of adequate baseline analgesia, this technique is not recommended for laparoscopic colorectal surgery, unless basic analgesia or intravenous lidocaine cannot be provided.

Surgical techniques

The majority of studies reported inadequate or unspecified baseline analgesia. Hand-assisted laparoscopic and straight laparoscopic colonic resection in the management of right sided colonic cancer showed similar postoperative pain scores.94,95 A study comparing robot-assisted to laparoscopically assisted colonic resection for right sided colonic cancer found no significant difference in VAS pain scores at 24, 72 and 120 h after surgery.96

Multiple studies compared single-incision laparoscopic surgery (SILS) to conventional laparoscopic surgery (CLS) in colorectal surgery.98–103 Poon et al.98 reported an analgesic benefit of single-instrument surgery specifically on days 1 and 2, whereas Song et al.102 found a marginal difference in pain scores on postoperative day 2. In the setting of background epidural analgesia for the first 72 h, Bulut et al.103 demonstrated improved pain scores on dynamic pain for SILS; however, three other studies found no acute pain advantage of SILS over CLS in colorectal surgery.99–101

Leung et al.104 compared their Hybrid Novel technique of natural orifice transluminal endoscopic surgery (NOTES) colectomy (HNC) to the conventional laparoscopic colectomy for left-sided tumours. The first week after surgery showed a significantly lower maximum pain score in the HNC group. For left sided colonic diseases, Wolthuis et al.105 investigated laparoscopic natural-orifice specimen extraction (NOSE) colectomy (n = 20) versus conventional laparoscopic colectomy (n = 20) with adequate baseline analgesia, and found analgesic and opioid-sparing effects with the NOSE procedure.

Two studies with a background of adequate baseline analgesia investigated the site of extraction. Tan et al.106 showed no difference in pain scores or analgesic consumption between a vertical peri-umbilical wound and a transverse left iliac fossa incision. Similarly, Lee et al.107 found no difference in pain scores between patients receiving midline versus transverse specimen extraction. Other studies highlighted the effect of different thermal conditions of CO2 insufflation on laparoscopic colorectal surgery,108 and the beneficial effect of intra-corporeal as compared with extracorporeal anastomosis.109

In summary, no surgical technique showed sufficient evidence to be recommended. There is inconsistent evidence for SILS, and two promising pilot studies on natural-orifice surgery, but that latter technique has not yet achieved widespread use.

Discussion

The current systematic review and guideline recommendations update our previous recommendations for laparoscopic colorectal surgery. Even though the volume of both clinical procedures and scientific evidence has increased substantially over the past decade, our procedure-specific recommendations remain limited to basic analgesia (combination of paracetamol and NSAIDs/COX-2 specific inhibitors), wound infiltration and rescue opioids (Table 1). No consensus could be achieved for the use of intrathecal morphine and intravenous lidocaine, thus no recommendation could be made for these interventions. However, intravenous lidocaine may be considered when basic analgesia cannot be provided (Table 2).

Table 1 - Overall recommendations for pain management following laparoscopic colorectal surgery Paracetamol and nonsteroidal anti-inflammatory drugs, administered preoperatively or intra-operatively (if no contraindications) Recommended Surgical port site wound infiltration Recommended Rescue opioids Recommended Intravenous lidocaine No consensus reached, may be used when basic analgesia cannot be provided Spinal morphine No consensus reached
Table 2 - Interventions that are not recommended for pain management following laparoscopic colorectal surgery Interventions Reasons for not recommending Intraperitoneal local anaesthetics Inconsistent evidence, may be used when basic analgesia or intravenous lidocaine cannot be provided Deep neuromuscular blockade Limited procedure-specific evidence Epidural analgesia Comprehensive risk–benefit assessment Truncal blocks Inconsistent procedure-specific evidence Specific surgical techniques Lack of procedure-specific evidence
Methodology

One of the main challenges in the creation of systematic reviews is that they require careful analysis of eligible studies with a clinical interpretation by both surgeons and anaesthesiologists, to identify for the reader the interventions that are recommended or not. This is especially important, given that for virtually all interventions, there are both positive and negative studies, and very few studies in peri-operative medicine with methodology and patient cohorts that are sufficiently similar to permit comparison. There exists a considerable heterogeneity in study design, which necessitates a balanced clinical view for interpretation of the overall evidence-base. In particular, some aspects of trial design, such as the provision of basic analgesia to all patients, are essential to interpreting results.5 In that sense, the PROSPECT approach is helpful in devising a basic plan, before taking individual patient characteristics and institutional idiosyncrasies into account.

Basic systemic analgesia

The provision of basic analgesia has become a key proposition of the PROSPECT group, both for patient care and for research settings.5,122 Especially with a view to the generation of clinically meaningful scientific data, we strongly advocate that all studies investigating analgesic interventions assess the intervention compared with, or superimposed on, a combination of drugs that is ubiquitously available and cost-effective. This would usually encompass simple analgesics such as paracetamol and NSAID/COX-2 specific inhibitors, and, if applicable, simple wound infiltration.5

In the case of colorectal surgery, however, the past years saw an active discussion on the risk–benefit ratio for NSAIDs. An association with a higher risk of anastomotic leakage after colorectal resection was first suggested more than a decade ago,123 later supported by mounting clinical and experimental evidence.124 Several large-scale studies and meta-analyses have since been published; some supported NSAID use as safe,125–128 while others found evidence of risk129,130 and were cautious.131 Some found that risk was substance-specific, with NSAIDs usually seen as more risky than COX-2-specific inhibitors.129,132–134 Together with goal-directed fluid therapy and delayed anastomoses in emergency surgery, avoidance of NSAID has been proposed to reduce the risk of leakage.135

The PROSPECT group follows these recommendations and does not recommend NSAID/COX-2-specific inhibitors for rectal surgery.

Neuromuscular blockade

The current recommendations list the evidence for deep neuromuscular blockade as ‘limited’, and this is a reflection of the current state of evidence, with both proponents and adversaries of deep blockade being able to make a good scientific argument, and evidence remaining equivocal.136 A collaboration of European researchers has started the EURO-RELAX trial, hoping to bring new large-scale evidence to the table (ClinicalTrials.gov registration NCT04124757).137

Intravenous lidocaine

Although there are a number of studies showing positive analgesic effects of intravenous lidocaine infusion, it should be noted that there has been a substantial shift in its contemporary clinical use. Initially met with enthusiasm following first studies showing impressive clinical results,138 lidocaine has seen a reframing of its indications with a meta-analyses in 2016 suggesting that it was most useful for laparoscopic and open abdominal surgery,139 but a subsequent analysis in 2018 was uncertain there was a relevant analgesic effect.113 More recent meta-analyses found that lidocaine may speed recovery of bowel function after elective colorectal surgery,140 but the degree of analgesia provided was not clinically relevant.140,141 Similarly, Sarakatsianou et al.142 in their meta-analysis, found no opioid-sparing effect of lidocaine infusions. A recent study suggested that lidocaine may need to be continued after surgery or postanaesthesia care unit (PACU) discharge to have a notable positive effect.143 Recently, several review articles have addressed the role of intravenous lidocaine infusion and consistently cautioned its use with regional analgesic techniques due to concerns of local anaesthetic systemic toxicity.144–146 The PROSPECT group entered into a comprehensive discussion on the question of intravenous lidocaine infusion, and consideration of the procedure-specific evidence; however, no consensus could be reached leading to no recommendation for its routine use for laparoscopic colorectal surgery.

Neuraxial (epidural/spinal) analgesia

In keeping with previous recommendations, our group does not recommend epidural analgesia and spinal analgesia.6 Peri-operative epidural analgesia has seen a declining trend in recent years,147 partly due to the emergence of viable alternatives such as truncal blocks and improved systemic analgesia,148 in part because of the realisation that it may not be necessary for pain following laparoscopic surgery,6 and in part because modern recovery protocols that emphasise fluid restriction, aggressive mobilisation and the avoidance of indwelling catheters, would argue against epidural analgesia in laparoscopic colorectal surgery. The question is, therefore, whether epidural analgesia has a net positive contribution, and, for this and many other types of surgery,149 the authors’ answer is ‘no’

The discussion amongst the PROSPECT Working Group was more lively around the topic of intrathecal opioids, where the evidence is much more equivocal than for epidural analgesia. All studies summarised in our recommendation showed analgesic benefits of morphine or diamorphine.55,61–63 In addition, both intrathecal morphine and diamorphine are cost-effective, widely available even in more resource-limited settings, and both a long and effective track-record spanning decades of clinical use.150 It should be noted that morphine and diamorphine are fundamentally different as their pharmacokinetic profile is concerned, but with adaptations in dosing, similar results in terms of clinical efficacy and adverse effects can be expected. Next to efficacy, the group also took into account frequent side-effects such as pruritus, nausea, and urinary retention, rare dose-dependent side-effects such as delayed respiratory depression,151,152 and the hypothetical risk associated with an additional spinal puncture in patients who will receive a general anesthetic by default. Given the lack of consensus, no recommendation could be made for the use of intrathecal morphine or diamorphine.

Truncal blocks

The topic which saw the most prolific growth of evidence was the introduction of truncal blocks to prevent postoperative pain after colorectal surgery. The evidence was, however, very equivocal, with only three64,65,69 of the eight studies64–71 comparing TAP block to placebo, sham or no intervention showing a clear analgesic benefit. Adequate baseline analgesia was only administered in three studies,64,67,69 while one study used port site infiltration.71 Similar confounded results were observed when TAP blocks were compared to other interventions such as wound infiltration, or epidural analgesia, or other blocks such as quadratus lumborum block. One area of contention was whether ultrasound-guided or laparoscopic blocks were more effective. Several studies suggest that surgeon-applied laparoscopically assisted injections were at least as effective as ultrasound-guided blocks,77–79 but given that the analgesic effect of truncal blocks is equivocal, the relevance of these findings is probably limited.

Due to the inconsistent procedure-specific evidence-base, our group does not recommend truncal blocks as a first-line treatment. Finally, we would like to highlight that the study by Dewinter et al. noted that 9 out of 50 patients receiving bilateral quadratus lumborum blocks reported metallic taste as a subjective early sign of local anaesthetic toxicity.53 This ties in with Rahiri et al.,153 who found that systemic local anaesthetic concentrations after single-dose bilateral truncal blocks can exceed commonly accepted thresholds of systemic toxicity in 9%.

Wound infiltration

Our previous recommendations did not list infiltration of the extraction wound among the recommended interventions. The two main studies, comparing continuous infiltration of the extraction wound to systemic analgesia showed neither an analgesic nor an opioid-sparing effect.88,89 Interestingly, although limited by the small number of patients included, Fustran et al.89 demonstrated a clear opioid-sparing and early analgesic effect of wound infiltration in a collective of laparotomy and laparoscopy patients, but on subgroup analysis, the effect was confined to the laparotomy patients. Another notable finding was that both intervention and control groups had low pain scores throughout the study period. Despite the inconsistent evidence, wound infiltration of laparoscopic port sites is recommended because of its simplicity and low costs. However, given the inconsistent evidence, continuous infiltration of the extraction site is not recommended after laparoscopic colorectal surgery.

Intraperitoneal local anaesthetics

IPLA has been recommended previously but not as first-line treatment.6 Given the equivocal results and the lack of studies having provided basic analgesia, the group notes the potential efficacy of IPLA, but the intervention is not recommended unless basic analgesia or intravenous lidocaine cannot be provided.

Nature of pain

Moving beyond a single pain score per time point, we need to differentiate further between pain at rest and dynamic pain, recognising that the latter is more relevant to early mobilisation and recovery. However, for laparoscopic colorectal surgery, we need to consider in addition that there are three distinct types of pain, which are distinct in location, intensity and time-course. First, there is the somatic pain from the port holes or hand-assisted incision, which in theory would respond best to regional anaesthesia. This type of pain, however, is light to moderate, and in most patients begins to fade by 6 to 8 h postoperatively.154 The second major source of pain is visceral pain, which for most patients is the main analgesic problem during the first 24 h. Thirdly, patients experience a variable, often low-level, referred pain in their shoulders, from diaphragmatic irritation due to the pneumoperitoneum.154 Future studies should incorporate these distinct types of pain into their assessment.

Limitations

This review is limited by the heterogeneous nature of the evidence, and specifically by the way that interventions are taken far down the road of evidence generation before more fundamental questions have been answered. We see this with truncal blocks, where despite a failure to firmly establish the procedure-specific evidence, some studies already investigate different pump settings for blocks, or compare different truncal blocks with each other without including a systemic analgesia control group. As suggested recently, it may be more helpful to channel our available resources to determine the general efficacy of an intervention first, before devoting ample resources to teasing out details of administration.155

In addition, there are several areas, which intersect with acute pain management. Firstly, preoperative optimisation and early recovery bundles of care have been shown to improve overall outcome,156 including, potentially, analgesic use.157 Secondly, recent evidence has underlined the considerable interindividual difference in pain susceptibility,158 which is influenced by readily identifiable risk factors such as preoperative opioid use,159 and also biopsychosocial factors that require specific testing to uncover.160 Thirdly, conversion of any laparoscopic abdominal surgery to open surgery will increase the complexity of postoperative pain management, increasing the need for more intense analgesic methods, such as multimodal analgesia, wound infiltration,161 or regional anaesthesia.64 Conversion risk is influenced by patient-related risk factors such as male sex, previous surgery of the abdomen, presence of adhesions, extremes of body mass index (high and low), by the proficiency of the surgeon, by procedural factors including indication (Crohn's disease) and site of resection (proctectomy more than colectomy) and by occurrence of complications such as bleeding.162–164 Fourthly, the spectrum of indications for laparoscopic colorectal surgery is wide, ranging from inflammatory diseases to malignant cancer. It is of note that patients suffering from Crohn's disease are thought to experience more intense acute postoperative pain than cancer patients and may require more intense analgesic management,165 but the currently available evidence does not allow for such detailed procedure-specific recommendations. Fifthl