Continuous intravenous perioperative lidocaine infusion for postoperative pain and recovery in adults

Stephanie Weibel; Yvonne Jelting; Nathan L Pace; Antonia Helf; Leopold HJ Eberhart; Klaus Hahnenkamp; Markus W Hollmann; Daniel M Poepping; Alexander Schnabel; Peter Kranke

doi:10.1002/14651858.CD009642.pub3

. 2018 Jun 4;2018(6):CD009642. doi: 10.1002/14651858.CD009642.pub3

Continuous intravenous perioperative lidocaine infusion for postoperative pain and recovery in adults

Stephanie Weibel ¹, Yvonne Jelting ¹, Nathan L Pace ², Antonia Helf ¹, Leopold HJ Eberhart ³, Klaus Hahnenkamp ⁴, Markus W Hollmann ⁵, Daniel M Poepping ⁶, Alexander Schnabel ¹, Peter Kranke ^1,^✉

Editor: Cochrane Anaesthesia Group

PMCID: PMC6513586 PMID: 29864216

Abstract

Background

The management of postoperative pain and recovery is still unsatisfactory in a number of cases in clinical practice. Opioids used for postoperative analgesia are frequently associated with adverse effects, including nausea and constipation, preventing smooth postoperative recovery. Not all patients are suitable for, and benefit from, epidural analgesia that is used to improve postoperative recovery. The non‐opioid, lidocaine, was investigated in several studies for its use in multimodal management strategies to reduce postoperative pain and enhance recovery. This review was published in 2015 and updated in January 2017.

Objectives

To assess the effects (benefits and risks) of perioperative intravenous (IV) lidocaine infusion compared to placebo/no treatment or compared to epidural analgesia on postoperative pain and recovery in adults undergoing various surgical procedures.

Search methods

We searched CENTRAL, MEDLINE, Embase, CINAHL, and reference lists of articles in January 2017. We searched one trial registry contacted researchers in the field, and handsearched journals and congress proceedings. We updated this search in February 2018, but have not yet incorporated these results into the review.

Selection criteria

We included randomized controlled trials comparing the effect of continuous perioperative IV lidocaine infusion either with placebo, or no treatment, or with thoracic epidural analgesia (TEA) in adults undergoing elective or urgent surgery under general anaesthesia. The IV lidocaine infusion must have been started intraoperatively, prior to incision, and continued at least until the end of surgery.

Data collection and analysis

We used Cochrane's standard methodological procedures. Our primary outcomes were: pain score at rest; gastrointestinal recovery and adverse events. Secondary outcomes included: postoperative nausea and postoperative opioid consumption. We used GRADE to assess the quality of evidence for each outcome.

Main results

We included 23 new trials in the update. In total, the review included 68 trials (4525 randomized participants). Two trials compared IV lidocaine with TEA. In all remaining trials, placebo or no treatment was used as a comparator. Trials involved participants undergoing open abdominal (22), laparoscopic abdominal (20), or various other surgical procedures (26). The application scheme of systemic lidocaine strongly varies between the studies related to both dose (1 mg/kg/h to 5 mg/kg/h) and termination of the infusion (from the end of surgery until several days after).

The risk of bias was low with respect to selection bias (random sequence generation), performance bias, attrition bias, and detection bias in more than 50% of the included studies. For allocation concealment and selective reporting, the quality assessment yielded low risk of bias for only approximately 20% of the included studies.

IV Lidocaine compared to placebo or no treatment

We are uncertain whether IV lidocaine improves postoperative pain compared to placebo or no treatment at early time points (1 to 4 hours) (standardized mean difference (SMD) −0.50, 95% confidence interval (CI) −0.72 to −0.28; 29 studies, 1656 participants; very low‐quality evidence) after surgery. Due to variation in the standard deviation (SD) in the studies, this would equate to an average pain reduction of between 0.37 cm and 2.48 cm on a 0 to 10 cm visual analogue scale . Assuming approximately 1 cm on a 0 to 10 cm pain scale is clinically meaningful, we ruled out a clinically relevant reduction in pain with lidocaine at intermediate (24 hours) (SMD −0.14, 95% CI −0.25 to −0.04; 33 studies, 1847 participants; moderate‐quality evidence), and at late time points (48 hours) (SMD −0.11, 95% CI −0.25 to 0.04; 24 studies, 1404 participants; moderate‐quality evidence). Due to variation in the SD in the studies, this would equate to an average pain reduction of between 0.10 cm to 0.48 cm at 24 hours and 0.08 cm to 0.42 cm at 48 hours. In contrast to the original review in 2015, we did not find any significant subgroup differences for different surgical procedures.

We are uncertain whether lidocaine reduces the risk of ileus (risk ratio (RR) 0.37, 95% CI 0.15 to 0.87; 4 studies, 273 participants), time to first defaecation/bowel movement (mean difference (MD) −7.92 hours, 95% CI −12.71 to −3.13; 12 studies, 684 participants), risk of postoperative nausea (overall, i.e. 0 up to 72 hours) (RR 0.78, 95% CI 0.67 to 0.91; 35 studies, 1903 participants), and opioid consumption (overall) (MD −4.52 mg morphine equivalents , 95% CI −6.25 to −2.79; 40 studies, 2201 participants); quality of evidence was very low for all these outcomes.

The effect of IV lidocaine on adverse effects compared to placebo treatment is uncertain, as only a small number of studies systematically analysed the occurrence of adverse effects (very low‐quality evidence).

IV Lidocaine compared to TEA

The effects of IV lidocaine compared with TEA are unclear (pain at 24 hours (MD 1.51, 95% CI −0.29 to 3.32; 2 studies, 102 participants), pain at 48 hours (MD 0.98, 95% CI −1.19 to 3.16; 2 studies, 102 participants), time to first bowel movement (MD −1.66, 95% CI −10.88 to 7.56; 2 studies, 102 participants); all very low‐quality evidence). The risk for ileus and for postoperative nausea (overall) is also unclear, as only one small trial assessed these outcomes (very low‐quality evidence). No trial assessed the outcomes, 'pain at early time points' and 'opioid consumption (overall)'. The effect of IV lidocaine on adverse effects compared to TEA is uncertain (very low‐quality evidence).

Authors' conclusions

We are uncertain whether IV perioperative lidocaine, when compared to placebo or no treatment, has a beneficial impact on pain scores in the early postoperative phase, and on gastrointestinal recovery, postoperative nausea, and opioid consumption. The quality of evidence was limited due to inconsistency, imprecision, and study quality. Lidocaine probably has no clinically relevant effect on pain scores later than 24 hours. Few studies have systematically assessed the incidence of adverse effects. There is a lack of evidence about the effects of IV lidocaine compared with epidural anaesthesia in terms of the optimal dose and timing (including the duration) of the administration. We identified three ongoing studies, and 18 studies are awaiting classification; the results of the review may change when these studies are published and included in the review.

Plain language summary

Intravenous infusion of lidocaine starting at the time of surgery for reduction of pain and improvement of recovery after surgery

Background

The most common problems immediately following surgery under general anaesthesia are pain, nausea and vomiting, delirium and slow or no movement of food through the digestive system. Opioid medications given to reduce postoperative pain may also be associated with nausea and constipation, also preventing a smooth recovery. It is of interest for patients and clinicians to reduce or prevent these complications leading to an early recovery so that patients can leave hospital earlier. One option for pain relief after surgery is epidural analgesia, where an opioid or local anaesthetic such as lidocaine is injected into the space surrounding the spinal cord. Not all patients may be suited to epidural analgesia, and so additional options such as intravenous non‐opioid analgesic medications that enable a rapid recovery are required.

The aim of this review was to assess the benefits and risks of intravenous infusion of lidocaine in patients undergoing various surgical procedures. Lidocaine is a medication used to numb tissue in a specific area.

Study characteristics

This review was published in 2015, and updated in 2017. We found 68 randomized controlled studies (RCTs), (clinical studies where people are randomly put into one of two or more treatment groups), with results from a total of 4525 participants. RCTs are used because they provide the most reliable evidence.

Intravenous lidocaine was compared with placebo or standard care in 66 of the studies, and with thoracic (chest area of spine) epidural analgesia in two studies. (A placebo is an inactive substance or procedure given to a participant in a medical trial to compare its effects with those of a real drug or other intervention). Lidocaine infusion was started during the surgery, before the first cut, and continued to at least the end of surgery. The included studies were moderately well conducted.

Key results

We are uncertain whether lidocaine infusion reduces pain, one to four hours after surgery when compared to placebo or usual care (29 studies, over 1600 participants). There was probably no difference in pain at 24 hours (33 studies, 1847 participants) and at 48 hours (24 studies, 1404 participants) between participants in the lidocaine and the placebo group. We are uncertain whether lidocaine infusion improves recovery of bowel function, with a reduction in the time to first defaecation or bowel movements (12 studies, 684 participants), and reduced risk of stopping the passage of food in the gut (4 studies, 273 participants). We are also uncertain whether lidocaine reduces postoperative nausea (35 studies, 1903 participants), and the requirement for opioids for pain relief (40 studies, 2201 participants). Only a limited number of studies systematically analysed adverse effects of intravenous lidocaine infusion. The side effects of intravenous lidocaine were unclear.

In the two studies that investigated intravenous lidocaine compared to epidural analgesia (102 participants), the effect on pain at 24 and 48 hours, and on the time to first bowel movement, remains unclear. The effect of lidocaine on the risk of stopping the passage of food in the gut and for postoperative nausea is also unclear, as only one small trial assessed these outcomes. Neither study investigated the effect on pain immediately after surgery, or on opioid consumption. Both studies looked at adverse effects associated with lidocaine, but the effect is uncertain.

Quality of the evidence

We rated the quality of evidence for most outcomes as very low. This was because of inconsistent findings across studies and the fact that the evidence came from small studies that were of moderate design quality or a limited number of studies. The quality of the evidence for minimal or no effect on pain at 24 and 48 hours was moderate quality. The studies involved a variety of surgical procedures. The dose of lidocaine used, and how long it was delivered for after the end of surgery, also varied between studies.

Summary of findings

Background

During the perioperative period alterations in haemodynamic, endocrine, metabolic and immune responses occur. Inflammatory processes are especially important for structural and functional wound repair. Conversely, excessive stimulation of the inflammatory response may lead to tissue damage (for example, reperfusion injury after cardiothoracic surgery), chronic postoperative pain, acute respiratory distress syndrome, systemic inflammatory response syndrome, and multiple organ failure. Typical and more common problems in the postoperative recovery are acute postoperative pain, postoperative nausea and vomiting (PONV), hypercoagulation, paralytic ileus, and postoperative cognitive dysfunction (Cassuto 2006). Fast‐track protocols aim to prevent or reduce these postoperative complications, facilitating early recovery. Evidence suggests that pain and paralytic ileus, causing a prolonged hospital stay, are major cost drivers in the postoperative period (Kehlet 2008).

Description of the condition

Local anaesthetics administered via the epidural route may reduce the catabolic stress response (Holte 2002); and provide sufficient pain therapy with a reduced need for opioids. Therefore, amongst other effects, an effective epidural analgesia can reduce the risk of developing a paralytic ileus, thus enabling enhanced recovery after surgery. However, recent research has demonstrated that systemic absorption of local anaesthetic, and not just drug interactions with the dorsal root ganglion neurons, also plays an important role in this protective action. Direct systemic (intravenous; IV) administration of lidocaine leading to low plasma levels in the range of 0.5 μg/mL to 5 μg/mL, which are comparable to concentrations after epidural administration, have been shown to achieve protective effects (Collinsworth 1974; Mayumi 1983).

Description of the intervention

Lidocaine (lignocaine), developed in 1948, is the first amino amide‐type short‐acting local anaesthetic. Originally it was used mainly via the IV route as an antiarrhythmic drug. Lidocaine has a very short half‐life and is therefore the local anaesthetic of choice for continuous IV administration. Drug accumulation because of delayed elimination due to hepatic or renal insufficiency might be a safety concern, limiting its usefulness in the perioperative setting. However, since only low plasma levels are required, in contrast to the therapy of chronic pain diseases like neuropathic pain syndromes, major complications following a continuous lidocaine infusion will not be expected.

How the intervention might work

Research in other pain entities, for example, peripheral neuropathic pain and complex regional pain syndromes, has shown that IV lidocaine administration produces prolonged analgesic effects (Kingery 1997). Inhibition of spontaneous impulse generation from injured peripheral nerves and dorsal root ganglions proximal to the injured fibres (Devor 1992a), as well as suppression of polysynaptic reflexes in the spinal dorsal horn (Woolf 1985), have been proposed as underlying mechanisms. Pain in the perioperative context is principally inflammatory pain, but could also be neuropathic or based on hyperalgesia. All these entities have been shown to be ameliorated by the administration of IV lidocaine (Koppert 2004). The anti‐inflammatory effects of local anaesthetic mediated through interactions with polymorphonuclear cells (Hollmann 2000a), and the inhibition of G protein‐coupled receptors (Hollmann 2000a; Hollmann 2001; Hollmann 2002; Hollmann 2005), may play a crucial role for the observed effects in the perioperative setting. Especially for the recovery of the gastrointestinal function and the prevention of the development of a paralytic ileus, which is thought to be the result of neurogenic inflammation, the anti‐inflammatory action of IV lidocaine can be beneficial (Herroeder 2007). Altogether, numerous clinically relevant outcomes may be influenced by IV administration of lidocaine; these include wound healing, analgesia, coagulation, postoperative cognitive dysfunction, paralytic ileus, and lung protection (Hollmann 2000b).

Why it is important to do this review

Epidural anaesthesia was once thought to be an anaesthetic strategy that improves outcomes after major surgery to a greater extent than has been confirmed recently (Popping 2014). However, recent evidence questions the risk‐benefit ratio for some patients and types of surgery (e.g. laparoscopic procedures, lower abdominal surgery or patients without pre‐existing lung disease). Serious neurologic complications after placement of an epidural catheter seem to occur more frequently than originally thought (Christie 2007; Cook 2009; Popping 2008). Thus, a growing number of patients and anaesthesiologists perform a proper risk‐benefit analysis in individual cases and also decide against epidural analgesia for some types of surgery except open thoracotomy and major abdominal surgery. This is notable with patients after coronary stenting, as they receive anticoagulant therapy and thus require careful assessment as to the risks and benefits of administering a regional anaesthetic technique. Nowadays anaesthesiologists are facing increasing numbers of these patients. In addition, for numerous types of surgery (e.g. surgeries involving the head), neuraxial techniques are not feasible at all. Therefore, alternative therapeutic interventions for optimal perioperative care are desirable. By characterizing the effects of IV lidocaine in the perioperative setting, lidocaine may be shown to offer a safe and alternative strategy for improving the perioperative outcome for patients unwilling or unable to receive epidural anaesthesia.

In spite of numerous preclinical studies in favour of systemic lidocaine, large published trials testing these effects in humans are not available. However, the number of rather small clinical studies has increased in recent years. Some of these trials have already been summarized in six systematic reviews (Chang 2017; Marret 2008; McCarthy 2010; Sun 2012; Vigneault 2011), including the original version of the current review (Kranke 2015). The current review update includes all new studies published until January 2017 and is the most comprehensive systematic review to date on the use of perioperative lidocaine for postoperative pain and recovery in adults.

Objectives

Methods

Criteria for considering studies for this review

Types of studies

We included all randomized controlled trials (RCTs) that evaluated the effect of perioperative lidocaine infusions versus no treatment, placebo treatment or versus epidural analgesia on relevant clinical outcomes.

We excluded cross‐over trials, since this study design is not relevant for the current review.

Types of participants

We included results obtained in adult (over 18 years) participants, independent of sex, undergoing any elective or urgent surgical procedure on any body part(s), and only if the procedure required general anaesthesia. Specifially, we excluded participants undergoing:

any kind of emergency procedure, and
minor surgical procedures, which are sometimes conducted using local or regional anaesthesia alone and do not provide a control event rate being high enough to demonstrate an effect of the investigated intervention.

Types of interventions

We included all studies comparing the effect of continuous perioperative lidocaine infusion, either with no treatment or placebo treatment, or with epidural analgesia. The IV lidocaine infusion must have been started intraoperatively (with or without an IV bolus) prior to incision and continued until the end of surgery. In trials of this intervention, standard care to enhance the postoperative recovery after surgery should also be provided.

Types of outcome measures

We analysed the following outcome measures.

Primary outcomes

Pain score at rest (0 to 10 cm, 0 to 100 mm visual analogue scale (VAS), numeric rating scale (NRS)), at 'early', 'intermediate', and 'late time points'
Gastrointestinal recovery: postoperative ileus (dichotomous), time to first defaecation/bowel movement (hours), time to first flatus (hours), and time to first bowel sounds (hours)
Adverse events (dichotomous; e.g. death, arrhythmias, other heart rate disorders or any sign of lidocaine toxicity)

Secondary outcomes

Length of hospital stay (inpatient ‐ days; outpatient ‐ minutes)
Functional postoperative neuropsychological status scales (e.g. quality of recovery (QoR) score or Mini Mental State Examination (MMSE))
Surgical complications (dichotomous; postoperative infections, thromboembolism, wound breakdown, etc.)
Patient satisfaction (0 to 10 cm VAS, 0 to 100 mm VAS, 0 to 10 NRS)
Cessation of the intervention (dichotomous; termination of the study before completion)
We investigated two separate outcomes for postoperative nausea and vomiting (PONV): First, postoperative nausea including PONV, if nausea was not separately reported in the study (referred to below as 'nausea') and, second, postoperative vomiting, both at 'early time points' (dichotomous; in postanaesthesia care unit (PACU)) and 'overall'
Intraoperative opioid consumption (remifentanil was separated from all other opioids due to an exceptional mode of action)
Postoperative opioid consumption, 'in PACU' and 'overall' (in mg morphine equivalents (MEQ)

When we reported 'early time points' for pain, nausea, vomiting, and opioid consumption, this referred to trials in which the outcome was reported approximately within the time period one to four hours postoperatively, or in the PACU. When we reported the 'intermediate' and 'late time points' for pain, this referred to pain ratings at 24 hours and 48 hours after surgery, respectively. In case of nausea, vomiting, and opioid consumption, 'overall' meant data that covered the time intervals from 0 to 24 hours, 0 to 48 hours, or 0 to 72 hours. We also accepted data that reported these outcomes for an interval from PACU (1 to 4 hours) to 24 hours, to 48 hours, or to 72 hours. If studies reported these outcomes for the 0 to 24‐hour time interval and later, we decided to analyse only the 0 to 24‐hour time interval. If studies did not explicitly report the time interval at which the outcome was documented, we grouped these data into the 'overall' category.

Search methods for identification of studies

Electronic searches

The search for the original review was performed in May 2014 (Kranke 2015), and the search for the update was performed in January 2017. We performed a further search in February 2018. We have added the February 2018 results to 'Studies awaiting classification' and we will incorporate them into the review at the next update.

We identified RCTs through literature searching with systematic and sensitive search strategies as outlined in Chapter 6.4 of the Cochrane Handbook for Systematic reviews of Interventions (Higgins 2011). We did not apply restrictions to language or publication status. We searched the following databases for relevant trials.

Cochrane Central Register of Controlled Trials (CENTRAL; 2017, Issue 1).
MEDLINE (Ovid SP, 1966 to 25 January 2017).
Embase (Ovid SP, 1980 to 25 January 2017).
CINAHL (EBSCO host, 1982 to 25 January 2017).

We developed a subject‐specific search strategy in MEDLINE and used that as the basis for the search strategies in the other databases listed. Where appropriate, we expanded the search strategy with search terms for identifying RCTs. All search strategies can be found in Appendix 1, Appendix 2, Appendix 3, Appendix 4.

We searched the American Society of Anesthesiologists (ASA) proceedings for relevant abstracts (16 March 2017).

We scanned the trial registry, Clinical Trials.gov for ongoing and unpublished trials to 16 March 2017 (ClinicalTrials.gov).

We developed the search strategy in consultation with the Information Specialist. We contacted researches in the field.

Searching other resources

We scanned the reference lists and citations of included trials and any relevant systematic reviews identified for further references to additional trials.

When necessary, we contacted trial authors for additional information.

Data collection and analysis

Selection of studies

Three review authors (original review: SW, JJ; update: SW, AH, YJ) independently scanned the titles retrieved by the initial search to exclude irrelevant trials. Then two review authors (original review: SW, JJ; update: AH, YJ) identified the studies that might be included in this review using a standardized study eligibility form developed by the authors (Appendix 5). If there were differences, we included a third review (original review: PK; update: SW) as arbiter. If necessary, we retrieved additional missing data and information about ongoing trials.

We resolved all differences by discussion among the authors. A PRISMA flow chart was prepared (Moher 2009).

Data extraction and management

Two authors (original review: SW as tandem with JJ, AS, LHJE, KH, DMP, MWH; update: AH, YJ) extracted the data using standardized data extraction forms developed by the authors (Appendix 6). If necessary, we retrieved additional data that were missing in published trials and information about ongoing trials by contacting the authors of the studies. We resolved all differences by discussion among the review authors at each step of data extraction.

Assessment of risk of bias in included studies

Two review authors (original review: SW, JJ; update: AH, YJ) independently performed the study quality assessment using a critical appraisal form provided by the Cochrane Anaesthesia, Critical and Emergency Care (ACE) Group with minor modifications (Appendix 7). We resolved any disagreements by discussion between the review authors, with a further review author acting as arbiter (original review: PK; update: SW).

We assessed the risk of bias of included studies using the Cochrane 'Risk of bias' tool (Higgins 2011). The standard domains include random sequence generation; allocation concealment; blinding of participants, personnel and outcome assessors; incomplete outcome data; selective reporting; and any other bias. Details of the risk of bias assessment were reported in the 'Critical Appraisal Form' (Appendix 7). We judged each component as being either low risk of bias, high risk of bias, or unclear. We included a 'Risk of bias' table as part of the table 'Characteristics of included studies' and a 'Risk of bias summary', which details all of the judgements we made for all included studies in the review.

Measures of treatment effect

For dichotomous outcomes, we obtained the risk ratio (RR) from the intervention and control group event rates. For continuous data, we obtained the mean difference (MD) from the difference between the intervention and control group mean values with associated standard deviations (SDs) if all studies measured data on the same scale. We used the standardized mean difference (SMD) when the studies all assessed the same outcome but measured it in a variety of ways (for example, studies measuring pain scores on different scales (visual analogue scale (VAS) 0 to 10 cm, 0 to 100 mm; numeric rating scale (NRS) 1 to 10)). We performed back‐transformation of SMD values into absolute values on a scale between 0 to 10 cm (VAS) to facilitate clinical interpretation. We used the smallest as well as the largest SD from the control groups of the pooled studies for back‐transformation (SMD * SD) to reflect the range of possible effects.

We transformed all opioid quantities into IV morphine equivalents (MEQ, mg) as described in the anatomic therapeutic chemical (ATC)/defined daily dose (DDD) Index (www.whocc.no/atc_ddd_index).

Unit of analysis issues

Multiple‐armed studies

We had planned to overcame a unit of analysis error for studies that contributed multiple comparisons by combining groups (by using the appropriate formula for adding SDs when required) to create a single pair‐wise comparison, if the presented data in the trials allow us to do so (Higgins 2011). Up to this update there were no studies with multiple comparisons of interest for this review.

Cluster‐randomized trials

We planned to include cluster‐randomized trials in the analyses along with individually‐randomized trials. However, for the present review we did not identify any relevant cluster‐randomized trials.

Dealing with missing data

If we encountered missing data, we contacted the relevant authors to obtain further information. If we obtained data, we included the data in the analyses. If data were missing, we included data in the analysis only on those participants whose results were known; we performed a complete‐case analysis. We subsequently excluded studies with incomplete reporting of their study flow or disputable exclusions in a sensitivity meta‐analysis to assess bias (Table 3). We considered the potential impact of the missing data on the results in the interpretation of the results of the review.

1. Sensitivity analyses ‐ risk of bias (incomplete outcome data).

		All studies		Without high/unclear risk of bias studies (incomplete outcome data)
Outcome	Statistical method	Studies	Effect estimate	Studies	Effect estimate
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	SMD (IV, Random, 95% CI)	29	−0.50 (−0.72 to −0.28)	17	−0.45 (−0.77 to −0.14)
Pain score, rest, 'intermediate time points' (24 hrs)	SMD (IV, Random, 95% CI)	33	−0.14 (−0.25 to −0.04)	18	−0.12 (−0.26 to 0.01)
Pain score, rest, 'late time points' (48 hrs)	SMD (IV, Random, 95% CI)	24	−0.11 (−0.25 to 0.04)	11	−0.06 (−0.27 to 0.15)
Postoperative ileus (dichotomous)	RR (MH, Random, 95% CI)	4	0.37 (0.15 to 0.87)	4	0.37 (0.15 to 0.87)
Time to first defaecation/bowel movement (hrs)	MD (IV, Random, 95% CI)	12	−7.92 (−12.71 to −3.13)	8	−7.5 (−14.38 to −0.63)
Time to first flatus (hrs)	MD (IV, Random, 95% CI)	13	−4.09 (−6.30 to −1.87)	9	−3.98 (−7.03 to −0.93)
Time to first bowel sounds (hrs)	MD (IV, Random, 95% CI)	2	−6.08 (−13.77 to 1.60)	2	−6.08 (−13.77 to 1.60)
Length of hospital stay (days)	MD (IV, Random, 95% CI)	32	−0.37 (−0.60 to −0.15)	17	−0.23 (−0.49 to 0.02)
Length of hospital stay (outpatient surgery, mins)	MD (IV, Random, 95% CI)	3	−10.81 (−36.93 to 15.31)	3	−10.81 (−36.93 to 15.31)
Surgical complications ‐ anastomotic leak	RR (MH, Random, 95% CI)	3	0.61 (0.08 to 4.80)	3	0.61 (0.08 to 4.80)
Surgical complications ‐ bleeding	RR (MH, Random, 95% CI)	3	1.79 (0.41 to 7.89)	3	1.79 (0.41 to 7.89)
Surgical complications ‐ postoperative infection	RR (MH, Random, 95% CI)	5	1.64 (0.41 to 6.52)	4	1.19 (0.25 to 5.67)
Patient satisfaction	MD (IV, Random, 95% CI)	6	0.76 (0.46 to 1.06)	2	0.59 (−0.09 to 1.26)
Postoperative nausea, early (PACU)	RR (MH, Random, 95% CI)	8	0.72 (0.53 to 0.98)	7	0.66 (0.47 to 0.91)
Postoperative nausea, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	RR (MH, Random, 95% CI)	35	0.78 (0.67 to 0.91)	19	0.87 (0.72 to 1.06)
Postoperative vomiting, early (PACU)	RR (MH, Random, 95% CI)	4	0.49 (0.16 to 1.48)	3	0.75 (0.15 to 3.80)
Postoperative vomiting, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	RR (MH, Random, 95% CI)	19	0.83 (0.63 to 1.08)	7	0.88 (0.58 to 1.31)
Intraoperative opioid consumption (MEQ, mg)	MD (IV, Random, 95% CI)	18	−2.14 (−3.87 to −0.40)	10	−1.52 (−4.13 to 1.09)
Intraoperative opioid consumption with remifentanil (MEQ, mg)	MD (IV, Random, 95% CI)	6	−14.17 (−35.27 to 6.92)	5	−16.08 (−41.41 to 9.25)
Postoperative opioid consumption, PACU (MEQ, mg)	MD (IV, Random, 95% CI)	21	−3.10 (−3.87 to −2.32)	12	−2.59 (−3.76 to −1.42)
Postoperative opioid consumption, overall (MEQ, mg)	MD (IV, Random, 95% CI)	40	−4.52 (−6.25 to −2.79)	25	−2.84 (−4.45 to −1.22)

		Mean + SD and median + IQR values		Only mean + SD values
Outcome	Statistical method	Studies	Effect estimate	Studies	Effect estimate
Pain score , rest, 'early time points' (1 hr to 4 hrs, PACU)	SMD (IV, Random, 95% CI)	29	−0.50 (−0.72 to −0.28)	23	−0.64 (−0.89 to −0.38)
Pain score , rest, 'intermediate time points' (24 hrs)	SMD (IV, Random, 95% CI)	33	−0.14 (−0.25 to −0.04)	27	−0.16 (−0.29 to −0.04)
Pain score , rest, 'late time points' (48 hrs)	SMD (IV, Random, 95% CI)	24	−0.11 (−0.25 to 0.04)	20	−0.12 (−0.29 to 0.04)
Time to first defaecation/bowel movement (hrs)	MD (IV, Random, 95% CI)	12	−7.92 (−12.71 to −3.13)	7	−6.03 (−10.98 to −1.08)
Time to first flatus (hrs)	MD (IV, Random, 95% CI)	13	−4.09 (−6.30 to −1.87)	10	−4.40 (−6.30 to −2.50)
Time to first bowel sounds (hrs)	MD (IV, Random, 95% CI)	2	−6.08 (−13.77 to 1.60)	2	−6.08 (−13.77 to 1.60)
Length of hospital stay (days)	MD (IV, Random, 95% CI)	32	−0.37 (−0.60 to −0.15)	16	−0.32 (−0.54 to −0.10)
Length of hospital stay (outpatient surgery, mins)	MD (IV, Random, 95% CI)	3	−10.81 (−36.93 to 15.31)	0	Not estimable
Patient satisfaction	MD (IV, Random, 95% CI)	6	0.76 (0.46 to 1.06)	1	0.30 (−0.21 to 0.81)
Intraoperative opioid consumption (MEQ, mg)	MD (IV, Random, 95% CI)	18	−2.14 (−3.87 to −0.40)	13	−2.32 (−4.33 to −0.32)
Intraoperative opioid consumption with remifentanil (MEQ, mg)	MD (IV, Random, 95% CI)	6	−14.17 (−35.27 to 6.92)	4	−20.45 (−52.10 to 11.19)
Postoperative opioid consumption, PACU (MEQ, mg)	MD (IV, Random, 95% CI)	21	−3.10 (−3.87 to −2.32)	15	−2.88 (−3.80 to −1.96)
Postoperative opioid consumption, overall (MEQ, mg)	MD (IV, Random, 95% CI)	40	−4.52 (−6.25 to −2.79)	28	−4.64 (−6.72 to −2.56)

Main meta‐analyses	Open abdominal surgery		Laparoscopic surgery		Other surgery		Test for subgroup difference (P)
Outcome	n	Effect estimate (I²)	n	Effect estimate (I²)	n	Effect estimate (I²)	Test for subgroup difference (P)
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	8	−0.54 (−0.82 to −0.26) (51%)	10	−0.78 (−1.34 to −0.21) (89%)	11	−0.21 (−0.44 to 0.02) (56%)	0.07
Pain score, rest, 'intermediate time points' (24 hrs)	No subgroup analysis performed (I² < 30%)
Pain score, rest, 'late time points' (48 hrs)	7	0.03 (−0.17 to 0.23) (0%)	7	−0.30 (−0.74 to 0.13) (74%)	10	−0.10 (−0.27 to 0.08) (20%)	0.35
Postoperative ileus (dichotomous)	No subgroup analysis performed (I² < 30%)
Time to first defaecation/bowel movement (hrs)	6	−7.09 (−10.33 to −3.86) (0%)	5	−6.23 (−18.07 to 5.62) (85%)	1	−6.10 (−24.49 to 12.29) NE	0.41
Time to first flatus (hrs)	6	−4.49 (−7.38 to −1.60) (6%)	5	−3.07 (−8.28 to 2.15) (78%)	2	−2.15 (−3.56 to −0.74) (0%)	0.36
Time to first bowel sounds (hrs)	1	−10.00 (−17.13 to −2.87) NE	1	−2.16 (−9.30 to 4.98) NE	0	NE	0.13
Length of hospital stay (days)	6	−0.59 (−0.99 to −0.18) (27%)	12	−0.15 (−0.58 to 0.28) (77%)	14	−0.48 (−0.84 to −0.11) (69%)	0.32
Length of hospital stay (outpatient surgery, mins)	0	NE	3	−10.81 (−36.93 to 15.31)	0	NE	NE
Surgical complications ‐ anastomotic leak	No subgroup analysis performed (I² < 30%)
Surgical complications ‐ bleeding	No subgroup analysis performed (I² < 30%)
Surgical complications ‐ postoperative infection	No subgroup analysis performed (I² < 30%)
Patient satisfaction	No subgroup analysis performed (I² < 30%)
Postoperative nausea, early (PACU)	No subgroup analysis performed (I² < 30%)
Postoperative nausea, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	No subgroup analysis performed (I² < 30%)
Postoperative vomiting, early (PACU)	No subgroup analysis performed (I² < 30%)
Postoperative vomiting, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	No subgroup analysis performed (I² < 30%)
Intraoperative opioid consumption (MEQ, mg)	7	−1.93 (‐4.61 to 0.75) (78%)	3	−0.71 (−7.95 to 6.53) (93%)	8	−2.03 (−4.14 to 0.07) (40%)	0.94
Intraoperative opioid consumption with remifentanil (MEQ, mg)	No subgroup analysis performed (I² < 30%)
Postoperative opioid consumption, PACU (MEQ, mg)	5	−3.03 (−4.82 to −1.23) (0%)	7	−3.84 (−4.57 to −3.11) (0%)	9	−2.66 (−4.19 to −1.13) (61%)	0.33
Postoperative opioid consumption, overall (MEQ, mg)	11	−3.56 (−6.76 to −0.35) (40%)	16	−4.85 (−7.46 to −2.23) (77%)	13	−5.54 (−9.35 to −1.72) (77%)	0.71

Main meta‐analyses	< 2 mg/kg/hr until end of surgery to PACU		≥ 2 mg/kg/hr until end of surgery to PACU		< 2 mg/kg/hr  for ≥ 24 hrs		≥ 2 mg/kg/hr  for ≥ 24 hrs		Test for subgroup  difference (P)
Outcome	n	Effect estimate (I²)	n	Effect estimate (I²)	n	Effect estimate (I²)	n	Effect estimate (I²)	Test for subgroup  difference (P)
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	8	−0.36 (−0.70 to −0.02) (67%)			21	−0.54 (−0.82 to −0.27) (82%)			0.42
Pain score, rest, 'intermediate time points' (24 hrs)	No subgroup analysis performed (I² < 30%)
Pain score, rest, 'late time points' (48 hrs)	5	−0.15 (−0.39 to 0.09) (0%)	13	−0.18 (−0.34 to −0.02) (12%)	5	0.03 (−0.45 to 0.51) (78%)	1	0.11 (−0.39 to 0.61) NE	0.66
Postoperative ileus (dichotomous)	No subgroup analysis performed (I² < 30%)
Time to first defaecation/bowel movement (hrs)	4	−7.06 (−11.37 to −2.75) (10%)	3	−7.27 (−13.54 to −1.00) (0%)	4	−6.97 (−20.09 to 6.16) (86%)	1	−20.00 (−50.62 to 10.62) NE	0.62
Time to first flatus (hrs)	4	−5.72 (−9.58 to −1.87) (28%)	4	−3.63 (−6.07 to −1.20) (64%)	4	−0.43 (−9.46 to 8.61) (84%)	1	−6.50 (−17.05 to 4.05) NE	0.65
Time to first bowel sounds (hrs)	1	−10.00 (−17.13 to −2.87) NE	1	−2.16 (−9.30 to 4.98) NE	0	NE	0	NE	0.13
Length of hospital stay (days)	7	−0.51 (−0.84 to −0.19) (2%)	16	−0.26 (‐0.50 to −0.03) (58%)	7	−0.25 (−1.04 to 0.54) (83%)	2	−1.29 (−4.47 to 1.89) (94%)	0.59
Length of hospital stay (outpatient surgery, mins)	1	−44.00 (−75.57 to −12.43) NE	2	−2.97 (−11.33 to 5.39) (0%)	0	NE	0	NE	0.01
Surgical complications ‐ anastomotic leak	No subgroup analysis performed (I² < 30%)
Surgical complications ‐ bleeding	No subgroup analysis performed (I² < 30%)
Surgical complications ‐ postoperative infection	No subgroup analysis performed (I² < 30%)
Patient satisfaction	No subgroup analysis performed (I² < 30%)
Postoperative nausea, early (PACU)	No subgroup analysis performed (I² < 30%)
Postoperative nausea, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	No subgroup analysis performed (I² < 30%)
Postoperative vomiting, early (PACU)	No subgroup analysis performed (I² < 30%)
Postoperative vomiting, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	No subgroup analysis performed (I² < 30%)
Intraoperative opioid consumption (MEQ, mg)	7	−0.68 (−3.51 to 2.15) (80%)			11	−3.11 (−5.74 to −0.47) (79%)			0.22
Intraoperative opioid consumption with remifentanil (MEQ, mg)	No subgroup analysis performed (I² < 30%)
Postoperative opioid consumption, PACU (MEQ, mg)	6	−3.55 (−5.43 to −1.67) (63%)			15	−3.02 (−3.86 to −2.18) (30%)			0.61
Postoperative opioid consumption, overall (MEQ, mg)	8	−2.33 (−5.05 to −0.13) (59%)	21	−7.41 (−10.91 to −3.91) (76%)	8	−2.88 (−6.25 to 0.49) (72%)	3	−3.90 (−10.18 to 2.38) (18%)	0.14

Outcome	Meta‐regression model (random‐effects model, tau² estimator: REML)	Open abdominal surgery		Laparoscopic surgery		Other surgery		Test of moderators (P)	Likelihood ratio test (P)
Outcome		n	Effect estimate (tau²)	n	Effect estimate (tau²)	n	Effect estimate (tau²)	Test of moderators (P)	Likelihood ratio test (P)
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	Individual tau²	8	−0.55 (−0.83 to −0.27) (0.08)	10	−0.78(−1.35 to −0.21) (0.76)	11	−0.21 to (−0.44 0.03) (0.09)	0.07	0.017
Pain score, rest, 'late time points' (48 hrs)	Individual tau²	7	0.03 (−0.17 to 0.23) (0.00)	7	−0.03 (−0.73 to 0.12) (0.24)	10	−0.10 (−0.27 to 0.08) (0.02)	0.34	0.049
Time to first defaecation/bowel movement (hrs)	Individual tau²	6	−8.16 (−12.44 to −3.87) (5.94)	5	−6.22 (−18.42 to 5.98) (127.06)	1	−6.10 (−24.49 to 12.29) (0.00)	0.94	0.097
Time to first flatus (hrs)	Individual tau²	6	−4.36 (−6.99 to −1.72) (0.00)	5	−2.55 (−9.31 to 4.21) (47.86)	2	−2.15 (−3.56 to −0.74) (0.00)	0.35	0.234
Length of hospital stay (days)	Individual tau²	6	−0.62 (−1.07 to −0.18) (0.11)	12	−0.16 (−0.66 to 0.33) (0.43)	14	−0.47 (−0.83 to −0.12) (0.18)	0.39	0.592
Intraoperative opioid consumption (MEQ, mg)	Individual tau²	7	−2.00 (−4.30 to 0.30) (4.34)	3	0.04 (−11.99 to 12.08) (107.18)	8	−1.86 (−3.34 to −0.38) (0.55)	0.95	0.027
Postoperative opioid consumption, PACU (MEQ, mg)	Individual tau²	5	−3.03 (−4.82 to −1.23) (0.00)	7	−3.84 (−4.57 to −3.11) (0.00)	9	−2.71 (−4.32 to −1.09) (3.45)	0.37	0.211
Postoperative opioid consumption, overall (MEQ, mg)	Individual tau²	11	−3.43 (−6.01 to −0.85) (3.08)	16	−5.78 (−9.33 to −2.23) (32.11)	13	−6.42 (−11.60 to −1.24) (50.54)	0.43	0.285

		Random‐effects model		Fixed‐effect model
Outcome	Statistical method	Studies	Effect estimate	Studies	Effect estimate
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	SMD (IV, Random, 95% CI)	29	−0.50 (−0.72 to −0.28)	29	−0.40 (−0.50 to −0.30)
Pain score, rest, 'intermediate time points' (24 hrs)	SMD (IV, Random, 95% CI)	33	−0.14 (−0.25 to −0.04)	33	−0.13 (−0.22 to −0.04)
Pain score, rest, 'late time points' (48 hrs)	SMD (IV, Random, 95% CI)	24	−0.11 (−0.25 to 0.04)	24	−0.09 (−0.19 to 0.02)
Postoperative ileus (dichotomous)	RR (MH, Random, 95% CI)	4	0.37 (0.15 to 0.87)	4	0.35 (0.15 to 0.82)
Time to first defaecation/bowel movement (hrs)	MD (IV, Random, 95% CI)	12	−7.92 (−12.71 to −3.13)	12	−6.01 (−8.53 to −3.49)
Time to first flatus (hrs)	MD (IV, Random, 95% CI)	13	−4.09 (−6.30 to −1.87)	13	−3.63 (−4.59 to −2.68)
Time to first bowel sounds (hrs)	MD (IV, Random, 95% CI)	2	−6.08 (−13.77 to 1.60)	2	−6.09 (−11.13 to −1.04)
Length of hospital stay (days)	MD (IV, Random, 95% CI)	32	−0.37 (−0.60 to −0.15)	32	−0.21 (−0.30 to −0.12)
Length of hospital stay (outpatient surgery, mins)	MD (IV, Random, 95% CI)	3	−10.81 (−36.93 to 15.31)	3	−5.66 (−13.74 to 2.43)
Surgical complications ‐ anastomotic leak	RR (MH, Random, 95% CI)	3	0.61 (0.08 to 4.80)	3	0.58 (0.08 to 4.24)
Surgical complications ‐ bleeding	RR (MH, Random, 95% CI)	3	1.79 (0.41 to 7.89)	3	1.86 (0.43 to 8.05)
Surgical complications ‐ postoperative infection	RR (MH, Random, 95% CI)	5	1.64 (0.41 to 6.52)	5	1.69 (0.53 to 5.33)
Patient satisfaction	MD (IV, Random, 95% CI)	6	0.76 (0.46 to 1.06)	6	0.76 (0.46 to 1.06)
Postoperative nausea, early (PACU)	RR (MH, Random, 95% CI)	8	0.72 (0.53 to 0.98)	8	0.72 (0.53 to 0.99)
Postoperative nausea, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	RR (MH, Random, 95% CI)	35	0.78 (0.67 to 0.91)	35	0.77 (0.68 to 0.88)
Postoperative vomiting, early (PACU)	RR (MH, Random, 95% CI)	4	0.49 (0.16 to 1.48)	4	0.51 (0.18 to 1.44)
Postoperative vomiting, overall (0 to 24 hrs, to 48 hrs, to 72 hrs)	RR (MH, Random, 95% CI)	19	0.83 (0.63 to 1.08)	19	0.78 (0.60 to 1.01)
Intraoperative opioid consumption (MEQ, mg)	MD (IV, Random, 95% CI)	18	−2.14 (−3.87 to −0.40)	18	−1.05 (−1.47 to −0.62)
Intraoperative opioid consumption with remifentanil (MEQ, mg)	MD (IV, Random, 95% CI)	6	−14.17(−35.27 to 6.92)	6	−13.68 (−33.53 to 6.17)
Postoperative opioid consumption, PACU (MEQ, mg)	MD (IV, Random, 95% CI)	21	−3.10 (−3.87 to −2.32)	21	−3.14 (−3.67 to −2.61)
Postoperative opioid consumption, overall (MEQ, mg)	MD (IV, Random, 95% CI)	40	−4.52 (−6.25 to −2.79)	40	−1.52 (−2.14 to −0.90)

		Without suspicious studies		With suspicious studies
Outcome	Statistical method	Studies	Effect estimate	Studies	Effect estimate
Pain score, rest, 'early time points' (1 hr to 4 hrs, PACU)	SMD (IV, Random, 95% CI)	29	−0.50 (−0.72 to −0.28)	37	−0.88 (−1.18 to −0.57)
Pain score, rest, 'intermediate time points' (24 hrs)	SMD (IV, Random, 95% CI)	33	−0.14 (−0.25 to −0.04)	41	−0.29 (−0.44 to −0.15)
Pain score, rest, 'late time points' (48 hrs)	SMD (IV, Random, 95% CI)	24	−0.11 (−0.25 to 0.04)	30	−0.22 (−0.40 to −0.03)
Time to first defaecation/bowel movement (hrs)	MD (IV, Random, 95% CI)	12	−7.92 (−12.71 to −3.13)	14	−7.09 (−10.06 to −4.11)
Time to first flatus (hrs)	MD (IV, Random, 95% CI)	13	−4.09 (−6.30 to −1.87)	16	−5.02 (−7.73 to −2.31)
Time to first bowel sounds (hrs)	MD (IV, Random, 95% CI)	2	−6.08 (−13.77 to 1.60)	4	−4.28 (−10.32 to 1.76)
Postoperative opioid consumption, PACU (MEQ, mg)	MD (IV, Random, 95% CI)	21	−3.10 (−3.87 to −2.32)	25	−3.51 (−4.88 to −2.15)
Postoperative opioid consumption, overall (MEQ, mg)	MD (IV, Random, 95% CI)	40	−4.52 (−6.25 to −2.79)	43	−4.81 (−6.55 to −3.07)

Study ID	Surgical procedure	Start infusion	End infusion	Duration of infusion	Bolus dose	Infusion dose	Total dose
Ahn 2015	Laparoscopic colectomy	2 mins before intubation	End of the operaton	216.60 mins (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Baral 2010	Abdominal surgery	30 mins before skin incision	1 hr after the end of surgery	157.80 min (infusion)	1.5 mg/kg	1.5 mg/kg/hr	NA
Bryson 2010	Abdominal hysterectomy	Prior to induction	Skin closure	105.0 mins (anaesthesia)	1.5 mg/kg	3 mg/kg/hr	NA
Cassuto 1985	Cholecystectomy	30 mins before skin incision	24 hrs postop	105 mins (surgery) + 30 mins (prior) + 24 hrs (postop)	100 mg	2 mg/min	NA
Chen 2015	Spine surgery	After induction of anaesthesia	End of surgery	129.2 mins (surgery)	1 mg/kg	1.5 mg/kg/hr	NA
Choi SJ 2012	Breast plastic surgeries	30 mins before skin incision	Skin closure	295 mins (surgery) + 30 mins (prior)	1.5 mg/kg	1.5 mg/kg/hr	NA
Choi GJ 2016	Elective total thyroidectomy	Prior to anaesthesia	End of surgery	135 mins (anaesthesia)	1.5 mg/kg	2 mg/kg/hr	NA
Choi KW 2016	Thyroidectomy	Immediately after induction	Extubation	148.9 mins (anaesthesia)	2 mg/kg	3 mg/kg/hr	NA
Cui 2010	Thoracic surgery	At induction	Skin closure	244 mins (anaesthesia)	No bolus	33 µg/kg/mins	NA
Dale 2016	Laparoscopic fundoplication	At induction	24 hrs after start of continuous infusion	24 hrs	1 mg/kg	2 mg/kg/hr	NA
De Oliveira 2012	Outpatient laparoscopic surgery	Prior to induction	End of the surgical procedure	105.5 mins (time of induction to skin incision)	1.5 mg/kg	2 mg/kg/hr	NA
De Oliveira 2014	Laparoscopic  bariatric surgery	Prior to induction	End of the surgical procedure	144 mins (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Dewinter 2016	Laparoscopic sterilisation in women	At induction	30 mins after arrival at PACU	77 mins	1.5 mg/kg	1.5 mg/kg/hr	240 mg
El‐Tahan 2009	Caesarean delivery	30 mins before induction	60 mins after skin closure	43.2 mins (anaesthesia) + 60 mins (postop)	1.5 mg/kg	1.5 mg/kg/hr	NA
Farag 2013	Spine surgery	At induction	Discharge from the PACU or a maximum of 8 hrs	8.5 hrs	No bolus	2 mg/kg/hr	NA
Grady 2012	Abdominal hysterectomy	At induction	24 hours postop	NA	1.5 mg/kg	2 mg/kg/hr	NA
Grigoras 2012	Surgery for breast cancer	Prior to induction	60 mins after skin closure	60.6 mins (surgery) + 60 mins (postop)	1.5 mg/kg	1.5 mg/kg/hr	328.1 mg
Groudine 1998	Radical retropubic prostatectomy	Prior to induction	60 mins after skin closure	NA	1.5 mg/kg	1.5 mg/kg/hr	NA
Herroeder 2007	Colorectal surgery	Prior to induction	4 hours  postop	194.3 mins (surgery) + NA (induction to skin incision) + 4 hrs (postop)	1.5 mg/kg	2 mg/mins	NA
Insler 1995	CABG	After induction of anaesthesia and before surgical incision	Up to 48 hours in the ICU  unless discharged earlier	NA	1.5 mg/kg	30 μg/kg/min	NA
Ismail 2008	Lumbar discectomy	30 mins before induction	Until 10 mins after extubation	NA	1.5 mg/kg	1.5 mg/kg	NA
Jain 2015	Laparoscopic cholecystectomy	10 mins prior to induction	End of first postop hr, max. 180 mins	NA	1.5 mg/kg	1.5 mg/kg/hr	NA
Kaba 2007	Laparoscopic colectomy	At induction	24 hrs postop	169 mins (anaesthesia) + 24 hrs (postop)	1.5 mg/kg	2 mg/kg/hr intraop and 1.33 mg/kg/h for 24 hrs  postop	NA
Kang 2011	Inguinal herniorrhaphy	2 mins before induction	End of the surgical procedure	66.03 mins (anaesthesia) + 2 mins (before induction)	1.5 mg/kg	2 mg/kg/hr	NA
Kasten 1986	CABG	2 mins before induction	NA	NA	3 mg/kg	0.05 mg/kg/min	NA
Kim TH 2011	Laparoscopic appendectomy	2 mins before induction	End of the surgical procedure	70.0 mins (anaesthesia) or 55.0 (surgery) + 2 mins (before induction)	1.5 mg/kg	2 mg/kg/hr	240.3 mg
Kim TH 2013	Laparoscopic gastrectomy	Preop	End of the surgical procedure	324 mins (anaesthesia) or 282.06 mins (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Kim HJ 2014	Coronary artery bypass graft	Before induction	24 hrs after end of surgery	339 mins (anaesthesia) + 24 hrs	1.5 mg/kg	2 mg/kg/hr	3917 mg
Kim HO 2014	Laparoscopic colectomy	Prior to incision	After 24 hrs	24 hrs	1 mg/kg	1 mg/kg/hr	NA
Kim KT 2014	Elective one‐level laminectomy and discectomy	Preop	End of surgery	110 min (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Koppert 2004	Major abdominal surgery	30 mins before skin incision	1 hr after the end of surgery	6.2 hrs (infusion)	1.5 mg/kg	5 mg/kg/hr	NA
Kuo 2006	Surgery for colon cancer	30 mins before surgery	End of the surgical procedure	157.8 mins (surgery) + 30 min (before surgery)	2 mg/kg	3 mg/kg/hr	NA
Lauwick 2008	Outpatient laparoscopic cholecystectomy	At induction	End of the surgical procedure	60 mins (surgery) + NA (induction to skin incision)	1.5 mg/kg	2 mg/kg/hr	NA
Lauwick 2009	Laparoscopic prostatectomy	At induction	End of the surgical procedure	262.5 mins (surgery) + NA (induction to skin incision)	1.5 mg/kg	2 mg/kg/hr	NA
Lee 2011	Off‐pump coronary artery bypass graft surgery	At induction	End of the surgical procedure	208.9 mins (surgery) + NA (induction to skin incision)	1.5 mg/kg	2 mg/kg/hr	NA
Maquoi 2016	Prostatectomy	Before induction	24 hrs postop	173 mins (anaesthesia) + 24 hrs	1.5 mg/kg	2 mg/kg/hr during surgery,  then 1.33 mg/kg/hr	NA
Martin 2008	Hip  arthroplasty	30 mins before skin incision	1 hr after the end of surgery	NA	1.5 mg/kg	1.5 mg/kg/hr	NA
Mathew 2009	Cardiac surgery	After induction	48 hrs postop	NA	1 mg/kg	4 mg/min for 1 hr,  2 mg/min for the second hr,  1 mg/min for the rest	NA
McKay 2009	Outpatient surgery	After induction	1 hr after arrival in the PACU	NA	1.5 mg/kg	2 mg/kg/hr	517 mg
Mitchell 1999	Cardiac surgery	At induction	48 hrs postop	NA	1 mg/kg	240 mg over the first hr and  120 mg over the second hr,  and then 60 mg/h thereafter if the patient  was receiving lidocaine	NA
Mitchell 2009	Cardiac surgery	At induction	Total  of 12 hours	NA	1 mg/kg	2 mg/min for 2 hrs, and 1 mg/min thereafter	NA
Oliveira 2015	Hysterectomy	At induction	End of surgery	145.1 mins (anaesthesia)	No bolus	2 mg/kg/hr	NA
Omar 2013	Functional endoscopic sinus surgery	After induction	End of the surgical procedure	87 mins (anaesthesia) or 62 mins (surgery)	1.5 mg/kg	1.5 mg/kg/hr	NA
Ortiz 2016	Laparoscopic cholecystectomy	Before incision	1 hr after end of surgery	105.23 mins (surgery) + 1 hr	1.5 mg/kg	3 mg/kg/hr	NA
Peng 2016	Supratentorial tumour surgery	After induction	End of surgery	254 mins (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Rimbäck 1990	Cholecystectomy	Prior to induction	24 hrs postop	109 mins (surgery) + NA (induction to skin incision) + 24 hrs (postop)	100 mg	3 mg/min	NA
Saadawy 2010	Laparoscopic cholecystectomy	Prior to induction	End of the surgical procedure	80.3 mins (surgery) + NA (induction to skin incision)	2 mg/kg	2 mg/kg/hr	NA
Samimi 2015	Abdominal hysterectomy	30 mins before incision	1 hr after surgery	30 mins + 95 min (surgery) + 60 mins	1.5 mg/kg	2 mg/kg/hr	NA
Slovack 2015	VATS	At induction	End of the surgical procedure	NA	1.5 mg/kg	3 mg/min if the  patient’s total body weight was more than 70 kg or 2 mg/min if weight was less than 70  kg	239.6 mg
Soltani 2013	Ophthalmologic surgeries	NA	Intraoperatively		No bolus	2.5 mg/kg/hr
Sridhar 2015	Open abdominal surgery	Time of intubation	1 hr after surgery	145.8 mins (surgery) + 60 min	1.5 mg/kg	1.5 mg/kg/hr	NA
Staikou 2014	Large bowel surgery	Before induction	Before skin suturing	122 mins	1.5 mg/kg	2 mg/kg/hr	NA
Striebel 1992	Tonsillectomy	30 mins before skin incision	24 hrs	57 mins (surgery) + 30 min (before skin incision) + 24 hrs (postop)	1.5 mg/kg	2 mg/kg/hr over 6 hrs and  0.5 mg/kg/hr for another 18 hrs	NA
Swenson 2010	Colon resection	Prior to induction	Until the day after return of bowel function or fifth postop  day	69 hrs 54 mins (infusion)	No bolus	11 patients: 2 mg/min in patients < 70 kg,  3 mg/min in patients > 70 kg,  and 11 patients: 1 mg/min in patients < 70 kg,  2 mg/min in patients > 70 kg	NA
Terkawi 2014	Breast cancer surgery	Before induction	2 hrs after arrival in PACU or at discharge from PACU	85 mins	1.5 mg/kg, max. 150 mg	2 mg/kg/h, max 200 mg/hr	NA
Tikuisis 2014	Laparoscopic colon resection	Prior to induction	24 hrs postop	115 mins (anaesthesia) + 24 hrs (postop)	1.5 mg/kg	2 mg/kg/hr during surgery, 1 mg/  kg/hr for 24 hrs	NA
Wallin 1987	Cholecystectomy	30 mins before skin incision	24 hrs postop	110 mins (surgery) + 30 min (before skin incision) + 24 hrs	100 mg	2 mg/min	NA
Wang 2002	CABG	At the opening of the  pericardium	End of the surgical procedure	NA	1.5 mg/kg, second dose (4 mg/kg) was administered to the priming solution  of CPB	4 mg/min	NA
Wang 2015	Hysterectomy	10 mins prior to induction	Discharge from the operating room	152.3 (anaesthesia) + 10 mins	1.5 mg/kg	1.5 mg/kg/hr	NA
Weinberg 2016	Radical retropubic prostatectomy	Before induction	End of surgery	NA	1.5 mg/kg	1.5 mg/kg/hr	NA
Wongyingsinn 2011	Laparoscopic colorectal surgery	Prior to induction	48 hrs postop	220 mins (surgery) + NA (induction to skin incision) + 48 (postop)	1.5 mg/kg, max: 100 mg	2 mg/kg/hr during surgery,  1 mg/kg/hr for 48 hrs	NA
Wu 2005	Laparoscopic cholecystectomy	30 mins before skin incision	End of the surgical procedure	81.4 mins (surgery) + 30 mins (before incision)	No bolus	3 mg/kg/hr	NA
Wuethrich 2012	Laparoscopic transperitoneal renal surgery	At induction	24 hrs postop	293 mins (anaesthesia) + 24 hrs (postop)	1.5 mg/kg	2 mg/kg/hr during surgery,  1.3 mg/kg/hr for 24 hrs	NA
Xu 2017	Abdominal hysterectomy	10 mins before induction	Wound closure	118.7 mins (anaesthesia)	1.5 mg/kg	1.5 mg/kg/hr	185.7 mg
Yang 2014	Laparoscopic  cholecystectomy	2 mins before induction	End of the surgical procedure	65 mins (anaesthesia) + 2 mins (before induction)	1.5 mg/kg	2 mg/kg/hr	228.71 mg
Yardeni 2009	Transabdominal hysterectomy	20 mins before skin incision	End of the surgical procedure	109 mins (surgery) + 20 mins (before skin incision)	2 mg/kg	1.5 mg/kg/hr	NA
Yon 2014	Subtotal gastrectomy	Preop (protocol: 2 mins before intubation)	End of surgery	271.27 mins (surgery)	1.5 mg/kg	2 mg/kg/hr	NA
Zengin 2015	Laparotomy	At induction	Wound closure	114.1 mins (surgery)	1.0 mg/kg	2 mg/kg/hr	NA

IV Lidocaine compared to placebo or no treatment in patients undergoing any elective or urgent surgical procedure under general anaesthesia
Patient or population: adult patients undergoing any elective or urgent surgical procedure under general anaesthesia  Settings: Asia (24 trials); USA, Canada, and South America (18 trials); Europe (15 trials); Middle East (7 trials); New Zealand and Australia (4 trials)  Intervention: IV lidocaine  Comparison: placebo or no treatment
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	Prediction interval  (95% PI)	No. of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk with placebo or no treatment	Corresponding risk with IV lidocaine
1. Pain (VAS 0 to 10 cm, 0 to 100 mm, NRS 0 to 10)	Pain score at rest, 'early time points' (1 hto 4 hpostoperatively, or in the PACU)		‐	(1.61 lower to 0.62 higher)	1656  (29 RCTs)	⊕⊝⊝⊝  Very low^a,b,c,d	A SMD of 0.50 fewer is equivalent to a range of 0.37 cm fewer (SD = 0.74 cm) to 2.48 cm fewer (SD = 4.95 cm) on a VAS 0 to 10 cm scale in the intervention group. The range of mean effects that can be expected in a future study (95% PI) includes both benefit and harm^k.
		The standardized mean pain score in the intervention group was 0.50 lower (0.72 lower to 0.28 lower)
	Pain score at rest, 'intermediate time points' (24 hpostoperatively)		‐	(0.44 lower to 0.16 higher)	1847  (33 RCTs)	⊕⊕⊕⊝  Moderate^e,f,g,h,i	A SMD of 0.14 fewer is equivalent to a range of 0.10 cm fewer (SD = 0.74 cm) to 0.48 cm fewer (SD = 3.42 cm) on a VAS 0 to 10 cm scale in the intervention group. The range of mean effects that can be expected in a future study (95% PI) is clinically not relevant^k.
		The standardized mean pain score in the intervention group was 0.14 lower (0.25 lower to 0.04 lower)
	Pain score at rest, 'late time points' (48 hpostoperatively)		‐	(0.60 lower to 0.38 higher)	1404  (24 RCTs)	⊕⊕⊕⊝  Moderate^e,f,g,h,i	A SMD of 0.11 fewer is equivalent to a range of 0.08 cm fewer (SD = 0.7 cm) to 0.42 cm fewer (SD = 3.8 cm) on a VAS 0 to 10 cm scale in the intervention group. The range of mean effects that can be expected in a future study (95% PI) is clinically not relevant^k.
		The standardized mean pain score in the intervention group was 0.11 lower (0.25 lower to 0.04 higher)
2. Gastrointestinal recovery	Postoperative ileus (dichotomous) The number of participants with postoperative ileus		RR 0.37 (0.15 to 0.87)	(0.05 lower to 2.43 higher)	273  (4 RCTs)	⊕⊝⊝⊝  Very low^a,b,c	The range of mean effects that can be expected in a future study (95% PI) includes both benefit and harm.
	131 per 1000	48 per 1000 (20 to 114)
	Time to first defaecation/bowel movement (h)		‐	(22.19 h shorter to 6.36 h longer)	684  (12 RCTs)	⊕⊝⊝⊝  Very low^a,b,c	The range of mean effects that can be expected in a future study (95% PI) includes both benefit and clinical non‐relevance.
	The mean time to first defaecation/bowel movement in the control group ranged from 24 h to 94 h	The mean time to first defaecation/bowel movement in the intervention group was 7.92 h shorter (12.71 h shorter to 3.13 h shorter)
3. Adverse events (e.g. the number of participants that died, or had arrhythmias, other heart rate disorders, or showed any signs of lidocaine toxicity)	See comment	See comment	‐	‐	See comment	⊕⊝⊝⊝  Very Low^j	Adverse events that were investigated in a few trials are death, arrhythmia, light‐headedness, perioral numbness, and dizziness. The effect of lidocaine on these adverse effects is uncertain.
4. Postoperative nausea, 'overall' (0 to 24 h, to 48 h, to 72 h)	350 per 1000	273 per 1000 (235 to 319)	RR 0.78 (0.67 to 0.91)	(0.49 lower to 1.23 higher)	1903  (35 RCTs)	⊕⊝⊝⊝  Very low^a,b,c,d	The range of mean effects that can be expected in a future study (95% PI) includes both benefit and clinical non‐relevance.
5. Postoperative opioid consumption, 'overall' (MEQ, mg)	The mean postoperative opioid consumption in the control group ranged from 1.13 mg to 233.93 mg	The mean postoperative opioid consumption in the intervention group was 4.52 mg lower (6.25 mg lower to 2.79 mg lower)	‐	(12.03 mg lower to 3.00 mg higher)	2201 (40 RCTs)	⊕⊝⊝⊝  Very low^a,b,c,d	The range of mean effects that can be expected in a future study (95% PI) includes both benefit and clinical non‐relevance.
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group* and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; MEQ: morphine equivalents; NRS: numeric rating scale; PACU: postanaesthesia care unit; RCT: randomized controlled trial; RR: risk ratio; SMD: standardized mean difference; SD: standard deviation; VAS: visual analogue scale
GRADE Working Group grades of evidence  High quality: we are very confident that the true effect lies close to that of the estimate of the effect.  Moderate quality: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.  Low quality: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.  Very low quality: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

IV lidocaine compared to TEA in adult patients undergoing any elective or urgent surgical procedure under general anaesthesia
Patient or population: adult patients undergoing any elective or urgent surgical procedure under general anaesthesia  Settings: USA and Canada (two trials)  Intervention: IV lidocaine  Comparison: TEA
Outcomes	*Illustrative comparative risks (95% CI)**		Relative effect (95% CI)	Prediction interval  (95% PI)	No. of Participants  (studies)	Quality of the evidence  (GRADE)	Comments
	Assumed risk with placebo or no treatment	Corresponding risk with IV lidocaine
1. Pain (VAS 0 to 10 cm, 0 to 100 mm, NRS 0 to 10	Pain score at rest, 'early time points' (1 hto 4 hpostoperatively, or in the PACU)		‐	‐	(0 RCTs)	‐	No trial assessed this outcome.
	See comment	See comment
	Pain score at rest, 'intermediate time points' (24 hpostoperatively) (VAS 0 to 10 cm)		‐	Not estimable*	102  (2 RCTs)	⊕⊝⊝⊝  Very low^a,b,c	The estimated effect (95% CI) includes both benefit and harm^g.
	The mean pain score 'intermediate time points' ranged across control groups from 0 to 3.3 cm	The mean pain score 'intermediate time points' in the intervention group was 1.51 cm higher (0.29 lower to 3.32 higher)
	Pain score at rest 'late time points' (48 hpostoperatively) (VAS 0 to 10 cm)		‐	Not estimable*	102  (2 RCTs)	⊕⊝⊝⊝  Very low^a,b,c	The estimated effect (95% CI) includes both benefit and harm^g.
	The mean pain score 'late time points' ranged across control groups from 0 to 2.7 cm	The mean pain score 'late time points' in the intervention group was 0.98 cm higher (1.19 lower to 3.16 higher)
2. Gastrointestinal recovery	Postoperative ileus (dichotomous) The number of participants with postoperative ileus		Not estimable	Not estimable*	60 (1 RCT)	⊕⊝⊝⊝  Very low^a,d	Only one small trial assessed this outcome.
	Two out of 30 participants in the control group and one out of 30 in the lidocaine group had postoperative ileus
	Time to bowel movements (h)		‐	Not estimable*	102  (2 RCTs)	⊕⊝⊝⊝  Very low^a,e	The estimated effect (95% CI) includes both benefit and harm.
	The mean time to first bowel movements (h) ranged across control groups from 39 h to 72 h	The mean time to first bowel movements (h) in the intervention group was 1.66 h shorter (10.88 shorter to 7.56 longer)
3. Adverse events (e.g. the number of participants that died, or had arrhythmias, other heart rate disorders, or showed any signs of lidocaine toxicity)	See comment	See comment	‐	‐	See comment	⊕⊝⊝⊝  Very low^f	All adverse events that are reported in the individual studies were listed in Table 14.
4. Postoperative nausea, 'overall' (0 to 24 h, to 48 h, to 72 h)	17 out of 30 participants in the control group and 11 out of 30 in the lidocaine group had nausea		Not estimable	Not estimable*	60 (1 RCT)	⊕⊝⊝⊝  Very low^a,d	Only one small trial assessed this outcome.
5. Postoperative opioid consumption, 'overall' (MEQ, mg)	See comment	See comment	‐	‐	(0 RCTs)	‐	No trial assessed this outcome.
The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group* and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MEQ: morphine equivalents; NRS: numeric rating scale; PACU: postanaesthesia care unit; RCT: randomized controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence  High quality: we are very confident that the true effect lies close to that of the estimate of the effect.  Moderate quality: we are moderately confident in the effect estimate; the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.  Low quality: our confidence in the effect estimate is limited; the true effect may be substantially different from the estimate of the effect.  Very low quality: we have very little confidence in the effect estimate; the true effect is likely to be substantially different from the estimate of effect.

Study ID	Type of adverse event/side effect ‐ lidocaine group	Type of adverse event/side effect ‐ control group	No adverse events/side effects detectable (statement)
Ahn 2015	NA	NA	“Not one patient had a postoperative complication related to lidocaine infusion.”
Baral 2010	Light headache (3), cardiac arrhythmias (0), perioral numbness (0), hypotension (0)	Cardiac arrhythmias (0), perioral numbness (0), hypotension (0)	NA
Bryson 2010	Light‐headedness, tinnitus, dysgeusia, etc. (11)	Light‐headedness, tinnitus, dysgeusia, etc. (21)	NA
Cassuto 1985	Light‐headedness (1)	Light‐headedness (1)	"No adverse reactions to lidocaine were observed."
Chen 2015	NA	NA	NA
Choi SJ 2012	Dizziness 24 hrs/ 48 hrs/ 72 hrs (1/1/1), itching 24 hrs/ 48 hrs/ 72 hrs (8/3/0), respiratory repression 24 hrs/ 48 hrs/ 72 hrs (0/0/0)	Dizziness 24 hrs/ 48 hrs/ 72 hrs (3/2/1), itching 24 hrs/ 48 hrs/ 72 hrs (6/2/1), respiratory repression 24 hrs/ 48 hrs/ 72 hrs (0/0/0)	NA
Choi GJ 2016	NA	NA	“There were no adverse events related to the investigational procedure or systemic administration of lidocaine such as arrhythmia, perioral numbness, visual disturbance, metal taste, or light‐headedness.”
Choi KW 2016	NA	NA	"In addition, none of the patients showed symptoms or signs associated with lidocaine toxicity during the perioperative period."
Cui 2010	Drowsiness (0), metal taste (0), perioral numbness (0), visual disturbances (0)	Drowsiness (0), metal taste (0), perioral numbness (0), visual disturbances (0)	"No patient reported any side effect of lidocaine toxicity."
Dale 2016	Severe bradycardia (1), perioral paraesthesia (1), restless legs (1)	Severe bradycardia (0), perioral paraesthesia (0), restless legs (0)	NA
De Oliveira 2012	NA	NA	"We did not observe any potential cardiovascular or neurological side effects associated with the infusion of systemic lidocaine in our investigation."
De Oliveira 2014	NA	NA	"We did not observe any potential cardiovascular or neurological side effects associated with the infusion of systemic lidocaine in our investigation."
Dewinter 2016	NA	NA	“Both groups did not differ with respect to the incidence of other AEs. No patient receiving lidocaine reported subjective symptoms of local anesthetic systemic toxicity.”
El‐Tahan 2009	Perioperative arrhythmia (0), light‐headedness (0), headache (0), perioral numbness (0), tunnel vision (0), seizures (0)	Perioperative arrhythmia (0), light‐headedness (0), headache (0), perioral numbness (0), tunnel vision (0), seizures (0)	"There were no reported serious side effects during the study."
Farag 2013	Pneumonia (0), respiratory failure (0), cardiac arrest (0), arrhythmia (0), heart failure (0), stroke (0), intravascular coagulopathy (0), thromboembolism (0), delirium (0), monoplegia (0), upper gastrointestinal bleeding (0), sepsis (0), readmission (2)	Pneumonia (0), respiratory failure (0), cardiac arrest (0), arrhythmia (0), heart failure (0), stroke (0), intravascular coagulopathy (0), thromboembolism (0), delirium (0), monoplegia (1), upper gastrointestinal bleeding (0), sepsis (0), readmission (3)	NA
Grady 2012	NA	NA	NA
Grigoras 2012	NA	NA	"No side effects related to lidocaine were observed."
Groudine 1998	NA	NA	"No patient experienced identifiable adverse events related to the lidocaine infusion."
Herroeder 2007	NA	NA	NA
Insler 1995	Death (1), myocardial infarction (0)	Death (1), myocardial infarction (0)	NA
Ismail 2008	NA	NA	NA
Jain 2015	Drowsiness (3)	Drowsiness (0)	“None of the patients complained of lignocaine‐related side effects such as perioral numbness or metallic taste. The incidence of light‐headedness and nausea was comparable in both the groups. Three patients in Group B demonstrated drowsiness in the postoperative period lasting between 10 and 17 mins.”
Kaba 2007	NA	NA	NA
Kang 2011	NA	NA	NA
Kasten 1986	NA	NA	NA
Kim TH 2011	NA	NA	"There was no adverse effect from intravenous lidocaine throughout the study."
Kim TH 2013	NA	NA	"In our study, no neuropsychiatric events were observed throughout the process."
Kim HJ 2014	NA	NA	“No specific complication or side effect regarding lidocaine or dexmedetomidine was reported.”
Kim HO 2014	Hospital mortality (0)	Hospital mortality (0)	“There were no significant lidocaine‐related adverse events during our trial.”
Kim KT 2014	NA	NA	“There were no side effects from the lidocaine, such as arrhythmia, hypotension, and hypersensitivity.”
Koppert 2004	NA	NA	"No anaesthesiologist noted adverse events related to the lidocaine infusion during surgery. Furthermore, no patient after having regained consciousness complained of lidocaine‐related side effects such as perioral numbness or metallic taste. The incidences of drowsiness, light‐headedness, and nausea were comparable in the lidocaine and control groups."
Kuo 2006	Bradycardia (3)	Bradycardia (0)	"No patient experienced an identifiable adverse event related to IV lidocaine infusion."
Lauwick 2008	NA	NA	NA
Lauwick 2009	Bleeding (1), sepsis (1), chest infection (1)	Bladder leakage (1)	NA
Lee 2011	Atrial fibrillation (9), other arrhythmia (7), myocardial infarction (0), death (0)	Atrial fibrillation (5), other arrhythmia (10), myocardial infarction (0), death (0)	"All patients started on lidocaine completed their full course of drug and did not experience any adverse events related to the local anaesthetic, such as severe bradycardia (< 40 beats min^‐1), asystole, or neurological symptoms."
Maquoi 2016	NA	NA	NA
Martin 2008	NA	NA	"No patient reported lidocaine toxicity side effects and no adverse events were reported in both groups"
Mathew 2009	Serious adverse events (12.3%), no detailed description	Serious adverse events (10.2%), no detailed description	"Adverse events were not significantly different between treatment groups."
McKay 2009	Dizziness and visual disturbances (1)	NA	"There were no serious adverse events recorded."
Mitchell 1999	Death (1)	Death (1)	NA
Mitchell 2009	Death due to multiorgan failure (3) and acute graft occlusion (1)	Death (0)	NA
Oliveira 2015	NA	NA	NA
Omar 2013	Hypotension (0)	Hypotension (0)	NA
Ortiz 2016	NA	NA	"There was no arrhythmia or adverse effect occurrence."
Peng 2016	Hypertension (3), coronary heart disease (0)	Hypertension (4), coronary heart disease (0)	“There were no seizures or other symptoms of potential lidocaine toxicity found in patients who received lidocaine infusion. There was no significant difference in the number of cases complicated by hypertension, tachycardia, dysphoria, or PONV between the normal saline group and the lidocaine group.”
Rimbäck 1990	Sedation (2)	NA	"No adverse reactions to lidocaine were reported."
Saadawy 2010	NA	NA	NA
Samimi 2015	NA	NA	"...also none of the patients experienced lidocaine‐related adverse effects."
Slovack 2015	Confusion (1), sedation (2), light‐headedness/dizziness (0), blurred vision (0), hypotension (0), respiratory depression (0), pruritus (0)	Confusion (0), sedation (0), light‐headedness/dizziness (0), blurred vision (1), hypotension (1), respiratory depression (0), pruritus (0)	NA
Soltani 2013	NA	NA	NA
Sridhar 2015	NA	NA	NA
Staikou 2014	Transient confusion in PACU (1), bradycardia requiring treatment (0)	Transient confusion in PACU (0), bradycardia requiring treatment (0)	NA
Striebel 1992	NA	NA	No signs of urticaria, dermatitis, asthma bronchiale, anaphylactic shock, restlessness, anxiety, lalopathy, tinnitus, metallic taste, dizziness, visual disturbance, and tremor.
Swenson 2010	Wound infection (0), anaemia (1), anxiety (1), supraventricular tachycardia (3), back pain (0), bradycardia (0), confusion (2), decreased oxygen saturation level (1), dizziness/light‐headedness (1), fever (1), hyperglycaemia (3), hypertension (3), itching (3), lower extremity numbness (1), intravascular device infection (0), syncope (0), arrhythmia severe (1), confusion severe (1), facial numbness severe (1), shortness of breath (1)	Wound infection (1), anaemia (1), anxiety (0), supraventricular tachycardia (1), back pain (1), bradycardia (1), confusion (0), decreased oxygen saturation level (0), dizziness/light‐headedness (1), fever (1), hyperglycaemia (0), hypertension (0), itching (3), lower extremity numbness (6), intravascular device infection (1), syncope (1), arrhythmia severe (1), confusion severe (0), facial numbness severe (0), shortness of breath (0)	NA
Terkawi 2014	NA	NA	"...no toxicity cases were reported in our cohort...."
Tikuisis 2014	Light‐headedness (0), perioral numbness (0), metallic taste (0), dizziness (0), and visual disturbances (0)	NA	"Lidocaine‐associated haemodynamic changes such as severe hypotension, bradycardia, and arrhythmia were not observed in any lidocaine group patient during surgery."
Wallin 1987	Drowsiness (2)	NA	"Aside from drowsiness in two patients of the lidocaine group, no side effects due to possible lidocaine overdosage were reported."
Wang 2002	Death (2)	Death (4)	NA
Wang 2015	NA	NA	NA
Weinberg 2016	Pruritus (6), dizziness (14), visual disturbances (4), perioral numbness (2), muscle weakness (1), constipation (4)	Pruritus (9), dizziness (20), visual disturbances (6), perioral numbness (2), muscle weakness (3), constipation (10)	NA
Wongyingsinn 2011	NA	NA	"No patients showed signs of lidocaine toxicity in the postoperative period."
Wu 2005	NA	NA	"No patient experienced an identifiable adverse event related to the lidocaine infusion, except that an occasional arrhythmia with stable vital signs was noted in one patient in both groups."
Wuethrich 2012	Light‐headedness (0), drowsiness (0), perioral numbness (0), visual disturbances (0), metal taste (0), pathological cardiac rhythm disturbances (0), and seizures (0)	NA	"No postoperative complications and no adverse events related to systemic administration of lidocaine were observed."
Xu 2017	NA	NA	NA
Yang 2014	Blurred vision (0), hearing problems (0), peripheral paraesthesia (0), dizziness (0), uncontrolled muscle contraction (0), convulsions (0), hypotension (0),  bradycardia (0), headache (0), and itching (0)	NA	NA
Yardeni 2009	NA	NA	NA
Yon 2014	Shivering (0), tinnitus (0)	Shivering (1), tinnitus (0)	NA
Zengin 2015	Pruritus (1)	Pruritus (4)	NA

Author (year)
Journal/Source
Title
Study design	RCT
Population	Experimental	Control
Age
Gender distribution
Weight
ASA physical status
Duration of anaesthesia
Duration of surgery
Main surgical procedure

Continuous outcome data  (unit of measurement)	Experimental (n)		Control (n)
Continuous outcome data  (unit of measurement)	n	Mean (SD)	n	Mean (SD)
Functional gastrointestinal recovery (e.g. time to defaecation)
Pain scores (e.g. VAS)
Length of hospital stay
Functional postoperative neuropsychological status scales
Intraoperative opioid consumption
Postoperative opioid consumption (early and overall)
Patient satisfaction

Methods	Randomized, placebo‐controlled trial. Participants, personnel, and outcome assessors were blinded. The purpose of this trial was to determine if intravenous lidocaine limited to the intraoperative period reduces length of hospital stay and improves functional recovery following abdominal hysterectomy. The study was conducted in Canada from June 2007 to October 2008 (NCT00382499).
Participants	Number assessed for eligibility: 279 Number randomized: 93→ 46:47 Number analysed: 44:46 Inclusion criteria Women, abdominal hysterectomy, ASA I to II Exclusion criteria ASA III, IV, and V, BMI < 18.5 or > 30 kgm^‐2, unable to use PCA, liver dysfunction, creatinine clearance < 50 mlmins^‐1, seizure disorder, hypersensitivity/allergy to amide‐type local anaesthetics study medication, chronic pain, opioid use more than once per week. Baseline details Experimental group (n = 44) Mean age (years): 46.3 M = 0%, F = 100% Mean weight (kg): 70.4 ASA I/II: 13:31 Mean duration of anaesthesia (mins): 105 Main surgical procedures: abdominal hysterectomy Control group (n = 46) Mean age (years): 45.4 M = 0%, F = 100% Mean weight (kg): 69.7 ASA I/II: 18:28 Mean duration of anaesthesia (mins): 108 Main surgical procedures: abdominal hysterectomy
Interventions	Experimental group (46 patients) Lidocaine subjects received prior to induction of anaesthesia an intravenous bolus of 1.5 mg/kg followed by an infusion of 3 mg/kg/hr until skin closure. Control group (47 patients) Control subjects received matching placebo.
Outcomes	The primary endpoint of the study was length of hospital stay. Dichotomous Length of hospital stay measured as number of patients discharged on POD 2 Subjective symptoms of local anaesthetic toxicity (lightheadedness, tinnitus, dysgeusia); PONV recording described, but results not reported Continuous Morphine requirements at PACU, PACU to 6 hrs, 6 to 24 hrs, 24 to 48 hrs, 0 to 48 hrs; intraoperative fentanyl Pain score (VAS 0 to 10) at rest and active at PACU, 6 hrs, 24 hrs, 48 hrs Subjective assessment of QoR score 0 to 18 at 6 hrs, 24 hrs, 48 hrs, and 7 days Brief pain intervention functional interference score at baseline, 24 hrs, 48 hrs, and 7 days First passage of flatus (POD) reported as median values with IQR; data with asymmetric distribution Recording of time to first bowel movements described, but results not reported
Notes	Small trial sample size (< 200 patients) Power analysis performed (patients discharged at POD 2, n = 42) All female patients Medication "All patients received antiemetic prophylaxis with dexamethasone 8 mg and ondansetron 4 mg. All wounds were infiltrated with 20 ml of 0.25% bupivacaine with adrenaline at skin closure. Postoperatively, all patients received celecoxib 200 mg po q12hr and acetaminophen 650 mg po q4hr until hospital discharge. Intravenous patient‐controlled morphine was prescribed with the following settings: boluses of 0.02 mg/kg, no continuous infusion, and a one‐hour maximum of 0.16 mg/kg/hr. Intravenous analgesia was discontinued when the patient tolerated a clear fluid diet. Morphine 5‐10 mg po q4hr prn was ordered for pain that was not controlled with celecoxib and acetaminophen." Anaesthesia All patients received a standardized balanced general anaesthetic. Funding "Trial expenses were funded by the Chair’s Research Fund, Department of Anesthesiology, University of Ottawa. Dr. Bryson was supported by the Ottawa Hospital Anesthesia Alternate Funds Association."
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Computer‐generated random numbers table.
Allocation concealment (selection bias)	Low risk	Central allocation. Quote: "campus‐specific randomization schedules were held by the research pharmacist at each campus. Study medications …prepared by the pharmacist in identical syringes labelled only with the patient’s unique study number".
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	Quote: "research personnel, patients, and attending anaesthesiologists were blinded to the contents of the syringes".
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "outcome measures were recorded by study personnel blinded to treatment allocation".
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate (experimental/control): 4%:2%. Quote: "five patients could not be contacted for follow‐up 7 days after surgery."
Selective reporting (reporting bias)	Low risk	The study protocol is available and all of the study's prespecified primary outcomes that are of interest in the review have been reported. NCT00382499
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled trial. No exact statement on blinding of personnel. This study aimed to evaluate the effects of lidocaine treatment on cognitive impairment in aged patients undergoing spine surgery and to explore the underlying mechanism. The study was conducted in China from September 2013 to February 2015.
Participants	Number assessed for eligibility: N/A Number randomized: 87→ N/A:N/A Number analysed: 80→ 40:40 Inclusion criteria ASA I to II, > 65 years old, scheduled for spine surgery Exclusion criteria Mini‐Mental State Examination score < 23 before surgery, history of neurological diseases (including Alzheimer’s disease and stroke history), psychological disorder, and drug or alcohol abuse, history of diabetes mellitus, severe hypertension, severe anaemia, hepatic or renal dysfunction; unwillingness to comply with the protocol or procedures, inability to speak and read Chinese Baseline details Experimental group (n = 40) Mean age(± SD) (years): 71.3 ± 2.0 M = 57.5%, F = 42.5% Mean weight (± SD) (kg): 64.7 ± 4.3 ASA I/II: 16:24 Mean duration of anaesthesia (± SD) (mins): N/A Mean duration of surgery (± SD) (mins): 129.2 ± 7.4 Main surgical procedures (n): spine surgery (40) Control group (n = 40) Mean age (± SD) (years): 71.8 ± 1.9 M = 62.5%, F = 37.5% Mean weight (± SD) (kg): 63.8 ± 4.3 ASA I/II: 18:22 Mean duration of anaesthesia (± SD) (mins): N/A Mean duration of surgery (± SD) (mins): 128.3 ± 7.3 Main surgical procedures (n): spine surgery (40)
Interventions	Experimental group (40 patients) Patients in the experimental group received a bolus of 1 mg/kg of lidocaine over 5 minutes administered after induction of anaesthesia and followed by a continuous infusion at 1.5 mg/kg/hr until the end of the surgery. Control group (40 patients) Normal saline administered as a bolus and an infusion with the same volume and rate changes as the lidocaine group.
Outcomes	The primary endpoint of the study were functional postoperative neuropsychological status scales. Dichotomous No dichotomous outcomes reported. Continuous Length of hospital stay (days, mean + SD) Functional postoperative neuropsychological status scales (Mini Mental State Examination, preoperative and after 3 days, data presented graphically with mean + SD) Serum assays (T1: preoperative, T2: end of surgery, T3: 3 days after end of surgery)
Notes	‐ Small trial sample size (< 200 patients) ‐ Power analysis not performed Medication N/A Anaesthesia All patients were anaesthetized using standard protocols. Funding This work was supported by grants from the Shandong Province Science and Technology Program.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: “patients were randomly allocated…” No exact statement on random sequence generation.
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: “both patients and the psychometrician were blinded to the treatment and group.” No statement on blinding before outcome assessment.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: “all study personnel were blinded to the results of the laboratory analysis.” Quote: “both patients and the psychometrician were blinded to the treatment and group.”
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate overall: 8% Quote: “seven patients were excluded because of refusal to neuropsychological evaluation after operation.”
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

Date	Event	Description
25 January 2017	New citation required but conclusions have not changed	The results of the meta‐analyses have remained similar to the original review, but the interpretation (GRADE) has changed methodically. We have introduced the 95% prediction interval (PI) to understand the uncertainty of the mean effect estimates associated with heterogeneity. The conclusion has changed for all GRADE‐relevant outcomes, and we graded quality of evidence as very low for: pain (early), postoperative ileus, time to first defaecation/bowel movement, adverse events, postoperative nausea (overall), and opioid consumption (overall). Quality of evidence was moderate for pain at 24 hours and at 48 hours and we are moderately confident that lidocaine has no effect on pain later than 24 hours. In contrast to the original review (Kranke 2015), we were no longer able to demonstrate a significant subgroup difference for the different surgical subgroups investigating pain (early).
25 January 2017	New search has been performed	We updated the search to January 2017. We found 23 new trials that we incorporated into this update and identified a further six trials that we have placed in ‘Studies awaiting classification’. We ran a top‐up search in February 2018, and added 12 trial reports to the six studies already in 'Studies awaiting classification' (Choi 2017; Dewinter 2017; Jendoubi 2017; Kendall 2017; Khalili 2017a; Khalili 2017b; Kim 2017; Kim 2018; Lee 2017; Metha 2017; Sherif 2017; Song 2017). We will incorporate these studies when we next update the review. We have omitted data of up to eight studies per meta‐analysis of continuous outcomes (pain, gastrointestinal recovery, and opioid consumption) from a total of 12 studies with suspected variance reporting. The list of authors has changed. Peter Kranke moves from first to last author (contact author). Stephanie Weibel is the new first author. Johanna Jokinen left the review team and Yvonne Jelting and Antonia Helf are newly added.

Author (Year)
Journal
Title
Relevant study design	RCT	Yes/No/Unclear
Relevant population	Adults of any age and either sex undergoing any surgical procedure	Yes/No/Unclear
Relevant intervention	Experimental group must have received an intravenous lidocaine infusion, that had to be started prior to surgical incision and to be continued at least until the end of surgery	Yes/No/Unclear
If you have not answered Yes to all of the questions, please exclude the study. If you answered Yes to all questions, please continue to data extraction form and critical appraisal.

Intervention	Experimental	Control
Duration of lidocaine infusion (if stated)
Dose of lidocaine infusion and bolus
Total administered dose of lidocaine infusion (if stated)
Starting point of administration before the operation (if stated)
Ending point of administration after the operation (if stated)
Dichotomous outcome data	Experimental (n)	Control (n)
Number of participants with postoperative ileus
Number of participants with occurrence of postoperative nausea (including PONV) or vomiting (early and overall)
Number of participants with surgical complications
Number of participants with cessation of the intervention
Number of participants with adverse effects following the lidocaine infusion (e.g. arrhythmia)

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Pain score at rest, 'early time points' (1 h to 4 h, PACU)	29	1656	Std. Mean Difference (IV, Random, 95% CI)	‐0.50 [‐0.72, ‐0.28]
1.1 open abdominal surgery	8	448	Std. Mean Difference (IV, Random, 95% CI)	‐0.54 [‐0.82, ‐0.26]
1.2 laparoscopic abdominal surgery	10	518	Std. Mean Difference (IV, Random, 95% CI)	‐0.78 [‐1.34, ‐0.21]
1.3 other surgery	11	690	Std. Mean Difference (IV, Random, 95% CI)	‐0.21 [‐0.44, 0.02]
2 Pain score at rest, 'intermediate time points' (24 h)	33	1847	Std. Mean Difference (IV, Random, 95% CI)	‐0.14 [‐0.25, ‐0.04]
3 Pain score at rest, 'late time points' (48 h)	24	1404	Std. Mean Difference (IV, Random, 95% CI)	‐0.11 [‐0.25, 0.04]
4 Postoperative ileus (dichotomous)	4	273	Risk Ratio (M‐H, Random, 95% CI)	0.37 [0.15, 0.87]
5 Time to first defaecation/bowel movement (h)	12	684	Mean Difference (IV, Random, 95% CI)	‐7.92 [‐12.71, ‐3.13]
6 Time to first flatus (h)	13	785	Mean Difference (IV, Random, 95% CI)	‐4.09 [‐6.30, ‐1.87]
7 Time to first bowel sounds (h)	2	110	Mean Difference (IV, Random, 95% CI)	‐6.08 [‐13.77, 1.60]
8 Length of hospital stay (days)	32	2077	Mean Difference (IV, Random, 95% CI)	‐0.37 [‐0.60, ‐0.15]
9 Length of hospital stay (outpatient surgery, mins)	3	191	Mean Difference (IV, Random, 95% CI)	‐10.81 [‐36.93, 15.31]
10 Surgical complications ‐ anastomotic leak	3	188	Risk Ratio (M‐H, Random, 95% CI)	0.61 [0.08, 4.80]
11 Surgical complications ‐ bleeding	3	222	Risk Ratio (M‐H, Random, 95% CI)	1.79 [0.41, 7.89]
12 Surgical complications ‐ postoperative infection	5	352	Risk Ratio (M‐H, Random, 95% CI)	1.64 [0.41, 6.52]
13 Patient satisfaction	6	306	Mean Difference (IV, Random, 95% CI)	0.76 [0.46, 1.06]
14 Postoperative nausea, 'early time points' (PACU)	8	511	Risk Ratio (M‐H, Random, 95% CI)	0.72 [0.53, 0.98]
15 Postoperative nausea, 'overall' (0 to 24 h, to 48 h, to 72 h)	35	1903	Risk Ratio (M‐H, Random, 95% CI)	0.78 [0.67, 0.91]
16 Postoperative vomiting, 'early time points' (PACU)	4	305	Risk Ratio (M‐H, Random, 95% CI)	0.49 [0.16, 1.48]
17 Postoperative vomiting, 'overall' (0 to 24 h, to 48 h, to 72 h)	19	1026	Risk Ratio (M‐H, Random, 95% CI)	0.83 [0.63, 1.08]
18 Intraoperative opioid consumption (MEQ, mg)	18	1116	Mean Difference (IV, Random, 95% CI)	‐2.14 [‐3.87, ‐0.40]
19 Intraoperative remifentanil consumption (µg)	6	490	Mean Difference (IV, Random, 95% CI)	‐14.17 [‐35.27, 6.92]
20 Postoperative opioid consumption, PACU (MEQ, mg)	21	1219	Mean Difference (IV, Random, 95% CI)	‐3.10 [‐3.87, ‐2.32]
21 Postoperative opioid consumption, overall (MEQ, mg)	40	2201	Mean Difference (IV, Random, 95% CI)	‐4.52 [‐6.25, ‐2.79]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Pain score (VAS 0 to 10 cm) at rest, 'intermediate time points' (24 h)	2	102	Mean Difference (IV, Random, 95% CI)	1.51 [‐0.29, 3.32]
2 Pain score (VAS 0 to 10 cm) at rest, 'late time points' (48 h)	2	102	Mean Difference (IV, Random, 95% CI)	0.98 [‐1.19, 3.16]
3 Time to first bowel movement (h)	2	102	Mean Difference (IV, Random, 95% CI)	‐1.66 [‐10.88, 7.56]
4 Length of hospital stay (days)	2	102	Mean Difference (IV, Random, 95% CI)	‐0.02 [‐0.38, 0.33]
5 Intraoperative opioid consumption (MEQ, mg)	2	100	Mean Difference (IV, Random, 95% CI)	7.27 [‐13.92, 28.47]

Methods	Prospective, randomized, placebo‐controlled study. Double‐blind. The purpose of this study was to evaluate the effectiveness of IV lidocaine in reducing postoperative pain for laparoscopic colectomy patients. The study was conducted in Korea. Date not published.
Participants	Number assessed for eligibility: 52 Number randomized: 50→ 25:25 Number analysed: 50→ 25:25 Inclusion criteria Adult patients scheduled to undergo a laparoscopic colectomy, age range 20 to 65 years Exclusion criteria Severe underlying cardiovascular, renal, or hepatic disease, allergic to local anaesthesia, weight < 45 kg or > 100 kg, patients who received opioid or nonsteroidal anti‐inflammatory drugs during the prior week or were taking these drugs chronically as a pain treatment, history of previous abdominal surgery Baseline details Experimental group (n = 25) Mean age (years): 64.48, SD = 11.68 M = 44%, F = 56% Mean weight (kg): 58.87, SD = 8.40 ASA I/II/III: 9:13:3 Mean duration of anaesthesia (mins): N/A Mean duration of surgery (mins): 216.60, SD = 56.29 Main surgical procedures (n): Laparoscopic colectomy (25) Control group (n = 25) Mean age (years): 66.20, SD = 8.88 M = 32%, F = 68% Mean weight (kg): 61.13, SD = 11.47 ASA I/II/III: 8:12:5 Mean duration of anaesthesia (mins): N/A Mean duration of surgery (mins): 204.20, SD = 75.69 Main surgical procedures (n): laparoscopic colectomy (25)
Interventions	Experimental group (25 patients) Group L received IV lidocaine. Two minutes before orotracheal intubation, patients in group L received an IV bolus of lidocaine, 1.5 mg/kg. After induction of anaesthesia, lidocaine (2 mg/kg/hr) was continuous infused during the operation. Control group (25 patients) Group C received normal saline as a placebo. Patients in group C received an IV normal saline bolus and then received the same amount of a continuous infusion of normal saline as that of group L.
Outcomes	The primary endpoint of the study was pain score at 2 hrs. Dichotomous ‐ Postoperative nausea and vomiting (exact time point not mentioned (overall)). Continuous Pain score at rest at 2, 4, 8, 12, 24, 48 hrs (VAS 0 to 100 mm, data presented graphically and as mean + SD) Patient satisfaction (NRS 0 to10) at 48 hrs postoperatively (median + IQR) Length of hospital stay (days, median + IQR) Total fentanyl at 2 hrs postoperatively (µg, mean + SD, data presented and extracted graphically) Total fentanyl cumulative (µg, mean + SD) Start of regular diet (unit unclear) Frequency of pushing PCA button (number, at 2, 4, 8, 12, 24, 48 hrs) Total frequency of pushing PCA button (number) Fentanyl consumption (µg, at 4, 8, 12, 24, 48 hrs, data presented graphically as mean + SD) CRP concentration after surgery (unit unclear, at day of surgery, day 1, 2, 4, 5)
Notes	Small trial sample size (< 200 patients) Power analysis performed (VAS pain score 2 hrs, n = 23 per group) Medication "All patients received fentanyl via PCA postoperatively. Postoperative nausea and vomiting were treated with 4 mg of intravenous ondansetron. " Anaesthesia All patients received the same anaesthetic protocol. Funding Ministry of Education, Science, and Technology
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “randomization into 1 of the 2 groups was based on a random table generated using PASS 11 (NCSS, Kaysville, Utah, USA). The randomization sequence was generated by a statistician who was not otherwise involved with the study.”
Allocation concealment (selection bias)	Unclear risk	Quote: “the details of the series were unknown to the investigators, and the group assignments were kept in sealed envelopes, each bearing only the case number on the outside.” Not explicitly mentioned opaque and sequentially envelopes (SNOSE).
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	Quote: “in order to keep the anesthesiologist ‘‘blind’’ to the patients’ assigned group, lidocaine or normal saline (placebo) was prepared in a syringe and a bottle that was only labeled with a case number. The preparations for the bolus and continuous infusion were arranged by an additional investigator who read the card.” Quote: “all parties involved, including the patients, surgeon, anesthesiologists, and investigator collecting the data, were unaware of the study drugs or the patients’ group assignment.”
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: “VAS scores were collected by 1 blinded investigator who had more than 2 years of experience interviewing patients regarding postoperative pain.” Quote:“all parties involved, including the patients, surgeon, anesthesiologists, and investigator collecting the data, were unaware of the study drugs or the patients’ group assignment.”
Incomplete outcome data (attrition bias)   All outcomes	High risk	Dropout rate (experimental/control): 4%/12% Quote: “we used an intention‐to‐treat strategy—that is, all participants were included in the analysis regardless of whether they had completed the study. Missing data were completed using a last‐observed carried‐forward (LOCF) analysis.” Quote: “four patients had incomplete data because 1 patient in group L and 1 patient in group C were treated with other drugs to control shivering and because 2 patients in group C discontinued the study after stopping the patient‐controlled analgesics because of nausea induced by the fentanyl infusion.” The imputation method (LOCF) was inappropriate and may introduce bias to relevant outcomes.
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, placebo‐controlled trial. No statement on sequence generation, allocation concealment and blinding. The study analysed the efficacy of a continuous low‐dose intravenous infusion of lidocaine on postoperative pain and the requirements for postoperative analgesics in patients after cholecystectomy. The study was conducted in Sweden. Date not published.
Participants	Number assessed for eligibility: N/A Number randomized: 20→ 10:10 Number analysed: N/A, probably 10:10 Inclusion criteria Adult women/man, cholecystectomy Exclusion criteria Patients with hepatic, renal, or cardiovascular disease were excluded. Baseline details Experimental group (n = 10) Median age (years): 44 M = 60%, F = 40% Median weight (kg): 72 ASA I/II: N/A Mean duration of surgery (mins): 105 Main surgical procedures: cholecystectomy Control group (n = 10) Median age (years): 55 M = 50%, F = 50% Median weight (kg): 70 ASA I/II: N/A Mean duration of surgery (mins): 112 Main surgical procedures: cholecystectomy
Interventions	Experimental group (10 patients) Half an hour before skin incision a bolus of lidocaine 100 mg was given followed by continuous infusion of lidocaine 2 mg/min for 24 hours postoperatively. Control group (10 patients) A placebo group received normal saline.
Outcomes	Dichotomous Postoperative nausea and vomiting reported (observation period not stated, but likely 24 hrs after surgery as reported for pain assessment) Adverse events (lightheadedness) Continuous Pain score (VAS 0 to 100) at 2 hr intervals (0 to 24 hrs), starting 1 hr after the return from the operating room (data presented graphically as mean with SEM), mean of the accumulated pain scores (0 to 24 hrs) Meperidine requirements at 0 to 24 hrs, 24 to 48 hrs (data presented graphically as mean with SEM)
Notes	Small trial sample size (< 200 patients) No sample size calculation reported Medication "No meperidine was administered before the first pain assessment. When patients complained of pain they were given injections of 50 mg of meperidine intramuscularly until pain was relieved. Each patient's requirements for meperidine were recorded for 48 hr after surgery." Anaesthesia The anaesthesia regime was standardized in both groups.  Funding No funding mentioned.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	No statement on sequence generation.
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	No statement on blinding of participants and personnel.
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	No statement on blinding of outcome assessors.
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Dropout rate: unclear. There is no statement as to whether the presented results are for all patients who entered the trial or otherwise.
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled trial. Double‐blind. The purpose of this study was to evaluate the effects of intravenous lidocaine on pain following thyroidectomy. The study was conducted in Korea. Date not published. In the protocol it is stated that the study started in July 2011 and was planned to be completed in December 2014 (not verified). (NCT01608360)
Participants	Number assessed for eligibility: 62 Number randomized: 56→ 28:28 Number analysed: 56→ 28:28 Inclusion criteria 18 to 65 years of age, scheduled for elective total thyroidectomy Exclusion criteria body weight < 45 kg or > 100 kg; severe respiratory, renal, or hepatic disease; psychological disorders; history of allergies to local anaesthetics; preoperative use of analgesics, modified radical neck dissection for lateral neck lymph node metastasis Baseline details Experimental group (n = 28) Mean age (years): 49.89, SD = 8.48 M = 28.6%, F = 71.4% Mean weight (kg): 58.62, SD = 7.95 ASA I/II/III: 20:7:1 Duration of anaesthesia (mins) (median): 135.00, IQR (112.25 ‐ 170.00). Duration of surgery (mins) (media, IQR): 100.00, IQR (86.25 ‐140.00) Main surgical procedures (n): elective total thyroidectomy (28) Control group (n = 28) Mean age (years): 50.61, SD = 15.02 M = 17.9%, F = 82.1% Mean weight (kg): 58.16, SD = 7.50 ASA I/II/III: 22:4:2 Duration of anaesthesia (mins) (median): 135.00, IQR (120.00 ‐ 182.50) Duration of surgery (mins) (median): 107.50, IQR (90.00 ‐ 152.50) Main surgical procedures (n): elective total thyroidectomy (28)
Interventions	Experimantal group (28 patients) Just prior to anaesthesia, patients in the lidocaine group received an intravenous bolus infusion of 1.5 mg/kg of lidocaine followed by a continuous infusion of 2 mg/kg/hr lidocaine until the end of surgery. Control group (28 patients) Patients in the control group received normal saline according to the same method.
Outcomes	The primary endpoint of the study was pain at PACU. Dichotomous Vomiting Use of rescue antiemetics Continuous Pain score at rest at PACU, 2 hrs, 4 hrs, 8 hrs, 12 hrs, 24 hrs, 48 hrs (VAS 0 to 100 mm, mean + SD, data presented and extracted graphically) Patient satisfaction (Likert scale 0 to10) at 48 hrs (median + IQR) Fentanyl consumption (µg, mean + SD) at 2 hrs and total (2 hr data presented graphically) Frequency of pushing PCA button (number) Nausea (NRS 0 to10) CRP (mg/L, preoperative, postoperative 2 hrs, POD 1 and 3)
Notes	Small trial sample size (< 200 patients) Power analysis performed (VAS pain at PACU, n = 25 per group) Medication All patients received ramosetron prior to the end of surgery. No additional analgesics or premedication were administered. For postoperative pain control, a fentanyl PCA was provided. Metoclopramide was used as the initial antiemetic rescue medication. Rescue medication was offered for persistent nausea with a NRS ≥ 4. For nausea scores < 4, rescue medication was administered when requested. Ondansetron 4 mg was administered as a second antiemetic, at the investigator’s discretion. Anaesthesia The anaesthesia regime was standardized in both groups. Funding Ministry of Education, Science and Technology
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “blocks of four were generated using the Wei’s Urn model, and randomization was performed using PASSTM 11 software (NCSS, Kaysville, UT, USA). The randomization code was generated by a statistician who was not otherwise involved in the study.”
Allocation concealment (selection bias)	Low risk	Quote: “different researchers completed group randomization and drug preparation, and these researchers were not involved in perioperative management or data collection.” Quote: “researcher A prepared sequentially numbered and sealed opaque envelopes containing patient group information.”
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	Quote: “researcher A delivered the allocation envelope to Researcher B, who opened and returned the envelope after identifying the assigned group, and then prepared trial medication (or placebo control) in a syringe pump labeled only with a case number. […] Researcher B delivered the prepared syringe pump to the anesthesiologist, who was not involved in the study.” Quote: “all surgical procedures were performed by an experienced endocrine surgeon who was unaware of patient group.” “All patients, investigators, and medical staff were blinded to group assignments during patient hospitalization.”
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: “rescue analgesics and antiemetics were administered by nursing staff following direction by an investigator responsible for post‐operative data collection.” Quote: “a trained investigator, who was not involved in the perioperative patient management, was responsible for data collection during the postoperative period. All patients, investigators, and medical staff were blinded to group assignments during patient hospitalization.”
Incomplete outcome data (attrition bias)   All outcomes	Low risk	No withdrawals or exclusions after randomization and no missing data
Selective reporting (reporting bias)	Unclear risk	There is a retrospective registered protocol available (NCT01608360, registered May 25, 2012; study start: July 2011). In the protocol, the primary outcome was pain at 2 hours. Secondary outcomes were among others pain at 4, 8, 12, 24, 48 hours. In the publication, the primary outcome was pain at all time points. The other secondary outcomes in the protocol were the same as in the publication.
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled study. Double‐blinded. The purpose of this study was to evaluate the effect of intravenously administered lidocaine on the quality of recovery and on acute and chronic postoperative pain after robot‐assisted thyroidectomy. The study was conducted in the Korea from July 2013 to January 2015. (NCT01907997)
Participants	Number assessed for eligibility: 94 Number randomized: 90→ 45:45 Number analysed: 84→ 41:43 Inclusion criteria ASA I to II, scheduled to undergo elective robot‐assisted thyroidectomy Exclusion criteria history of chronic pain, chronic use of analgesics, allergy to local anaesthetics, severe cardiopulmonary, hepatic or renal disease, diabetes, and neuropsychiatric disease Baseline details Experimental group (n = 41) Mean age (years): 34.0, SD = 7.3 M = 9.8%, F = 90.2% Mean weight (kg): 58.7, SD = 8.4 ASA I/II: 39:2 Mean duration of anaesthesia (mins): 148.9, SD = 54.1 Mean duration of surgery (mins): 121.6, SD = 54.2 Main surgical procedures (n): thyroidectomy (41) Control group (n = 43) Mean age (years): 34.4, SD = 8.4 M = 2.3%, F = 97.7% Mean weight (kg): 58.0, SD = 9.0 ASA I/II: 37:6 Mean duration of anaesthesia (mins): 152.4, SD = 57.4 Mean duration of surgery (mins): 125.9, SD = 54.1 Main surgical procedures (n): thyroidectomy (43)
Interventions	Experimantal group (41 patients) In the lidocaine group (Group L), 0.1 ml/kg of 2% lidocaine (2 mg/kg) was infused intravenously for 10 mins immediately after anaesthesia induction, and then, it was continuously infused at a rate of 0.15 ml/kg/hr of 2 % lidocaine (3 mg/kg/hr) until the patients were extubated. Control group (43 patients) The control group (Group C) received the same volumes of 0.9% normal saline during the same time periods.
Outcomes	The primary endpoint of the study were functional postoperative neuropsychological status scales (QoR‐40). Dichotomous Rescue antiemetics Rescue antiemetics on the ward Chronic postsurgical pain at 3 months Sensory disturbance at 3 months Continuous Pain score at rest (NRS 0 to 10, median + IQR) on discharge from PACU and at 24 hrs Functional postoperative neuropsychological status scale (QoR‐40 40 to 200 (global), mean + SD) before surgery and at 24 hrs Fentanyl in the PACU (µg, mean + SD) Tramadol on the ward (mg) (mean + SD) Mean amount of remifentanil administered intraoperatively (µg, mean + SD) Maximum pain (NRS 0 to 10, at 48 hrs) Mean volume of fluid administered (ml) QoR‐40 (Emotional status, physical comfort, psychological support, physical independence, pain) Sensory score (24 hrs, 3 months after surgery)
Notes	Small trial sample size (< 200 patients) Power analysis performed (QoR‐40, n = 45 per group) Medication Before the end of the operation, patients received propacetamol 2 g for postoperative pain control and ramosetron 0.3 mg to prevent PONV. On the ward, 200 mg of oral ibuprofen was routinely administered three times per day to all of the patients until they were discharged. If a patient’s vascular endothelial growth factor score for pain was greater than four or if the patient requested an analgesic, he/she was intravenously administered fentanyl (50 µg) in the PACU or tramadol (25 mg) on the ward. Metoclopramide was administered as a rescue antiemetic agent if a patient suffered severe nausea or retching/vomiting. Anaesthesia The anaesthesia regime was standardized in both groups.  Funding Ministry of Science, ICT and Future planning
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “…the principal investigator (J.H.L.) randomly allocated the patients to either the control or the lidocaine group, using a randomization sequence generated by the web site www.randomizer.org/form.htm.”
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	Quote: “the other investigators, including the anesthesiologists responsible for the patients’ intraoperative care, the surgeons, and the nursing staffs, and the patients were blinded with regard to the groups to which the patients were assigned during the entire study period.” Quote: “an anesthetic nurse, who did not participate in the study, prepared the 2% lidocaine or the 0.9% normal saline in 50‐ml syringes in accordance with the principal investigator’s instructions. These injections were administered to the patients by the attending anesthesiologists who did not know the patients’ group allocations.”
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: “the other investigators, including the anesthesiologists responsible for the patients’ intraoperative care, the surgeons, and the nursing staffs, and the patients were blinded with regard to the groups to which the patients were assigned during the entire study period.” Quote: “all of the preoperative and postoperative data for this study were obtained by one investigator who was unaware of the groups to which the patients had been allocated.”
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate (experimental/control): 10%:5% One patient in the control group did not receive the allocated intervention because of conversion to open radical neck dissection and another one was lost to follow up after three months. Four patients in the intervention group didn't receive the allocated intervention due to decline to participate (n = 1), conversion to modified open radical neck dissection (n = 2) or surgical complication (n = 1). Reasons for missing data are unlikely to be related to true outcome (before start of the intervention).
Selective reporting (reporting bias)	Low risk	The study protocol is available (NCT01907997) as well as one conference abstract (Lee 2015). All of the study’s prespecified primary outcomes that are of interest in the review have been reported in the prespecified way. The protocol was prospectively registered (July 2013).
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled trial. No information provided on random sequence generation and allocation concealment. Participants and outcome assessors were blinded. Anaesthesiologists were unblinded. The study aimed to examine whether intraoperative systemic lidocaine may present beneficial effects on the recovery of bowel function, pain intensity, and analgesic consumption in patients undergoing various breast plastic surgeries. The study was conducted in Korea. Date not published.
Participants	Number assessed for eligibility: N/A Number randomized: 60→ 30:30 Number analysed: (30:28:26)/(30:27:22) at 24 hrs: 48 hrs: 72 hrs postoperatively. Inclusion criteria Female patients, aged 20 to 60 years, ASA I to II, elective breast plastic surgeries. Exclusion criteria Severe hepatic, renal, cardiac, respiratory, or endocrine diseases, morbid obesity, or allergies to local anaesthetics. Episodes of intraoperative hypotension (mean BP < 60 mmHg) or bradycardia (heart rate < 40 beats/min), arrhythmia or urticaria associated with lidocaine infusion were also criteria for exclusion. Baseline details Experimental group (n = 30) Mean age (years): 41 M = 0%, F = 100% Mean weight (kg): 56 ASA I/II: N/A Mean duration of surgery (min): 295 Main surgical procedures (n): Augumentation mammaplasty (8), reduction mammaplasty (1), tissue expander removal + augmentation mammaplasty (single/both, 3:4), breast reconstruction with flap (10), mastectomy with implant (4) Control group (n = 30) Mean age (years): 40 M = 0%, F = 100% Mean weight (kg): 55 ASA I/II: N/A Mean duration of surgery (min): 288 Main surgical procedures (n): augumentation mammaplasty (5), reduction mammaplasty (1), tissue expander removal + augmentation mammaplasty (single/both, 5:2), breast reconstruction with flap (11), mastectomy with implant (6)
Interventions	Experimental group (30 patients) 1.5 mg/kg bolus of lidocaine approximately 30 min before incision followed by continuous infusion of lidocaine (1.5 mg/kg/hr) until skin closure. Control group (30 patients) The control group was untreated.
Outcomes	The primary endpoint of the study was restoration of bowel function after surgery. Dichotomous Postoperative nausea and vomiting reported at 24 hrs, 48 hrs, and 72 hrs Side effects (dizziness, itching, respiratory depression) reported at 24 hrs, 48 hrs, and 72 hrs Level of satisfaction for pain control (excellent/satisfied/poor) at 72 hrs Number of patients with extra pain rescue analgesic medication Continuous Time to first flatus and first defaecation (hrs) Length of hospital stay (days) Pain score (VAS 0 to 10) at 24 hrs, 48 hrs, and 72 hrs Pethidine requirement during PACU (mg/patient); PCA (fentanyl/ketorolac) requirement at PACU ‐ 24 hrs, 24 to 48 hrs (ml) Intraoperative anaesthetic requirement (end tidal sevoflurane)
Notes	All female patients Small trial sample size (< 200 patients) Power analysis performed (bowel function, n = 20) Medication "No supplemental opioid was given to patients in either group during the maintenance of anaesthesia. For postoperative pain control, pethidine 0.5 mg/kg was provided within 30 min of the end of surgery and supplemented after recovery as needed with further boluses of 0.25 mg/kg at 20‐30 min intervals. Upon arrival to the post anaesthetic care unit (PACU), patients were connected to an intravenous patient controlled analgesic system (IVPCA) with fentanyl 1,500 μg and ketorolac 180 mg in 64 ml of saline (100 ml of total volume) to deliver a bolus of 1 ml of the above analgesics with a lockout time of 15 min and a basal rate of 1 ml/hr. After transfer to the ward, all patients received IVPCA, and extra rescue medications such as pethidine or NSAID according to body weight, if required." Anaesthesia The anaesthesia regime was standardized in both groups. Funding No funding mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Unclear risk	Quote: " …patients…were randomly and equally divided to two groups". No information provided.
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: " participating patients … were all blinded to the patient's group assignment". "For the safety of patients, anaesthesiologists involved in the anaesthetic managements were not blinded to the groups. However, they were not involved in further management of postoperative pain control or data collection associated with this study".
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "...surgeons, and medical investigators who were involved in the data collection, were all blinded to the patient’s group assignment".
Incomplete outcome data (attrition bias)   All outcomes	High risk	Dropout rate (experimental/control): at 48 hrs: 7%:10%; at 72 hrs: 13%:27% Outcome data 48 hrs and 72 hrs after surgery were incomplete. Withdrawals were not described. High dropout rate at 72 hrs.
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

PERMALINK

Continuous intravenous perioperative lidocaine infusion for postoperative pain and recovery in adults

Stephanie Weibel

Yvonne Jelting

Nathan L Pace

Antonia Helf

Leopold HJ Eberhart

Klaus Hahnenkamp

Markus W Hollmann

Daniel M Poepping

Alexander Schnabel

Peter Kranke

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

Summary of findings

Background

Description of the condition

Description of the intervention

How the intervention might work

Why it is important to do this review

Objectives

Methods

Criteria for considering studies for this review

Types of studies

Types of participants

Types of interventions

Types of outcome measures

Primary outcomes

Secondary outcomes

Search methods for identification of studies

Electronic searches

Searching other resources

Data collection and analysis

Selection of studies

Data extraction and management

Assessment of risk of bias in included studies

Measures of treatment effect

Unit of analysis issues

Multiple‐armed studies

Cluster‐randomized trials

Dealing with missing data

1. Sensitivity analyses ‐ risk of bias (incomplete outcome data).

2. Sensitivity analyses ‐ median + interquartile range.

Assessment of heterogeneity

Assessment of reporting biases

Data synthesis

Subgroup analysis and investigation of heterogeneity

3. Subgroup analyses ‐ type of surgery.

4. Subgroup analyses ‐ time and dosing of lidocaine.

5. Subgroup analyses with independent tau2 (type of surgery).

'Summary of findings' tables and GRADE

Sensitivity analysis

6. Sensitivity analyses ‐ risk of bias (selection bias).

7. Sensitivity analyses ‐ risk of bias (blinding).

8. Sensitivity analyses ‐ random‐effects versus fixed‐effect model.

9. Sensitivity analyses ‐ with studies with 'suspected variance reporting'.

Results

Description of studies

Results of the search

1.

Included studies

Comparators

Surgical procedures

Details on lidocaine administration (dose and timing)

10. Study drug administration.

Inclusion and exclusion criteria

Study conduct (location)

Study sample size

Source of funding

Reported outcomes

Excluded studies

Ongoing studies

Studies awaiting classification

5. Subgroup analyses with independent tau² (type of surgery).

Methods	Randomized, placebo‐controlled trial. Sequence generation based on date of admission. No information provided on allocation concealment. Participants and personnel were blinded. No statement on blinding of outcome assessors. This study was designed to evaluate whether a continuous low‐dose lidocaine infusion reduces postoperative pain and anxiety in patients undergoing CABG and to retrospectively examine time to extubation, ICU stay, and hospital length of stay. The study was conducted in the USA. Date not published.
Participants	Number assessed for eligibility: N/A Number randomized: 100 → 50:50 Number analysed: 44:45 Inclusion criteria Hospital patients undergoing first‐time CABG Exclusion criteria Patients > 75 years, hepatic dysfunction, vitamin K deficiency, serum albumin < 3.0 mg/dl, serum bilirubin > 2.0 mg/dl. Renal impairment, serum creatinine > 1.8 mg/dl, severe left ventricular dysfunction, concomitant valvular surgery, CABG reoperation, patients with pacemakers or atrial and/or ventricular arrhythmias Baseline details Experimental group (n = 44) Mean age (years): 62.65 (mean age of all experimental and control patients) M = 81%, F = 19% Mean weight (kg): 64.5 to 85.3 kg (range of weight for all experimental and control patients) ASA I/II: N/A Duration of surgery (min): N/A Main surgical procedure: CABG Control group (n = 45) Mean age (years): 62.65 (mean age of all experimental and control patients) M = 75%, F = 25% Mean weight (kg): 64.5 to 85.3 kg (range of weight for all experimental and control patients) ASA I/II: N/A Duration of surgery (min): N/A Main surgical procedure: CABG
Interventions	Experimental group (44 patients) An infusion was begun after induction of anaesthesia and before surgical incision. An intravenous dose of 1.5 mg/kg was administered over a 10‐minute period, followed by an infusion of 30 µg/kg/min throughout surgery and for up to 48 hours in the ICU unless discharged earlier. Control group (45 patients) The control group received a placebo substitute.
Outcomes	The primary endpoint of the study was postsurgical pain. Dichotomous Adverse events (hospital mortality, exploration for re‐bleeding, myocardial infarction) Continuous Heart rate at 16 hrs, central venous pressure at 2 hrs and 8 hrs, pulmonary artery pressure at 2 hrs Length of hospital stay (days), (data were presented as median with IQR) Length of ICU stay (hrs) Pain score (visual analogue pain scoring system 0 to 10) at 4 hrs, 8 hrs, 16 hrs, 24 hrs, 48 hrs, and 96 hrs (data presented graphically) Sedation score at 1 hrs, 2 hrs, 4 hrs, 8 hrs, 16 hrs, 24 hrs, 48 hrs, and 96 hrs (data presented graphically) Total postoperative fentanyl consumption (µg) Total postoperative midazolam consumption (mg) Total postoperative propranolol consumption (mg)
Notes	Small trial sample size (< 200 patients) Power analysis performed (pain, n = 44) Medication "If the patient experienced pain, residual neuromuscular blockade or anxiety as evidenced by hypertension (systolic blood pressure greater than 150 mmHg), or tachycardia (heart rate greater than 100 beats per minute), if a conscious patient was unable to maintain a sustained 5‐second head lift, or if a patient experienced a direct communication of pain or discomfort in response to questioning, he or she was treated with intravenous fentanyl and/or midazolam via the following standardized regimen. Fentanyl was administered an 250‐1xg intravenous increments every 15 minutes until a total of I mg was reached. If these demonstrated features were still evident, then intravenous midazolam was administered in 0.5‐mg increments every 5 minutes until the patient was judged comfortable according to the previously cited criteria or a total of 5 mg was reached. If a hyperdynamic situation persisted, then propranolol was administered in 0.25‐rag intravenous increments until the situation abated or a total of 1.0 mg was administered." Anaesthesia  The anaesthesia regime was standardized in both groups. Funding  No funding mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	High risk	Quote: "a double‐blinded, randomized, and prospective approach." "Patients accepted into the study were numbered sequentially 1 through 100."
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: "…pharmacy‐prepared lidocaine infusion (L) in an 8 mg/ml concentration, or placebo substitute (P), numbered 1 through 100, was sent to the operating room on the day of surgery…" It is not clear who was responsible for randomisation and informed the pharmacy how to prepare the study drugs (i. e. which number referred to which group). Therefore, it is unclear if blinding of personnel and participants was adequate.
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	No statement on blinding of outcome assessors.
Incomplete outcome data (attrition bias)   All outcomes	High risk	Dropout rate (experimental/control): 12%:10% Eleven patients were excluded and reasons were described. Reasons for exclusion (e.g. ventricular arrhythmia) may be related to the intervention. One patient in the lidocaine group was excluded due to death (multi‐organ system failure).
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	High risk	Patients in the placebo group received an intravenous bolus of lidocaine (1.5 mg/kg) if ventricular ectopy or fibrillation occurred during surgery (CABG).

Methods	Randomized, controlled trial. Single‐blinded. The aim of this study was to compare the effect of lidocaine and dexmedetomidine infusion during off‐pump CABG. The study was conducted in Korea from September 2012 to August 2013. (NCT01688648)
Participants	Number assessed for eligibility: 174 Number randomized: 160→ 40:40:40:40 Number analysed: 153→ 36:40:39:38 Four groups, two not of interest (dexmedetomidine, lidocaine and dexmedetomidine (combined)) Inclusion criteria Patients undergoing off‐pump CABG by a single surgical team Exclusion criteria Surgery with pre‐planned CPB; patients diagnosed with arrhythmia with medication or pacemaker; unexpected conversion to CPB during the surgery Baseline details Experimental group (n = 36) Age (years) (median): 67, IQR (61 ‐ 72) M = 69.4%, F = 30.6% Mean weight (kg): 66, SD = 11 ASA I/II (n): N/A Mean duration of anaesthesia (min): 339, SD = 52 Mean duration of surgery (min): 283, SD = 50 Main surgical procedures (n): CABG (36) Control group (n = 38) Age (years) (median): 65, IQR (57 ‐ 72) M = 73.7%, F = 26.3% Mean weight (kg): 67, SD = 9 ASA I/II (n): N/A Mean duration of anaesthesia (min): 302, SD = 52 Mean duration of surgery (min): 247, SD = 50 Main surgical procedures (n): CABG (38)
Interventions	Experimental group (36 patients) For the lidocaine infusion group (Group LIDO), lidocaine was infused at a dose of 2 mg/kg/hr from the start of anaesthesia induction after a bolus dose of 1.5 mg/kg. Both lidocaine and dexmedetomidine were infused until 24 hrs after the end of surgery on (POD 1). Control group (38 patients) Neither lidocaine nor dexmedetomidine was infused in the control group.
Outcomes	The primary endpoint of the study were cTnI levels. Dichotomous Inotrope use LEVF 35 << 55% Left main stenosis > 50% Surgical complications (event of myocardial ischaemia, pleural effusion, surgical wound infection, pulmonary consolidation, neurologic deterioration) Adverse events (one year mortality) Continuous Length of hospital stay (postoperative days, median + IQR) Amount of remifentanil infused during surgery (µg, mean + SD) Amount of dexmedetomidine infused (mg, mean + SD) Administered volume (crystalloid, colloid, packed red blood cells cell saver, median + IQR) Estimated blood loss (ml, median + IQR) ICU stay (days, median + IQR) Creatinine kinase myocardial band (ng/ml, median + IQR) cTnI levels (ng/ml, median + IQR)
Notes	Small trial sample size (< 200 patients) Power analysis performed (cTnI, n = 37 per group) Medication Packed red blood cells were transfused when the hematocrit level was less than 25 % during the surgery. In all groups, remifentanil was infused at a dose range of 0.05 – 0.30 μg/kg/min during the surgery. Anaesthesia: The anaesthesia regime was standardized in both groups. Funding No external fund received.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “subjects were randomly assigned at a 1:1:1:1 allocation ratio into one of four groups, Group LIDO, Group DEX, Group Combined or control group, using the random numbers generated by an internet‐based computer program (www.randomizer.org) […].”
Allocation concealment (selection bias)	Unclear risk	Quote: “[…] and sealed envelope technique.” Not explicitly mentioned sequentially numbered, opaque envelopes.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: “while the anesthesiologists were not blinded to the study drug, the participants, surgeon, and data analyst were kept blinded to the assigned group.” The attending anaesthesiologists were not blinded. It is unclear if outcomes are influenced by lack of blinding.
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	No statement on blinding of outcome assessment.
Incomplete outcome data (attrition bias)   All outcomes	Unclear risk	Dropout rate (LIDO/DEX/COMB/Control): 10%:0%:3%:5% Quote: “of the 160 patients initially randomized, four patients were excluded from the study due to unexpected conversion to surgery with CPB. Three patients were further excluded from the analysis because there was a missing laboratory value. We analyzed 36 patients in Group LIDO, 40 in Group DEX, 39 in the Group Combined, and 38 in the control group.” Reasons for missing outcome data are explained in the text. According to the flow diagram, the seven missing patients were excluded because of unexpected conversion to surgery with CBP and not due to missing laboratory values. It is not clear from the description, if patients refused blood draw or follow up. It remains unclear if reasons for missing outcome data are related to true outcome.
Selective reporting (reporting bias)	Low risk	The study protocol is available and all of the study’s prespecified (primary and secondary) outcomes that are of interest have been reported in the prespecified way (NCT01688648). The secondary outcomes blood sodium, potassium and calcium level as well as hypokalaemia and the incidence of arrhythmia have been prespecified but have not been reported in the publication (these outcomes are not of interest for the review). The study has been prospectively registered (first received: 10 September 2012).
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled trial. No statement on blinding of participants. The anaesthesiologists were unblinded. The outcome assessors were blinded. The purpose of this study was to determine whether intraoperative lidocaine infusion reduces opioid consumption in the PACU in patients undergoing laparoscopic cholecystectomy. The study was conducted in Canada from May 2007 to February 2008.
Participants	Number assessed for eligibility: 63 Number randomized: 50 → 25:25 Number analysed: 25:24 Inclusion criteria Outpatient laparoscopic cholecystectomy. Exclusion criteria: Age < 18 yrs or > 85 yrs, ASA physical status III and greater, history of hepatic, renal or cardiac failure, organ transplant, diabetes, morbid obesity (BMI > 40 kg/m²), chronic use of opioids, allergy to local anaesthetics, or inability to comprehend pain assessment. Baseline details Experimental group (n = 25) Mean age (years): 50.2 M = 20%, F = 80% Mean weight (kg): 66.9 ASA I/II: 17:8 Duration of surgery (min): 60 Main surgical procedure: laparoscopic cholecystectomy Control group (n = 24) Mean age (years): 53.8 M = 48%, F = 52% Mean weight (kg): 75 ASA I/II: 11:14 Duration of surgery (min): 70 Main surgical procedure: laparoscopic cholecystectomy
Interventions	Experimental group (25 patients) At induction of anaesthesia the lidocaine group received fentanyl 1.5 µg/kg and a bolus of lidocaine 1.5 mg/kg followed by a continuous infusion of lidocaine 2 mg/kg/hr until the end of surgery. Control group (24 patients) At induction of anaesthesia the control group received fentanyl 3 µg/kg.
Outcomes	The primary endpoint of the study was fentanyl consumption. Dichotomous Postoperative nausea and vomiting during PACU and 24 hrs after surgery Use of ondansetron and number of patients requiring ondansetron (0:2:4:8 mg) Number of patients with White‐Song score > 12 at 1st: 30th: 60th min Continuous Pain score (VRS 0 to 10) during rest at 1 min, 30 min, 60 min, and 90 min, as well as pain score during rest, coughing, walking at 24 hrs (data presented as median with IQR, in part with asymmetric distribution) Shoulder pain (VRS 0 to 10) at 24 hrs (data presented as median with IQR) Fatigue (VRS 0 to 10) at 24 hrs (data presented as median with IQR) White‐Song score at 1 min, 30 min, 60 min, and 90 min (data presented as median with IQR) Fentanyl consumption (µg), intraoperatively and at PACU Time from arrival PACU to discharge home (min), (data presented as median with IQR) Acetaminophen consumption (mg) in 24 hrs Naproxen consumption (mg) in 24 hrs Oxycodone consumption (mg) in 24 hrs
Notes	Small trial sample size (< 200 patients) Power analysis performed (fentanyl consumption, n = 25) Medication "No supplemental opioids were given during surgery. All patients received acetaminophen, ketorolac, dexamethasone, droperidol and local anaesthetics in the skin incision. Patients received fentanyl and ondansetron in the PACU. Before induction of anaesthesia, patients in the control group received fentanyl 3.0 µg/kg iv, while patients in the lidocaine group received fentanyl 1.5 µg/kg iv. No supplemental fentanyl was given to patients in either group during maintenance of anaesthesia. Ketorolac 15 mg and droperidol 0.625 mg were also given intravenously. Ten millilitres of bupivacaine 0.25% with epinephrine was injected into the surgical incisions. According to study protocol, the PACU nursing staff administered fentanyl 25 µg iv boluses for postoperative pain relief, to be administered every five minutes up to a maximum of 200 µg/hr only if the VRS score for pain (0–10 scale, where 0 = no pain, and 10 = excruciating pain) was > 3, at rest. Ondansetron 2 mg iv was prescribed for persistent nausea (lasting > five minutes) or vomiting, and it could be repeated up to four times over a three‐hour period if necessary." Anaesthesia  The anaesthesia regime was not fully standardized. The control group received more fentanyl. Funding  "This work was supported by internal funds, Department of Anesthesia, McGill University Health Centre."
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "...randomly assigned, using a computer‐generated randomization schedule, into two groups of 25 patients…"
Allocation concealment (selection bias)	Low risk	Quote: "...sequentially numbered sealed brown envelopes…"
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	The anaesthesiologist was not blinded, but (quote:) "the anaesthesia record was not made available to the recovery room nurse, to avoid bias.", "...the anaesthesiologists (S.L. and F.C.) who executed the study protocol, were not involved in either the preoperative or the postoperative data collection."
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "...monitored and recorded by nurses who were blinded to the randomization sequence."; "The anaesthesia record was not made available to the recovery room nurse, to avoid bias."
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate (experimental/control): 0%:4% Quote: "One patient in the control group was excluded from the analysis because his surgery was converted from laparoscopy to laparotomy."
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Unclear risk	The study groups differ with a greater proportion of males (48% versus 20%) in the control group.

Methods	Randomized, placebo‐controlled trail. No statement on blinding of personnel. Participants and outcome assessors were blinded. This study assessed the potential of intravenously administered lidocaine to reduce postoperative cognitive dysfunction following cardiac surgery employing cardiopulmonary bypass. The study was conducted in the USA from March 1999 to April 2003.
Participants	Number assessed for eligibility: 2681 Number randomized: 277 → 133:144 Number analysed: 88:94 Inclusion criteria Patients scheduled to undergo CABG and/or an open chamber procedure with CPB. Exclusion criteria Patients undergoing circulatory arrest or had a history of symptomatic cerebrovascular disease (e.g. stroke with a residual deficit), psychiatric illness (any clinical diagnoses requiring therapy), renal failure (serum creatinine > 2 mg/dl), liver disease (liver function tests > 1.5 times the upper limit of normal), higher alcohol consumption (> 2 drinks/day), or were unable to read or had less than a seventh grade education. Baseline details Experimental group (n = 114) Mean age (years): 61.7 M = 72.8%, F = 27.2% Mean weight (kg): 86.1 ASA I/II: N/A Duration of surgery (min): N/A Main surgical procedure (n): CABG (51), CABG with valve (22), valve (40), other (1) Control group (n = 127) Mean age (years): 61.4 M = 66.9%, F = 33.1% Mean weight (kg): 81.6 ASA I/II: N/A Duration of surgery (min): N/A Main surgical procedure (n): CABG (52), CABG with valve (23), valve (47), other (5)
Interventions	Experimental group (114 patients) Lidocaine was administered after induction of anaesthesia as a 1 mg/kg bolus followed by a continuous infusion (4 mg/min for 1 hr, 2 mg/min for the second hr, 1 mg/min for the rest) through 48 hours postoperatively. Control group (127 patients) Placebo bolus and infusion for 48 hrs.
Outcomes	The primary endpoint of the study was incidence of cognitive deficit. Dichotomous Number of patients with cognitive deficits at 6 weeks and 1 year after surgery Serious adverse events (mortality) Continuous Cognitive score (5 cognitive tests producing 10 scores) at 6 weeks and 1 year after surgery Length of hospital stay (days), (data presented as median with IQR, asymmetric distribution) Plasma level of caspase‐3, C‐reactive protein, IL‐8, matrix metalloproteinase‐9, vascular endothelial growth factor, S‐100ß at baseline, at end of CPB, 4.5 hrs and 24 hrs after CPB
Notes	Large trial sample size (> 200 patients) Power analysis performed (incidence of cognitive deficit, n = 112) Medication N/A Anaesthesia  Standardization of the anaesthesia regime is unclear. Funding  "Supported in part by grants #9970128N (Dr. Newman) from the American Heart Association, Dallas, TX, USA, #M01‐RR‐30 from the National Institutes of Health, Washington, D.C., USA, and by the Division of Cardiothoracic Anesthesiology and Critical Care Medicine, Department of Anesthesiology, Duke University Medical Center, Durham, NC, USA." Conflict Of Interest "Dr. Laskowitz is a consultant for Biosite Diagnostics."
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "a group assignment schedule was prepared using a randomization function in sedation‐agitation status…"
Allocation concealment (selection bias)	Unclear risk	Quote: "...and stored in consecutively numbered sealed envelopes until allocation." Not mentioned if envelopes were opaque.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: "...identical volume and rate changes as the treatment group such that blinding was preserved." No statement on blinding of personnel and participants.
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: "experienced psychometricians blinded to the treatment group examined subjects..."
Incomplete outcome data (attrition bias)   All outcomes	High risk	Dropout rate after received intervention (experimental/control): 23%:26% Exclusions, withdrawals, and dropouts were described. It is unclear from the description whether the reasons (e.g. lack of interest, health, other) may be related to true outcome.
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry and no published study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Multicentric, double‐blind, randomized, and placebo‐controlled trial. This trial aimed to compare postoperative analgesia, opioid consumption, duration of ileus and hospital stay, and cytokine levels in patients undergoing laparoscopic cholecystectomies who received intravenous lidocaine in comparison with a control group. The study was conducted in Brazil from July 2013 to February 2014. (NCT02363699)
Participants	Number assessed for eligibility:87 Number randomized: 44→ 22:22 Number analysed: 43→ 21:22 Inclusion criteria patients scheduled to undergo elective laparoscopic cholecystectomy, 18 years and older, ASA I and II Exclusion criteria Patients older than 75 years, patients with heart disease, and patients with history of kidney failure, liver failure, psychiatric disorder, chronic use of opioids, or medications that could cause induction of liver enzymes (anticonvulsants) were not included in the study. In addition, the presentation of adverse effects during the intervention or postoperative complications, and the conversion to open surgery were used as exclusion criteria. Baseline details Experimental group (n = 22) Mean age (years): 43.77, SD = 12.55 M = 22,7%, F = 77,3% Mean weight(kg): 76.91, SD = 16.41 ASA I/II (n): NA Mean duration of anaesthesia (min): N/A Mean duration of surgery (min): 105.23, SD = 38.25 Main surgical procedures (n): elective laparoscopic cholecystectomy (21) Control group (n = 22) Mean age (years): 46.09, SD = 11.50 M = 40,9%, F = 59,1% Mean weight (kg): 86.86, SD = 19.76 ASA I/II (n): NA Mean duration of anaesthesia (min): N/A Mean duration of surgery (min): 112.50, SD = 47.58 Main surgical procedures (n): elective laparoscopic cholecystectomy (22)
Interventions	Experimental group (22 patients) Lidocaine was administered in bolus of 1.5 mg/kg at the start of the procedure and maintained at a dose of 3 mg/kg/h until 1 hour after the end of the surgery. A solution of 0.3 % lidocaine was used, so that the infusion rate was equal to the patient's weight. Answer from author upon request: lidocaine infusion started before incision Control group (22 patients) Saline solution was administered in the control group with the same infusion rates
Outcomes	The primary endpoint of the study was pain within the first 24 hrs. Dichotomous Adverse effect (arrhythmia) Continuous Pain score at rest at 1, 2, 4, 12, 24 h (VNS 0 to 10, mean + SD) First flatus (hrs, median, P = 0.75) Length of hospital stay (hrs) Opioid consumption total (morphine) (mg, median + IQR) Inflammatory markers (IL1, IL6, IL10, Interferon‐gamma, TNF‐alpha) at 1 hr and 24 hrs Pain score when coughing 1, 2, 4, 12, 24 hrs (VNS 0 to10, mean + SD)
Notes	Small trial sample size (< 200 patients) Power analysis not performed Medication During surgery, all patients received 4 mg ondansetron for prophylaxis of nausea and vomiting and dipyrone 30 mg/kg, 30 minutes before the end of the procedure. Postoperatively, patients received dipyrone 1 g IV every 6 hours and ondansetron 8 mg IV every 8 hrs. For patients who reported pain at rest equal or greater than 4, morphine titration was started with 1 mg increments every 5 minutes until the pain was reported as less than 4. At this point, patients were encouraged to manage their own medication. The patient‐controlled analgesia pumps were programmed to bolus of 4 ml (morphine solution 0.5 mg/ml) followed by 8 minutes of security lock between doses. The maximum dose in 4 hrs was 30 mg. No continuous maintenance dose of morphine was used in the postoperative period. Fifteen percent increases in mean arterial pressure or heart rate values greater than 100 beats/min, with bispectral index between 40 and 60, allowed for supplementary administration of 5 μg sufentanil. Anaesthesia The anaesthesia regime was standardized in both groups. Funding No external funding source was used.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “randomization was performed by a computer program (random.org) that generated a random number sequence from 1 to 44, divided into 2 columns. “
Allocation concealment (selection bias)	Unclear risk	Quote: “an employee of the Surgical Center pharmacy, previously trained for that, appointed each column with a group (lidocaine and placebo) and stored the results in 44 envelopes. “ From the description of the text is it unclear if the allocation was fully concealed.
Blinding of participants and personnel (performance bias)   All outcomes	Low risk	Quote: “this same collaborator was responsible for preparing the solutions according to the dilution protocol. These solutions were kept in identical color and volume containers for both groups and were provided just before anesthetic induction.” Quote: “The study was double blind. Patients were not informed about the solution they were receiving. Likewise, both the research team and the auxiliary anesthetists were unaware of which group each patient belonged to.”
Blinding of outcome assessment (detection bias)   All outcomes	Low risk	Quote: “the study was double blind. Patients were not informed about the solution they were receiving. Likewise, both the research team and the auxiliary anesthetists were unaware of which group each patient belonged to.”
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate (experimental/control): 5%:0% Quote: “the patients who had hospital discharge before 24 hours were asked to wait until the final pain assessment and blood collection. In this way, we avoided losses and all the data were included in the analysis.” Quote: “the lidocaine group lost one patient who had his surgery converted due to problems during the procedure. All patients who started receiving lidocaine continued the infusion according to the protocol.” Reasons for missing data (1 patient) are unlikely to be related to true outcome.
Selective reporting (reporting bias)	Unclear risk	The study protocol is available and the primary outcome as well as some secondary outcomes have been reported. The secondary outcomes “length of hospital stay” and “time to first flatus” have not been prespecified. The protocol has been retrospectively registered (4 February 2015) (NCT02363699).
Other bias	Low risk	The study appears to be free of other sources of bias.

Methods	Randomized, controlled trial. No blinding of participants, personnel, and outcome assessors. This study compared postoperative epidural analgesia and IV infusion of local anaesthetic on ileus duration and hospital stay in patients after colon surgery. The study was conducted in the USA from April 2005 to July 2006 (NCT00600158).
Participants	Number assessed for eligibility: N/A Number randomized: 45 → 24:21 Number analysed: 22:20 Inclusion criteria Patients aged 18 to 75 years of ASA I to III, scheduled for elective colon resection. Exclusion criteria Allergy to local anaesthetics, myocardial infarction within 6 months before surgery, liver disease (aspartate aminotransferase, alanine aminotransferase, or bilirubin 92.5 times the upper limit of normal), renal impairment (creatinine clearance 60 ml/min), systemic corticosteroid use, chronic use of opiates, unwillingness or contraindication to epidural analgesia, pregnancy, or active breast‐feeding. Baseline details Experimental group (n = 22) Median age (years): 52 M = 45%, F = 55% Median BMI (kg/m²): 25 ASA I/II/III: 1:14:7 Duration of surgery (min): 181 Main surgical procedure (n): subtotal colectomy (2), total abdominal colectomy (0), low‐anterior resection/abdominal perineal resection/ileal pouch‐anal anastomosis (20), lyses of adhesion, small‐bowel resection with primary anastomosis and ileostomy (0), closure of end ileostomy with bowel resection (0) Control (epidural) group (n = 20) Median age (years): 49 M = 80%, F = 20% Median BMI (kg/m²): 28 ASA I/II/III: 1:18:0 Duration of surgery (min): 175 Main surgical procedure (n): subtotal colectomy (4), total abdominal colectomy (1), low‐anterior resection/abdominal perineal resection/ileal pouch‐anal anastomosis (12), lyses of adhesion, small‐bowel resection with primary anastomosis and ileostomy (1), closure of end ileostomy with bowel resection (1)
Interventions	Experimental group (22 patients) Before induction of general anaesthesia, patients received IV lidocaine (11 patients: 2 mg/min in patients < 70 kg, 3 mg/min in patients > 70 kg, and 11 patients: 1 mg/min in patients < 70 kg, 2 mg/min in patients > 70 kg). The day after return of bowel function, the lidocaine infusion was turned off. If flatus had not occurred on the fifth POD, IV lidocaine was discontinued. Control group (20 patients) Patients received an epidural analgesia (bupivacaine 0.125% and hydromorphone 6 Kg/ml were started at 10 ml/hr within 1 hr of the end of surgery).
Outcomes	The primary endpoint of the study was time to first bowel movement. Dichotomous Postoperative nausea and vomiting within 5 days after surgery Side effects (wound infection, anaemia, anxiety, supraventricular tachycardia, back pain, bradycardia, confusion, decreased oxygen saturation level, dizziness/light headedness, fever, hyperglycaemia, hypertension, itching, lower extremity numbness, intravascular device infection, syncope, arrhythmia severe, confusion severe, facial numbness severe, shortens of breath) Continuous Time to first flatus (days), (data presented as median with IQR) Time to first bowel movement (days), (data presented as median with IQR, asymmetric distribution) Length of stay (inpatient time, days), (data presented as median with IQR, asymmetric distribution) Time of advancement to clear liquid diet, (data presented as median with IQR) Intraoperative fentanyl dose (µg) Intraoperative morphine dose (mg) Morphine equivalents during surgery Daily opioid consumption (morphine, mg), operation day, POD 1, POD 2, POD 3, POD 4, (data presented as median with IQR) Median average daily pain score (VAS 0 to 10) at day 1, day 2, day 3, day 4, and day 5 (data presented as median with IQR)
Notes	The authors reported "Patients randomized to the IV local anaesthetic group received an IV infusion of lidocaine starting after anaesthesia induction. We initially administered 2 mg/min in patients less than 70 kg and 3 mg/min in patients 70 kg or greater, as reported in the literature. However, several patients reached potentially toxic plasma levels, and therefore, we reduced the dose in the remaining 11 patients to 1 mg/min in patients less than 70 kg and 2 mg/min in patients 70 kg or greater. Subgroup analysis showed no difference in the primary end point between the 2 dosing schemes, and we therefore pooled the data from the groups for further analysis" The authors used two different local anaesthetics: bupivacaine for epidural administration and lidocaine for IV administration The starting time of local anaesthetics infusion was different between the groups (lidocaine: before induction of general anaesthesia, epidural: within 1 hr of the end of surgery) There were differences in the proportion of female patients (20% in the epidural group and 55% in the IV lidocaine group; P = 0.021) and distribution of ASA scores (P = 0.014: the IV lidocaine arm included all the ASA III patients) Small trial sample size (< 200 patients) Power analysis performed (bowel function, n = 19) Medication "When fentanyl was used rather than morphine, it was converted to morphine equivalents using a conversion ratio of 100 µg fentanyl = 10 mg morphine." "In the recovery area, pain was assessed using an 11‐point verbal scale (0 to 10) every 15 min, and scores greater than 3 were treated with either fentanyl 50 µg every 10 min or morphine 4 mg every 20 min as needed. After transfer to the ward, all patients received PCA for breakthrough pain. Initial PCA setting included morphine 2 mg IV demand dose with 6‐min lockout interval (10 mg/hr maximum). Fentanyl was used in an appropriate dose if the patient reported an allergy to morphine." Anaesthesia  The anaesthesia regime was standardized. Funding  No funding mentioned
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: "patient assignments were generated using a published table of random numbers."
Allocation concealment (selection bias)	Unclear risk	Quote: "...stored in sealed envelopes before initiation of the study protocol." Not mentioned sequentially numbered and opaque envelopes.
Blinding of participants and personnel (performance bias)   All outcomes	High risk	No blinding possible due to study design.
Blinding of outcome assessment (detection bias)   All outcomes	High risk	No blinding possible due to study design.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Dropout rate (experimental/control): 8%:5% Withdrawals were described. Quote: "if therapy outside the standard protocol was required, the patient was withdrawn from the study and followed in an intent‐to‐treat manner for assessment of primary outcomes."
Selective reporting (reporting bias)	Unclear risk	The study protocol is available and all of the study's prespecified primary and secondary outcomes that are of interest in the review have been reported in the prespecified way. However, trial registry occurred retrospectively (registered January 2008, study completed July 2006). (NCT00600158)
Other bias	High risk	Fifty percent of the patients in the lidocaine group received higher doses of lidocaine infusion than the other 50%. The authors used two different local anaesthetics: bupivacaine for epidural administration and lidocaine for IV administration. The starting time of local anaesthetics infusion was different between the groups (lidocaine: before induction of general anaesthesia, Epidural: within 1 hr of the end of surgery). There were differences in the proportion of female patients (20% in the epidural group and 55% in the IV lidocaine group; P value = 0.021) and distribution of ASA scores (P value = 0.014: the IV lidocaine arm included all the ASA III patients).

Methods	Randomized, controlled trial. Blinding unclear. The aim of this study was to investigate if lidocaine has a beneficial effect on anti‑cell‑mediated immunity during the postoperative period in patients with cervical cancer undergoing radical hysterectomy. The study was conducted in China between August 2013 and January 2014.
Participants	Number assessed for eligibility: N/A Number randomized: 30→ 15:15 Number analysed: 30→ 15:15 Inclusion criteria Aged between 25 and 65 years old, undergoing radical hysterectomy Exclusion criteria Weight < 45 kg or > 65 kg; a history of allergies to local anaesthetics, bradycardia or heart block; severe respiratory, renal or hepatic disease, previous history of opioid medication use or a psychiatric medical history Baseline details Experimental group (n = 15) Mean age (years): 44.2, SD = 11.8 M = 0%, F = 100% Mean weight (kg): 56.0, SD = 6.5 ASA I/II (n): 11:4 Mean duration of anaesthesia (min): 152.3, SD = 14.1 Mean duration of surgery (min): 132.3, SD = 25.1 Main surgical procedures (n): hysterectomy (15) Control group (n = 15) Mean age (years): 48.6, SD = 5.6 M = 0%, F = 100% Mean weight (kg): 56.9, SD = 7.6 ASA I/II (n): 10:5 Mean duration of anaesthesia (min): 158.0, SD = 16.9 Mean duration of surgery (min): 129.3, SD = ± 24.4 Main surgical procedures (n): hysterectomy (15)
Interventions	Experimental group (15 patients) The patients assigned to the lidocaine group received an intravenous bolus infusion of 1.5 mg/kg lidocaine 10 min prior to the induction of anaesthesia, followed by continuous infusion at 1.5 mg/kg/hr until discharge from the operating room. Control group (15 patients) The patients in the control group received the same volume of normal saline.
Outcomes	The primary endpoint of the study was not reported and power analysis was not performed. Dichotomous No outcomes reported Continuous Proliferation rate of peripheral blood lymphocytes (450 nm optical density), apoptosis, expression of cytokines, serum protein levels of HMGB1
Notes	All female patients (100%) in both groups Small trial sample size (< 200 patients) Power analysis not performed Medication Premedication 0.1 mg/kg midazolam Anaesthesia The anaesthesia regime was standardized in both groups. Funding This study was supported by the Qilu Hospital Science Research Foundation (grant no. 26010175616032), and in part, by Shandong Provincial Natural Science Foundation, China.
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Quote: “all participants were randomized into two groups, according to a computer‑generated random number table […].”
Allocation concealment (selection bias)	Unclear risk	No statement on allocation concealment.
Blinding of participants and personnel (performance bias)   All outcomes	Unclear risk	Quote: “the solutions were prepared in a 20 cc syringe and labeled only with a case number by a nurse in a blinded‑manner.” No statement on blinding of participants and other research team.
Blinding of outcome assessment (detection bias)   All outcomes	Unclear risk	No statement on blinding of outcome assessment. All outcomes were dependent on blood values. It remains unclear who assessed the blood draws.
Incomplete outcome data (attrition bias)   All outcomes	Low risk	Quote: “all 30 patients recruited in the present study completed the study.” No missing outcome data.
Selective reporting (reporting bias)	Unclear risk	There is no reference to a trial registry or study protocol.
Other bias	Low risk	The study appears to be free of other sources of bias.

Study	Reason for exclusion
Bartlett 1961	No control group, not a randomized controlled trial
Birch 1987	Lidocaine was given only after the operation
Cepeda 1996	Lidocaine was given as patient‐controlled analgesia only after the operation
Chia 1998	Lidocaine was given only after the operation
Couceiro 2015	Lidocaine was given after incision
De Clive‐Lowe 1958	No control group, not a randomized controlled trial
De Kock 1994	Lidocaine was given as repeated bolus, not as continuous infusion
Feld 2003	Lidocaine was only one of a group of non‐opioid drugs which were compared with fentanyl
Hans 2010	Infusion was only given over the first 30 minutes
Harvey 2009	Lidocaine was given for 24 hours only after the operation
Joppich 2010	Review article (relating to McCarthy 2010; no secondary publication)
Juarez‐Pichardo 2009	Lidocaine infusion was terminated 10 minutes before end of surgery
Kavak 2014	Wrong control intervention (remifentanil)
Knight 1980	No control group, not a randomized controlled trial
Marret 2008	Review article
McCarthy 2010	Review article
Olivares 2012	No control group, the other group received magnesium
Perniola 2014	Licocaine was given after incision
Rinne 1998	Infusion was started after skin incision
Sun 2012	Review article
Vigneault 2011	Review article
Zhu 2015	Wrong intervention (additional ketamine infusion in intervention group)

Methods	Randomized controlled trial. No statement on blinding. We evaluate the effects of perioperative infusion of lidocaine and dexmedetomidine IV on postoperative pain control and analgesics consumption after laparoscopic cholecystectomy. The study was conducted in the Republic of Korea. It is not stated when the study was conducted. Trial identifier: n/a
Participants	Sample size: 84 (3 groups, n = 28, respectively) Three groups, one not of interest (dexmedetomidine) Inclusion criteria Aged 20 to 60 years, elective LC Exclusion criteria N/A
Interventions	Experimental group (28 patients) The patients in group L received an IV lidocaine bolus of 1.5 mg/kg and then continuous infusion of 2 mg/kg/h. Bolus doses were given during 10 minutes before the induction of anaesthesia,  followed by continuous infusion until end of the surgery. Control group (28 patients) The group N received saline by same method as group L (bolus of 1.5 mg/kg and then continuous infusion of 2 mg/kg/h). Bolus doses were given during 10 minutes before the induction of anaesthesia,  followed by continuous infusion until end of the surgery.
Outcomes	VAS pain score during first 24 h after LC, postoperative analgesics consumption (amount of fentanyl consumption in PACU) were evaluated during 24 h after the surgery.
Notes	Small trial sample size (< 200 patients) Anaesthesia N/A Funding This work is supported by the 2011 Inje University research grant. Notes Only abstract available

Methods	Randomized, double‐blinded, controlled trial. The aim of this randomized, double‐blinded, controlled study was to evaluate the effect of IV administered lidocaine on the QoR and on acute and chronic postoperative pain after robot‐assisted thyroidectomy. The study was conducted in the Republic of Korea between July 2013 and January 2015. Trial identifier: NCT01907997
Participants	Sample size: 90 (2 groups, n = 45, respectively) Inclusion criteria The patients were aged between 20 and 65 years, had ASA grades of 1 or 2, and were scheduled to undergo elective robot‐assisted thyroidectomy. Exclusion criteria History of chronic pain, chronic use of analgesics, allergy to local anaesthetics, severe cardiopulmonary, hepatic or renal disease, diabetes, and neuropsychiatric disease.
Interventions	Experimental group (45 patients) In the lidocaine group (Group L), 0.1 mL/kg of 2% lidocaine (2 mg/kg) was infused IV for 10 mins immediately after anaesthesia induction, and then, it was continuously infused at a rate of 0.15 mL/kg/h of 2% lidocaine (3 mg/kg/h) until the patients were extubated. Control group (45 patients) The control group (Group C) received the same volumes of 0.9% normal saline during the same time periods.
Outcomes	QoR‐40, pain at admission to and discharge of PACU, at 24 and 48 hours, analgesic consumption, chronic postsurgical pain
Notes	Small trial sample size (< 200 patients) Anaesthesia Anaesthesia was induced using IV propofol (1.5 mg/kg) and remifentanil (1 µg/kg). Endotracheal intubation was performed after sufficient muscle relaxation had been achieved by administering 0.6 mg/kg of rocuronium. Anaesthesia was maintained with sevoflurane and remifentanil. Funding This work supported by a research grant. Notes N/A