Pharmacotherapy for the prevention of chronic pain after surgery in adults

Luis Enrique Chaparro; Shane A Smith; R Andrew Moore; Philip J Wiffen; Ian Gilron

doi:10.1002/14651858.CD008307.pub2

. 2013 Jul 24;2013(7):CD008307. doi: 10.1002/14651858.CD008307.pub2

Pharmacotherapy for the prevention of chronic pain after surgery in adults

Luis Enrique Chaparro ¹, Shane A Smith ², R Andrew Moore ³, Philip J Wiffen ⁴, Ian Gilron ^5,^✉

Editor: Cochrane Pain, Palliative and Supportive Care Group

PMCID: PMC6481826 PMID: 23881791

Abstract

Background

Chronic pain can often occur after surgery, substantially impairing patients’ health and quality of life. It is caused by complex mechanisms that are not yet well understood. The predictable nature of most surgical procedures has allowed for the conduct of randomized controlled trials of pharmacological interventions aimed at preventing chronic postsurgical pain.

Objectives

The primary objective was to evaluate the efficacy of systemic drugs for the prevention of chronic pain after surgery by examining the proportion of patients reporting pain three months or more after surgery. The secondary objective was to evaluate the safety of drugs administered for the prevention of chronic pain after surgery.

Search methods

We identified randomized controlled trials (RCTs) of various systemically administered drugs for the prevention of chronic pain after surgery from CENTRAL, MEDLINE, EMBASE and handsearches of other reviews and trial registries. The most recent search was performed on 17 July 2013.

Selection criteria

Included studies were double‐blind, placebo‐controlled, randomized trials involving adults and evaluating one or more drugs administered systemically before, during or after surgery, or both, which measured pain three months or more after surgery.

Data collection and analysis

Data collected from each study included the study drug name, dose, route, timing and duration of dosing; surgical procedure; proportion of patients reporting any pain three months or more after surgery, reporting at least 4/10 or moderate to severe pain three months or more after surgery; and proportion of participants dropping out of the study due to treatment‐emergent adverse effects.

Main results

We identified 40 RCTs of various pharmacological interventions including intravenous ketamine (14 RCTs), oral gabapentin (10 RCTs), oral pregabalin (5 RCTs), non‐steroidal anti‐inflammatories (3 RCTs), intravenous steroids (3 RCTs), oral N‐methyl‐D‐aspartate (NMDA) blockers (3 RCTs), oral mexiletine (2 RCTs), intravenous fentanyl (1 RCT), intravenous lidocaine (1 RCT), oral venlafaxine (1 RCT) and inhaled nitrous oxide (1 RCT). Meta‐analysis suggested a modest but statistically significant reduction in the incidence of chronic pain after surgery following treatment with ketamine but not gabapentin or pregabalin. Results with ketamine should be viewed with caution since most of the included trials were small (that is < 100 participants per treatment arm), which could lead to the overestimation of treatment effect.

Authors' conclusions

Additional evidence from better, well designed, large‐scale trials is needed in order to more rigorously evaluate pharmacological interventions for the prevention of chronic pain after surgery. Furthermore, available evidence does not support the efficacy of gabapentin, pregabalin, non‐steroidal anti‐inflammatories, intravenous steroids, oral NMDA blockers, oral mexiletine, intravenous fentanyl, intravenous lidocaine, oral venlafaxine or inhaled nitrous oxide for the prevention of chronic postoperative pain.

Plain language summary

Systemic drugs for the prevention of chronic pain after surgery

Pain associated with surgery generally resolves within one to two weeks, however in some situations surgical patients are left with longstanding pain for months or even years after the surgical procedure. Researchers have studied the ability of various drug treatments to prevent the development of chronic pain after surgery and this systematic review evaluated published studies in this field. Available studies suggest a modest effect of ketamine, compared to placebo, for prevention of chronic pain after surgery, however small study size could lead to an overestimation of this effect. Studies of other drugs such as gabapentin and pregabalin did not suggest the same preventative effect. Additional large studies using improved research methods are necessary to more clearly identify treatments that are beneficial for preventing chronic postsurgical pain.

Background

Description of the condition

Surgical procedures, as a cause of chronic pain, are unique among various kinds of tissue injury because the injury (for example surgical incision) is, to some degree, planned such that preincisional interventions are possible. This provides an opportunity to implement therapies that could prevent the development of the long‐term, costly and devastating complication of chronic postsurgical pain. Studies that follow postsurgical patients over time suggest that various surgical procedures are associated with a high incidence of chronic postsurgical pain, including limb amputation (30% to 50%), breast cancer surgery (20% to 30%), thoracotomy (30% to 40%) and coronary bypass surgery (30% to 50%) (Perkins 2000). Furthermore, data from outpatient pain treatment clinics suggest that surgery is a contributor to chronic pain in 23% of patients (Crombie 1998).

Laboratory and clinical investigations have contributed to a better understanding of the mechanisms and pathogenesis of chronic postsurgical pain (Kehlet 2006). Although chronic postsurgical pain may not be exclusively neuropathic, clinical presentation (for example hyperalgesia, an exaggerated response to painful stimulus at or near the site of surgery; allodynia, pain following an innocuous stimulus at or near the site of surgery) and the association of chronic postsurgical pain with clearly nerve‐injuring procedures (for example amputation) suggest that nerve injury is an important inciting event (Kehlet 2006). Laboratory studies of nerve injury have identified several mechanisms that contribute to persistent pain, including sensitization (leading to exaggerated pain responsiveness) of peripheral nociceptors (that is receptors preferentially sensitive to a noxious stimulus) and associated primary afferent neurons (nerves transmitting sensory information towards the central nervous system), new growth of neuromas (tumours arising in nerve tissue) on pain‐sensing fibres (whose spontaneous activity can cause unexpected shooting pain) and sensitization of spinal cord and nerve cells in the brainstem and brain (again leading to exaggerated pain responsiveness).

Description of the intervention

The pharmacological mechanisms of these complex changes are such that they may be suppressed by several pain relieving drugs including opioids, cyclo‐oxygenase inhibitors (for example non‐steroidal anti‐inflammatory drugs (NSAIDs)), local anesthetics and other sodium channel blockers (for example mexiletine), glutamate antagonists, anticonvulsant drugs (for example gabapentin) and antidepressant drugs (for example amitriptyline) (D'Mello 2008; Woolf 2000).

How the intervention might work

In considering the consequences of injury to nerves or other tissues, or both, it is important to note that the above pharmacological mechanisms may be involved in the induction or the maintenance, or both, of a persistent pain condition (Woolf 1991). This distinction is important because treatments that successfully suppress the induction of chronic postsurgical pain may serve to prevent its occurrence (that is they would only need to be administered during the induction period) whereas treatments which merely suppress maintenance will only reduce pain while the treatment is being actively administered.

Why it is important to do this review

In several animal studies, administration of different drugs in the setting of nerve or tissue injury has been reported to prevent the development of long‐term pain‐relevant sequelae (for example pain transmitter upregulation, hyperalgesia), including gabapentin (Cesena 1999), ketamine (Burton 1999), amitriptyline (McCarson 2005) and local anesthetic nerve blockade (Wen 2007). However, several human postoperative drug trials have failed to demonstrate such benefits (Hayes 2004; Nikolajsen 1997; Nikolajsen 2006), perhaps because of species differences, other complex issues related to the development of chronic pain (for example psychosocial issues, genetic predisposition and previous pain problems) and, most relevant to this review, differences in dose, duration and timing of the intervention.

There remains considerable uncertainty in this area leading to the focus of this review, which was to evaluate the efficacy of various drugs, administered systemically, for the purpose of preventing the development of, or transition to, chronic postsurgical pain. Given recently expressed uncertainty as to whether any systemic drugs have the potential to prevent chronic pain after surgery, the purpose of this broad review was to identify evidence from multiple trials suggesting that any specific drug prevents chronic pain after any specific surgical procedure.

Two systematic review protocols of narrower focus, recently published in the Cochrane Database of Systematic Reviews, are related to this protocol. Krisanaprakornkit 2007 propose to review pharmacological trials for the prevention of phantom limb pain and Andreae 2008 propose to review trials of local anaesthetics for the prevention of chronic pain after surgery.

Objectives

The primary objective was to evaluate the efficacy of systemic drugs for the prevention of chronic pain after surgery by examining the proportion of patients reporting pain three months or more after surgery.
The secondary objective was to evaluate the safety of drugs administered for the prevention of chronic pain after surgery.

Methods

Criteria for considering studies for this review

Types of studies

Double‐blind, placebo‐controlled, randomized trials of one or more drugs administered systemically before, during or after surgery, or both, which measured pain (using a validated pain assessment instrument) three months or more after surgery. All considered trials were graded for risk of bias using the Cochrane risk of bias tool.

Types of participants

Participants of both genders, 18 years of age and older, undergoing planned surgical procedures involving tissue injury.

Types of interventions

Drugs administered immediately before, during or after the procedure by any dose, route or frequency.

Types of outcome measures

Patient‐reported measure of pain three months or more after the procedure.

Primary outcomes

Proportion of participants reporting any pain at the anatomical site of the procedure or pain referred to the surgical site, or both (for example phantom limb pain, shoulder pain referred from the diaphragm etc.), three months or more after the procedure.

Secondary outcomes

Number of participants reporting moderate or severe pain at the anatomical site of the procedure six months or more after the procedure.
Number of participants dropping out of the study due to treatment‐related adverse effects.

Search methods for identification of studies

Electronic searches

The following databases were searched for trials:

Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library (from inception to July 17, 2013);
MEDLINE (from inception to July 17, 2013);
EMBASE (from inception to July 17, 2013).

The search strategy used to search these databases can be seen in Appendix 1.

Searching other resources

Reference lists of identified published review articles on the subject of chronic pain after surgery were also searched for eligible clinical trials, as were the reference lists of any included studies or relevant excluded studies.

Data collection and analysis

Selection of studies

Studies were selected as per the criteria listed above.

Data extraction and management

Data extracted from each study included: study drug name(s), dose(s), route(s), timing and duration; surgical procedure; proportion of patients:

reporting any pain three months or more after surgery,
reporting at least 4/10 or moderate to severe pain three months or more after surgery, and proportion of participants dropping out of the study due to treatment‐emergent adverse effects.

Data extraction was performed by LC and IG by reading each included study report and completing a data extraction form for each article.

Assessment of risk of bias in included studies

All considered studies were graded for risk of bias using the Cochrane risk of bias tool.

Measures of treatment effect

The primary comparison of interest was between study drug(s) and placebo. Comparisons of study drug(s) and any other active treatment comparators were also to be made. Studies were combined if they evaluated the same study drug(s) at roughly similar doses and durations of treatment (for example a study evaluating a single preoperative drug dose would not be compared to another study evaluating several weeks of treatment with the same drug). RevMan 5 was used to analyse the study data for binary outcomes. Sensitivity analyses were used to evaluate the robustness of a particular result by repeating primary analyses without any studies considered to be outliers with respect to study quality, drug dose and duration, or pain measurement scales.

Subgroup analysis and investigation of heterogeneity

Subgroup analyses were performed to compare the frequencies of positive trial results across differing:

surgical procedures,
timing of the intervention,
duration of intervention, and
study participants with respect to preoperative pain at the surgical site.

Results

Description of studies

See the Characteristics of included studies and Characteristics of excluded studies tables for further information.

We identified 40 clinical trials that fulfilled the review inclusion criteria.

Results of the search

In July 2013, our systematic search identified 7640 citations, which were independently screened by two of the review authors (SS, LEC) based on the title and the abstract information. The first screening for obvious exclusions yielded 107 records that were retrieved and reviewed in full text. Finally, 40 studies fulfilled the inclusion criteria. See Figure 1.

Our search for trials registered on the clinicaltrials.gov and controlled‐trials.com databases yielded over 20 ongoing and unpublished studies. See Characteristics of ongoing studies.

Included studies

Study selection

We identified 40 papers that fulfilled the inclusion criteria of the review. Fourteen studies (1388 participants) used ketamine (Chaparro 2010; Crousier 2008; De Kock 2001; Duale 2009; Dullenkopf 2009; Hayes 2004; Katz 2004a; Malek 2006; Perrin 2009; Remerand 2009; Sen 2009a; Suzuki 2006; Sveticic 2008); one study evaluated (S)‐ketamine (77 participants) (Spreng 2010). Ten studies (791 participants) evaluated gabapentin (Amr 2010; Brogly 2008; Clarke 2009; Fassoulaki 2002; Kinney 2011; Moore 2011; Nikolajsen 2006; Sen 2009; Sen 2009a; Ucak 2011). Five studies (509 participants) evaluated pregabalin (Burke 2010; Buvanendran 2010; Gianesello 2012; Kim 2010; Pesonen 2011). Three studies evaluated corticosteroids: one (50 participants) for dexamethasone (Bergeron 2009), another (219 participants) that compared methylprednisolone (Romundstad 2006) versus parecoxib and versus placebo, and a third one (36 participants) using hydrocortisone (Weis 2006). Two studies (166 participants) evaluated mexiletine (Fassoulaki 2001; Fassoulaki 2002). Two studies (932 participants) evaluated ibuprofen (Fransen 2006; Lakdja 1997). Single studies evaluated: amantadine (22 participants) (Eisenberg 2007); dextromethorphan (50 participants) (Ilkjaer 2000); lidocaine (36 participants) (Grigoras 2012); memantine (19 participants) Schley 2007; venlafaxine (150 participants) (Amr 2010); nitrous oxide (640 participants) (Chan 2011); and opioids (65 participants) (Karanikolas 2011).

Study design

This review included placebo‐controlled studies except for one trial which compared several different active treatments (Karanikolas 2011). Several studies included a third arm or more for comparison: two studies included a dose response evaluation (De Kock 2001; Dullenkopf 2009); one study included four comparisons of low dose and high dose of intravenous and epidural ketamine (De Kock 2001). Two studies compared pre versus postincisional administration of the same medication: one study was performed with ketamine (Katz 2004a) and another one with gabapentin (Clarke 2009). We identified 3 head‐to‐head (comparative) studies that compared in a placebo‐controlled fashion the administration of gabapentin versus ketamine (Sen 2009a), gabapentin versus venlafaxine (Amr 2010) and gabapentin versus mexiletine (Fassoulaki 2002). One head‐to‐head study compared methylprednisolone and parecoxib (Romundstad 2006). We identified one study that compared administration of oral mexiletine combined with regional block with local anesthetics versus each intervention alone (Fassoulaki 2001). A five arm trial compared several strategies for the administration of opioids, giving the medication by PCA or through an epidural catheter (Karanikolas 2011). Of the 40 included studies, trial sizes varied from eight (Perrin 2009) to 452 (Fransen 2006) participants per treatment arm.

Surgical procedures

Ten trials evaluated persistent pain after breast surgery: three with ketamine (Chaparro 2010; Crousier 2008; Malek 2006), one with gabapentin and venlafaxine (Amr 2010), one comparing a regional block versus mexiletine (Fassoulaki 2001), another from the same research group comparing gabapentin versus mexiletine (Fassoulaki 2002), one using ibuprofen (Lakdja 1997), one using methylprednisolone versus parecoxib (Romundstad 2006), and one using lidocaine (Grigoras 2012), one using amantadine (Eisenberg 2007). Six trials evaluated persistent pain after hip or knee arthroplasty: two using ketamine (Perrin 2009; Remerand 2009), one gabapentin (Clarke 2009), one pregabalin (Buvanendran 2010), one ibuprofen (Fransen 2006), and one dexamethasone (Bergeron 2009). Five trials assessed chronic pain after abdominal or pelvic surgery: three ketamine studies (De Kock 2001; Katz 2004a; Sen 2009a), one trial using gabapentin (Moore 2011), and one clinical trial using dextromethorphan (Ilkjaer 2000). Four randomized controlled trials (RCTs) assessed patients following limb amputation: one study of ketamine (Hayes 2004), one of gabapentin (Nikolajsen 2006), one of memantine (Schley 2007), and one of fentanyl (Karanikolas 2011). Chronic pain following thoracotomy was evaluated in three trials: two ketamine trials (Duale 2009; Suzuki 2006) and one study that used gabapentin (Kinney 2011). Three studies have evaluated persistent pain after heart surgery: one with pregabalin (Pesonen 2011), one with gabapentin (Ucak 2011), and one evaluating hydrocortisone (Weis 2006). Two studies evaluated persistent pain after spine surgery, both involving treatment with pregabalin (Burke 2010; Gianesello 2012). Long‐term pain outcomes were reported in two thyroidectomy trials, involving treatment with gabapentin (Brogly 2008) or pregabalin (Kim 2010). Single analgesic trials that reported pain outcomes out to three months or longer involved other surgical procedures including inguinal herniorrhaphy (Sen 2009) and hemorrhoidectomy (Spreng 2010). Three trials involved a combination of different surgical procedures (Chan 2011; Dullenkopf 2009; Sveticic 2008).

Characteristics of participants: preoperative pain and analgesic use

Only seven of the 40 clinical trials only included patients that were free of pain or analgesic use before surgery (Brogly 2008; Chaparro 2010; Duale 2009; Eisenberg 2007; Ilkjaer 2000; Pesonen 2011; Sen 2009a). Patients taking any analgesic were excluded from 11 other trials (Amr 2010; Crousier 2008; Fassoulaki 2001; Fassoulaki 2002; Katz 2004a; Kim 2010; Lakdja 1997; Moore 2011; Romundstad 2006; Spreng 2010; Sveticic 2008). Patients taking more than 10 to 20 mg of morphine equivalent were excluded from four trials: two trials in major joint replacement surgery (Perrin 2009; Remerand 2009), one study in thoracotomy (Kinney 2011), and one in spine surgery (Gianesello 2012). Of note, we assumed that patients had at least mild preoperative pain in 10 trials given the indication for surgery: four studies involving joint replacement (Bergeron 2009; Buvanendran 2010; Clarke 2009; Fransen 2006), four studies for limb amputation (Hayes 2004; Karanikolas 2011; Nikolajsen 2006; Schley 2007), and two studies for decompressive spine surgery (Burke 2010; Gianesello 2012). Information about preoperative pain or analgesic use was unclear or difficult to infer in seven studies (Chan 2011; De Kock 2001; Dullenkopf 2009; Grigoras 2012; Malek 2006; Sen 2009; Suzuki 2006; Ucak 2011; Weis 2006).

Drug interventions

Ketamine (14 trials) (Table 1): 12 of the 14 studies used a preincisional loading dose of ketamine, ranging from 0.15 to 1 mg/kg, plus an intraoperative infusion; two continued the infusion for the first 24 hours (Duale 2009; Remerand 2009), one for 48 hours (Malek 2006) and two additional studies maintained the ketamine infusion for three days (Hayes 2004; Suzuki 2006). One study mixed ketamine with morphine in a patient‐controlled analgesia (PCA) device (Sveticic 2008). The cumulative dose of ketamine reached less than 1 mg/kg in 6/14 studies (Chaparro 2010; Crousier 2008; Dullenkopf 2009; Katz 2004a; Sen 2009a; Spreng 2010); 4/14 studies administered a cumulative dose of ketamine that ranged between 1 and 2 mg/kg (De Kock 2001; Malek 2006; Perrin 2009; Sveticic 2008) and four studies used a cumulative dosage > 2 mg/kg (Duale 2009; Hayes 2004; Remerand 2009; Suzuki 2006).

Gabapentin (10 trials) (Table 3): nine of the 10 studies administered a preincisional dose of gabapentin; this dosage ranged between 300 mg and 1200 mg; 6/10 studies evaluated a single dose of 600 mg (Clarke 2009; Kinney 2011; Moore 2011) or 1200 mg (Brogly 2008; Sen 2009; Sen 2009a); Of the other four multidose studies, one used gabapentin for two days in cardiac surgery (Ucak 2011), two evaluated 10 days of gabapentin administration in breast surgery (Amr 2010; Fassoulaki 2002) and one studied 30 days of gabapentin administration in an amputation trial (Nikolajsen 2006).

Pregabalin (five trials) (Table 3): all five pregabalin trials used a preincisional dose of 150 mg (Kim 2010; Pesonen 2011) or 300 mg (Burke 2010; Buvanendran 2010; Gianesello 2012) and at least one additional postoperative dose. The postoperative administration scheme for the drug varied across the trials from a single postoperative dose (Burke 2010; Kim 2010) to 14 days (Buvanendran 2010).

Corticosteroids (three trials): two of these three trials evaluated single preoperative doses, dexamethasone (40 mg) for total hip arthroplasty (Bergeron 2009) or methylprednisolone (125 mg) for augmentation mammaplasty (Romundstad 2006). In the third trial hydrocortisone was administered using a loading dose, intraoperative infusion and a tapering scheme up to the fourth postoperative day (Weis 2006). Further details about these studies are described in the Characteristics of included studies.

Ibuprofen (two trials): we found two ibuprofen trials, one that administered the analgesic for two weeks in patients undergoing total hip arthroplasty (Fransen 2006) and the other for only two days, including a preoperative dose, for breast surgery (Lakdja 1997). Further details about these studies are in the Characteristics of included studies.

Mexiletine (two trials): in two studies of patients undergoing breast cancer surgery (Fassoulaki 2001; Fassoulaki 2002), both trials had an active control arm, brachial plexus infiltration with ropivacaine (Fassoulaki 2001) or gabapentin (Fassoulaki 2002); mexiletine was administered for six or 10 days respectively.

Other drugs: we identified various other single drug trials, one study that evaluated amantadine administration for two weeks (Eisenberg 2007) and another that administered memantine for four weeks after traumatic amputation (Schley 2007); one study evaluated the intraoperative use of intravenous lidocaine (Grigoras 2012); one trial evaluated a single preoperative dose of dextromethorphan (Ilkjaer 2000) and another evaluated the intraoperative administration of nitrous oxide (70%) (Chan 2011). A single dose of parecoxib (40 mg) was administered as the active control arm of the aforementioned study of methylprednisolone (Romundstad 2006). Only one study using an antidepressant was identified (Amr 2010); venlafaxine was started before surgery and continued for 10 days at 37.5 mg/night. One five‐arm study evaluated different strategies of intravenous versus epidural administration of fentanyl for two postoperative days (Karanikolas 2011). Further details about these studies are described in the Characteristics of included studies.

Outcome measures

Primary outcome

Any surgical site pain three or more months after surgery: the incidence of 'any pain' at any time point was reported in 26/40 studies (Chan 2011; Chaparro 2010; Clarke 2009; Crousier 2008; Duale 2009; Fassoulaki 2001; Fassoulaki 2002; Gianesello 2012; Grigoras 2012; Hayes 2004; Ilkjaer 2000; Katz 2004a; Kim 2010; Kinney 2011; Lakdja 1997; Malek 2006; Moore 2011; Nikolajsen 2006; Perrin 2009; Remerand 2009; Romundstad 2006; Schley 2007; Spreng 2010; Suzuki 2006; Sveticic 2008; Ucak 2011). Twenty‐four trials collected this information using a 0 to 10 pain intensity scale (Bergeron 2009; Burke 2010; Clarke 2009; Duale 2009; Dullenkopf 2009; Fassoulaki 2001; Fassoulaki 2002; Fransen 2006; Grigoras 2012; Hayes 2004; Karanikolas 2011; Katz 2004a; Moore 2011; Nikolajsen 2006; Remerand 2009; Romundstad 2006; Schley 2007; Sen 2009a; Sen 2009; Spreng 2010; Suzuki 2006; Sveticic 2008; Ucak 2011; Weis 2006); 7/40 studies used a categorical pain scale for pain measurement (Chan 2011; Dullenkopf 2009; Gianesello 2012; Malek 2006; Nikolajsen 2006; Perrin 2009; Pesonen 2011). Additionally, only seven studies reported movement‐evoked pain at three or more months after surgery (Amr 2010; Burke 2010; Dullenkopf 2009; Grigoras 2012; Pesonen 2011; Romundstad 2006; Spreng 2010).

Secondary outcomes

Moderate or severe surgical site pain six months after surgery: trials reporting data on this secondary outcome for this systematic review included three studies that followed up the patients at three months (Gianesello 2012; Nikolajsen 2006; Pesonen 2011); three studies that evaluated this outcome at six months (Buvanendran 2010; Malek 2006; Nikolajsen 2006); and one trial that included a one year follow‐up (Gianesello 2012). Two trials quantified the incidence of mild or moderate pain, one at three months (Chan 2011) and the other at six months (Perrin 2009). Only one study reported the number of patients that persisted with severe pain (VAS > 5/10) (Chan 2011). A single study opted for the evaluation of pain control as poor, acceptable, or excellent (Dullenkopf 2009). The information on moderate to severe pain was retrieved on two occasions by the authors (Buvanendran 2010; Nikolajsen 2006).

Proportion of participants dropping out of the study due to treatment‐related adverse effects: only 6/40 studies reported this outcome. One patient had to cancel the surgery due to severe sedation after 300 mg of pregabalin (Burke 2010); a similar situation occurred with one patient that received pregabalin 300 mg before total knee arthroplasty (Buvanendran 2010); furthermore, one patient that received 600 mg of gabapentin before cesarean section dropped out from the study due to severe sedation (Moore 2011). In another gabapentin trial, in patients undergoing lower limb amputation, 2/23 in the active group and 2/23 in the placebo group dropped out from the study due to drug‐related adverse events (Nikolajsen 2006). In another trial, 51/452 in the ibuprofen group versus 37/450 discontinued the treatment due to side effects or intolerance to the intervention (Fransen 2006). Of note, one trial that evaluated nitrous oxide reported that 217/640 patients were deceased before the follow‐up (Chan 2011); 85% died secondary to cancer complications.

Other relevant outcome data: the incidence of pain or sensory abnormalities was reported at the three to four months time point by 24/40 studies (Burke 2010; Buvanendran 2010; Chan 2011; Crousier 2008; Duale 2009; Dullenkopf 2009; Eisenberg 2007; Fassoulaki 2001; Fassoulaki 2002; Gianesello 2012; Grigoras 2012; Ilkjaer 2000; Kim 2010; Kinney 2011; Moore 2011; Nikolajsen 2006; Pesonen 2011; Remerand 2009; Sen 2009a; Sen 2009; Spreng 2010; Suzuki 2006; Sveticic 2008; Ucak 2011); at six months by 21/40 (Amr 2010; Brogly 2008; Buvanendran 2010; Clarke 2009; De Kock 2001; Eisenberg 2007; Fransen 2006; Hayes 2004; Karanikolas 2011; Katz 2004a; Lakdja 1997; Malek 2006; Nikolajsen 2006; Perrin 2009; Remerand 2009; Schley 2007; Sen 2009a; Sen 2009; Suzuki 2006; Sveticic 2008; Weis 2006) and six studies reported outcomes at the 12 months time point (Bergeron 2009; Chaparro 2010; De Kock 2001; Gianesello 2012; Romundstad 2006; Schley 2007). A significant number of studies (15/40) reported pain‐related descriptors or the number of chosen words from the McGill pain questionnaire (MPQ) (Amr 2010; Brogly 2008; Chan 2011; Chaparro 2010; Crousier 2008; De Kock 2001; Eisenberg 2007; Fassoulaki 2001; Fassoulaki 2002; Hayes 2004; Katz 2004a; Kim 2010; Moore 2011; Romundstad 2006; Suzuki 2006); the questionnaires were collected by mail or telephone interviews. Only a few studies opted for a clinical visit for physical examination and evaluation of sensory abnormalities (Buvanendran 2010; Crousier 2008; Duale 2009; Eisenberg 2007; Grigoras 2012; Ilkjaer 2000; Lakdja 1997). Validated scores for detection of neuropathic pain were used in three studies: one study opted for the neuropathic pain diagnostic questionnaire (DN2) score (Brogly 2008), another chose the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) score (Buvanendran 2010) and a third one evaluated the pain using the neuropathic pain scale (Clarke 2009). Information about analgesic requirement was collected in 16/40 studies (Amr 2010; Buvanendran 2010; Chan 2011; De Kock 2001; Duale 2009; Eisenberg 2007; Fassoulaki 2001; Fassoulaki 2002; Fransen 2006; Hayes 2004; Nikolajsen 2006; Pesonen 2011; Remerand 2009; Romundstad 2006; Spreng 2010; Suzuki 2006). Functional outcomes or interference with activities of daily living by pain were evaluated in 18 studies (Bergeron 2009; Burke 2010; Buvanendran 2010; Chan 2011; Clarke 2009; Crousier 2008; Duale 2009; Fransen 2006; Gianesello 2012; Katz 2004a; Moore 2011; Perrin 2009; Remerand 2009; Romundstad 2006; Sen 2009a; Sen 2009; Ucak 2011; Weis 2006). Mood disorder scales were evaluated in three studies only (Clarke 2009; Gianesello 2012; Grigoras 2012).

Excluded studies

Given this review's exclusion criteria, trials evaluating regional anaesthesia techniques (Arcioni 2007; Bell 2001; Chiu 2008; Cohen 2006; Essving 2009; Essving 2010; Gupta 2006; Honigmann 2007; Jirarattanaphochai 2007; Kadic 2009; Kampe 2003; Katz 2004; Lambert 2001; Martin 2008; Morin 2005; Nikolajsen 2000; Obata 1999; Perniola 2009; Ryu 2011; Sanders 2009; Senturk 2002; Singh 2007) and that targeted injection therapies (Hartrick 2011; Wai 2010) or topical analgesia (Fassoulaki 2000) were excluded. We also excluded a clinical trial with one arm that combined topical analgesia and gabapentin (Fassoulaki 2005) since it was unknown how much the oral gabapentin effect contributed to the reported efficacy results of this multimodal intervention. Another study was excluded given that gabapentin was administered only to patients who developed moderate or severe pain (Elkaradawy 2012). Other studies were excluded because they had pain outcomes which were earlier than three months after surgery (Buvanendran 2003; Fassoulaki 2006; Fassoulaki 2007; Hartrick 2011; Kim 2011; Nissman 2008; White 2007). For more details please see the Characteristics of excluded studies table and Figure 1.

Risk of bias in included studies

Each study was assessed independently for risk of bias by two of the review authors (LC and IG) using the 'Risk of bias' tool (ROB) (Higgins 2011); 28/40 included studies had at least 4/7 items that qualified as low risk of bias. Several studies did not state the method for sequence generation or allocation concealment. Since most study medications were associated with recognizable adverse effects (for example sedation) in the acute setting, methods to prevent knowledge or evaluate (for example blinding questionnaires) the quality of blinding of outcomes assessors were not adequately addressed. Details are in the Characteristics of included studies table. Risk of bias assessments are presented in Figure 2 and Figure 3.

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Allocation

Seventeen of the 40 included trials reported the method used to generate a random sequence and to keep the allocation concealed (Amr 2010; Buvanendran 2010; Chan 2011; Clarke 2009; Dullenkopf 2009; Fassoulaki 2002; Fransen 2006; Gianesello 2012; Grigoras 2012; Karanikolas 2011; Katz 2004a; Moore 2011; Nikolajsen 2006; Pesonen 2011; Remerand 2009; Romundstad 2006; Sen 2009a); and 15 additional studies (Brogly 2008; Burke 2010; Chaparro 2010; De Kock 2001; Duale 2009; Hayes 2004; Kinney 2011; Malek 2006; Perrin 2009; Sen 2009; Spreng 2010; Suzuki 2006; Sveticic 2008; Ucak 2011; Weis 2006) appropriately reported one or the other item.

Blinding

The vast majority of the trials reported how participants were blinded; however, 16/40 did not adequately described methods for blinding the outcomes assessors (see Figure 3).

Incomplete outcome data

Attrition bias was assessed as 'low risk' for studies where the dropout rate was below 20% (Bhandari 2005). Studies with higher dropout rates but including ITT analyses were assessed as 'unclear' or 'high risk of bias'.

Selective reporting

Although few studies indicated pre‐trial registration on a clinical trial registry, all reported at least one of the outcomes that are considered to be clinically relevant (Dworkin 2005). This section was difficult to assess without access to the study protocol.

Other potential sources of bias

Given that the primary outcome was rarely the incidence of chronic pain, most of the included trials were underpowered for this particular outcome. This item was assessed as high risk in studies that had fewer than 50 participants per arm (Moore 1998).

Effects of interventions

Forty trials were included in this review.

Ketamine

Fourteen studies evaluating the perioperative effectiveness of ketamine were identified (Chaparro 2010; Crousier 2008; De Kock 2001; Duale 2009; Dullenkopf 2009; Hayes 2004; Katz 2004a; Malek 2006; Perrin 2009; Remerand 2009; Sen 2009a; Spreng 2010; Suzuki 2006; Sveticic 2008).

The studies are described in Table 1 in terms of the number of perioperative periods where the participants were exposed to ketamine: 1. preincisional loading dose; 2. intraoperative infusion; and 3. postoperative infusion (see Table 1). Three studies, one in thoracotomy (Duale 2009), a second one in amputation (Hayes 2004) and another one in total hip arthroplasty (Remerand 2009), administered ketamine in each of the periods of perioperative treatment; seven clinical trials used an incisional loading dose plus intraoperative infusion (Chaparro 2010; Crousier 2008; De Kock 2001; Katz 2004a; Perrin 2009; Sen 2009a; Spreng 2010); two studies started with an intraoperative infusion and continued the treatment for two (Malek 2006) or three (Suzuki 2006) postoperative days; only two studies limited the administration of ketamine to one of the three potential periods of treatment, one study used two different preincisional loading doses (Dullenkopf 2009) and one study administered ketamine mixed in the patient‐controlled analgesia (PCA) device with morphine (Sveticic 2008).

1. Ketamine trials.

Study ID	Surgery	Preoperative Dose (μg/kg)	Intraoperative Dose (μg/kg/h)	Postoperative Dose (μg/kg/h)	Cumulative dose (μg/kg)
Chaparro 2010	Breast surgery	420	200	none	420 + (200 x 2h) = 820*
Crousier 2008	Breast surgery	500	250	none	500 + (250 x 1.63h) = 908*
De Kock 2001	Rectal carcinoma resection	250	125	none	250 + (125 X 4.3h) = 772*
De Kock 2001	Rectal carcinoma resection	500	250	none	500 + (250 X 4.38h) = 1,595*
Duale 2009	Thoracotomy	1000	1000	1000 for 24h	3,000*
Dullenkopf 2009	Orthopedic surgery	150	none	none	150
Dullenkopf 2009	Orthopedic surgery	500	none	none	500
Hayes 2004	Amputation	500	150 for 72h		11,300
Katz 2004a	Radical prostatectomy	200	175 in 70 minutes	none	375
Katz 2004a	Radical prostatectomy	None	200 + 175 in 70 minutes	none	375
Malek 2006	Breast surgery	None	1,000/day for 48 hours		2,000
Perrin 2009	Total knee arthroplasty	500	240	none	500 + (4x125.4)= 1,001.6*
Remerand 2009	Total hip arthroplasty	500	120 for 24h		3,380*
Sen 2009a	Hysterectomy	300	50	none	300 + (50 x 1.25h)= 362.5*
Spreng 2010	Hemorrhoidectomy	350	300	none	350+ (300 x 0.33)= 451.5*
Suzuki 2006	Thoracotomy	none	50 for 72h		3,600*
Sveticic 2008	Orthopedic surgery	none	none	1.5 mg**	1,198***

Study ID	Surgery	Preoperative Dose (mg)	Postoperative Dose (mg/day)	Cumulative Dose (mg)	Cumulative Days
Amr 2010	Breast surgery	300	300	3,000	10
Brogly 2008	Thyroidectomy	1,200	none	1,200	single dose
Clarke 2009	Total hip arthroplasty	600	none	600	single dose
Clarke 2009	Total hip arthroplasty	none	600	600	single dose
Fassoulaki 2002	Breast surgery	800	1200	12,000	10
Kinney 2011	Thoracotomy	600	none	600	single dose
Moore 2011	Caesarean	600	none	600	single dose
Nikolajsen 2006	Amputation	None	300 to 2,400	75,600	30
Sen 2009a	Hysterectomy	1,200	none	1,200	single dose
Sen 2009	Inguinal herniorrhaphy	1,200	none	1,200	single dose
Ucak 2011	Cardiac surgery	1,200	1,200	3,600	2

Study ID	Drug	Surgery	Preoperative Dose (mg)	Postoperative Dose (mg/day)	Cumulative Dose (mg)	Cumulative Days
Amr 2010	Venlafaxine	Mastectomy	37.5	37.5	375	10 days
Bergeron 2009	Dexamethasone	Hip arthroplasty	40	none	40	Single dose
Romundstad 2006	Methyl‐prednisolone	Breast augmentation	125 mg	none	125	Single dose
Weis 2006	Hydrocortisone	Cardiac surgery	100	240 to 30	550	4 days
Chan 2011	Nitrous oxide	Numerous	Intraoperative 70%		N.A.	Intraoperative
Eisenberg 2007;	Amantadine	Mastectomy	100	200	2,800	14 days
Ilkjaer 2000	Dextro‐methorphan	Hysterectomy	750	none	750	Single dose
Schley 2007	Memantine	Amputation	none	10 to 40	630	28 days
Grigoras 2012	Lidocaine	Breast surgery	Bolus 1.5 mg/kg + 1.5 mg/kg/hour up to the first hour after surgery.
Fassoulaki 2001	Mexiletine	Mastectomy	200	400	2,400	6 days
Fassoulaki 2002	Mexiletine	Breast surgery	400	600	6,000	10 days
Fransen 2006	Ibuprofen	Hip arthroplasty	None	1,200	16,800	14 days
Lakdja 1997	Ibuprofen	Mastectomy	400	1,600	2,000	2 days
Romundstad 2006	Parecoxib	Breast augmentation	40 mg	none	40 mg	Single dose
Karanikolas 2011	Fentanyl	Amputation	58.3 mcg/h	54.5 mcg/h	Variable	2 days

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1.1 Incidence of any pain at 3 months (all studies)	5	384	Risk Ratio (M‐H, Fixed, 95% CI)	0.82 [0.61, 1.11]
1.1.1 Breast Surgery	1	30	Risk Ratio (M‐H, Fixed, 95% CI)	1.15 [0.79, 1.69]
1.1.2 Hemorrhoidectomy	1	72	Risk Ratio (M‐H, Fixed, 95% CI)	1.00 [0.15, 6.72]
1.1.3 Orthopedic Surgery	2	238	Risk Ratio (M‐H, Fixed, 95% CI)	0.85 [0.54, 1.32]
1.1.4 Thoracotomy	1	44	Risk Ratio (M‐H, Fixed, 95% CI)	0.50 [0.25, 1.00]
1.2 Incidence of any pain at 3 months (drug administration ≤ 24 hours)	3	249	Risk Ratio (M‐H, Fixed, 95% CI)	0.96 [0.69, 1.35]
1.2.1 Breast Surgery	1	30	Risk Ratio (M‐H, Fixed, 95% CI)	1.15 [0.79, 1.69]
1.2.2 Hemorrhoidectomy	1	72	Risk Ratio (M‐H, Fixed, 95% CI)	1.00 [0.15, 6.72]
1.2.3 Orthopedic Surgery	1	147	Risk Ratio (M‐H, Fixed, 95% CI)	0.88 [0.55, 1.41]
1.3 Incidence of any pain at 3 months (drug administration > 24 hours)	2	135	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.28, 1.00]
1.3.1 Orthopedic Surgery	1	91	Risk Ratio (M‐H, Fixed, 95% CI)	0.64 [0.15, 2.71]
1.3.2 Thoracotomy	1	44	Risk Ratio (M‐H, Fixed, 95% CI)	0.50 [0.25, 1.00]
1.4 Incidence of any pain at 4 months	1	69	Risk Ratio (M‐H, Fixed, 95% CI)	0.84 [0.48, 1.46]
1.5 Incidence of any pain at 6 months (all studies)	8	516	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.47, 0.83]
1.5.1 Abdominal and/or pelvic surgery	2	164	Risk Ratio (M‐H, Fixed, 95% CI)	0.40 [0.19, 0.88]
1.5.2 Amputation	1	32	Risk Ratio (M‐H, Fixed, 95% CI)	0.66 [0.35, 1.23]
1.5.3 Breast surgery	1	86	Risk Ratio (M‐H, Fixed, 95% CI)	0.87 [0.47, 1.60]
1.5.4 Orthopedic surgery	3	190	Risk Ratio (M‐H, Fixed, 95% CI)	0.65 [0.40, 1.06]
1.5.5 Thoracotomy	1	44	Risk Ratio (M‐H, Fixed, 95% CI)	0.55 [0.25, 1.21]
1.6 Incidence of any pain at 6 months (drug administration ≤ 24 hours)	4	318	Risk Ratio (M‐H, Fixed, 95% CI)	0.55 [0.36, 0.83]
1.6.1 Abdominal and/or pelvic surgery	2	164	Risk Ratio (M‐H, Fixed, 95% CI)	0.40 [0.19, 0.88]
1.6.2 Orthopedic Surgery	2	154	Risk Ratio (M‐H, Fixed, 95% CI)	0.64 [0.39, 1.04]
1.7 Incidence of any pain at 6 months (drug administration > 24 hours)	4	198	Risk Ratio (M‐H, Fixed, 95% CI)	0.72 [0.49, 1.06]
1.7.1 Amputation	1	32	Risk Ratio (M‐H, Fixed, 95% CI)	0.66 [0.35, 1.23]
1.7.2 Breast Surgery	1	86	Risk Ratio (M‐H, Fixed, 95% CI)	0.87 [0.47, 1.60]
1.7.3 Orthopedic Surgery	1	36	Risk Ratio (M‐H, Fixed, 95% CI)	1.12 [0.08, 16.52]
1.7.4 Thoracotomy	1	44	Risk Ratio (M‐H, Fixed, 95% CI)	0.55 [0.25, 1.21]
1.8 Incidence of Moderate or severe pain at 6 months	3	259	Risk Ratio (M‐H, Fixed, 95% CI)	0.50 [0.26, 0.95]
1.8.1 Amputation	1	31	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.32, 3.52]
1.8.2 Breast Surgery	1	86	Risk Ratio (M‐H, Fixed, 95% CI)	0.48 [0.18, 1.32]
1.8.3 Total hip arthroplasty	1	142	Risk Ratio (M‐H, Fixed, 95% CI)	0.29 [0.08, 1.02]
1.9 Incidence of any pain at 12 months	2	104	Risk Ratio (M‐H, Fixed, 95% CI)	0.26 [0.06, 1.15]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
2.1 Incidence of any pain at 3 months (all studies)	5	280	Risk Ratio (M‐H, Fixed, 95% CI)	0.99 [0.80, 1.21]
2.1.1 Amputation (phantom limb pain)	1	38	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.61, 2.03]
2.1.2 Breast Surgery	1	46	Risk Ratio (M‐H, Fixed, 95% CI)	0.94 [0.56, 1.56]
2.1.3 Caesarean section	1	36	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.13, 2.99]
2.1.4 Cardiac Surgery	1	40	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.36, 1.68]
2.1.5 Thoracotomy	1	120	Risk Ratio (M‐H, Fixed, 95% CI)	1.05 [0.83, 1.34]
2.2 Incidence of any pain at 3 months (drug administration ≤ 24 hours)	2	156	Risk Ratio (M‐H, Fixed, 95% CI)	1.02 [0.79, 1.30]
2.2.1 Caesarean section	1	36	Risk Ratio (M‐H, Fixed, 95% CI)	0.63 [0.13, 2.99]
2.2.2 Thoracotomy	1	120	Risk Ratio (M‐H, Fixed, 95% CI)	1.05 [0.83, 1.34]
2.3 Incidence of any pain at 3 months (drug administration > 24 hours)	3	124	Risk Ratio (M‐H, Fixed, 95% CI)	0.94 [0.66, 1.34]
2.3.1 Amputation (phantom limb pain)	1	38	Risk Ratio (M‐H, Fixed, 95% CI)	1.11 [0.61, 2.03]
2.3.2 Breast surgery	1	46	Risk Ratio (M‐H, Fixed, 95% CI)	0.94 [0.56, 1.56]
2.3.3 Cardiac surgery	1	40	Risk Ratio (M‐H, Fixed, 95% CI)	0.78 [0.36, 1.68]
2.4 Incidence of any pain at 6 months	2	116	Risk Ratio (M‐H, Fixed, 95% CI)	1.10 [0.72, 1.68]
2.4.1 Amputation (phantom limb pain)	1	34	Risk Ratio (M‐H, Fixed, 95% CI)	1.12 [0.64, 1.97]
2.4.2 Total hip arthroplasty	1	82	Risk Ratio (M‐H, Fixed, 95% CI)	1.09 [0.60, 1.98]
2.5 Mean Pain Score at 6 months (continuous data)	1	100	Mean Difference (IV, Fixed, 95% CI)	‐1.00 [‐5.16, 3.16]

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
3.1 Incidence of any pain at 3 months follow up (all studies)	4	439	Risk Ratio (M‐H, Fixed, 95% CI)	0.70 [0.51, 0.95]
3.1.1 Cardiac surgery	1	57	Risk Ratio (M‐H, Fixed, 95% CI)	0.14 [0.02, 1.04]
3.1.2 Total Knee Arthroplasty	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.64 [0.40, 1.02]
3.1.3 Spine Surgery	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.63, 1.81]
3.1.4 Thyroidectomy	1	94	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.38, 1.58]
3.2 Incidence of any pain at 3 months (drug administration ≤ 24 hours)	1	94	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.38, 1.58]
3.2.1 Thyroidectomy	1	94	Risk Ratio (M‐H, Fixed, 95% CI)	0.77 [0.38, 1.58]
3.3 Incidence of any pain at 3 months (drug administration > 24 hours)	3	345	Risk Ratio (M‐H, Fixed, 95% CI)	0.68 [0.48, 0.96]
3.3.1 Cardiac Surgery	1	57	Risk Ratio (M‐H, Fixed, 95% CI)	0.14 [0.02, 1.04]
3.3.2 Total knee arthroplasty	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.64 [0.40, 1.02]
3.3.3 Spine surgery	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	1.07 [0.63, 1.81]
3.4 Incidence of any pain at 6 months	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.30, 0.93]
3.4.1 Total Knee Arthroplasty	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.53 [0.30, 0.93]
3.5 Incidence of moderate to severe pain at 3 months	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.43 [0.20, 0.94]
3.5.1 Total Knee Arthroplasty	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.43 [0.20, 0.94]
3.6 Incidence of moderate to severe pain at 6 months	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.23 [0.07, 0.80]
3.6.1 Total knee arthroplasty	1	228	Risk Ratio (M‐H, Fixed, 95% CI)	0.23 [0.07, 0.80]
3.7 Incidence of any pain at 12 months follow up	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	0.62 [0.23, 1.69]
3.7.1 Major spinal surgery	1	60	Risk Ratio (M‐H, Fixed, 95% CI)	0.62 [0.23, 1.69]

*Study characteristics*
Methods	Single center, randomized, double blind, 3 arms, placebo‐controlled trial for 6 months
Participants	150 patients scheduled for either partial or radical mastectomy with axillary dissection
Interventions	In the (Group 1) venlafaxine group, patients received 37.5 mg of venlafaxine extended release along with another capsule (identical in appearance to the gabapentin capsules) once daily at bed time. Patients in the (Group 2) gabapentin group received 300 mg of gabapentin and another capsule (identical in appearance to the venlafaxine extended release capsules) once daily at bed time. Patients in the (Group 3) placebo group received 2 capsules that were filled with thin sugar (1 identical in appearance to the gabapentin capsules and the other identical to venlafaxine extended release capsules) once daily at bed time. Administration of the drugs started the evening before surgery and continued for the first 10 postoperative days, including the day of the surgery.
Outcomes	Pain scores at rest and movement‐induced, morphine consumption, and side effects profile
Notes	Co‐analgesia: morphine the first 24 hours; acetaminophen + codeine during the rest of the follow‐up
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Shuffling envelopes: "A prospective, randomized (sealed envelopes indicate the group of assignment)...
Allocation concealment (selection bias)	Low risk	"An independent anesthesiologist, who did not participate in the study or data collection, read the number contained in the envelope and made group assignments".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Control (sugar filled) and/or treatment capsules for each group were packaged in group‐specific bottles and coded as Bottle 1, Bottle 2, and Bottle 3 for Groups 1, 2, and 3, respectively. Yellow hard gelatin capsules (identical in appearance to the gabapentin capsules) and gray/peach capsules (identical in appearance to the venlafaxine extended release capsules) were filled with thin sugar. Envelopes, bottles with capsules, and coding were prepared by the pharmacy of the hospital.
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"double‐blind design was used, with both patients and postoperative assessors blinded to management protocol"
Incomplete outcome data (attrition bias) All outcomes	Low risk	The authors reported no missing outcome data
Selective reporting (reporting bias)	Low risk	Outcomes reported are clinically meaningful and validated
Other bias	Low risk	No other potential sources of bias were detected

*Study characteristics*
Methods	Single center, randomized, double‐blind, placebo‐controlled trial for 6 months
Participants	50 patients aged 18‐75 yr, ASA I‐III, undergoing scheduled total or partial thyroidectomy without lymph node dissection
Interventions	Gabapentin 1200 mg PO 2 hours before surgery or matching placebos (placebos are not described)
Outcomes	Primary outcome: analgesic drug consumption was assessed during the procedure and postoperatively in the postanesthesia care unit and after discharge to the ward. Over the first 24 h, pain levels at rest and during swallowing were measured. The day before operation and 6 mo after patients were asked to answer a neuropathic pain diagnostic questionnaire (DN2).
Notes	Coanalgesia: Superficial cervical plexus block for all participants. analgesic rescue: paracetamol + tramadol 50 mg
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomization was performed by subgroups of 10 patients with a randomization table".
Allocation concealment (selection bias)	Unclear risk	Not reported
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Not reported
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Not reported
Incomplete outcome data (attrition bias) All outcomes	Low risk	Missing outcome data (3/50) was minimum
Selective reporting (reporting bias)	Unclear risk	Insufficient information to permit judgment
Other bias	High risk	Fewer than 50 patients per arm

*Study characteristics*
Methods	Multicenter, randomized, double‐blind, controlled trial with long‐term follow‐up for at least three months
Participants	640 patients that were 18 yr or older scheduled, under general anesthesia, for surgery that included a skin incision and that was anticipated to exceed 2 hours, and were expected to be in the hospital for at least 3 days after surgery
Interventions	Intraoperative 70% nitrous oxide
Outcomes	The primary outcome of this follow‐up study was chronic postsurgical pain according to the definition by the International Association for the Study of Pain. Secondary outcomes were the impact of chronic postsurgical pain on daily living.
Notes	This study had a very high mortality rate at follow‐up (34%)
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"….using a computer‐generated code, accessed via an automated telephone voice recognition service".
Allocation concealment (selection bias)	Low risk	"…..automated telephone voice recognition service. Treatment assignment was stratified by site and elective/emergency status of the surgery, using permuted blocks".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"Attending anesthesiologists were required to have knowledge of group identity for the safe administration of anesthesia, but group identity was concealed from the surgeon using drapes or cardboard to screen the anesthesia machine".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"At the end of the procedure, the intraoperative case report form and documentation of group identity were faxed to the data management center and then placed in an opaque envelope by the anesthesiologist. The envelope was then sealed to ensure blinding of research staff conducting the postoperative follow‐ups".
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	This secondary analysis included the Hong Kong population only
Selective reporting (reporting bias)	Low risk	Outcomes reported for this secondary analysis are clinically meaningful and validated
Other bias	High risk	"Pain intensity was not described as an outcome in the original 2007 report".

*Study characteristics*
Methods	Multicenter, randomized, double‐blind, placebo‐controlled trial for a variable follow up time (6‐14 months)
Participants	902 Patients identified within 24 hours of completed THR (primary or revision), irrespective of age, reason for surgery, or procedure performed
Interventions	Ibuprofen 1200 mg Daily (TID) for 14 days
Outcomes	Primary outcomes: changes from baseline to follow‐up in WOMAC. We standardised scores to a range of 0‐10, with 0 indicating no hip pain or no difficulty with daily activities and 10 indicating severe hip pain or severe difficulty with daily activities. Secondary outcomes: Short form 36, compared with before surgery; hip status today with 5 response levels; frequency of use of analgesia for hip pain during the past week; ability to get “about the house” and ability to get “out of the house” with 5 response levels ranging from “not at all” to “no difficulty”; time spent participating in physical activity during the past week; objective measures of physical performance (hip flexion, time to walk 50 feet (about 15 metres), and timed “up and go”; radiographic evidence of ectopic bone formation according to the Brooker classification and major bleeding complications during hospital admission.
Notes	Co‐analgesia was up to treatment group. No changes in the NSAIDs were allowed
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	"Randomisation was performed centrally by using a computer based system"
Allocation concealment (selection bias)	Low risk	"We used a minimisation program to stratify treatment by study centre and type of surgery performed (primary or revision). Treatment allocation was blinded and concealed from patients and study staff until the database was locked".
Blinding of participants and personnel (performance bias) All outcomes	Low risk	"All study tablets were packaged in identical blister packs".
Blinding of outcome assessment (detection bias) All outcomes	Low risk	"All assessments were standardised and performed blind to randomised treatment allocation".
Incomplete outcome data (attrition bias) All outcomes	Low risk	No statistical significant difference in the rate of dropouts. (P=0.06). The proportion of missing outcome data < 20%
Selective reporting (reporting bias)	Low risk	Protocol was registered and no differences were noticed compared with the publication
Other bias	Low risk	No other potential sources of bias were detected

*Study characteristics*
Methods	Single center, randomized, double‐blind (patients and outcomes assessor nurse), placebo‐controlled trial for 6 months
Participants	100 women scheduled for breast cancer surgery
Interventions	Ketamine 1 mg/k/day in IV infusion for 2 days
Outcomes	After 6 months of surgery patients received a questionnaire concerning the presence of chronic pain, its intensity (1 ‐ mild, 2‐medium, 3 ‐ strong, 4 ‐ Unbearable), quality (constant vs intermittent), location (in scars, in the whole breast or chest wall, armpit, elsewhere) and nature (using the modified McGill Pain Questionnaire).
Notes	Co‐analgesics: preoperative and postoperative meperidine + intraoperative sufentanil
*Risk of bias*
Bias	Authors' judgement	Support for judgement
Random sequence generation (selection bias)	Low risk	Drawing lots
Allocation concealment (selection bias)	Unclear risk	Insufficient information to permit judgement
Blinding of participants and personnel (performance bias) All outcomes	Unclear risk	Patients were blinded. However, the person that prepared the solutions is unclear if he/she was blinded
Blinding of outcome assessment (detection bias) All outcomes	Unclear risk	Patient filled out a questionnaire at 6 months. However is unclear if already knew what received during surgery
Incomplete outcome data (attrition bias) All outcomes	Unclear risk	Insufficient information to permit judgement
Selective reporting (reporting bias)	Low risk	Outcomes reported are clinically meaningful and validated
Other bias	Low risk	No other potential sources of bias were detected

Study	Reason for exclusion
Arcioni 2007	The study evaluated regional/local techniques for analgesia
Bell 2001	The study evaluated regional/local techniques for analgesia
Biyik 2009	Open label study
Bone 2002	Not preventive study; pain is already established
Buvanendran 2003	Patients were followed for less than 3 months
Camilleri 2001	The study used a population out of the scope of the review (non surgical/late postsurgical setting)
Capdevila 1999	The study reported outcomes in the acute setting only
Chiu 2008	The study evaluated regional/local techniques for analgesia
Clarke 2010	The study is a secondary analysis of Clarke 2009
Cobellis 2004	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Cohen 2006	The study evaluated regional/local techniques for analgesia
Elkaradawy 2012	The study randomized a regional/local technique for analgesia. Gabapentin was administered for patients reporting moderate/severe pain
Essving 2009	The study evaluated regional/local techniques for analgesia
Essving 2010	The study evaluated regional/local techniques for analgesia
Fassoulaki 2000	The intervention consisted of a topical intervention that is out of the scope of the review
Fassoulaki 2005	This study combines a systemic analgesic (gabapentin) and topical analgesia. It is not possible to 'weight' the effectiveness of each intervention due to methodological issues
Fassoulaki 2006	Patients were followed for less than 3 months
Fassoulaki 2007	Patients were followed for less than 3 months
Fransen 2004	This is the publication of a protocol
Guan 2005	The study reported pain scores in the short term, only; for the long‐term follow‐up, the study reported range of motion
Gupta 2006	The study evaluated regional/local techniques for analgesia
Harding 1991	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Hartrick 2011	Patients were followed for less than 3 months
Honigmann 2007	The study evaluated regional/local techniques for analgesia
Huse 2001	Cross‐over study that explored the impact of morphine sulfate in pain scores and sensory findings. Patients included were already amputated
Jaeger 1992	Cross‐over study
Jirarattanaphochai 2007	The study evaluated regional/local techniques for analgesia
Kadic 2009	The study evaluated regional/local techniques for analgesia
Kamencic 2008	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Kampe 2003	The study evaluated regional/local techniques for analgesia
Katz 2004	The study evaluated regional/local techniques for analgesia
Kim 2011	Patients were followed for less than 3 months
Kollender 2008	The study reported outcomes in the acute setting only
Koninckx 2008	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Lambert 2001	The study evaluated regional/local techniques for analgesia
Lampl 2002	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Lavand'homme 2005	The study has no placebo group
Lynch 2003	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Maier 2003	Not preventive study; pain is already established
Martin 2008	The study evaluated regional/local techniques for analgesia
Morin 2005	The study evaluated regional/local techniques for analgesia
Nehler 2007	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Nesher 2008	The study reported outcomes in the acute setting only
Nesher 2009	The study reported outcomes in the acute setting only
Nikolajsen 1997	The study evaluated regional/local techniques for analgesia
Nikolajsen 2000	Not preventive study; pain is already established
Nissman 2008	Patients were followed up for less than 3 months
Obata 1999	The study evaluated regional/local techniques for analgesia
Parker 2006	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Perniola 2009	The study evaluated regional/local techniques for analgesia
Reuben 2004	Retracted publication
Reuben 2005	Retracted publication
Reuben 2006	Retracted publication
Reuben 2007	Retracted publication
Reuben 2008	Retracted publication
Rhodes 1988	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Ryu 2011	The study evaluated regional/local techniques for analgesia
Salengros 2010	The study is not placebo controlled
Sanders 2009	The study evaluated regional/local techniques for analgesia
Senturk 2002	The study evaluated regional/local techniques for analgesia
Sesti 2007	The study used a population out of the scope of the review (non‐surgical/late postsurgical setting)
Singh 2007	The study evaluated regional/local techniques for analgesia
Solak 2007	Not preventive study; pain is already established
Tramer 1996	Patients were followed up for less than 3 months
Wai 2010	The study evaluated the local administration of morphine, this review is focused in the impact of systemic analgesia
Waikakul 1999	Open label trial
White 2007	Patients were followed up for less than 3 months
Wu 2008	Not preventive study; pain is already established

Methods	Double blind, randomized trial
Participants	One hundred and forty term pregnant women undergoing elective Cesarean delivery
Interventions	Participants were randomized into four groups (N=35 each), placebo (0.9% normal saline), ketamine 0.25, 0.5, or 1 mg kg(‐1) intravenously
Outcomes	Patients were evaluated for persistent postoperative pain at 2 weeks, 1 and 6 months, and 1 year.
Notes

Methods	Multicenter, randomized, controlled trial
Participants	Seventy‐seven patients undergoing inguinal hernia surgery
Interventions	Subcutaneous etanercept 50 mg administered 90 minutes before surgery vs. saline
Outcomes	The primary outcome measure was a 24‐hour numerical rating scale pain score. Secondary outcome measures were postanesthesia care unit pain scores, 24‐hour opioid requirements, time to first analgesic, and pain scores recorded at 1 month, 3 months, 6 months, and 12 months.
Notes

Methods	Randomised controlled trial.
Participants	Eighty patients scheduled for abdominal hysterectomy or myomectomy
Interventions	The pregabalin group received 150 mg of pregabalin 8‐hourly, starting on the afternoon before surgery and continued until the fifth postoperative day. The control group was similarly treated, but received placebo capsules instead.
Outcomes	One and 3 months postoperatively patients were interviewed for the presence of pain and analgesic needs due to surgery.
Notes

Methods	Randomized, double‐blind, controlled study.
Participants	One hundred patients undergoing scheduled lumbar discectomy
Interventions	The minocycline group received 100mg minocycline orally, twice daily, beginning the evening before surgery and continuing for 8days.
Outcomes	The primary outcome was the change in lower limb pain intensity at rest between baseline and 3months. Secondary outcomes were pain intensity on movement, the incidence of persistent pain and chronic neuropathic pain, back pain intensity at rest and on movement, and changes in Neuropathic Pain Symptom Inventory, Brief Pain Inventory, and Roland‐Morris scores at 3months.
Notes

Methods	Randomized, double‐blind study.
Participants	Sixty‐six patients undergoing thoracotomy
Interventions	Thirty‐three patients received an i.v. infusion of S(+)‐ketamine (Group S(+)K) for 60 hours and 33 patients received i.v. placebo (Group PLAC)
Outcomes	Neuropathic Pain Symptom Inventory (NPSI) was evaluated at 1, 3 and 6 months.
Notes

Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients eligible for thoracotomy or pneumectomy
Interventions	Drug: Ketamine
Outcomes	Primary outcome measures: post thoracotomy pain at 2 months assessed using a french equivalent of the McGill Pain score and pain area measurements. Secondary outcome measures: Analogic pain scores at rest and after coughing
Notes	This study has been completed.

Methods	Randomized, double‐blind, placebo‐controlled dose‐finding trial.
Participants	One hundred and thirty‐two women undergoing elective cesarean delivery
Interventions	Participants were randomized into 3 groups to receive 300 or 600 mg oral gabapentin, or placebo, 1 hour before surgery.
Outcomes	The primary outcome was pain on movement at 24 hours. Secondary outcomes included satisfaction with analgesia, supplemental opioid consumption, lactation difficulties, neonatal outcomes, maternal sedation, and other adverse effects. Three months after delivery, patients were contacted for assessment of chronic pain.
Notes

Study name	Effect of Ketalar to Prevent Postoperative Chronic Pain After Mastectomy
Methods	Randomized, double‐blind, clinical trial
Participants	Adult Patients (18‐80), ASA I‐II Exclusion criteria: heart, hepatic or renal failure; allergy to study medications; psychiatric disease; chronic antalgic treatment
Interventions	Ketalar (further information is unclear)
Outcomes	Primary outcome measures: pain intensity 3 months after mastectomy Secondary outcome measures: impact on life quality; area of hyperalgesia
Starting date	December 2004
Contact information	Vincent, PIRIOU, MD
Notes	The recruitment status of this study is unknown because the information has not been verified recently. Verified March 2007 by Hospices Civils de Lyon. Recruitment status was active, not recruiting

Study name	Efficacy And Safety Of Pregabalin For Pain Following Total Knee Replacement
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: Adult patients (ASA I‐III)with osteoarthritis (OA) undergoing elective primary TKA under regional anesthesia (neuroaxial with or without peripheral nerve block) Exclusion criteria: bilateral procedure or revision; subjects with chronic inflammatory conditions or any other chronic pain disorder
Interventions	Pregabalin 150 mg/day, 300 mg/day or placebo
Outcomes	Primary Outcome Measures: Subject Reported Worst 0‐10 Pain Score in Daily Diaries Using the Worst Pain Item of the Modified Brief Pain Inventory ‐ Short Form (m‐BPI‐sf) Long term Secondary Outcome Measures (at 3 & 6 months): Pain Interference Index Score as Measured by the m‐BPI‐sf; Pain Interference With Relations With People; Pain Interference With Enjoyment of Life; pain Interference With General Activity; pain Interference With Mood; Pain Interference With Walking Ability; pain Interference With Normal Work; Pain Interference With Sleep: number of Subjects With Persistent Pain Based on 11‐Point Verbal Rating Scale (VRS); Neuropathic Pain Symptom Inventory (NPSI).
Starting date	May 2007
Contact information	Pfizer CT.gov Call Center
Notes	This study has been completed (First Received on March 1, 2007. Last Updated on February 15, 2011)

PERMALINK

Pharmacotherapy for the prevention of chronic pain after surgery in adults

Luis Enrique Chaparro

Shane A Smith

R Andrew Moore

Philip J Wiffen

Ian Gilron

Abstract

Background

Objectives

Search methods

Selection criteria

Data collection and analysis

Main results

Authors' conclusions

Plain language summary

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

Subgroup analysis and investigation of heterogeneity

Results

Description of studies

Results of the search

1.

Included studies

Study selection

Study design

Surgical procedures

Characteristics of participants: preoperative pain and analgesic use

Drug interventions

Outcome measures

Primary outcome

Secondary outcomes

Excluded studies

Risk of bias in included studies

2.

3.

Allocation

Blinding

Incomplete outcome data

Selective reporting

Other potential sources of bias

Effects of interventions

Ketamine

1. Ketamine trials.

Ketamine three month postoperative pain outcome

Ketamine four month postoperative pain outcome

Ketamine six month postoperative pain outcome

4.

Ketamine 12 month postoperative pain outcome

Gabapentin

2. Gabapentin trials.

Gabapentin three month postoperative pain outcome

5.

Gabapentin six month postoperative pain outcome

Pregabalin

3. Pregabalin trials.

Pregabalin three month postoperative pain outcome

6.

Pregabalin six month postoperative pain outcome

Pregabalin 12 month postoperative pain outcome

Study name	Study Of The Efficacy And Safety Of Pregabalin Compared To Placebo For Treatment Of Post‐Surgical Pain From Hysterectomy
Methods	Randomized, double‐blind, clinical trial
Participants	Patients (ASA I‐II)undergoing elective total abdominal hysterectomy using a transverse incision with or without bilateral salpingo‐oophorectomy. The subject is expected to remain at the hospital (or intermediate care facility) for a minimum of 2 days following surgery; Exclusion Criteria: vaginal hysterectomy; use of nerve block, spinal anesthesia or epidural anesthesia for post‐surgical pain control; Subjects who have been using any opioid medications 2 weeks or more continuously within 3 months prior to the screening visit; NSAIDs users.
Interventions	Pregabalin (Lyrica) 150 mg/day, 300 mg/day double‐blind or matched placebos
Outcomes	Primary Outcome Measures: Worst Pain Using the Modified Brief Pain Inventory at day POP 2. Secondary Long‐term Outcome Measures: Incidence of Chronic Post‐operative Pain [ Time Frame: 3 and 6 Months PS ]Chronic post‐operative pain as a result of abdominal hysterectomy as reported by participants on PS questionnaire of pain within last 24 hours in area affected by surgery.
Starting date	June 2007
Contact information	Pfizer
Notes	This study has been completed. First received on May 1, 2007. Last updated on June 7, 2011

Study name	Surgical Pain After Inguinal Hernia Repair (SPAIHR)
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients undergoing primary open unilateral inguinal herniorrhaphy (hernia repair), using mesh Lichtenstein surgery procedure and under general anesthesia; patients must be able to use and tolerate non‐steroidal anti‐inflammatory drugs (such as naproxen), tramadol, oxycodone, and acetaminophen/paracetamol for pain control after surgery. Exclusion criteria: emergency surgery; use of nerve block or spinal/epidural/paravertebral anesthesia; subjects that are not allowed to receive the anesthesia agents indicated per protocol and general anesthesia.
Interventions	Pregabalin 25 mg BID, or 75 mg BID or 150 mg BID or placebo
Outcomes	Primary outcome measures: Modified Brief Pain Inventory‐Short Form (mBPI‐sf): Worst Pain 24 Hours Post Surgery. A self‐administered 11‐point Likert rating scale to rate pain in the past 24 hours. Secondary long‐term outcome measures (3 & 6 months after surgery): Number of participants who reported surgery‐related pain at assessment; Pain Severity Index Score and Pain Interference Index Score on the Modified Brief Pain Inventory‐Short Form (mBPI‐sf); Total Score and Subscale Scores Using the Neuropathic Pain Symptom Inventory (NPSI).
Starting date	January 2008
Contact information	Pfizer
Notes	This study has been completed. First Received on October 26, 2007. Last Updated on June 8, 2010

Study name	Role of Pregabalin in Treatment of Post‐Op Pain in Fracture Patients (LYRICA)
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients (19‐70) with fractures requiring operative treatment during a single operative episode Exclusion criteria: history of opioid abuse/misuse; contraindications to pregabalin or narcotic analgesics; closed head injury; psychiatric illness requiring medical treatment; surgery for other injuries (splenectomy, etc); history of seizures requiring current anticonvulsant therapy.
Interventions	Placebo group: Two hours before surgery, patients will receive 75mg of pregabalin. Patients will be placed on a patient‐controlled anesthesia pump (PCA) for 24 hours. On post‐operative day one, the patients will be switched to oral oxycodone as needed with supplementation with IV Demerol for breakthrough pain. In addition, patients will receive a placebo PO BID or pregabalin 75 mg or pregabalin 150 mg beginning on the day of surgery until discharge.Upon discharge, the patient will be given study medication (placebo PO BID or pregabalins). Rescue medications will be allowed during the study (including post‐operative and outpatient periods). Outpatient rescue medications will consist of hydrocodone/APAP 7.5mg PO Q6H PRN.
Outcomes	Primary outcome measures: Amount of pain medication in morphine equivalent units used during the hospitalization Secondary outcome measures: pain scores at 3 months
Starting date	May 2007
Contact information	David A. Volgas, MD, Associate Professor of Surgery, The University of Alabama at Birmingham
Notes	This study is ongoing, but not recruiting participants. First received on December 21, 2007. Last updated on September 15, 2010

Study name	The Effect of Perioperative Ketamine on Acute and Chronic Pain After Major Back Surgery (KetaDol)
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: Adults, ASA I‐III, undergoing back surgery (laminectomy, lumbar arthrodesis (Posterior Lumbar Interbody Fusion ‐ PLIF, Transforaminal Lumbar Interboby Fusion ‐ TLIF, Postero‐lateral Fusion, semi‐rigid fixation. Exclusion Criteria: heart disease, glaucoma, allergy to study drugs, dementia, failed back surgery syndrome, postraumatic paraplegia.
Interventions	50 ml syringes provided by the HUG pharmacy will contain 1% ketamine or 0.9% NaCl. After induction and before start of surgery, patients will receive an intravenous bolus of 0.025 ml/kg of the study solution (corresponding to 0.25 mg/kg ketamine). Maintenance will be with a syringe driver at a rate of 0.025 ml/kg/h (corresponding to 0.25 mg/kg/h ketamine) until one hour before the end of surgery, and will then be decreased to a rate of 0.01 ml/kg/h (corresponding to 0.1 mg/kg/h ketamine) throughout the stay in the recovery room (usually 2 to 3 hours). The infusion will be stopped when the patient leaves the recovery room.
Outcomes	Primary outcome measures: 6 and 12 months effect of perioperative intravenous low‐dose ketamine on chronic neuropathic pain in patients undergoing major back surgery. Secondary long‐term outcome measures: short‐term (during hospitalisation) effect of perioperative intravenous low‐dose ketamine in patients undergoing major back surgery: tolerability and safety, opioid‐sparing effect, pain intensity, morphine‐related adverse effects. To study psychosocial factors that may be involved in the perception of acute and chronic postoperative pain in patients with or without chronic back pain undergoing back surgery.
Starting date	October 2007
Contact information	Study Chair: Martin Tramèr, Prof, MD, PhD. Christoph Czarnetzki, Responsable Investigator, University Hospital, Geneva
Notes	This study is ongoing, but not recruiting participants. First received on February 6, 2008. Last updated on April 3, 2012

Study name	Pregabalin in Treating Pain in Women Undergoing Mastectomy or Lumpectomy
Methods	Randomized, double‐blind, clinical trial
Participants	Patients (ASA I‐III) undergoing unilateral modified radical mastectomy or lumpectomy with axillary node dissection. Exclusion criteria: allergy to study drugs, drug/alcohol abuse; kidney failure; concurrent use of other analgesics.
Interventions	Oral placebo or pregabalin 1‐2 hours prior to surgery, at 12 hours after surgery, and then twice daily for 14 days
Outcomes	Primary outcome measures: pain scores at 3 months; PCA morphine consumption; side effects profile; and modified Brief Pain Inventory‐short form
Starting date	December 2006
Contact information	Babatunde Ogunnaike, Simmons Comprehensive Cancer Center at University of Texas Southwestern Medical Center ‐ Dallas
Notes	The recruitment status of this study is unknown because the information has not been verified recently. Verified May 2009 by National Cancer Institute (NCI). Recruitment status was: Recruiting

Study name	Pregabalin and Post‐thoracotomy Pain
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients (18‐75; ASA I‐III)undergoing elective thoracotomy (ET) or video assisted thoracotomy (VAT). Exclusion criteria: intolerance to study drugs; Contraindication to thoracic epidural placement; Renal insufficiency (serum creatinine > 1.5 x upper limit of normal); Body Mass Index > 40; surgery extending to the chest wall; alcohol abuse; chronic pain/analgesic use; History of congestive heart failure; Major psychiatric disorder; Insufficient safety data in a specific patient population; Pregnant or breastfeeding.
Interventions	Pregabalin 150mg or 300 mg or placebo administered 1 hour prior to surgery and 12 hours after surgery, then continued BID until day 7 post‐op
Outcomes	Primary outcome measures: The primary outcome measure for the final study will be the incidence of CPTPS at 2 months. [Time Frame: 2, 4, and 6 months]
Starting date	April 2008
Contact information	Dr Jorge Enrique Zamora, Department of Anesthesiology, Queen's University
Notes	The recruitment status of this study is unknown because the information has not been verified recently. Verified September 2009 by Queen's University. Recruitment status was active, not recruiting

Study name	Perioperative Pregabalin Use, Rehabilitation, Pain Outcomes and Anxiety Following Hip Surgery (RCT)
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: Patients (18‐75), ASA I‐III undergoing total hip arthroplasty. Exclusion criteria: allergy to study drugs; drug/alcohol/opioid abuse; rheumatoid arthritis; psychiatric disorders; renal failure; obesity.
Interventions	Pre‐operative pregabalin 150 mg or placebo plus post‐operative dose 75 mg bid or placebo
Outcomes	Movement‐evoked pain scores at 6 weeks and 3 months post surgery
Starting date	May 2009
Contact information	Colin McCartney, MD. colin.mccartney@sunnybrook.ca
Notes	This study is currently recruiting participants. Verified April 2011 by Sunnybrook Health Sciences Centre

Study name	The Effects of Peri‐Operative Pregabalin on Post‐Operative Pain Following Breast Cancer Surgery With Axillary Node Dissection: A Pilot Study
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients (18‐60), ASA I‐III undergoing breast surgery with axillary dissection for the treatment of breast cancer. Exclusion criteria: persons with a history of allergy to gabapentin or pregabalin, morphine, NSAIDs, acetaminophen or oxycodone; pregnancy; Body Mass Index >40; liver or renal failure; chronic opioid users (30 mg per day of morphine equivalent); gabapentin or pregabalin users within 3 months of surgery; persons with a history of drug abuse.
Interventions	Participants will receive pregabalin (75 mg BID) or placebo (sugar pills) for BID for 14 days starting 1 hour before surgery
Outcomes	Primary Outcome Measures: numeric rating pain Score at rest and with movement 24 hours after surgery.  Secondary Outcome Measures: Incidence of chronic post‐mastectomy pain at 3 months defined as persistent pain or discomfort not present prior to surgery and not present as a result of new or recurrent tumour growth.
Starting date	May 2008
Contact information	Peter MacDougall, MD. pcmacdou@gmail.com
Notes	This study is currently recruiting participants. Verified January 2009 by Capital District Health Authority, Canada

Study name	Study To Investigate The Effectiveness Of Pregabalin For Management Of Patients Undergoing Total Hip Replacement
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients undergoing total primary or secondary hip replacement surgery performed under spinal anesthesia. Exclusion criteria: revision surgery, hip replacement secondary to trauma; history of uncontrolled chronic disease or a concurrent clinically significant illness or medical condition, which in the Investigator's opinion, would contraindicate study participation or confound interpretation of the results.
Interventions	Placebo or pregabalin 75 mg BID for 14 days
Outcomes	Primary outcome measures: O‐10 Pain score on Movement obtained over postoperative Days 1‐5. Secondary long‐term outcome measures: Incidence of neuropathic pain as detected using the ID Pain questionnaire 3 & 6 months following the surgery.
Starting date	November 2009
Contact information	Pfizer
Notes	This study is currently recruiting participants. Verified April 2012 by Pfizer

Study name	Efficacy of Perioperative Pregabalin in Reducing the Incidence of Chronic Neuropathic Pain and Postthoracotomy Syndrome.
Methods	Randomized, double‐blind, clinical trial
Participants	Inclusion criteria: patients aged 18 to 80 years, ASA I‐III, undergoing elective thoracotomy. Exclusion criteria: contraindication to pregabalin or epidural technique; current use of opioids, NMDA receptor blockers, membrane stabilizing agents (lidocaine mesylates, flecainide) or topical coanalgesics (capsaicin cream, lidocaine patch); previous use of pregabalin or gabapentin; preexisting pain at the surgical site or any other chronic pain syndrome; creatinine clearance of less than 60 mL/min; previous thoracotomy; recent history of alcohol and/or drug abuse; known allergy to local anesthetics or hydromorphone.
Interventions	Pregabalin 150 mg bid or matching placebos during 5 days, starting the days before surgery
Outcomes	Primary outcome measures: development of neuropathic pain and intensity of pain assessed using the Leeds Assessment of Neuropathic Symptoms and Signs (LANSS) scale and Brief Pain Inventory questionnaire (BPI) at 3 months
Starting date	June 2010
Contact information	Centre hospitalier de l'Université de Montréal (CHUM)
Notes	This study has been completed. First received on August 26, 2009. Last updated on April 11, 2012