Characteristics of excluded studies [ordered by study ID]

Study	Reason for exclusion
Abdel-Salam 1975	Study comparing different epidural LA mixtures for analgesic effect, 2 days after surgery. No long-term outcomes recorded
Aveline 2011	Participants undergoing day-case open inguinal hernia repair with mesh given TAP block or ilioinguinal/iliohypogastric nerve block. No control group. VAS scores at 3 and 6 months
Bach 1988	Pseudo-clinical RCT (sequence generation by means of patients’ year of birth) investigating epidural analgesia before limb amputation for chronic phantom pain with a follow-up of 12 months
Bamigboye 2013	Outcome was attenuation of (pre-existing) chronic pelvic pain. The primary outcome of interest for this review, (new onset wound pain persisting for > 3 months after surgery) was not measured
Baral 2010	Study assessing effectiveness of preoperative IV lidocaine infusion on post-op pain, however, no chronic pain outcomes assessed
Batoz 2009	Follow-up only 2 months in this RCT of scalp infiltration for craniotomy
Blumenthal 2011	Comparing regional technique against combination of regional techniques
Borgeat 2001	Outcome: regional anaesthesia complications associated with interscalene block
Borghi 2010	Non-randomized prospective trial of perineural catheter for phantom limb pain
Brull 1992	Non-randomized observational study of continuous infusion through an iliac crest catheter for postoperative analgesia after ICBG harvesting
Cerfolio 2003	Preincision epidural anaesthesia vs none for thoracotomy, but no control (as both groups had post-op epidural anaesthesia)
Chelly 2011	All participants received local wound infiltration and there was no control group without application of local or regional anaesthesia
Corsini 2013	Article in French. Single-dose intraincisional infiltration of levobupivacaine or placebo into wound after scheduled C-section. Longest pain outcome at 2 months
da Costa 2011	Excluded for pseudo-randomization, this prospective trial investigated different anaesthetic techniques for the prevention of regional pain syndrome after carpal tunnel release
De Kock 2001	Comparing IV ketamine to epidural ketamine to control as adjuvant therapy; all patients receiving LAs via epidural catheter
Duale 2009	Comparison of ketamine or placebo in people undergoing thoracotomy. All participants received local ropivacaine administration at the edges of the thoracotomy and chest drainage orifices and in the inter pleural space postoperatively (thus no control group)
Eisenach 2010	RCT comparing intrathecal bupivacaine with ketoralac vs saline for prevention of postoperative pain. All participants received intrathecal bupivacaine thus no control group
El-Morsy 2012	Randomized, blinded study comparing outcome of paravertebral block vs thoracic epidural block for post-thoracotomy incision pain in paediatric patients. The primary objective was evaluation of immediate postoperative analgesia. Secondary objectives included hormonal responses, side effects, failure rate, and pulmonary function. No long-term outcomes were measured
Elman 1989	Comparing different doses of bupivacaine intrapleurally, no long-term pain outcomes were measured
Farag 2013	Patient on chronic opioids preoperatively
Gottschalk 1998	Follow-up only 9.5 weeks, in a double-blind clinical RCT of 100 people undergoing elective radical retropubic prostatectomy for the treatment of prostate cancer. Epidural bupivacaine, epidural fentanyl, or no epidural drug was administered prior to induction of anaesthesia and throughout the entire operation resulting in more pain-free participants at 9.5 weeks
Haythornthwaite 1998	Study on prostatectomy with 3 groups: epidural anaesthesia only, combined epidural and general anaesthesia and general anaesthesia only. Total of 6-month follow-up. However, excluded because epidural PCA was provided with bupivacaine and fentanyl for all participants in the postoperative period, thus no control group
Hirakawa 1996	Not randomized
Hivelin 2011	Not a randomized trial but only a prospective blinded study of TAP block in breast reconstruction
Howell 2001	Study designed to investigate differences in backache as complication/adverse effect of labour epidural
Ilfeld 2004	Not a clinical RCT, but only case reports on 3 paediatric patients with continuous regional anaesthesia catheters, 2 patients with pain outcomes at 3 months
Ilfeld 2015	Comparison of continuous vs single shot (regional vs regional) anesthesia
Jahangiri 1994	Prospective, but not randomized study of preoperative epidural anaesthesia for phantom pain after limb amputation
Jirarattanaphochai 2007	Excluded because chronic pain present at baseline and is reason for surgery
Joseph 2012	RCT in which all participants received epidural catheter with participant-controlled ropivacaine administration, comparing IV ketamine vs no ketamine in people undergoing thoracotomy. Follow-up of 3 months post-op
Kairaluoma 2010	Comparing paravertebral block against local infiltration for hernia repair under SA
Kindberg 2009	RCT comparing use of ear acupuncture vs LA in primiparous women with a vaginal delivery at term undergoing surgical repair of lacerations to the labia or the vagina, perineal lacerations of first or second degree or mediolateral episiotomies. Excluded because of traumatic reason for ‘surgical’ intervention (suturing), not an elective procedure
Kumar 1989	Non-randomized pilot study of 20 patients to examine post-cholecystectomy pain relief of paravertebral block with bupivacaine, with or without adrenaline added. Alternating participants received adrenaline or did not
Kumar 2009	Men undergoing totally extra-peritoneal repair of groin hernia were randomized to pre-peritoneal bupivacaine vs saline after mesh placement. All prospective trocar sites were infiltrated by bupivacaine in all cases, thus no control group without regional analgesia
Lambert 2001	Comparing regional against regional technique: clinical RCT comparing preoperative epidural vs postoperative perineural catheter for risk reduction of phantom pain after limb amputation
Lebreux 2007	Not comparing regional vs nonregional anaesthesia. 20 healthy parturients undergoing elective caesarean section under SA were randomized to receive spinal clonidine. Outcome was pain up to 6 months and hyperalgesia
Lee 2012	RCT of patients undergoing video-assisted thoracic surgery, with all participants receiving epidural ropivacaine and fentanyl, with or without magnesium sulphate
Loughnan 2002	Controlled clinical trial designed to detect difference in backache as complication/adverse effect of labour epidural
Mendola 2012	RCT evaluating use of S(+)-ketamine for prevention of post thoracotomy pain syndrome at 6 months. Patients undergoing thoracotomy under general anaesthesia, with thoracic epidural catheter placed +/− IV infusion of ketamine vs IV placebo with 6 months post-op follow-up. All participants received epidural catheter with levobupivacaine, thus no control group
Milligan 2002	Comparison of LA vs LA
Muthukumar 2012	Prospective-double blind RCT investigating haemodynamic effects, quality of surgical field and postoperative analgesia following surgical field infiltration with different concentrations of adrenaline with and without lignocaine in children undergoing cleft lip repair. Only immediate postop pain was recorded, no long-term outcomes measured
Nabhan 2011	Patients undergoing endoscopic carpal tunnel release under LA (prilocaine) vs IV regional anaesthesia (prilocaine)
Nikolajsen 1997	Study excluded for pseudo-randomization as discussed in (Appendix 9). Double-blinded (patients and outcome assessors), pseudo-randomized (sequence generationwas by “the toss of a coin”) controlled clinical trial on preoperative epidural analgesia for limb amputation with a follow-up of 12 months including 60 adults in a university setting in Aarhus, Denmark
Obata 1999	Comparing preincisional vs postincisional epidural anaesthesia for thoracotomy
Ochroch 2006	Comparing preincisional vs postincisional epidural anaesthesia for thoracotomy
Ouaki 2009	Prospective study examining continuous infusion of ropivacaine at iliac crest donor site in paediatric patients undergoing ICBG. However, non-randomized with only 1 study group, all with same treatment (no control group)
Panos 1990	RCT comparing IV vs epidural fentanyl, not LA vs control
Perniola 2009	RCT of intra-abdominal LA for abdominal hysterectomy. Follow-up 3 months. Excluded because all 3 groups used LA infusions
Pompeo 2007	Comparison of awake video-assisted thoracoscopic bullectomy with pleural abrasion using thoracic epidural anaesthesia vs general anaesthesia (control) in treatment of spontaneous pneumothorax. No long-term pain outcomes measured; follow-up at 12 months was to elicit recurrences of pneumothorax
Rosen 2009	Patients undergoing laparoscopic ventral hernia repair randomized to receive elastomeric pain pump with continuous LA vs saline. Each trocar site injected with LA in either group thus both groups received LAs. Total follow-up 3 months
Royse 2007	Measured outcome was a depression score, no chronic postsurgical pain measured
Ryu 2011	Comparison of pre-emptive thoracic epidural analgesia with or without ketamine in people undergoing operations using classic posterolateral thoracotomy incisions. Thus, no control group. Total follow-up of 3 months post-op
Saber 2009	Follow-up only 2 months
Salengros 2010	RCT investigating pre- vs postoperative epidural anaesthesia after thoracotomy
Schaan 2004	Pain outcomes measured < 3 months
Schley 2007	Study on effect of adjuvants for LAs to prevent chronic postsurgical pain. All 19 participants received a continuous brachial plexus block for 1 week after the amputation of an upper extremity. In addition they were treated with the NMDA antagonist memantine or placebo for 4 weeks
Sen 2009	RCT of 60 men aged 20-40 years undergoing inguinal herniorrhaphy, comparing preoperative oral gabapentin to placebo and the effects on acute and long-term pain. All participants received intrathecal bupivacaine. Follow-up total of 6 moths post-op
Shikano 1994	RCT looking at the effect of wound infiltration with bupivacaine before insertion of trocars on post-op pain and respiratory impairment in people undergoing laparoscopic cholecystectomy. No long-term pain outcomes measured
Sim 2012	Randomized trial investigating pre- vs postincisional pre-emptive thoracic epidural analgesia for thoracotomy with outcomes at 6 months, but with no control group without regional anaesthesia
Suvikapakornkul 2009	Pain outcomes measured only until 24 h post-op; 3-month follow-up was only for recurrence and complications
Suzuki 2006	Studying the adjuvant effect of IVketamine vs placebo in 49 thoracotomy patients, all participants receiving ropivacaine with morphine via epidural analgesia for 2 days
Verma 2006	Patients with chronic cholecystitis divided into 4 groups, to receive either saline or different combinations of bupivacaine at gallbladder bed and trocar sites. No long-term pain outcome measures
Vigneau 2011	Pain outcomes measured only up to 2-month follow-up in this RCT on would infiltration after breast surgery
Wang 1992	Article in Mandarin. No comparison group without regional anaesthesia
Weihrauch 2005	Comparing block vs block with no pain outcome measured
Wilson 2008	RCT on patients undergoing lower limb amputation received combined intrathecal/epidural anaesthetic for surgery followed by epidural infusion with bupivacaine with ketamine vs bupivacaine with placebo (saline). No control group as both received LA
Yang 2012	We acknowledge the study author’s response to our inquiry; pain data only measured until 2 months postop

ICBG: iliac crest bone graft; IV: intravenous; NMDA: N-methyl-D-aspartate receptor; PCA: patient controlled analgesia; RCT: randomized controlled trial; SA: spinal anaesthetic; TAP: transabdominal plane block; VAS: visual analogue scale