Facilitatory effects of perineural dexmedetomidine on neuraxial and peripheral nerve block: a systematic review and meta-analysis

British Journal of Anaesthesia 110 (6): 915 25 (2013) Advance Access publication 15 April 2013. doi:10.1093/bja/aet066 Facilitatory effects of perineural dexmedetomidine on neuraxial and peripheral nerve : a systematic review and meta-analysis F. W. Abdallah 1 and R. Brull 2 * 1 Department of Anesthesia and Pain Management, St Michael s Hospital, and Women s College Hospital, University of Toronto, Toronto, Canada 2 Department of Anesthesia and Pain Management, Toronto Western Hospital, University Health Network, and Women s College Hospital, University of Toronto, Toronto, Canada * Corresponding author: Department of Anesthesia, Toronto Western Hospital, 399 Bathurst Street, Toronto, ON, Canada M5T 2S8. E-mail: richard.brull@uhn.ca Editor s key points Dexmedetomidine has been used to prolong the of local anaesthetics (LAs). In this meta-analysis, nine randomized controlled trials on perineural dexmedetomidine in neuraxial and peripheral nerve s were selected. Dexmedetomidine prolonged. More studies are required to establish the safety of using dexmedetomidine as a perineural adjunct to LAs. Summary. Nerve s improve postoperative analgesia, but their benefits may be shortlived. This quantitative review examines whether perineural dexmedetomidine as a local anaesthetic (LA) adjuvant for neuraxial and peripheral nerve s can prolong the of analgesia compared with LA alone. All randomized controlled trials (RCTs) comparing the effect of dexmedetomidine as an LA adjuvant to LA alone on neuraxial and peripheral nerve s were reviewed., motor, onset times, analgesic consumption, time to first analgesic request, and sideeffects were analysed. Results were combined using random-effects modelling. A total of 516 patients were analysed from nine RCTs. Five trials investigated dexmedetomidine as part of spinal anaesthesia and four as part of a brachial plexus (BP). was prolonged by 150 min [95% confidence interval (CI): 96, 205, P,0.00001] with intrathecal dexmedetomidine. Perineural dexmedetomidine used in BP may prolong the mean of sensory by 284 min (95% CI: 1, 566, P¼0.05), but this difference did not reach statistical significance. Motor and time to first analgesic request were prolonged for both intrathecal and BP. Dexmedetomidine produced reversible bradycardia in 7% of BP patients, but no effect on the incidence of hypotension. No patients experienced respiratory depression. Dexmedetomidine is a potential LA adjuvant that can exhibit a facilitatory effect when administered intrathecally as part of spinal anaesthesia or peripherally as part of a BP. However, there are presently insufficient safety data to support perineural dexmedetomidine use in the clinical setting. Keywords: acute pain; regional techniques; anaesthetic techniques; regional; brachial plexus; analgesic techniques; subarachnoid; analgesics; postoperative; sympathetic nervous system; dexmedetomidine Regional anaesthesia techniques provide important advantages compared with general anaesthesia and systemic analgesia, including excellent pain control, reduced sideeffects, and shortened stay in the post-anaesthesia care unit. 1 3 However, these early advantages can be short-lived 3 and limited by the relatively brief of action 45 of currently available local anaesthetics (LAs), 6 potentially resulting in resolution before the period of worst postoperative pain. 7 8 Increasing the volume (dose) of LAs may prolong the of analgesia, 9 but may also increase the risk of LA systemic toxicity. 10 Although continuous catheter-based nerve s can extend postoperative analgesia, 11 12 their placement requires additional time, cost, and skill. 13 While a novel sustained-release encapsulated (liposomal) preparation of is presently undergoing investigation in phase III trials, 14 a variety of perineural adjuvants, 15 including buprenorphine, 16 clonidine, 17 dexamethasone, 18 magnesium, 19 and midazolam, 20 21 have been used to prolong the of analgesia of nerve s with varying degrees of success. Dexmedetomidine, an a 2 adrenoreceptor agonist, 22 was first proposed as an adjuvant capable of prolonging of sensory and motor produced by nerve s by Memiş and colleagues. 23 However, the series of clinical trials that followed produced contradictory results. 24 27 Some trials have shown that perineural dexmedetomidine reduces the onset time and prolongs the of sensory and motor. 23 25 26 Conversely, other trials have demonstrated either a delay in & The Author [2013]. Published by Oxford University Press on behalf of the British Journal of Anaesthesia. All rights reserved. For Permissions, please email: journals.permissions@oup.com

BJA Abdallah and Brull sensory and motor onset time 27 or no effect on sensory and motor 24 with the use of perineural dexmedetomidine. The primary objective of this quantitative review is to determine whether the administration of perineural dexmedetomidine as an LA adjuvant for neuraxial Table 1 Trial characteristics. Dex, dexmedetomidine; n, number of trials Author/year Quantity (n) Percentage Source database Medline 4 44.4 Google scholar 2 22.2 Hand search 2 22.2 Grey literature 1 11.1 Trial source journal Listed in Index Medicus 4 44.4 Not listed in Index Medicus 5 55.5 Trial source country Egypt 2 22.2 India 3 33.3 Jordan 1 11.1 Lebanon 1 11.1 Turkey 2 22.2 Jadad score 5 (excellent quality) 6 66.6 4 3 33.3 3 0 0 2 0 0 1 (poor quality) 0 0 Number of subjects,50 1 11.1 50 100 8 88.8 Age Adult ( 18) 9 100 Paediatric (,18) 0 0 Gender Female 0 0 Male 1 11.1 Both 8 88.8 Outcomes assessed Analgesia 9 100 Block characteristics 9 100 Dex side-effects 9 100 Location of surgery Abdominal 2 22.2 Extremity, upper 4 44.4 Extremity, lower 2 22.2 Combination 1 11.1 Disposition Inpatient 3 33.3 Outpatient 2 22.2 Both 2 22.2 Unspecified 2 22.2 and peripheral nerve s can prolong the of analgesia compared with LA alone. Methods The PRISMA 28 recommendations were followed in the preparation of this manuscript. Eligibility criteria We sought to identify all randomized controlled trials (RCTs) that examined the effects of adding perineural dexmedetomidine to LA (dexmedetomidine group) compared with LA alone (control group) on neuraxial or peripheral nerve characteristics, postoperative analgesia, and dexmedetomidine-related side-effects in patients undergoing regional anaesthesia. Blocks performed for either anaesthesia or postoperative analgesia were included. RCTs were excluded if dexmedetomidine was used as a stand-alone perineural agent without LA, 29 or administered via a nonperineural route, 23 24 38 if continuous nerve s were performed, 39 40 and if s were performed in paediatric patients where characteristics could not be assessed. 41 44 Only trials that explicitly mentioned obtaining approval from the local ethics committee or institutional review board were considered. The use of dexmedetomidine as part of i.v. regional anaesthesia (Bier ) was not considered for the purposes of this review. Literature search We retrieved RCTs from the US National Library of Medicine database, MEDLINE; the Excerpta Medica database, Embase; Cochrane Database of Systematic Reviews; and Cochrane Central Register of Controlled Trials databases (January 1985 August 2012). The search terms dexmedetomidine and medetomidine were used in combination with the search terms perineural, adjuvant, adjunct, and admixture. Searches were combined using the Boolean operator AND with medical subject headings analgesia/pain relief/pain control/pain prevention/and pain management and the medical subject headings regional anaesthesia/nerve / /neuraxial /central /peripheral. The search was limited to trials published in the English language. We also reviewed the reference lists of selected trials for additional RCTs. Trials that are unpublished or in progress were not included. Data collection and presentation The two authors (F.W.A. and R.B.) independently evaluated the methodological quality of the included trials using the Jadad score; 45 and a final score was designated by consensus for each RCT. We selected sensory as the primary endpoint, while motor, sensory and motor onset time, analgesic consumption, time to first analgesic request, pain scores, 46 and dexmedetomidinerelated adverse effects (hypotension, bradycardia, respiratory depression, and postoperative sedation) 47 48 were defined as secondary endpoints. The authors each used a standardized 916

Perineural dexmedetomidine BJA data sheet to extract and record trial results, which were compared and any differences were resolved by reexamination of the source trials. Meta-analysis Data entry was performed by one author (F.W.A.) and rechecked by another (R.B.). Meta-analytic techniques (Revman 5.1, Cochrane Library, Oxford, UK) were used to pool data whenever possible. Data from trials with more than two intervention groups receiving different doses of dexmedetomidine via the same route were combined into a single group as recommended by the Cochrane Handbook. 49 Dichotomous and continuous outcomes were analysed using random-effects modelling. The odds ratio (OR) and 95% confidence intervals (CIs) are reported for dichotomous outcomes, while the standardized mean difference and 95% CI are reported for continuous outcomes. Differences were considered statistically significant when the 95% CI did not include 1 for OR and 0 for the standardized mean difference. Heterogeneity of the pooled results was assessed using the I 2 statistic. 50 Results We retrieved 37 articles, of which nine met our inclusion criteria. 25 27 51 56 Tables 1 and 2 present the trial characteristics and outcomes assessed for each trial, respectively. The methodological quality 45 of all nine trials was good; six trials 26 51 52 54 56 achieved a Jadad score of 5 out of 5, while the remaining three 25 27 53 achieved a score of 4 out of 5. The countries of origin for all eight trials were Middle Eastern; all protocols were approved by the local ethics committee or institutional review board of their respective institution. Figure 1 summarizes the search results, including the RCTs retrieved, excluded, and presently reviewed. Twenty-eight trials were excluded because of the interventions examined (n¼14), populations studied (n¼7), active comparators (n¼5), study design (n¼1), and language of publication (n¼1) (Appendix). The trials reviewed included a total of 516 patients for analysis; 274 patients in the dexmedetomidine group and 242 in the control group. Five trials examined the effect of neuraxial dexmedetomidine administered intrathecally as part of spinal anaesthesia, 26 51 52 54 57 and four assessed peripheral dexmedetomidine 25 27 55 56 administered as part of a brachial plexus (BP). All nine trials reviewed herein used a long-acting LA, namely ropivacaine, 53, 26 27 51 52 54 56 and levo (Table 1). We did not identify any RCTs 25 55 that investigated the use of peripheral dexmedetomidine for truncal or lower extremity s, or neuraxial dexmedetomidine in epidural or caudal s. Two trials included two dexmedetomidine groups 51 52 ( low dose and high dose ). Data regarding sensory were available from all trials reviewed and are presented in Figure 2. Intrathecal administration of dexmedetomidine as part of spinal anaesthesia prolonged the mean sensory by 150 min (95% CI: 95.94, 204.75, P,0.00001), which is a relative increase of 72% compared with LAs alone. The heterogeneity among the pooled studies in this subgroup was significant (I 2 ¼0.95; P,0.00001). The data signal that peripheral dexmedetomidine used in a BP at the axillary, 25 55 supraclavicular, 27 and infraclavicular 56 levels may prolong the mean of sensory by 284 min (95% CI: 1.39, 565.68, P¼0.05), a relative increase of 76% compared with LAs alone, but this difference did not reach statistical significance. Heterogeneity among trials in this subgroup was also high (I 2 ¼1; P,0.00001). Block characteristics When used intrathecally, dexmedetomidine hastened sensory onset by 2 min (95% CI: 22.96, 20.11, P¼0.04) or 19%, prolonged motor by 132 min (95% CI: 87.69, 176.74, P¼0.00001) or 88%, and delayed the time to first analgesic request by 293 min (95% CI: 174.32, 411.41, P¼0.00001) or 127% compared with LA alone. The motor onset time was similar between the dexmedetomidine and control groups (Table 3). Administering perineural dexmedetomidine as part of a BP resulted in a prolongation of motor by 268 min (95% CI: 15.47, 520.06, P¼0.04) or 87%, and an increase in time to first analgesic request by 345 min (95% CI: 102.68, 587.23, P¼0.005) or 70% compared with LAs alone. The sensory and motor onset times were similar between the dexmedetomidine and control groups (Table 3). Dexmedetomidine-related adverse effects Because of the diversity in the definitions of dexmedetomidine-related adverse effects in the reviewed trials, the results of these outcomes are reported as standardized units. The incidence of hypotension was similar between the dexmedetomidine and the control groups (Table 3). The incidence of bradycardia was higher in patients who received dexmedetomidine as part of a BP (7% vs 0%, P¼0.03), but there was no difference with intrathecal administration. The observed bradycardia was transient, successfully reversed by i.v. atropine administration, and did not recur later during the postoperative period. Respiratory depression was explicitly assessed in four trials. 25 27 52 55 None of the patients in these trials experienced respiratory depression. The incidence of postoperative sedation was evaluated in four trials 26 51 53 that examined the intrathecal administration of dexmedetomidine; a six-point measurement scale was used in two of the trials 51 52 and a four-point scale was used in the other two trials. 26 53 The heterogeneity of scales used and the inconsistency in reporting rates of occurrence precluded any quantitative analyses. Qualitatively, one trial 52 reported higher sedation levels in a group of patients receiving high-dose intrathecal dexmedetomidine (15 mg) without specifying the actual rate of occurrence, while 917

918 Table 2 Trial outcomes. ACL, anterior cruciate ligament repair; AXB, axillary nerve ; Dex, dexmedetomidine; ICB, infraclavicular ; Intraop, intraoperative; N/D, not defined; NS, normal saline; Postop, postoperative; SCB, supraclavicular ; TURBT, transurethral resection of bladder tumour; TURP, transurethral resection of prostate tumour; TVT, tension-free vaginal tape Study Jadad score Surgery Neuraxial Intrathecal Kanazi and colleagues 26 5 TURBT, TURP Al-Mustafa and colleagues 51 5 TURBT, TURP, TVT Block/ use n Groups (n) Local anaesthetic Spinal/ 60 (16) (2) Clonidine+ (16) (3) Bupivacaine (19) Spinal/ 66 (1) Dex 5 mg+ (21) (2) Dex 10 mg+ (21) (3) NS+ (22) Eid and 5 ACL Spinal/ 48 colleagues 52 (1) Dex 10 mg+ (15) (2) Dex 15 mg+ (16) (3) NS+ (16) Gupta and 4 Lower colleagues 53 extremity Spinal/ 60 ropivacaine (2) NS+ ropivacaine 12 mg of 0.75% hyperbaric 12.5 mg of 0.5% isobaric 3 ml of 0.5% hyperbaric 3 ml of 0.75% ropivacaine Adjuvant (1) Dex 3 mg (2) Clonidine (3) None (1) Dex 5 mg (2) Dex 10 mg (3) NS (1) Dex 10 mg (2) Dex 15 mg (3) NS (1) Dex 5 mg (2) NS Primary outcome Time to first analgesic request Block characteristics Analgesic outcomes Dex-related adverse effects onset time Motor onset time Motor Postop pain Time to first analgesic request Analgesic Hypotension Bradycardia Postop Respiratory consumption sedation depression BJA Abdallah and Brull

919 Shukla and 5 Abdominal, colleagues 54 lower extremity Peripheral Brachial plexus Esmaoglu and colleagues 25 4 Forearm, hand Gandhi and 4 Forearm, colleagues 27 hand Kaygusuz 55 5 Forearm, hand Ammar and 5 Forearm, Mahmoud 56 hand Spinal/ 90 (2) Magnesium+ (3) NS+ 2 ml of 0.75% hyperbaric AXB/ 60 40 ml of 0.5% levo levo (2) NS+ levo SCB/ 75 (35) (2) NS+ (35) 38 ml 0.25% AXB/ 64 39 ml 0.5% levo (2) NS+ levo ICB/ 60 (2) NS+ (1) Dex 10 mg (2) Magnesium (3) NS (1) Dex 100 mg in 1ml (2) NS 1 ml (1) Dex mg in 2ml (2) NS 2 ml levo (1) Dex 1 mgkg 21 ml 0.33% in 1 ml (2) NS 1 ml (1) Dex 0.75 mg kg 21 in 1 ml (2) NS 1 ml N/D Motor N/D Time to first analgesic request Perineural dexmedetomidine BJA

BJA Abdallah and Brull Dexmedetomidine OR Medetomidine Analgesia OR Pain Relief OR Pain Control OR Pain Prevention OR Pain Management 37 Records identified through other sources 144 Records 139 Records after removing duplicates 139 Records screened 37 Articles assessed 9 Trials included AND AND 107 Records Perineural OR Adjuvant OR Adjunct OR Admixture Regional Anaesthesia OR Nerve Block OR Block OR Neuraxial Block OR Central Block OR Peripheral Block 102 Excluded records 28 Excluded articles (Appendix) Fig 1 Flowchart summarizing retrieved, included, and excluded RCTs. three trials 26 51 53 reported no difference in sedation between the patients who received intrathecal dexmedetomidine as an adjunct to spinal anaesthesia and those who did not. We could not quantitatively analyse our remaining secondary endpoints (pain scores, analgesic consumption) due to inconsistent reporting and heterogeneous assessment protocols within and between the source trials. Qualitatively, these results favour the dexmedetomidine group and are presented in Table 4. Discussion Our review of the literature suggests that the use of dexmedetomidine as a perineural adjuvant can prolong the s of both sensory and motor produced by long-acting LAs in spinal s. For BP s, perineural dexmedetomidine can prolong the of motor ; however, the trend towards prolonged sensory did not reach statistical significance. Dexmedetomidine also hastens the onset of sensory in spinal anaesthesia and prolongs the time to first analgesic request in the setting of both spinal anaesthesia and BP. The advantages of dexmedetomidine may be offset by an increased likelihood of transient, reversible bradycardia, and the prolongation of motor when it is undesirable. In the context of perineural adjuvants, the efficacy of dexmedetomidine appears to be comparable with buprenorphine 16 and dexamethasone 18 58 59 when administered peripherally, and exceeds that of clonidine, magnesium, and midazolam for both intrathecal 19 20 26 54 60 and peripheral 17 21 61 applications. However, unlike clonidine, another a 2 adrenoreceptor agonist shown capable of prolonging the Study or Subgroup Neuraxial-intrathecal Kanazi 2006 26 Al-Mustafa 2009 51 Eid 2011 52 Gupta 2011 53 Shukla 2011 54 Subtotal (95% CI) Dex Mean [Min] 3 283 365.78 468.3 352 SD [Min] 75 66.65 79.76 36.8 45 Total 16 42 31 149 Heterogeneity: τ 2 = 3607.38; χ 2 = 88.02, df = 4 (P<0.00001); l 2 = 95% Test for overall effect: Z = 5.42 (P < 0.00001) Control Mean difference Mean [Min] SD [Min] Total Weight I.V. random, 95% CI [Min] 190 165.5 238 239.3 194 48 32.9 57 16.8 55 19 22 16 117 11.0% 11.1% 11.0% 11.2% 11.1% 55.5% 113.00 [70.38, 155.62] 117.50 [93.10, 141.90] 127.78 [88.18, 167.38] 229.00 [214.52, 243.48] 158.00 [132.57, 183.43] 150.35 [95.94, 204.75] Mean difference I.V. random, 95% CI [Min] Peripheral-brachial plexus Esmaoglu 2010 25 Gandhi 2012 27 Kaygusuz 2012 55 Ammar 2012 56 Subtotal (95% CI) 887 732.4 924.15 197.4 66.23 48.9 78.27 14.4 35 125 673 146.5 664.62 122.7 Heterogeneity: τ 2 = 82697.47; χ 2 = 2223.07, df = 3 (P<0.00001); l 2 = 100% Test for overall effect: Z = 1.97 (P = 0.05) 73.77 36.4 61.7 15.2 35 125 11.1% 11.2% 11.1% 11.2% 44.5% 214.00 [178.52, 249.48] 585.90 [565.70, 606.10] 259.53 [223.87, 295.19] 74.70 [67.21, 82.19] 283.54 [1.39, 565.68] Total (95% CI) 274 Heterogeneity: τ 2 = 31010.66; χ 2 = 2348.66, df = 8 (P<0.00001); l 2 = 100% Test for overall effect: Z = 3.55 (P = 0.0004) Test for subgroup differences: χ 2 = 0.83, df = 1 (P < 0.36); l 2 = 0% 242 100.0% 208.98 [93.51, 324.44] 500 250 0 250 500 Favours control Favours Dex Fig 2 Forest plot showing sensory. The sample size, mean, standard deviations, and the pooled estimates of the mean difference are shown. The 95% CIs are shown as lines for individual studies and as diamonds for pooled estimates. 920

Perineural dexmedetomidine BJA Table 3 Quantitative results. N/A, not applicable Block type/outcome Neuraxial Intrathecal onset (min) Motor onset (min) Motor (min) Time to first analgesic request (min) Incidence of hypotension (n/n) Incidence of bradycardia (n/n) Peripheral Brachial plexus onset (min) Motor onset (min) Motor (min) Time to first analgesic request (min) Incidence of hypotension (n/n) Incidence of bradycardia (n/n) Studies included Dex mean or n/n Control mean or n/n Odds ratio or weighed mean (95% confidence interval) P-value for statistical significance P-value for heterogeneity 26 51 54 5.53 6.82 21.54 (22.96, 20.11) 0.04 0.00001 90% 26 51 54 9.33 13.15 24.52 (29.17, 0.13) 0.06 0.00001 95% 26 51 52 54 294.46 156.45 132.22 (87.69, 176.74) 0.00001 0.00001 93% 52 53 525.68 231.37 292.87 (174.32, 411.41) 0.00001 0.0003 92% 26 51 54 9/156 8/118 0.73 (0.20, 2.71) 0.64 0.26 25% 26 51 54 3/156 2/118 0.97 (0.04, 21.49) 0.98 0.11 60% 25 27 55 56 13.19 14.89 21.9 (25.08, 1.28) 0.24 0.00001 97% 25 27 55 56 12.58 14.15 21.67 (24.93, 1.58) 0.31 0.00001 97% 25 27 55 56 600.72 321.20 267.76 (15.47, 520.06) 0.04 0.00001 100% 25 27 55 56 850.97 500.21 344.95 (102.68, 587.23) 0.005 0.00001 99% 25 27 55 56 2/127 0/127 5. (0.25, 114.47) 0.29 N/A N/A 25 27 55 56 9/127 0/127 10.52 (1.27, 87.08) 0.03 0.54 0% I 2 test for heterogeneity Table 4 Qualitative results. +, favours Dex;, no difference; Dex, dexmedetomidine; VAS, visual analogue scale Study Postoperative pain Analgesic consumption Comments Neuraxial Intrathecal Kanazi and colleagues 26 Al-Mustafa and colleagues 51 Eid and colleagues 52 + + Dex reduces rest VAS pain scores at 8, 12, 24 h (P,0.05). Dex reduces dynamic VAS pain scores at 4, 8, 12, 24 h (P,0.05). Dex reduces i.v. diclofenac consumption by 45% at 24 h; Dex 77.4 mg, control 140.6 mg (P,0.05) Gupta and colleagues 53 + + Dex reduces maximum VAS pain scores during first 24 h by 35%; Dex 4.4, control 6.8 (P,0.001). Dex reduces i.m. diclofenac consumption by 64% at 24 h; Dex 72.8 mg, control 202.5 mg (P,0.001) Peripheral Brachial plexus Ammar and Mahmoud 56 + + Dex reduces rest VAS pain scores at 1, 2, 12, 24, 36, 48 h (P,0.05). Dex reduces i.v. morphine consumption by 64% at 48 h; Dex 4.9 mg, control 13.6 mg (P¼0.005) 921

BJA Abdallah and Brull of shorter acting 62 64 but not long-acting 65 66 LAs, this review demonstrated that dexmedetomidine clearly prolonged the of long-acting LAs. Also, while clonidine produces preferential extension of sensory, 67 our review suggests that dexmedetomidine prolongs both sensory and motor, a difference that may be disadvantageous by delaying rehabilitation and/or discharge, or worse, precipitating falls. The results of our review are subject to several limitations. The trials included herein were small and characterized by high levels of heterogeneity, factors that limit the clinical combinability of the source trials, and the generalizability of our results. Similarly, the present safety data, however limited, may not apply to other types as local neurotoxicity and systemic uptake are both influenced by site-specific perfusion levels. 15 The generalizability of this review is further limited by publication bias as all source studies originated from Middle Eastern countries, which may or may not reflect less stringent Institutional Review Board and/or editorial board policies. 68 The lack of United States (US) Food and Drug Administration (FDA) approval for the perineural application of dexmedetomidine 69 almost certainly explains the scant appearance of trials on this topic in US-based indexed journals. 70 73 The lack of available data from trials already in progress, including at least one phase III trial, 74 may further contribute to publication bias. Finally, considerable differences existed in the doses of perineural dexmedetomidine; doses varied between 3, 26 5, 51 53 10, 51 52 54 or 15 mg 52 for the intrathecal route, and, 27 100 mg, 25 0.75, 56 or 1 mg kg 21 55 for the peripheral route. While these dosing inconsistencies most likely reflect the absence of human dose response studies and/or extrapolation from animal studies, 75 82 the variable doses of dexmedetomidine, the different types of LAs used, and the variation between trials in selecting endpoints that defined hypotension, 51 53 bradycardia, 27 55 and onsets of sensory 25 27 and motor 25 27 may have contributed to the statistically significant heterogeneity of the pooled results. Despite the variability in dexmedetomidine doses used, it is nonetheless noteworthy that significantly prolonged s of sensory and motor s and also time to first analgesic request were achieved even with the lowest doses of dexmedetomidine, namely, 3 mg for intrathecal and mg of peripheral administration. While we could not demonstrate any association between perineural dexmedetomidine and the frequency of hypotension or respiratory depression, the trials examined herein were not specifically designed to assess safety. While dexmedetomidine may appear safe in the short term, 83 84 systematic preclinical and subsequent human neurotoxicity data, including the investigation of potential delayed adverse neurological effects and effects related to prolonged perineural exposure, are lacking. 85 Indeed, relevant neurotoxicity data seem contradictory; while a number of reports suggest that dexmedetomidine is protective against hypoxic ischaemic neuronal injury in rat and human neonatal asphyxia 86 92 models, dexmedetomidine has also been shown to cause moderate to severe demyelination in white matter when doses as high as 6.1 mg kg 21 were administered via an epidural route in rabbits. 93 While the doses used in the trials reviewed herein did not exceed 0.2 mg kg 21 for intrathecal and 1 mg kg 21 for peripheral administration, the hazards of drawing conclusions of safety based on isolated small animal data are self-evident. 15 Finally, none of the reviewed trials justified the safety and compatibility of their dexmedetomidine LA mixture for peripheral and neuraxial use. In summary, dexmedetomidine is a potential LA adjuvant that can exhibit a facilitatory effect when administered intrathecally as part of spinal anaesthesia or peripherally as part of a BP. However, there are presently insufficient safety data to support the use of perineural dexmedetomidine in the clinical setting. Declaration of interest None declared. Funding This work was supported by departmental funding. References 1 Liu SS. A comparison of regional versus general anesthesia for ambulatory anesthesia: a meta-analysis of randomized controlled trials. Anesth Analg 2005; 101: 1634 42 2 Liu SS. The effect of analgesic technique on postoperative patient-reported outcomes including analgesia: a systematic review. Anesth Analg 2007; 105: 789 808 3 McCartney CJ, Brull R, Chan VW, et al. Early but no long-term benefit of regional compared with general anesthesia for ambulatory hand surgery. Anesthesiology 2004; 101: 461 7 4 Covino BJ, Wildsmith JA. Clinical pharmacology of local anesthetic agents. In: Cousins MJ, Bridenbaugh PO, eds. Neural Blockade in Clinical Anesthesia and Management of Pain. Philadelphia, PA: Lippincott-Raven, 1998; 97 128 5 Kuzma PJ, Kline MD, Calkins MD, Staats PS. Progress in the development of ultra-long-acting local anesthetics. Reg Anesth 1997; 22: 543 51 6 Grant SA. Continuous peripheral nerve for ambulatory surgery. Reg Anesth Pain Med 2001; 26: 209 7 Gramke HF, de Rijke JM, van Kleef M, et al. The prevalence of postoperative pain in a cross-sectional group of patients after day-case surgery in a university hospital. Clin J Pain 2007; 23: 543 8 8 Sommer M, de Rijke JM, van Kleef M, et al. The prevalence of postoperative pain in a sample of 1490 inpatients. Eur J Anaesthesiol 2008; 25: 267 74 9 Schoenmakers KP, Wegener JT, Stienstra R. Effect of local anesthetic volume (15 vs 40 ml) on the of ultrasound-guided single shot axillary brachial plexus : a prospective randomized, observer-blinded trial. Reg Anesth Pain Med 2012; 37: 242 7 10 Scott DB. Acute toxicity of ropivacaine compared with that of. Anesth Analg 1989; 69: 563 9 11 Ilfeld BM. Continuous peripheral nerve s: a review of the published evidence. Anesth Analg 2011; 113: 904 25 12 Capdevila X. Continuous peripheral nerve s in clinical practice. Curr Opin Anaesthesiol 2008; 21: 619 23 922

Perineural dexmedetomidine BJA 13 Salinas FV. The effect of single-injection femoral nerve versus continuous femoral nerve after total knee arthroplasty on hospital length of stay and long-term functional recovery within an established clinical pathway. Anesth Analg 2006; 102: 1234 9 14 Golf M, Daniels SE, Onel E. A phase 3, randomized, placebocontrolled trial of DepoFoam w (extended-release local analgesic) in bunionectomy. Adv Ther 2011; 28: 776 88 15 Yaksh TL, Ilfeld BM, Wiese AJ. Perineural local anesthetic and adjuvant action: the meaning of an ex vivo data set for efficacy and safety. Reg Anesth Pain Med 2012; 37: 366 8 16 Candido KD. Buprenorphine added to the local anesthetic for axillary brachial plexus prolongs postoperative analgesia. Reg Anesth Pain Med 2002; 27: 162 7 17 McCartney CJ, Duggan E, Apatu E. Should we add clonidine to local anesthetic for peripheral nerve ade? A qualitative systematic review of the literature. Reg Anesth Pain Med 2007; 32: 3 8 18 Parrington S, O Donnell D, Chan VW, et al. Dexamethasone added to mepivacaine prolongs the of analgesia after supraclavicular brachial plexus ade. Reg Anesth Pain Med 2010; 35: 422 6 19 Buvanendran A, McCarthy RJ, Kroin JS, Leong W, Perry P, Tuman KJ. Intrathecal magnesium prolongs fentanyl analgesia: a prospective, randomized, controlled trial. Anesth Analg 2002; 95: 661 6 20 Murali Krishna T, Panda NB, Batra YK, Rajeev S. Combination of low doses of intrathecal ketamine and midazolam with improves postoperative analgesia in orthopaedic surgery. Eur J Anaesthesiol 2008; 25: 299 6 21 Laiq N, Khan MN, Arif M, Khan S. Midazolam with for improving analgesia quality in brachial plexus for upper limb surgeries. J Coll Physicians Surg Pak 2008; 18: 674 8 22 Gabriel JS, Gordin V. Alpha 2 agonists in regional anesthesia and analgesia. Curr Opin Anaesthesiol 2001; 14: 751 3 23 Memiş D, Turan A, Karamanlioğlu B, Pamuku Z, Kurt I. Adding dexmedetomidine to lidocaine for intravenous regional anesthesia. Anesth Analg 2004; 98: 835 40 24 Esmaoglu A, Mizrak A, Akin A, Turk Y, Boyaci A. Addition of dexmedetomidine to lidocaine for intravenous regional anaesthesia. Eur J Anaesthesiol 2005; 22: 447 51 25 Esmaoglu A, Yegenoglu F, Akin A, Turk C. Dexmedetomidine added to levo prolongs axillary brachial plexus. Anesth Analg 2010; 111: 1548 51 26 Kanazi GE, Aouad MT, Jabbour Khoury SI, et al. Effect of low-dose dexmedetomidine or clonidine on the characteristics of spinal. Acta Anaesthesiol Scand 2006; 50: 222 7 27 Gandhi R, Shah A, Patel I. Use of dexmedetomidine along with for brachial plexus. National J Med Res 2012; 2: 67 9 28 Moher D. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009; 6: e1000097 29 Al-Metwalli RR, Mowafi HA, Ismail SA, et al. Effect of intra-articular dexmedetomidine on postoperative analgesia after arthroscopic knee surgery. Br J Anaesth 2008; 101: 395 9 El-Hamamsy M, Dorgham D. Intra-articular adjuvant analgesics following knee arthroscopy: comparison between dexmedetomidine and fentanyl. Res J Medicine Med Sci 2009; 4: 355 60 31 Cheung CW, Ng KF, Liu J, Yuen MY, Ho MH, Irwin MG. Analgesic and sedative effects of intranasal dexmedetomidine in third molar surgery under local anaesthesia. Br J Anaesth 2011; 107: 4 7 32 Paul S, Bhattacharjee DP, Ghosh S, Dawn S, Chatterjee N. Efficacy of intra-articular dexmedetomidine for postoperative analgesia in arthroscopic knee surgery. Ceylon Med J 2010; 55: 111 5 33 Paswan AK, Prakash S. The effect of intra-articular dexmedetomidine and morphine on postoperative analgesia for arthroscopy knee surgery. Indian J Res 2011; 5: 6 10 34 Mizrak A, Gul R, Ganidagli S, Karakurum G, Keskinkilic G, Oner U. Dexmedetomidine premedication of outpatients under IVRA. Middle East J Anesthesiol 2011; 21: 53 60 35 Jaakola ML. Dexmedetomidine premedication before intravenous regional anesthesia in minor outpatient hand surgery. J Clin Anesth 1994; 6: 204 11 36 Kol I, Ozturk H, Kaygusuz K, Gursoy S, Comert B, Mimaroglu C. Addition of dexmedetomidine or lornoxicam to prilocaine in intravenous regional anaesthesia for hand or forearm surgery: a randomized controlled study. Clin Durg Investig 2009; 29: 121 9 37 Mizrak A, Gul R, Erkutlu I, Alptekin M, Oner U. Premedication with dexmedetomidine alone or together with 0.5% lidocaine for IVRA. J Surg Res 2010; 164: 242 7 38 Nasr Y, Waly SH. Lidocaine tramadol versus lidocaine dexmedetomidine for intravenous regional anesthesia. Eg J Anaesth 2012; 28: 37 42 39 El-Hakim M, Abdelhamid D, Abdelfattach H, Magdy H, El Sayed A, El Shafei M. Effect of epidural dexmedetomidine on intraoperative awareness and post-operative pain after one-lung ventilation. Acta Anaesthesiol Scand 2010; 54: 703 9 40 Sinha S, Mukherjee M, Chatterjee S, Vijay MK, Hazra A, Ray M. Comparative study of analgesic efficacy of ropivacaine with ropivacaine plus dexmedetomidine for paravertebral in unilateral renal surgery. Anesth Pain Intensive Care 2012; 16: 38 42 41 El-Hennawy AM, Abd Elwahab AM, Abd Elmaksoud AM, El Ozairy HS, Boulis SR. Addition of clonidine or dexmedetomidine to prolongs caudal analgesia in children. Br J Anaesth 2009; 103: 268 74 42 Bajwa SJ, Arora V, Kaur J, Singh A, Parmar S. Comparative evaluation of dexmedetomidine and fentanyl for epidural analgesia in lower limb orthopedic surgeries. Saudi J Anaesth 2011; 5: 365 70 43 Neogi M, Bhattacharjee D, Dawn S, Chatterjee N. A comparative study between clonidine and dexmedetomidine used as adjuncts to ropivacaine for caudal analgesia in paediatric patients. J Anaesthesiol Clin Pharmacol 2010; 26: 149 53 44 Saadawy I, Boker A, Elshahawy MA, et al. Effect of dexmedetomidine on the characteristics of in a caudal in pediatrics. Acta Anaesthesiol Scand 2009; 53: 251 6 45 Jadad AR, Moore RA, Carroll D, et al. Assessing the quality of reports of randomized clinical trials: is blinding necessary? Control Clin Trials 1996; 17: 1 12 46 Liu SS, Warren D, Wu CL, et al. A lovely idea: forming an ASRA acute postoperative pain (AcutePOP) database. Reg Anesth Pain Med 2006; 31: 291 3 47 Aantaa RE. Dexmedetomidine premedication for minor gynecologic surgery. Anesth Analg 1990; 70: 407 13 48 Ebert TJ, Hall JE, Barney JA, Uhrich TD, Colinco MD. The effects of increasing plasma concentrations of dexmedetomidine in humans. Anesthesiology 2000; 93: 382 94 49 Higgins JP, Deeks JJ, Altman DG, eds. Chapter 16: Special topics in statsitics. In: Higgins JP, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions. Baltimore, MD: The Cochrane Collaboration, 2011 923

BJA Abdallah and Brull 50 Higgins JP. Quantifying heterogeneity in a meta-analysis. Stat Med 2002; 21: 1539 58 51 Al-Mustafa MM, Abu-Halaweh SA, Aloweidi AS, et al. Effect of dexmedetomidine added to spinal for urological procedures. Saudi Med J 2009; : 365 70 52 Eid H, Shafie M, Youssef H. Dose-related prolongation of hyperbaric spinal anesthesia by dexmedetomidine. ASJA 2011; 4: 83 95 53 Gupta R, Bogra J, Verma R, Kohli M, Kushwaha J, Kumar S. Dexmedetomidine as an intrathecal adjuvant for postoperative analgesia. Indian J Anaesth 2011; 55: 347 51 54 Shukla D, Verma A, Agarwal A, Pandey H, Tyagi C. Comparative study of intrathecal dexmedetomidine with intrathecal magnesium sulfate used as adjuvants to. J Anaesthesiol Clin Pharmacol 2011; 27: 495 9 55 Kaygusuz K. Effects of adding dexmedetomidine to levo in axillary brachial plexus. Curr Ther Res Clin Exp 2012; 73: 103 11 56 Ammar AS, Mahmoud KM. Ultrasound-guided single injection infraclavicular brachial plexus using alone or combined with dexmedetomidine for pain control in upper limb surgery: a prospective randomized controlled trial. Saudi J Anaesth 2012; 6: 109 14 57 Gupta R, Verma R, Bogra J, Kohli M, Raman R, Kushwaha J. A comparative study of intrathecal dexmedetomidine and fentanyl as adjuvants to. J Anaesthesiol Clin Pharmacol 2011; 27: 339 43 58 Vieira P, Pulai I, Tsao G, Manikantan P, Keller B, Connelly N. Dexamethasone with increases of analgesia in ultrasound-guided interscalene brachial plexus ade. Eur J Anaesthesiol 2010; 27: 285 8 59 Cummings KC, Napierkowski DE, Parra-Sanchez I, et al. Effect of dexamethasone on the of interscalene nerve s with ropivacaine or. Br J Anaesth 2011; 107: 446 53 60 Ozalevli M, Cetin TO, Unlugenc H, Guler T, Isik G. The effect of adding intrathecal magnesium sulphate to fentanyl spinal anaesthesia. Acta Anaesthesiol Scand 2005; 49: 1514 9 61 Abosedira MA. Adding clonidine or dexmedetomidine to lidocaine during biers : a comparative study. J Med Sci 2008; 8: 660 4 62 Singelyn FJ, Dangoisse M, Bartholome S, Gouverneur JM. Adding clonidine to mepivacaine prolongs the of anesthesia and analgesia after axillary brachial plexus. Reg Anesth 1992; 17: 148 50 63 Singelyn FJ, Gouverneur JM, Robert A. A minimum dose of clonidine added to mepivacaine prolongs the of anesthesia and analgesia after axillary brachial plexus. Anesth Analg 1996; 83: 1046 50 64 Bernard JM. Dose-range effects of clonidine added to lidocaine for brachial plexus. Anesthesiology 1997; 87: 277 65 Culebras X, Van Gessel E, Hoffmeyer P, Gamulin Z. Clonidine combined with a long acting local anesthetic does not prolong postoperative analgesia after brachial plexus but does induce hemodynamic changes. Anesth Analg 2001; 92: 199 204 66 Duma A, Urbanek B, Sitzwohl C, Kreiger A, Zimpfer M, Kapral S. Clonidine as an adjuvant to local anaesthetic axillary brachial plexus : a randomized, controlled study. Br J Anaesth 2005; 94: 112 6 67 Rhee K, Kang K, Kim J, Jeon Y. Intravenous clonidine prolongs spinal anesthesia. Acta Anaesthesiol Scand 2003; 47: 1001 5 68 Abou-Zeid A, Afzal M, Silverman HJ. Capacity mapping of national ethics committees in the eastern Mediterranean region. BMC Med Ethics 2009; 10: 8 15 69 US Food and Drug Administration. Precedex (dexmedetomidine hydrochloride) prescribing information October 2010. Available from http://www.accessdata.fda.gov/drugsatfda_docs/label/ 2010/021038s017lbl.pdf (accessed 20 August 2012) 70 Neal JM, Rathmell JP, Rowlingson JC. Publishing studies that involve off-label use of drugs: formalizing Regional Anesthesia and Pain Medicine s policy. Reg Anesth Pain Med 2009; 34: 391 2 71 Editorial Board. 2010 Anesthesia & Analgesia guide for authors. Anesth Analg 2010; 111: 525 38 72 Rowbotham MC. PAIN s policy on the spinal administration of drugs. Pain 2010; 149: 415 6 73 Editorial Board. 2011 Canadian Journal of Anesthesia guide for authors. Can J Anaesth 2011; 58: 668 96 74 ClinicalTrials.gov. Efficacy and safety study of dexmedetomidine as an additive to local anesthetics in shoulder surgery. Available from http://clinicaltrials.gov/ct2/show/nct01557270?term= dexmedetomidine&rank=45 (accessed 20 August 2012) 75 Takano Y, Yaksh TL. Characterization of the pharmacology of intrathecally administered alpha-2 agonists and antagonists in rats. J Pharmacol Exp Ther 1992; 261: 764 72 76 Eisenach JC, Shafer SL, Bucklin BA, Jackson C, Kallio A. Pharmacokinetics and pharmacodynamics of intraspinal dexmedetomidine in sheep. Anesthesiology 1994; 80: 1349 59 77 Xu M, Kontinen VK, Kalso E. Effects of radolmidine, a novel alpha2-adrenergic agonist compared with dexmedetomidine in different pain models in the rat. Anesthesiology 2000; 93: 473 81 78 Onttonen T, Pertovaara A. The mechanical antihyperalgesic effect of intrathecally administered MPV-2426, a novel alpha2-adrenoceptor agonist, in a rat model of postoperative pain. Anesthesiology 2000; 92: 1740 5 79 Horvath G, Joo G, Dobos I, Klimscha W, Toth G, Benedek G. The synergistic antinociceptive interactions of endomorphin-1 with dexmedetomidine and/or S(+)-ketamine in rats. Anesth Analg 2001; 93: 1018 24 80 Talke P, Xu M, Paloheimo M, Kalso E. Effects of intrathecally administered dexmedetomidine, MPV-2426 and tizanidine on EMG in rats. Acta Anaesthesiol Scand 2003; 47: 347 54 81 Shimode N, Fukuoka T, Tanimoto M, Tashiro C, Tokunaga A, Noguchi K. The effects of dexmedetomidine and halothane on fos expression in the spinal dorsal horn using a rat postoperative pain model. Neurosci Lett 2003; 343: 45 8 82 Lo WC, Harris J, Clarke RW. Endogenous opioids support the spinal inhibitory action of an alpha 2-adrenoceptor agonist in the decerebrated, spinalised rabbit. Neurosci Lett 2003; 340: 95 8 83 Brummett C, Norat M, Palmisano J, Lydic R. Perineural administration of dexmedetomidine in combination with enhances sensory and motor ade in sciatic nerve without inducing neurotoxicity in rat. Anesthesiology 2008; 109: 502 11 84 Brummett CM, Padda A, Amodeo F, Welch K, Lydic R. Perineural dexmedetomidine added to ropivacaine causes a dosedependent increase in the of thermal antinociception in sciatic nerve in rat. Anesthesiology 2009; 111: 1111 9 85 Hodgson PS, Neal JM, Pollock JE, Liu SS. The neurotoxicity of drugs given intrathecally (spinal). Anesth Analg 1999; 88: 797 809 86 Hoffman WE, Kochs E, Werner C, Thomas C, Albrecht RF. Dexmedetomidine improves neurologic outcome from incomplete ischemia in the rat. reversal by the alpha 2-adrenergic antagonist atipamezole. Anesthesiology 1991; 75: 328 32 87 Maier C, Steinberg GK, Sun GH, Zhi GT, Maze M. Neuroprotection by the alpha 2-adrenoreceptor agonist dexmedetomidine in a focal model of cerebral ischemia. Anesthesiology 1993; 79: 6 12 924

Perineural dexmedetomidine BJA 88 Halonen T, Kotti T, Tuunanen J, Toppinen A, Miettinen R, Riekkinen PJ. Alpha 2-adrenoceptor agonist, dexmedetomidine, protects against kainic acid-induced convulsions and neuronal damage. Brain Res 1995; 693: 217 24 89 Kuhmonen J, Pokorny J, Miettinen R, et al. Neuroprotective effects of dexmedetomidine in the gerbil hippocampus after transient global ischemia. Anesthesiology 1997; 87: 371 7 90 Jolkkonen J, Puurunen K, Koistinaho J, et al. Neuroprotection by the a2-adrenoceptor agonist, dexmedetomidine, in rat focal cerebral ischemia. Eur J Pharmacol 1999; 372: 31 6 91 Laudenbach V, Mantz J, Lagercrantz H, Desmonts JM, Evrard P, Gressens P. Effects of alpha(2)-adrenoceptor agonists on perinatal excitotoxic brain injury: comparison of clonidine and dexmedetomidine. Anesthesiology 2002; 96: 134 41 92 Ma D, Hossain M, Rajakumaraswamy N, et al. Dexmedetomidine produces its neuroprotective effect via the a2a-adrenoceptor subtype. Eur J Pharmacol 2004; 502: 87 97 93 Konakci S, Adanir T, Yilmaz G, Rezanko T. The efficacy and neurotoxicity of dexmedetomidine administered via the epidural route. Eur J Anaesthesiol 2008; 25: 403 9 Appendix: Excluded trials First author Reference Reason for exclusion Abosedira J Med Sci 2008; 8: 660 4 Comparator Al-Ghanem Am J Appl Sci 2009; 6: 882 87 Population Al-Metwalli Br J Anaesth 2008; 101: 395 9 Intervention Anand Indian J Anaesth 2011; 55: 340 Population Bajwa Indian J Anaesth 2011; 55: 116 21 Comparator Bajwa Saudi J Anaesth 2011; 5: 365 70 Comparator Cheung Br J Anaesth 2011; 107: 4 7 Population, intervention El-Hakim Acta Anaesthesiol Scand 2010; 54: 703 9 Intervention El-Hamamsy Res J Medicine Med Sci 2009; 4: 355 60 Intervention El-Hennawy Br J Anaesth 2009; 103: 268 74 Population Esmaoglu Eur J Anaesthesiol 2005; 22: 447 51 Intervention Gupta J Anaesthesiol Clin Pharmacol 2011; 27: 339 Comparator Jaakola J Clin Anesth 1994; 6: 204 11 Intervention Jain South Afr J Anaesth 2012; 18: 105 9 Intervention Kol Clin Durg Investig 2009; 29: 121 9 Intervention Memiş Anesth Analg 2004; 98: 835 40 Intervention Mizrak J Surg Res 2010; 164: 242 7 Intervention Mirzak Middle East J Anesthesiol 2011; 21: 53 60 Intervention Nasr Eg J Anaesth 2012; 28: 37 42 Intervention Obayah Eur J Anaesthesiol 2010; 27: 280 4 Population Paswan Indian J Res 2011; 5: 6 10 Intervention Paul Ceylon Med J 2010; 55: 111 5 Intervention Saadawy Acta Anaesthesiol Scand 2009; 53: 251 6 Population Salagado Rev Assoc Méd Bras 2008; 54: 110 5 Language Schnaider Rev Bras Anestesiol 2005; 55: 525 31 Design: non-randomized Sinha Anaesth Pain Intensive Care 2012; 16: 38 42 Intervention Neogi J Anaesthesiol Clin Pharmacol 2010; 26: 149 53 Population Vieira Rev Bras Anestesiol 2004; 54: 473 8 Comparator Handling editor: R. P. Mahajan 925