Intravenous Dexamethasone and Perineural Dexamethasone

ORIGINAL ARTICLE

Intravenous Dexamethasone and Perineural DexamethasoneSimilarly Prolong the Duration of Analgesia After

Supraclavicular Brachial Plexus BlockA Randomized, Triple-Arm, Double-Blind, Placebo-Controlled Trial

Faraj W. Abdallah, MD,* James Johnson, MD,† Vincent Chan, MD, FRCPC,†Harry Murgatroyd, MB, ChB, BSc, FRCA,† Mohammad Ghafari, MD,† Noam Ami, MSc,†

Rongyu Jin, MD,† and Richard Brull, MD, FRCPC†

Background and Objectives: Perineural dexamethasone prolongs theduration of single-injection peripheral nerve block when added to the localanesthetic solution. Postulated systemic mechanisms of action along withtheoretical safety concerns have prompted the investigation of intravenousdexamethasone as an alternative, with decidedly mixed results. We aimedto confirm that addition of intravenous dexamethasonewill prolong the du-ration of analgesia after single-injection supraclavicular block comparedwith conventional long-acting local anesthetic alone or in combinationwithperineural dexamethasone for ambulatory upper extremity surgery.Methods: Seventy-five patientswere randomized to receive supraclavicularblock using 30-mL bupivacaine 0.5% alone (Control), with concomitantintravenous dexamethasone 8 mg (DexIV), or with perineural dexametha-sone 8mg (DexP). Duration of analgesiawas designated as the primary out-come. To test our hypothesis, the superiority of DexIV was first comparedwith Control and then with DexP. Motor block duration, pain scores, opioidconsumption, opioid-related side effects, patient satisfaction, and block-related complications were also analyzed.Results: Twenty-five patients per group were analyzed. The duration ofanalgesia (mean [95% confidence interval]) was prolonged in the DexIVgroup (25 hours [17.6–23.6]) compared with Control (13.2 hours [11.5–15.0]; P < 0.001) but similar to the DexP group (25 hours [19.5–27.4];P = 1). The DexIV group experienced longer motor block (30.1 hours)compared with DexP (25.5 hours) and Control (19.7 hours) groups. BothDexIV and DexP had reduced pain scores, reduced postoperative opioidconsumption, and improved satisfaction compared with Control.Conclusions: In a single-injection supraclavicular blockwith long-actinglocal anesthetic, the effectiveness of intravenous dexamethasone in pro-longing the duration of analgesia seems similar to perineural dexamethasone.

(Reg Anesth Pain Med 2015;40: 125–132)

Avariety of novel local anesthetic adjuncts have been studiedin an attempt to prolong the effects of single-injection periph-

eral nerve blockade (PNB) often with variable success1–6 andunproven safety.7,8 Among these adjuncts, dexamethasone hasgained considerable interest because it has been shown to prolong

From the Departments of Anesthesia at *St. Michael’s Hospital and †TorontoWestern Hospital; and University of Toronto, Toronto, Ontario, Canada.Accepted for publication December 3, 2014.Address correspondence to: Richard Brull, MD, FRCPC, Department of

Anesthesia, Toronto Western Hospital, 399 Bathurst St, Toronto, Ontario,Canada M5T 2S8 (e‐mail: [email protected]).

Dr Vincent Chan receives equipment support from BK Medical, PhilipsMedical Systems, SonoSite, and Ultrasonix.

The other authors declare no conflict of interest.Drs Faraj Abdallah and Richard Brull are supported by the Merit Award

Program, Department of Anesthesia, University of Toronto.Copyright © 2015 by American Society of Regional Anesthesia and Pain

MedicineISSN: 1098-7339DOI: 10.1097/AAP.0000000000000210

Regional Anesthesia and Pain Medicine • Volume 40, Number 2, March

Copyright © 2015 American Society of Regional Anesthesia and Pain

significantly the duration of analgesia when applied perineurallyin combination with local anesthetics.4,9 Although dexameth-asone has been used both peripherally10 and centrally11 in thesetting of chronic pain management for decades, its true mech-anism and site of action (ie, local or central) are not fully under-stood. Indeed, dexamethasone has even been shown to reducepostsurgical pain after a variety of surgical procedures when ad-ministered systemically in the absence of any nerve block.12

Given the theoretical concerns of neurotoxicity associated with lo-cally applied perineural dexamethasone,8 understanding the site ofaction for dexamethasone as a local anesthetic adjunct, and themost effective and safe route of administration, is ever more im-portant. As such, several investigators have attempted recently tocompare the effects of perineural and systemic dexamethasoneon the duration of analgesia after PNB, with decidedly mixed re-sults. Two separate studies13,14 found that only perineural dexa-methasone, and not intravenous dexamethasone, prolongs theduration of analgesia when used in combination with interscaleneblock and sciatic block, respectively. By contrast, 2 other stud-ies15,16 demonstrated that the effects of perineural and systemicdexamethasone on the duration of analgesia are equivalent in thesetting of interscalene block and sciatic block, respectively. There-fore, we aimed to confirm that the administration of systemicdexamethasone in the setting of a long-acting local anesthetic so-lution prolongs the analgesic duration of a single-injection bra-chial plexus block compared with long-acting local anestheticalone or in combination with perineural dexamethasone. Specifi-cally, we hypothesized that the administration of intravenousdexamethasone will prolong the duration of analgesia aftersingle-injection bupivacaine for supraclavicular brachial plexusblock compared with bupivacaine alone or in combination withperineural administration of dexamethasone for ambulatory upperextremity surgery.

METHODS

Study ParticipantsThis study received Health Canada (US Food and Drug

Administration equivalent) and University Health Network Re-search Ethics Board (REB) approval. This trial was not prospec-tively registered on clinicaltrials.gov. Although it is our currentinternal practice to prospectively register all clinical trials at thetime of REB approval, the present study was granted REB ap-proval in June 2011, when trial registration had not yet becomeour uniform practice. After providing written informed consent,patients scheduled for unilateral upper extremity surgery undera supraclavicular brachial plexus block at the Toronto WesternHospital between April 2013 and June 2014 were enrolled.English-speaking American Society of Anesthesiologists (ASA)

-April 2015 125

Medicine. Unauthorized reproduction of this article is prohibited.

mailto:[email protected]

Abdallah et al Regional Anesthesia and Pain Medicine • Volume 40, Number 2, March-April 2015

classes I to III patients aged 18 to 80 years with body mass index(BMI) of 35 kg m�2 or less undergoing elective ambulatory fore-arm or hand surgery were considered eligible to participate in thisprospective, randomized, double-blind, parallel triple-arm, placebo-controlled, superiority clinical trial. Exclusion criteria includedinability to provide informed consent; significant cognitive or psy-chiatric history; pregnancy; diabetes mellitus; clavicular fracture;surgical procedure duration of 180 minutes or longer; severe respi-ratory disease; chest or shoulder deformities on the operative side;preexisting chronic pain requiring daily use of 30 mg or more ofoxycodone (or equivalent); preexisting neurological deficit or neu-ropathy in the upper extremities; allergy to local anesthetics ordexamethasone or a component of the multimodal analgesic regi-men; and contraindication(s) to peripheral nerve blocks such as lo-cal skin infection, coagulopathy, or bleeding diathesis.

Randomization and BlindingA computer-generated random numbers list was used to ran-

domize consented study participants on a 1:1:1 ratio to receive asupraclavicular block with local anesthetic (30 mL bupivacaine0.5%) alone in the Control group (Control), with local anesthetic(30 mL bupivacaine 0.5%) in conjunction with intravenous dexa-methasone 8 mg (2 mL dexamethasone 0.4%, DexamethasoneOmega; Omega Laboratory, Montreal, Quebec, Canada) in the in-travenous dexamethasone group (DexIV), or combination withperineural dexamethasone 8 mg (2 mL dexamethasone 0.4%,Dexamethasone Omega) in the perineural dexamethasone group(DexP), as illustrated in Table 1. The allocation sequence was gen-erated by a research assistant not involved in study interventionsor data collection and was concealed in sealed opaque envelopes.Envelopes were provided to the attending anesthesiologist or di-rectly supervised regional anesthesia fellow assigned to the blockprocedure room on the day of surgery, who performed all nerveblocks. The intravenous and perineural study solutions were pre-pared to appear identical by a trained anesthesia assistant accord-ing to the protocol in Table 1 and labeled only for “intravenous”or “perineural” use. The patients, anesthesiologists performingblocks, intraoperative anesthesiologists, surgeons, nurses, andresearch assistants collecting outcome data were blinded as togroup allocation.

Supraclavicular BlockBefore commencement of the nerve block, standardmonitors

were applied, intravenous access was secured and supplementaloxygen was administered at 6 to 8 Lmin−1 via face mask. Proce-dural sedation was provided intravenously by midazolam 0.03 to0.04 mg kg−1, as needed. Ultrasound-guided single-injectionsupraclavicular block was performed using a 6- to 13-MHz38-mm linear US probe (M-TURBO; SonoSite Inc, Bothell,Washington) placed in a sterile sheath and using the previously

TABLE 1. Study Groups

Control

Perineural solution 30 mL bupivacaine 0.5%+

2 mL normal saline 0.9%Intravenous solution 100 mL normal saline 0.9%

DexIV indicates intravenous dexamethasone; DexP, perineural dexamethason

126


described17,18 in-plane technique. After identifying the brachialplexus trunks and/or divisions over the first rib, lateral to the sub-clavian artery, and following skin infiltration with 1 mL of lido-caine 1%, a sterile 22-gauge blunt Stimuplex needle 50 mm (B.Braun, Bethlehem, Pa) was advanced to the junction of the firstrib and subclavian artery. After negative aspiration, the perineuralsolution was injected in 5-mL aliquots ensuring spread in the cor-ner pocket and distension of the brachial plexus sheath. Concur-rently, the corresponding intravenous study solution was infusedover 10 minutes.

Block AssessmentAfter completion of injection of the perineural solution, a

blinded research assistant evaluated sensory and motor block on-set every 5 minutes up to 30 minutes. The extent of sensory blockwas assessed in the median, radial, ulnar, and musculocutaneousnerve distributions using a 3-point score: 0 = loss of sensation tolight touch, 1 = loss of sensation to pinprick, and 2 = normalsensation. The extent of motor block was also tested in the distri-bution of the median (thumb opposition), radial (thumb abduc-tion), ulnar (thumb adduction), and musculocutaneous (flexionof the elbow in supination and pronation) nerves using a 3-pointscale, where 0 = nomovement, 1 = paresis, 2 = normal movement.Block successwas defined as the achievement of sensory and mo-tor scores of 1 or less in all 4 nerve distributions within 30minutesof injection of the perineural solution. At the discretion of the at-tending block room anesthesiologist, a supplemental terminalbranch nerve block (5 mL of lidocaine 2% per nerve, as needed)at the elbow or forearm was administered to patients with a suc-cessful block in whom surgical anesthesia (defined as sensoryand motor scores = 0 in the distribution of surgery) was notachieved by the time the operating theater was ready to receivethem. Patients who received a supplemental terminal branch nerveblock were included in all analyses based on the principle of inten-tion to treat. Patients who did not meet criteria for block success at30 minutes were excluded from the data analyses.

Intraoperative CareAn anesthesiologist blinded to group allocation administered

conscious sedation intraoperatively, titrated to patient comfortusing low intravenous doses of midazolam (1–3 mg), fentanyl(1–2 μg kg−1), and/or propofol (25–75 μg kg−1 min−1). In the eventof inadequate analgesia during surgery, surgical site infiltrationwith lidocaine 2% by the surgeon was followed by conversion togeneral anesthesia as necessary. These patients were included inthe analyses based on the principle of intention to treat.

Postoperative CarePatients were transferred to the postanesthetic care unit

(PACU) after surgery. Moderate or severe pain (visual analogue

DexIV DexP

30 mL bupivacaine 0.5% 30 mL bupivacaine 0.5%+ +

2 mL normal saline 0.9% 2 mL dexamethasone 0.4%98 mL normal saline 0.9% 100 mL normal saline 0.9%

+2 mL dexamethasone 0.4%

e.

© 2015 American Society of Regional Anesthesia and Pain Medicine


Regional Anesthesia and Pain Medicine • Volume 40, Number 2, March-April 2015 Intravenous or Perineural Dexamethasone

scale [VAS] ≥4) or patient request for additional analgesics inPACU was treated with fentanyl in 25-μg intravenous incrementsevery 5 minutes, as needed, up to 200 μg h−1 administered byblinded PACU nurses. Once oral intake was permitted, patientsrequiring additional analgesics received acetaminophen (1 g),followed by oxycodone (5 mg), as needed. Nausea and vomitingwere treated with dimenhydrinate 25 mg intravenously and/orondansetron 4 mg intravenously, as required. Patients were dis-charged home when they met institutional discharge criteria.Discharged patients received a prescription for oral Tylenol No. 3(acetaminophen 300 mg/codeine 30 mg) or Percocet (acetamino-phen 325 mg/oxycodone HCl 5 mg) if intolerant to codeine, asneeded. On discharge from the hospital, patients were providedwith and instructed to complete a standardized postoperative di-ary19 at home and return to the investigators using a stampedreturn-addressed envelope. All patients received a telephone callby a blinded research assistant on postoperative day 1 to remindthem to complete and return their diary, as well as a telephone callat 2 weeks postoperatively to inquire about block-related compli-cations. Any block-related complications identified were followeduntil resolution.

Outcome MeasuresDuration of analgesia, defined as time in hours to the first re-

port of postoperative pain at the surgical site, was designated asthe primary outcome. Secondary outcomes included (1) durationof motor block, defined as time (in hours) to return to normal (orbaseline) motor strength in the operative limb; (2) severity of painat the surgical site (VAS) at 8 and 24 hours, as well as at 7 and14 days; (3) cumulative intraoperative opioid consumption (con-verted to milligrams of intravenous morphine equivalent)20;(4) cumulative postoperative opioid consumption (converted tomilligrams of oral morphine equivalent)20 at 24 hours; (5) inci-dence of postoperative nausea and vomiting during the first24 hours postoperatively; (6) patient satisfaction with pain relief(expressed as VAS) at 24 hours; and (7) occurrence of any block-related complications, including any new paresthesia (numb-ness or tingling) or weakness in the operative limb at 2 weekspostoperatively.

Statistical AnalysisOur hypothesis was that intravenous dexamethasone (DexIV

group) prolongs the duration of analgesia compared with conven-tional long-acting local anesthetic alone (Control group) and thatintravenous dexamethasone (DexIV group) prolongs the durationof analgesia compared with perineural dexamethasone (DexPgroup) after supraclavicular brachial plexus block.We used the se-quential testing approach as a serial gatekeeping procedure for thishypothesis testing.21 As such, the duration of analgesia observedin the DexIV group was first tested for superiority over the Controlgroup and then tested for superiority over the DexP group.

The mean duration of sensory block associated withultrasound-guided supraclavicular block performed with 30 mLbupivacaine 0.5% is reported to be 13.7 ± 1.5 hours.22 In the ab-sence of any published data available at the outset of the presenttrial, it had been our anecdotal experience that addition of in-travenous dexamethasone to PNB can prolong the duration of sen-sory block by as much as 2-fold. This corresponds to a size effectequivalent to 0.98.

Assuming that intravenous dexamethasone increases the du-ration of analgesia after supraclavicular block by a size effect of0.98 compared with long-acting local anesthetic alone, we calcu-lated that 66 patients (22 patients per group) would be required todetect a statistically significant difference between groups with



α = 0.05 and 80% power. Allowing a 10% loss caused by patientdropout or incomplete follow-up, we planned to recruit 25 patientsper group or a total of 75 patients for the 3 groups. Based on pre-viously published data,23 we also estimated that this sample sizewould provide 80% power to detect differences as small as 15%in the duration of analgesia between the DexIV and DexP groups.

We used the SPSS Statistics for Windows (Version 22; IBM,Armonk, NY) to perform the analysis. Our calculations were per-formed under the assumptions that the 3 study groups were inde-pendent, that the parent population from which the data werederived is normally distributed, and that within-group variancesare equal. The normality of data distribution was tested usingthe Shapiro-Wilk test. Except for the exclusion of patients withan unsuccessful block at 30 minutes, all analyses used anintention-to-treat approach, with patients evaluated according tothe groups to which they were initially randomized, regardless ofthe actual treatment they received during the study.

Continuous data are presented as mean (SD) or mean (95%confidence interval [95% CI]), whereas categorical data are pre-sented as numbers or percentages. We used a 1-way analysis ofvariance combinedwith post hoc testing using the T test to analyzecontinuous data. We used a contingency table with either χ2 orFisher exact test, as appropriate, combined with post hoc testingusing the Mann-Whitney U test for categorical data. We usedthe Kruskal-Wallis test combined with post hoc testing using theMann-Whitney-Wilcoxon U test for ordinal data. For time-to-event outcomes, we analyzed data using the Kaplan-Meier sur-vival method and compared groups using the log-rank test, withadjustment for multiple comparisons.

The 2-tailed P value threshold of statistical significancefor the 1-way analysis of variance comparison among groupsas well as for the log-rank test was calculated using the Bonferronicorrection, and the P value was set at 0.017. The P values forrepeated measurements of the secondary end points were ad-justed using the Bonferroni-Holm24 step-down correction forrepeated comparisons.

RESULTSA total of 185 patients were approached to participate in this

study; and 75 patients were randomized and completed the studyprotocol (Control group, n = 25; DexIV group, n = 25; and DexPgroup, n = 25). Figure 1 represents the Consolidated Standardsof Reporting Trials (CONSORT) flow diagram depicting patientprogress through the study phases. Block success was docu-mented in all patients; however, 14 patients (Control, 5; DexIV,2; DexP, 7) required a supplemental terminal branch nerve blockto achieve surgical anesthesia in time for operating theater readi-ness. Intraoperatively, none of the study patients required local an-esthetic supplementation by the surgeon or conversion to generalanesthesia. Sensory and motor blockade resolved in all patientswithin the study period, and none of the patients were censoredfrom the survival analysis. The DexIV group included less ASAI patients, but all other characteristics were similar between thestudy groups (Table 2).

The duration of analgesia (mean [95% CI]) was significantlyprolonged in both the DexIV (25 hours [17.6–23.6]) and DexP(25 hours [19.5–27.4]) groups compared with the Control group(13.2 hours [11.5–15.0]; P < 0.001). There was no detectable dif-ference in the duration of analgesia between the DexIV and DexPgroups (P = 1.0). The Kaplan-Meier survival analysis of the dura-tion of analgesia for the 3 study groups is shown in Figure 2. Thelog-rank test also suggested prolongation of the duration of anal-gesia for both the DexIV and DexP groups compared with the

127


FIGURE 1. Consolidated Standards of Reporting Trials (CONSORT) flow diagram showing patient progress through the study phases.


Control (P < 0.001) and no detectable difference between theDexIV and DexP groups (P =1.0).

The duration of motor block (mean [95% CI]) was signifi-cantly prolonged in both the DexIV (30.1 hours [22.0–24.2]) andDexP (25.5 hours [22.2–26.0]) groups compared with the Control(19.7 hours [13.5–23.0]; P < 0.001). In addition, the DexIV groupexperienced motor block duration that was significantly longerby (4.6 hours [3.7–5.5]; P < 0.00001) compared with the DexPgroup. The Kaplan-Meier survival analysis of the duration of mo-tor block for the 3 study groups is shown in Figure 3. The log-ranktest also suggested prolongation of the duration of motor blockfor both DexIV and DexP groups compared with the Control(P < 0.001), as well as for the DexIV group compared with theDexP group (P = 0.01).

TABLE 2. Patient Characteristics

Control (n = 25) DexIV (n = 25) DexP (n = 25)

Age, y 46.1 (14.3) 46.1 (15.9) 51.8 (15.3)Sex (male/female) 10/15 15/10 13/12ASA (I/II/III) 14/10/1 14/9/2 6/16/3Weight, kg 74.6 (18.2) 75.1 (12.5) 75.2 (18.0)Height, cm 167.1 (9.4) 170.2 (17.8) 168.4 (9.5)BMI, kg m−2 26.5 (4.9) 27.2 (11.8) 26.4 (5.5)Surgical siteRight/left 14/11 12/13 9/16Elbow 1 1 2Forearm 2 4 1Wrist 5 5 4Hand 17 15 18

Values are expressed as the mean (SD) or absolute numbers.

ASA indicates American Society of Anesthesiologists; BMI, bodymass index.

128


Patients in the DexIV and DexP groups reported less pain at24 hours postoperatively compared with those in the Control(P < 0.0001); however, no differences in pain were detected be-tween the DexIV and DexP groups at 24 hours (Fig. 4). The 3groups had similar pain scores at all the other time points.

Postoperative cumulative opioid consumption was reducedin the DexIV and DexP groups at 24 hours postoperatively com-pared with that in the Control (Table 3). The 3 groups experi-enced a similar incidence of postoperative nausea and vomiting(Table 3); however, patients in the DexIVand DexP groups reportedhigher satisfaction with pain relief compared with the Controlgroup. Finally, none of the patients reported any block-relatedcomplications.

DISCUSSIONThis study confirms that addition of intravenous dexametha-

sone is an effective alternative to perineural dexamethasone toprolong the duration of single-injection supraclavicular brachialplexus block compared with long-acting local anesthetic alone.

The analgesic effects of dexamethasone as a local anestheticadjunct have long been investigated; however, the actual site ofaction and potential systemic effect has only recently been ad-dressed. In a trial of patients undergoing shoulder surgery, Desmetand colleagues15 recently demonstrated that the analgesic durationof a single-injection interscalene block with ropivacaine 0.5% issimilarly prolonged with the addition of 10 mg dexamethasoneadministered either perineurally or intravenously. Curiously, how-ever, the evaluation of interscalene block success in Desmet’sstudy was performed postoperatively and was dependent on handstrength rather than any evidence of sensory blockade in theshoulder. Moreover, instead of presenting actual pain scoresthroughout the study period, these authors only present the propor-tion of patients who reported verbal rating pain scores of 2 or lesscompared with higher than 2. The difficulty in interpreting theclinical significance of such data is self-evident because painscores of 3 and 10 were treated as equivalent in this type of



FIGURE 2. Kaplan-Meier survival plot representing duration of analgesia in the 3 study groups. Dex indicates dexamethasone.


analysis. Next, Fredrickson and colleagues16 compared the addi-tion of 8 or 10 mg of dexamethasone administered either perineu-rally or intramuscularly with 8 mg of intravenous dexamethasonein a mixed surgical population receiving single-injection sciaticnerve blocks or ankle blocks. Although Fredrickson concludedthat intramuscular dexamethasone confers similar analgesic ben-efits to perineural application, the lack of statistical power and

FIGURE 3. Kaplan-Meier survival plot representing duration of motor bl



absence of a negative control group further complicate the in-terpretation of these results. In contrast, Rahangdale and col-leagues14 recently demonstrated that addition of 8 mg ofperineural or intravenous dexamethasone to a single-injection sci-atic nerve block with 0.45 mL kg−1 bupivacaine 0.5% and epi-nephrine 1:300,000 did not significantly improve the quality ofrecovery or opioid consumption compared with local anesthetic

ock in the 3 study groups. Dex indicates dexamethasone.

129


FIGURE 4. Box plots of postoperative visual analogue scale (VAS) pain scores in the 3 study groups during the first 2 weeks. *A statisticallysignificant difference between the 3 study groups (F test).


alone. Unfortunately, however, the relatively large doses of bu-pivacaine and the influence of epinephrine may have at leastpartially masked the true effects of the dexamethasone. Last,Kawanishi and colleagues13 compared the perineural and intrave-nous routes for 4 mg of dexamethasone in the setting of single-injection interscalene block with 20 mL ropivacaine 0.75%.Although these authors reported that only the perineural routecan prolong the analgesic duration of interscalene block, thisstudy lacked adequate statistical power to definitively differentiateoutcomes between the 2 experimental groups.

The prolongation of analgesia observed herein with intrave-nous dexamethasone was also accompanied by a prolongation inmotor block duration when compared with perineural dexametha-sone. Although previous trials have reported that intravenousdexamethasone and perineural dexamethasone similarly prolongthe duration of motor block after PNB,14,15 the present study isthe first to demonstrate that intravenous dexamethasone extendsmotor block duration to a greater degree than the perineural appli-cation. The clinical importance of this finding is questionable be-cause it may preclude ambulation and delay discharge but mayprove valuable for select orthopedic inpatients who require contin-uous passive motion postoperatively.

The mechanism and site of action by which dexamethasoneexerts its analgesic and anesthetic effects remain speculative andhave been discussed in detail elsewhere.9,25–27 In brief, locallyapplied perineural dexamethasone is thought to suppress the ex-citability of nociceptive C fibres10 via glucocorticoid receptorchanges and changes to ion channel function.28,29 The local anes-thetic effect may also be prolonged by dexamethasone-inducedvasoconstriction.13 In addition, there is likely a local suppressionof inflammatory mediators and subsequent and ectopic neuraldischarges in pain fibres.30 Similarly, the analgesic effects of

TABLE 3. Results

Outcome

Cumulative intraoperative intravenous morphine equivalent consumption,Cumulative postoperative oral morphine equivalent consumption at 24 h,Incidence of postoperative nausea and vomiting during first 24 hPatient satisfaction with pain relief at 24 h on an VAS scale

*P value for the overall F test is set at 0.05.

Values are expressed as the mean (95% confidence interval) or absolute num

DexIV indicates intravenous dexamethasone; DexP, perineural dexamethason

130


systemically administered dexamethasone likely arises from avariety of mechanisms, including peripheral and central anti-inflammatory effects by binding to cellular receptors, modifyinggene transcription and protein synthesis, and ultimately inhibitingproduction of prostaglandins, leukotrienes, and proinflammatorycytokines.31–34 Dexamethasone is also believed to suppress theneuropeptide immune response in injured tissue, thus lesseningthe extent of pain.35

There are a number of limitations associated with the presentstudy. First, a number of patients required distal supplementationwith lidocaine 2% to hasten block onset because of operatingroom readiness and patient flow efficiencies. Although it is possi-ble that such supplementation could have influenced early postop-erative outcomes, it is improbable that our primary outcomemeasure was affected. On designing the present study, we wereaware of the potential for delayed block onset (relative to our insti-tutional standard of 1:1 lidocaine 2%:bupivacaine 0.5% admixturefor surgical anesthesia); however, we believed that it was impor-tant to evaluate the effect of dexamethasone in the setting of con-ventional long-acting local anesthetic alone. Furthermore, theminimum therapeutic dose of intravenous dexamethasone capableof prolonging the duration of analgesia after PNB has not beenelucidated. It is therefore possible that the dose of dexamethasoneused herein may have been sufficiently large so as tomask the truesite of action (ie, local or central). Nonetheless, the current state ofevidence supports a dose of 8 mg for the perineural application ofdexamethasone.4,8,9,36 Also, although our trial is adequatelypowered to detect differences as small as 15% that may exist be-tween the DexIV and DexP groups, a sample size of at least 240patients (ie, 80 per group) would be required to detect very small(ie, <10%) differences in the duration of analgesia between theDexIV and DexP groups. Our results suggest that intravenous

Control (n = 25) DexIV (n = 25) DexP (n = 25) P*

mg 3.3 (1.2–5.4) 3.9 (1.5–6.2) 4.3 (2.0–6.6) 0.3mg 22.1 (7.6–36.6) 12.5 (2.4–22.6) 13.3 (1.1–25.5) 0.013

4/25 1/25 1/25 0.357.2 (4.6–9.8) 9.0 (8.0–10.0) 9.0 (7.9–10.0) <0.001

bers.

e; VAS, visual analogue scale.




dexamethasone and perineural dexamethasone possess very simi-lar treatment effects in the present context; although investigatingsuperiority is likely futile, seeking to demonstrate noninferiorityor even equivalence may be a worthwhile future scholarly en-deavor. Finally, the present study was not powered to detectcomplications of perineural dexamethasone. Bench studies inves-tigating the safety of perineural dexamethasone suggest a dose-dependent adverse effect on peripheral nerves,35 although a morerecent study by Williams and colleagues suggests that the neuro-toxicity of a dexamethasone-ropivacaine combination is similarto that of ropivacaine alone.8 An additional concern includes thepossibility of a reduction in blood flow associated with the peri-neural application of dexamethasone; this may have important im-plications if perineural dexamethasone is coadministrated withepinephrine and/or used in diabetic patients.31 Many formulationsof dexamethasone also contain additives. The dexamethasone so-dium phosphate solution used in this study contained benzyl alco-hol 1%; although this preservative had once been implicated intransient paraparesis after epidural injection,37 there is little evi-dence of harm associated with its perineural use,38 especially insuch a diluted concentration as that used herein. Importantly, pre-vious reports of nerve injury associated with perineural dexameth-asone have been attributed to direct needle trauma rather thanthe drug itself.39

CONCLUSIONSWhen added to a single-injection supraclavicular brachial

plexus block with long-acting local anesthetic, the effectivenessof intravenous dexamethasone in prolonging the duration of an-algesia seems similar to perineural dexamethasone. Comparableeffectiveness and theoretical safety concerns support the use ofintravenous dexamethasone over perineural dexamethasone insettings where prolonged analgesia after single-injection PNBis desired.

REFERENCES1. Abdallah FW, Brull R. The facilitatory effects of perineural

dexmedetomidine on neuraxial and peripheral nerve blockade: a systematicreview and meta-analysis. Br J Anaesth. 2013;110:915–925.

2. Candido KD. Buprenorphine added to the local anesthetic for axillarybrachial plexus block prolongs postoperative analgesia. Reg Anesth PainMed. 2002;27:162–167.

3. McCartney CJ, Duggan E, Apatu E. Should we add clonidine to localanesthetic for peripheral nerve blockade?A qualitative systematic reviewofthe literature. Reg Anesth Pain Med. 2007;32:330–338.

4. Choi S, Rodseth R, McCartney CJ. Effects of dexamethasone as a localanaesthetic adjuvant for brachial plexus block: a systematic reviewand meta-analysis of randomized trials. Br J Anaesth.2014;112:427–439.

5. Laiq N, Khan MN, Arif M, Khan S. Midazolam with bupivacaine forimproving analgesia quality in brachial plexus block for upper limbsurgeries. J Coll Physicians Surg Pak. 2008;18:674–678.

6. Lee AR, Yi H, Chung IS, et al. Magnesium added to bupivacaine prolongsthe duration of analgesia after interscalene nerve block. Can J Anaesth.2012;59:21–27.

7. Candido KD, Knezevic NN. All adjuvants to local anesthetics were notcreated equal: animal data evaluating neurotoxicity, thermal hyperalgesia,and relevance to human application. Reg Anesth Pain Med. 2011;36:211–212.

8. Williams BA, Hough KA, Tsui BY, Ibinson JW, Gold MS, Gebhart GF.Neurotoxicity of adjuvants used in perineural anesthesia andanalgesia in comparison with ropivacaine. Reg Anesth Pain Med.2011;36:225–230.



9. Noss CD, MacKenzie LD, Kostash MA. Adjuvant dexamethasone:Innovation, farce, or folly? Reg Anesth Pain Med. 2014;39:540–545.

10. Johannsson A, Hao J, Sjolund B. Local corticosteroid application blockstransmission in normal nociceptive c-fibers. Acta Anaesthesiol Scand.1990;34:335–338.

11. Abdi S, Datta S, Trescot AM. Epidural steroids in the management ofchronic spinal pain: a systematic review. Pain Physician. 2007;10:185–212.

12. De Oliveira GS Jr, Almeida MD, Benzon HT, McCarthy RJ. Perioperativesingle dose systemic dexamethasone for postoperative pain: ameta-analysis of randomized controlled trials. Anesthesiology. 2011;115:575–588.

13. Kawanishi R, Yamamoto K, Tobetto Y, et al. Perineural but not systemiclow-dose dexamethasone prolongs the duration of interscalene blockwith ropivacaine: A prospective randomized trial. Local Reg Anesth.2014;7:5–9.

14. Rahangdale R, Kendall MC, McCarthy RJ, et al. The effects of perineuralversus intravenous dexamethasone on sciatic nerve blockade outcomes: arandomized, double-blind, placebo-controlled study. Anesth Analg. 2014;118:1113–1119.

15. Desmet M, Braems H, Reynvoet M, et al. I.V. and perineuraldexamethasone are equivalent in increasing the analgesic duration of asingle-shot interscalene block with ropivacaine for shoulder surgery:a prospective, randomized, placebo-controlled study. Br J Anaesth.2013;111:445–452.

16. Fredrickson MJ, Danesh-Clough TK, White R. Adjuvant dexamethasonefor bupivacaine sciatic and ankle blocks: results from 2 randomizedplacebo-controlled trials. Reg Anesth Pain Med. 2013;38:300–307.

17. Soares LG, Brull R, Lai J, Chan VW. Eight ball, corner pocket: the optimalneedle position for ultrasound‐guided supraclavicular block.Reg Anesth Pain Med. 2007;32:94–95.

18. Perlas A, Lobo G, Lo N, Brull R, Chan VW, Karkhanis R.Ultrasound-guided supraclavicular block: outcome of 510 consecutivecases. Reg Anesth Pain Med. 2009;34:171–176.

19. Parrington SJ, O’Donnell D, Chan VW, et al. Dexamethasone added tomepivacaine prolongs the duration of analgesia after supraclavicularbrachial plexus blockade. Reg Anesth Pain Med. 2010;35:422–426.

20. Canadian Pharmacists Association. Compendium of Pharmaceuticals andSpecialties: The Canadian Drug Reference for Health Professionals.45th ed. Ottawa, Ontario, Canada: Canadian PharmacistsAssociation, 2010.

21. Dmitrienko A, Tamhane AC. Gatekeeping procedures with clinical trialapplications. Pharm Stat. 2007;6:171–180.

22. Kapral S, Krafft P, Eibenberger K, Fitzgerald R, Gosch M, Weinstabl C.Ultrasound-guided supraclavicular approach for regional anesthesia of thebrachial plexus. Anesth Analg. 1994;78:507–513.

23. Shrestha BR, Maharjan SK, Shrestha S, et al. Comparative study betweentramadol and dexamethasone as an admixture to bupivacaine insupraclavicular brachial plexus block. JNMA J Nepal Med Assoc. 2007;46:158–164.

24. Bixler EO, Vgontzas AN, Ten Have T, Tyson K, Kales A. Effects of age onsleep apnea in men: I. Prevalence and severity. Am J Respir Crit Care Med.1998;157:144–148.

25. Yilmaz-Rastoder E, Gold MS, Hough KA, Gebhart GF, Williams BA.Effect of adjuvant drugs on the action of local anesthetics in isolated ratsciatic nerves. Reg Anesth Pain Med. 2012;37:403–409.

26. Wang PH, Tsai CL, Lee JS, Wu KC, Cheng KI, Jou IM. Effects of topicalcorticosteroids on the sciatic nerve: an experimental study to adduce thesafety in treating carpal tunnel syndrome. J Hand Surg Eur Vol.2011;36:236–243.

131



27. Shishido H, Kikuchi S, Heckman H,Myers RR. Dexamethasone decreasesblood flow in normal nerves and dorsal root ganglia. Spine(Phila Pa 1976). 2002;27:581–586.

28. Attardi B, Takimoto K, Gealy R, Severns C, Levitan ES. Glucocorticoidinduced up-regulation of a pituitary K+ channel mRNA in vitro andin vivo. Receptors Channels. 1993;1:287–293.

29. Kopacz DJ, Lacouture PG, Wu D, Nandy P, Swanton R, Landau C. Thedose response and effects of dexamethasone on bupivacaine microcapsulesfor intercostal blockade (T9 to T11) in healthy volunteers.Anesth Analg. 2003;96:576–582.

30. Devor M, Govrin-Lippmann R, Raber P. Corticosteroids suppress ectopicneural discharge originating in experimental neuromas. Pain.1985;22:127–137.

31. Gilron I. Corticosteroids in postoperative painmanagement: future researchdirections for a multifaceted therapy. Acta Anaesthesiol Scand.2004;48:1221–1222.

32. Holte K, Kehlet H. Perioperative single-dose glucocorticoid administration:pathophysiologic effects and clinical implications. J Am Coll Sur.2002;195:694–712.

33. Iskandar H, Benard A, Ruel-Raymond J, Cochard G, Manaud B. Femoralblock provides superior analgesia compared with intra-articular

132


ropivacaine after anterior cruciate ligament reconstruction.Reg Anesth Pain Med. 2003;28:29–32.

34. Sapolsky RM, Romero LM, Munck AU. How do glucocorticoids influencestress responses? Integrating permissive, suppressive, stimulatory, andpreparative actions 1. Endocr Rev. 2000;21:55–89.

35. Williams BA, Schott NJ, Mangione MP, Ibinson JW. Perineuraldexamethasone and multimodal perineural analgesia: how much is toomuch? Anesth Analg. 2014;118:912–914.

36. Ma R, Wang X, Lu C, et al. Dexamethasone attenuatedbupivacaine-induced neuron injury in vitro through a threonine-serineprotein kinase B–dependent mechanism. Neuroscience. 2010;167:329–342.

37. Craig DB, Habib GG. Flaccid paraparesis following obstetrical epiduralanesthesia: possible role of benzyl alcohol. Anesth Analg. 1977;56:219–221.

38. Rowlingson JC, Neal JM. Adjuvant and preservavtive toxicity. In:Neal JM, Rathmel JP. Complications in Regional Anesthesia and PainMedicine. 2nd ed. Philadelphia, PA: Lippincott WilliamsWilkins, 2007.

39. Stahl S, Rozen N, Michaelson M. Ulnar nerve injury following midshaftforearm fractures in children. J Hand Surg Br. 1997;22:788–789.



Documents

Intravenous Dexamethasone and Perineural Dexamethasone