Full text of DOI:10.1007/BF03020416

32
REGIONAL ANESTHESIA AND PAIN
A systematic review of adjuncts for intravenous
regional anesthesia for surgical procedures
[Étude méthodique des traitements d’appoint à l’anesthésie régionale
intraveineuse pendant les interventions chirurgicales]
Andrew Choyce MBCHB FRCA,* Philip Peng MBBS FRCPC†
Purpose: To review the use of adjuncts to intravenous regional
anesthesia (IVRA) for surgical procedures in terms of their intraop-
erative effects (efficacy of block and tourniquet pain) and postoper-
ative analgesia.
Objectif : Passer en revue l’usage des traitements d’appoint à
l’anesthésie régionale intraveineuse (ARIV) pour les interventions
chirurgicales, en termes des effets peropératoires (efficacité du bloc et
douleur du garrot et de l’analgésie postopératoire.
Source: A systematic search (Medline, Embase, reference lists) for
randomized, controlled and double-blinded studies using adjuncts
to IVRA for surgical procedures was conducted. Data were collect -
ed on intraoperative effects (onset/offset and quality of block and
tourniquet pain), postoperative effects (pain intensity and analgesic
consumption) and side effects recorded. Statistical significance as
indicated in the original report and likely clinical relevance were
taken into account to arrive at a judgment of overall benefit.
Source : Une recherche systématique (bases de données Medline,
Embase, listes de références) des études randomisées, contrôlées et à
double insu utilisant les traitements d’appoint à l’ARIV pendant les
interventions chirurgicales a été réalisée. Les données concernant les
effets peropératoires ont été recueillies (début/fin et qualité du bloc et
douleur de garrot) ainsi que les effets postopératoires (intensité de la
douleur et consommation analgésique) et les effets secondaires. La
signification statistique, tel qu’indiqué dans l’article original, et la per-
tinence clinique possible ont été retenues afin d’en arriver à un juge -
ment sur les bienfaits généraux.
Principal findings: Twenty-nine studies met all inclusion criteria.
Data on 1,217 study subjects are included. Adjuncts used were opi-
oids (fentanyl, meperidine, morphine, sufentanil), tramadol, non-
steroidal anti-inflammatory drugs (NSAIDs; ketorolac, tenoxicam,
acetyl-salicylate), clonidine, muscle relaxants (atracurium, pancuro-
nium, mivacurium), alkalinization with sodium bicarbonate, potassi-
um and temperature. There is good evidence to recommend
NSAIDs in general and ketorolac in particular, for improving post-
operative analgesia. Clonidine 1 µg·kg–1 also appears to improve
postoperative analgesia and prolong tourniquet tolerance. Opioids
are poor by this route; only meperidine 30 mg or more has sub-
stantial postoperative benefit but at the expense of postdeflation
nausea, vomiting and dizziness. Muscle relaxants improve intraop-
erative motor block and aid fracture reduction.
Constatations principales : Vingt-neuf études répondaient à tous
les critères d’inclusion. Les données sur 1 217 sujets d’étude ont été
compilées. Les traitements d’appoint étaient des opioïdes (fentanyl,
mépéridine, morphine, sufentanil), le tramadol, les anti-inflamma-
toires non stéroïdiens (AINS; kétorolac, ténoxicam, acétylsalicylique),
la clonidine, les myorelaxants (atracurium, pancuronium, mivacuri-
um), l’alcalinisation avec le bicarbonate de sodium, le potassium et la
température. Il existe de bonnes indications pour l’utilisation des AINS
en général et du kétorolac en particulier concernant l’analgésie
postopératoire. L’administration de 1µg·kg–1 de clonidine semble
améliorer l’analgésie postopératoire et la tolérance prolongée du gar-
rot. Les opioïdes font piètre figure dans ces circonstances; seule une
dose de 30 mg ou plus de mépéridine présente des bienfaits
postopératoires substantiels, mais aux dépens de nausées, de vo-
missements et d’étourdissement au relâchement du garrot. Les
myorelaxants améliorent le bloc moteur peropératoire et favorisent la
réduction de fracture.
Conclusion: Using NSAIDs or clonidine as adjuncts to IVRA
improves postoperative analgesia and muscle relaxant improves
motor block.
Conclusion : L’utilisation de AINS ou de clonidine comme traitement
d’appoint de l’ARIV améliore l’analgésie postopératoire tandis que les
myorelaxants améliore le bloc moteur.
From the Departments of Anesthesia, King’s College Hospital,* Denmark Hill, London, UK; and the Toronto Western Hospital,†
University Health Network, University of Toronto, Toronto, Ontario, Canada.
Address correspondence to: Dr. Philip Peng, Department of Anesthesia, Toronto Western Hospital, University Health Network,
University of Toronto, 399 Bathurst Street, Toronto, Ontario M5T 2S8, Canada. Phone: 416-603-5118; Fax: 416-603-6494;
E-mail: philip.peng@uhn.on.ca
Accepted for publication July 4, 2001.
Revision accepted September 24, 2001.
CAN J ANESTH 2002 / 49: 1 / pp 32–45

Choyce et al.: ADJUNCTS FOR IVRA
33
NTRAVENOUS regional anesthesia (IVRA)
was first described in 1908 for anesthesia of the
hand and forearm.1 The earliest agent injected
as either “none”, “not analyzed” (where no statistical
analysis was performed) or the actual side effects if sta-
tistically different between groups. Study outcomes
were classed as “supportive” if they showed significant
benefit in either the intraoperative or postoperative vari-
ables examined, or “negative” if they did not.
Both authors using the above criteria reviewed each
study independently. The clinical relevance of any dif-
ferences is considered in the discussion.
I
into the isolated vascular space was procaine.
The technique regained popularity in the 1960’s when
Holmes used lidocaine.2 Lidocaine remains the stan-
dard local anesthetic (LA) agent for surgical proce-
dures in North America3 and prilocaine is used widely
in Europe.4
IVRA is simple to administer, reliable and cost-effec-
tive.5 It is ideal for short operative procedures on the
extremities performed on an ambulatory basis.
Disadvantages include concerns about LA toxicity, slow
onset, poor muscle relaxation, tourniquet pain and
minimal postoperative pain relief. The ideal IVRA solu-
tion should have the following features: rapid onset,
reduced dose of LA, reduced tourniquet pain and pro-
longed postdeflation analgesia. At present, this may
only be achieved by the addition of adjuncts to LA.
The purpose of this article is to review the available
literature and provide the operator with a guide to the
likely benefits of adjuncts added to LA for IVRA for
surgical procedures, i.e., improved block efficacy,
decreased tourniquet pain or prolonged duration of
postdeflation analgesia.
Results
A total of 29 studies, all involving adults, met the crite-
ria for inclusion. Two of these studies looked at the use
of two adjuncts in combination with LA,6, 7 and one
study compared four separate adjuncts to control.8 Ten
studies investigated opioids6–15 and one tramadol1 6
(Table I). Six studies investigated non-steroidal anti-
inflammatory drugs (NSAIDs)8,17–21 and five cloni-
dine8,22–25 (Table II). Five studies investigated muscle
relaxants,6,7,26–28 three alkalization,29–31 one potassium
3 2
and two temperature33,34 (Table III). None of the stud-
ies reported sample size estimation and power analysis.
Opioids
Ten studies involving 412 patients have investigated
opioids either as sole adjunct8–15 or in combination
with a muscle relaxant6,7 (Table I). One of the combi-
nation studies also included fentanyl as a sole adjunct.7
The opioids investigated were fentanyl6,7,9–11 (five
studies), meperidine13,15 (two studies), morphine12,14
(two studies) and sufentanil8 (one study). No study
included a systemic control group. Six out of ten stud-
ies were supportive of the addition of the
adjunct(s)6–8,13–15 (two with fentanyl plus pancuroni-
um,6,7 two with meperidine,13,15 one with morphine1 4
and one with sufentanil).8
Methods
Studies were identified in a search of Medline
(between July 1966 and January 2001) and Embase
(from 1980) by using the following Medical Subject
Headings (MeSH) terms: IVRA; intravenous regional
analgesia, anesthesia or anaesthesia; Bier’s block. The
reference sections of eligible and related review articles
were then examined for further relevant publications
that might have been missed by the computer search.
Articles that were unpublished, abstracts, letters,
and non-peer-reviewed or non-English language were
excluded from the final systematic review. Of the stud-
ies identified, only those that were of a prospective,
randomized, controlled and double-blinded design
were included. Individual authors were not contacted
for further information.
For each study, the concentration and volume of LA
and type and dose of adjunct were recorded. Additional
data included information on study design, numbers
recruited, assessments made, side effects and outcomes.
Data are presented in tabulated form. Each type of
adjunct is examined for its potential intraoperative ben-
efit (speed of onset and recovery of sensory and motor
block, effect on tourniquet pain), postoperative anal-
gesic benefit (visual analogue score (VAS), analgesic
consumption (AC), time to first analgesic (TTFA) and
side effect profile. The incidence of side effects is given
FENTANYL
Four studies on a total 157 subjects (including two
volunteer studies), investigated fentanyl as sole
adjunct.6,9–11 The dose range was 50–200 µg.
No study demonstrated any intraoperative advan-
tage in terms of onset (motor or sensory). Tourniquet
pain was not investigated and none of the studies
looked at postoperative analgesia. Two of the studies
reported significant nausea after the deflation of the
tourniquet cuff in the fentanyl-treated groups.9,11
Two further studies looked at the combination of
fentanyl plus pancuronium.6, 7 Abdulla et al.6 found
that the combination of fentanyl 50 µg plus pancuro-
nium 0.5 mg to a dilute solution of LA (0.25% lido-
caine) provided good or excellent intraoperative
analgesia in 100% of cases. No comparison of onset of

34
CANADIAN JOURNAL OF ANESTHESIA
anesthesia or postoperative analgesia was made. Sztark
et al.7 compared 0.25% lidocaine plus fentanyl 1
µg·kg–1 and pancuronium 0.5 mg with conventional
LA dose (0.5% lidocaine, 0.6 mL·kg–1). The plain
lidocaine group had complete sensory block four min-
utes earlier but the overall success rate was similar.
demonstrated. Light-headedness after tourniquet
deflation was reported in 8/15 but this was not ana-
lyzed statistically.
TRAMADOL
One study looked at the intraoperative effects of
adding tramadol 100 mg to LA (Table I).1 6The resul-
tant sensory (pinprick, touch and temperature) block
was faster in onset compared to plain LA. However,
only touch sensation was slower to recover compared
to plain LA. Onset and recovery of motor block was
not affected. Skin rash below the tourniquet that dis-
appeared within one hour of deflation was the only
significant side effect when tramadol was added to LA.
Possible benefits in terms of tourniquet pain and post-
operative course were not investigated.
MEPERIDINE
Two studies (one involving volunteers) on a total 80
subjects, investigated meperidine as sole adjunct.13,15
The dose range was 10–100 mg.
Armstrong et al.1 3 looked at the intraoperative effect
of adding meperidine 100 mg to a dilute solution of
LA. The resultant sensory and motor block was faster in
onset and slower to recover. Furthermore, forearm pain
at 20 min inflation and tourniquet pain at ten minutes
(but not at 20 min) inflation were significantly less in
the meperidine group than in the control group.
Recovery in several of the meperidine-containing
groups was complicated by light-headedness or nausea
despite giving naloxone at deflation. The authors con-
cluded that postdeflation complications precluded the
use of meperidine in routine IVRA.
NSAIDS
Six studies involving 370 patients have investigated var-
ious NSAIDs as the sole adjunct8,17–21 to IVRA (Table
II). The agents investigated were ketorolac,17,19,21
tenoxicam8,18 and aspirin.2 0One study compared IVRA
ketorolac to local infiltration of ketorolac into the
wound margin.19 Three studies had a systemic control
group.17,18,20 All of these studies were supportive in
terms of postoperative outcome.
A dose-response study by Reuben et al.1 5 looked at
the effect of 10 to 50 mg of meperidine on postopera-
tive analgesia. The VAS at one hour and the need for
analgesic (acetaminophen/codeine) in the first 24 hr
were lower with 30 mg or more. TTFA increased lin-
early up to 30 mg. Side effects (nausea, vomiting, seda-
tion and dizziness) were more likely with 30 mg or
more. The incidence of side effects up to 20 mg
meperidine was statistically similar to the control group,
but the duration of analgesia significantly less compared
to 30 mg (269 67 compared to 419 123 min).
KETOROLAC
Three studies on a total 190 subjects, have investigat-
ed ketorolac (5–60 mg) as the sole adjunct to
IVRA.17,19,21
Only the study by Reuben et al.1 7 looked at the
potential intraoperative benefit of NSAIDs added to
LA. They compared tourniquet pain scores and found
significantly fewer patients had a pain score of >3/10
during the first 30 min when ketorolac 60 mg was
added.
MORPHINE
Two studies on a total 60 subjects, investigated mor-
phine as the sole adjunct.12,14 The doses studied were
1 and 6 mg.
Reuben et al.3 6 reported a number of significant
postoperative benefits from IVRA-ketorolac compared
to systemic control. Patients had less pain during the
first postoperative hour, required no supplemental anal-
gesia in the postanesthesia care unit (PACU) and con-
sumed fewer analgesics during the first postoperative
day when added to standard LA. In a follow-up study,
Reuben et al.1 9 found that ketorolac 60 mg was equal-
ly effective either infiltrated into the surgical site before
incision or when given as an adjunct to IVRA.
Steinberg et al.2 1 added varying doses of ketorolac to
IVRA and found a linear dose-response relationship up
to 20 mg. Between 20 and 60 mg, there appeared to be
no additional analgesic benefit. Side effects with IVRA
ketorolac were not reported, and fears about wound
hematoma remained unfounded.
Erciyes et al.1 4 added 6 mg morphine to LA and
reported a significantly faster onset and slower offset of
sensory block but only by approximately one minute
each. Gupta et al.1 2 focussed on the postoperative
course and found no benefit from the addition of mor-
phine 1 mg to LA. At the doses studied, there were no
significant side effects reported in either study.
SUFENTANIL
Hoffman et al.8 studied a number of single adjuncts to
LA for IVRA. The other agents are discussed else-
where. The addition of sufentanil 25 µg shortened the
onset of complete sensory block compared to control
by about three minutes. No postoperative benefit was

Choyce et al.: ADJUNCTS FOR IVRA
35
TENOXICAM AND ACETYL-SALICYLATE
(first 24 hr) was less in the clonidine group compared
to the control group. IV clonidine (systemic control
group) conferred no advantage.
Two studies involving 120 patients have looked at 20
mg tenoxicam as an adjunct.8,18 Both looked at the
postoperative effects only. Jones et al.1 8found patients
had less pain and consumed fewer analgesics during
the first 24 hr. Hoffman et al.8 failed to demonstrate
significant benefit from tenoxicam beyond 30 min
compared to control, but they did not record anal-
gesic intake and follow-up lasted only 45 min.
Two of five studies reported side effects.22,23 Both
these studies used a larger dose of clonidine (2 µg·kg– 1
or 150 µg) that resulted in sedation or hypotension
upon tourniquet deflation. The smaller dose of cloni-
dine (1 µg·kg–1) used by Reuben et al.2 4 and Lurie et
al.2 5 appears to be well tolerated.
One study20 involving 60 patients looked at acetyl-
salicylate 90 mg as an adjunct. Again, only the postop-
erative course was investigated. Duration of
postoperative analgesia was prolonged by over three
hours and significantly less morphine was consumed
during the first six hours compared to placebo; between
6–24 hr there was no difference between groups.
Muscle relaxants
Five studies involving 186 patients have investigated
muscle relaxants either in combination with fentanyl6,7
or as sole adjuncts26–28 (Table III). The agents investi-
gated were pancuronium (combined with fentanyl –
two studies),6,7 atracurium26,27 and mivacurium.2 8
Four out of five studies were supportive overall.6,7,26,27
Elhakim et al.2 7 looked at sensory onset and found
no difference. No study reported the speed of onset of
motor block. Elhakim et al.27 reported a significantly
greater degree of muscle relaxation and, in agreement
with McGlone et al.,26 reported that the addition of
atracurium 2 mg to IVRA significantly improved both
intraoperative analgesia and operating conditions for
performing closed and open reduction of wrist and
hand fractures.
Clonidine
Five studies (one with volunteers) involving 246
patients have investigated clonidine as the sole adjunct
for IVRA8,22–25 (Table II). The dose range was 1–2
µg·kg–1, with one study using 150 µg in all patients.
Two out of five studies included a systemic control
group.22,24 Four out of five of these studies were sup-
portive overall.8,23–25
No study has demonstrated a difference in terms of
onset of block. Two studies have investigated tourni-
quet pain and tolerance.23,25 Gentili et al.2 3showed that
clonidine 150 µg produced a significant increase in
tourniquet tolerance (median [range]: 22 [10–50] vs
10 [5–10] min). In a study in volunteers, Lurie et al.2 5
added clonidine 1 µg·kg– 1 and found that the onset of
intolerable tourniquet pain from the distal cuff (inflated
when the pain score from the proximal cuff reached
6/10) was delayed for over seven minutes.
Results of studies reporting on the postoperative
analgesic effects of clonidine have been mixed.
Kleinschmidt et al.2 2 found no difference in VAS and
AC when observing regression of the block for 45
min. Hoffmann et al.8 found reduced VAS at 30 and
45 min after tourniquet deflation, at which point
recording stopped. Gentili et al.2 3 were unable to
demonstrate any difference in VAS or AC although
these were only followed for 60 min and VASs were
low in both treatment and control groups.
Elhakim et al.2 7 measured postoperative pain and
found it to be reduced at five and 15 min. Two stud-
ies reported a delay in return of fine motor con-
trol.26,27 Torrance et al.2 8 investigated 0.6 mg
mivacurium, the shortest acting of the agents investi-
gated, and found a prolonged motor block after cuff
deflation (median recovery to 90% control 80 min,
range 60 min to more than eight hours).
Four out of five studies did not find any significant
side effects.6,7,26,27 Torrance et al.2 8 reported signs of
LA toxicity (light-headedness, perioral paresthesia,
tinnitis, diplopia) in the group given mivacurium 0.6
mg compared to none in the control group given a
similar dose of LA. This apparent interaction between
mivacurium and prilocaine has not been reported in
other studies.
Alkalinization
Three studies involving 131 patients have investigated
alkalinization with bicarbonate as an adjunct for
IVRA2 9 – 3 1 (Table III). Two studies were supportive.30,31
The three studies have each used different concen-
trations of LA from 0.5–1.0%. Armstrong et al.,31 in a
study in volunteers, demonstrated a slightly faster onset
of block when 0.5% prilocaine (pKa=7.9) had bicarbon-
ate added to adjust the pH from 6.40 to 7.75. This pH
adjustment theoretically raised the amount of free-base
Reuben et al.2 4 studied the effect of clonidine 1
µg·kg–1 on postoperative analgesia and presented a
number of significant findings. TTFA was longer in
the IVRA-clonidine group compared to control
(median [range]: 460 [215–1440] vs 115 [14–390]
min). VAS was reduced at one and two hours postop-
eratively in the clonidine group compared to control.
Less fentanyl was administered in the PACU and AC

36
CANADIAN JOURNAL OF ANESTHESIA
have demonstrated reduced tourniquet pain23,25 and
improved postoperative pain relief2 4 without adverse
effects. Improved muscle relaxation during IVRA can
be achieved by the addition of a non-depolarizing neu-
romuscular blocking agent such as atracurium 2 mg.
This can facilitate fracture reduction and also improve
overall analgesia particularly in young, muscular
patients.2 6 However there is a risk of residual muscle
weakness that can last several hours.2 8A muscle relaxant
plus an opioid have been shown to be a useful combi-
nation to allow the use of a non-toxic dose of lidocaine
(1.5 mg·kg–1) to provide a satisfactory block, albeit with
a slower onset of action.6,7 The alkalinization31 and
warming3 3of LA for IVRA offered little benefit beyond
reduced pain during injection.
Drugs selected as potential adjuncts to convention-
al LA agents could theoretically potentiate a block by
either altering nerve conduction or via peripheral
nociceptor binding. The observation that some drugs
are analgesic at the spinal level has led researchers to
examine if the same is true in the periphery. A variety
of receptors mediate nociceptor response and, there-
fore, peripherally administered agents may have an
analgesic benefit, perhaps avoiding systemic side
effects.3 5 IVRA isolates the limb from the rest of the
circulation and is a useful model for studying the
peripheral actions of a drug in the absence of central
effects. Well-designed studies looking at the duration
of postdeflation analgesia also included a systemic
control group to elucidate if the analgesic effect is
local or systemic.
from 3% to nearly 50%. In a follow-up clinical study,
Armstrong et al.3 0found no difference in time to estab-
lish an adequate block when 0.75% prilocaine under-
went similar pH adjustment. However, the more
alkaline solution was less painful to inject and produced
a denser block (decreased pain reported during
surgery). Benlabed et al.2 9 studied the effect of pH
adjustment on 1.0% lidocaine (pKa=7.8) for IVRA. The
change in pH of the solution from 6.63 to 7.34 result-
ed in an increase in the proportion of free base from 5%
in the control group to 22% in the study group.
Unfortunately, this study failed to demonstrate any dif-
ference with respect to sensory and motor onset or off-
set following alkalinization of LA.
Postoperative outcome following alkalinization of
LA has not been well investigated. Armstrong et al.3 0
noted less pain at five minutes post cuff deflation in
the treatment group, after which no further assess-
ments were made. Benlabed et al.2 9 could find no dif-
ference in the time to first report pain after surgery.
No side effects were reported in any of these studies.
Other agents
Table III also includes studies investigating the addi-
tion of potassium3 2 and alteration of temperature.33,34
McKeown et al.3 2 hypothesized that a physiological
concentration of potassium added to LA would poten-
tiate IVRA. Yet there was little clinical advantage to
this addition. Heavner et al.34 found no advantage
from cooling the limb by 5–10°C. Paul et al.3 3 inject-
ed LA solution at three different temperatures in vol-
unteers: 0°C, 22°C and 37°C. The cold solution was
significantly more painful and the warmed solution
was significantly less painful to inject compared to the
room temperature solution. It seems that warming LA
solutions may be of practical benefit in reducing the
discomfort associated with injection.
Opioids
Anesthesiologists have been striving for many years to
improve the efficacy and duration of regional anesthe-
sia by injecting opioids close to nerve trunks or nerve
endings.3 6 A peripheral action of opioids could theo-
retically be mediated via either a peripheral opioid
receptor3 7 or by a LA action of their own.3 8 It has
been suggested that previously inactive neuronal opi-
oid receptors may become active in painful inflamma-
tory conditions resulting in reduced neuronal
excitability, inhibited propagation of action potentials,
and the release of excitatory, pro-inflammatory neu-
ropeptides.3 9
Discussion
The results of this systematic review suggest that
NSAIDs have the most to offer as adjuncts to IVRA. A
dose-response study investigating ketorolac2 1 has
shown that the optimal dose is 20 mg, beyond which
no further postoperative benefit was accrued.
1 8
Tenoxicam and lysine-acetylsalicylate2 0 were also
shown to be beneficial for postoperative analgesia but
dose-response studies have not been performed, so the
ideal dose remains unknown. The results of studies
investigating opioids as adjuncts to IVRA have been
disappointing. Reuben et al.1 5demonstrated prolonged
postoperative analgesia with meperidine 30 mg but the
side effect profile was much better with 20 mg. Studies
investigating the addition of clonidine 1 µg·kg–1 to LA
Clinical evidence for the efficacy of peripherally
administered opioids is mixed. A recent systematic
review of opioids added to brachial plexus blocks
exposed a lack of well-conducted trials employing a
systemic control group, making overall evidence
equivocal.4 0 There appears to be better evidence for
the use of intraarticular morphine4 1 than for opioids
given by other peripheral routes.3 6

Choyce et al.: ADJUNCTS FOR IVRA
37

38
CANADIAN JOURNAL OF ANESTHESIA

Choyce et al.: ADJUNCTS FOR IVRA
39

40
CANADIAN JOURNAL OF ANESTHESIA

Choyce et al.: ADJUNCTS FOR IVRA
41

42
CANADIAN JOURNAL OF ANESTHESIA

Choyce et al.: ADJUNCTS FOR IVRA
43
NSAIDs
There is no evidence that clonidine affects the
speed of onset or quality of IVRA. However, two
studies suggested that clonidine improved tourniquet
pain tolerance.23,25 Previous studies have demonstrat-
ed that clonidine prolongs tourniquet tolerance under
spinal anesthesia via the intrathecal route5 0 and oral-
ly.5 1 The mechanism of action is unclear because
Gentili et al.2 3 kept the distal tourniquet inflated
throughout, deflating only the proximal cuff, whereas
Lurie et al.2 5 inflated the distal cuff and deflated the
proximal cuff when pain from the proximal cuff
became moderate.
There are conflicting results for the postoperative
benefit of clonidine and it is difficult to reconcile the
contradictory results. The study by Reuben et al.1 9
demonstrated improved postoperative analgesia for
two hours and reduced analgesic intake for 24 hr. This
was not shown by Gentili et al.,2 3 but the pain scores
they quoted were low making it difficult to detect a
difference from a small sample size. Kleinschmidt et
al.2 2 also failed to demonstrate any postoperative ben-
efit, this may be due to the use of a relatively insensi-
tive four-point pain scale and the lack of a well defined
analgesic protocol. A smaller dose of clonidine (1
µg·kg–1) is well tolerated.
Surgical trauma results in release of intracellular con-
tents from damaged and inflammatory cells.
Nociceptor stimulation causes a neurogenic response
with release of mediators such as substance P and neu-
rokinin A. This results in an “inflammatory soup”
containing histamine, serotonin, bradykinin and
metabolites of the cyclooxygenase and lipooxygenase
pathways.4 2 NSAIDs inhibit the production of
prostaglandins from arachidonic acid in phospholipid
membranes. The result is decreased afferent nocicep-
tive signals arising from the site of surgery. Whether
interfering with the synthesis of inflammatory media-
tors has a preemptive analgesic role in preventing sen-
sitization of nociceptors remains controversial.4 3
The role of NSAIDs in the management of postop-
erative pain is well established.4 4 Clinical studies have
demonstrated an enhanced analgesic effect from
NSAIDs when concentrated at a peripheral site com-
pared to the systemic administration of the same
drug.45,46 This would suggest a predominantly periph-
eral site of action.
It is interesting to note that the plasma half-life of
ketorolac is four to six hours yet the duration of anal-
gesia reached a plateau at over ten hours.1 7 It may be
that by concentrating the dose of NSAID at the site of
surgery, either as part of IVRA or wound infiltration,1 9
the resulting analgesic benefit is longer lasting than
the same dose administered parenterally. Presumably
there is a persistent drug level in the tissues, and this
coupled to the lower dosage could result in reduced
systemic side effects.
Muscle relaxants
Both LA and muscle relaxants exert an effect at the
neuromuscular junction. In addition, muscle relaxants
probably interfere with the muscle spindle activity
resulting in loss of muscle tone and spasm.2 7The spin-
dle is the sensory end-organ of skeletal muscles, send-
ing information about fibre length to the brain. The
resulting loss of tone and spasm may improve both
intraoperative pain and operating conditions.
Clonidine
a -adrenoceptors are located on primary afferent ter-
2
minals (both at peripheral and spinal endings), on
neurons in the superficial laminae of the spinal cord
and within several brainstem nuclei implicated in anal-
gesia.4 7 The mechanism by which clonidine, an ago-
nist at these receptors, produces analgesia is not fully
understood but is likely to be by a number of mecha-
nisms.4 7 Peripherally, reduced release of norepineph-
rine may contribute to analgesia4 8 and there is an
inhibitory effect on nerve-fibre action potentials that
Alkalinization
LA exist in two forms: the non-ionized, lipid-soluble
free base and the water-soluble ionized form. The rel-
ative proportions of each depend upon the pKa of the
drug and the pH of the environment. The pKa of a LA
is fixed but, by increasing the pH of a solution, it is
possible to increase the percentage of free base and
thus improve the nerve penetration and the rate of
onset of blockade. Unfortunately, raising the pH too
high will result in precipitation of the drug.
Several mechanisms are thought to be involved in
the potentiation of LA by alkalinization with bicar-
bonate.5 2 Firstly, bicarbonate reduces the net inward
current (decreased sodium influx, increased potassium
efflux) by a direct effect on membrane channels.
Secondly, ion trapping occurs when membrane-per-
meable CO2 combines with axoplasmic water to pro-
is not mediated by the a -receptor.4 9
2
In clinical studies, clonidine-containing LA solu-
tions have been shown to prolong and intensify anal-
gesia, compared to plain solutions, when used for
spinal, epidural or peripheral nerve blocks.40,47 In a
recent systematic review, clonidine was reported to be
the most promising analgesic adjunct for brachial
plexus block, with minimal risk of adverse effects in
doses up to 150 µg.40

44
CANADIAN JOURNAL OF ANESTHESIA
Anaesthesia 1991; 46: 278–80.
duce carbonic acid that releases protons. The neutral
form of LA that has crossed the membrane will then
accept these protons and become a charged, active
channel-blocking molecule. Thirdly, affinity of the LA
molecule for the sodium channel increases. This is well
demonstrated by the substantial potentiation of ben-
zocaine, an unionizable, uncharged LA.52
10 Arthur JM, Heavner JE, Mian T, Rosenberg PH.
Fentanyl and lidocaine versus lidocaine for Bier block.
Reg Anesth 1992; 17: 223–7.
11 Pitkänen MT, Rosenberg PH, Pere PJ, Tuominen MK,
Seppälä TA. Fentanyl-prilocaine mixture for intra-
venous regional anaesthesia in patients undergoing
surgery. Anaesthesia 1992; 47: 395–8.
12 Gupta A, Björnsson A, Sjöberg F, Bengtsson M. Lack of
peripheral analgesic effect of low-dose morphine dur-
ing intravenous regional anesthesia. Reg Anesth 1993;
18: 250–3.
13 Armstrong PJ, Morton CPJ, Nimmo AF. Pethidine has
a local anaesthetic action on peripheral nerves in vivo.
Addition to prilocaine 0.25% for intravenous regional
anaesthesia in volunteers. Anaesthesia 1993; 48:
382–6.
14 Erciyes N, Akturk G, Solak M, Dohman D.
Morphine/prilocaine combination for intravenous
regional anesthesia. Acta Anaesthesiol Scand 1995; 39:
845–6.
15 Reuben SS, Steinberg RB, Lurie SD, Gibson CS. A dose-
response study of intravenous regional anesthesia with
meperidine. Anesth Analg 1999; 88: 831–5.
16 Acalovschi I, Cristea T, Margarit S, Gavrus R.
Tramadol added to lidocaine for intravenous regional
anesthesia. Anesth Analg 2001; 92: 209–14.
17 Reuben SS, Steinberg RB, Kreitzer JM, Duprat KM.
Intravenous regional anesthesia using lidocaine and
ketorolac. Anesth Analg 1995; 81: 110–3.
18 Jones NC, Pugh SC. The addition of tenoxicam to
prilocaine for intravenous regional anaesthesia.
Anaesthesia 1996; 51: 446–8.
Conclusion
There is good evidence to recommend NSAIDs in
general, and ketorolac in particular, for improving
postoperative analgesia after IVRA. Clonidine also
appears to improve postoperative analgesia and pro-
long tourniquet tolerance. Opioids are disappointing
by this route; only 30 mg meperidine has substantial
postoperative benefit but at the expense of postdefla-
tion side effects. Muscle relaxants improve motor
block and aid fracture reduction.
References
1 Brown BL, Fink BR. The history of neural blockade
and pain management. In: Cousins MJ, Bridenbaugh
PO (Eds.). Neural Blockade in Clinical Anesthesia and
Management of Pain, 3rd ed. Philadelphia: Lippincott-
Raven, 1998: 3–34.
2 Holmes CMCK. Intravenous regional neural blockade.
In: Cousins MJ, Bridenbaugh PO (Eds.) Neural
Blockade in Clinical Anesthesia and Management of
Pain, 3rd ed. Philadelphia: Lippincott-Raven, 1998:
395–410.
3 Henderson CL, Warriner CB, McEwen JA, Merrick PM.
A North American survey of intravenous regional anes-
thesia. Anesth Analg 1997; 85: 858–63.
19 Reuben SS, Duprat KM. Comparison of wound infiltra-
tion with ketorolac versus intravenous regional anes-
thesia with ketorolac for postoperative analgesia
following ambulatory hand surgery. Reg Anesth 1996;
21: 565–8.
20 Corpataux J-B, Van Gessel EF, Donald FA, Forster A,
Gamulin Z. Effect on postoperative analgesia of small-
dose lysine acetylsalicylate added to prilocaine during
intravenous regional anesthesia. Anesth Analg 1997;
84: 1081–5.
4 Bader AM, Concepcion M, Hurley RJ, Arthur GR.
Comparison of lidocaine and prilocaine for intravenous
regional anesthesia. Anesthesiology 1988; 69: 409–12.
5 Chilvers CR, Kinahan A, Vaghadia H, Merrick PM.
Pharmacoeconomics of intravenous regional anaesthe-
sia vs general anaesthesia for outpatient hand surgery.
Can J Anaesth 1997; 44: 1152–6.
6 Abdulla WY, Fadhil NM. A new approach to intra-
venous regional anesthesia. Anesth Analg 1992; 75:
597–601.
21 Steinberg RB, Reuben SS, Gardner G. The dose-
response relationship of ketorolac as a component of
intravenous regional anesthesia with lidocaine. Anesth
Analg 1998; 86: 791–3.
22 Kleinschmidt S, Stöckl W, Wilhelm W, Larsen R. The
addition of clonidine to prilocaine for intravenous
regional anaesthesia. Eur J Anaesthesiol 1997; 14:
40–6.
7 Sztark F, Thicoipe M, Favarel-Garrigues JF, Lassie P,
Petitjean ME, Dabadie P. The use of 0.25% lidocaine
with fentanyl and pancuronium for intravenous region-
al anesthesia. Anesth Analg 1997; 84: 777–9.
8 Hoffmann V, Vercauteren M, Van Steenberge A,
Adriaensen H. Intravenous regional anesthesia.
Evaluation of 4 different additives to prilocaine. Acta
Anaesthesiol Belg 1997; 48: 71–6.
23 Gentili M, Bernard J-M, Bonnet F. Adding clonidine to
lidocaine for intravenous regional anesthesia prevents
9 Armstrong P, Power I, Wildsmith JAW. Addition of fen-
tanyl to prilocaine for intravenous regional anaesthesia.

Choyce et al.: ADJUNCTS FOR IVRA
45
review. Anesth Analg 2000; 90: 1122–8.
41 Kalso E, Tramèr MR, Carroll D, McQuay HJ, Moore
tourniquet pain. Anesth Analg 1999; 88: 1327–30.
24 Reuben SS, Steinberg RB, Klatt JL, Klatt ML.
Intravenous regional anesthesia using lidocaine and
clonidine. Anesthesiology 1999; 91: 654–8.
25 Lurie SD, Reuben SS, Gibson CS, DeLuca PA, Maciolek
HA. Effect of clonidine on upper extremity tourniquet
pain in healthy volunteers. Reg Anesth Pain Med
2000; 25: 502–5.
26 McGlone R, Heyes F, Harris P. The use of a muscle
relaxant to supplement local anaesthetics for Bier’s
blocks. Arch Emerg Med 1988; 5: 79–85.
27 Elhakim M, Sadek RA. Addition of atracurium to lido-
caine for intravenous regional anaesthesia. Acta
Anaesthesiol Scand 1994; 38: 54–4.
28 Torrance JM, Lewer BMF, Galletly DC. Low-dose
mivacurium supplementation of prilocaine i.v. regional
anaesthesia. Br J Anaesth 1997; 78: 222–3.
29 Benlabed M, Jullien P, Guelmi K, Hamza J, Bonhomme
L, Benhamou D. Alkalinization of 0.5% lidocaine for
intravenous regional anesthesia. Reg Anesth 1990; 15:
59–60.
30 Armstrong P, Brockway M, Wildsmith JA. Alkalinisation
of prilocaine for intravenous regional anaesthesia.
Anaesthesia 1990; 45: 11–3.
RA. Pain relief from intra-articular morphine after knee
surgery: a qualitative systematic review. Pain 1997; 71:
127–34.
42 Siddall PJ, Cousins MJ. Introduction to pain mecha-
nisms. Implications for neural blockade. In: Cousins
MJ, Bridenbaugh PO (Eds.) Neural Blockade in
Clinical Anesthesia and Management of Pain.
Philadelphia: Lippincott-Raven, 1998: 675.
43 Woolf CJ, Chong M-S. Preemptive analgesia–treating
postoperative pain by preventing the establishment of
central sensitization. Anesth Analg 1993; 77: 362–79.
44 Dahl JB, Kehlet H. Non-steroidal anti-inflammatory
drugs: rationale for use in severe postoperative pain. Br
J Anaesth 1991; 66: 703–12.
45 Ben-David B, Katz E, Gaitini L, Goldik Z. Comparison
of i.m. and local infiltration of ketorolac with and with-
out local anaesthetic. Br J Anaesth 1995; 75: 409–12.
46 O’Hanlon JJ, McCleane G, Muldoon T. Preoperative
application of piroxicam gel compared to a local anaes-
thetic field block for postoperative analgesia. Acta
Anaesthesiol Scand 1996; 40: 715–8.
47 Eisenach JC, De Kock M, Klimscha W.
2-Adrenergic
31 Armstrong P, Watters J, Whitfield A. Alkalinisation of
prilocaine for intravenous regional anaesthesia.
Suitability for clinical use. Anaesthesia 1990; 45:
935–7.
32 McKeown DW, Scott DB. Influence of the addition of
potassium to 0.5% prilocaine solution during i.v.
regional anaesthesia. Br J Anaesth 1984; 56: 1167–70.
33 Paul DL, Logan MR, Wildsmith JAW. The effects of
injected solution temperature on intravenous regional
anaesthesia. Anaesthesia 1988; 43: 362–4.
34 Heavner JE, Leinonen L, Haasio J, Kytta J, Rosenberg
PH. Interaction of lidocaine and hypothermia in Bier
blocks in volunteers. Anesth Analg 1989; 69: 53–5.
35 Carlton SM, Coggeshall RE. Nociceptive integration:
does it have a peripheral component? Pain Forum
1998; 7: 71–8.
36 Picard PR, Tramèr MR, McQuay HJ, Moore RA.
Analgesic efficacy of peripheral opioids (all except
intra-articular): a qualitative systematic review of ran-
domised controlled trials. Pain 1997; 72: 309–18.
37 Stein C, Yassouridis A. Peripheral morphine analgesia
(Editorial). Pain 1997; 71: 119–21.
agonists for regional anesthesia. A clinical review of
clonidine (1984-1995). Anesthesiology 1996; 85:
655–74.
48 Sato J, Perl ER. Adrenergic excitation of cutaneous
pain receptors induced by peripheral nerve injury.
Science 1991; 251: 1608–10.
49 Gaumann DM, Brunet PC, Jirounek P. Clonidine
enhances the effects of lidocaine on C-fiber action
potential. Anesth Analg 1992; 74: 719–25.
50 Bonnet F, Diallo A, Saada M, Belon M, Guilbaud M,
Boico O. Prevention of tourniquet pain by spinal isobar-
ic bupivacaine with clonidine. Br J Anaesth 1989; 63:
93–6.
51 Liu S, Chiu AA, Neal JM, Carpenter RL, Bainton BG,
Gerancher JC. Oral clonidine prolongs lidocaine spinal
anesthesia in human volunteers. Anesthesiology 1995;
82: 1353–9.
52 Wong K, Strichartz GR, Raymond SA. On the mecha-
nisms of potentiation of local anesthetics by bicarbon-
ate buffer: drug structure-activity studies on isolated
peripheral nerve. Anesth Analg 1993; 76: 131–43.
38 Power I, Brown DT, Wildsmith JAW. The effect of fen-
tanyl, meperidine and diamorphine on nerve conduc-
tion in vitro. Reg Anesth 1991; 16: 204–8.
39 Stein C. The control of pain in peripheral tissue by opi-
oids. N Engl J Med 1995; 332: 1685–90.
40 Murphy DB, McCartney CJL, Chan VWS. Novel anal-
gesic adjuncts for brachial plexus block: a systematic