Anaesthesia 2014, 69, 954–960 doi:10.1111/anae.

12774

Original Article
Analgesic efficacy of pre-operative stellate ganglion block on
postoperative pain relief: a randomised controlled trial
N. Kumar,1 D. Thapa,2 S. Gombar,3 V. Ahuja4 and R. Gupta5

1 Junior Resident, 2 Associate Professor, 3 Professor, 4 Assistant Professor, Department of Anaesthesia and Intensive
Care, 5 Professor, Department of Orthopaedics, Government Medical College and Hospital, Chandigarh, India

Summary
We undertook a randomised, double-blind, placebo-controlled study to compare the analgesic efficacy of pre-opera-
tive stellate ganglion block on postoperative pain relief after upper limb orthopaedic surgery. Patients were adminis-
tered a 3-ml injection during ultrasound-guided stellate ganglion block; 15 patients received lidocaine 2% and 15
patients received 0.9% saline. Following the block, all patients received standardised general anaesthesia. Postoperative
analgesia included regular intravenous diclofenac, paracetamol and patient-controlled analgesia with tramadol for
24 h. Patients were observed at 0, 2, 4, 6, 8, 12 and 24 h after surgery for tramadol consumption, cardiovascular vari-
ables and visual analogue scale pain scores at rest and on movement. The mean (SD) hourly tramadol consumption
was significantly reduced in the lidocaine group compared with the saline group at 4 h (8.0 (10.1) mg vs 28.0
(12.6) mg, respectively; p = 0.001), 6 h (5.3 (10.8) mg vs 17.3 (12.7) mg, respectively; p = 0.013) and 8 h (5.3
(11.8) mg vs 21.3 (9.1) mg, respectively; p = 0.001). The cumulative 24-h tramadol consumption was 97.3 (16.6) mg
in the lidocaine group and 150.6 (26.0) mg in the saline group (p = 0.001). There were significant differences in the
pain visual analogue scale at rest at two time points; at 4 h the median (IQR [range]) visual analogue scale scores
were 4 (4–6 [2–8]) in the lidocaine group and 5 (4–6 [2–7]) in the saline group (p = 0.03), and at 6 h visual analogue
scale scores were 3 (3–4 [3–6]) and 4 (4–6 [2–7]), respectively (p = 0.04). Pain visual analogue scale on movement
was lower in the lidocaine group at all time intervals compared with the saline group, but this did not reach statistical
significance. The present study has demonstrated a postoperative tramadol-sparing and analgesic effect of pre-opera-
tive stellate ganglion block in patients undergoing upper limb orthopaedic surgery under general anaesthesia.
.................................................................................................................................................................
Correspondence to: V. Ahuja
Email: vanitaanupam@yahoo.co.in
Accepted: 14 May 2014

Introduction but this may be associated with significant side-effects

Postoperative pain relief is a major concern for the
anaesthesiologist after upper limb surgery. Inade-
quately managed postoperative pain is associated with
increased autonomic instability, and acute pain may
progress to chronic pain if left untreated [1].
High-dose opioids have been used during the intra-
and postoperative periods for fractured humerus [2],
such as nausea, vomiting, sedation and respiratory
depression. Nerve blocks have been used for orthopaedic
procedures on the upper limb for both surgery and post-
operative pain relief, but are associated with loss of
tactile sensation and motor function which hampers
functional assessment, as well as nerve injuries in the
early postoperative period [2, 3].

954 © 2014 The Association of Anaesthetists of Great Britain and Ireland

Kumar et al. | Stellate ganglion block and postoperative pain relief
The role of the sympathetic nervous system is well
established in patients with chronic pain states such as
complex regional pain syndrome [2]. Recently,
McDonnell et al. described the use of pre-operative
ultrasound-guided stellate ganglion block in four
patients having upper limb surgery for trauma, with
promising results in terms of low postoperative pain
scores and morphine consumption [2]. An editorial
accompanying their study suggested that these findings
should be confirmed using a well-designed randomised
controlled clinical trial [4]. We conducted the present
randomised trial to compare the efficacy of pre-opera-
tive stellate ganglion block with lidocaine or saline,
and its effect on tramadol consumption during the first
postoperative 24 h, in patients following upper limb
orthopaedic surgery.
Methods
We performed a prospective, randomised, double-
blinded, placebo-controlled study between April 2012
and September 2013 after gaining approval from our
Institutional Ethics Committee. We included patients
of ASA physical status 1–2, aged between 18 and
60 years with an upper limb fracture scheduled for
surgery under general anaesthesia. We did not study
patients with a body mass index > 30 kg.m 2, a his-
tory of substance abuse, inability to understand a pain
visual analogue scale (VAS), inability to use patient-
controlled analgesia (PCA), allergy to lidocaine,
chronic sympathetic-mediated pain syndromes affect-
ing the upper limb or trauma other than to the
affected limb.
Patients were evaluated a day before surgery to
assess their fitness for the proposed surgical procedure
under general anaesthesia, and written informed
consent was obtained at this stage. Patients were
instructed on the use of the PCA pump and the pain
VAS the day before surgery, and were reminded before
transfer to the operating theatre and in the post-
anaesthesia care unit. All patients were fasted for 8 h
pre-operatively and premedicated with oral alprazolam
0.25 mg and ranitidine 150 mg the night before and
2 h before surgery.
In the operating theatre, mandatory monitoring
including non-invasive blood pressure, heart rate, con-
tinuous ECG, oxygen saturation and end-tidal carbon
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Anaesthesia 2014, 69, 954–960
dioxide (AS5; Datex Ohmeda, Helsinki, Finland) was
placed. Intravenous access was established and patients
received 500 ml saline 0.9% and midazolam 1–2 mg.
Randomisation of patients was performed using a
computer-generated random number table. An anaes-
thesiologist who was not participating in the study
opened an opaque envelope and prepared the study
drug. The blocks were performed by trained consultant
anaesthesiologists (DT, VA), or by a resident (NK)
under their direct supervision. The anaesthesiologist
performing the block and the patient were blinded to
the study group. Stellate ganglion block was performed
under ultrasound guidance with a SonoSite TITANâ
with 5–10 MHz probe (SonoSite Inc, Bothell, WA,
USA) using the para-tracheal, out-of-plane technique.
Doppler imaging was used to rule out any aberrant
vessels along the plane of needle placement. Three mil-
lilitres of either lidocaine 2% (lidocaine group) or sal-
ine 0.9% (saline group) was injected at the level of the
C7 vertebra underneath the fascia of the longus colli
muscle. Correct localisation of the injection was dem-
onstrated by expansion of the fascia over the muscle
[2]. Patients in both groups were observed for any
possible adverse effects of the block such as Horner’s
syndrome, motor or sensory blockade of the upper
limb, recurrent laryngeal nerve or phrenic nerve block-
ade or any other effects for 10 min before induction of
general anaesthesia.
General anaesthesia was induced with intravenous
fentanyl 2 lg.kg 1 and propofol 2–3 mg.kg 1, and ve-
curonium 0.1 mg.kg 1 was administered to provide
muscle paralysis. After tracheal intubation, anaesthesia
was maintained with nitrous oxide:oxygen 60:40 and
isoflurane 1–2%, and intermittent boluses of vecuroni-
um were administered as required. Twenty minutes
before the end of surgery, tramadol 1 mg.kg 1 and
ondansetron 0.1 mg.kg 1 were administered intrave-
nously. At the end of surgery, residual neuromuscular
blockade was reversed with intravenous neostigmine
50 lg.kg 1 plus glycopyrronium 10 lg.kg 1.
Routine postoperative analgesia included tramadol
PCA, intravenous paracetamol 1 g every 6 h and intra-
venous diclofenac 75 mg every 12 h. Tramadol was
administered via a PCA pump (Master PCA; Fresenius
Kabi, Bad Homburg, Germany) with a bolus of 20 mg,
lockout time of 20 min and a maximum dose
955
Anaesthesia 2014, 69, 954–960 Kumar et al. | Stellate ganglion block and postoperative pain relief
≤ 400 mg.24 h 1 [5]. Intravenous ondansetron 4 mg
three times.day 1 and ranitidine 50 mg twice.day 1
were also administered.
Postoperative measurements were made at 0, 2, 4,
6, 8, 12 and 24 h postoperatively consisting of tram-
adol consumption, cardiovascular variables, pain VAS
at rest and movement, sedation, nausea, vomiting, and
any adverse effects. Sedation was assessed using a
four-point scale (0: completely alert; 1: sleepy occa-
sionally but rousable; 2: asleep often but rousable; 3:
asleep and unrousable) [6]. Nausea and vomiting were
assessed using a four-point scale (0: no nausea and
vomiting; 1: slight nausea resolving without treatment;
2: slight nausea and/or vomiting responding to treat-
ment; 3: nausea and/or vomiting not responding to
treatment) [7].
Sample size was calculated based on a previous
study [8], in which the mean (SD) 24-h consumption of
PCA tramadol was 267 (91) mg. We considered a 50%
reduction in tramadol consumption as significant. To
demonstrate this difference at p < 0.05 with a power of
90%, the study required 12 patients per group. To cater
for a 20% dropout rate, we enrolled 15 patients per
group. The statistical analysis was performed using Sta-
tistical Package for Social Sciences (version 15.0 for
Windows; SPSS Inc, Chicago, IL, USA). For normally
distributed data, means were compared using an inde-
pendent sample t-test. For skewed data, the Mann–
Whitney test was applied. For time-related variables,
repeated measures ANOVA was applied. Proportions
were compared using chi-squared or Fisher’s exact test
as appropriate. All statistical tests were two-sided and a
value of p ≤ 0.05 was considered statistically significant.
Results
A total of 40 patients were screened during the trial,
of whom 10 patients were not studied as they did not
fulfil the inclusion criteria (Fig. 1), leaving 15 patients
in the lidocaine group and 15 in the saline group. The
lidocaine group included 11 patients with a fractured
shaft of humerus and four with fractures of radius and
ulna, and the saline group included seven and eight
patients in the respective groups. Table 1 shows the
physical characteristics of the patients.
The mean (SD) total tramadol consumption over
24 h was 97.3 (16.6) mg in the lidocaine group, which
956
was significantly lower than that in the saline group
150.6 (26.0) mg (p = 0.001). There were significant
differences in hourly tramadol consumption between
the two groups at 4, 6 and 8 h (Fig. 2). The VAS
scores at rest were lower in the lidocaine group at 4
and 6 h only (Table 2).
Heart rate was lower in the lidocaine group at 2, 4
and 8 h, and mean arterial pressure at 4 h (Fig. 3).
There were no differences in respiratory rate. None of
the patients demonstrated motor or sensory blockade
in the arm, shivering, pruritus, Horner’s syndrome or
other adverse effects.
Discussion
Postoperative pain has both neuropathic and inflam-
matory components [2]. We suggest that in the pres-
ent study, stellate ganglion block relieved the
neuropathic (sympathetic) component and the sys-
temic analgesics relieved the inflammatory component.
Kakazu & Julka have suggested the possible involve-
ment of the sympathetic nervous system in acute pain
following humerus fracture [9]. They found reduction
in pain VAS scores from 10 to 0 within 5 min of stel-
late ganglion block placement [9]. In the case series of
McDonnell et al., excellent postoperative analgesia with
low morphine consumption was provided for 48 h fol-
lowing stellate ganglion block with lidocaine [2]. The
expected duration of analgesia of stellate ganglion
block is up to 72 h when performed in patients with
chronic pain [10, 11]. Our study was intended to ver-
ify the role of sympathetic block in treatment of early
postoperative pain and, therefore, we did not continue
observations beyond 24 h. Further studies will be
required to establish any longer term benefits of this
approach.
We found a significant reduction in pain VAS at
rest in the lidocaine group compared with the saline
group. A plausible explanation is that the sympathetic
nervous system, which is normally inactive, becomes
activated following injury or surgery [12]. The role of
stellate ganglion block is well established in patients
with chronic pain as it interrupts the pain cycle,
reduces sympathetic tone, prevents central sensitisation
and helps to restore normal somatic sensation [12, 13].
The reduction in resting pain in the lidocaine group
might be due to improved blood flow and washing out
© 2014 The Association of Anaesthetists of Great Britain and Ireland
Kumar et al. | Stellate ganglion block and postoperative pain relief Anaesthesia 2014, 69, 954–960

Figure 1 CONSORT diagram of study recruitment.

of inflammatory mediators in the blocked arm [2], producing motor or sensory blockade, allowing the
thus modulating and attenuating the neuropathic com- surgeon to assess the operated arm in the immediate
ponent of Ad and C-fibres in the deeper muscles [2, postoperative period [3]. The use of a low volume of
14–17]. Our observation that there was only marginal
difference in pain VAS on movement could have mul- Table 1 Comparison of characteristics of 30 patients
tiple mechanisms related to a greater pain response in receiving stellate ganglion block with lidocaine or sal-
postoperative patients [18, 19]. A continuous release of ine. Values are mean (SD) or number (proportion).
inflammatory mediators in the peripheral tissue might
Lidocaine Saline
sensitise functional and dormant nociceptors responsi- (n = 15) (n = 15)
ble for somatic pain on movement [20]; these mecha- Age; years 37.5 (10.2) 38.6 (13.2)
nisms would not be directly influenced by stellate Male 10 (66%) 12 (80%)
ganglion block. Height; cm 168.1 (7.1) 176.6 (25.6)
2
Body mass index; kg.m 24.6 (3.3) 23.8 (4.7)
Stellate ganglion block provides an additional Weight; kg 69.4 (7.3) 71.9 (10.2)
advantage over conventional nerve block of not

© 2014 The Association of Anaesthetists of Great Britain and Ireland 957

Anaesthesia 2014, 69, 954–960 Kumar et al. | Stellate ganglion block and postoperative pain relief

60 to produce a sympathetic block [24, 25]. The safety of

50 the patients was ensured throughout the placement of
* the block as trained personnel performed the block
40
Tramadol (mg)

* * under ultrasound guidance with strict asepsis. None of

30
20
10
0
0 2 4 6 8 12
–10
Time (h)
Figure 2 Comparison of mean hourly tramadol
consumption (mg) after surgery in patients receiving
lidocaine ( ) or saline ( ) for stellate ganglion block.
Error bars are SD. *p < 0.05.
local anaesthetic to ensure the absence of brachial
plexus block and other adverse effects of stellate gan-
glion block is supported by the literature [2, 21]. The
dose that we used is too small to have had a systemic
analgesic effect [22].
Recently, McGuirk et al. questioned the use of
invasive placebos in randomised controlled trials
because of the potential for harm [23]. They described
a ‘SHAM’ (Serious Harm and Morbidity) scale to
assess the risk (0: no risk (no intervention); 1: minimal
risk (non-invasive placebo action); 2: minor risk (mini-
mally invasive placebo); 3: moderate risk (invasive pla-
cebo intervention); 4: major risk (invasive placebo
procedure)) [23, 24]. The present study would be clas-
sified as SHAM grade 4. A stellate ganglion injection
of saline 3 ml was used to ensure maximum internal
validity, blinding and unbiased assessment [23, 24].
Saline for stellate ganglion block has been shown not
the patients had any procedure-related complications.
Potentially life-threatening complications such as
subarachnoid or vertebral artery misplacement may
occur following stellate ganglion block at a frequency
24 of 1.7 in 1000 [26]. The development of Horner’s syn-
drome has been used to indicate success when per-
forming stellate ganglion block at the C6 vertebra
using a blind technique with a large volume injection.
Volumes > 5 ml spread to distant structures such as
the carotid baroreceptor and vagus nerve leading to
Horner’s syndrome and hoarseness of voice [27, 28].
The blind technique also carries greater risk of injury
to the vertebral artery and pneumothorax [29]. Other
complications of stellate ganglion block include head-
ache, nausea, blurred vision, dysphagia, hoarseness and
haematoma at the puncture site. Kastler et al. reported
that a stellate ganglion block with 10 ml bupivacaine
0.25% at C6–C7 using fluoroscopy resulted in transient
side-effects in 84% of patients including 76% with dys-
phagia, but with no serious complications [30]. A
recent study using computed tomography-guided stel-
late ganglion block at the C7 and T1 vertebral levels
found no complications in the block group [31].
Some studies suggest that Horner’s syndrome is
not an essential indicator of correct placement of the
local anaesthetic during stellate ganglion block [11,
27]. We did not observe Horner’s syndrome in our
patients, possibly due to the use of a low volume, place-
ment of the block at the C7 level, use of ultrasound

Table 2 Comparison of visual analogue scale pain scores at rest and during movement in 30 patients receiving
lidocaine or saline for stellate ganglion block. Values are median (IQR [range]).

Rest Movement

Lidocaine (n = 15) Saline (n = 15) Lidocaine (n = 15) Saline (n = 15)

0h 8 (8–9 [0–10]) 8 (7–8 [6–9]) 9 (8–9 [0–10]) 9 (8–9 [7–9])
2h 4 (4–6 [2–8]) 6 (4–7 [2–9]) 6 (5–8 [3–9]) 6 (5–7 [3–9])
4h 4 (3–5 [3–7])* 5 (4–6 [2–7])* 5 (4–6 [3–8]) 6 (5–6 [3–8])
6h 3 (3–4 [3–6])** 4 (4–6 [2–7])** 4 (4–5 [3–7]) 5 (4–6 [3–8])
8h 3 (3–4 [3–6]) 4 (4–5 [2–7]) 4 (4–4 [3–7]) 5 (4–5 [3–8])
12 h 5 (3–6 [3–7]) 4 (4–5 [2–6]) 5 (4–7 [4–8]) 4 (4–5 [3–6])
24 h 3 (3–4 [2–5]) 3 (3–4 [2–4]) 4 (3–4 [2–5]) 4 (3–4 [2–4])

*p = 0.03.
**p = 0.04.

958 © 2014 The Association of Anaesthetists of Great Britain and Ireland

Kumar et al. | Stellate ganglion block and postoperative pain relief Anaesthesia 2014, 69, 954–960

(a) 95 undergoing upper limb orthopaedic surgery under

* general anaesthesia.
Heart rate (beats.min–1)

90 * *
85
80 Competing interests
75 No external funding and no competing interests
70 declared.
65
60
0 2 4 6 8 12 24 References
Time (h) 1. Kehlet H. Multimodal approach to control postoperative path-
ophysiology and rehabilitation. British Journal of Anaesthesia
(b) 105 1997; 78: 606–17.
2. McDonnell JG, Finnerty O, Laffey JG. Stellate ganglion blockade
100 * for analgesia following upper limb surgery. Anaesthesia 2011;
66: 611–4.
MAP (mmHg)

95
90
85
80
75
0 2 4 6 8 12
Time (h)
Figure 3 Cardiovascular variables (mean) after surgery
in patients receiving lidocaine ( ) or saline ( ) for stel-
late ganglion block: (a) heart rate; (b) mean arterial
pressure (MAP). *p < 0.05. Error bars are SD.
guidance and induction of general anaesthesia soon
after placement of the block; Horner’s syndrome usu-
ally manifests in 15–30 min [32]. In addition, after
induction of general anaesthesia, the assessment of
cardinal features of Horner’s syndrome such as
enophthalmos, ptosis and miosis becomes difficult and
inconclusive due to the use of neuromuscular blocking
drugs and opioids [3].
This study obtains its strength due to the unavail-
ability of any other randomised controlled trial on
ultrasound-guided stellate ganglion block for postoper-
ative pain in upper limb orthopaedic surgery. How-
ever, there are some limitations. Firstly, lidocaine has a
short duration of action. Secondly, the sample size is
small. Further multi-centre studies will be required to
strengthen the findings of our study and the potential
benefits of stellate ganglion block in patients with ASA
grade ≥ 2 and those where early functional assessment
of the arm is required by the surgeon.
In conclusion, our study has established a post-
operative tramadol-sparing and pain-relieving effect
of pre-operative stellate ganglion block in patients
3. Hurley RW, Wu CL. Acute post-operative pain. In: Miller RD,
Eriksson LI, Fleisher LA, Wiener-Kronish JP, Young WL, eds.
Miller’s Anaesthesia, Vol. 2, 7th edn. Philadelphia, PA: Elsevier
Churchill Livingstone, 2010: 2757–82.
4. Chambers WA, Smith WCS. Case reports of novel treatments
– proper evaluation before clinical use. Anaesthesia 2011; 66:
539–40.
24
5. Vickers MD, Morgan M, Spencer PSJ, Read MS. Drugs in Anaes-
thetic and Intensive Care Practice, 8th edn. Boston, MA: But-
terworth Heinemann, 1999.
6. Day FJ. Stellate ganglion block: normal saline as placebo.
Anesthesiology 1988; 68: 819.
7. Chen CK, Tan PC, Phui VE, Teo SC. A comparison of analgesic
efficacy between oblique subcostal transversus abdominis
plane block and intravenous morphine for laparoscopic cho-
lecystectomy. A prospective randomized controlled trial. Kor-
ean Journal of Anesthesiology 2013; 64: 511–6.
8. Kemal SO, Sahin S, Apan A. Comparison of tramadol, tram-
adol-metamizol and tramadol-lornoxicam administered by
intravenous PCA in management of postoperative pain. Agri
2007; 19: 24–31.
9. Kakazu CZ, Julka I. Stellate ganglion blockade for acute post-
operative upper extremity pain. Anesthesiology 2005; 102:
1288–9.
10. Arne
r S, Lindblom U, Meyerson BA, Molander C. Prolonged
relief of neuralgia after regional anesthetic blocks. A call for
further experimental and systematic clinical studies. Pain
1990; 43: 287–97.
11. Price DD, Long S, Wilsey B, Rafii A. Analysis of peak magnitude
and duration of analgesia produced by local anesthetics
injected into sympathetic ganglia of complex regional pain syn-
drome patients. Clinical Journal of Pain 1998; 14: 216–26.
12. Bantel C, Trapp S. The role of the autonomic nervous system
in acute surgical pain processing – what do we know? Anaes-
thesia 2011; 66: 541–4.
13. Rho RH, Brewer RP, Lamer TJ, Wilson PR. Complex regional
pain syndrome. Mayo Clinic Proceedings 2002; 77: 174–80.
14. Bennett GJ. Neuropathic pain. In: Wall PD, Melzack R, eds.
Textbook of Pain, 3rd edn. Edinburgh: Churchill Livingstone,
1994: 201–24.
15. Ogonda L, Wilson R, Archbold P, et al. A minimal-incision tech-
nique in total hip arthroplasty does not improve early postop-
erative outcomes: a prospective, randomized, controlled trial.
Journal of Bone and Joint Surgery. American volume 2005; 87:
701–10.
16. Dorr LD, Maheshwari AV, Long WT, Wan Z, Sirianni LE. Early
pain relief and function after posterior minimally invasive and

© 2014 The Association of Anaesthetists of Great Britain and Ireland 959

Anaesthesia 2014, 69, 954–960
conventional total hip arthroplasty: a prospective, random-
ized, blinded study. Journal of Bone and Joint Surgery. Ameri-
can volume 2007; 89: 1153–60.
17. Møiniche S, Dahl JB, Erichsen CJ, Jensen LM, Kehlet H. Time
course of subjective pain ratings, and wound and leg tender-
ness after hysterectomy. Acta Anaesthesiologica Scandinavica
1997; 41: 785–9.
18. Carli F, Mayo N, Klubien K, Schricker T, Trudel J, Belliveau P.
Epidural analgesia enhances functional exercise capacity and
health-related quality of life after colonic surgery: results of a
randomized trial. Anesthesiology 2002; 97: 540–9.
19. Carr DB, Goudas LC. Acute pain. Lancet 1999; 353: 2051–8.
20. Dworkin RH, Backonja M, Rowbotham MC, et al. Advances in
neuropathic pain: diagnosis, mechanisms and treatment rec-
ommendations. Archives of Neurology 2003; 60: 1524–34.
21. Lee MH, Kim KY, Song JH, et al. Minimal volume of local anes-
thetic required for an ultrasound-guided SGB. Pain Medicine
2012; 13: 1381–8.
22. De Oliveira GS, Fitzgerald P Jr, Streicher LF, Marcus R-J, McCar-
thy RJ. Systemic lidocaine to improve postoperative quality of
recovery after ambulatory laparoscopic surgery. Anesthesia
and Analgesia 2012; 115: 262–7.
23. McGuirk S, Fahy C, Costi D, Cyna AM. Use of invasive placebos
in research on local anaesthetic interventions. Anaesthesia
2011; 66: 84–91.
24. Sites BD, Neal JM. Placebo or intervention? Is it all a sham?
Anaesthesia 2011; 66: 73–5.
960
Kumar et al. | Stellate ganglion block and postoperative pain relief
25. Haddox JD, Kettler RE. Stellate ganglion block: normal saline
as placebo. Anesthesiology 1987; 67: 832–4.
26. Wulf H, Maier C. Complications and side effects of stellate
ganglion blockade. Results of a questionnaire survey. Der
Anaesthesist 1992; 41: 146–51.
27. Hardy PAJ, Wells JCD. Extent of sympathetic blockade after
stellate ganglion block with bupivacaine. Pain 1989; 36:
193–6.
28. Guntamukkala M, Hardy PAJ. Spread of injectate after stellate
ganglion block in man: an anatomical study. British Journal of
Anaesthesia 1991; 66: 643–4.
29. Malmqvist EL-A, Bengtsson M, So €rensen J. Efficacy of stellate
ganglion block: a clinical study with bupivacaine. Regional
Anesthesia and Pain Medicine 1992; 17: 340–7.
30. van Eijs F, Geurts J, van Kleef M, et al. Predictors of pain
relieving response to sympathetic blockade in complex
regional pain syndrome type 1. Anesthesiology 2012; 116:
113–21.
31. Kastler A, Aubry S, Sailley N, et al. CT-guided stellate ganglion
blockade vs. radiofrequency neurolysis in the management of
refractory type I complex regional pain syndrome of the
upper limb. European Radiology 2013; 23: 1316–22.
32. Erdine S. Sympathetic blocks of the head and neck. In: Raj PP,
Lou L, Erdine S, Staats PS, Waldman SD, Racz G, Hammer M,
Niv D, Ruiz-Lopez R, Heavner JE, eds. Interventional Pain Man-
agement: Image-guided Procedures, 2nd edn. Philadelphia,
PA: Saunders Elsevier, 2008: 108–26.
© 2014 The Association of Anaesthetists of Great Britain and Ireland