papr_331 435..

442

EVIDENCE-BASED MEDICINE
Evidence-Based Interventional Pain Medicine
according to Clinical Diagnoses

2. Cluster Headache

Maarten van Kleef, MD, PhD, FIPP*; Arno Lataster, MSc†;

Samer Narouze, MD, MSc, FIPP‡; Nagy Mekhail, MD, PhD, FIPP‡;
José W. Geurts, MSc*; Jan van Zundert, MD, PhD, FIPP*§
*Department of Anesthesiology and Pain Management, Maastricht University Medical
Centre, Maastricht, The Netherlands; †Department of Anatomy and Embryology, Maastricht
University, Maastricht, The Netherlands; ‡Pain Management Department, Cleveland Clinic,
Cleveland, Anesthesiology Institute, Ohio, U.S.A.; §Department of Anesthesiology and Pain
Management, Ziekenhuis Oost Limburg, Genk, Belgium

䊏 Abstract: Cluster headache is a strictly unilateral head- In conclusion, the primary treatment is medication.
ache that is associated with ipsilateral cranial autonomic Radiofrequency treatment of the ganglion pterygopalati-
symptoms and usually has a circadian and circannual pattern. num should be considered in patients who are resistant to
Prevalence is estimated at 0.5 to 1.0/1,000. The diagnosis of conservative pain therapy. In patients with cluster headache
cluster headache is made based on the patient’s case history. refractory to all other treatments, occipital nerve stimulation
There are two main clinical patterns of cluster headache: the may be considered, preferably within the context of a clinical
episodic and the chronic. Episodic is the most common study. 䊏
pattern of cluster headache. It occurs in periods lasting 7 days
to 1 year and is separated by at least a 1-month pain-free Key Words: cluster headache, headache, evidence-based
interval. The attacks in the chronic form occur for more than medicine, radiofrequency ablation
1 year without remission periods or with remission periods
lasting less than 1 month. INTRODUCTION
Conservative therapy consists of abortive and preventa-
This review on cluster headache is part of the series
tive remedies. Ergotamines and sumatriptan injections, sub-
“Evidence-Based Interventional Pain Medicine accord-
lingual ergotamine tartrate administration, and oxygen
inhalation are effective abortive therapies. Verapamil is an ing to Clinical Diagnoses.” Recommendations formu-
effective and the safest prophylactic remedy. When pharma- lated in this chapter are based on “Grading Strength of
cological and oxygen therapies fail, interventional pain Recommendations and Quality of Evidence in Clinical
treatment may be considered. The effectiveness of radiofre- Guidelines,” described by Guyatt et al.1 and adapted by
quency treatment of the ganglion pterygopalatinum and of van Kleef et al.2 in the editorial accompanying the first
occipital nerve stimulation is only evaluated in observational article of this series (Table 1).
studies, resulting in a 2 C+ recommendation.
The latest literature update was performed in July
Address correspondence and reprint requests to: Maarten van Kleef, 2009.
MD, PhD, FIPP, Maastricht University Medical Centre, Department of Anes- Cluster headache is a primary neurovascular head-
thesiology and Pain Management, PO Box 5800, 6202 AZ Maastricht, The
Netherlands. E-mail: [email protected].
ache. It is a strictly unilateral headache that is associated
DOI. 10.1111/j.1533-2500.2009.00331.x with cranial autonomic symptoms and usually has cir-
cadian and circannual patterns. A cluster headache is
© 2009 World Institute of Pain, 1530-7085/09/$15.00
characterized by the clustered nature of the attacks. The
Pain Practice, Volume 9, Issue 6, 2009 435–442 attacks can be provoked by vasodilators such as alcohol
436 • van kleef et al.

Table 1. Summary of Evidence Scores and Implications for Recommendation

Score Description Implication

1 A+ Effectiveness demonstrated in various RCTs of good quality. The benefits clearly outweigh risk and burdens
1 B+ One RCT or more RCTs with methodologic weaknesses, demonstrate effectiveness. The benefits clearly
outweigh risk and burdens Positive recommendation
2 B+ One or more RCTs with methodologic weaknesses, demonstrate effectiveness. Benefits closely balanced
with risk and burdens

2 B⫾ Multiple RCTs, with methodologic weaknesses, yield contradictory results better or worse than the control
treatment. Benefits closely balanced with risk and burdens, or uncertainty in the estimates of benefits,
Considered, preferably
risk and burdens.
study-related
2 C+ Effectiveness only demonstrated in observational studies. Given that there is no conclusive evidence of the
effect, benefits closely balanced with risk and burdens

0 There is no literature or there are case reports available, but these are insufficient to suggest effectiveness
Only study-related
and/or safety. These treatments should only be applied in relation to studies.

2 C- Observational studies indicate no or too short-lived effectiveness. Given that there is no positive clinical
effect, risk and burdens outweigh the benefit
2 B- One or more RCTs with methodologic weaknesses, or large observational studies that do not indicate any
superiority to the control treatment. Given that there is no positive clinical effect, risk and burdens Negative recommendation
outweigh the benefit
2 A- RCT of a good quality which does not exhibit any clinical effect. Given that there is no positive clinical
effect, risk and burdens outweigh the benefit

RCT, randomized controlled trial.

and nitroglycerine. Despite the low prevalence of cluster Table 2. Autonomic Characteristics of Cluster Headache
headaches, it has been shown that this affliction has
Ipsilateral to site of pain
a large socioeconomic impact. Almost 80% of the Lacrimation or conjunctival injection
patients report restrictions in daily activities, which, in Rhinorrhea or nasal congestion
Cranial and/or facial sweating
13%, were even present outside of attack periods.3 Miosis and/or ptosis
Noticeably, patients with cluster headaches have often Edema of the eyelid or orofacial tissues (including the gingiva and
palate)
endured a long course prior to diagnosis. Sometimes, Facial flushing or pallor
they have even undergone operations on the sinus Swelling around the eye and orofacial tissues (including the mouth)
Thermography determined “cold spot” at the site of pain (usually
maxillaris/septum nasi or had their teeth removed. The supraorbital)
prevalence of cluster headaches can be estimated at Systemic
approximately 0.5 to 1.0/1,000.4 The first attacks Bradycardia
Vertigo and ataxia
appear between the ages of 20 and 40 years. In contrast Syncope
with migraine, cluster headaches affect mainly men in a Hypertension
Increased gastrointestinal acid production
ratio of 5:1.5 There are indications that the risk for
cluster headache is notably higher in patients with a From Balasubramaniam and Klasser.6

family history of cluster headaches.6

During attacks, there is a high degree of restlessness,
and the patient often moves around, unable to remain
I. DIAGNOSIS still. The attacks are often of short (15 minutes) or
sometimes long duration (3 hours) and, strangely
I.A HISTORY enough, usually occur at night. The attacks appear to be
The attacks consist of a strictly unilateral, strong, pierc- associated with REM sleep.
ing pain around or behind the eye. The pain often occurs In the most typical form, there will be periods of
simultaneously with ipsilateral signs of autonomic dys- episodic headaches (clusters), lasting from a few weeks
regulation, whereby a red watering eye, nasal conges- to months, whereby the attack frequency can vary from
tion and/or rhinorrhea, and even miosis and/or ptosis one every 2 days to eight per day. After such a period, a
can occur. The autonomic characteristics of cluster patient may often experience one or more attack-free
headache are given in Table 2.6 Apart from miosis years before a new attack period occurs. In a small
and/or ptosis, which may persist between attacks, auto- number of patients, the attack pattern changes from
nomic features tend to cease with the pain.7 episodic to chronic, wherein daily attacks occur for
 2. Cluster Headache • 437

years and years. The chronic form of headache occurs in cluster headache such as nasopharyngeal carcinoma,
about 15% of the patients with cluster headache and is sphenoidal meningioma, carotid artery dissection,
therefore relatively rare. vertebral artery dissection, pituitary adenoma, or
aneurysm.6 Balasubramaniam and Klasser discuss
I.B PHYSICAL EXAMINATION “red flags,” which should be taken into account. These
The neurological examination usually does not reveal encompass a chronic unremitting headache, a lasting
any peculiarities in these patients. residual headache of a low intensity after cessation of
the cluster headache attack, minimal response to stan-
I.C ADDITIONAL TESTS dard treatment, the presence of abnormal physical find-
The diagnosis of cluster headache is made based on the ings such as vital signs, and abnormal cranial nerve
patient’s case history. Additional tests, blood tests, or examination with the exception of miosis and ptosis.6
X-rays to exclude other causes of the headache are
seldom necessary because cluster headache is distinctive. II. TREATMENT OPTIONS
I.D DIFFERENTIAL DIAGNOSIS II.A CONSERVATIVE MANAGEMENT
The International Headache Society’s International As with migraine, the therapy for cluster headache con-
Classification of Headache Disorders 2nd Edition sists of symptomatic/abortive treatment to alleviate
(ICHD-II) has stated diagnostic criteria for cluster head- symptoms and shorten the duration of the attacks,
ache and classified them into two forms: episodic and and preventative/prophylactic treatment to prevent the
chronic (Tables 3 and 4).8 attacks and reduce their number. Ergotamines and
Episodic cluster headaches occur in periods lasting sumatriptan injections are effective abortives for cluster
from 7 days to 1 year and are separated by at least a 1- headache, as for migraine. It is important to realize that
month pain-free interval. The attacks in the chronic form cluster headache attacks are usually episodic, and pro-
occur for more than 1 year without remission periods or phylactic medications like verapamil must be rapidly
with remission periods lasting less than 1 month. administered. Those treatments should be stopped once
The clinical picture of cluster headache is very spe- the cluster period is over. Although some prophylactic
cific. Yet, attention must be paid to the possible under- medications, such as tegretol, carbamazepine, and pro-
lying head and neck pathology that causes symptomatic panolol, have no effect on cluster headache, they are
frequently prescribed nonetheless.
Table 3. Diagnostic Criteria for Cluster Headache Symptomatic/Abortive Therapy
A. At least five attacks fulfilling B through D Abortive therapy of cluster headache that consists of
B. Severe or very severe unilateral orbital, supraorbital and/or temporal
pain lasting 15 to 180 minutes if untreated
oxygen inhalation, 100% oxygen 7 L/minute via a facial
C. Headache is accompanied by at least one of the following: mask, is one of the most effective methods and, by far,
1. Ipsilateral conjunctival injection and/or lacrimation
2. Ipsilateral nasal congestion and/or rhinorrhea the safest. In about 70% of cases, the attack is halted
3. Ipsilateral eyelid edema within 15 to 30 minutes. A recent Cochrane review
4. Ipsilateral forehead and facial sweating
5. Ipsilateral miosis and/or ptosis found limited evidence for this therapy.9
6. A sense of restlessness or agitation
D. Attacks have a frequency from one every other day to eight per day.
The effectiveness of subcutaneous sumatriptan on
E. Not attributed to another disorder cluster headache has been demonstrated in a double-
blind, placebo-controlled study. In 70% of the patients
treated with subcutaneous sumatriptan, 6 mg, the head-
Table 4. International Classification of Headache
Disorders 2nd Edition Criteria for Episodic and Chronic
ache disappeared within 15 minutes compared with
Cluster Headache 26% treated with placebo.10
The third abortive therapy is 2 to 4 mg of sublingual
Episodic cluster headache
A. All fulfilling criteria A through E of Table 3 ergotamine tartrate. Tablets of ergotamine are not, or
B. At least two cluster periods lasting from 7 to 365 days and insufficiently, effective most likely because of limited
separated by pain free remissions of > 1 month.
Chronic cluster headache absorption. Subcutaneous or intravenous administra-
A. All fulfilling criteria A through E of Table 3 tion is probably more effective. Currently, there are
B. Attacks recur for > 1 year without remission periods or with
remission periods lasting < 1 month no randomized studies investigating the effects of this
medication.
438 • van kleef et al.

Preventative/Prophylactic Therapy In open-label trials, a favorable effect is reported

Verapamil, 120 to 480 mg per day, is an effective and the for lithium by 80% of the patients. This effect has
safest prophylactic. The efficacy of verapamil is known been confirmed in a double-blind, placebo-controlled
from clinical practice, but there are few controlled study.11
studies. In an open-label study, 69% of patients reported
an improvement after the administration of verapamil.6 II.B INTERVENTIONAL MANAGEMENT
It is claimed that prednisone has a favorable effect
on cluster headache. Yet there is actually little evi- The Role of Radiofrequency (RF) Treatment of
dence to support this. Most of the studies are open label, the Ganglion Pterygopalatinum (PPG)
some studies are retrospective, and, at the moment, When pharmacological and oxygen therapies are
few reasons exist to prescribe prednisone for cluster ineffective, an interventional treatment can be carried
headaches.6 out (Figure 3). This interventional treatment involves

Figure 1. Anatomy of the ganglion

pterygopalatinum (ganglion spheno-
palatinum) within the fossa pterygo-
palatina.

N. ophtalmicus
Fossa ptygopalatina N. maxilaris
N. infraorbitalis Ganglion gasseri
Ganglion pterygopalatinum
N. mandibularis
(former sphenopalatinum)
Vidian nerve N. petrosus major
N. petrosus profundus Figure 2. Anatomy of the ganglion
pterygopalatinum. Illustration: Roger
N. carotis interna Trompert, Medical Art, www.medical-
art.nl.
 2. Cluster Headache • 439

blocking the ganglion pterygopalatinum (sphenopala- treatment of the PPG may improve episodic cluster
tinum (PPG) and should preferably be carried out at headache but not chronic cluster headache. However,
the beginning of a cluster period. recently, Narouze et al. reported favorable outcome
The PPG is located in the fossa pterygopalatina, after intractable chronic cluster headache as well. In this
which is a small “upside-down” pyramidal space, 2 cm retrospective study of 15 patients, they reported signifi-
high and 1 cm wide. The fossa pterygopalatina is cant improvement in both mean attack intensity and
located behind the posterior wall of the sinus maxilla- mean attack frequency for up to 18 months.16
ris and is bordered posteriorly by the medial plate After RF treatment of the PPG, postoperative
of the processus pterygoideus, superiorly by the sinus epistaxis, bleeding in the jaw and unintentional partial
sphenoidalis, and medially by the perpendicular plate lesion of the nervus maxillaris were observed. In one
of the os palatinum, and laterally communicates with report, nine out of the 10 patients experienced hypes-
the fossa infratemporalis.12,13 Superolaterally lies the thesia of the palatum.10,17
foramen rotundum with the exiting nervus maxillaris,
and inferomedially there is the vidian nerve (nervus
Occipital Nerve Stimulation (ONS)
petrosus major and nervus petrosus profundus) within
the canalis pterygoideus. The fossa pterygopalatina con- ONS in patients with refractory cluster headache has
tains the arteria maxillaris interna and its branches, been described in few case series.18–21 One systemic
the nervus maxillaris, and the PPG and its afferent and review of ONS for chronic headache (including cluster
efferent branches. The PPG is located posterior to the headache) has been found.22 This treatment appears to
concha nasalis media (middle turbinate) and is few mil- mainly decrease the intensity of the attacks. Noticeably,
limeters deep to the lateral nasal mucosa. It is suspended there is a relatively long (2 months or more) period of
from the nervus maxillaris by the nervi pterygopalatini; latency between the implantation of the electrode and a
inferiorly, it is connected to the nervus palatinus major clinical effect.
and minor, and posteriorly, it is connected to the vidian
nerve. Efferent branches of the PPG form the three rami II.C COMPLICATIONS OF INTERVENTIONAL
nasales posteriores ganglii pterygopalatini (posterior MANAGEMENT
nasal branches of the PPG) and the nervus pharyn- Total destruction of the PPG could result in dryness
geus13,14 (Figures 1 and 2). of the eyes. However, under normal conditions, the RF
The rationale for the RF treatment of the PPG in treatment only aims at a partial lesion of the ganglion.
cluster headache is influenced by the parasympathetic A possible complication is hypesthesia of the palatum
symptoms during the attack and perhaps by vasoactive molle, which generally disappears after 6 to 8 weeks.
substances such as calcitonin gene-related peptide.15 Another complication is nosebleeding and swelling of
In a retrospective analysis of patients with refractory the cheek as the result of a hematoma. A bothersome
cluster headache treated by RF treatment of the PPG, 56 complication is the accidental lesion of the nervus max-
patients with episodic and 10 patients with chronic illaris, which can occur when the technique is not pro-
cluster headache were followed up over a period of 12 perly carried out. A thorough understanding of the
to 70 months. In the group with episodic cluster head- anatomy allows the clinician to predict correct needle
ache, 60.7% experienced complete pain relief, while placement during RF treatment, according to the result
only 3 out of 10 patients with chronic cluster headache of the stimulation, and hence can reduce the incidence of
had the same result.13 The above report showed that RF complications (Table 5).23

Table 5. Different Possible Scenarios of Stimulation before Applying Radiofrequency Treatment of the Ganglion
Pterygopalatinum23

Location of Paresthesia Nerves Stimulated Location of Needle Tip Action Needed

Upper teeth and gums Maxillary branches Superolateral Redirect the needle; caudally and medially.
Hard palate Greater and lesser palatine nerves Anterior, lateral, caudal Redirect the needle; posteromedially and cephalad.
Root of the nose PPG efferents posterior lateral nasal nerves Correct needle placement None

PPG, ganglion pterygopalatinum.

440 • van kleef et al.

II.D EVIDENCE FOR INTERVENTIONAL A line is drawn on the skin over the fossa and the
MANAGEMENT introduction point is chosen just below the arcus zygoma-
A summary of the available evidence is given in Table 6. ticus (Figure 4). The skin is anesthetized and a 100-mm
radiofrequency electrode with a 2-mm active tip is slowly
introduced. This needle is carefully inserted in a super-
Table 6. Summary of Interventional Pain Treatments
ior and anterior direction using lateral fluoroscopy
Technique Score toward the anterosuperior point of the fossa pterygopa-
Radiofrequency treatment of the pterygopalatine ganglion 2 C+
latina (Figures 5 and 6). The C-arm is now placed in
Occipital nerve stimulation 2 C+ an anteroposterior position; the tip of the canula should
be lying just lateral to the nasal wall. The stylet is
removed, and a thermocouple RF-probe is placed. The
position of the electrode is confirmed by electrostimula-
III. RECOMMENDATIONS
tion using 50 Hz. It is important to use a 2-mm active tip,
In patients with therapy-resistant cluster headache, otherwise damage can occur to the nervus maxillaris
study-related RF treatment of the PPG can be considered. during the lesion. Generally, the patient feels paresthesia
With cluster headache, which is refractory to on the lateral side and back of the nose at a threshold
other treatment options, ONS may be considered, pref- of 0.4 V. There should be no paresthesia felt in the soft
erably in relation to a study in specialized centers palate or the upper jaw because this indicates the stimu-
(Table 6). lation of the nervus maxillaris or its branches (Table 5).
After a limited amount of local anesthesia (maximum
of 1 mL), a lesion is carried out for 60 s at 80°C, and this
III.A CLINICAL PRACTICE ALGORITHM lesion is repeated twice whereby the electrode is inserted
further.
Attackwise appearance of headaches
IV. SUMMARY
The diagnosis of cluster headache is based on the case
Red flags excluded: residual headache, abnormal vital signs, history and indications from the physical examination.
abnormal tests for the cranial nerves

Pharmacological treatment fails

Oxygen therapy fails

Consider radiofrequency (RF) therapy of

the pterygopalatine ganglion (PPG)

Refractory pain

Consider occipital nerve stimulation

(ONS)

Figure 3. Clinical practice algorithm for cluster headache.

III.B TECHNIQUES
The RF Treatment of the PPG Procedure
Figure 4. Radiofrequency treatment of the pterygopalatine
The patient is positioned lying on the back and the fossa ganglion: projection of the metal bar indicates the line over the
pterygopalatina is identified by using lateral fluoroscopy. fossa pterygopalatina.
 2. Cluster Headache • 441

Figure 5. Radiofrequency treatment of the ganglion pterygopa- Figure 6. Radiofrequency treatment of the pterygopalatine gan-
latinum: AP view. glion: lateral view. Needle high in the fossa pterygopalatina.

The primary treatment is medication, but, in patients 3. Jensen RM, Lyngberg A, Jensen RH. Burden of
who are therapy resistant, RF treatment of the PPG cluster headache. Cephalalgia. 2007;27:535–541.
should be considered. This treatment should be carried 4. Leroux E, Ducros A. Cluster headache. Orphanet J
out at the beginning of a cluster period. In patients with Rare Dis. 2008;3:20.
cluster headache refractory to all other treatments, ONS 5. Sjaastad O, Bakketeig LS. Cluster headache preva-
lence. Vaga study of headache epidemiology. Cephalalgia.
may be considered, preferably within a clinical study in
2003;23:528–533.
specialized centers.
6. Balasubramaniam R, Klasser GD. Trigeminal auto-
nomic cephalalgias. Part 1: cluster headache. Oral Surg Oral
ACKNOWLEDGEMENTS Med Oral Pathol Oral Radiol Endod. 2007;104:345–
The authors would like to thank Nicole van den Hecke 358.
7. Lance JW, Anthony M. Migrainous neuralgia or
for extensive literature search, editorial assistance, and
cluster headache? J Neurol Sci. 1971;13:401–414.
suggestions regarding the article. This review was ini-
8. Silberstein SD, Olesen J, Bousser MG, et al. The
tially based on practice guidelines written by Dutch and international classification of headache disorders. 2nd ed.
Flemish (Belgian) experts that are assembled in a hand- (ichd-ii)—revision of criteria for 8.2 medication-overuse head-
book for the Dutch-speaking pain physicians. After ache. [erratum appears in cephalalgia. 2006 mar;26(3):360].
translation, the article was updated and edited in coop- Cephalalgia. 2005;25:460–465.
eration with U.S./International pain specialists. 9. Schnabel A, Bennet M, Schuster F, et al. Hyper-
bzw. Normobare sauerstofftherapie zur behandlung von
REFERENCES migrane und clusterkopfschmerzen (hyper- or normobaric
oxygen therapy to treat migraine and cluster headache
1. Guyatt G, Gutterman D, Baumann MH, et al. pain). Cochrane review. Schmerz. 2008;22:129–132, 134–
Grading strength of recommendations and quality of evidence 126.
in clinical guidelines: report from an American college of chest 10. Group TSCHS. Treatment of acute cluster headache with
physicians task force. Chest. 2006;129:174–181. sumatriptan. N Engl J Med. 1991;325:322–326.
2. van Kleef M, Mekhail N, van Zundert J. Evidence- 11. Steiner TJ, Hering R, Couturier EG, et al. Double-
based guidelines for interventional pain medicine according to blind placebo-controlled trial of lithium in episodic cluster
clinical diagnoses. Pain Pract. 2009;9:247–251. headache. Cephalalgia. 1997;17:673–675.
442 • van kleef et al.

12. Salar G, Ori C, Iob I, et al. Percutaneous thermoco- 18. Magis D, Allena M, Bolla M, et al. Occipital
agulation for sphenopalatine ganglion neuralgia. Acta Neuro- nerve stimulation for drug-resistant chronic cluster head-
chir (Wien). 1987;84:24–28. ache: a prospective pilot study. Lancet Neurol. 2007;6:314–
13. Sanders M, Zuurmond WW. Efficacy of sphenopa- 321.
latine ganglion blockade in 66 patients suffering from cluster 19. Schwedt TJ, Dodick DW, Trentman TL, et al.
headache: a 12- to 70-month follow-up evaluation. J Neuro- Occipital nerve stimulation for chronic cluster headache and
surg. 1997;87:876–880. hemicrania continua: pain relief and persistence of autonomic
14. Day M. Neurolysis of the trigeminal and sphenopa- features. Cephalalgia. 2006;26:1025–1027.
latine ganglions. Pain Pract. 2001;1:171–182. 20. Burns B, Watkins L, Goadsby PJ. Treatment of medi-
15. Edvinsson L. Blockade of CGRP receptors in the cally intractable cluster headache by occipital nerve stimula-
intracranial vasculature: a new target in the treatment of head- tion: lng-term follow-up of eight patients. [see comment].
ache. Cephalalgia. 2004;24:611–622. Lancet. 2007;369:1099–1106.
16. Narouze S, Kapural L, Casanova J, et al. Sphenopa- 21. Burns B, Watkins L., Goadsby PJ. Treatment of
latine ganglion radiofrequency ablation for the management intractable chronic cluster headache by occipital nerve stimu-
of chronic cluster headache. Headache. 2009;49:571–577. lation in 14 patients. Neurology. 2009;72:341–345.
17. Filippini-de Moor G, Barendse G, van Kleef M, 22. Jasper JF, Hayek SM. Implanted occipital nerve
et al. Retrospective analysis of radiofrequency lesions of the stimulators. Pain Physician. 2008;11:187–200.
sphenopalatine ganglion in the treatment of 19 cluster head- 23. Narouze S. Complications of head and neck proce-
ache patients. Pain Clin. 1999;11:285–292. dures. Tech Reg Anesth Pain Manag. 2007;11:171–177.