a

SCIENTIFIC ARTICLE

Typical Brachial Neuritis (Parsonage-Turner

Syndrome)With Hourglass-Like Constrictions in the

Affected Nerves

Yong-wei Pan, MD, PhD, ShufengWang, MD, PhD, Guanglei Tian, MD, Chun Li, MD,Wen Tian, MD,Mengmeng Tian, MD, PhD

Purpose To report on 5 patients who had acute brachial neuritis (Parsonage-Turner syn-drome) with hourglass-like constriction in the affected nerves.

Methods We retrospectively reviewed 5 patients who were treated in our department fromDecember 2003 to December 2008. Acute, intense pain around the shoulder girdle and upper armwas the first symptom and was followed by muscle weakness and atrophy. Clinical and EMGexaminations showed involvement of 2 or more nerves in the affected extremity. Those severelyaffected nerves that had no response to conservative treatment were explored, and an hourglass-like constriction was identified. Neurolysis was performed at the sites of constrictions in 2 radialnerves and 1 median nerve. The constricted portion was resected, and direct coaptation wasperformed in 1 radial nerve and 1 musculocutaneous nerve. The constricted portion was resected,and nerve graft was performed in 2 radial nerves and 1 median nerve.

Results All patients were followed up for 24 to 84 months after surgery. Of 3 nerves treatedwith external neurolysis, all attained full recovery. Of 2 nerves treated with resection andneurorrhaphy, 1 attained full recovery, and the other had an incomplete recovery. Of 3 nervestreated with resection and nerve graft, 1 (4-cm nerve graft) attained full recovery, and 2(4-cm and 13-cm nerve graft, respectively) had incomplete recovery.

Conclusions The site of nerve lesion of brachial neuritis was not necessarily within thebrachial plexus. Our finding of hourglass-like constrictions in individual peripheral nervessuggest that multifocal involvement of terminal branch lesions may underlie the complexpatterns of paralysis often encountered clinically. (J Hand Surg 2011;36A:1197–1203.Copyright © 2011 by the American Society for Surgery of the Hand. All rights reserved.)

Type of study/level of evidence Prognostic IV.

Key words Brachial neuritis, hourglass-like constriction, neuralgic amyotrophy, Parsonage-Turner syndrome.

alm

ACUTE BRACHIAL NEURITIS, also known as neural-

gic amyotrophy and Parsonage-Turner syn-drome, is an uncommon clinical problem. This

bnormality typically has a characteristic presentation:

From the Department of Hand Surgery, Beijing Jishuitan Hospital, Beijing, China; Department of Pathol-ogy, Beijing Jishuitan Hospital, Beijing, China.

Received for publication January 26, 2011; accepted in revised form March 23, 2011.

The authors thank Michael J. Epstein, MD, Professor Emeritus, Department of Orthopaedic Surgery,

Department of Surgery, Baylor College of Medicine, for his editorial assistance.

©

cute onset of severe shoulder and/or arm pain, fol-owed shortly thereafter by weakness and atrophy ofuscles in the shoulder girdle and arm.1,2 Although the

syndrome is now well recognized, the etiology remains

No benefits in any form have been received or will be received related directly or indirectly to thesubject of this article.

Corresponding author: Yong-wei Pan, MD, PhD, Beijing Jishuitan Hospital, Department of HandSurgery, 31# Xinjiekou Dongjie, Beijing, China, 100035; e-mail: panyongwei@yahoo.com.cn.

0363-5023/11/36A07-0011$36.00/0

doi:10.1016/j.jhsa.2011.03.041

ASSH � Published by Elsevier, Inc. All rights reserved. � 1197

1198 BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION

unknown, and the site of lesions is still debated.3 Wehave recently encountered 5 patients who had acutebrachial neuritis. The patients were treated conserva-tively. Some involved nerves recovered gradually.However, palsy persisted in some severely affectednerves, signs of clinical recovery were absent, and

TABLE 1. Preoperative Status

Case Age Gender Side

1 21 F L Left shoulder

2 26 M L Whole upper

3 45 F R Around the ri

4 29 M L Both shoulde

5 38 M L Left shoulderand radial a

RN, radial nerve; MN, median nerve; MCN, musculocutaneous nerve

TABLE 2. Finding at Surgery and Final Results

CasePeriods from Onset

to Surgery (mo) Nerves Explored

1 4 Left MN (left RN was explored ianother hospital)

2 4 Left RNLeft MN

3 8 Right MCNRight RN(RN had been previously explor

another hospital)

4 2 Left RN

5 10 Left RN

RN, radial nerve; MN, median nerve; MCN, musculocutaneous nerve

EMG examination showed complete denervation

JHS �Vol A

of the muscles innervated by these nerves. Todefine the etiology, the nerves were explored, andan hourglass-like constriction without any externalcompression in the affected nerves was discovered.Our findings suggest that hourglass-like constric-tive lesion in the nerve might be a pathologic basis

Pain at Onset Cause of Pain

upper arm None

None

houlder girdle None

Flu-like symptoms

with radiation into lateral upper armt of forearm

Strenuous work

ulnar nerve; AN, axillary nerve; SN, suprascapular nerve.

Appearance of NerveNumber of Constricted

Lesions

Edematous and hardened Many, along the nervefascicles

Edematous and hardened RN: 4MN: 1

MCN: edematous andhardened

RN: severely scarred

MCN: 1

Edematous and hardened 2

Edematous and hardened 1

and

limb

ght s

rs

painspec

; UN,

n

ed in

.

of brachial neuritis.

, July

BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION 1199

MATERIALS AND METHODSAll patients were evaluated at our center between De-cember 2003 and December 2008. The criteria for in-clusion for each patient were a sudden onset of severepain in or about the shoulder girdle, without a systemicdisorder that might affect the musculoskeletal system,

TABLE 1. (Continued)

Interval betweenPain and Palsy (d)

Duration ofPain (d) Sensory Disturbance

15 60 The dorsal of thumbweb space

The thumb, index, andmiddle fingers

M

3 7 The thumb, index, andmiddle fingers

T

1 4 The dorsal of thumb webspace

T

T

4 4 Normal D

20 180 Normal M

TABLE 2. (Continued)

Surgical Method Histologic ExaminationT

Foll

RN: neurolysisMN: nerve graft (13 cm)

N/A

Neurolysis N/A

RN: nerve graft (4 cm)MCN: Neurorrhaphy

Edematous, lymphocytes andsome neutrophilsinfiltrated the walls of thesmall feeding arteries inthe perineurium

Nerve graft (4 cm) A slightly thickenedperineurium, mildperivascular lymphoid cellinfiltration, and obviouscentral edema

Neurorrhaphy Fibrosis, edema, andscattered lymphocyteinfiltration

and an onset of weakness in the involved limb within a

JHS �Vol A

few weeks after the onset of the pain. Physical and/orEMG examination demonstrated findings that wereconsistent with brachial neuritis.

Laboratory investigations included white blood cellcount with differential, blood urea nitrogen, liver func-tion tests, and antinuclear antibody titer. All tests were

otor Strength (Clinical) Muscle Involved (EMG)

es innervated by RN (M0)MN (M0)

Muscles innervated by UN, AN,SN, and MCN

uscles innervated by MNRN (M0)

Muscles innervated by MN,RN, UN, AN, SN, and MCN

uscles innervated by MCNRN (M0)

erratus anterior muscle (M3)

Biceps and serratus anterior.Muscles innervated by RN,MN, UN, and ANs

d (M3) muscles innervatedN (M0)

N/A

es innervated by RN (MO) Muscles innervated by RN,MN, UN, and MCN

ofp (m)

Results

Motor Recovery Sensory Recovery

RN: full recoveryMN: poor recovery

RN: full recoveryMN: 2-point

discrimination �12 mm

Full recovery RN: full recoveryMN: 2-point

discrimination � 6 mm

RN: incomplete recoveryMCN: full recovery

RN: some degree ofsensory loss

MCN: some degree ofsensory loss

Full recovery Some degree of sensoryloss

Incomplete recovery Some degree of sensoryloss

M

uscland

he mand

he mand

he s

eltoiby R

uscl

imeow-u

24

25

30

77

84

normal unless specifically stated.

, July

1200 BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION

Needle EMG was performed in muscles representingall the suspected nerves that were involved. The distri-bution of neuropathic features was recorded.

A diagnosis of brachial neuritis was made in allpatients. They were treated conservatively with vitaminB12, physiotherapy, and rehabilitative exercises. Somemildly involved nerves recovered gradually. However,in some severely affected nerves, palsy persisted, andthere was no clinical or EMG sign of recovery afterconservative treatment for 2 to 11 months. Surgicalexploration was then performed.

On the basis of careful analysis of clinical findingsand EMG examination, we found that neurologic defi-cits were more compatible with focal lesions withinindividual peripheral nerves rather than within the bra-chial plexus. Therefore, we decided to explore individ-ual peripheral nerves. The types of treatment for hour-glass-like constrictive lesion in the nerve were chosenon the basis of the degrees of constriction found atsurgical exploration, the results of intraoperative nervestimulation, and personal experience of the surgeon.

RESULTSClinical signs and symptoms of the 5 patients are shownin Table 1. All 5 patients had a typical presentation ofsudden onset of shoulder pain, which subsided in 4 daysto 4 months, followed by a flaccid paralysis of musclesin the shoulder girdle and arm.

Physical and EMG examination at the time of pre-sentation revealed multiple nerve involvement. Therewere marked differences in grade of weakness anddenervation between muscles innervated by differentnerves. We found complete paralysis and denervationin some muscles and moderate to mild paralysis anddenervation in muscles supplied by other nerves. Sub-clinical mild denervation was also detected in somemuscles.

After conservative treatment for 2 to 11 months, 8severly affected nerves (5 radial nerves, 2 mediannerves, and 1 musculocutaneous nerve) had no clinicalor EMG sign of recovery. We explored these nerveswith the findings shown in Table 2. There were noconstriction bands or muscles compressing the nerves.An hourglass-like constriction in the involved nervedwas present (Figs. 1, 2A, 2B, 3A, 4A, 5A). The areas ofstricture were not consistent with overlying musclegroups or sites of known compression. The nerves werefound to be grossly edematous and thickened for alength of 3 cm to 5 cm beyond the constrictions. Slightadhesions were found around the constrictions.

Neurolyses were performed at the sites of constric-

tions in 2 radial nerves and 1 median nerve. The con-

JHS �Vol A

stricted portion was resected and the ends directly co-apted in 1 radial nerve and 1 musculocutaneous nerve.The constricted portion was resected and nerve graftperformed in 2 radial nerves and 1 median nerve. Theresected portions were examined histologically; the re-sults are shown in Table 2 and Figures 3B, 4B, and 5B.

All patients were followed up for 24 to 84 monthsafter surgery; the final results are shown in Table 2. Of3 nerves treated with external neurolysis, all obtainedrecovery of at least grade 4 muscle power in affectedmuscle groups using the Medical Research Councilgrading system. Of 5 nerves treated with resection andneurorrhaphy or nerve graft, 2 obtained full recovery,and 3 had incomplete recovery (grade 0–3). The shoul-ders of all patients had full movement, with M4–5muscle strength of the shoulder girdle muscles. Therecovery of sensation is shown in Table 2. The initialpain of all patients disappeared completely, except oc-casional, uncomfortable feeling in the periscapular areaor upper arm.

DISCUSSIONRecently, intraoperative cases of hourglass-like nerveconstriction lesions without any visible source of exter-nal compression have been reported. The involvednerves have included the main trunk of the radial nerveand posterior interosseous nerve,4–9 the main trunk ofthe median nerve and anterior interosseous nerve,9–12

the axillary nerve,13,14 and the suprascapular nerve.14

FIGURE 1: Finding at surgery in case 1. A 30-cm segment ofthe median nerve above the elbow was swollen and hard. Afterthe epineurium was dissected, intra-epineurial constrictions ofnerve fascicles were evident. Constrictions occurred in eachfascicle of the nerve trunk at multiple levels. Some constrictionswere mild; the others were so severe that no nerve fascicles werefound with loupe visualization.

Most of these reported cases presented as an involve-

, July

o the

he sm

BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION 1201

ment of an isolated peripheral nerve, and to the best ofour knowledge, only 3 cases published in the Englishliterature had multiple nerve involvement.9,14,15 It iswell accepted that this phenomenon is not caused byexternal compression. However, the etiology of thisphenomenon is still unclear.9,11

In our series, all 5 patients satisfied the clinical andEMG criteria of brachial neuritis1,2 and were thought tobe typical cases of brachial neuritis. Surgical explora-tion of severely involved nerves in all 5 cases showedhourglass-like constriction in the fascicular or maintrunk of the nerve, similar to lesions that have beenpreviously described. Therefore, we consider that thisremarkable and unexpected nerve pathology could bethe basis of lesions in brachial neuritis.

Brachial neuritis is a misleading name because the

FIGURE 2: Finding at surgery in case 2. A The radial nerve alike constrictions 9 cm and 10 cm proximal to the lateral ephard, and a severe hourglass-like constriction 10 cm proximal t

FIGURE 3: Finding in case 3. A The musculocutaneous nerve wconstriction was evident. B Pathological examination of thelymphocytes and some neutrophils had infiltrated the walls of t

brachial plexus might not be the site of pathology.

JHS �Vol A

Detailed clinical and EMG examinations of patientsoften indicate lesions of individual nerves or nervebranches, many of which are not compatible with abrachial plexus localization.3,16–18 England et al3 be-lieved that patients with neuralgic amyotrophy had apattern of neurologic deficits that was most compatiblewith focal lesions within nerve fascicles or individualperipheral nerves, rather than within the brachialplexus. They were dependent on clinical and electro-physiologic analysis to locate the site of nerve lesionwithout surgical observations. The clinical presentationand course of our 5 cases were similar to those ofEngland’s cases. Surgical exploration of severely af-fected nerves showed them to be swollen, and hour-glass-like constrictions were evident. In case 3, weexplored the brachial plexus, which was macroscopi-

spiral groove was edematous and hard. Two severe hourglass-yle were evident. B The median nerve was also swollen andmedial epicondyle was noted.

wollen and thickened on palpation, and a severe hourglass-likericted nerve revealed that the nerve was edematous, and theall feeding arteries in the perineurium.

t theicond

as sconst

cally normal, whereas we discovered a severe constric-

, July

scula

1202 BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION

tion in the musculocutaneous nerve. Our finding sug-gested that the basic abnormality of brachial neuritiswas this hourglass-like nerve constriction, and it pro-vided evidence suggesting that multifocal involvementof terminal branch lesions might underlie the complexpatterns of paralysis that are often encountered clini-cally.

The etiology of this condition remains unclear, andreports of pathological study are scant. Suarez et al19

reported brachial plexus biopsy findings from 4 patientswith brachial plexus neuropathy. There were prominentcollections of inflammatory cells within the brachialplexus. The authors therefore suggested that brachialneuropathies have an inflammatory–immune pathogen-esis. However, our pathologic findings revealed that,though present, lymphocyte infiltration was limited.

FIGURE 4: Finding in case 4. A The radial nerve around the srevealed 2 severe hourglass-like constrictions 8 cm and 9 cmspecimen showed a slightly thickened perineurium, mild periva

FIGURE 5: Finding in case 5. A The radial nerve around the sconstriction was discovered 5 cm proximal to the lateral epscattered lymphocyte infiltration.

Suarez et al performed a brachial plexus biopsy from

JHS �Vol A

patients with brachial plexus neuropathy. However,many authors regard brachial neuritis as a form ofmononeuropathy multiplex, involving individual nervesor nerve branches, rather than the brachial plexus.3,16

Many cases might have been inappropriately diagnosedbecause biopsies might have been taken from an unaf-fected portion of the nerve.

Many authors believe that brachial neuritis is a self-limited disease, and nonsurgical treatment is the ac-cepted treatment.1,2 However, recovery from this dis-order can be quite protracted, and complete restorationof strength is not always achieved.1,20 van Alfen et al1

reported that the majority of patients in their 246 casesexhibited persisting functional deficits after an averagefollow-up of more than 6 years. In Misamore’s20 series,3 of 7 patients had some residual deficit at long-term

groove appeared slightly edematous and hard. Epineurectomymal to the lateral epicondyle. B Pathologic examination of ther lymphoid cell infiltration, and central edema.

groove appeared edematous and hard. A severe hourglass-likeyle. B Histologic examination revealed fibrosis, edema, and

piralproxi

piralicond

evaluation. The reasons for absent recovery remain

, July

BRACHIAL NEURITIS WITH HOURGLASS-LIKE CONSTRICTION 1203

unknown. England16 speculated that patients with se-vere nerve injuries or involvement of nerves with longregenerative distance might have poor recovery. Ourfindings at surgical exploration confirmed the specula-tion of England. An hourglass-like constriction wasdiscovered in the fasciculus or main trunk of the nerve,and some constricted lesions were so severe that thefascicules appeared to be completely ruptured. It isquestionable whether nerve regeneration can be ex-pected in such situations. We speculate that the sameseverely constricted lesion might have existed in vanAlfen’s1 and Misamore’s20 patients who had poor re-covery.

On the basis of poor clinical recovery in some casesreported in the literature and our own surgical findings,we consider that the strategy for treating brachial neu-ritis should be reconsidered. Surgical interventionshould be carried out in cases that do not respond toconservative treatment after several months. Nagano11

recommended that exploration of the nerve be offeredto patients who did not show any signs of recovery by3 months after onset. He recommended only interfas-cicular neurolysis and believed that nerve grafting wasunnecessary. He believed that more fibers would regen-erate after neurolysis. In our present study, 3 of 8 nerveshad incomplete recovery, and all the resection caseswere the ones that did not regain function. Of them, 1patient was treated with neurorrhaphy 11 months afterthe onset of the symptoms, 1 was treated with nervegraft 8 months after the onset of symptoms, and 1 wastreated with a 13-cm nerve graft. We believe that thesepatients received treatment that was too late or tooaggressive. Case 1 was a comparison of varying treat-ment methods at surgery. It seemed that the severity ofconstrictive lesions in both the left radial nerve and themedian nerve was similar, so the radial nerve wastreated with neurolysis, whereas the median nerve wastreated with resection and a 13-cm nerve graft. Theradial nerve had a full recovery, whereas the mediannerve had a poor recovery. Therefore, we agreed withNagano’s opinion that the period of “wait and see”should not exceed 3 months, and neurorrhaphy or nervegrafting should be avoided. However, we are reluctantto draw any firm conclusions on the basis of such a

small number of cases studied retrospectively.

JHS �Vol A

REFERENCES1. van Alfen N, van Engelen BGM. The clinical spectrum of neuralgic

amyotrophy in 246 cases. Brain 2006;129:438–450.2. Sathasivam S, Lecky B, Manohar R, Selvan A. Neuralgic amyotro-

phy. J Bone Joint Surg 2008;90B550–553.3. England JD, Sumner AJ. Neuralgic amyotrophy: an increasingly

diverse entity. Muscle Nerve 1987;10:60–68.4. Yongwei P, Guanglei T, Jianing W, Shuhuan W, Qingtai L, Wen T.

Nontraumatic paralysis of the radial nerve with multiple constric-tions. J Hand Surg 2003;28A:199–205.

5. Yamamoto S, Nagano A, Mikami Y, Tajiri Y. Multiple constrictionsof the radial nerve without external compression. J Hand Surg2000;25A:134–137.

6. Vastamaki M. Prompt interfascicular neurolysis for the successfultreatment of hourglass-like fascicular nerve compression. Scand JReconstr Surg Hand Surg 2002;36:122–124.

7. Kotani H, Miki T, Senzoku F, Nakagawa Y, Ueo T. Posteriorinterosseous nerve paralysis with multiple constrictions. J Hand Surg1995;20A:15–17.

8. Hashizume H, Inoue H, Nagashima K, Hamaya K. Posterior in-terosseous nerve paralysis related to focal radial nerve constrictionsecondary to vasculitis. J Hand Surg 1993;18B:757–760.

9. Omura T, Nagano A, Murata H, Takahashi M, Ogihara H, Omura K.Simultaneous anterior and posterior interosseous nerve paralysiswith several hourglass-like fascicular constrictions in both nerves.J Hand Surg 2001;26A:1088–1092.

10. Yasunaga H, Shiroishi T, Ohta K, Matsunaga H, Ota Y. Fasciculartorsion in the median nerve within the distal third of the upper arm:three cases of nontraumatic anterior interosseous nerve palsy. J HandSurg 2003;28A:206–211.

11. Nagano A. Spontaneous anterior interosseous nerve palsy. J BoneJoint Surg 2003;85B:313–318.

12. Nagano A, Shibata K, Tokimura H, Yamamoto S, Tajiri Y. Sponta-neous anterior interosseous nerve palsy with hourglass-like fascicu-lar constriction within the main trunk of the median nerve. J HandSurg 1996;21A:266–270.

13. Oberlin C, Shafi M, Diverres J, Silberman O, Adle H, Belkheyar Z.Hourglass-like constriction of the axillary nerve: report of two pa-tients. J Hand Surg 2006;31A:1100–1104.

14. Vigasio A, Marcoccio I. Homolateral hourglass-like constrictions ofthe axillary and suprascapular nerves: case report. J Hand Surg2009;34A:1815–1820.

15. Cohen BE, Cukier J. Simultaneous posterior and anterior interosse-ous nerve syndromes. J Hand Surg 1982;7:398–400.

16. England JD. The variations of neuralgic amyotrophy. Muscle Nerve1999;22:435–436.

17. Watson BV, Rose-Innes A, Engstrom JW, Brown JD. Isolated bra-chialis wasting: an unusual presentation of neuralgic amyotrophy.Muscle Nerve 2001;24:1699–1702.

18. Cruz-Martínez A, Barrio M, Arpa J. Neuralgic amyotrophy: variableexpression in 40 patients. J Peripher Nerv Syst 2002;7:198–204.

19. Suarez GA, Giannini C, Bosch EP, Barohn RJ, Wodak J, Ebeling P,et al. Immune brachial plexus neuropathy: suggestive evidence for aninflammatory-immune pathogenesis. Neurology 1996;46:559–561.

20. Misamore GW, Lehman DE. Parsonage-turner syndrome (acute bra-

chial neuritis). J Bone Joint Surg 1996;78A:1045–1048.

, July