Letters to the Editor

foot and toes. During an episode, whichlasted for one to five seconds, the patientcomplained of sharp pain in the left extrem-ities that could not be explained solely bythe muscle contraction causing the dystonicposturing. There were no alteration of con-sciousness, loss of sphincter control, orclonic movements in the limbs. The attackshad started without obvious precipitatingfactors 10 days before admission andoccurred with a frequency of 5-10 per day.The patient reported that at the age of 23

she had experienced a single similar attackof "painful spasm" in her left extremitiesthat had not been treated and had notrecurred.

Mild depressive symptoms had been pre-sent since the age of 60, being treated withfluvoxamine and benzodiazepines intermit-tently. There was no family history ofepilepsy or other neurological diseases.Her interictal general and neurological

examination was unremarkable. Bloodcounts and serum biochemistry were nor-mal. The routine EEG recording was nor-mal. After sleep deprivation the awake EEGdisclosed no abnormalities; during drowsi-ness and sleep an almost continuous focalepileptic activity was apparent in the rightfrontal-frontal sagittal region (figure).Computed tomography with and withoutcontrast and MRI of the brain were per-formed and showed mild symmetric brainatrophy without any evidence of focal struc-tural lesions or changes suggestive of ademyelinating disease. Treatment with car-bamazepine (200 mg three times daily) ledto dramatic cessation of attacks and disap-pearance of the epileptic activity on theEEG.

Painful paroxysmal dystonia was previ-ously described by different terms, such aspainful tonic seizures and painful tonicspasm.' It represents one of the distinctparoxysmal features of multiple sclerosis. 'Ephaptic activation of axons withindemyelinated plaques or other structurallesion may be responsible for both motorand sensory components of the paroxysms.'

Bennett2 attributed painful paroxysmaldystonia to the heterogeneous group ofparoxysmal dyskinesiae, stressing their non-epileptic character and emphasising theirimportant place in the differential diagnosisof epilepsy. The most characteristic featuresof these fits are their painful nature and typ-ical pattern of unilateral limb posturing.' 2

Both ictal and interictal EEGs were reportedto be normal or showed non-specificchanges' 2supporting the non-epileptic ori-gin of painful paroxysmal dystonia. By con-trast, dystonic posturing may occur inepilepsy, albeit it is usually not painful.3

Paroxysmal pain associated with motorphenomena may also be a rare manifesta-tion of epilepsy.45 It has been suggested thatepileptic pain originates from the contralat-eral rolandic area4 or superior and medialparts of the parietal lobe.5 The motor com-ponent of painful epileptic seizures wasreported to be represented by unilateralclonic or tonic-clonic convulsions with orwithout march, tonic deviation of the headand eyes, bilateral clonic movements of theextremities, generalised tonic-clonic convul-sions, transient motor weakness withouttonic or clonic phenomena, and "stiffness"of the arm or leg.45 Painful epilepticseizures with a characteristic pattern of uni-lateral posturing of the extremities meetingthe criteria of painful paroxysmal dystonia

to our knowledge have apparently neverbeen described.

Our patient had typical unprovokedpainful paroxysmal dystonia associated withfocal cortical epileptic activity in the con-tralateral frontal-frontal sagittal region. Thisapproximates to the premotor and supple-mentary motor region that, when activated,may result in postural changes.2

Although we have not obtained an ictalrecording, the association of the clinicalevents with a very active epileptic focus inthe EEG and the disappearance of bothafter carbamazepine treatment, suggest acausal relation between the electrographicphenomena and the bouts supporting theirepileptic nature. They may represent a dis-tinct type of painful paroxysmal dystonia ofcortical epileptic origin.

F BOKSTEINM Y NEUFELDP NISIPEANU

A D KORCZYNDepartment of Neurology,Tel Aviv Medical Center,

Sackler Faculty ofMedicine,Tel Aviv University, Israel

Correspondence to: Dr Felix Bokstein, Depart-ment of Neurology, Tel Aviv Souraski MedicalCenter, 6 Weizmann St, Tel Aviv, 64239 Israel.

1 Osterman PO, Westberg CE. Paroxysmalattacks in multiple sclerosis. Brain 1975;98:198-202.

2 Bennett DA. Paroxysmal dyskinesias. In:Resor SR, ed. The medical treatment ofepilepsy. New York: Marcel Dekker Inc,1992:219-21.

3 Korczyn AD, Inzelberg R. Dystonia. CurrOpin Neurol Neurosurg 1993;6:350-7.

4 Young GB, Blume WT. Painful epilepticseizures. Brain 1983;106: 537-54.

5 Wilkinson HA. Epileptic pain. An uncommonmanifestation with localizing value.Neurology 1973;23:518-20.

Treatment of spasmodic torticollis withintramuscular phenol injection

Phenol is a caustic agent that produces tis-sue destruction and has been used to weak-en muscle in patients with spasticity byeither nerve block, motor point block, orintramuscular neurolysis. ' 5 Local musclepain and tenderness but no systemic or longterm side effects have been reported aftersuch use. There is no report of the use ofintramuscular phenol in the treatment ofspasmodic torticollis. I have used this agentin patients with spasmodic torticollis whohave not responded well to intramuscularinjections of botulinum toxin A (BTX).Both patients gave informed consent to par-ticipate in the trial with the approval of theDuke University Investigational ReviewBoard.The muscles responsible for abnormal

head movements were determined by clini-cal examination and by EMG recordingsmade with a concentric EMG needle in thesternomastoid, splenius capitus, or otherneck muscles involved in head turning.Phenol solution (1% weight/volume USPphenol crystals in sterile aqueous solution)was injected into the involved muscles witha recording monopolar injection electrodeto determine that injections were made inmuscle active in the abnormal movement.Injections were made at two to six sites ineach muscle, in sites where motor unitaction potentials with sharply rising compo-nents were recorded, indicating that theneedle tip was close to muscle fibres thatwere activated during the abnormal move-

ment. Areas near major vessels or nerveswere avoided. After 18 hours or more, sub-sequent injections were given if improve-ment was incomplete as determined by thepatient's symptoms and by examination. Ateach visit, examination of strength in theinvolved muscles and functional assessmentof the patient's torticollis were recorded onvideotape for comparison with examinationsmade before phenol treatment.

Patient 1 is a 53 year old man who hashad torticollis since the age of 43. Medicaltreatment had produced no improvement.At the age of 48, he had shown moderateimprovement with BTX injections but, after18 months, the response to repeated injec-tions diminished despite increasing doses.Muscles no longer developed weakness,atrophy, or denervation changes on EMGtesting after BTX was injected. Antibody tobotulinum could not be detected in theserum. He underwent selective peripheraldenervation surgery when aged 50 and 51,with only mild improvement. Severalmonths later, BTX injections were repeatedwithout benefit.He had tonic, uncontrollable turning of

the head to the left with a phasic compo-nent and mild retrocollis before phenolinjection. He maintained his head in themidline with great difficulty while sitting.Standing and any attempt to talk, walk, oruse his hands produced immediate uncon-trollable head turning. He had constantposterior neck and intrascapular pain.

Within 18 hours after the initial injectionof 100 mg phenol into the left splenius capi-tis, splenius cervicis, and longissimus capi-tus muscles, he noted a fuller range ofmotion in the neck and mild improved headcontrol while walking. The injected muscleswere tender and oedematous but the neckand intrascapular pain was considerablyless. After subsequent injections he notedprogressive improvement. After receiving atotal dose of 500 mg of phenol over onemonth, he could walk, drive, or sit to eateven in public with only occasional involun-tary head turning. Manoeuvres that previ-ously had exacerbated the abnormal headmovements now were performed with littleor no difficulty. He estimated that pain wasreduced by 90%. His only side effects weremild erythema and tenderness over theinjected muscles lasting one to two daysafter each injection. The improvement wassustained for five months after the initialinjection. He then experienced gradualworsening of head control and some returnof pain. He has subsequently received phe-nol injections at intervals of six months tomaintain improvement.

Patient 2 is a 43 year old white malephysician who had onset of involuntaryhead turning to the left at the age of 33.Medical treatment produced intolerableside effects and no appreciable improve-ment. At the age of 41, he received BTXinjections every three to four months withmoderate improvement but symptoms ofhead turning and pain returned four to sixweeks after each injection. After each BTXinjection, he developed severe dysphagiathat persisted for two weeks. Before phenolwas injected, he had moderately severe tor-ticollis that was greatly exacerbated by pub-lic speaking and manual activities.

Eighteen hours after the initial injectionof 120 mg phenol into the left splenius capi-tis, splenius cervicis, and longissimus capi-tus muscles, he noted improvement in neck

258 on A

pril 26, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.58.2.258 on 1 February 1995. D

ownloaded from

Letters to the Editor

pain but little change in head control. After a__subsequent injections of 260 mg given over ..

two days, he noted considerable improve-ment in the ability to drive and carry outother manual tasks in his work. He gradedimprovement in head control and pain as50% and 80%, respectively, above baseline, A

compared with 50% and 50% after the pre-vious BTX injections. He had transientmild tendemess in the injected muscles.After the first month he experienced a slightdecline (10%) in function and pain control X 04^that remained constant for the next fourmonths and then gradually declined again.With subsequent injections of phenol atabout six monthly intervals, he has main-tained his maximal level of improvement.

These two patients had moderatelysevere spasmodic torticollis that hadimproved only partially after previous treat-ment. In the first patient, BTX initially pro-vided relief but became ineffective. In thesecond patient, BTX provided improve-ment but the side effect of dysphagia wasnearly intolerable. Within 18 hours afterphenol injections into cervical muscles,there was definite reduction of involuntarymovements and pain, with functionalimprovement. Improvement was greaterthan after all previous treatments and per-sisted for six and five months respectively, lIafter the initial series of phenol injections.The only side effect was transitory-namely,mild tendemess in the injected muscles.

In patients who become resistant torepeated injections of BTX, presumablydue to formation of antibody to the toxin, itwould be of great benefit to have an altema-tive treatment. Phenol may be of benefit inthis situation and has the additional advan-tage of being inexpensive. If the promisingresponse in these two patients is confirmedin a larger series of patients I am currently ...

studying, EMG guided intramuscular phe-nol injections may prove to be an effectivetreatment for some patients with spasmodictorticollis.

JANICE M MASSEYDivision of Neurology, Department of Medicine,

Duke University Medical Center, Durham,North Carolina, USAl

Correspondence to: Dr Janice M Massey, Box3403, Duke University Medical Center, Durham,North Carolina 27710, USA.

1DeLateur BJ. A new technique of intra-muscular phenol neurolysis. Arch Phys Med

L

Rehabil 1972;53:179-81.2 Easton JK, Ozel T, Halpem D. Intramuscular 5

neurolysis for spasticity in children. ArchPhys Med Rehabil 1979;60:155-8.

3 Garland DE, Lilling M, Keenan M4A.Percutaneous phenol blocks to motor poimtsof spastic forearm muscles in head-injuredadults. Arch Phys Med Rehabil 1 984;65:243-5.

4 Gibson II. Phenol block in the treatment ofspasticity. Gerontology 1987;33:327-30. . I)

5 Keenan MAE. Management of the spasticupper extremity in the neurologically (A) Sagittal Tl weightedMRI of the cervical spine before and (B) after gadolinium DTPAimpaired adult. Clin Orthop Rel Res 1988; showing enhancement in the spinal meninges. (C) Sagittal Ti weightedMRI of cervical and dorsal233:116-25. spine after gadolinium-DTPA four months after the initial scan. Multiple gadolinium ring

enhancing lesions are seen within the cervical cord. (D) Sagittal Tl weightedMRI of cervical cordshowingfurther changes in the intramedullary cavities eight months after the initial scan. Intensegadolinium enhancement continues in some loculi, others have resolved, and a new locule has

Tuberculous myelopathy: a serial MRI appeared at the cervicomedullary junction.study

In two definitive publications in 1969Wadia and Dastur delineated spinal menin- cord involvement in tuberculosis can be had been in the United Kingdom for onegitides with associated radiculomyelopathy studied with serial MRI. The current study year presented with a three month history ofwith particular reference to tuberculosis.'I involves a single case followed up with a nausea and vomiting associated withThe advent of MRI has meant that the series of MRI examinations over an eight increasingly severe headaches. He had nonature of the intramedullary lesions can for month period. It shows a previously unpub- previous history of or exposure to tubercu-the first time be defined during life. lished pattem of cord involvement. losis. On admission to hospital he had neckFurthermore, the natural history of spinal A 26 year old Samalian male refugee who stiffness only. Initial CSF examination

259 on A

pril 26, 2020 by guest. Protected by copyright.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.58.2.258 on 1 February 1995. D

ownloaded from