Journal ofNeurology, Neurosurgery, and Psychiatry 1995;58:649-654 649

Journal of J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from NEUROLOGY NEUROSURGERY & PSYCHIATRY

Editorial

The cystic

The spinal cord with cystic cavities and spaces within it is lum. There is a likelihood, however, that free fluid truly an intriguing sight. These cavities, when they are exchange occurs across the walls of the cystic cord in symptomatic, are an exclusively surgical problem.' spinal tumours. This is suggested by the speed with Modem surgery can transect the cord, make apertures in which intrathecal contrast material enters the cystic syrinxes, cut off the filum terminale, block up holes, open cavities as seen on postmyelographic CT. out subarachnoid spaces, and drain CSF from ventricles, subarachnoid spaces, or the syrinxes themselves to many other sites with adjustable control of pressure differen- tials. Surgery done in a timely way, and done well, may The cord: achieve enormous success in preventing crippling disor- cystic spinal classification according to associated conditions der and pain. Classification % Understanding of the nature of any phenomenon may Hindbrain related 72 be aided by classification. Because the Hindbrain herniation physical causation Idiopathic herniation (Chiari type 1) 32 of intracord cystic cavities is ill understood, the classifica- Secondary to birth injury 39 tion I propose is based on the Secondary to tumours 1-2 existence of associated Bony or meningeal tumours of the posterior fossa lesions. These associated anomalies are therefore not Tumours forming the hindbrain hernia causes the Intrinsic brain tumours above the lower fourth necessarily of cysts, but the associations may Secondary to bony abnormality suggest causes (table). Basilar invagination (idiopathic or birth injury) 10 Microscopy of the cavities mostly shows gliosis of the Sclerosteosis lining with the direction of fibres mostly annular. The Rickets Acro-osetolysis http://jnnp.bmj.com/ septations are not histologically do not Osteogenesis imperfecta intriguing. They Associated with hydrocephalus 10 always contain vessels in their edges, and there is no Intracranial arachnoid pouches 2 evidence that they contain fibres of the Dandy-Walker cysts intraspinally Early onset hydrocephalus (aqueduct stenosis) decussating spinothalamic tracts. Meningeal fibrosis of cisterna magna The fluid seems to Birth injury related 9 develop the cavities by cutting Postinflammatory through sites of structural weakness.23 Associated lesions Post-traumatic (postnatal) such as hindbrain herniation may be associated with the Infections Non-hindbrain related cases 22 presence of excess fluid inside the cervical cord before a Spinal tumours 1 on September 28, 2021 by guest. Protected copyright. clear cavity Intramedullary 1 develops. Cysts wholly or partly within the tumour The location and lining of the cavities has provoked Cysts outside the tumour extending in the cord discussion relating to whether or not the Extramedullary intradural tumours central canal is Extradural tumours, including disc disease involved. Nomenclature has varied and the suggestion Meningeal fibrosis has been made that Idiopathic "hydromyelia" might describe the Postinflammatory cavities when lined with ependyma. Human cases with Pyogenic meningitis intrauterine hydrocephalus and Epidural abscess 13 hydromyelia, usually Tuberculous meningitis associated with spina bifida, may have the cavities partly Myodil (Pantopaque) lined with ependyma but in most Post-traumatic human cases the cen- Secondary to spinal bony deformities 11 tral canal and the ependyma are irrelevant. The fluid Post-traumatic tracks readily either within or outside Tuberculous bone disease the canal. The term Idiopathic scoliosis "hydrosyringomyelia" has been suggested. Syringomyelia Dysraphic lesions has the advantage of familiarity, however, and authors With hydrocephalus 2 With hindbrain hernia 1 who prefer to use other terms should define them. With both the above 4 Tumours may be associated with syringomyelia cavi- With neither of the above 7 ties. These are usually intramedullary tumours and in Unknown cause (no associated conditions) some of them, notably haemangioblastomas, it seems Percentages are from a database of syringomyelia and related diseases. Diagnosis with no percentage given have an incidence of less than 1%. They likely that the cystic cord is due to secretion of proteinous do not add up to 100% because of intersections-for example, post-traumatic fluid such as may be thought responsible for similar cysts syringomyelia is associated with meningeal fibrosis and bony deformity. Some factors such as head injury are difficult to interpret. Others such as hydro- that are found with haemangioblastomas in the cerebel- cephalus, spina bifida occulta, or basilar impression are matters of degree. 650 Williams J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from A further class of difficult cases occurs in association equilibrium at work in the established syringomyelia with with spina bifida.4 Hindbrain hemiation is not a part of some substances, certainly including water, moving freely the dysraphic process but consequent on it, and when it both ways across syrinx walls. is present the cases are usually variants of hindbrain One variety of syringomyelia that seems perhaps the related syringomyelia and treatment is straightforward, simplest of all is the syrinx found in association with a ventricular shunting for associated hydrocephalus being single web, usually dorsal to the cord, which forms an the most useful first step. Difficulties arise with spina almost complete transverse blockage across the subarach- bifida occulta, or sometimes spina bifida manque; that is noid space. These webs are usually idiopathic. to say, dysraphic lesions which lie between the occult and Commonly the morphology of the web shows the effect the full myelocele or myelomeningocele. Such of pressure surges in the CSF with a convexity, usually syringomyelia may be part of the dysraphic process-that directed upwards. Fluid which surges violently into the is, the failure of the cord to close and the place of concave side of the web tends to close off the surround- surgery is sometimes problematic. Intraspinal lipoma at ing spaces as the pouch like structure expands. They do the plaque site seems to be a cause of syringomyelia. The not show well on MRI directly, but they may be associ- place of "tethering" remains unclear in both the dys- ated with a displacement of the cord. Sometimes there raphic and posttraumatic varieties of syringomyelia. are several webs tending to retain fluid and the pathology How do these cystic cavities fill? There is somewhere may come to resemble a pouch in the subarachnoid within the human CNS, a group of neurons that derives space. Myelography shows them best but the associated satisfaction from a good explanation. Many people will syrinx and displacement of the cord are well appreciated not do things unless they have an explanation which pro- by MRI. Such webs may have a syrinx either above or vides them with reinforcement for their motivation. This below, sometimes both. The syrinx commonly does not human anfractuosity is the basis of religions, of interpre- reach up to, or apparently start from the level of the web tations of history, criticisms of the arts, and the livings but often starts 1 cm to 2 cm away from it (fig 1). gained by those who tell us exactly which were the moves The analogies with a cervical syrinx related to the that led Short to be defeated by Kasparov. In neuro- hindbrain that extends to within a few cm of the surgery there is no better illustration of the delights of impacted hindbrain are clear. In the normal human the proposing an explanation than the medical literature on upward surge of CSF produced by raised intraspinal the causation of syringomyelia. Consider the hindbrain pressure is vented via the cisterna magna to the intracra- related cases first. I recollect the satisfaction I gained nial compartment. This has a capacitance afforded to it from the "communicating hypothesis" in 1967.6 The fact mostly by compressibility of the venous system. If the that the proposal was wrong about the way in which the foramen magnum is plugged by tonsils, or by dense venous pressure impinges on the neuraxis and that subse- meningeal fibrosis then the situation may be similar to quent investigations have shown that the "communica- that illustrated by fig 1. Instead of the intracranial capaci- tion" from the fourth ventricle to the syrinx is not present tance acting as a damper the fluid is immediately in most cases, barely dims the recollection of the pleasure bounced back down the spine. The continuing arrival of at providing such a splendid, rational, and temporarily the pressure wave from the lower spine is likely to lead to convincing explanation. This same group of neurons chaotic disturbances of fluid below the site of blockage continues to give satisfaction to those who write new with consequences that may produce local conditions of explanations. Such are published often and commonly pressure such that fluid may be forced into the spinal are preceded by an attack on Gardner's views7 8 or on the cord.

25 year old hypothesis mentioned.7 One feature that The pressure inside the cord with a syrinx is almost http://jnnp.bmj.com/ many of these articles have in common is an unrealistic identical with that outside; the pressure may be raised a attempt to simplify the aetiological factors. They are cer- trifle above that in the subarachnoid space. This is clear tainly complex and multifactorial, with different mecha- from the porosity of the cord. If the spaces of Virchow- nisms in individual cases.9 " One unlikely sounding Robin are freely patent to allow inflation why is equal theory was that from Ball and Dayan'2 who criticised the and corrective deflation not going on? When the cord is "communicating hypothesis". They suggested that the artificially inflated, the fluid will tear out or leak easily pressure differences between the head and the spine were and quickly; the postmortem room and the animal labo- insufficient to fill the cord cavities along the presumed ratory attest to this. Why is it then that fluid in syrinxes on September 28, 2021 by guest. Protected copyright. communication, even though these differentials had been produces such damage? The tissue may be thought of as measured in the test situation as being up to almost 100 being like a lace curtain. The passage of a gentle waft of mm Hg.'3 Their proposal was that the hindbrain hernia- air may be unimpeded, but a breeze will move the curtain tion blocked the outlet of the spinal compartment so that and a gust of wind may damage it. the CSF in the subarachnoid space, when attempting to enter the head in response to an upward surge of fluid, had to be diverted and that therefore the fluid preferen- tially entered the spinal cord through the spaces of Virchow-Robin. The idea that raising the pressure around the cord should make it fill up with fluid was, and is, patently absurd. The pressures tending to force fluid into the potential syrinx are the same as those tending to compress the cord and to squeeze fluid out of the cavities through those same spaces of Virchow-Robin. Observation of the behaviour of water soluble contrast suggests, however, that Ball and Dayan were correct in their finding that there is a route from the subarachnoid pathway to the inside of the syrinx that must be through Figure 1 Idiopathic arachnoid webs tend to balloon out under stress and the walls of the cord. Filling of the inside of the cavities to cause energy to be inflicted on the underlying cord. There is often a syringomyelia cavity on one or other, sometimes both, sides of the web. seen on postmyelographic CT affects the tumour cases as Chaotic conditions of this sort may lead to syringomyelia in response to well as the hindbrain related syrinxes. There is thus an surges offluid pressure. Editorial 651 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from

. ...a-.---A

Figure 2 Vigorous movementfrom left to right (top arrows) probably results in the development ofwavelike undulations. During the passage ofan oscillation it is conceivable that pressure differences may develop infavour ofmigration across the cord (top empty arrows). It is also likely that pulsation within the cord is characterised by vortices and smalljets offluidpassing through holes in septa. A wavelength from A-B would mean a minimal change in the volume ofthat section. The development ofoscillating segments may predispose to the occurrence ofseptations. Energy may be transmitted through sections ofthe cord not in free communication with the main cavities (heavy arrows right) and may lead to the production of new cavities (empty arrows right).

It may be simplistic to suggest that there is necessarily Surgery is able to deal with most syringomyelia with a a "filling mechanism" for these cavities that may be complication rate that is steadily declining. Most cases explained by fluid conduits or valves. Perhaps it is more can be successfully dealt with by one operation and an sensible to ask the question in this form: improved consensus on treatment will surely come. Why are the laws of fluid dynamics better satisfied for Despite the bafflement which may exist about these ques- this person's spine by having fluid resting or moving up tions it may be best to recall what Gibbs said about and down inside the spinal cord rather than on the out- surgery for aneurysm in 1962, "The surgery of side? aneurysms is here to stay. It needs improving not Given the pleasure of expounding yet another explana- analysing". tion I cannot forbear repeating my current thoughts on We are perhaps better employed therefore in the mat- pathogenesis. Starting off with the non-hindbrain related ter of surgical management, in observing, empirically, the types of case, the likely scenario seems to be that impul- behaviour of syringomyelia when the CSF pathways are sive and sometimes violent venous distension compresses surgically modified. the dura. This leads to squeezing of the subarachnoid Magnetic Resonance Imaging has provided improved spaces and longitudinal pressure changes in the CSF and visualisation of these cavities with the advantage that the thus the spinal cord. This occurs in the normal subject; pictures can be repeated either in a period of preoperative the prerequisite for syringomyelia seems to be the block- indecision or to monitor postoperative progress. The age of the subarachnoid pathways. Blockage of the sub- availability of such a method makes an expectant policy arachnoid pathways presumably favours relocation of the more useful than it used to be.15-'9 Not all the MRI fluid from the subarachnoid space to the inside of the appearances are understood and particularly in the recur- cord; such cavities often seem to start close to a zone of rent or partly treated cases the appearances may be http://jnnp.bmj.com/ obstruction. Once fluid containing spaces develop then unclear. Perhaps the most important benefit that MRI these cavities also oscillate, often with a short wave- has brought is the ability to follow cases over years with length, thus allowing the formation or preservation of emphasis on postoperative results that may show objec- septations. The presence of multiple septations in the tive evidence of continuing reduction in syrinx size and cord indicates a likelihood of oscillating zones possibly correction of associated abnormalities. Neurological with the development of "standing waves" inside or out- assessments or subjective reports are not such a good

side the cavitated cord. It seems probable that there may guide to surgical success. on September 28, 2021 by guest. Protected copyright. be zones or times during such oscillations where the pres- One feature of MRI has been that there have been sure in the subarachnoid spaces is higher than that in the small cystic cavities disclosed which would not have been cord thus forcing fluid inwards across the wall (fig 2). easily detected by earlier methods. The symptomatology The explanation for the causation of syringomyelia of many of these is unclear. This problem is pronounced related to the hindbrain remains mysterious. Altered CSF in the field of spinal injuries: up to 50% of paraplegic dynamics has to be the explanation. The communicating patients have small cysts at the site of injury.3 1516 In par- hypothesis probably remains true, but only for a small tial paraplegia when there is a suggestion of neurological proportion of cases. About 10% of patients have a visible advance or severe pain the question of whether the cyst is communication on radiological investigation. Some cases contributing to the patient's problems is sometimes may have had a communication that has been closed off unaswerable. Small cysts may persist at the site of injury by later hindbrain compression but it seems that for the after successful surgery for big adjacent cavities. They greater number of cases of hindbrain related may be called "primary cysts" and their pathogenesis is syringomyelia there is no communication between the obviously different from clamant syringomyelia. For fourth ventricle and the syrinx. For some of these there intermediate varieties, surgery may be chosen as a test as never could have been such a communication. What then much as a treatment although for severe post-traumatic is the explanation? The nature of the closure at the top syringomyelia surgery has a high success rate.'4 18 29 end of the spine may be a factor and the hydrodynamic The MRI also helps in showing atypical small syrinxes, changes of relevance may all occur in the spine. often low down in the cord, sometimes clearly sympto- Reading the collection of communications from the matic with segmental disturbance of sensation or pro- Centenary Syrinx Workshop in Heidelberg, collected by gressive impairment of long tract function, even Donnauer, 4 might suggest that the surgical community is progressing to complete paraplegia. One pathology that in total disarray. This is, however, too pessimistic a view. has come to light in recent years is that a proportion of 652 Williams J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from these cases have a small subarachnoid web, or even features of hindbrain related syringomyelia.7'01314363843 several webs, interfering with free surging of fluid in the Failures occur, as well as complications46 and although it subarachnoid space as discussed earlier. is sometimes possible to suggest that the failure is due to There are some idiopathic small syrinxes the relevance some operative inaccuracy, there are other cases where of which is unclear. Gadolinium enhancement may be the correction of the hindbrain abnormalities seems fault- non-contributory and the CSF protein may be normal less and yet the associated syrinx remains full. Direct making a tumour unlikely for these cases. Symptoms syrinx drains may have a value in this situation. such as pain are not certainly related to the existence of Correcting the anatomical abnormalities of primarily these small cavities and exploration may sometimes show spinal syringomyelia is easy for the cases with a small no causative features and produce no useful therapeutic zone of blockage such as a web of idiopathic meningeal result. fibrosis. Most cases, however, have spinal fractures and Given that syringomyelia is a predominantly surgical the meningeal fibrosis may not only be severe but may condition the question is what operation to do. The sim- also extend over more than one segment. Correcting the plest is to put in a drain to allow the fluid to get out. This blockage is therefore more difficult and my present prac- has a century of tradition and optimistic modem reports tice includes opening up of the CSF pathways from to support it.202420 29-36 On even a moment's reflection, above to below by the creation of a surgical meningocele. however, it is clear that the placement of drains has a It is difficult to prevent further postoperative fibrosis limited value. A hole in the spinal cord may allow an from closing this off and a technique using stents has intracord high pressure to abate, but at least at rest, the been described which at present seems to be success- pressure in the cord is only minutely higher than in the fUl.21 23 Transection of the cord will usually correct and subarachnoid space and myelotomy may sometimes prevent the advance of syringomyelia but its use, of allow fluid to enter the cord, dependent on the local course, is limited to completely paralysed patients. Where anatomy. paraplegia is not present and meningeal fibrosis is present Also, any myelotomy carries a risk. Septations may be over many segments then the problems may be difficult. a problem when passing a drainage tube. There is a ten- Drains have a high failure rate and lowering the overall dency for a hole in the cord to heal. If the cord incision is CSF pressure may be the best stratagem. held open by any sort of hollow tube there is a tendency The value of lowering overall CSF pressure has been to fibrosis in the subarachnoid space. If the drain is to a noted since the writings of Benini and Krayenbuhl48 who low pressure area such as the pleural or peritoneal cavity, found symptomatic improvement after the shunting of then success of the procedure means that the syrinx cavi- the ventricles to extrathecal sites. They thought that the ties collapse around the end of the tube, usually with communicating hypothesis was correct for many cases blockage of the drain by glial overgrowth (fig 3). If the and that after ventricular drainage the pressure of the factors leading to syrinx formation are still operative then ventricles might be prevented from acting along the com- recurrence is probable, the drain becomes immured in munication to fill the syrinx. In cases such as those men- the wall, and is then useless.37 Drainage has a failure rate tioned, when the primary abnormality has been corrected of about 50% for syringopleural or syringoperitoneal and there is an element of hydrocephalus, then correction shunts over the first six years and results for syringosub- of the hydrocephalus by a valved shunt may lead to sub- arachnoid shunts or myelotomies are worse.37 The cor- stantial flattening of the syrinx.4849 rection of anatomical abnormalities in my view, is for the Recent enthusiasm has been expressed for the view present at least, the most dependable method, sufficing that the lowering of the pressure within the subarachnoid in most cases and appealing because it is the most "phys- spaces of the spine might prevent filling across the walls http://jnnp.bmj.com/ iological" in not being dependent on any kind of drains. of the syrinx.50 The idea that the fluid is forced across the The correction of hindbrain abnormalities is usually walls as expressed by Ball and Dayan remains unconvinc- successful in lessening both the radiological and clinical ing and yet the stratagem plainly has been successful. It is even successful when a hindbrain abnormality has not been corrected5' 52 although in that situation the use of a method to lower the pressure below the hindbrain hernia is dangerous. If the lowering of intrathecal pressure can cause a hindbrain hernia then lowering it when such her- on September 28, 2021 by guest. Protected copyright. niation is present must increase the risk of serious prob- lems from impacted hindbrain heia46473 56 The phenomenon of CSF drainage from outside the syrinx cavities favourably influencing cavity size is inter- esting. This might at first be thought to be paradoxical, lowering the pressure outside the syrinx might be consid- ered to increase the pressure difference distending the syrinx. The effect of lowering the CSF pressure by diminishing the volume is to predistend the veins and thus to limit the amount of energy which can be inflicted upon the CSF pathways by coughing, sneezing, straining, and so on (fig 4). Thus the amplitude and force of the longitudinal movements are likely to be lessened. It has the same effect as lowering the overall CSF pressure by ventricular shunting. The strategy of lowering the intrathecal CSF pressure is one of the few methods known to work for syringomyelia in the presence of Figure 3 Spinal cord tissues will naturally collapse around an effective drain. This case shows blockage of the lumen by ingrowth ofglia into the severe meningeal fibrosis along the length of the spine. syringopleural drain. The syrinx remains collapsed but this may in large The future holds the prospect of the elimination of part be due to the craniovertebral decompression and the ventricular shunt disabling syringomyelia from industrialised societies. which had previously been tried and had initially failed to control syrinx size. This may be achieved by early diagnosis, particularly by Editorial 653 J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from 6 Williams B. The distending force in the production of "communicating syringomyelia". Lancet 1969;ii: 189-93. M:qvimiim vpnnijc nrpcs:iirp 7 Gardner WJ. The dysraphic states: from syringomyelia to anencephaly. developed during straining Amsterdam: Exerpta Medica, 1973. 8 Pillay PK, Award IA, Hanh J. Gardner's hydrodynamic theory of syringomyelia revisited. Cleve Clin J Med 1992;59:373-80. 9 Milhorat TH, Johnson WD, Miller JI, Bergland RM, Hollenberg-Sher J. H Surgical treatment of syringomyelia based on magnetic resonance imag- CL~~~~~~~~~ ing criteria. Neurosurgery 1992;31:231-45. 10 Oldfield EH, Muraszko K, Shawker TH, Patronas NH. Pathophysiology of syringomyelia associated with Chiari I malformation of the cerebellar tonsils. J Neurosurg 1994;80:3-15. 11 Milhorat TH, Miller JI, Johnson WD, Adler DE, Heger IM. Anatomical Range of equilibration between CSF and ,-' C 5 ^ basis of syringomyelia occurring with hindbrain lesions. Neurosurgery venous pressure Baselin' 1993;32:748-54. 12 Ball MJ, Dayan AD. Pathogenesis of syringomyelia. Lancet 1993;fi: 799-801. 13 Williams B. Syringomyelia. In: Stein BM, McCormick PC, eds. Neurosurgery Clinics of North America. Intradural Spinal Surgery. Philadelphia: Saunders, 1990:653-85. 14 Donnauer E. 100 years syrinx-surgery, proceedings of a workshop, Dec 13-15 1992 at Heidelberg. Acta Neurochir (Wien) 1993;123:157-225. 15 Backe HA, Betz RR, Mezgarzadeh M, Beck T, Clancy M. Post traumatic spinal cord cysts evaluated by magnetic resonance imaging. Paraplegia 199 1;29:607-12. 16 Sett P, Crockard HA. The value of magnetic resonance imaging in the fol- low up management of spinal injury. Paraplegia 1991;29:396-410. 17 Enzmann DR, O'Donohue J, Rubin JB, Shuer L, Cogen P, Silverberg G. CSF Pulsation within non-neoplastic spinal cord cysts. AJNR Am Y c/ L Neuroradiol 1987;8:517-25. 18 Stevens JM, Olney JS, Kendall BE. Post traumatic cystic and non-cystic . Neuroradiology 1985;27:48-56. 19 Quencer RM. The injured spinal cord. Evaluation with magnetic reso- CSF volume No nance and intraoperative sonography. Radiol Clin of North Am 1988; 26:1025-45. 20 Shannon N, Symon L, Logue V, Cull D, Kang J, Kendall BE. Clinical 4 The volume curve the neuraxis is over features, investigation and treatment of post-traumatic syringomyelia. Figure pressure for largelyflat J Neurol Neurosurg Psychiatry 1981;44:35-42. the section where the veins are in pressure equilibrium with the CSF. At 21 Williams B. Post traumatic syringomyelia and cystic myelopathy. In: the tight side ofthe diagram (H) where the pressure is high the veins are Vinken PJ, Bruyn EW, eds. Handbook of neurology Vol 17 (61). Spinal flat and on the left (L) where it is low they are distended. Note that the cord trauma. Amsterdam: Elsevier, 1992. pressure difference between these two points is slight because ofthe zone of 22 Williams B. Post traumatic syringomyelia: surgical management and case equtlibration between the venous and CSFpathways. 7he amount of reports. In: Floman Y, Farcy J-P, Argenson C, eds. Thoraco-lumbar energy that may be added to the system by, for example a cough, is fractures. New York: Raven Press, 1993:429-48. maximised at the side. Movement to the is to 23 Williams B. Non-hindbrain related and post traumatic syringomyelia. In: therefore right left likely Schmidek HH, Sweet WH, eds. Operative neurosurgical techniques: indi- lessen theforces imposed in the CSF compartments ofthe neuraxis. Thus cations, methods and results. 3rd ed. Philadelphia: Saunders, 1995 (in the efficiency with which energyfrom the coelom is inflicted on the press). neuraxis may be represented by the lower curve. The higher the pressure, 24 Silberstein M, Hennessy 0. Cystic cord lesions and neurological deterio- within the normal range, the greater the imposition ofenergy imposed on ration in spinal cord injury: operative considerations based on magnetic the CSFpathways, by coughing or similar strenuous activity. resonance imaging. Paraplegia 1991;30:661-8. 25 Rossier AB, Foo D, Shillito J, Dyro FM. Post traumatic cervical syringomyelia. Brain 1985;108:429-61. 26 Barnett HJM, Jousse AT. Nature, prognosis and management of post- traumatic syringomyelia. In: Barnett HJM, Foster JB, Hudgson P, eds. Syringomyelia, major problems in neurology. Vol 1. London: Saunders, identification of populations at risk, such as those injured 1983. at birth,517 adults with spina bifida, patients with 27 Lacert P, Trottier S, Durand J, Pannier S, Grossiord A. Syndromes syringomyeliques tardifs chez les paraplegiques. Rev Neurol (Paris) kyphoscoliosis, and those with spinal meningeal fibrosis 1977;133:325-38. caused or due 28 Umbach A. cord by meningitis, epidural abscess, paraplegia I, Heilporn Post-spinal injury syringomyelia. Paraplegia http://jnnp.bmj.com/ 1991;29:219-21. to injury. Necessarily this optimism supposes the devel- 29 Abbe R, Coley WB. Syringomyelia, operation, exploration of cord, opment of adequate surgical techniques. 'Me future will withdrawal of fluid, exhibition of patient. J Nerv Ment Dis 1892; 19:512-20. certainly yield the necessary information, provided that 30 Barbaro NM. Surgery for primarily spinal syringomyelia in Batzdorf U, the surgical community shows self discipline in not mak- ed. Syringomyelia: current concepts in diagnosis and treatment. Baltimore: fallacious or claims for Williams and Wilkins, 1991:183-98. ing overoptimistic inadequately 31 Pousepp L. Traitment operatoire dans deux cas de syringomyelie. tried methods. Maybe those "explanation satisfied neu- Amelioration notable. Rev Neurol 1926;26:1171-9. 32 Padovani R, Cavallo M, Gaist G. Surgical treatment of syringomyelia: rons" should be consciously quelled for a while and favourable results with syringosubarachnoid shunting. Surg Neurol replaced by "audit satisfied neurons". These cells, admit- 1989;32: 173-80. on September 28, 2021 by guest. Protected copyright. 33 Tator CH, Briceno C. Treatment of syringomyelia with a syringosub- tedly, are more difficult to satiate; the business of careful arachnoid shunt. Can JNeurol Sci 1988;15:48-57. audit is at best tedious and at worst unproductive. 34 Irger IM, Paramonov LV. New method for draining a syringomyelia cyst. some Zh VoprNeirokhir 1979;3:3-9. Nevertheless there is much to gain in this field. In 35 Huewel N, Perneczky A, Urban V, Fries G. Neuroendoscopic technique ways MRI is our best weapon in the fight against this dis- for the operative treatment of septated syringomyelia. Acta Neurochir (Wein) 1992;54:59-62. order. The optimism of the introduction is not unrealis- 36 Rhoton AL. Microsurgery of Arnold-Chiari malformation in adults with tic. and without hydromyelia. J Neurosurg 1976;45:473-83. BERNARD W1IILLJAMS 37 Sgouros S, Williams B. A critical appraisal of syringomyelia drainage. Midland Centrefor JNeurosurg 1995;82:1-10. Neurosurgery and Neurology, 38 Batzdorf U. Syringomyelia related to abnormalities at the level of the Holly Lane, Warley, WMidlands craniovertebral junction. In: Batzdorf U, ed. Syringomyelia: current con- B67 73', UK cepts in diagnosis and treatment. Baltimore: Williams and Wilkins, 1991:163-83. 39 Batzdorf U. Chiari malformation and syringomyelia. In: Apuzzo MJ, ed. Brain surgery: complication avoidance and management. New York: Churchill Livingstone, 1993:1985-2001. 40 Dyste GN, Menezes AH, VanGilder JC. Symptomatic Chiari malforma- tions. Y Neurosurg 1989;71:1590-68. 41 Milhorat TH, Johnson WD, Miller JI. Syrinx shunt to posterior fossa cis- I Pearce J. Surgical management of syringomyelia. BM3F 1981ii:1204-5 terns (syringocistemostomy) for bypassing obstructions of upper cervi- 2 Greenfield JG. Syringomyelia and . In Greenfield JG, ed. cal theca. JNeurosurg 1992,77:871-4. Neuropathology. London: Edward Arnold, 1976:306-12. 42 Morgan DW, Williams B. Syringobulbia: a surgical appraisal. Jf Neurol 3 Vinters H. Neuropathology of syringomyelia. In Batzdorf U, ed. Neurosurg Psychiatry 1992;55:1132-41. Syringomyelia: current concepts in diagnosis and treatment. Baltimore: 43 Raftopolous C, Sanchez A, Matos C, Baleriaux D, Bank WO, Brotchi J. Williams and Wilkins, 1991. Hydrosyringomyelia-Chiari I complex. Prospective evaluation of a mod- 4 Park TS, Kaufman A. Syringohydromyelia associated with spinal dys- ified foramen magnum decompression procedure; preliminary results. raphism. In: Parks TS, ed. Spinal dysraphism. Oxford: Blackwell, Surg Neurol 1993;39:163-9. 1993:126-41. 44 Rice-Edwards M. Syringomyelia and its surgical management. In: Rice- 5 Oakes WJ, Gaskill SJ. Symptomatic Chiari malformations in childhood. Edwards, M, ed. Topical reviews in neurosurgery. Bristol: Wright, In: Park TS, ed. Spinal dysraphism. Oxford: Blackwell, 1992:104-25. 1972:161-84. 654 Williams J Neurol Neurosurg Psychiatry: first published as 10.1136/jnnp.58.6.649 on 1 June 1995. Downloaded from 45 Vaquero J, Martinez R, Arias A. Syringomyelia-Chiari complex: magnetic Doshi PK, Kamat MM. Percutaneous thecoperitoneal shunt for resonance imaging and clinical evaluation of surgical treatment. syringomyelia. Neurosurg 199 1;74:827-3 1. Neurosurg 1990;73:64-8. 52 Vassilouthis J, Papandreou A, Anagnostaras S, Pappas J. Theco-peritoneal 46 Williams B. A critical appraisal of posterior fossa surgery for communicat- shunt for syringomyelia: report of three cases. Neurosurgery 1993; ing syringomyelia. Brain 1978;10:223-50. 33:324-8. 47 Payner TD, Prenger E, Berger TS, Crone KR. Acquired Chiari malfor- 53 Chumas P, Drake JM, Bigio MR. Death from chronic tonsillar herniation mations: incidence, diagnosis and management. Neurosurgery in a patient with lumboperitoneal shunt and Crouzon's disease. Br 1994;34:429-34. Neurosurg 1992;6:595-9. 48 Benini A, Krayenbiihl H. Ein neuer chirurgischer Weg zur Behandlung 54 Chumas P, Armstrong DC, Drake JM, Kulkarni AV, Hoffman HJ, der Hydro-und Syringomyelie. Embryologische Grundlage und erste Humphreys RP, Rutka JT, Hendrick EB. Tonsillar herniation: the rule Ergebnisse. Schweiz Med Wochchenschr 1969;99:1137-42. rather than the exception after lumboperitoneal shunting in the paedi- 49 Kruse A, Rasmussen G, Borgesen SE. CSF dynamics in syringomyelia: atric population. Jf Neurosurg 1994;78:568-73. Intracranial pressure and resistance to outflow. Br J7 Neurosurg 55 Welch K, Shillito J, Strand R, Fischer EG, Winston KR. Chiari I 1987;1:477-84. "Malformation"-an acquired disorder? Jf Neurosurg 1981;55:604-9. 50 Park TS, Cail WS, Broaddus WC, Walker MG. Lumboperitoneal shunt 56 Park TS. LP shunting in case of Chiari malformation. J7 Neurosurg combined with myelotomy for treatment of syringohydromyelia. 1989;71:953. Neurosurg 1989;70:721-7. 57 Williams B. Difficult labour as a cause of communicating syringomyelia. 51 Vengsarkar US, Panchal VG, Tripathi PD, Patkar SV, Agarwal A, Lancet 1977;ii:51-53.

NEUROLOGICAL STAMP

Datura stramomium (Jimsonweed)

This plant, also known as thorn apple, devil's apple, mad apple, stinkweed, apple of Peru and Jamestown weed, is a member of the solanaceae (nightshade) family. Like other members of that family it is extremely poisonous and may cause death. Marc Antony's troops ate Datura when retreating from Parthia in 38 AD. Delirium, stupor, and S~~~~~~61 death occurred as a consequence. The word Jimsonweed is believed to be a corruption of the common name Jamestown weed. That name refers to an incident (about 1676) when soldiers, sent to quell a rebellion in the Jamestown Colony in North America, put some of the herb into their cooking pot and spent the next 11 days in a state of incoherence. The American physician and

botanist Charles Millspaugh stated in his medicinal plants http://jnnp.bmj.com/ (1892) that Jamsonweed was employed as "a narcotic, soothing drug" for epilepsy and neuralgia. He also noted that it was recommended as an ointment in bums and scalds, and that it had been used externally in folk medicine to treat boils and cuts and asthma. Sufferers MA YAR PO SAT 11-A&A&&.-IfXt :rt A& A. I= .0 .21 A.. A inhaled the smoke of the leaves or smoked the dried leaves for relief. -LUA- -

I on September 28, 2021 by guest. Protected copyright. Datura stramomium is portrayed on a set of stamps showing medicinal plants published by Hungary in 1961 (Stanley Gibbons 1776, Scott 1421). L F HAAS