the centrifugal pathway for micturition within …part in the control andcoordination ofmicturition....

J. Neurol. Neurosurg. Psychiat., 1958, 21, 177.

THE CENTRIFUGAL PATHWAY FOR MICTURITIONWITHIN THE SPINAL CORD

BY

P. W. NATHAN and MARION C. SMITHFrom the National Hospital for Nervous Diseases, Queen Square, London

In previous papers we have described the locationin man of the spinal pathways subserving sensationfrom the lower bowel and the motor aspects ofdefaecation (Nathan and Smith, 1953), and the spinalpathways subserving sensation from the bladder andurethra (Nathan and Smith, 1951; Nathan, 1952).It remains now to show the location in the spinalcord of the pathways used in the conscious controlof micturition and those used for the coordinationof bladder functioning.

HistoricalThe location of these fibres in the spinal cord both

of man and other animals has been relativelyneglected. Most observers who have studied thefunctional anatomy of micturition have consideredthe peripheral nerves of the bladder, the effects onthe bladder of total transverse lesions of the spinalcord, and they have studied those centres in the brain,stimulation or lesions of which affect the tone andother reflex activity of the bladder. The first workerto investigate the location of the tracts subservingmicturition was Budge; he worked on the subject in1842 and published his results and conclusions in1864. He obtained bladder contractions in the dog,cat, and rabbit by electrically stimulating the cerebralpeduncles, the inferior cerebellar peduncles, andthe region between these peduncles and the calamusscriptorius. In one animal, he cut through theposterior columns, and obtained contraction of thebladder, rectum, and Fallopian tubes from stimu-lating the rest of the spinal cord; he concluded thatthe descending pathway must lie in the anterior halfof the cord. Mosso and Pellacani (1882) criticizedBudge's work, as they found that his mode ofstimulating the cord was bound to cause muchspread of the current. They induced bladder con-tractions in the dog by painfully compressing thepaws. To determine the spinal course of the tractconveying impulses giving rise to these contractions,they carried out various partial sections of the cord.Their conclusions were: " Dans les cordonsanterieures de la moelle et dans la partie anterieure

des cordons lateraux, il ne passe pas de fibres motricesde la vessie." Stewart (1899), who worked on thecat, found that " bladder contractions can be pro-duced by stimulating the cord at any point"; andhe also found that " stimulation of the cord oftencauses an already contracted bladder to relax ".He came to the conclusion that, " Above the vesicalcentre . . . impulses pass down the cord in theposterior half of the lateral columns only. . . Belowthe level of the centre the impulses were shown topass in the lateral columns." He placed the vesicalcentre in the lumbar cord. Spiegel and MacPherson(1925) found that in the cat complete section of theanterior and posterior columns failed to interruptthe pathway; and they concluded, by increasing thearea of cord sectioned, that the fibres subservingbladder contraction must run in the lateral columns.Further work on this subject in the cat led Hunsickerand Spiegel (1933) to conclude that the descendingcorticofugal tracts to the bladder are conveyed bothby pyramidal and extrapyramidal pathways.

In our opinion, the best work on the location ofthe tracts subserving micturition is that of Bar-rington (1933). He summed up the results of hiswork, in 1948, as follows:

" In the cat, micturition is affected in the same wayas by a spinal transection if the periphery of the dorsalhalf of the lateral column is destroyed on both sidesin the upper lumbar region, but it remains unaffectedif symmetrical lesions are made in other parts of thecord at this level. It is probable that the peripheryof the dorsal half of the lateral columns is the partof the cord containing the tracts used in micturitionin the segments above the lumbar region."He also showed (1933) that:" The centrifugal and the centripetal paths of mic-

turition undergo extensive crossing in the segmentsfrom which the pelvic nerves arise and possibly in thesegment next in front of these."

Wang and Ranson (1939) caused bladder con-tractions in the cat by stimulating the hypothalamus,and sought for the descending pathway conveyingimpulses giving rise to the bladder contractions bymaking various lesions; their most caudal lesion was

177

Protected by copyright.

on October 6, 2020 by guest.

http://jnnp.bmj.com

/J N

eurol Neurosurg P

sychiatry: first published as 10.1136/jnnp.21.3.177 on 1 August 1958. D

ownloaded from

http://jnnp.bmj.com/

P. W. NATHAN AND MARION C. SMITH

at the first and second cervical segments. Theycame to the conclusion that the descending tractmediating impulses to the bladder lies at this levelin the antero-lateral column. Levin and Langworthy(1937) showed that lesions in the cat's tegmentum,not involving the cerebral peduncles and thussparing the cortico-spinal tracts, cause increasedresponses of the detrusor musculature to stretchand to " peripheral stimulation in the sensorydomain of the sacral and coccygeal segments ",changes which eventually lead to a decreased vesicalcapacity. Later, Langworthy (1940) showed thatelectrical stimulation of the cortico-spinal tractcauses bladder contraction. Thus, Langworthy hasbrought forward evidence that there are pathwaysboth within and without the cortico-spinal tract inthe cat which are concerned with bladder function-ing. Tower (1940), however, found in the monkeythat dividing the cortico-spinal tract bilaterallycaused " no urinary or faecal incontinence orretention ".

Thus, the evidence for the location of the descend-ing pathways to the bladder centre in the cat isfairly consistent: the pathways lie in the lateralcolumn of the cord. It is not agreed whether thefibres lie in the dorsal or ventral parts of thesecolumns. The evidence indicates that some of thefibres run with the cortico-spinal tract and some runoutside this tract.

Before the turn of this century, no mention of thelocation of these tracts within the human spinalcord had been made. In 1900, von Czyhlarz andMarburg wrote that the tracts connecting the cere-brum with the sacral cord " seem to run in theneighbourhood of the pyramids. But in order toget permanent disorder of bladder function, it seemsthat both pyramids must be damaged, and damagedto a large extent ". Nineteen years later, Muller(1918) still had to state:

" We have no information concerning the paths inthe spinal cord by which the bladder centres receivetheir stimulation or inhibition from the brain. Wedo not know in what part of a section of the spinalcord we should seek them. We have no proofthat the impulses to the smooth muscles of the bladdermake use of the centrifugal pyramidal tract."

And in 1926, Dennig wrote that doubt might becast on the hypothesis that the motor tract in manlies in a similar position to that which it occupiesin animals. He wrote that lesions of the spinal cordgiving rise to severe disorders of micturition maynot involve the lateral cortico-spinal tracts, andthat lesions of the lateral cortico-spinal tract maycause no disorders of micturition. He pointed out,however, that it is hazardous to draw conclusionsfrom clinical conditions, for the upper motor neurone

lesion might not involve the nerve supply to allmuscle groups.The first workers to present any experimental

work on man designed to show the location of thesetracts were Foerster and Gagel (1932). They statedin their text that bilateral cordotomy " repeatedly"gave rise to retention of urine, although the retentionwas always temporary. They drew the conclusionthat:

" the descending tract in the spinal cord subservingbladder innervation runs partly in the antero-lateralcolumn. This conclusion has to be drawn from theretention of urine that initially follows the operation.But vesical tracts must also run in other columns,for even after the most complete bilateral antero-lateral cordotomies, the bladder behaviour showscomplete recovery."It must be pointed out, however, that this part

of their study of the effects of cordotomy is inade-quate, and that the very short case histories ap-pended to their text do not always support theconclusions they drew. In their paper there are17 cases of bilateral cordotomy; in two of these casesthe state of micturition of the bladder is not men-tioned. In a further six cases the evidence presenteddoes not give support to the statements they made.In one case, due to collapse of the vertebrae, therewas retention of urine before the cordotomy; andin another case the collapse of the vertebrae occuLrredone week after the cordotomy, making furtherobservations on the effects of cordotomy impossible;another patient died one week after the operation.Two patients died three months and eight monthsrespectively after the cordotomy, from cystopyelitis-a condition which usually follows paralysis of thebladder. A further patient two months after thecordotomy still had retention of urine.Langworthy and Lewis (1935) concluded from

two cases of disseminated sclerosis which they studiedin detail (unfortunately without obtaining post-mortem evidence), that their patients' urgency andfrequency of micturition was related to " damage tothe lateral columns of the spinal cord in the neigh-bourhood of the cortico-spinal tracts ". In 1940,Langworthy, Kolb, and Lewis stated that the path-way " controlling tone in the vesical muscle . . .

probably lies in the lateral columns close to thecortico-spinal tracts ". Presumably, they drew thisconclusion from the two cases already mentionedand the evidence from experimental work onanimals.Wruck (1943) stated that extensive antero-lateral

cordotomies do not cause bladder disturbance; andfrom this he concluded that the centrifugal tracts tothe bladder probably do not lie between the antero-lateral and postero-lateral columns. No evidencewas given in his paper apart from the bald state-

178



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


PATHWAY FOR MICTURITION WITHIN SPINAL CORD

ment; it seems that his views were based on Foerster'sexperience, already quoted.Thus we see that in man there is no adequate

evidence apart from that provided by Foerster andGagel; and this is most unsatisfactory. Langworthyet al. wrote in 1940:

" Pathological studies in man are needed to delimitmore clearly the pathways controlling micturition.It will be of interest to observe whether these studieswill confirm the anatomical localization suggested byanimal experimentation."Such studies in man are presented here.

Effects on Bladder Functioning of Dividing theSpinal Cord

When all the spinal tracts subserving micturitionhave been divided, the resulting disturbance offunction resembles, though it is not identical with,that which follows when the spinal cord has beentotally divided. Before considering the presentseries of cases, the functioning of the bladder whenthe cord is divided will be briefly reviewed. It isimmaterial at what level such a transection occursprovided it divides the sacral centre from thosecentres above the medulla oblongata that play apart in the control and coordination of micturition.

Transection of the cord interrupts the sensationof pain from the bladder and the normal sensationunderlying the desire to micturate, and the controlof micturition permanently; but the behaviour ofthe automatic bladder, cut off from impulses arisingabove the spinal cord, is a changing one. Thischanging course of reflex and postural activity isusually considered to pass through three stages;these will now be briefly considered.Most writers on this subject write of the state of

the visceral musculature of the bladder as beingregulated by the same factors that regulate the stateof the somatic musculature. This is an assumption.Whether this assumption is justified will not beconsidered here, as it does not influence the observedbehaviour of the bladder.The bladder, like all muscle, has tone or posture.

Posture was designated by Sherrington (1915) as:" that property of the bladder by virtue of which itsolves the problem of acting as a reservoir for quanti-ties of fluid of very varying volume from one occasionto another without allowing the intravesical pressureto attain the reflex stimulus threshold height with oneparticular fluid quantity only."This postural activity is under the constant

influence of centres situated cranial to the spinal cord.When the spinal cord is suddenly divided, the

bladder is then, according to most observers, in astate of flaccidity or atonicity. The cystometrographshows the flat type of curve typical of low pressure.In fact, the viscus certainly possesses some intrinsictone, for it is not completely flaccid even when all3

the nerves between it and the sacral centres havebeen divided. Further, some recent workers con-sider that the type of curve seen in the early stageafter division of the cord results from stretching themusculature of a bladder in which the micturitionreflex is absent. At this time, there is no vesical reflexactivity; there are no detrusor contractions and nomicturition reflex, and these cannot be induced bythe injection of carbaminoyl choline. The sphinctersare tightly shut. Just as with the somatic musculature,in which the stage of flaccidity with spinal shockeventually gives way to a stage of increased reflexactivity characteristic of the spinal cord functioningindependently of the centres cranial to the medullaoblongata, so in the bladder there follows a stageof increased reflex activity; reflexes are increased inthe sense that they fire off with lesser stimuli andthat the resultant contraction is larger and longerlasting. There is spasticity-an increased resistanceto passive movement. In the bladder, this can bemade apparent by filling the viscus so as to stretchthe musculature, while the pressure is recorded.This spasticity is thought to be due to increasedreflex activity, the result of heightened stretchreflexes. Cystometrography at this time shows amore active detrusor musculature; the tone isincreased, the pressure curve rises very steeplycompared with the normal, and there are sharpwaves of increased pressure. When these sharpwaves occur, the sphincter mechanism of the bladderand urethra relaxes. Periodic micturition nowoccurs; and the urine can be manually expressedwith relative ease. Thus, retention of urine withdecreased reflex activity changes to periodic auto-matic micturition. This periodic micturition can beso efficient that the bladder may expel all its contents.Eventually, posture, as designated by Sherringtonabove, becomes conspicuously absent; for muchspontaneous change in pressure occurs with aconstant volume of fluid. The pressure is con-stantly high with a small volume of fluid;forinstance,75 ml. of fluid may be held at a pressure of 100 cm.of water. The reaction of the bladder musculatureis now greatly enhanced. Sudden contractions of themusculature often expel the catheter as well asthe urine. Anything causing the slightest increase intension on the bladder musculature, such as a risein intra-abdominal pressure, or a slight movement,gives rise to the micturition reflex-contraction ofthe detrusor with opening of the sphincters andexpulsion of urine; but this reflex is inadequatelysustained, so that although small amounts of urineare spurted out, large amounts often remain in thebladder. When the pressure of the bladder is veryhigh, these reflex expulsions of urine may lead tothe wrong conclusion that urine is dribbling away

179



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



continuously, as with a patulous sphincter. Cysto-metrography at this time shows a steeply rising curvein nearly all cases; however, in a few cases the curveremains flat once a certain high level has beenreached.The normal coordination between the detrusor

musculature and the sphincter musculature isseverely disturbed. Denny-Brown and Robertson(1933) have suggested that this is partly due to thefact that intercurrent stimuli from the lower partof the body cause a sudden reflex shutting of thesphincter with consequent cessation of micturitionbefore the bladder has been emptied.The progress through these recognized three

stages is gradual. Stage I is the stage of absence ofdetrusor contractions and of the micturition reflex;Stage II is the stage of spasticity with a steeply risingcystometrographic curve; and Stage III shows thefeatures of Stage II in more pronounced form; thereis intense spasticity, vigorous contractions of thedetrusor being evoked by small quantities of urineand by intercurrent stimulation of the more caudalparts of the body. This manner of considering theprogress of bladder functioning as passing throughthree stages will be followed here, as it allows ourwork to be compared with previous work, and asit permits more concise presentation of our results.

Effects on Bladder Functioning of Cutting SpinalPathways Subserving Micturition

It was said above: " When all the tracts subservingmicturition have been divided, the resulting dis-turbance of function resembles that which occurswhen the spinal cord has been totally divided."The slight differences in bladder functioning whenthe cord has been divided and when the spinal tractsspecifically subserving micturition have been dividedwill now be considered.

(a) Sensation.-Patients with division of the tractconveying impulses subserving the normal sensationof the desire to micturate have nevertheless variousother ways of finding out when the bladder is fulland when it is going to evacuate its contents auto-matically. This aspect of the subject has beeninvestigated and discussed by Nathan and Smith(1951) and by Nathan (1956 a and b).

(b) Conscious Control.-The ability to controlconsciously the levatores ani, and in particular theiranterior parts, the pubo-coccygeus muscles, and theexternal sphincter is lost. How far this loss interfereswith the ability to start micturition is not reallyknown. In normal micturition the neck of thebladder descends in the pelvis and the anteriorparts of the levatores ani contract and the detrusorcontracts at about the same time; Nesbit and

Lapides (1948) have shown that bilateral pudendalnerve block does not interfere with the ability tostart micturition. With regard to stopping mic-turition, the external sphincter is important, andthey showed that bilateral pudendal nerve blockdoes interfere with this mechanism. It seems thatone may state that without control of the perinealmusculature the patient cannot start or stop mictu-rition in a normal manner. However, such a patientwhen micturition has reached Stages II and III,may be able to empty his bladder, at least partially,by inducing a reflex contraction of the detrusor.Various ways of doing this are found by patients;they may press on the abdomen, hold the breathand strain, or they may flex the thighs and trunk.Although such a patient cannot actively stopmicturition, it can often happen that micturitioncontinues only as long as straining is sustained, andin such a case the patient can stop micturition merelyby ceasing to strain. The normal inhibition of therising pressure exerted by the bladder musculature,which is a feature of bladder control, can in no waybe substituted when the relevant tract is divided.It is important to realize that the paralysis of themuscles of the pelvic floor and of the perineum canoccur without any concomitant paralysis of thelower limbs, and with some neurological lesions onecan find severe paralysis of the lower limbs withminimal involvement of the movements of themusculature of the pelvis and perineum.

(c) Postural and Reflex Activities.-There is nodifference in these activities in those cases in whichthe cord is totally divided and in those in whichonly the relevant tracts in the cord are divided.Such are the features of the functioning of the

bladder to be found when the tracts within the spinalcord subserving micturition are divided. They canbe made manifest in an acute experiment-observingthe posture or tone of the bladder during the opera-tion of dividing the relevant pathways in the cord,and in a chronic experiment-day-to-day observa-tions of the postural and reflex activities after thesetracts have been divided. Observations will not bemade here on the sensory aspect of the subject asit has been considered in previous papers.

Methods of InvestigationBoth the clinical and the histological methods of

investigation employed were described in ourprevious paper on the afferent pathway in the cord(Nathan and Smith, 1951).

MaterialOut of a series of 60 cases ofcordotomy performed

for the relief of pain in which full histologicalexamination of the spinal cord and brain has been

180



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


PATHWA Y FOR MICTURITION WITHIN SPINAL CORD

carried out, 21 cases ofbilateral cordotomy and D

15 cases of unilateral cordo- uwtomy were studied in detail 36 -

with regard to the function- 0~33-aing of the bladder and mic- 3

0~~~~turition; one case of throm- °n 30 1lbosis of the posterior in-

2

ferior cerebellar artery was ; 27also investigated, although mX 24-lno post-mortem examina- /

21 -tion was made, the patient 0 21 -still living. The evidence mafrom the other cases of this C /series supports the conclu- m ISsions drawn from this 12 Zselected material; but as 0these patients were not 9investigated in such detail, 25 5no further mention of them ML.will be made.Although we now have a large number of cases,

it is surprisingly difficult to collect a sufficientnumber of suitable cases for such a study as this.All the patients, except the one with a thrombosis ofthe posterior inferior cerebellar artery, had cancer.Patients in whom the cancer involved the centralnervous system were excluded from the series; butthose in whom only the peripheral nerves were in-volved had to be used. For in most patients needingcordotomy for the relief of pain the peripheral nervesare involved by the tumour. In some of thesepatients there was involvement of the sacral nerveson one side. It might be thought that as there was alesion of the sacral nerves, the effects of dividing thepathway in the spinal cord could not be properlyestimated; and it could further be objected thatcomparisons between cases cannot properly be madewhen some of the patients have the sacral nervesinvolved by growth and others have the peripheralnerve supply intact. And in addition to theseobjections, it might be considered that as the lesionsmade in the spinal cord vary considerably from caseto case, the number of factors playing a role is solarge that one case can hardly be compared withanother. Nevertheless, it seems to us that theinclusion of patients with involvement of the sacralnerves on one side is justified. There is in this seriesa sufficient number of cases without any involvementof the sacral nerves, and so it is quite clear what theresults of dividing the tracts in the cord are. Afurther reason is that each patient was investigatedbefore and after the cordotomy; and so, whateversort of bladder function was found before thisoperation, any sudden changes in it could beattributed to the effects of dividing the tracts in the

FiG. .-Cystometrogram made during the cordotomy. At eacharrow an incision was made into the antero-lateral part of thespinal cord.

TOT L AMOUNT180 ML.

75 100 125 ISO 200OF FLUID RUN IN

spinal cord. And finally, the kind of bladder functionthat results from a lesion of the peripheral nervesupply is different from that kind which results froma lesion of the spinal tracts.

The Immediate Effects of Dividing the TractsAn acute experiment demonstrating the loss of

tone caused by sudden division of the lateral andantero-lateral parts of the cord is shown in Fig. 1(Case 47). Cystometrography was performed duringthe course of the bilateral cordotomy; the anaesthesiawas kept as light as possible to avoid any effects ofthe anaesthetic on the centres coordinating mic-turition. The pressure in the bladder was risingrather rapidly before the incision was made; thismay have been due to the inevitable interferencewith the spinal cord. When one side of the cord wasincised, the bladder pressure fell from 36 cm. to18 cm. of water; when the second side of the cord

R L

TSFIG. 2.-Case 47: Transverse section through lesion (Marchi prepara-

tion).

181



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


P. W. NATHAN AND MARION C. SMITHFIG. 3.-The cystometrogram made three days before the cordotomy

is shown by the dotted line, and that made one month after theoperation is shown by the continuous line. M followed by +indicates the desire to micturate.

ER EXTRUDED TOTAL AMOUNT 200 ML.HEADACHE GONE.

I'

I'.

I 'I "TOTAL AMOU NT

600 ML.

.

1,,,J . M4-~

was incised, the bladder pressure fell a further 5 cm.,and it remained at this level for the rest of the timeof the operation, in spite of filling of the viscus.The extent and location of the lesions made in this

spinal cord, in the thoracic segment, can be seenin Fig. 2.

Chronic Effects of Dividing the Tracts BilateraUyTwo cases will be described in which there was

good evidence of bilateral division of all descendingtracts.

Case 1.-Mrs. M. had a bilateral cordotomy for thepain due to carcinoma of the cervix uteri. Followingthe operation there was analgesia throughout the rightside of the body below the sixth thoracic dermatomeand throughout the left side of the body below the ninththoracic dermatome. The function of the bladder wasthat of Stage I; and three weeks after the operation025 mg. ofcarbaminoyl choline chloride had no effect onthe bladder. One month after the operation Stage II

was reached. Fig. 3 shows the cystometrogram per-formed before the cordotomy, and that performed afterthe cordotomy during Stage II. Comparing the two, the

182

A

0In

-4

m(a

m

ML. FLU ID PUT INTO BLADDER

l



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


R

PATHWA Y FOR MICTURITION WITHIN SPINAL CORD

L

T4FIG. 4.-Case 1: Transverse section through lesion (Marchi prepara-

tion).

pressure on starting, when only enough fluid is in thesystem to take a reading, was 4 cm. before the operation,and 16 to 18 cm. after it. Before the cordotomy thepressure remained under 12 cm. as long as the contentswere below 375 ml.; then, with the increasing desire tomicturate, the pressure began to rise steeply. Strongdetrusor contractions started when the bladder contained500 ml. In Stage II, the curve showed no horizontalcomponent; it rose steeply as the bladder was filled, andthere were strong detrusor contractions throughout.Finally, when only 200 ml. was in the bladder, there wassuch a strong contraction that the catheter was extruded.Before the operation, as the bladder was filled the patientwas making a conscious effort to hold urine; after theoperation, she felt she could not influence the bladderin any way. At necropsy 68 days after the operation, thebladder was seen to be trabeculated.The location and extent of the bilateral lesions in this

case are shown in Fig. 4.

Case 7.-Mrs. Y had a bilateral cordotomy for thepain due to a colloid carcinoma of the cervix uteri.Following the operation there was analgesia throughoutthe right side of the body below the tenth thoracicdermatome and throughout the left below the eighththoracic dermatome. The function of the bladder wasthat of Stage I; Stage II was reached 12 days after theoperation.The location and extent of the bilateral lesions in this

case are shown in Fig. 5.

From these three cases-the two described im-mediately above and the one given to show theacute effects of dividing the tracts-deductionscan be drawn on the location of the fibres used inthe control and coordination of micturition. Thesecases were typical of the group; 16 of the 21 studiedshowed this complete syndrome. The smallest areaof degenerating fibres common to the six sides ofthe three cords presented must contain the majorityof the centrifugal fibres concerned in micturition.It can be seen from Figs. 2, 4, and 5 that this area

lies on the equatorial plane of the cord. In all thecases in which there was the complete syndrome,

TSFIG. 5.-Case 7: Transverse section through lesion (Marchi prepara-

tion).

the lesions involved this region of the cord. In fiveother cases the typical syndrome was present forweeks, and then there was more recovery from thisstate than is usually found with complete lesions.It was interesting to note that in these cases thelesions made at operation did not involve this regionas completely on both sides of the cord as it did inthose cases with the complete clinical syndrome.We thus conclude that the majority of the centri-

FIG. 6.-The region within the spinal cord of the centrifugal pathwayfor micturition. The majority of the fibres probably lie withinthe region indicated by cross-hatching.

R L

T5

183

44



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



fugal fibres descending from the brain must lie inthis region of the lateral columns shown in Fig. 6.

In most of our cases, and in those presented sofar, the lesions were made in the thoracic cord. Inone case the incisions were made at the fifth andsixth cervical segments, and in two cases at the firstand second lumbar segments. The location of thelesions in these three cases showed that the efferentpathway lies in the same region of the lateral columnsin these parts of the cervical and lumbar cord as itdoes in the thoracic cord.More evidence that the pathways lie in this part

of the cord can be obtained from those patientswho did not have the syndrome of complete divisionof the tracts but showed a part of the syndrome ormore recovery from the stages of the syndrome thanis possible with total division of the relevant tracts.Further evidence is also gathered from those patientswho had the incisions made in the neighbourhoodof these tracts, and yet had no disturbance of bladderfunction.

Chronic Effects of Partial Division of the TractsBilaterally

The following case shows that bilateral lesionsanterior to the region shown in Fig. 6 do not causethe complete syndrome.

Case 21.-Mr. M. had a bilateral cordotomy for thepain due to a chondrosarcoma of the ilium. Althoughthe operation was bilateral, the incision on the left sideof the cord was made purposely superficial and somewhatanterior. Following operation there was completeanalgesia throughout the left side of the body below theninth thoracic dermatome, and hypoalgesia on the rightside of the body below the ninth thoracic dermatome.For the first week after the cordotomy there was completeretention of urine. By the end of the first fortnight,however, except for some frequency of micturition,bladder functioning was normal. He could micturatewithout straining and he could stop micturating in themiddle of the act; he could hold urine for half an houror more after feeling some desire to micturate. Thisstate continued for five months. After this time, it wasseen through the cystoscope that the base of the bladderwas involved in the spreading tumour; and so furtherconclusions on bladder functioning could not be attri-buted solely to the lesions in the spinal cord. The lesionsin the cord of this case are shown in Fig. 7. It is probablethat on the left side some of the relevant fibres have beencut and that on the right side they have not been involved.

No Effects after Bilateral Lesions in Neighbourhoodof the Tracts

The following case shows that bilateral lesionsposterior to the region shown in Fig. 6 do not causethe syndrome.

Case 22.-Although the operation was bilateral, theincision on the left side of the cord did not involve any

FIG. 7.-Case 21: Transverse section through lesion (Marchi prepara-tion).

sensory fibres. Following operation, the only part of thebody where pain sensibility was diminished was on theleft in the first and second sacral dermatomes; it wasslightly diminished in the third sacral dermatome. The

FIG. 8.-Case 22: Transverse section through lesion (Marchi prepara-tion).

184

N. .As-

M.V

41-

AN2...

M 4



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



R

FIG. 9.-Case 11:preparation).

Transverse section through lesion (Weig

L concerned with the conscious control and coordina-tion of bladder functioning.

Degeneration in fibres to the sacral micturitioncentre is accompanied by degeneration in descendingfibres to the neurones for the lower limbs. But as theneurones for the limbs are situated mainly cranialto the third sacral segment, there are, among thosedegenerating fibres still present at, and caudal to thissegment, the fibres concerned with micturition,defaecation, and the pelvic musculature. In ourseries of cases only a few fibres are to be seen at

F8 these levels; and even in the cases in which thegert Pal cordotomies were performed in the lumbar region

more fibres are not seen. This reduction in the

patient developed no disturbance of micturition of anysort; he never had to be catheterized, he could start andstop micturition when he wanted, and he passed a normalamount of urine at normal intervals of time. The lesionsin the cord are shown in Fig. 8. It will be seen that theequatorial plane is not involved, with the possible excep-tion of a few fibres on the right side.

Thus Cases 21 and 22 serve to show the boundariesof this region where the centrifugal fibres are thoughtto run. In Case 21 the lesion on the right side is atthe anterior limit of the region; and in Case 22 bothlesions probably lie just posterior to it. Other caseswith lesions not involving this region of the spinalcord showed no disturbance of micturition or itsconscious control.

In a previous paper (Nathan and Smith, 1951) onthe afferent pathway from the bladder, we showedthat the afferent pathway lies in the same region asthat which we are now proposing for the efferentpathway. In one patient, Case 11, there was clinicalevidence of bilateral division of the afferent tractwith only partial involvement of the efferent path-way. The lesions in the cord, shown in Fig. 9,involved the lateral part of the region shown inFig. 6. Thus, this one case may be regarded asevidence suggesting that the centripetal pathway lieslateral to the centrifugal pathway.We have now shown where the tracts lie in the

cervical, thoracic, and lumbar regions of the cord.It remains to consider their position in the sacralregion. Fig. 10 shows the distribution of thedegenerating fibres in the lower lumbar and thesacral segments, following a unilateral cordotomyin the third cervical segment. Degenerating fibresare seen in the lateral column of the lumbar sectionand considerably fewer fibres are to be seen in thesame region of the sacral segments. When a com-parison is made between this case and other caseswith lesions involving slightly different parts of thelateral columns, it is seen that the more anterior ofthe degenerating fibres shown in Fig. 10 are those

R L.

C3

L3

S3FIG. 10.-Case 37: Transverse section through lesion in cervical

segment, and sections showing resultant degeneration in lumbarand sacral segments (Marchi preparation).

185



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



number of fibres between the lumbar and sacralregions may seem surprising. The reasons for thesmall number of degenerating fibres are probablythe following. As is well known, the degenerationof the terminal parts of the axon is completed muchsooner than that throughout the rest of the axon,and so the terminal parts cannot always be seen inthose cases where degeneration can be tracedthroughout the rest of the tract. Also, it could bethat a few of the centrifugal fibres relay in the greymatter of the cord between cervical or thoracicsegments and the sacral segments; in this case, therewould be no degeneration in the axons having theircells of origin caudal to the axons divided in thecordotomy lesion.

Chronic Effects of Division of Tracts UnilaterallyThe clinical picture resulting from division of the

tracts on one side of the cord can now be given;for in certain cases it can be shown histologicallythat the relevant tracts have certainly been divided.

All patients who had the pathway subserving con-trol of micturition divided unilaterally had no dis-turbance of voluntary control. Thus, one mayconclude that a tract on one side of the cord isenough to ensure the normal function of startingand stopping micturition. (Needless to say, thisstatement is true only if the peripheral pathway andthe structures, such as the bladder, urethra, andlevatores ani, are themselves free from abnormali-ties.) Normal voluntary control is not necessarilypresent for the first four or five days after thecordotomy; for during this time (and immediatelyafter any operation on the spinal cord) there maybe retention of urine. But after a few days normalcontrol is regained.

All patients who had the pathway subserving thepostural and reflex activities of the bladder dividedunilaterally showed an alteration in these functions.Although the patients were themselves unaware ofany abnormal functioning of the bladder, an ab-normality could be demonstrated by comparing thepre- and the post-operation cystometrograms: inall these patients the post-operation differed from thepre-operation one, even though the post-operationone came within normal limits. In all cases thedifference consisted of a higher vesical pressure onstarting filling and a high pressure throughoutfilling.

Case 37.-This will serve as an illustration of the clinicalpicture found with division of the tracts unilaterally. Thecystometrograms made a week before and a fortnightafter the operation are shown as Figs. 1 A and B;and the extent of the lesion made at operation is shownas Fig. 10. The cystometrogram (Fig. 1 A) made beforethe operation is essentially normal; and the patient had

no history to suggest any sort of abnormality of bladderfunctioning. Immediately after the operation, he hadalmost complete retention of urine; after an injectionof carbaminoyl choline, he passed 120 ml., but a further800 ml. remained in the bladder and had to be removedby catheter. For five days he was treated by tidaldrainage. On the sixth day, he had a definite normaldesire to micturate, he strained, and urine was passedbeside the catheter. Tidal drainage was discontinued.Micturition from this time was apparently normal. Thepatient had a normal desire to micturate, he could feelurine passing and could feel the muscles of his perineumacting. He could micturate without straining, could startwhen he wanted to, and could stop in the middle of theact. The cystometrogram (Fig. llB) made a fortnightafter the operation is, however, far from normal. Thisis the typical cystometrogram of Stage II. On everyoccasion that the pressure rose, the patient had the desireto micturate, but felt it unilaterally. (This feature wasdescribed in our previous paper, 1951.) That he couldcontrol micturition is reflected in the rise of pressure to20 cm. when he was told to try and micturate. The com-parison between the cystometrograms performed beforeand after the operation is striking. After the operation,the smooth slowly rising curve of the normal has gone,the base line, such as it is, is raised throughout, andspontaneous contraction occurs repeatedly, startingwhen there is less than 50 ml. of fluid in the bladder.When this cystometrogram was made, the patient wasunaware of having any disorder of bladder functioning,for he could control micturition normally. But thedivision of the pathway unilaterally has disturbed normalbladder tone and has heightened the micturition reflexso that it is triggered off by inadequate stimulation.

Thus unilateral division of the centrifugal tractconveying impulses serving to inhibit bladder tonegives a change in vesical pressure-a change similarto but less than that which follows bilateral divisionof the tract. This might have been expected. Forpresumably less impulses arrive at the sacral centrefrom the centres in the brain, and so the total be-haviour of the bladder resembles more nearly thatof the bladder in cases of total cord division.

Cases of thrombosis of the posterior inferiorcerebellar artery also show the syndrome of divisionof the tracts unilaterally. Although we havenaturally not had the opportunity of performingcystometrograms before the thrombosis in suchcases, cystometrograms performed after the throm-bosis are similar to that given in Fig. llB. Thepatient is usually unaware of any abnormality ofbladder function, but the pressure is abnormallyhigh, and the micturition reflex tends to occurfrequently and with a relatively small amount ofurine in the bladder. It is clear from such casesthat in the medulla oblongata the fibres associatedwith the postural and reflex activities of the bladdermust lie somewhere in the region between theinferior cerebellar peduncle and the olive, lateral

186



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



I-Tde

=0

wI

at

I ML. OF FLUID RUN INLET ITCOME

FIG. llA and FIG. II B.-Cystometrogramsmade one week before and two weeksafter a unilateral cordotomy. The abilityto control micturition on request isclearly seen to be present after thecordotomy: the pressure rose as soonas the patient was asked to let the urinecome, and it fell again when he wasrequested to stop micturating.

TOTAL AMOUNT500 ML.

224

w4

I--

R-.

wa

W

w

Z

us

_lzI diu)

w

w

U)o 7

AIS-0

In

-4''v

9

m

#A

I-U

30 0

2 7 - tn trK 24 a

O 21-

In

m 12

to 9 _

m 6

3--r

BO 28 56 112 i4O 168

ML. OF FLUID RUN IN

187

I I I I--- -



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from



to the twelfth nerve. This is a region known tocontain many fibres arising in the reticular substance,including reticulo-spinal fibres. At this level themore exact location of the fibres concerned withmicturition cannot be worked out from our material.It may be concluded, however, that the fibres con-cerned do not form a part of the cortico-spinaltract.

ConclusionsThe following conclusions may be drawn from

our evidence. In man, the majority of descendingfibres concerned with micturition lie in the lateralcolumn, on an equatorial plane passing through thecentral canal. This location remains the same inthe cervical, thoracic, lumbar, and sacral segments.These are the fibres mediating conscious controlof the perineal and pelvic musculature, the fibresnecessary for the starting and stopping of micturi-tion, those concerned in inhibiting micturition, andthose that inhibit the independent activity of thesacral bladder centre. It may be concluded thatthese fibres do not run in the median longitudinalbundle, the tecto-spinal tract, nor the vestibulo-spinal tract, for these tracts do not lie in theequatorial plane.The centrifugal fibres run in close association

with the centripetal fibres from the bladder sub-serving the sensation underlying the desire tomicturate, and with the fibres subserving defaeca-tion. We have previously shown (Nathan and Smith,1951, 1953) that the afferent fibres from the bladderand rectum lie in the lateral part of the lateralcolumn, approximately on the plane of the centralcanal. The centrifugal fibres concerned with mic-turition are now shown to lie in the same plane, in aslightly more medial position; in the sacral segments,however, some of the efferent fibres lie on theperiphery of the cord. We have been unable toseparate the fibres concerned with conscious controlfrom those concerned with the automatic, uncon-scious aspects of micturition. We have found someindication of a separation of the spinal fibres sub-serving defaecation from those subserving mic-turition. But the nature of our material-where thelesions made by the surgeon have to be large, andin which many of the patients have carcinoma inthe pelvis-does not allow us to draw satisfactoryconclusions about any differences in location of thefibres subserving these two functions.

DiscussionIt has been shown that in man a bilateral operation

on the spinal cord which divides the region on anequatorial plane passing through the central canalhas an effect upon micturition similar to that pro-

duced by a complete transection. Using the sametechnique, Barrington (1933, 1948) found the tractsrun in the same region in the cat. Czyhlarz andMarburg's (1900) suggestion that these descendingfibres run in man in the neighbourhood of thecortico-spinal tract is supported by our evidence, buttheir belief that " in order to get permanent disorderof bladder function " both cortico-spinal tracts haveto be damaged to a large extent was not justifiedby our experience. Langworthy and Lewis's (1935)conclusion from two cases of disseminated sclerosiswithout post-mortem evidence that these tractsprobably run in the lateral columns of the spinalcord in the neighbourhood of the cortico-spinaltracts is shown to be correct.We observed in some patients more evidence of

recovery than occurs after total division of the cord.Foerster and Gagel (1932) and Foerster (1936) statethat they found this also. They concluded that theremust be other tracts in the cord subserving micturi-tion which had not been divided by the bilateralcordotomy incisions, and they thought that suchtracts must run elsewhere than in the anterior or theantero-lateral columns. The argument, it wouldseem, is not so simple. To show that other efferentfibres are functioning, it is necessary to show thatin all cases micturition recovers more than it wouldafter complete section of the cord. However, evenwith total cord transection, bladder functioningeventually appears to be fairly efficient. And further-more, somewhat better functioning after division ofonly the lateral columns is to be expected than aftertotal cord transection, for it is likely that othermechanisms are at the disposal of these patients,analogous to the trick movements used by patientswith total divisions of peripheral nerves. For in-stance, they may have adequate information from theurethra to tell them that micturition is starting,progressing, and stopping, and they can take stepsaccordingly. In fact, experience has shown us thatsuch patients sometimes have the impression thatmicturition is normal, when in fact all the spinalcord pathways are divided. They may notice onlysome degree of urgency. To the questions, Do youknow when you need to pass urine, can you startmicturition when you want to, can you stop passingurine in the middle of the act, the answer in eachcase can be, yes. And yet the patient may knowwhen he needs to pass urine by paying attention toone of the substitute sensations, having lost thenormal sensation underlying the desire to micturate;he may be able to start passing urine by means ofincreasing the intra-abdominal pressure or flexingforward while pressing his hands on his lowerabdomen, having lost the normal mechanism; andhe may be able to stop passing urine in the middle

188



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


PAlHWAY FOR MICTURITION WITHIN SPINAL CORD

of the act, as he may have that kind of micturitionwhich needs straining to be kept up throughout theact; if he ceases to strain, he ceases to pass urine;but he has lost the normal active contraction of theperineal muscles and sphincters that is needed tointerrupt micturition in the normal way. If in anypatient there is definite evidence that the recoveryof bladder function has gone beyond what is possiblewith total division of the pathways subservingmicturition in addition to the aid provided by anotherwise intact nervous system, then one is justifiedin believing that some pathways are intact. It ismore reasonable to suppose that the fibres whichhave not been divided lie immediately adjacent tothe area of the lesion made at operation than toconclude that they do not lie in the anterior or theantero-lateral columns as Foerster did. The reasonfor saying this is that all the patients we have seenwent through the typical stages of bladder functionthat occur following complete division; if somethen recovered to a more normal condition, it wouldseem that originally all the fibres were affected, butthat some on the periphery of the lesion were onlytemporarily put out of action.

It might be objected to our argument that as inalmost all our cases the centripetal pathway in thecord was divided at the same time as the centrifugalpathways, it is wrong to attribute all the effects we

have described to the division of the centrifugalpathways, for they might be due to an inadequacyof information on the state of the bladder beingreceived at the centres cranial to the spinal cord.But this objection seems to us to be invalid for threereasons. The behaviour of the bladder followingdivision of these tracts is identical with that ofsomatic musculature following division of therelevant descending tracts. Both show a period ofspinal shock, followed by a period of reflex hyper-activity and spasticity, and both show paralysis orsevere paresis of consciously controlled movements.It is known that these effects on the somatic muscu-lature result from the removal of the influence of thedescending tracts. Further, the fact that in thosecases in which the relevant tracts were cut unilaterallythere was an increase in the tone of the bladdermusculature would tend to implicate the descendingtracts as responsible for these changes rather thanthe centripetal tracts; for it seems very likely thatone afferent pathway intact on one side of the cordwould suffice to convey information from the blad-der to cerebral centres adequate for the normalcoordination and control of micturition. Finally,

Case 11, of which the cord is shown in transversesection as Fig. 9, is important from this point ofview; for this patient never regained the normaldesire to micturate, the afferent pathways beingdivided, yet the tone and the reflex activities of thebladder returned to normal; and so here we had acase showing division of the spinal afferent path-ways with normal bladder posture and coordination.

SummaryThe location within the human spinal cord of the

tracts subserving conscious control of micturitionand the tracts subserving coordination of bladderfunctioning is determined.These tracts are within the lateral column through-

out the length of the spinal cord. Their approximatelocation is on an equatorial plane reaching from one

periphery of the cord to the other, and passingthrough the central canal.We wish especially to thank Mr. Wylie McKissock

for his wholehearted cooperation. We are also gratefulto the other neurosurgeons who have helped us andparticularly to Mr. J. O'Connell. We would also liketo thank Dr. E. A. Carmichael who gave us encourage-

ment in this work and provided the facilities for it. Toour technical assistants, Miss Anne Ebborn and Mr.K. L. Frampton, we would like to express our gratitudefor their careful and painstaking work.

REFERENCESBarrington, F. J. F. (1933). Brain, 56, 126.

(1948). In Winsbury-White, H. P. Textbook of Genito-urinarySurgery. Livingstone, Edinburgh.

Budge, J. (1864). Z. rat. Med., 21, 1.Czyhlarz, E. von, and Marburg, 0. (1900). Wien. klin. Rdsch.,

14, 933.

Dennig, H. (1926). Innervation der Blase. Berlin.Denny-Brown, D., and Robertson, E. G. (1933). Brain, 56, 149

and 397.

Foerster, 0. (1936). In Bumke, O., and Foerster, 0. Handbuch derNeurologie. Berlin.and Gagel, 0. (1932). Z, ges. Neurol. Psychiat., 138, 1.

Hunsicker, W. C., and Spiegel, E. A. (1933). Proc. Soc. exp. Biol.(N. Y.), 31, 974.

Langworthy, 0. R. (1940). Res. Publ. Ass. Res. nerv. ment. Dis.,20, 617.Kolb, L. C., and Lewis, L. G. (1940). Physiology of Micturition.Williams & Wilkins, Baltimore.and Lewis, L. G. (1935). Bull. Johns Hopk. Univ., 56, 211.

Levin, P. M., and Langworthy, 0. R. (1937). Amer. J. Physiol.,118, 483.

Mosso, A., and Pellacani, P. (1882). Arch. ital. Biol., 1, 291.Muller, L. R. (1918). Dtsch. Arch. klin. Med., 128, 81.Nathan, P. W. (1952). J. Neurol. Neurosurg. Psychiat., 15, 148.

(1956a). Ibid., 19, 101.-(1956b). Brit. J. Urol., 28, 126.

and Smith, M. C. (1951). J. Neurol. Neurosurg. Psychiat.,14, 262.

(1953). Ibid., 16, 245.Nesbit, R. M., and Lapides, J. (1948). J. Urol. (Baltimore), 59, 726.Sherrington, C. S. (1915). Brain, 38, 191.Spiegel, E. A., and MacPherson, D. J. (1925). Pflugers Arch. ges.

Physiol., 208, 570.Stewart, C. C. (1899). Amer. J. Physiol., 2, 182.Tower, S. S. (1940). Brain, 63, 36.Wang, S. C., and Ranson, S. W. (1939). J. comp. Neurol., 71, 457.Wruck, J. (1943). Z. ges. Neurol. Psychiat., 176, 179.

189



http://jnnp.bmj.com

/J N

eurol Neurosurg P


ownloaded from


the centrifugal pathway for micturition within …part in the control andcoordination ofmicturition....

Documents