J. clin. Path. (1947), 1, 10.

MENINGOCOCCAL MENINGITIS WITH PARTICULARREFERENCE TO EPIDEMIOLOGY AND

PATHOGENESISBY

R. W. FAIRBROTHERDepartment of Clinical Pathology, Manchester Royal Infirmary

(RECEIVED FOR PUBuCATION, JULY 8, 1947)

In this country cerebrospinal meningitis (fever) isan endemic disease found mainly in children andadolescents; local outbreaks occur, but they are rareand usually involve few individuals. The diseasehas, however, a marked predilection for war-timeconditions; relatively extensive outbreaks occurredduring the 1914-18 war, and again to an evengreater extent during the 1939-45 war. Theseoutbreaks involved individuals of all ages, includingService personnel, but it is interesting to note that,even though there was a great increase in the totalnumber of cases notified, these tended to occursingly and an epidemic, in the generally acceptedsense, did not develop.Although such general factors as the enhanced

virulence of the meningococcus and the resistanceof the individual are well recognized as playing animportant role in the epidemiology of the disease,a precise explanation of the peculiar nature of theoutbreaks is still lapking. A popular theory is thatoutbreaks of cerebrospinal meningitis result from awidespread meningococcal infection of the naso-phayhx in the general population (Ministry ofHealth Memorandum, 1940). Many workers acceptthe hypothesis developed by Dopter (1921) thatmeningococcal infection occurs in three progressivestages: (1) catarrhal, (2) septicaemic, and (3) menin-gitic; in the majority of cases the infection isarrfsted at the catarrhal stage and the occurrenceof meningitis is a rare phenomenon (Lundie andothers, 1915; Herrick 1918; Murray, 1929;Brinton, 1941). The evidence offered to supportthis hypothesis is, however, inconclusive andunconvincing. Furthermore this theory does notprovide a solution of many problems presented inthe various outbreaks, and it seemed desirable thatthe question should be re-examined in the light ofinformation gained during the 1940-42 outbreak.

During the past seven years an opportunity hasbeen available of studying the disease under Serviceand civilian conditions during both so-calledepidemic and interepidemic periods and it waspossible to direct attention particularly to theepidemiology and pathogenesis of the disease.The results are discussed below; they do not supportthe above hypothesis and an alternative is offered.

TechniqueCerebrospinal fluid.-Samples of cerebrospinal fluid

were centrifuged as soon after collection as possible for5 or 10 minutes; the supematant fluid was decantedand the deposit seeded heavily on to Loeffler serumslopes and blood-agar plates, films were then made andthe remainder of the deposit added to 10 ml. of glucose-broth, usually containing a small amount of para-aminobenzoic acid. Incubation was carried out at370 C.; the blood-agar plates were usually placed in anatmosphere containing approximately 10 per cent carbondioxide.

Nasopharyngeal swabs.-Material was collected fromthe nasopharynx by means of the West swab, which waspressed and then stroked firmly against the mucousmembrane. Cultures were prepared on blood-agarplates; after overnight incubation under ordinaryatmospheric conditions, suspicious colonies were sub-cultured on to Loeffler slopes for further investigation;the plates were invariably incubated a further twenty-fourhours and re-examined.Blood cultures.-Ten ml. of blood were collected by

venepuncture; 2 ml. were added to a small amount oftrypsin broth before mixing with 10 ml. agar, 2 ml.to Robertson's meat medium, and the remainder wasplaced in 25 or 50 ml. of glucose broth. Incubation at370 C. was continued for seven to fourteen days before anegative report was given, subcultures on Loeffler serumslopes being made after forty-eight hours and later asindicated.The identification of the meningococcus was carried

out by fermentation tests with sucrose, maltose, and

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

MENINGOCOCCAL MENINGITIS

glucose in Hiss's serum water and finally by agglutinationtests using Standard Group sera. The suspensions forthe agglutination tests were made by emulsifying thegrowth from a Loeffler slope in normal saline and adjust-ing to a light opacity; the tube method with incubationat 550 C. was used. Standard Group sera, from Armyand M.R.C. sources, were used throughout.

Preliminary tests were frequently carried out by theslide agglutination technique; useful information wasobtained when the strains readily agglutinated, but thismethod proved unsatisfactory with relatively inagglutin-able organisms.

Scope of the InvestigationMaterial was collected from various sources during

this investigation, particularly from cases, closecontacts of cases, and members of the generalcommunity (Table I).

TABLE ISOURCE OF MATERIAL

Type of case Number

Meningitis <\ciilicean4.105

Purulent conjunctivitis .. .. .. 2

Chronic septicaemia .. .. .. 3

? Carrier/close contacts .. .. .. 144'general community .. .. 530

Cerebrospinal fluid only was examined from thecivilian cases of meningitis, but the Service caseswere investigated in greater detail. Cerebrospinalfluid was examined from all, while blood culturesand nasopharyngeal swabs were collected from alimited number. The search for carriers wascarried out by the collection of nasopharyngealswabs. Close contacts (Service personnel) wereconsidered to be individuals occupying bedsadjoining the case or all occupants of small roomsor tents in which the case was sleeping (War OfficeMemorandum, 1940). While such a definition didnot embrace all close contacts, it did ensure that allmembers of this group had been exposed to thenasopharyngeal discharge of the patient a shorttime before the collection of the swab.

Clinical Course- and TherapyThe various clinical manifestations of meningo-

coccal infections have been described in detail byRolleston (1919), Priest (1941), and Brinton (1941).The cases seen during this investigation followed

the general pattern. The cases of meningitis duringthe 194042 outbreak were mainly of a severe type,the initial symptoms being headache, neck rigidity,vomiting, and malaise. Some examples of thefulminating meningeal type were seen; thesewithin a few hours of the onset of the diseaseexhibited mental confusion and delirium. Duringthe later stages of the outbreak the mild abortiveform of the disease was occasionally encountered.No examples of adrenal involvement (Waterhouse-Friderichsen syndrome) or of encephalitis (Banksand McCartney, 1942) were seen.

Intensive chemotherapy with a sulphonamide,usually sulphapyridine, was begun immediately thediagnosis was established on the lines recommendedby Banks (1939). Serum was not used. The usualpractice was to examine the cerebrospinal fluid beforechemotherapy was instituted; the fluid Was invari-ably turbid and meningococci generally demon-strable.

In the fulminating types of disease the solubleform of the sulphonamide was administered by theintravenous route.The results were excellent; in the group of 105

Service cases there was only one death. This isvery much lower than the general mortality rate forthe disease, but it must be appreciated that thecases were adults of military age and that therapywas usually instituted early in the disease; it iswell known that the mortality rate is highest at theextremes of life (Jubb, 1943; Beeson and Westerman,1943).

Several cases of chronic meningococcal septi-caemia were suspected on clinical grounds, but inonly three instances was it possible to confirm thediagnosis by blood culture. When this was accom-plished sulphonamide therapy was instituted withthe anticipated dramatic effect.

BacteriologyMeningococci isolated during this investigation

generally behaved in the characteristic mannerwhich was described in detail by Murray (1929).A few interesting features were, however, observed,and these merit special consideration.The standard solid media used throughout the investi-

gation were 6 per cent blood-agar plates and Loefflerserum slopes. Both media gave a satisfactory growthbut it was observed that, under ordinary atmosphericconditions, strains isolated in pure culture directly fromthe cerebrospinal fluid grew readily on the Loefflerserum slopes in screw-cap bottles, while growth onblood-agar plates was not infrequently poor; incubationof the blood agar plate in 10 per cent carbon dioxide,however, gave a luxurious growth. It therefore follows

11

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

R. W. FAIRBROTHER

tlat, if blood-agar is used for the isolation of themeningococcus from cerebrospinal fluid, incubationshould be carried out in an atmosphere containing carbondioxide.

Biochemical tests were limited to the fermentation ofglucose, maltose, and sucrose in serum-water. The greatmajority of strains gave consistent results, glucose andmaltose being fermented but not sucrose; irregularitieswere occasionally found.The agglutination tests followed the generally accepted

rule that strains isolated during an "epidemic" periodagglutinate readily with standard sera, but during"non-epidemic" periods inagglutinable strains tend toappear. During the main 1940-42 outbreak, cerebro-spinal fluid strains agglutinated readily with the StandardGroup sera; many gave clear-cut agglutination onlywith Group I serum, but some did not exhibit any markedantigenic specificity and agglutinated also to varyingdegrees with Group II serum. Many indications of thecomplex antigenic structure of the meningococcus wereobtained during this investigation, but it was not possibleto examine this problem. A simple classification bymeans of Standard Group sera only was attempted.When the large outbreak had subsided and

infrequent sporadic cases only were reported, adifferent serological picture was seen. Thus duringthe 1946-7 winter several severe cases, mainly ininfants, were investigated; meningococci wereisolated from the cerebrospinal fluid but thesetended to agglutinate poorly, usually giving a slightreaction only with the Group II serum; one strainproved to be inagglutinable with the sera available(Table II).Nasopharyngeal swabs gave similar results;

strains isolated during the main outbreak were moreeasily grouped than those obtained during thenon-epidemic period. Throughout the investigationGroup II strains were predominant.

EpidemiologyThe epidemiology of cerebrospinal meningitis

presents several unusual features for which noadequate explanation has been offered. Themarked predilection of the disease for war-timeconditions has long been recognized, and thereforethe sudden widespread outbreak of the disease atthe beginning of 1940 was not unexpected. It isinteresting to note that the incidence of the diseasein this country during 1940 reached the unprece-dented total of 12,771 cases; this figure is actuallygreater than the total figures for the 1914-18 period.There was not, however, an epidemic in the generallyaccepted sense of the term; cases were widelydistributed, occurred sporadically, and involvedindividuals of all ages. Fortunately the sulphon-amides proved excellent therapeutic agents andsome of the serious problems of 1914-18, such as

the high mortality rate and long periods of hospital-ization, did not arise.During the 1914-18 outbreak special research

teams carried out extensive investigations of thedisease and the following important fundamentalpoints were established: many normal individualsharbour the meningococcus in their nasopharynx;cases of cerebrospinal meningitis develop almostentirely from contact with a carrier (instances ofinfection from other cases of the disease occur butthey are relatively rare); the disease is transmittedby droplet infection, and consequently overcrowdingand poor ventilation play prominent roles in thespread of infection.The precise significance of the carrier is uncertain.

Glover (1920) claimed that a carrier rate of over20 per cent in a community was not only an indexof overcrowding but also a warning of an impendingoutbreak of meningitis, and he stressed the import-ance of the search for carriers as a prophylacticmeasure. Subsequent observations have, however,failed to confirm this rather extreme view. Dudleyand Brennan (1934) in a naval garrison found acarrier rate of 13 per cent associated with 11 casesof meningitis, while at a later period the carrierrate was 55 per cent but no case of meningitisdeveloped. Other workers have also found carrierrates over 20 per cent without any increased incidenceof meningitis (Straker et al., 1939). It is, therefore,obvious that the development of meningitis in anindividual is dependent on factors other than thepresence of carriers in the community, and there islittle doubt thai these are closely concerned with(1) the virulence of the meningococcus, and (2) theresistance of the host.

Virulence of meningococcus.-The absence ofsuitable animals has prevented a detailed studyof this problem and it has been necessary to dependmainly on indirect serological tests for information.During the 1914-18 outbreak four serological typeswere identified (Gordon and Murray, 1915); thesetypes were not sharply demarcated and considerableantigenic overlapping occurred. In view of thisother workers (Griffith, 1918; Scott, 1918) con-sidered that the most practical serological classifica-tion was into two main groups-Group I embracingtypes I and III, and Group II comprising types IIand IV. It was generally recognized that inagglutin-able strains were particularly prominent in thenasopharynx during interepidemic periods.The use ofGroup sera is now the accepted method

of subdividing meningococci, but knowledge of theantigenic structure of the meningococcus is stillincomplete and irregularities are not uncommon.

12

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

MENINGOCOCCAL MENINGITIS

In the 1914-18 outbreak, the majority of caseswere caused by types I and II, but since that timethere has been a close relationship between Group Istrains and major outbreaks. It is therefore notsurprising to find that, while Group II strains havepredominated in carriers, some 90 per cent of strainsisolated in this country from cerebrospinal fluidduring the 194042 outbreak belonged to Group I(Harries, 1942; Ministry of Health Report, 1946).During this investigation there was striking

serological difference in the strains isolated duringthe 1940-42 outbreak and those isolated from theinfrequent cases seen during the 1946-7 winter(Table II).

TABLE IISEROLOGY OF MENINGOCOCCI ISOLATED FROM

CEREBROSPINAL FLUID

1940-2 1946-7

Group1 .. .. .. .. 86 1

Groupll .. .. .. .. 3 9

Inagglutinable .. .. .. 1

Total .. .. .. .. 90 11

The close association of Group I strains withmajor outbreaks or "epidemics" suggests stronglythat such strains are more virulent than those ofGroup II. It is, however, the general experiencethat there is little or no difference in the severity ofthe disease produced by these strains. It must beappreciated that normal cerebrospinal fluid is anexcellent culture medium for most bacteria and,provided a sufficient number is introduced, multi-plication will occur irrespective of the intrinsicpathogenicity of the organism.An important difference in the two serological

groups of meningococci is therefore the greaterfacility with which Group I strains are able to passfrom the nasopharynx to the cerebrospinal fluid.The factors concerned with the virulence of themeningococcus are obscure, but they appear to beassociated with the surface, probably the capsule,of the organism. The production of toxic sub-stances by the meningococcus has been demon-strated by various workers (Branham, 1940);knowledge of these is, however, vague, and it hasnot been proved that they play any part in thepassage of the organism to the subarachnoid space.The low incidence of meningitis in spite of the

wide distribution of the meningococcus indicatesclearly that the passage from the nasopharynx to

the meninges is not readily accomplished. Animportant reason for this is the barrier to invasionpresented by the nasopharynx; the responsiblefactors consequently merit close consideration.

Resistance of the host.-During the 1939-45war conditions favoured the wide dissemination ofthe meningococcus, particularly in the Services, asovercrowding was common and ventilation ofrooms often inadequate as a result of stringentblack-out precautions. Although there was agreatly increased incidence of meningitis during themain outbreak of 1940 42, the majority of caseswere single and widely distributed, having no directcontact with each other. Most individuals would,therefore, appear to possess a high degree ofresistance against the meningococcus, and thefactors concerned in this have been the subject ofmuch speculation.Many advocate the theory that infection with the

meningococcus occurs in three progressive stages:(1) catarrhal, (2) septicaemic, and (3) meningitic;the common form of the infection is a nasopharyn-gitis, and only in relatively few cases does theorganism penetrate the pharyngeal barrier to reachthe blood stream, from which the meninges areattacked. The nature of this pharyngeal barrierhas, however, never been satisfactorily explained.The so-called catarrhal stage is obviously the one

of greatest epidemiological significance in view ofthe potential danger to the community. It isconsequently important that the criteria on whichthis theory is postulated should be criticallyexamined.The main arguments put forward to support this

theory are: (1) the isolation of the homologousmeningococcus from both the nasopharynx andcerebrospinal fluid of early cases (Flack, 1917),(2) the history of premeningitic sore-throat orcatarrh given by some cases, and (3) the widespreaddistribution of carriers.These views have not passed unchallenged, and

the following points have been raised in opposition:(1) the failure of many workers to isolate regularlythe meningococcus from the nasopharynx of earlycases of meningitis (Andrewes and others, 1916);it is also interesting to note in this connexion thatFildes and Baker (1918) reported the occurrenceof meningitis in twenty-six naval ratings althoughtheir nasopharyngeal swabs had been negative atvarying periods from two to seventy-five days beforethe onset of symptoms, and concluded that casescan seldom be carriers before the onset ofthe disease;(2) a history of sore-throat or catarrh is frequentlyunobtainable from cases or carriers (Foster and

13

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

R. W. FAIRBROTHER

Gaskell, 1916; Worster-Drought and Kennedy,1919); (3) cerebrospinal meningitis occurs mainlyduring the early months of the year when upperrespiratory infections are common, and theirassociation with meningitis is likely to be coinci-dental (it hag even, been suggested that this non-specific catarrh might constitute a predisposingfactor to the development of meningitis); (4) case-to-case infection is extremely rare; the source ofinfection is almost invariably an undetected carrier;(5) in the majority of instances the carrier state istemporary.As a reply to these arguments protagonists of the

catarfhal theory claim that the local meningococcalinfection is generally mild and transient, giving riseto little or no personal discomfort, but clinical andbacteriological confirmation is seldom produced.

There is general agreement that in every case ofmeningitis the portal of entry of the meningococcusis the nasopharynx. There is, however, a lack ofconvincing evidence to indicate that the productionof a nasopharyngitis, even in a mild form, is anessential stage in the development of meningitis.It must be pointed out that the examination ofnasopharyngeal swabs from early cases is rarelycarried out, and knowledge of this important aspectof the subject is consequently limited.During this investigation the validity of the

"catarrhal"'theory was examined by the collectionof swabs from the following groups of thecommunity:

1. Close contacts.-These were soldiers sleeping inbeds adjoining the case, or all occupants incases arisg in tents or billets. All had hadample opportunity of receiving the upperrespiratory flora from the cases as overcrowdingwas often present and ventilation, particularlyat night, usually poor; the swabs were collectedwithin a short time, often a few hours, of theremoval of the case.

2. General community.-Swabs were collected frompersons having no direct contact with cases inorder to obtain some indication of the generalcarrier rate. These have been sub-divided intotwo groups-one contaiing swabs collected atthe time of the main outbreak (1940-42) andthe other comprising those taken during an"interepidemic" period (1946-47).

3. Cases of meningitis.-Swabs were collected from22 cases, in 16 of these within 48 hours of onset.Group I strains were isolated from the cerebro-spinal fluid of 13 cases but no growth wasobtained from the other 9, 3- of which tere ofthe mild, abortive type. The nasopharyngealswab from one of these mild cases gave a heavygrowth of a Group II strain.

4. Cases ofchronic meningococcal septicaemia.-Twoonly were investigated.

The results obtained from these swabs are pre-sented collectively in Table III

TABLE IIITHE ISOLATION OF THE MENINGOCOCCUS FROM NASO-

PHARYNGEAL SWABS

Num-Group ber Number GopIGop1exa- positive Grp I Grop II

ined

Close contacts 145 21 (14%) 6 (4%) 15 (10°/.)(1940-41)

General com-munity .. 410 75 (18%) 11(3%) 64 (15%)(epidemic period1940-42)

General com-munity 120 8 (7%/O) 1 (1%) 7 (6%)(non-epidemicperiod 1946-47)

Cases of menin-gitis 22 6 (27%) 4 (18%/) 2 (9%/0)Cases ofchronicsepticaemia 2 1 0 1 *

*A few colonies of a Group II meningococcus were isolated fromthe nasopharyngeal swab some 21 days after the onset of symptomsa Group I strain was isolated by blood-culture.

These results provide little bacteriological evidenceto support the "catarrhal" theory. Meningococciwere isolated from the nasopharynx of only 6 of22 cases and in the majority of these the organismswere not numerous; the 1940-42 outbreak wascaused almost entirely by Group I strains, butduring this period the majority ofcarriers harbouredGroup II strains; there was no significant differencein the carrier rates of Group I strains given by theclose contacts (4 per cent) and the general com-munity (3 per cent), in both the rates were low;few of the cases or carriers gave a definite history ofsore throat or catarrh.The meningococcus is transmitted by droplet

infection, and therefore in all persons developingmeningitis these organisms must have been presentin the nasopharynx during the initial stages of theinfection. Their subsequent isolation from this siteis dependent not only on the interval betweenimplantation of the organisms and the collection ofthe swab but also on the nature of the local reaction.If a local infection develops it is reasonable toexpect (1) the isolation of meningococci from thenasopharyngeal swabs in many early cases-notinfrequently in practically pure culture, and (2) arelatively high carrier rate of the infecting type of

14

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

MENINGOCOCCAL MENINGITIS

organism among close contacts, as an interchangeof the upper respiratory flora takes place readilyamong occupants of overcrowded and badlyventilated rooms. Such results have rarely beenobtained.

It is recognized that a small number of carriersare of the "persistent" type and that the persistenceof the carrier state is associated with some pharyn-geal abnormality (Cleminson, 1918; Embleton andSteven, 1919). During this investigation persistentcarriers were found to have some abnormality, suchas enlarged adenoids, and the majority harbouredGroup II strains often in relatively pure culture.These organisms were usually associated with achronic form of local infection, but they were oflittle epidemiological significance as the main out-break was caused by Group I strains; they werereadily cleared by a course of sulphonamides(Fairbrother, 1940).The failure to detect the homologous meningo-

coccus in swabs from the two chronic septicaemiacases is not surprising. Logan (1946), in an excellentreview of meningococcal septicaemia, points outthat the pathogenesis of the disease is uncertain;there is no bacteriological proof that the portal ofentry of the organisms is the same in meningococ-caemia as in meningitis, although this is probable,while a persistent focus of infection has seldombeen found. In the two cases under review, theswabs were collected when the disease was wellestablished and the negative results indicate onlythat the focus of the existing infection was not thenasopharynx.

PathogenesisThe path of the meningococcus from the naso-

pharynx to the cerebrospinal fluid has not beenconclusively determined, and in consequence it hasbeen the subject of much speculation. Two mainviews have been formulated; the first accepts thedirect extension of the meningococcus from thepharynx to the meninges, the second postulatesthat the organisms are transmitted by way of theblood stream.The theory of direct extension was advocated

during the 1914-18 outbreak but convincing proofwas not produced. There has consequently beenincreasing support for the "haematogenous" theory,which is now generally accepted (Brinton, 1941;War Office Memorandum, 1942; Strong, 1943;Banks, 1947). It must, however, be appreciatedthat the available evidence to support this view isinconclusive and merits examination.The main arguments advanced in favour of the

"haematogenous" route are: (1) the development

of a so-called premeningitic stage with involvementof the blood stream; (2) the rare isolation of theorganism by blood-culture during the premeningiticphase while the cerebrospinal fluid was normal;(3) the isolation of the meningococcus by bloodculture from early cases of meningitis; and (4) theoccurrence of a septicaemic form of meningococcalinfection in which there is no involvement of themeninges-these infections may occasionally be of afulminating character.

It has already been pointed out that a satisfactoryexplanation has not been produced to indicate howthe meningococcus enters the blood stream from thenasopharynx. Considerable doubt has, moreover,been cast on the "catarrhal" theory and there islittle evidence to indicate that there is a persistentfocus of infection in the nasopharynx from whichthe organisms pass into the blood. Also a satis-factory explanation of the method or route bywhich the meningococcus passes from the blood tothe meninges has not been forthcoming. In thisrespect, it is interesting to note that there is adefinite blood-central nervous system barrier; thepassage of substances through the choroid plexusor other vessels into the cerebrospinal fluid is notautomatic but is controlled by some form of selectivemechanism. It is well known that penicillin injectedinto the blood does not readily reach the theca;while experimental infection with the poliomyelitisvirus is extremely difficult to produce by the intra-venous route, although it occurs readily by theintrathecal or intraneural route.

There is no evidence to suggest that the menin-gococcus has any special affinity for the choroidplexus or other part of the theca. It is indeedaccepted that in most cases of meningococcalsepticaemia the development of meningitis does notoccur even though, in pre-sulphonamide days, theblood infection often persisted for months.

Moreover, a positive blood culture in the earlyand acute stages of a severe infection cannot beaccepted as evidence of primary blood invasion.In such infections as pneumonia and enteric fevera positive blood culture is frequently present in theearly stages as a secondary phenomenon, being anoverflow of the organisms from the primary focusof the infection. It is, therefore, reasonable toclaim that the positive blood culture obtained inmany cases of meningococcal meningitis has been asecondary development.A more serious argument is the rare occurrence

of a positive blood culture before the developmentof meningitis; it must, however, be noted that inseveral of these cases the primary focus of infectionwas not established, the presence of a rhinopharyn-

15

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

R. W. FAIRBROTHER

gitis being assumed, and/or the presence of menin-gitis was not excluded by lumbar puncture.The occurrence of a definite premeningitic clinical

stage can also be challenged. Cases of moderateseverity tend to give a twenty-four-hour history ofgeneral malaise, weakness, headache of varying butusually progressive intensity, and slight neck rigidity.On lumbar puncture a turbid fluid is almost invari-ably obtained from these cases; this is proof thatthe invasion of the meninges occurred many hourspreviously. In fulminating meningitic cases, aturbid fluid is obtained by lumbar puncture eventhough the history of the infection is short, usually amatter of hours; the possibility of such infectionsdeveloping in progressive stages (viz., catarrhal andsepticaemic) is remote.There is no specific sign or symptom to indicate

involvement of the meninges; the initial symptomsof moderate meningeal infections are often those ofan indefinite character, and such cases are notinfrequently diagnosed tentatively as influenza. Thepresence of meningitis can be established only byan examination of the cerebrospinal fluid. Thedevelopment of a premeningitic phase of the infec-tion cannot therefore be accepted on clinicalgrounds alone.

In order to test the "haematogenous" theory,blood cultures were prepared as early as possible inthe infection from a number of cases, usually atthe time of collection of the cerebrospinal fluidand nasopharyngeal swabs. The results are givenin Table IV.

TABLE IVInOLAT[6N OF TIE MENINGOCOCCUS FROM THE BLOODAND CEREBROSPINAL FLUID IN CASES OF MENINGMS

Time of No. No. C.S.F.collection examined positive positive in

Within 24 hours ofonset .. .. 12 1 12

24-48 hours of on-set .. .. 8 1 6

Later than 48 hoursof onset.. .. 5 1 3

Total .. 25 3 21

The meningococcus was isolated from the bloodstream in only 3 out of 25 cases (12 per cent). In20 of the 25 cases the blood was collected withinforty-eight hours of the onset of the disease; inall cases a. well-established meningitis was presentand meningococci were isolated from the cerebro-spinal fluid of 21 of the 25.

These results are not consistent with the develop-ment of a premeningeal septicaemic stage, butindicate merely the secondary invasion of the bloodstream from the primary infection of the meninges.A positive blood culture, obtained when meningitisis present and the cerebrospinal fluid contains largenumbers of meningococci, cannot be accepted asevidence of either the passive transfer by the bloodstream of the organisms to the subarachnoid spacefrom an indefinite focus of infection in the naso-pharynx or the presence of a distinct premeningealphase.

It is interesting to note that the majority ofpositive blood-cultures from cases of cerebrospinalmeningitis, as distinct from chronic septicaemia,reported in the literature have been obtained whenthe meningitis was well established.

DiscussionThe results of this investigation do not provide

evidence, clinical or bacteriological, to support thewidely accepted theory that the meningococcusproduces a primary pharyngeal catarrh, from which,in relatively few cases, it passes to the meninges byway of the blood stream.

This is not surprising, as evidence advanced tosupport this theory is far from conclusive. Toaccount for the failure of many cases to provide ahistory of sore-throat or catarrh, it was suggestedthat the local pharyngeal infection was oftentransient and trivial, but bacteriological or clinicalproof of this infection was seldom obtained.Moreover there has been no satisfactory explanationof the method by which the organisms enter theblood stream from such a trivial primary lesion orby which they subsequently pass from the bloodstream to the meninges.

In view of the fundamental weaknesses of thehaematogenous theory it is essential that an alternatemethod of spread, namely by direct extension,should be carefully examined. This theory receivedmuch support during the 1914-18 outbreak, whenthe following routes were suggested: (1) by theperineural lyniphatics of the olfactory nervesthrough the cribriform plate of the ethmoid; (2)through the sphenoidal sinuses; (3) via the middleear. Convincing evidence to support any of theseroutes was not produced, and this theory conse-quently became neglected.

In 1929, however, Clark produced conclusiveproof that material could pass directly to the brainfrom the nasal passages. Investigating,the develop-ment of nervous complications following vaccina-tion, Clark showed that a solution of potassium

16

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

MENINGOCOCCAL MENINGITIS

ferrocyanide and iron ammonium citrate reachedthe subarachnoid spaces of the brain of rabbitswithin one hour of being dropped into their nasalcavities. The route of spread was not by thelymphatics but by the perineural sheaths of theolfactory nerves, which provide continuity betweenthe subarachnoid space and the olfactory sensoryepithelium; in these spaces a centripetal flow waspostulated. It was also found that some of thesolution entered the blood capillaries. On treatmentof the tissues, the deposit of Prussian blue in thenasal mucosa was found to be considerably moreabundant in the olfactory area, which is innervatedby the olfactory nerves, than in the "respiratory"portion which is not reached by the olfactorynerves. Clark considered that a similar sequente ofevents, apart from variations in degree of absorptionfrom the mucosa, was probable in man, and suggestedthat these findings had an important bearing on thepassage of infective material from the nasal cavitiesto the brain.

It is also interesting to note that, during experi-mental work on poliomyelitis, Faber (1933) demon-strated that the hairs of the olfactory nerves lie freein the olfactory portion of the nasal mucosa andare covered only by mucus. In spite of this it isimportant to note that intranasal application of thepoliomyelitis virus into the nares of monkeys doesnot invariably lead to the development of experi-mental infection even though the monkey is arelatively susceptible animal (Fairbrother andHurst, 1930). The poliomyelitis virus passes to thecentral nervous sytem along the axons of theolfactory nerve. This is a different route from thatdescribed above, but the area of nasal mucosa atwhich the neural tract is entered is probably thesame in both cases.

There is thus reliable evidence of definite butcomplex communication:tiacpeen the nasal cavitiesand the central nervous system; these are situatedin the olfactory portion of the nasal mucosa andoccupy a relatively small area on the septal andlateral walls of the superior nasal meatus. Passageof infective material along the olfactory tract isnot, however, readily accomplished.

Applying these observations to the pathogenesisof meningococcal meningitis, the following hypothe-sis is postulated.The meningococcus is widely disseminated by

droplet infection and gains entrance to the upperrespiratory tract by the nostrils and mouth. Thesubsequent sequence of events is dependent onmany factors, particularly important being thevirulence and the number of the organisms, their

implantation on the olfactory portion of the nasalmucosa, and the subsequent overpowering of thelocal defence mechanism.

In the majority of individuals these requirementsare not fulfilled; the meningococcus becomes atransient member of the pharyngeal flora, probablyas a commensal, and the individual becomes atemporary carrier. Under circumstances particu-larly favourable to the meningococcus, the localdefence barrier is overcome and the organisms may,by direct extension, reach either the theca orpossibly the blood capillaries; in the latter eventthe septicaemic type of infection might result.Many factors are concerned with the tendency

for outbreaks to appear under war-time conditions.During the war 1939-45 opportunities for thedissemination of the meningococcus were greatlyincreased as a result of overcrowding and poorventilation, which tended to be accentuated byblack-out precautions, while in certain groups, suchas new recruits, general resistance was probablylowered by fatigue and exposure, especially duringthe winter. It is also interesting to note thatGroup I strains were responsible for the mainoutbreak and there is much indirect evidence toindicate that these strains are more virulent thanGroup II.A possible explanation for occurrence of sporadic

Group II infections, found mainly in infants andchildren, during inter-epidemic periods is the closecontact of infants with adults, in particular mothersand nurses, and the consequent greater risk offrequent and large dosage with these strains.Adults may possess some form of specific resistanceto the meningococcus, but there is no convincingevidence that the development of specific immunityplays a major role in the defence against menin-gococcal infections. It is, however, possible thatthe local defence mechanism is less efficient in infantsthan in adults, and is more readily overpoweredby the relatively avirulent Group II strains.

SummaryThe results of this investigation do not support

the theory that meningococcal infections occur inthree successive stages-catarrhal, septicaemic, andmeningitic. Support is given to the theory of directspread of infection from the nasopharynx to thetheca by the perineural sheaths of the olfactorynerves.

I am indebted to many colleagues, in particularDr. D. C. Liddle, for assistance in the collection ofmaterial for this investigation, and wish to offer mysincere thanks for their help.

17

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from

R. W. FAIRBROTHER

REFERENCESAndrewes, F. W., Bulloch, W., and Hewlett, R. T. (1916). Med.

Res. Cncl. Sp. Rep. Ser., No. 2. P. 29.Banks, H. S. (1939). Lancet, 2, 921.Banks, H. S. (1947). Bull. Min. Hlth. Publ. Hlth. Lab. Serv., 6, 138.Banks, H. S., and McCartney, J. E. (1942). Ibid., 1, 219.Beeson, P. B., and Westerman, E. (1943). Brit. med. J., 1, 497.Branham, S. E. (1940). Bact. Rev., 4, 59.Brinton, D. (1941). "Cerebrospinal Fever." Livingstone, Edin-

burgh.Clark, W. E. Le Gros. (1929). Min. Health Rep. Publ. Hlth. Med.

Sub., No. 54.Cleminson, F. J. (1918). Brit. med. J., 2, 51.Dopter, C. (1921). "L'Infection M6ningococcique." Balliere et Fils,

Paris.Dudley, S. F., and Brennan, J. R. (1934). J. Hyg., Ca.mb., 34, 525.Embleton, D., and Steven, G. H. (1919). Lancet, 2, 682.Faber, H. K. (1933). Medicine, 12, 83Fairbrother, R. W. (1940). Brit. med. J., 2, 859.Fairbrother, R. W., and Hurst, E. W. (1930). J. Path. Bact., 33, 17.Fildes, P., and Baker, S. L. (1918). Med. Res. Cncl. Sp. Rep. Ser.,

No. 17.Flack, M. (1917). Med. Res. Cncl. Sp. Rep. Ser., No. 3, p. 31.Foster, M., and Gaskell, J. F. (1916). "Cerebrospinal Fever,"

University Press, Cambridge.

Glover, J. A. (1920). Med. Res. Cncl. Sp. Rep. Ser., No. 50,p. 133.

Gordon, M. H., and Murray, E. G. (1915). J. R.A.M.C., 25, 411.Griffith, F. (1918). J. Hyg., Camb., 17, 124.Harries, G. E. (1942). Brit. med. J., 2, 423.Herrick, W. W. (1918). J. Amer. med. Ass., 71, 612.Jubb, A. A. (1943). Brit. med. J., 1, 501.Logan, W. R. (1946). Edinb. med. J., 53, 183, 235.Lundie, A., Thomas, D. J., and Fleming, S.'(1915). Brit. med. J., 1,

466.Ministry of Health Memorandum No. 234 (1940). "Cerebrospinal

Fever."Ministry of Health (1946). "Report of Chief Medical Officer, 1939-

1945."Murray, E. G. D. (1929). Med. Res. Cncl. Sp. Rep. Ser., No. 124.Priest, R. (1941). J. R.A.M.C., 76, 249.Rolleston, Sir H. (1919). Lancet, 1, 541, 593, 645.Scott, W. M. (1918). J. Hyg., Camb., 17,191.Straker, E., Hill, A. B., and Lovell, R. (1939). Rep. Pub. Hlth. Med.

Sub. No. 90.Strong, P. S. (1943). Amer. J. med. Sci., 206, 561.War Office Memorandum (1940 and 1942). "Cerebrospinal Fever

among Troops."Worster-Drought, C. C., and Kennedy, A. M. (1919). "Cerebro-

spinal Fever." Black, Ltd., London.

18

copyright. on F

ebruary 6, 2022 by guest. Protected by

http://jcp.bmj.com

/J C

lin Pathol: first published as 10.1136/jcp.1.1.10 on 1 N

ovember 1947. D

ownloaded from