J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

Association ofClinical Pathologists Broadsheet 121. June 1989

Investigation of meningococcal disease

Meningococci cause a spectrum of disease, ranging sion to hospital. This is optimal management and is to from asymptomatic colonisation of the nasopharynx, be encouraged, but it makes the task of the laboratory through virulent bacterial meningitis, to fulminating somewhat more difficult. Emphasis has been given in septicaemia. The organisms also cause septic arthritis, this Broadsheet to methods that may assist in the conjunctivitis, and pneumonia. Meningitis is the com- isolation of an organism in these circumstances. monest clinical syndrome seen, occurring predomin- antly in infants and young children, though teenagers Specimens and adults may also be affected, particularly during epidemics. Meningococci vary in their fastidiousness; CEREBROSPINAL FLUID specimens must be transmitted as rapidly as possible Though about 10% of cases of invasive meningococ- to the laboratory. cal disease have septicaemia without clear evidence of Strains with decreased sensitivity to penicillin have meningitis, a combination of the two conditions is by been reported from Spain, South Africa, and the far the most common presentation. Examination of United a producing strain cerebrospinal fluid (CSF) is likely to provide Kingdom'"3; P-lactamase copyright. causing meningitis has also been reported from South immediate confirmation of the diagnosis for the Africa.2 It is therefore imperative that careful sen- clinician in about 90% of cases of meningitis; CSF is sitivity testing is undertaken on all clinically important also the specimen most likely to yield a meningococcus isolates. for sensitivity testing and epidemiological typing. Meningococci are grouped according to the antigenic nature of the capsular polysaccharide. Cell count

Groups B and C are currently most prevalent in the Make a cell count in a Neubauer counting chamber in http://jcp.bmj.com/ United Kingdom while strains of group A are com- the routine way. If desired (and if the specimen is of mon in Africa and the Middle East. Groups Y and sufficient size) a stain such as Azure A MacNeal, which WI 35 are associated with only a very small percentage aids differentiation of white cell nuclei, may be used. of infections. Organisms may be typed and subtyped Add one drop of stain to 0 5 ml of CSF. Mix and leave by variations in the outer membrane proteins (porins). for five minutes before charging the counting cham- Sensitivity to sulphonamide is a further useful dis- ber. Count in the usual way; the dilution factor is not important. A neutrophil leucocytosis is commonly criminatory test. Changes in the epidemiology of on September 26, 2021 by guest. Protected meningococcal disease occur constantly and all clini- seen with counts varying between a few cells and up to cally important isolates should therefore be sent to a 40 000 cells/mm3 CSF. Red blood cells are often also reference laboratory for full identification and sen- present in smaller numbers, even in a specimen sitivity testing. Meningitis is a notifiable disease, collected atraumatically. though substantially undernotified at present.4 The laboratory is likely to be aware of most local Chemistry cases. Good liaison with clinical colleagues and with A portion of the specimen should be reserved for the consultant responsible for communicable disease estimation oftotal protein and glucose. CSF protein.in control enables a clinical diagnosis of meningococcal bacterial meningitis is usually greater than 0 45 g/l. meningitis to be confirmed as rapidly as possible, with CSF glucose is compared with a blood glucose sample follow up of contacts in the community where neces- taken at the same time; the normal CSF glucose should sary. be 70% or more of the blood glucose value an An increasing proportion of cases of suspected relative reduction is usually seen in bacterial menh meningococcal infection are being treated with gitis. antibiotics by the general practitioner before admis- Microscopyfor organisms Accepted for publication 12 January 1989 If organisms are abundant in the CSF they may be 634 J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

Association ofClinical Pathologists; Broadsheet 121, June 1989 635 seen during the cell count, though this is unusual in zae type b,7 but was less sensitive in our hands than meningococcal disease. microscopy in meningococcal meningitis due to group One or two loopsful ofneat CSF are transferred to a B organisms. The group B polysaccharide is a poor clean microscope slide, fixed, and Gram stained. At antigen and detection systems for the polysaccharide the same time a larger aliquot ofCSF is centrifuged at are less sensitive than with groups A, C, Y and W135. 2000 rpm for five minutes. The supernatant is There is even less information on the acute examina- aspirated aseptically to a sterile container and tion ofother fluids such as blood and urine but they are preserved at - 25°C. (It may be used for viral culture in likely to be less satisfactory owing to the lower the event of bacterial culture proving negative). The concentration of meningococcal antigen. deposit after centrifugation is used to make a second In general, latex agglutination is the simplest and Gram film and to inoculate culture plates. Either or most convenient of these methods as well as being at both films may be examined according to the concen- least as sensitive as enzyme linked immunosorbent tration of cells in the original specimen. assay (ELISA) and better than CIE. The limulus assay Careful examination of the Gram film is the most is technically demanding. important step in examination of a CSF specimen exhibiting neutrophil pleocytosis. Up to 80% of cases Culture in which a meningococcal aetiology is subsequently Inoculate blood agar and chocolate agar plates. A confirmed will show Gram negative diplococci if drop of centrifuged deposit should be placed on each examined carefully. Examination may take half an plate and allowed to dry in before being spread with a hour or longer if'bacteria are very scanty. Often loop. This is particularly important in patients who bacteria are predominantly intracellular with have received parenteral antibiotics before admission occasional cells showing one to dozens of pairs of to hospital as the area of original inoculation may diplococci while most polymorphs contain no bacteria retain an inhibitory concentration ofantibiotic. Plates at all. If the patient has received antibiotics more than should be incubated in 5% carbon dioxide at 37°C for an hour or two before the lumbar puncture was 48 hours, though almost all meningococci will grow

undertaken swollen degenerating pairs of diplococci after overnight incubation. Even though plates have copyright. and a greater proportion of single organisms may be been inoculated in the middle of the night, careful seen. Occasionally it may be difficult or impossible to examination with a hand lens the following morning be certain that some Gram negative, intracellular may show developing colonies. inclusions are truly bacteria. In 5% or more of patients with clinically suspected Alternative staining techniques, such as Sandiford's meningococcal meningitis the organism is grown from modification ofGram's stain in which malachite green a CSF specimen which contains no cells and in which is used as a counterstain,5 or simple staining with organisms are not seen (Cartwright KAV, unpublished http://jcp.bmj.com/ methylene blue may be tried, but probably do not add observations). CSF protein and glucose are also greatly to the chances ofproviding a definitive answer normal in these cases. Negative CSF findings must not for the clinician on a difficult specimen. be used to defer antibiotic treatment in a patient with headache and a suspicious rash. A pure lymphocytic Antigen detection cellular response in CSF, normally associated with If there are polymorphs in the CSF and organisms are viral meningitis, is occasionally seen in culture positive not seen after careful examination of a Gram film, meningococcal meningitis (and is not uncommon in on September 26, 2021 by guest. Protected meningococcal meningitis is a more likely diagnosis meningococcal septicaemia, resulting from meningeal than haemophilus meningitis in children aged between irritation without invasion). 3 months and 3 years. In these circumstances examina- tion of CSF for the presence of bacterial antigen may Blood culture be tried. Various methods for the detection of free antigen in CSF are available including latex agglutina- At best, blood cultures are positive in only a third to a tion (Wellcogen, Wellcome Diagnostics; Directigen, halfofthe untreated patients, despite the presence ofa Becton Dickinson), enzyme immunoassay (Pharmacia characteristic meningococcal rash often accompanied Diagnostics), and counter-current immunoelectro- by positive CSF on microscopy or culture, or both. phoresis (CIE). The limulus amoebocyte assay which Despite this surprisingly low figure blood cultures are detects endotoxin, denoting infection with a Gram a mandatory part of the investigation ofpatients with negative organism, has also been used.6 Good com- suspected meningococcal disease either septicaemia parative trials ofthe sensitivity and specificity of these or meningitis. Meningococci will often grow only in methods in meningococcal disease are lacking. Latex the aerobic bottle ofa two-bottle blood culture system. agglutination seems to be superior to both CIE and the Ideally aerobic cultures should be incubated in an limulus test in meningitis due to Haemophilus influen- atmosphere of 5% carbon dioxide for at least five J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

636 Association ofClinical Pathologists; Broadsheet 121, June 1989 days, though most positive results will be obtained are used the plates should be inoculated immediately within the first 48 hours. In patients who have received after the swab has been taken, although a transport parenteral antibiotics before admission to hospital the medium that contains charcoal (Amies) achieves chances of a positive blood culture are remote.89 acceptable but reduced survival.'0 Using the Bactec radiometric blood culture system we have found occasional strains which have failed to PERNASAL SWAB use sufficient labelled substrate to give a positive Pernasal swabs are necessary in small children and reading but which have survived. We therefore also in older children and adults who are so disorien- routinely subculture daily the aerobic bottles from tated as to be unable to cooperate with the collection patients suspected of having meningococcal disease of a postnasal swab. A pernasal swab consisting of a whether or not they give a positive reading. Isolates small cotton wool pledget mounted on flexible wire in obtained from blood cultures should be identified in a blue tube (Medical Wire & Equipment Co., Cor- the same way as for strains from CSF. sham, Wiltshire) is needed. The child's head should be steadied by one nurse while a second nurse holds the Throat swab patient's hands. The swab is passed gently via either Though the isolation of a meningococcus from the nasal orifice directly backwards towards the occiput. throat does not confirm a diagnosis of invasive The posterior nasopharyngeal wall is encountered at a meningococcal disease, it may be a useful finding in a depth ofabout two inches. The swab is quickly rotated patient with suggestive symptoms. It offers a good and withdrawn. The procedure is moderately irritant chance to obtain an isolate for sensitivity testing and and babies and small children will try to pull the swab for epidemiological purposes in two particular situa- out if not prevented from doing so. If possible two tions. The first is when the patient has been given swabs should be obtained so that one may be used for antibiotics before the collection of diagnostic virus culture. specimens. Blood cultures will almost certainly be Postnasal or pernasal swabs should be plated out negative and even if a lumbar puncture is performed immediately on to a selective medium such as

about 30-40% of all cases of invasive meningococcal Modified New York City medium or GC agar medium copyright. disease will not yield a positive culture from blood or to suppress other components of the normal oro- CSF. The second situation, encountered increasingly pharyngeal flora. often, is when a lumbar puncture is not undertaken. Lumbar puncture is contraindicated in the presence of Smears from purpuric skin lesions raised intracranial pressure, usually diagnosed clin- ically by the finding of papilloedema on fundoscopy. We do not have personal experience ofthis technique, Some clinicians, however, take signs of increasing rarely used nowadays. Nevertheless, it seems to give http://jcp.bmj.com/ confusion and disorientation, or progression towards positive results in a substantial proportion of cases'2 a comatose state, as evidence of raised intracranial and may permit the demonstration oforganisms from pressure and will not undertake a lumbar puncture in patients with negative blood and CSF cultures. The these circumstances. A well taken throat swab then method below is that of Tompkins.'3 assumes great importance. Swabs can be collected A purpuric skin lesion is pinched between finger and most reliably by the medical microbiologist, ensuring thumb to exclude circulating blood. The lesion is correct swabbing technique and rapid plating of the "picked" with a sterile hypodermic needle or fine on September 26, 2021 by guest. Protected specimen. A postnasal (transoral) swab is preferable'" scalpel blade. More pressure is applied to squeeze out a but often difficult to obtain in the ill patient and in drop oftissue fluid and blood which is then smeared on young children. Because up to 25% of young adults a glass slide. Several small smears of3-4 mm are better may carry nasopharyngeal meningococci a throat than one large one. After fixation in methanol the isolate will need to be grouped and typed before the smears are better stained with Wright's or Giemsa possible relevance of the strain can be assessed. stain than with the Gram stain. Dilute Giemsa stain (one part of stock to 50 parts of buffer at pH 7-2 is POSTNASAL (TRANSORAL) SWAB recommended for optimal differentiation of bacteria The patient is asked to open the mouth as wide as and polymorph nuclei). Organisms stain blue-black possible and the swab is swept over the posterial and can be detected easily within polymorphs. pharyngeal wall directly behind the uvula and avoid- Extracellular organisms should be ignored as possible ing the tonsillar areas. Fluffy charcoal impregnated contaminants. swabs are preferred as meningococci do not remain viable on plain swabs for very long; they can be used in Confirmation of identity of presumptive colonies conjunction with Stuart's transport medium" but immediate plating gives better results. If plain swabs Meningococci appear as 2-k4mm grey, round entire J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

Association ofClinical Pathologists; Broadsheet 121, June 1989 637 colonies after 24-48 hours' incubation. An oxidase test staphylococcus may be used instead for testing sensi- using aqueous 1% tetramethyl-p-phenylenediamine tivity to penicillin, rifampicin, and chloramphenicol. prepared freshly should be performed. A known, The Oxford staphylococcus is not suitable as a control oxidase positive, control organism should be used. A for sulphonamide sensitivity testing and a known Gram film of a colony should show characteristic sensitive meningococcus must be used. Gram negative diplococci. Low level penicillin resistance will not be detected Colonies of meningococci produce y-glutamyl by this method but ,B-lactamase production (if and aminopeptidase as a preformed enzyme which when it occurs in the United Kingdom) should be hydrolyses y-glutamyl p-nitroanilide to release yellow apparent. Clinically important resistance to the other p-nitroaniline. This forms the basis of a commercial agents should also be detectable. Neisseria identification kit (Gonocheck I1 -EY Laboratories Inc., San Mateo, California, USA). GROUP AND SEROTYPE DETERMINATION This kit will differentiate satisfactorily strains of N Cultures may be grouped by bacterial agglutination lactamica, N gonorrhoeae, and N meningitidis. A using monoclonal or polyclonal anti-polysaccharide second tube containing a chromogenic a lactam is used antibodies (DMRQC, Difco, or Wellcome). These to detect ( lactamase production. Careful determina- occasionally yield misleading results. Latex prepara- tion of the penicillin minimum inhibitory concentra- tions for antigen detection may also be used to obtain a tion should be considered ifthere is any indication that presumptive result. The group may be confirmed and the penicillin sensitivity of an isolate from a systemic the serotype determined by the reference laboratory. infection is reduced. SEROLOGY RAPID CARBOHYDRATE UTILISATION TEST Invasive meningococcal disease is accompanied by an (RCUT) antibody response detectable in a variety ofways using Growth on a medium containing glucose induces tests with antigens ofdifferent specificities. While these carbohydrate degrading enzymes which may be used tests are not routine, and may only be available at

to confirm the identity of putative meningococcal reference laboratories, they may occasionally be of copyright. colonies in a rapid test. (Appendix). value to confirm the nature of an infection for epidemiological purposes, or to investigate patients MENINGOCOCCAL REFERENCE LABORATORIES thought to have some immunological deficiency. All strains of meningococci isolated from deep sites (blood, CSF etc) should be sent to a meningococcal STORAGE OF ORGANISMS reference laboratory (Appendix). This is important to Meningococci can be stored in the short term on provide epidemiological data on the incidence of Dorset egg slopes. Strains should be inoculated on to http://jcp.bmj.com/ meningococcal disease and also to monitor the slopes, incubated overnight at 37°C, and then held at prevalence of meningococcal serogroups, serotypes, 30°C. Viability should be maintained for at least six and drug resistance in the United Kingdom. Menin- months. Strains on Dorset egg slopes are suitable for gococcal disease is substantially undernotified and the sending through the post. For longer term storage submission of strains to reference laboratories liquid nitrogen is satisfactory; scrape the growth from provides invaluable additional data. an overnight non-selective culture plate into Oxoid

nutrient broth with 16% added glycerol. on September 26, 2021 by guest. Protected SENSITIVITY TESTING Freeze dried cultures remain viable for 10-15 years. Sensitivity testing presents problems in laboratories which encounter only a few meningococcal isolates Appendix each year. Disc sensitivity testing is inherently unrelia- ble. The method below is acceptable for occasional NON-SELECTIVE MEDIA use, but results should be regarded as provisional and Blood agar: Columbia agar base with 5% sterile should be confirmed as soon as possible by determina- defibrinated horse blood. tion ofantibiotic minimum inhibitory concentrations, Chocolate agar: as for blood agar but with preferably by a reference laboratory. chocolated horse blood. Oxoid DST agar CM261 with 5% lysed horse blood and Mast antibiotic discs with the following concen- SELECTIVE MEDIA trations should be used: penicillin 1 pg; chloramphen- A GC agar icol 10 ig; rifampicin 5 pg; sulphafurazole 5 pg. The Oxoid GC agar base: CM367 Stokes method'4 with a rotary plater is used. A Oxoid sterile GC Supplement: SR56 documented, fully sensitive strain ofmeningococcus is Contains yeast autolysate 10-0 g/l the best control, but if none is available the Oxford dextrose l-5g/l J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

638 Association ofClinical Pathologists; Broadsheet 121, June 1989 sodium bicarbonate 0- 15 g/l To make Modified NYC Agar: vancomycin 3 0 mg/l Melt 300 ml starch agar; cool to 65'C. colistin 7 5 mg/I Mix with 200 ml NYC base. nystatin 12500 units/i trimethoprim 5 0 mg/l Pour plates. Suspend 36 g of Oxoid GC agar base in 880 ml of Final concentrations of antibiotics: distilled water and bring gently to the boil to dissolve trimethoprim 3 pg/ml the agar. Autoclave at 1 OC for 20 minutes. Cool to vancomycin 3 pg/mI 50°C. Add 100 ml Iysed sterile defibrinated horse colistin 6 pg/ml blood and two phials of sterile GC supplement reconstituted with 10 ml sterile distilled water to give RAPID CARBOHYDRATE UTILISATION TEST final concentrations as above. Mix gently to avoid Purity plate trapping air bubbles. Oxoid GC agar base 100 ml Vitox solution I ml B Modified NYC agar 10% glucose I ml (a) NYC Base (to make 1000 ml base) Constituents: 5% lysed horse red blood cells300 ml Sugar solutions distilled water 330 ml 10% w/v solutions of lactose, glucose, maltose and horse serum 300 ml sucrose in sterile distilled water prepared in advance yeast extract 0 5% (Difco) 62-5 ml and stored as 1 ml volumes in bijoux at - 20C. Use trimethoprim 10 mg/ml 0-76 ml BDH extra pure maltose to avoid trace contamination vancomycin 10 mg/ml 0-76 ml with glucose. For use distribute 20 p1 of 10% sugar colistin 10 mg/ml 1 87 ml solutions into screw-capped phials and store at glucose 12-5 g - 20°C. peptone (Evans) 37-5 g

K2HPO4 11-5 g Buffer solution copyright. KH2PO4 0 65 g 01MK2HPO4 40ml NaCI 12-5 g 0-1MKH2PO4 12ml L-glutamine 0 5 g 8% KCI 100ml Method: 1% aqueous phenol red 10 ml Make 5% lysed red cells; clarify with Seitz grade 5 Sterile distilled water 838 ml filter. Make bulk ofmedium in bucket; clarify with Distribute as 20 ml volumes in universals and store at Seitz grade 5 filter; add red blood cells. - 20°C, For use distribute in bijoux as 2 5 ml volumes http://jcp.bmj.com/ and store at - 20°C. Reclarify total medium using Seitz grade S9 filter. Check pH; adjust to 7-2 if necessary. The test Sterilise by filtration, such as Carlson Ford 10 litre Inoculate a single colony of a suspected meningococ- filter holder with EKS filter pad (No XE 675). cus on to modified GC agar and incubate overnight. Bottle aseptically in 200 ml volumes. Take from the deep freeze the appropriate number of sugars and a bijou ofbuffer solution. Distribute 0 6 ml on September 26, 2021 by guest. Protected Prepare test batch of finished medium using final of buffer into a labelled clean tube. Make a heavy bottle. suspension ofthe test organism in the buffer (at least a Store at - 20°C. loopful). Dispense 120 pi of suspension into each sugar. Incubate in a 37°C waterbath for four hours. (b) NYC Starch Agar (to make 3000 ml) Positive reactions are indicated by development of a Distilled water 2800 ml yellow colour. Tubes remaining red are negative. Each Davis agar 60 g batch of sugars is controlled with a known menin- Autoclave to dissolve. gococcus and gonococcus but control organisms are Distilled water 200 ml not used with each individual test. Meningococci give Starch log positive reactions with glucose and maltose, negative Boil to dissolve. reactions with lactose and sucrose. Mix agar and starch. REFERENCE LABORATORIES Bottle in 300 ml amounts; autoclave at 121°C for England 20 minutes. Public Health Laboratory, Withington Hospital, Store. Manchester M20 8LR J Clin Pathol: first published as 10.1136/jcp.42.6.634 on 1 June 1989. Downloaded from

Association ofClinical Pathologists; Broadsheet 121, June 1989 639 Scotland Duguid JP, Marmion BP, et al eds. The practice of medical microbiology. Vol 2. Edinburgh: Churchill Livingstone, Department of Bacteriology, Ruchill Hospital, Glas- 1975:36-7. gow G20 9NB 6 McCracken GH. Rapid identification of specific etiology in meningitis. J Paediatr 1976;88:706-8. We are grateful to Mr R Shannon, FIMLS, Bristol 7 Hoban DJ, Witwicki E, Hammond GW. Bacterial antigen detec- tion in cerebrospinal fluid of patients with meningitis. Diagn Public Health Laboratory, for the method of prepara- Microbiol Infect Dis 1985;3:373-9. tion of modified NYC medium. 8 Ferguson R. The effect of antibiotic therapy before admission to hospital on recognition of the causative organism in acute KAV CARTWRIGHT,* bacterial meningitis. J Infect 1986;13:241-4. 9 Bohr V, Rasmussen N, Hansen B, et al. 875 cases of bacterial DM JONES meningitis: Diagnostic procedures and the impact ofpreadmis- *Public Health Laboratory, sion antibiotic therapy. Part III of a three-part series. J Infect Gloucestershire Royal Hospital, 1983;7:193-202. Great Western Road, 10 Olcen P, Kjellander J, Danielsson D, et al. Culture diagnosis of meningococcal carriers. Yield from different sites and influence GlQucester, GLI 3NN of storage in transport medium. J Clin Pathol 1979;32:1222-5. Public Health Laboratory, 11 Stuart RD. Transport medium for specimens in Public Health Withington Hospital, Manchester, M20 8LR bacteriology. Public Health Report (Washington) 1959;74:43 1. 12 McLean S, Caffey J. Endemic purpuric meningococcus bacteremia in early life. The diagnostic value of smears from the purpuric lesions. Am J Dis Child 1931;42:1053-74. 13 Tompkins VN. The diagnostic value of smears from purpuric References lesions of the skin in meningococcic disease. JAMA 1943;123:31-2. I Van Esso D, Fontanals D, Uriz S, et al. Neisseria meningitidis 14 Stokes EJ, Ridgway GL. Clinical bacteriology. 5th ed. London: strains with decreased susceptibility to penicillin. J Paediatr Edwin Arnold, 1980:215. Infect Dis 1987;6:438-9. 2 Botha P. Penicillin-resistant Neisseria meningitidis in Southern Africa. Lancet 1988;i:54. This Broadsheet has been prepared by the authors at the 3 Sutcliffe EM, Jones DM, El-Sheikh S, et al. Penicillin-insensitive invitation of the Association of Clinical Pathologists who meningococci in the UK. Lancet 1988;i:657-8. reserve the copyright. Further copies ofthis Broadsheet may be copyright. 4 Cartwright KAV, Stuart JM, Noah ND. An outbreak of menin- gococcal disease in Gloucestershire. Lancet 1986;ii:558-61. obtained from the Publishing Manager, Journal of Clinical 5 Quoted by Duguid JP. Staining Methods. In: Cruickshank R, Pathology, BMA House, Tavistock Square, WCIH 9JR. http://jcp.bmj.com/ on September 26, 2021 by guest. Protected