Current Topics of Interest in Meningitis

Robert Kacprowicz, MD
Melvin J Marque, MD
Department of Emergency Medicine
Wilford Hall USAF Medical Center, San Antonio, TX.

(Provided in cooperation with EmedHome.com. Go to www.EMedHome.com to take the CME quiz for this article for free category 1 CME credit).

Introduction:

Meningitis remains a complicated problem for emergency physicians. Diagnosis and management are fraught with controversy. Not only is meningitis a highly morbid disease for the patient, but it can also be a cause for serious litigation for the physician.

In this review, we hope to clarify some of the current controversies and topics of interest in meningitis, particularly in the post-Hib immunization era.

Epidemiology and Microbiology:

The introduction of the Haemophilus influenzae type B (Hib) vaccine has caused a dramatic shift in both the epidemiology and the most common pathogens responsible for bacterial meningitis. Prior to the introduction of the Hib vaccine, 2/3 of the cases of meningitis were diagnosed in children less than 5 years of age. In less than 10 years since the introduction of the vaccine, the median age at diagnosis shifted from 15 months in 1986, to 25 years of age in 1995. More importantly, the total incidence of bacterial meningitis has been cut in half.¹ According to CDC data, the rate of invasive Hib disease has decreased from a rate of 40/100,000 children in 1987 to <2 cases/100,000 in 1994 (Fig 1).²

This dramatic reduction has largely shifted the leading cause of meningitis to Streptococcus pneumoniae. In 1986, Hib was responsible for 29 cases of meningitis per million persons (44% of the total cases), and S. pneumo was responsible for 11 cases per million persons (18% of total cases). As of 1994-95, Hib was responsible for only 2 cases per million and S.pneumo remained constant at 11 cases per million. Incidence of meningitis including the top 5 pathogens dropped from 12,920 cases total to 5,755 cases total, a 66% reduction in total meningitis cases.³

A recent review of 12 years experience with meningitis in the post-Hib vaccine era in Alberta, Canada, further illustrates this trend. Culture data showed the causative organism to be S. pneumo in 52% of cases, Listeria monocytogenes in 12.5%, Haemophilus in 7.5%, Streptococcus agalactiae in 4%, E. coli in 4% and Neisseria meningitidis in 2.5%.⁴

A similar experience was reported in New England in 1995, with S. pneumo again the top causative agent, responsible for 47% of cases of meningitis. Interestingly, the same New England data showed N. meningitidis to be the second most common cause, at a rate of 25%. Group B Strep rounded out the top three at a rate of 12%. Overall, there was a 94% reduction in the incidence of H. flu . By age group, the most common causative organisms were: neonates: Group B Strep, Infants 1-23 mos: S. pneumo (45%) and Neisseria (31%), 2-18 y.o.: Neisseria (59%), and Adults > 18: S. pneumo (62%).⁵

Clearly, this data will have a profound effect on clinical practice in the years to come. With the introduction of the polyvalent S. pneumo vaccine, the coming years promise a further reduction in the incidence of bacterial meningitis. For now, however, many of the decisions made in emergency departments must take into account the overall predominance of S. pneumo in bacterial meningitis.

Resistance Trends:

As S. pneumo has ascended to become the predominant cause of bacterial meningitis, concerns over antibiotic resistance have also assumed a greater role in discussions of management. Once considered the drug of choice for bacterial meningitis, penicillin has been supplanted by other antibiotics in light of these trends.

One of the most pressing concerns has been the emergence of regionally high levels of resistance of S. pneumo to penicillin. Penicillin resistant Streptococcus pneumoniae (PRSP) has been reported in many areas, with some regions reporting levels of resistance to penicillin approaching 40% (Connecticut data).⁶ Trends clearly point to increasing resistance over the last decade. Overall levels of resistance in the United States (both intermediate and highly resistant strains) have increased from 14% in 1993-94 to 25% in 1997.⁶

Furthermore, resistance of S. pneumo to ceftriaxone (Rocephin) has also increased. Composite data from 8 states showed an increase in rates of resistance from 1.4% in 1994 to 15% in 1996.⁷

Obviously, these are worrisome developments. Multiple theories have emerged to explain this trend, but by far the most common explanation has been over utilization of antibiotics for non-bacterial infections. Whatever the cause, however, changes in management must be dictated by these trends and will be discussed in the management section of this review.

Of note, local resistance trends vary widely, so it is of utmost importance to know the data for the area in which you practice.

Diagnosis:

One of the many challenges facing emergency physicians remains the successful diagnosis of bacterial meningitis. Case reports detail many varied and unusual presentations of meningitis, but several trends noted from clinical studies can be helpful to the practicing physician.

The classic triad of headache, fever and neck stiffness presents little diagnostic dilemma for the seasoned EP. Present in at least 2/3 of patients, this constellation makes treatment decisions easy.⁴

Of this triad, fever is by far the most common complaint, present in 85% of children and up to 97% of adults.⁸ Neck stiffness is also highly sensitive, apparent in 82-99% of adults.^10,11,12 In children, it is a bit less helpful, but occurs in 78% of children by the age of 2.⁸ Headache also occurs in most cases, but in one report 16% of patients eventually diagnosed with meningitis denied headache at presentation.⁴

Other features may also be helpful. Vomiting is present in approximately 50% of adults.¹⁰ While only 10% of adults present with seizures, up to 30% of children may present in this fashion.⁹

Regarding rash, approximately 10% of cases will present with petechiae. In children, 20% of cases of fever and petechial rash have invasive bacterial disease.⁴

The most challenging cases remain those in which the patient suffers from impaired immunity. Patients with a history of altered immunity, diabetes mellitus, alcohol abuse or prior neurosurgery demand extra vigilance as their presentations are often quite atypical. Further, those with shunts seem to be at increased risk for bacterial meningitis and warrant extra attention.¹³

Laboratory Evaluation:

The standard for laboratory diagnosis of bacterial meningitis remains the CSF culture, but, unfortunately, this time-consuming test is of little value to the emergency physician. Therefore, surrogate markers of infection are critical in the first few hours of diagnosis and treatment of meningitis. Additionally, differentiation of bacterial from aseptic meningitis is of considerable importance, as patients with aseptic meningitis, with a few exceptions (notably herpes) can usually be treated in the outpatient setting.

The classic CSF picture of WBCs > 1,000 with polymorphonuclear leukocyte predominance, elevated protein and depressed glucose is highly predictive of bacterial meningitis, but the ambiguous CSF analysis presents the greatest challenge for emergency physicians. Differentiating bacterial from aseptic meningitis can be a particularly challenging problem and is one which almost everyone practicing emergency medicine for any length of time will experience.

Recent literature experience from both Canada and Pittsburgh cloud the picture and illustrate that clinical impression remains the most important aspect in the diagnosis of meningitis.

In the recent Canadian experience, CSF analyses of bacterial meningitis provide chilling data. Of 103 cases, only 56 had WBC counts greater than 1,000. 35% had counts from 100-1000, and a startling 10% had WBC counts under 100. All but one, however, had protein levels above 45, and 65% had levels of greater than 200.⁴ These cases represent primarily community acquired cases of meningitis in Alberta from 1985-1997. Of note, however, 43 out of a total of 103 patients had received previous antibiotics. This seems to suggest that if a patient has been treated with antibiotics previously, extreme caution is warranted when interpreting the results of the CSF. These patients should generally be admitted until the picture is clarified.

Classic teaching states that those with aseptic meningitis may have an early predominance of PMNs on CSF analysis which shifts to mononuclear cells within 24 hours. This is clearly helpful to the EP in cases which are clinically consistent with aseptic meningitis. However, in less clear cases, repeat LP in 24 hours has been advocated as a diagnostic maneuver to differentiate bacterial from aseptic meningitis. Experience in Pittsburgh, however, calls this strategy into doubt. In their experience of 138 cases of aseptic meningitis, greater than 50% had a PMN predominance more than 24 hours after disease onset.¹⁴ Therefore, it may be prudent to admit any patient with an ambiguous CSF presentation for treatment with IV antibiotics until culture results are available, given the significant morbidity and mortality of bacterial meningitis.

The use CSF bacterial antigen diagnostic testing (BADT) in the emergency department is an appealing concept for early diagnosis and identification of pathogens in meningitis. Numerous techniques exist, all of which are designed to detect the presence of bacteria in the CSF prior to the availability of culture data. The most commonly used technique uses latex particle agglutination. Briefly, latex particles are coated with IgG specific for the polysaccharide capsule of several common meningeal pathogens. When a specific antigen is present in sufficient numbers, a lattice forms, with polysaccharide as the bridge between latex particles. This lattice then becomes visible within 2-3 minutes and can be used as a surrogate marker for the presence of the specific bacterium.²²

Limitations to this technique exist, however, and may limit the utility of LPA BADT. The test is only useful for bacteria, and specifically those that produce large capsular polysaccharides. These include S. pneumo, N. meningitidis A,B,C,Y and W135, H. flu B, and Group B strep.²² Of note, N. meningitidis B does not produce large amounts of antigen, and, therefore, may be missed by LPA testing. Furthermore, the test was not designed as a screening tool but rather as a therapeutic adjunct to be used in those patients who had CSF pictures consistent with bacterial meningitis. Because of this, a significant false-positive rate exists. When used as a diagnostic tool, LPA BADT can actually show more false-positives than true-positives.²² Narchi, et al. found a sensitivity of 57% overall for BADTs, versus 81.5% for gram stain and 80% for culture.¹⁵ The University of North Carolina also found BADTs to be no more sensitive than gram stain.¹⁶ Finally, in a review of 5,169 cases of meningitis, Duke University researchers found no instances in which CSF BADT changed either antimicrobial treatment or clinical management.¹⁷ Based on these data, the routine use of LPA BADT cannot be recommended in the emergency department, and, in fact, a number of laboratories nationwide have discontinued their use.²²

PCR, or polymerase chain reaction, is another technique of pathogen identification which may have a role in the ED, primarily in the detection of viral pathogens. PCR is used to amplify DNA particles present in the CSF and identify the offending organism and is much more sensitive than viral culture.²³

Utility of PCR for Pathogen Identification in Viral CNS Infections

Pathgoen	Sensitivity (% )	Specificity (% )
Herpes simplex virus type 1	>95	100
Cytomegalovirus	80-100	75-100
Varicella zoster virus	N/A	100
Epstein-Barr virus	97	100
JC virus	74-92	92-96
Enterovirus	97	100

(from: Zunt JR - Neurol Clin - 01-Nov-1999; 17(4): 675-89)

Obvious limitations to the routine use of PCR exist. First, PCR is a labor intensive test which is not available in all laboratories at all hours. Second, PCR is known to have a significant false-positive rate which may confuse diagnosis.²³ It may have a role, however, if it is available, as it can provide early diagnosis of those who may benefit from acyclovir therapy for herpetic infections.

Taken in aggregate, these findings underscore what EPs have always known: the clinical picture is the most important diagnostic tool available. No laboratory tests should influence the decision to treat a patient suspected of having bacterial meningitis, nor should they affect the overall management and disposition of the patient.

Treatment:

Early treatment of bacterial meningitis remains the standard of care, and the available literature supports a 30 minute door to antibiotic time standard.¹⁸ Resistance trends discussed previously must influence the choice of antibiotics, however.

The historical mainstay has been, and remains, either cefotaxime or ceftriaxone for bacterial meningitis. Recent resistance data, however, have led to the recommendation that vancomycin be added to empiric therapy until resistance data are available from culture.¹⁹

For certain special situations, other choices or additions may be appropriate. In the neonate and the elderly (>60 y.o.), the addition of ampicillin is warranted due to increased rates of infection with Listeria monocytogenes, enterococci, and Streptococcus agalactiae.¹⁹ For immunocompromised patients, therapy should be expanded to cover for gram negative organisms, including Pseudomonas aeruginosa. Options for therapy in this situation include ampicillin and ceftazidime, or meropenem, with an aminoglycoside.¹⁹ Finally, in those with a recent history of neurosurgery or in the presence of a shunt, extra attention should be directed to Staphylococcus aureus with the addition of oxacillin/nafcillin or vancomycin.¹⁹ If the patient has an altered mental status, focal neurologic signs, or seizures, herpetic encephalitis should be considered. Furthermore, hemorrhagic encephalidites, most notably herpes encephalitis, may result in a significant number of RBCs in the cell count of an otherwise atraumatic lumbar puncture and should alert the EP to the presence of a possibly treatable viral etiology. In these cases, pending PCR results (where available) empiric therapy with acyclovir is warranted as this is the only treatment available to these patients The usual dosage is 10 mg/kg every 8 hours .²⁴

Antibiotics may be further tailored in the inpatient setting as culture data becomes available, however, these choices should provide adequate empiric therapy in the emergency department.

One of the most controversial issues in medicine had been the use of adjunctive steroids for the treatment of bacterial meningitis. Much of the research in management of meningitis has been directed toward the inflammatory cascade, believed to be responsible for much of the morbidity and mortality associated with meningitis. In-vitro studies suggest that much of the pathophysiology associated with bacterial meningitis occurs when bacterial lysis, caused by antibiotic therapy, releases inflammatory mediators, which, in turn, cause increased CSF permeability and decreased CSF outflow. The use of steroids, notably dexamethasone, has been investigated extensively as a possible adjunctive therapy to interrupt this inflammatory cascade.

The majority of studies regarding the use of dexamethasone have been conducted in children. Several studies have shown a decrease in serious neurologic sequelae, most notably bilateral severe hearing loss, when dexamethasone is given either before, or with, antibiotic therapy.²⁰ Unfortunately, the majority of benefit has been shown in the subset of children suffering from invasive H.flu, a pathogen which has been largely eliminated due to the high prevalence of the Hib vaccine. Therefore, the applicability of this data to current practice has rightly been called into question.

A few studies have looked at the use of dexamethasone as adjunctive treatment of pneumococcal meningitis. The largest meta-analysis of dexamethasone therapy found a significant decrease in severe neurologic sequelae in meningitis due to S.pneumo, but the benefit was present only if dexamethasone was given prior to the initiation of antibiotic therapy.²⁰ One study from Egypt was able to show a statistically significant reduction in overall mortality due to S. pneumo with dexamethasone therapy, (13.5% vs 40.7%, p<.002).²¹ These benefits came with no difference between groups in terms of time to defervescence, 24 hour CSF leukocyte counts, glucose or protein.²¹ Although it is difficult to say with certainty based on one study whether dexamethasone actually reduces mortality, it does seem to improve neurologic outcome, but only if given prior to the initiation of antibiotic therapy.

Disposition of those with meningitis is generally straightforward. In those who have clear-cut aseptic meningitis, i.e. those without prior antibiotic therapy, a gradual onset of viral disease syndrome, non-toxic appearance and compatible CSF testing results, no therapy is indicated. These patients may generally be discharged to home with oral pain control if suitable follow-up is available.

Those with a confusing diagnostic picture should be admitted to an inpatient setting pending further work-up, with empiric therapy begun in the ED.

If clear cut bacterial meningitis is present, the patient will most likely require ICU level care for the immediate future.

Antibiotic prophylaxis:

Few diseases inspire more fear among lay persons and hospital personnel alike than meningitis. If a diagnosis of meningitis is made in the ED, rest assured dozens of personal contacts and hospital employees, from physicians to housekeepers will present requesting prophylaxis against meningitis.

With few exceptions, prophylaxis can be handled with a simple rule: only CLOSE contacts need to be treated. Close contacts are defined as "individuals who frequently sleep or eat in the same dwelling with the index case,"²⁵ (including daycare centers and military barracks) or those in contact with patient secretions. Schoolmates are not considered close contacts, even if they share the same classroom (unless seated next to index case), but boy/girlfriends should receive chemoprohylaxis.²⁶

Hospital personnel are not at increased risk, unless they have performed mouth-to-mouth resuscitation or have had contact with the patient’s secretions. During a large outbreak of N. meningitidis group A meningitis in Finland, not one case of secondary meningitis was seen in hospital personnel.²⁷ Therefore, casual contacts, those in the same general area (or ED) as the index case, and those involved in simple patient care do not require antibiotic prophylaxis.

If indicated, chemoprophylaxis can be accomplished with several different antibiotics.

CHEMOPROPHYLAXIS OF MENINGOCOCCAL AND HiB MENINGITIS
TO WHOM?	Patient and close contacts*
ALTERNATIVES	Rifampin (orally)
	ADULTS	600 mg b.i.d. for 2 days
	CHILDREN
	Meningococcal disease	10 mg/kg b.i.d. for 2 days
	Hib disease	20 mg/kg once daily for 4 days
	OR
	Ceftriaxone (intramuscularly)
	ADULTS	250 mg, 1 single dose
	CHILDREN	125 mg, 1 single dose
	OR
	Ofloxacin400 mg, 1 single dose orally (adults)
	OR
	Ciprofloxacin500 mg, 1 single dose orally (adults)
	OR
	Azithromycin500 mg, 1 single dose orally (adults, meningococcal disease)
	OR
	Sulfonamide (orally) If sensitivity is known
	ADULTS	1000 mg b.i.d. for 2 days
	CHILDREN	500 mg b.i.d. for 2 days
	OR
	Minocycline100 mg b.i.d. for 5 days (adults)
SWABBING	Not indicated (not known how to deal with the result)
KEEP CALM	Danger of hysteria

 *Individuals who frequently sleep and eat in the same dwelling with an index case (own family, day care, boy or girlfriend)           

Not primarily recommended for children

(From: Peltola H - Infect Dis Clin North Am - 01-Sep-1999; 13(3))

Clearly the easiest regimen includes a single dose fluoroquinolone, however this is not suitable for a pregnant woman or children, who may be most easily treated with an IM dose of ceftriaxone (250 mg adults/125 mg children).²⁸ Note that the efficacy of fluoroquinolones in preventing HiB disease has not been studied.²⁸ Given the recent low prevalence of HiB disease, however, it is probably safe to assume that this is not a concern.

The most critical part of prophylaxis, however, is education. All persons in contact with the index case should be instructed to see a physician immediately should any signs or symptoms of illness develop during the week to ten days following identification of the index case.²⁸

Conclusions:

Meningitis is a complex problem, both for the patient and the practicing EP. The Hib vaccine has dramatically changed both the epidemiology and microbiology of meningitis in the United States. Due to these changes, S. pneumo has emerged as the most common cause of meningitis today.

With the prevalence of S. pneumo, changes in the treatment of meningitis have become necessary, but the approach to diagnosis has changed little. Above all, the clinical history and exam remain the most important tools for the emergency physician.


References:


¹ Gold R. Epidemiology of Bacterial Meningitis. Infectious Disease Clinics of North America. 13:3, 1999.

² Haemophilus influenzae Serotype b (Hib) Disease. www.cdc.gov. 2001.

³ Spach D and L Jackson. Bacterial Meningitis. Neurologic Clinics. 17:4, 1999.

⁴ Hussein A and S Shafran. Acute Bacterial Meningitis in Adults: A 12 Year Review. Medicine. 79:6, 2000.

⁵ Schuchat A, Robinson K, Wenger J, et al. Bacterial Meningitis in the United States in 1995. NEJM. 337:970-6. 1997.

⁶ Centers for Disease Control and Prevention. Geographic variation in Pencillin Resistance in Streptococcus pneumoniae: Selected Sites, United States, 1997. MMWR Morb Mortal Wkly Rep. 48:656-661.

⁷ Arditi M, Mason E, Bradley S, et al. Three-Year Multicenter Surveillance of Pneumococcal Meningitis in Children: Clinical Characteristics, and Outcome Related to Penicillin Susceptibility and Dexamethasone Use. Pediatrics. 102:5, 1998.

⁸ Kilpi T, Anttila M, Kallis MJT, et al. Severity of Childhood Bacterial Meningitis and Duration of Illness before Diagnosis. Lancet. 338:406, 1991.

⁹ Kornelisse R, Westerbeek C, Spoor A, et al. Pneumococcal Meningitis in Children: Prognostic Indicators and Outcome. Clin Infect Dis. 21:1390, 1995.

¹⁰ Andersen J, Backer V, Voldsgaard P, et al. Acute Meningiococcal Meningitis: Analysis of Features of the Disease According to the Age of 225 Patients. J Infect. 34:227, 1997.

¹¹ Sigurdardottir B, Bjornsson O, Jonsdottir K, et al. Acute Bacterial Meningitis in Adults: A 20 Year Overview. Arch Intern Med. 157: 425, 1997.

¹² Durand M, Calderwood S, Weber D, et al. Acute Bacterial Meningitis in Adults. A Review of 493 Episodes. NEJM. 328:21, 1993.

¹³ Kaplan S. Clinical Presentations, Diagnosis, and Prognostic Factors of Bacterial Meningitis. Infect Dis Clinics Of North Am. 13:3, 1999.

¹⁴ Negrini B, Kelleher K, Wald E. Cerebrospinal Fluid Findings in Aseptic Versus Bacterial Meningitis. Pediatrics. 105:2, 2000.

¹⁵ Narchi H. CSF Antigen Detection Testing in the Diagnosis of Meningitis. Annals of Saudi Medicine. 17:1, 1997.

¹⁶ Kisha D, Jones M, Manzum E, et al. Quality Assurance Study of Bacterial Antigen Testing of Cerebrospinal Fluid. Journal of Clinical Microbiology. 33:1141-1144, 1995.

¹⁷ Perkins M, Mirrett S, Reller L. Rapid Bacterial Antigen Detection is not Clinically Useful. Journal of Clinical Microbiology. 33:1486-1491, 1995.

¹⁸ Tunkel A, Scheld WM. Acute Meningits. Principles and Practice of Infectious Diseases, 4^th ed. Mandell G, Bennett J, Dolin R, eds. New York:Churchill Livingstone, pp. 831-65, 1995.

¹⁹ Saez-Lorens X, McCracken G. Antimicrobial and Anti-Inflammatory Treatment of Bacterial Meningitis. Infectious Disease Clinics of North America. 13:3, 1999.

²⁰ McIntyre P, Berkey C, King S, et al. Dexamaethasone as Adjunctive Therapy in Bacterial Meningitis: A Meta-Analysis of Randomized Clinical Trials Since 1988. JAMA. 278:11, 1997.

²¹ Girgis N, Farid Z, Mikhail I, et al. Dexamethasone Treatment for Bacterial Meningitis in Children and Adults. Pediatric Infectious Disease Journal. 8:12, 1989.

²² Snyder E and P Rainey. The Clinical Microbiology Laboratory Discontinues Routine Bacterial Antigen Testing. Lab News, Yale Department of Laboratory Medicine. 38:1, 1996.

²³ Zunt JR. Cerebrospinal fluid testing for the diagnosis of central nervous system infection. Neurol Clin 17(4): 1999.

²⁴ Roos KL. Encephalitis. Neurol Clin 17(4): 1999.

²⁵ Kaiser AB, Hennekens CH, Saslaw MS, et al. Seroepidemiology and Chemoprophylaxis of Diseases due to Sulfonamide-resistant Neisseria Meningitidis in a Civilian Population. J Infect Dis 130:217, 1974

²⁶ Jacobson JA, Camargos PAM, Ferreira JT, et al: The Risk of Meningitis Among Classroom Contacts During an Epidemic of Meningococcal Disease. Am J Epidemiol 104:552, 1976  

²⁷ Salmi I, Pettay O, Simola I, et al: An Epidemic due to Sulphonamide-resistant Group A Meningococci in the Helsinki Area (Finland). Epidemiological and clinical observations. Scand J Infect Dis 8:249, 1976  

²⁸ Peltola H. Prophylaxis of Bacterial Meningitis. Infect Dis Clin North Am 13:3,1999.