Meningitis Podcast

What's New in the Management of Bacterial Meningitis?

Felicia C. Chow, MD, MAS; Rodrigo Hasbun, MD, MPH


May 09, 2023

This transcript has been edited for clarity. For more episodes, download the Medscape app or subscribe to the podcast on Apple Podcasts, Spotify, or your preferred podcast provider.

Felicia Chow, MD: Hello, I'm Dr Felicia Chow. Welcome to Medscape's InDiscussion series on meningitis. Today, we'll discuss what's new in the management of bacterial meningitis with Dr Rodrigo Hasbun. Dr Hasbun is a professor in the division of Infectious Diseases at McGovern Medical School at UTHealth. Welcome to InDiscussion, Rodrigo, it's fantastic to have you on today. I don't think you and I have met before, but I've certainly seen your work and read a lot of your studies, and it's really wonderful to have you on today talking about bacterial meningitis. I was wondering if you could start by talking about how the epidemiology of community-acquired bacterial meningitis has changed in the United States over the past few decades. Are pneumococcal and meningococcal meningitis still the most common causes of community-acquired bacterial meningitis among adults?

Rodrigo Hasbun, MD, MPH: In the last four decades, there's been a shift in the epidemiology of community-acquired bacterial meningitis in the United States. The introduction of the Haemophilus influenzae conjugate vaccine in 1985, and then the pneumococcal conjugate vaccine and the meningococcal conjugate vaccines have changed the epidemiology substantially. H influenzae was the main cause of bacterial meningitis in the 1980s, and it has now become very rare. Now, pneumococcal and meningococcal meningitis are still number one, but the epidemiology has shifted from kids to adults. In the 1980s, the median age of bacterial meningitis were infants with H influenzae. Now the median age is around 45, and pneumococcal leads the way with meningococcal in second place.

Chow: We saw a patient with H influenzae meningitis not too long ago. I have to admit, I was really blown away. Like you said, it's something that we don't see all that much anymore. It was a real surprise to us when we made that diagnosis. Are there any emerging pathogens that we need to be aware of as potential causes of bacterial meningitis?

Hasbun: Other bacterial pathogens that can cause bacterial meningitis are staphylococcus (staph) and viridans streptococci, especially Streptococcus intermedius, that can cause bacterial meningitis in the United States. In some parts of the world, like in Vietnam, one of the most common causes of bacterial meningitis is Streptococcus suis that accounts for up to one-third of cases, especially in patients who are in contact with pigs or work with slaughterhouses in Southeast Asia.

Chow: Very interesting. I want to switch now and talk about the approach to diagnosis of bacterial meningitis. We talk a lot about which patients should get head imaging before lumbar puncture (LP), particularly because of concerns that head imaging inevitably delays initiation of antibiotics, which is strongly associated with worse outcomes in bacterial meningitis. I think there remains some disagreement in terms of the approach to this. Also, I get the sense that what we generally do in the United States may not quite mirror what's done in other countries, even in Europe. My general sense is that in the United States we tend to be a little more conservative in terms of whom we image before LP. What are the big categories that you think about in terms of patients who absolutely have to get imaging before LP? And what do you consider to be some of the softer indications for imaging before LP?

Hasbun: The Infectious Diseases Society of America (IDSA) guidelines that were published in 2004 were very clear in which patients should get an image before the LP. These are patients who are immunosuppressed, have focal neurological symptoms, present with altered mental status or new-onset seizures, or papilledema. Despite these recommendations in the United States, there's still a lot of overuse of cranial imaging, even in patients who don't have an indication. We did a study of 614 adults here in Houston, and we saw that two-thirds of the patients who got a CT scan before an LP did not have any of those indications, and none of them had any major intracranial abnormalities. In contrast, in Europe, there are different recommendations. The European Society, the UK guidelines, and the Scandinavian guidelines all differ in what they recommend. Depending on which guidelines you look at, instead of just being altered mental status, one recommends the Glasgow Coma Scale (GCS) less than 12, the other one recommends the GCS less than 10, and the Scandinavian guidelines recommend a CT before LP only if you have coma and focal neurological signs. There are more stringent criteria in Europe. Compliance, even in Europe with those guidelines, is around 50%. Compliance is definitely an issue. Immunosuppression is something that has been dropped from two of those four guidelines in Europe. We're doing a follow-up study right now, and we're seeing that patients who are immunosuppressed and have no other indications still can have some major intracranial abnormalities, so we advocate following the IDSA guidelines as much as we can.

Chow: It sounds like we should continue to rely on the IDSA guidelines in terms of the categories of patients whom we should be imaging before LP. What about in terms of specific studies that you send from cerebral spinal fluid (CSF) in patients presenting with an acute meningitis picture? For example, do you routinely send lactate from CSF, and does lactate from CSF help you to distinguish between bacterial and viral meningitis?

Hasbun: There's a lot of serum and CSF biomarkers that can be of big utility in differentiating viral vs bacterial meningitis. Lactate is one of them. Serum procalcitonin and C-reactive protein (CRP) have been studied and showed good diagnostic accuracy in trying to distinguish who has viral vs bacterial meningitis. But the reality is they're not being used in clinical practice as much as we would like. For example, we did a study looking at healthcare-associated ventriculitis where CSF lactate is an important predictor of having infectious meningitis, and it was only done in 50% of patients. CRP was only done in 1% of our patients. They are not being implemented, at least in our hospital system, but they can be of big use to differentiate viruses from bacteria.

Chow: Why do you think we're not using some of these studies as much as we potentially could be, given the fact that they can be helpful, at least in certain situations, in distinguishing between different categories of pathogens that can cause meningitis?

Hasbun: The IDSA guidelines are almost 20 years old. They have not been updated, and I think a new revision of the IDSA guidelines would strongly recommend those serum biomarkers because now there's much more data. I think it's awareness of clinicians in the emergency room and clinicians taking care of patients. Bacterial meningitis has become a rare entity, so physicians are not well versed in seeing those patients. I think it's a matter of awareness of clinicians.

Chow: The next thing I want to turn to is how to interpret CSF results in patients who have received antibiotics prior to LP. We know that CSF can sterilize quickly, even within hours after administration of antibiotics for some of the more common causes of bacterial meningitis, including, of course, Streptococcus pneumoniae. I think we always try to drill into our trainees that it's important to know whether a patient has received antibiotics before a LP to really interpret the results that you're seeing. If CSF cultures are sterile for a patient, but they saw antibiotics before the LP, can you talk about how the rest of the CSF profile and the clinical status of the patient might help you to tease apart whether a patient really has bacterial meningitis and, importantly, whether you might consider stopping empiric antibiotics for that patient?

Hasbun: Excellent question. There are several studies, one by Dr Nigrovic at Boston Children's, where she looked at the impact of the antibiotic therapy on the CSF, Gram stain, culture, and profile. She saw that there's an impact on the sensitivity of the Gram stain and culture but not on the profile. We also did a very large study here in Houston of 326 patients, and we validated those results. It decreases sensitivity of the Gram stain and culture, but does not affect the CSF white count, glucose, and protein. We just published a very large study that we derived and validated a model with the help of a Dutch meningitis cohort in clinical micron infection this year that created a model to differentiate viral vs bacterial meningitis with 99.5%-100% sensitivity. This model included CSF white count, CSF neutrophil count, CSF protein, CSF glucose, serum white count and fever — things that are widely available worldwide that can help you distinguish those who have viral vs bacterial meningitis despite receiving antibiotic therapy before.

Chow: In terms of, for example, the CSF glucose, is that something where after receiving antibiotics, that should normalize more quickly typically than, say, the pleocytosis in the CSF? Is that something that you can still rely on in terms of making the distinction between bacterial and viral meningitis?

Hasbun: Yes. There's only one study, to my knowledge, looking at repeat spinal taps in bacterial meningitis, and it was done by Dr van de Beek and Matthijs Brouwer from the Netherlands, and they actually looked at CSF at baseline and then went to 3 days later, 4 days, 5 days, and all the way to 14 days. They could see a trend in how the CSF white cell, glucose, and proteins behave. It takes a while for the glucose, protein, and CSF white count to get better. In patients who just got antibiotics within a day or so, you can still reliably utilize those CSF profiles to make an informed decision.

Chow: Do you ever find yourself in a situation where cultures are sterile in someone who's received antibiotics prior to LP and they're starting to improve clinically? And you're not entirely sure, even with the information that you have from the profile, so you end up treating for a full course empirically for bacterial meningitis because of the uncertainty of the cause of the process?

Hasbun: I think that has decreased the amount of the patients whom I need to do that for, because now we have multiplex PCR, and we have Streptococcus pneumoniae antigen and CSF that's also 99% sensitive and specific to rule in or rule out the most common cause of bacterial meningitis. That's pneumococcal. Even if you have prior antibiotic therapy, your PCR or your pneumococcal antigen would be positive. If you add the multiplex PCR or the pneumococcal antigen in context of the model, I would feel very comfortable stopping antibiotic therapy in that patient, especially if you are able to diagnose a viral etiology or something else to account for the meningitis. That's why having a complete workup of what the etiology of your meningitis is paramount, because you want to deescalate, you want to treat appropriately, and you want to stop acyclovir if the patient doesn't have herpes or varicella-zoster virus (VZV).

Chow: We'll switch gears again and talk about treatment for bacterial meningitis. In the United States, the general recommendation is to start vancomycin and a third-generation cephalosporin, usually ceftriaxone, for empiric treatment of bacterial meningitis in individuals between the ages of 2 and 50 who are not known to be immunocompromised. This is not uniform in all practice settings and differs from guidelines in other countries. I think a lot of people think that vancomycin covers methicillin-resistant Staphylococcus aureus (MRSA), but that's not at all why vancomycin is used. Could you remind us of the use of vancomycin as part of the empiric regimen that we use?

Hasbun: Sure. This is based on a rabbit model that was done by Dr George McCracken at UT Southwestern in Dallas, where he saw that using vancomycin and ceftriaxone together was synergistic to treat rabbits with ceftriaxone-resistant pneumococcal meningitis. And because ceftriaxone-resistant pneumococcal meningitis is very rare in several European countries, you can see that the UK guidelines and the European guidelines state that only in countries where the resistance rate to ceftriaxone is greater than 1% you should add vancomycin. In the United States, our ceftriaxone rates are greater than 1%, and that's why we use vancomycin and ceftriaxone empirically until you get your cultures. If it's susceptible to penicillin, you can definitely downgrade to penicillin.

Chow: In terms of the addition of ampicillin in the United States as part of our empiric antibiotic regimen in patients older than 50 years of age is when we are recommended to add that for Listeria coverage, some guidelines I know in other countries push that age higher to 60 years of age and older. Are there any other populations that are potentially at higher risk for Listeria, for whom you would also add ampicillin?

Hasbun: Yes. Listeria meningitis does happen in patients who have cellular immunodeficiency. Patients who are on chronic steroids, more than 20 mg of prednisone a day for a month. Also, patients who have AIDS with low CD4 counts, patients who are solid-organ transplant recipients, or anybody else who is on any medication that lowers your cellular immunodeficiency. Those patients should empirically start ampicillin until you get your culture results.

Chow: Let's talk about dexamethasone. A question that comes up all the time is in terms of the use of dexamethasone for bacterial meningitis and the timing of initiation of dexamethasone in patients with bacterial meningitis. We know from the trial that was published decades ago in 2002 in The New England Journal of Medicine that dexamethasone in high-resource settings like the United States improves outcomes in individuals with bacterial meningitis and, in particular, pneumococcal meningitis. In that particular study they gave the dexamethasone about 15 minutes before antibiotics, so that is the recommendation as well. But in practice, this doesn't always happen. Do you think there's still benefit in terms of giving dexamethasone after antibiotics have been initiated? How do you approach that in your own practice or when this question comes up for patients you're seeing in the hospital?

Hasbun: Yes, dexamethasone can definitely decrease mortality and adverse outcomes, not only in pneumococcal meningitis but also meningococcal meningitis. The recently published study from Diederik van de Beek showed that it also decreases mortality in Listeria meningitis. This is in contrast to a study done by the French in the MONALISA trial where they showed an increased mortality. Now, at least with Listeria, there's this new study from Dr van de Beek that shows decreased mortality with steroids. Steroids should be used empirically on all adults with bacterial meningitis. Unfortunately, like you stated, the timing is a problem with the 15-minute window. We have a study here that we presented at IDWeek last year showing only 15% of our patients get steroids within 15 minutes. And we have a study in Clinical Infectious Diseases in 2017 that shows that only one-third of patients with pneumococcal meningitis in the US get steroids within 24 hours. Steroids are underutilized, unfortunately. The IDSA guidelines recommend 15 minutes before the antibiotic, just based on the clinical trials, the European guidelines recommend up to 4 hours based on expert opinion, and the UK guidelines recommend up to 12 hours. The UK guidelines base their recommendations on a study that was done in Critical Care Medicine 2003 by a group in France where they were able to show that if you give steroids within 12 hours, you could decrease mortality in adults with pneumococcal meningitis in the ICU. So, ideally, the earlier, the better. Considering the risk-benefit ratio, I would go up to 12 hours like the UK guidelines recommend.

Chow: What about in individuals who have bacterial meningitis and evidence of systemic infection and sepsis, for example? Is it also safe to apply these same guidelines for that population in terms of using high-dose dexamethasone?

Hasbun: Yes. The big proportion of the patients with pneumococcal meningitis came in with sepsis and bacteremia, and there was a decrease in mortality in those patients. I would definitely not shy away if the patient is septic and the primary source of sepsis is pneumococcal meningitis.

Chow: We focused most of our time on community-acquired bacterial meningitis. But I was hoping in the last couple of minutes we could also talk about health-care-associated bacterial meningitis and post-neurosurgical infection patients. Are there any data to suggest that it makes sense to use dexamethasone in those populations of patients?

Hasbun: I'm not aware of any clinical trials or large studies looking at steroids in healthcare-associated ventriculitis. That's a completely different syndrome. I've used steroids in healthcare-associated ventriculitis very seldomly. They're not being used. To my knowledge, there are no large studies looking at steroids in healthcare-associated meningitis.

Chow: What about the selection of an empiric antibiotic regimen for healthcare-associated meningitis? Can you talk through how you think about those patients and make a decision in terms of what type of empiric regimen to use?

Hasbun: The epidemiology, the microbiology of healthcare-associated ventriculitis is either staph, strep, or Gram negatives like Pseudomonas aeruginosa and Enterobacteriaceae. The recommendation is to start either vancomycin for MRSA and then either cefepime or meropenem to cover Gram negatives, including Pseudomonas. Once you have your cultures, you can adjust accordingly. But vancomycin and cefepime or vancomycin and meropenem are the most used regimens here.

Chow: We talked a little bit about immunocompromised patients earlier and the use of ampicillin in some of those patients to cover Listeria. What about a patient who's immunocompromised potentially coming in with community-acquired bacterial meningitis? What's the thought process in terms of the empiric regimen that you would use for a patient like that?

Hasbun: It would be the same. If you have somebody coming in with bacterial meningitis that's community acquired, vancomycin, ceftriaxone — and then, if the patient's immunosuppressed, ampicillin — that's the recommendation. That's what's usually done. And then, based on your PCR and your cultures, you can de-escalate.

Chow: In terms of the approach to treatment, I want to pick your brain a bit on the role of intraventricular and intrathecal antibiotics in the treatment of meningitis. I think that this question tends to come up in patients who have a shunt, and there may be concern of a shunt infection. I was hoping you could speak generally on what the role is, if any, of using antibiotics administered intrathecally or intraventricularly meningitis.

Hasbun: In the IDSA guidelines published in 2017, it's recommended to use intraventricular or intrathecal antibiotics in those who are failing intravenous antibiotics. You start with vancomycin and cefepime or vancomycin and meropenem. If the patient's not improving, especially if the bacteria is multidrug resistant like Acinetobacter, which is resistant to a lot of different intravenous antibiotics, you need to do intravenous colistin or intravenous gentamicin. We see this being done the most is in multi-drug resistant Gram-negative rods. The other thing is that the hardware, such a shunt, has to be removed. Any device must be removed to cure healthcare-associated ventriculitis.

Chow: Finally, it would be great to hear about whether there are any new or novel therapies being considered or tested for the treatment of bacterial meningitis. I think there is a trial of adjunctive daptomycin for pneumococcal meningitis, maybe more for the potential anti-inflammatory effects of daptomycin. Anything that's coming down the pike that we have to look forward to for treatment?

Hasbun: Daptomycin has been used in the rabbit model. Stephen Leib at the Institute for Infectious Diseases in Bern was the first one doing the studies. Daptomycin is a bacteriolytic antibiotic, so it doesn't generate inflammatory cytokines, and that's why in comparing daptomycin vs ceftriaxone in the rabbit model, daptomycin decreases hearing loss and especially apoptosis in the hippocampus. The problem is that if you use that with steroids, then you're blunting the effect of daptomycin. Also, daptomycin doesn't have excellent central nervous system penetration. There's been a lot of failures with daptomycin in healthcare-associated ventriculitis. There is a study that's being enrolled in Europe looking at that, but the results are not out.

Chow: Wonderful. Thank you so much for joining us — it's been such a pleasure today. We've had Dr Hasbun discuss the management of bacterial meningitis. Thank you for joining us. This is Dr Felicia Chow for InDiscussion.


Meningitis Treatment & Management

Epidemiology of Community-acquired Bacterial Meningitis

Haemophilus Influenzae Type B Vaccine (Rx)

Pneumococcal Vaccine 13-valent (Rx)

Meningococcal A C Y and W-135 Diphtheria Conjugate Vaccine (Rx)

Bacterial Meningitis Imaging

Practice Guidelines for the Management of Bacterial Meningitis

Cranial Imaging Before Lumbar Puncture in Adults With Community-acquired Meningitis: Clinical Utility and Adherence to the Infectious Diseases Society of America Guidelines

Comparison of Four Recommendations Guiding the Use of Neuroimaging in the Management of Bacterial Meningitis

Glasgow Coma Scale

Microbial Aspects and Potential Markers for Differentiation Between Bacterial and Viral Meningitis Among Adult Patients


C Reactive Protein

Healthcare-associated Ventriculitis: Current and Emerging Diagnostic and Treatment Strategies

2017 Infectious Diseases Society of America's Clinical Practice Guidelines for Healthcare-associated Ventriculitis and Meningitis

Effect of Antibiotic Pretreatment on Cerebrospinal Fluid Profiles of Children With Bacterial Meningitis

Impact of Antibiotic Therapy in the Microbiological Yield of Healthcare-associated Ventriculitis and Meningitis

A Risk Score for Identifying Patients at a Low Risk of Bacterial Meningitis Amongst Adults With Cerebrospinal Fluid Leucocytosis and a Negative Gram Stain Result: A Derivation and Validation Study

Cerebrospinal Fluid Pleocytosis Level as a Diagnostic Predictor? A Cross-sectional Study

Repeat Lumbar Puncture in Adults With Bacterial Meningitis

Multiplex Polymerase Chain Reaction

Dexamethasone in Adults With Bacterial Meningitis

Adjunctive Dexamethasone Treatment in Adults With Listeria Monocytogenes Meningitis: A Prospective Nationwide Cohort Study

Colistin (Rx)


Follow Medscape on Facebook, Twitter, Instagram, and YouTube


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.