CASE
On his 23rd day of life (DOL), a baby boy was transferred to our medical center from a community emergency department for treatment and management of presumed neonatal sepsis and meningitis.
He was born full term via normal spontaneous vaginal delivery. His mother was primigravida with no medical complications during her pregnancy. Her prenatal lab results were negative for group B streptococci (GBS) at 36 weeks of gestation, and testing for HIV, syphilis, gonorrhea, chlamydia, and hepatitis B and C was negative. She reported no history of herpes simplex virus (HSV) infection, and no genital lesions were noted on examination. She did not recall having any fever or influenza-like symptoms at any point during her pregnancy. The baby emerged vigorous, was discharged after 2 days in the hospital, was exclusively breastfed, and regained his birth weight by DOL 14.
One day prior to his admission on DOL 22, the patient developed a tactile fever, decreased feeding, and fussiness. He was taken to his outpatient pediatrician, who noted an afebrile, well-appearing baby. The baby was given acetaminophen and discharged from the clinic. Over the subsequent night, the patient became fussier and again had tactile fevers. His mother then took him to an emergency department, where he was found to be irritable, tachycardic, and febrile to 39.2°C. No other abnormal examination findings were identified.
Blood and cerebrospinal fluid (CSF; obtained by lumbar puncture [LP]) were collected prior to administration of antimicrobials and submitted for culture and laboratory studies (Table 1). His initial CSF was significant for a pleocytosis, as well as abnormally elevated protein (143 mg/dl) and low glucose (16 mg/dl), highly suggestive of bacterial meningitis. The patient was empirically started on intravenous ampicillin (300 mg/kg/day, 4 divided doses), gentamicin (7.5 mg/kg/day, 3 divided doses), and acyclovir (60 mg/kg/day, 3 divided doses) and transferred to our medical center. Upon arrival, he was fussy but afebrile. As the Gram stain was negative for a causative organism, his gentamicin was exchanged for cefotaxime (300 mg/kg/day, 4 divided doses) for more complete empirical Gram-negative coverage and he was continued on ampicillin for empirical treatment of GBS and Listeria, as well as on acyclovir for HSV. Magnetic resonance imaging with contrast of his brain was performed on DOL 24, and mild enhancement of the leptomeninges was noted without any other findings such as abscess or infarction.
Laboratory values on days 1 and 3 after hospital admission
Blood cultures and CSF cultures remained negative at the community medical center, and an HSV CSF PCR assay was also negative. Repeat blood cultures and CSF were collected on DOL 25. One aerobic blood culture bottle (BD Bactec Peds Plus/F) was collected and incubated in the Bactec automated blood culture system (Becton Dickinson, Franklin Lakes, NJ, USA), and no growth was detected after 5 days. Gram staining of the CSF revealed many white blood cells and no organisms. The specimen was inoculated onto 5% sheep blood agar (blood agar plate [BAP]), chocolate agar, and MacConkey agar and incubated at 35°C with 5% CO2 for 72 h. In addition, 0.5 ml of CSF was also inoculated into thioglycolate broth and incubated for 5 days. Despite consistent abnormal CSF chemistry, all of the cultures remained negative (Table 1) and the patient was continued on empirical therapy for presumed bacterial meningitis.
Communication between the infectious disease team and microbiology laboratory prompted testing of CSF with the FilmArray Meningitis/Encephalitis (ME) panel (BioFire Diagnostics, Salt Lake City, UT, USA), and Listeria monocytogenes was detected. On the basis of this finding, acyclovir was discontinued and the patient was treated with a 21-day course of ampicillin and gentamicin. He fully recovered because of his excellent clinical status, and no follow-up studies were performed.
DISCUSSION
Bacterial meningitis is the infection of the protective membranous linings around the brain by bacterial pathogens and can occur at any age. Neonates are a particularly susceptible population. Bacterial meningitis in the neonate may occur in as many as 1 in 2,000 live births in developed nations and in a significantly higher percentage of neonates in underdeveloped nations (1, 2). Of the serious bacterial illnesses of childhood, meningitis carries the highest morbidity and mortality rates. The ongoing inflammation can cause irreversible injury to the underlying parenchyma, thrombosis of vessels, and subsequent ischemia/infarction, hearing and vision loss, and intellectual and physical disability, and if left untreated, bacterial meningitis is usually fatal. While older children and adults typically present with fever, neck stiffness, headache, and confusion, initial symptoms in neonates are frequently nonspecific, including fever, irritability, excessive crying or sleepiness, and difficulty with feeding. Only occasionally do infants develop posturing, seizures, or an unusually firm, tender, or bulging fontanelle that would direct a clinician to suspect a CNS infection (1, 2).
Although clinical symptoms may not be readily distinguishable from other serious bacterial infections in the neonate, CSF obtained from an LP can quickly and reliably identify concerns for bacterial meningitis (3). Abnormally low glucose levels in the CSF, termed hypoglycorrhacia, is the result of severe inflammation of the meninges causing dysfunction of glucose transport across the blood-brain barrier and an increase in metabolic activity by both the infecting bacteria and infiltrating WBCs. Significant recruitment of activated macrophages and neutrophils, as well as capillary leak and acute inflammatory mediators, causes an elevation in CSF protein levels. This constellation of CSF pleocytosis, hypoglycorrhacia, and elevated CSF protein is the classic pattern of findings in bacterial meningitis, as was seen in our patient. This pattern is rarely encountered with viral meningitis (3). Gram staining and culture of CSF obtained prior to the administration of antibiotics are the standards used to define bacterial meningitis and are sufficient to identify the causative organism in up to 85% of suspected cases (3). The sensitivity of CSF Gram staining alone is highly variable, ranging from 20 to 93%, depending on the organism (4). Acridine orange is a potential adjunct when a Gram stain is negative or as a first-line screening test, as sensitivity has been reported to be 21% higher than that of Gram staining (5). The yield of CSF cultures can significantly decrease up to 41% in patients treated with antibiotics prior to LP, whereas the yield of CSF Gram staining pretreatment decreases only slightly (4). Despite laboratory evidence highly suggestive of bacterial meningitis and CSF obtained prior to antibiotic administration, cultures remained negative in our patient's case, creating a diagnostic dilemma.
A broad range of bacteria is able to cause meningitis, and the most frequent organisms encountered vary according to patient age and comorbidities. The most common organisms that cause neonatal bacterial meningitis are GBS and Escherichia coli, which together account for >70% of the cases in this age group (1, 2). The remaining bacterial etiologic agents include other enteric Gram-negative organisms, as well as enterococci. L. monocytogenes is a relatively rare yet clinically significant cause of meningitis, affecting 0.5 to 3 per 100,000 live births (1, 2).
There are several species in the family Listeriaceae, which are environmental organisms typically found in soil and water. L. monocytogenes is the only major human pathogen in the family and can colonize mammals, including humans. It is resistant to cold temperature, high salinity, and alkalinity, allowing the organism to replicate both in the environment and especially in unpasteurized or contaminated foods even when they are refrigerated (6). Indeed, cold enrichment is one method of isolating Listeria species from other bacteria. L. monocytogenes is most commonly encountered as a food-borne pathogen, accounting for ∼1,600 cases and 260 deaths in the United States annually. Most healthy individuals have either asymptomatic infections or very mild symptoms, such as abdominal cramping and self-limited diarrhea, when exposed to the bacterium. Those most at risk for severe disease are pregnant women, infants <2 months of age, the elderly, and those with immunodeficiencies, particularly with impairment of T-cell function such as in HIV/AIDS. Pregnant women have a decline in T-cell function that allows them to accommodate a fetus, which allows the bacteria to proliferate in the placenta and may be the reason for the increased risk of invasive disease in this population (6).
In the laboratory setting, L. monocytogenes is a Gram positive, non-spore-forming rod or coccobacillus (Fig. 1) that displays tumbling motility in broth incubated at 18 to 25°C. This activity is reduced at lower or higher temperatures. In addition to the Gram stain findings and demonstration of tumbling motility, definitive identification of L. monocytogenes can be confirmed by positive catalase and hippurate tests, as well as the presence of a narrow zone of beta-hemolysis on BAP (Fig. 1) (7). Recovery of Listeria spp. from clinical specimens and food samples can be challenging and frequently requires a pre-enrichment step and selective media. Listeria spp. can grow over a wide temperature range (1 to 45°C), but optimal growth is at 37°C. Prolonged incubation of agar medium at 4°C can potentially increase the yield but can take up to several weeks and will not allow the isolation of injured bacterial cells, as the organism will not grow under environmental stress (8). Several alternative methods are also able to identify L. monocytogenes, including commercial identification systems, 16S rRNA sequencing, or matrix-assisted laser desorption ionization–time of flight mass spectrometry. Importantly, L. monocytogenes is intrinsically resistant to broad-spectrum cephalosporins, a beta-lactam drug class commonly employed to empirically treat sepsis and meningitis of unknown etiology.
Macroscopic and microscopic morphology of L. monocytogenes. (a) Demonstration of a narrow zone of beta-hemolysis surrounding colonies after 24 h of incubation at 37°C. (b) Gram stain revealing characteristic small, Gram-positive rods or coccobacilli (×100 magnification).
Identification of Listeria spp. is important to allow for appropriate de-escalation or optimization of antimicrobial therapy. However, recovery from clinical specimens can be challenging. In one study, only 83% of CSF cultures and 64% of blood cultures of samples from patients with probable L. monocytogenes meningitis grew the organism (9), emphasizing the need for a more sensitive diagnostic approach. A false-negative result by culture may be detrimental, as it may prompt clinicians to incorrectly rule out bacterial meningitis and discontinue antimicrobials or select suboptimal empirical antimicrobials. Furthermore, as a false-negative result cannot be distinguished from a true negative result, empirical therapy may be continued unnecessarily. The availability of molecular panels provides a means to augment current practices. Testing is significantly more rapid, shortening the time to a diagnosis compared to that of standard culture, and can identify a variety of pathogens that may potentially be missed, depending on the clinician's standard testing practices. The FilmArray ME panel is an FDA-approved multiplex PCR assay for the simultaneous detection of 14 common bacterial, viral, and yeast pathogens implicated in infections of the central nervous system in approximately 1 h. Multicenter evaluation of the panel with 1,560 clinical specimens revealed a range of 85.7 to 100% sensitivity and specificity, depending on the target (10). In another study of 342 adults and pediatric patients, the FilmArray ME panel had overall positive and negative agreement levels of 92.9 and 91.9%, respectively. There were significant discrepancies between the panel and cryptococcal antigen results, with correlation in only 57% (6/14) of the cryptococcal antigen-positive specimens. Of note, no culture results were available and confirmatory testing by sequencing was also negative (11). Because of the rarity of the infection, L. monocytogenes was not detected in either study; importantly, there were also no false-positive cases identified (10, 11).
Syndromic panels have the potential to be useful tools but are not appropriate for every patient and must be interpreted in the context of the clinical picture, patient risk factors, and other laboratory and clinical data. Specifically, FilmArray ME panel testing must not replace standard culture of CSF, as the targets included in the panel are not representative of all of the pathogens implicated in CNS infections (e.g., other members of the family Enterobacteriaceae, Pseudomonas species, and enterococci), and susceptibility testing of culture-positive organisms is imperative to further optimize antimicrobial therapy. Another limitation of the panel is the potential risk of contamination with normal oral flora, such as Streptococcus pneumoniae, that may result in false-positive results (10). Thus, strict adherence to appropriate molecular practices and correlation of results with clinical findings are imperative.
In conclusion, L. monocytogenes is a significant cause of meningitis and a definitive diagnosis is imperative for appropriate optimization of antimicrobial therapy. Laboratory diagnosis of listeriosis by conventional culture approaches can be limited. The advent of novel molecular assays may increase the detection of L. monocytogenes, as well as other pathogens commonly associated with bacterial meningitis, and directly impact patient management.
SELF-ASSESSMENT QUESTIONS
What are the two most common organisms that cause neonatal bacterial meningitis?
(a) Listeria monocytogenes and GBS.
(b) Haemophilus influenzae type B and Streptococcus pneumoniae.
(c) GBS and Escherichia coli.
(d) Neisseria meningitidis and Streptococcus pneumoniae.
According to CLSI guidelines M35-A2, which of the following tests is required for the definitive identification of Listeria monocytogenes in the laboratory?
(a) Oxidase.
(b) CAMP test.
(c) Hippurate.
(d) Motility on semisolid medium.
To which of the following antibiotics is Listeria monocytogenes intrinsically resistant?
(a) Ampicillin.
(b) Gentamicin.
(c) Cefotaxime.
(d) Trimethoprim-sulfamethoxazole.
(For answers to the self-assessment questions and take-home points, see page 3075 in this issue [doi:10.1128/JCM.01160-16].)
- Copyright © 2016, American Society for Microbiology. All Rights Reserved.
