Listeria meningitis associated with tonsillar herniation in an immunocompetent school-aged child

Abstract

Listeria monocytogenes can cause severe illnesses such as gastroenteritis, sepsis and neurolisteriosis, especially in infants, the elderly and immunocompromised patients. We report a case of a previously healthy school-aged girl presenting with severe neurological deficits found to have Listeria meningoencephalitis. Her potential exposure to L. monocytogenes was consumption of contaminated cheese. She had some clinical improvement after initiation of tailored Listeria anti-microbial coverage with ampicillin and gentamicin; however, she developed hydrocephalus requiring external ventricular drain placement and tonsillar herniation requiring emergent posterior fossa decompression. The patient made significant improvements after neurosurgical intervention, and along with continued antibiotics and subsequent rehabilitation services, she improved to near full recovery within a year. The case highlights that neurolisteriosis can affect even immunocompetent children, and aggressive neurosurgical interventions should be considered in patients who develop severe complications such as hydrocephalus and tonsillar herniation to improve outcomes.

Background

Listeria monocytogenes is a gram-positive rod intracellular bacterium that causes uncommon but serious infections, including neurolisteriosis, bacteraemia and maternal–neonatal infection especially in infants, the elderly and immunocompromised patients.¹ Its incidence in developed countries is about 6 per 1 000 000. It may also cause a self-limiting febrile gastroenteritis in healthy patients.² A surveillance study by the Center for Disease Control (CDC) and Food Net shows the incidence of listeriosis is about 0.25–0.32 per 100 000 patients in the USA.³ There were 24 Listeria outbreaks during 1998–2008, of which food vehicles were associated in 20 outbreaks, with serotype 4b being the most common.⁴ Recent outbreaks reported by the CDC have been linked to soft cheeses, celery, sprouts, cantaloupe, ice cream and delicatessen ham. The following is a case report of a healthy school-aged child who contracted Listeria meningoencephalitis complicated by hydrocephalus and tonsillar herniation.

Case presentation

A school-aged girl with no significant medical history presented to an outside emergency department with frontal headaches, emesis, subjective fever and chills. She had no recent travel, but her diet was noted for eating a brand of cheese from South America and may have consumed a piece laid out for an unknown period. After initial resuscitation, her symptoms worsened, so she was transferred to a children’s hospital emergency department for further care. With worsening encephalopathy, meningitis was suspected but a lumbar puncture (LP) was unsuccessfully attempted. The patient was empirically treated with vancomycin, ceftriaxone and acyclovir, and subsequently transferred to the paediatric intensive care unit (PICU), where an LP was successfully performed (table 1).

Table 1.

Initial CSF studies (outside hospital)

White cell count	199 cells/mm3
Segmented neutrophils	24%
Lymphocytes	66%
Monocytes	8%
Red cell count	3240 cells/mm3
Glucose	2.5 mg/dL (serum glucose reported normal)
Protein	246 mg/dL
HSV PCR	Negative

CSF, cerebrospinal fluid; HSV, herpes simplex virus.

During the initial period, she had a waxing and waning pattern of encephalopathy with intermittent unintelligible speech and hallucination. On hospital day (HD) 3, an MRI of the brain performed revealed scattered perfusion abnormalities without hydrocephalus. On HD-5, the patient deteriorated clinically requiring emergent endotracheal intubation and mechanical ventilation. A CT of the brain revealed hydrocephalus in the lateral and third ventricles, and the follow-up CT of the brain the subsequent day showed worsening ventriculomegaly (figure 1). The patient was transferred to the PICU at a tertiary care children’s hospital for emergent paediatric neurosurgery evaluation and intervention.

Figure 1.

CT scans of the brain prior to external ventricular drain placement. (A) A normal scan of the patient’s brain upon presentation. Repeat brain CT scan (B) on hospital day 5 reveals new ventriculomegaly of the lateral and third ventricles.

Upon arrival, she was not following commands, intermittently withdrew to pain in all four extremities and had pinpoint sluggish pupils with left eye deviation, requiring full ventilatory support. Her reflexes were initially noted to be 2+ in all extremities. An external ventricular drain (EVD) was placed at bedside, with initial intracranial pressure (ICP) around 6–9 mm Hg. Repeat laboratory studies were obtained (table 2). Her antibiotic regimen of vancomycin (20 mg/kg/dose every 6 hours, initial trough 17.3 µg/mL), ceftriaxone (100 mg/kg/day divided every 12 hours) and acyclovir (15 mg/kg/dose every 8 hours) was continued.

Table 2.

Admission laboratory studies (tertiary care PICU)

Basic metabolic panel
Sodium	164 mEq/L
Potassium	3.4 mEq/L
Chloride	130 mEq/L
Bicarbonate	28 mEq/L
Creatinine	0.35 mg/dL
Blood urea nitrogen	7 mg/dL
Calcium	9.2 mg/dL
Glucose	131 mg/dL
C-reactive protein	221 mg/L
Complete blood count
White cell count	14.9×109/L
Segmented neutrophils	72%
Bands	3%
Lymphocytes	12%
Haemoglobin	103 g/L
Haematocrit	29.8%
Platelets	264×109/L
Repeat CSF studies
White cell count	70 cells/mm3
Segmented neutrophils	88%
Lymphocytes	7%
Monocytes	5%
Red cell count	3150 cells/mm3
Glucose	13 mg/dL
Protein	67 mg/dL

CSF, cerebrospinal fluid; PICU, paediatric intensive care unit.

Within 48 hours of obtaining the first overall cerebrospinal fluid (CSF) sample at the outside facility and despite pretreatment, the CSF culture grew gram-positive rods. In discussion with paediatric infectious disease, the differential expanded to include L. monocytogenes and tuberculosis (TB). High-dose ampicillin (300 mg/kg/day divided every 6 hours) was added for Listeria coverage.

Despite low ICP and minimal CSF drainage, the patient’s clinical condition did not improve after EVD placement. She grimaced of pain, intermittently opened her eyes spontaneously and remained without purposeful movements. A repeat MRI of the brain and full MRI of the spine were performed, which revealed cerebellar tonsillar herniation with 15 mm extension below the foramen magnum, severe compression of the cervicomedullary junction, and resultant ischaemia of the inferior cerebellum, cerebellar tonsils, and in the spinal cord at the level of C1 (figure 2). Given these findings and lack of clinical improvement, she underwent emergent posterior fossa decompression, C1-laminectomy and duraplasty.

Figure 2.

MRI of the brain T1-weighted images of our patient after external ventricular drain (EVD) placement, but before decompression and laminectomy. (A) The axial view shows dilated lateral and third ventricle, though improved after EVD placement. The sagittal view (B) shows cerebellar herniation with severe compression of the cervicomedullary junction and cerebellar ischaemia.

Blood and CSF cultures resulted positive for L. monocytogenes, sensitive to both ampicillin and vancomycin. CSF was negative for enterovirus and herpes simplex virus through PCR, and serum T-SPOT was negative for TB. Vancomycin had been reportedly at subtherapeutic trough levels at the outside hospital. Antibiotics were changed to ampicillin (400 mg/kg/day divided every 4 hours) with gentamicin (7.5 mg/kg/day divided every 8 hours) for synergy to optimise Listeria coverage. Repeat CSF and blood cultures were subsequently negative.

The patient clinically improved postoperatively, able to follow commands with improvement of motor function, right greater than left. She was successfully extubated on HD-11. Several EVD weaning trials were performed, but the CT of the brain on HD-22 revealed progressive hydrocephalus (frontal horn diameter increasing from 3.6 cm to 3.9 cm), so on HD-27, the patient underwent ventriculoperitoneal shunt (VP shunt) placement.

Differential diagnosis

The most common bacterial pathogens of meningitis to consider in this patient’s age group are Streptococcus pneumoniae and Neisseria meningitidis, accounting for over 80% of cases in the USA, followed by Haemophilus influenzae, group B streptococcus and then L. monocytogenes that account for less than 3% of cases.⁵ Listeria became highest on the differential once the gram-stain of the CSF revealed gram-positive rods and the patient’s history revealed cheese exposure.

Outcome and follow-up

With physical medicine and rehabilitation services, her clinical condition improved with increased motor strength and tolerance of a pureed and thin liquid diet; subsequently, she was transferred to inpatient rehabilitation on HD-32. She completed 3 weeks of intravenous ampicillin and gentamicin combined therapy and an additional 3 weeks of intravenous ampicillin monotherapy. After a year of follow-up, the patient did remarkably well with full recovery of cognition and making good grades in school. She was ambulating again with only mild motor strength deficits on the left.

Discussion

The Multicentric Observational NAtional Study on LISteriosis and ListeriA (MONALISA) in France during 2009–2013 among 372 centres with 818 cases showed less than 15% of Listeria infection as maternal–neonatal infection, 50% bacteraemia and about 30% as neurolisteriosis.⁶ A meta-analysis in the USA involving 16 studies between 1999 and 2014 showed that among infants aged ≤90 days with serious bacterial infections, Listeria infection is rare, with bacteraemia seen in 0.03% of blood cultures and meningitis seen in 0.02% of CSF cultures.⁷ Despite being an uncommon disease, listeriosis has significant mortality and morbidity. In the MONALISA Study, among patient with bacteraemia or neurolisteriosis, 31 (5%) did not receive antimicrobial therapy and all died within 3 days.⁶ There was a 3-month mortality of 46% for bacteraemia and 30% for neurolisteriosis, with 44% of these patients having persistent neurological deficit.⁶ Neurolisteriosis can be complicated by hydrocephalus requiring ventriculostomy, brain abscess, brain stem encephalitis or meningoencephalitis.^{8 9}

Our patient’s presentation is rare, developing Listeria meningitis with significant neurological deficits, complicated by hydrocephalus requiring emergent EVD placement and tonsillar herniation with subsequent neurosurgical decompression. Her known predisposing factor was exposure to cheese,¹⁰ and contributory factor was initial delay in optimum antibiotic coverage by about 24 hours. In addition, after this case, it was noted a South American cheese product was recalled by the Food and Drug Administration for possible contamination with L. monocytogenes, which may have been the culprit for our patient’s illness.¹¹ This case is unique given our patient’s age and immunocompetency, as severe listeriosis is seen more often in infants, the elderly and immunocompromised patients.^{1 7}

The MONALISA Study suggests ampicillin with gentamicin for synergy as first-line therapy for invasive listeriosis for at least 21 days.^{6 12} Our patient began to receive this therapy only 5 days after the onset of illness. The patient was initially on vancomycin which has Listeria coverage, but vancomycin treatment failures have been reported, such as a case of a stem cell transplant recipient found to have a vancomycin-resistant species of Listeria, L. grayi.¹³ She was also reported to have subtherapeutic vancomycin trough levels.

Theoretically, the antibiotics used for our patient all had adequate CSF penetration. Nau et al described CSF penetration of different antibiotics based on the area under the drug concentration time curve in CSF to serum ratio (AUC_CSF/AUC_s).¹⁴ Initial broad-spectrum antibiotics used were ceftriaxone, with a ratio of 0.01 in uninflamed meninges and 0.15 in strongly inflamed meninges, and vancomycin, with a ratio of 0.18 and 0.30 in uninflamed and strongly inflamed meninges, respectively.¹⁴ Targeted treatment with ampicillin (AUC_CSF/AUC_s ratios of 0.02 and 0.2, uninflamed and strongly inflamed meninges, respectively) and gentamicin (ratio of 0.2 in uninflamed meninges) also had adequate CSF penetration.¹⁴ Increased vancomycin clearance as demonstrated by subtherapeutic trough levels probably led to decreased CSF penetration.

Studies have documented loss of information and clinical opportunity during hospital transfer.¹⁵ Our patient underwent transfer to multiple healthcare centres including multiple emergency rooms, PICUs and paediatric floors with possibility of missed clinical opportunity for early optimised antibiotics during the transition.

Emergent neurosurgical decompression and laminectomy were crucial in our patient, as further increases in ICP could have potentially resulted in further brain herniation and brain death. Tonsillar herniation, protrusion of the cerebellar tonsils through the foramen magnum, is a rare complication of meningoencephalitis. If left untreated, this can result in neurologically devastating outcomes. Currently, there is no consensus on surgical intervention in these cases. Foramen magnum decompression was performed on two reported paediatric cases of meningoencephalitis associated with tonsillar herniation, one in an early adolescent girl with meningococcal meningitis¹⁶ and one in an early adolescent girl with culture-negative meningitis.¹⁷ Another case describes a female patient in her 20s found to have TB meningitis associated with a Chiari I malformation who underwent posterior fossa decompression in addition to anti-TB treatment.¹⁸ Cerebellar tonsillar displacement was 20 mm, 12 mm and 10 mm, respectively, for each case; while in our patient, the tonsillar displacement was 15 mm.^16–18 All the patients proceeded to have excellent neurological recovery, though the female patient in her 20s had the longest recovery time, taking about 4 years.¹⁸ In another case of tonsillar herniation, a patient in early childhood with Listeria meningitis underwent only VP shunt placement and did not undergo any posterior fossa decompression.¹⁹ This patient had gradual but incomplete recovery, with still significant gross motor impairments on follow-up.¹⁹ Our patient in this vignette had near-complete recovery with the therapies and interventions she underwent, which include posterior fossa decompression. Given the potential for improved outcomes, decompressive surgery should be considered in patients with meningitis associated with tonsillar herniation.

For this patient, the need for long-term CSF diversion with a VP shunt for an acute infection was weighed against the ongoing infection risk of an EVD. The incidence of EVD-related infections widely ranges from 3.4% to 21.9%.²⁰ Though internalised VP shunts tend to have a lower incidence of infection, 2%–4.2% in the paediatric population, they are placed for a greater period, if not indefinitely.^21–23 After CSF clearance of Listeria, our patient underwent several EVD clamping trials that resulted in progressive hydrocephalus, confirming the need for chronic CSF diversion with a VP shunt. These factors, including infection risks in the short term and long term, should all be considered when evaluating for VP shunt placement.

This case report provides an opportunity to have high clinical suspicion of Listeria infection among patients presenting with severe meningitis and neurological deficit, especially those with exposure to known food vectors. Early initiation of antibiotic therapy, recognition of worsening symptoms with appropriate interventions that include decompressive neurosurgery and a multidisciplinary approach are paramount to improve outcomes.

Patient’s perspective.

Written in the perspective of the patient’s father:

Our daughter woke up complaining of a headache. I assumed she was still tired and told her to get a little more sleep. She continued to sleep most of the morning, and then she started to get nauseous and vomit in the afternoon; this continued into the evening. My wife suggested we take her to the local emergency room because she was vomiting any food or drink she consumed. The doctors at the local hospital said that it sounded like she was dehydrated with a high fever of 102°F (38.9°C). After giving her a saline bag, they told us that she should be up moving around if she was dehydrated, and they wanted to transfer her to a hospital with a pediatric emergency care unit because they did not have that level of care, which I completely understood. Unfortunately, her fever went up to 103°F (39.4°C) at the second hospital, where they admitted her right away.

No one seemed to be checking on her except for a nurse that looked around and then would leave. I noticed our daughter would look at me and then drift off like she was falling asleep several times. I made the nurse look at her and told her this did not match her normal behavior, and finally, a doctor came to see her. At this point, many doctors and nurses were whispering the words meningitis, spinal tap, and the name of another hospital. My wife and I were finally let in on what was going on. They told me that they were transferring her to another hospital with a pediatric intensive care unit, but they needed to do a spinal tap to start the process of finding out if she had meningitis. I carried her now fragile body to a room and laid her on a metal table where my daughter looked not at me but through me, and they could not collect a sample from the spinal tap. Shortly after this, she was transferred to a third hospital by air, where a successful spinal tap was done, and a confirmation that she had meningitis was already found out by the time my wife and I got there. Unfortunately, we were told after a few days that she was out of the woods, in the sense that she was going to live, but they did not know the damage that could have happened to her brain due to her having multiple strokes. The following day when I lifted her to put her back on her pillow, her head fell to the side, and she stopped breathing. They had to intubate her because the pressure in her head had caused her to stop breathing. At this point, we were told she needed to be sent to another hospital to relieve the pressure. Once at the fourth hospital in less than a week, I could see the magnitude of the situation. A team of doctors and nurses took her to a room to drill a hole in the top of our daughter’s head. Another group analyzed an image of her brain, while a nurse tried to explain things to me. I was completely zoned out from lack of sleep and all that was happening. The following day, I was approached about a major surgery to relieve pressure from her brain stem and to take a sample of the brain tissue to find out exactly what kind of bacteria was causing all these problems. This was the change into a positive trajectory after a week of what felt like nothing but downward spirals.

After the surgery, my wife and I saw our daughter open her eyes and look at her mother with tears and a facial expression that we had not seen in a week of being in hospital. A couple of days later, they were taking her off the ventilator. Every day after that, she slowly improved with the guidance and action plan of the incredible teams she had from general physicians, physical therapists, occupational therapists, neurosurgeons, and nurses. Finally, after three months of rehabilitation to learn to walk, talk, and gain the general use of her body again, she held my hand as she walked out of the hospital to come home after three and a half months. It has been almost five years since all of this occurred. She is now older, getting good marks in school, playing volleyball, and just had her first dance recital. I have to hold back tears every time she accomplishes something because I know how close it was to none of these things ever being a reality, and I am forever grateful for those who had a hand in giving my beautiful daughter a future.

Learning points.

• Neurolisteriosis can be considered as an infectious aetiology for meningitis even in healthy immunocompetent children, especially with exposure to certain food vectors such as soft cheeses.

• Optimal antibiotic therapy towards suspected aetiologies is paramount in any case of bacterial meningitis, and care should be taken to reduce antibiotic delays when patients are transferred to other facilities and higher levels of care.

• Neurosurgical decompression can be considered as a therapeutic intervention for patients with bacterial meningitis and signs of tonsillar herniation to improve outcomes.

Ethics statements

Patient consent for publication

Parental/guardian consent obtained.