Bacterial meningitis during the neonatal period is still one of the most devastating conditions, with a morbidity rate of 20% to 60%.[1] The nationwide mortality can be as high as 40% in treated cases in the first month of life, and up to 10% beyond the neonatal period. Multiple factors contribute to the susceptibility of infants to this illness. The immune immaturity of infants is the biggest contributor, especially pre-term infants. Because infants do not receive their first set of immunizations until 2 months of age, the risk is high for bacteremia, possibly resulting in bacterial meningitis.
Those populations at highest risk are preterm infants, males, the indigent population, and infants in daycare. Also, children of mothers with a history of a sexually transmitted infection, including genital herpes, and mothers who test positive for group B streptococcus are at high risk. Mothers who have eaten certain types of foods may be at risk for passing Listeria infection to their newborns, another pathogen found in the neonatal population. Gram-negative rods, most commonly Escherichia coli, contribute to significant mortality. Group B streptococcus continues to be the most common pathogen causing meningitis in the neonatal period.
Neonates are especially prone to this disease due to their immune immaturity. Different pathogens are responsible depending on the age of the child, gestational age and location. The distribution of organisms seen in neonatal meningitis is similar to neonatal sepsis.[2] The disease is classified as either early or late onset. Early onset occurs within the first 72 hours of life. Late-onset is predominantly seen in premature infants, and a different array of pathogens is found in this population.
The incidence of early-onset meningitis has been greatly reduced by the initiation of intrapartum antibiotics to combat Group B streptococcus (GBS) infection. However, GBS remains the most common cause of both meningitis and neonatal sepsis, causing greater than 40% of all early-onset infections.[2] The next common pathogen in this group is E. coli and has emerged as the most common cause of early-onset sepsis and meningitis among very low birth weight (VLBW, less than 1500 g) infants.[2]
In the late onset group, the incidence is directly related to gestational age and birth weight. The most common offenders here are coagulase-negative staphylococci and Staphylococcus aureus, followed by E. coli and Klebsiella.
Another bacteria found in early-onset meningitis is Listeria, and antibiotic coverage should consider this as well. Late-onset illness should cover additional organisms in the nosocomial environment, especially those in neonatal intensive care units, including Pseudomonas aeruginosa and methicillin-resistant Staphylococcus aureus.[3]
Viral illnesses should be considered, and include herpes simplex virus (HSV) infection and enterovirus. With a detailed maternal history indicating her infection with HSV, antiviral coverage is strongly suggested.
Neonates are especially prone to meningitis and sepsis due to their cellular and humoral immune immaturity. They are at high risk for bacterial infections, with 10% to 20% of febrile infants younger than 3 months having a serious bacterial infection. Bacteremia is twice as likely to occur in the first month of life.
In developed countries, culture-proven neonatal meningitis is estimated at 0.3 per 1000 live births, but this is likely underestimated. For infants in the neonatal intensive care unit (NICU), of those evaluated for sepsis, only 30% to 50% have a lumbar puncture done, and 75% of the time it occurs after the initiation of broad-spectrum antibiotics. As such, the culture results may be affected by this.[2] The mortality rate is about 10% to 15%, and the morbidity remains high. Up to 50% of infants who survive the illness develop chronic neurological sequelae, including seizures, cognitive deficiencies, motor problems, as well as hearing and visual impairment.[3]
In developing countries, the incidence is higher, at 0.8 to 6.1 per 1000 live births, with a mortality rate of up to 58%.[2] Reporting in some of these countries is suspect, and the incidence is likely higher.
Multiple sources report that in the last 40 years, the mortality of this disease has dropped tremendously. However, despite the multitude of advances in neonatology and medicine, morbidity has not changed.
The most common mechanism this illness develops is via primary bloodstream infection with seeding into the central nervous system. Early and late onset infection can manifest in different ways, with early-onset infection being primarily maternal in origin. The birth process exposes the infant to a multitude of pathogens. It can be transmitted through the vagina to ruptured amniotic membranes, or due to contact of the neonate’s skin during passage through the birth canal. Organisms such as Listeria monocytogenes can be transmitted through the placenta.[2] Late-onset infection is largely nosocomial in nature. Poor hand hygiene between infected and uninfected infants is one cause.[2] Foreign devices such as endotracheal tubes, ventilators, catheters, and feeding tubes can also transmit infection.
Introduction of the pathogen through the blood-brain barrier (BBB) is much easier in the neonate. From late gestation until the postnatal period, the barrier is still in development, and this leads to increased susceptibility to infection.[4] Because the infant's brain is still developing after birth, the cerebral vessels are more fragile, leading to more devastating effects from the illness.
The immune system of the neonate is immature, including the phagocytic response. While the phagocytic response is impaired only transiently, thought to reach adult levels by day 3 of life, the neonatal immune system is deficient in complement and antimicrobial proteins and peptides.[4] After infection, a multitude of inflammatory responses begins, including the production of TNF-alpha, which is an inflammatory marker, and IL1B and IL6 cytokines. In a study of 54 infants, higher levels of these inflammatory markers were found in the cerebrospinal fluid (CSF) of those with bacterial meningitis.[4]
Meningitis in neonates is difficult to diagnose as symptoms can be vague. Fever is the obvious red flag; however, it can sometimes present as a multitude of complaints. Irritability, or “fussiness,” and poor feeding should raise suspicions. The clinician should take a good maternal and pregnancy history as well. Questions should include those of the mother’s sexual history, whether she received antibiotics during delivery, and how long was it from rupture of membranes until delivery. Were there complications during delivery, were forceps or vacuums used, or other interventions that could introduce infection? All of these are risk factors and will help guide workup and treatment.
Neonates may look well, but can also be toxic appearing. Tachypnea, a petechial rash, poor neonatal reflexes, and a floppy infant should prompt the clinician to suspect meningitis or any other cause of sepsis. Other findings on the physical exam include a bulging fontanelle, although this is a late finding.
Any infant 28 days old or younger who presents with a fever (100.4 F) should undergo a septic workup. This includes a complete blood count (CBC) with differential, blood culture, catheterized urine with culture, chest radiograph and lumbar puncture. Orders for the lumbar puncture should include cell count, glucose, protein, gram stain, culture and, if suspected, HSV polymerase chain reaction (PCR) study.
The lumbar puncture with cell count, protein, gram stain, and culture is essential in pinpointing this diagnosis. The culture of the CSF continues to be the gold standard. Typical white blood cell (WBC) counts in CSF for bacterial meningitis range from 200 to 100,000 per mL, and 25 to 1000 per mL for viral meningitis.[5] In the differential, there may be 80% to 100% neutrophils in bacterial illness, and less than 50% in viral illness.[5] Some sources indicate that the cell count in CSF can be unreliable. Typically any WBC count over 20 per mL should raise concern; however, some studies show that meningitis can be present despite a normal WBC count. A study evaluating 9111 infants showed that using the 20 per mL cutoff missed 13% of meningitis cases.[2] This indicates that culture continues to be the gold standard for diagnosis.
PCR may be a more sensitive and real-time tool to diagnose meningitis in the future. A real-time PCR assay to detect multiple pathogens, including Streptococcus pneumonia, E. coli, GBS, S. aureus and L. monocytogenes, had an overall higher detection rate compared to culture (72% vs. 48%). Even if antibiotics had been started, PCR detected pathogens that cultures did not (58% vs. 29%).[2] More studies are needed before PCR is widely used.
Another test to detect SBI in infants includes C-reactive protein (CRP) and procalcitonin. Studies involving CRP in diagnosis have been promising, but its use is limited because it takes 8 to 10 hours to synthesize, so its sensitivity varies. Procalcitonin shows promise, as it increases within 2 hours of infection. It has a high sensitivity (92.6%) and specificity (97.5%) if drawn after the first hours of life.
Because of the high morbidity and mortality of meningitis in neonates, treatment is aggressive. Infants should be hospitalized and cultures followed until negative for 72 hours. Broad-spectrum antibiotics should be started as soon as possible. Toxic patients may require care in a pediatric intensive care setting.
Antibiotic choices for neonatal meningitis include ampicillin and gentamicin or cefotaxime. For infants younger than 8 days old, the dose for Ampicillin is 150 mg/kg per day divided every 8 hours, plus gentamicin 4 mg/kg daily or cefotaxime 100 to 150 mg/kg per day divided every 8 to 12 hours. From 8 to 28 days old, the antibiotics are the same, but the dosing if slightly different. The ampicillin dose is 200 mg/kg/day divided q6 hours, plus the same dose for gentamycin or cefotaxime 150 to 200 mg/kg per day divided every 6 to 8 hours.
If concern for HSV is high, starting acyclovir is highly recommended. The dose is 60 mg/kg per day divided every 8 hours, or 20 mg/kg per dose. Symptoms that trigger this include seizures, skin lesions, and abnormal liver function tests.
There are many reasons infants can be febrile. The number one consideration is infections. Most commonly, a viral infection is the culprit. However, since the morbidity and mortality of both neonatal meningitis and sepsis are high, other causes of fever need to be ruled out. Other causes of neurological symptoms in infants include primary brain tumor, head injury, intraventricular hemorrhage in the premature infant population, a toxin, hyponatremia due to watered-down formula, genetic diseases, primary metabolic disorders, among others. Non-accidental trauma, including shaken baby syndrome, should be considered as well.
Despite the decrease in mortality, neonatal meningitis continues to have high morbidity. Those include profound neurological deficits such as learning disabilities, seizures, behavioral abnormalities, visual and hearing deficits, and profound mental retardation.
One study in Tunisia showed that 21.9% of the children reviewed in their retrospective study had neurological sequelae. Respiratory distress, low birth weight, shock and a pleocytosis of fewer than 500 cells/mL were indicators of a worse prognosis. The addition of Ofloxacin to the antibiotic regimen was associated with decreased neurological sequelae in survivors.[6] High CSF protein, both during and after the acute illness, has also been linked to poorer outcomes.[7]
The biggest pitfall of meningitis in infants is not considering it in the first place. Well-appearing febrile infants can become toxic quickly, and are at high risk for SBI due to their immature immune systems.
The lumbar puncture with a culture of the CSF continues to be the gold standard of diagnosis. PCR may be a test that can be used in the future. Broad-spectrum antibiotics are the standard of care and should include ampicillin plus gentamicin or cefotaxime. Cefotaxime is currently preferred. If suspected, consider adding acyclovir to the regimen. In children with neurologic symptoms consider other etiologies; however, meningitis must be ruled out, due to its high morbidity and mortality.
The diagnosis of neonatal meningitis can be a devastating one. However, today mortality is lower, due to aggressive treatment with antibiotics, antiviral and advanced neonatal medicine. An interprofessional team approach including physicians, nurses, pharmacists, and caseworkers can help not only treat the patient but the parents as well. Neurologic sequelae that can result will require, in many cases, lifelong care. Over his or her lifetime, they will need physical therapy, cognitive therapy, medication, and social support. This will center on the child’s primary pediatrician, who can coordinate therapies and specialist consultations, including neurology, and if the sequelae are severe enough, home health care.
There are clinical decision rules to help determine which children should be admitted and who can be safely discharged. The Bacterial Meningitis Score, which can identify very low risk (VLR) patients, has been tested and reaffirmed in several studies and shows it can help decrease costs and increase patient safety by decreasing unnecessary, and possibly harmful, treatments and testing. However, it cannot be used in the less than 60-day-old age group, as there has been uncertainty regarding its validity in this population. (Level 1)
In a more successful trial in reducing unnecessary testing and improving outcomes in infants, a team of pediatricians at the University of Utah devised an evidence-based care process model (EB-CPM). Applying this to not only well-appearing infants at the tertiary care center but also the other medical facilities in the region, they were able to decrease costs and lengths of stay, and improve outcomes over 2 years across all facilities. This involved determining evidence-based approaches to laboratory testing, imaging, and treatment. They also used web-accessible tools including algorithms, orders, references, and antibiotic recommendations. (Level 1)
[1] | Berardi A,Lugli L,Rossi C,China MC,Vellani G,Contiero R,Calanca F,Camerlo F,Casula F,Di Carlo C,Rossi MR,Chiarabini R,Ferrari M,Minniti S,Venturelli C,Silvestrini D,Dodi I,Zucchini A,Ferrari F, Neonatal bacterial meningitis. Minerva pediatrica. 2010 Jun [PubMed PMID: 21089719] |
[2] | Ku LC,Boggess KA,Cohen-Wolkowiez M, Bacterial meningitis in infants. Clinics in perinatology. 2015 Mar [PubMed PMID: 25677995] |
[3] | Gordon SM,Srinivasan L,Harris MC, Neonatal Meningitis: Overcoming Challenges in Diagnosis, Prognosis, and Treatment with Omics. Frontiers in pediatrics. 2017 [PubMed PMID: 28670576] |
[4] | Barichello T,Fagundes GD,Generoso JS,Elias SG,Simões LR,Teixeira AL, Pathophysiology of neonatal acute bacterial meningitis. Journal of medical microbiology. 2013 Dec [PubMed PMID: 23946474] |
[5] | Norris CM,Danis PG,Gardner TD, Aseptic meningitis in the newborn and young infant. American family physician. 1999 May 15 [PubMed PMID: 10348069] |
[6] | Ben Hamouda H,Ben Haj Khalifa A,Hamza MA,Ayadi A,Soua H,Khedher M,Sfar MT, [Clinical outcome and prognosis of neonatal bacterial meningitis]. Archives de pediatrie : organe officiel de la Societe francaise de pediatrie. 2013 Sep [PubMed PMID: 23829970] |
[7] | Tan J,Kan J,Qiu G,Zhao D,Ren F,Luo Z,Zhang Y, Clinical Prognosis in Neonatal Bacterial Meningitis: The Role of Cerebrospinal Fluid Protein. PloS one. 2015 [PubMed PMID: 26509880] |