This article is more than 5 years old. Information may no longer be current.
A preterm neonate presents with seizures and meningitis
Click Here to Manage Email Alerts
A 7-day-old, 34-week premature male in the neonatal ICU became irritable with fever and underwent a full sepsis workup. His exam was unremarkable, but the lab results revealed a complete blood count with 23,700 WBCs, a normal urinalysis and spinal fluid that was significantly abnormal with 9,067 white blood cells (80% Segs), 182,000 red blood cells, a protein of 480 mg/dL, a glucose of zero and a negative Gram stain.
While cultures of blood, urine and cerebrospinal fluid (CSF) and herpes simplex virus PCR of CSF are pending, empiric antimicrobial therapy was begun with ampicillin, gentamicin and acyclovir. Due to the severity, imaging was done with an MRI revealing subgaleal, subdural, subarachnoid and intraventricular blood with multiple, scattered deep white matter infarcts (Figures 1, 2, 3).
Due to the severity, imaging was done with an MRI revealing subgaleal, subdural, subarachnoid and intraventricular blood with multiple, scattered deep white matter infarcts (Figures 1 and 2).
Source: Brien JH
The pregnancy was complicated by a teenage mother, prematurity and eclampsia. The mother had good prenatal care, and all prenatal screening was negative. Delivered by cesarean section due to fetal distress, with Apgar scores of 2, 6 and 8 at 1, 5 and 10 minutes, the infant received his hepatitis B vaccine and was well until the current illness.
The examination was fairly normal except for the anterior fontanel being somewhat full and the infant being very sensitive to touch, which triggered much irritability.
All of the cultures and HSV PCR were negative, as well as CSF enterovirus PCR, cytomegalovirus urine screen and Toxoplasma serologies. Empiric antibiotics were continued despite the lack of an organism being identified because of the infant’s overall condition.
Due to technical difficulties, the lumbar puncture was not repeated until the infant began having seizure-like activity 1 week after the onset of the illness. At that time the CSF revealed a WBC count of 2,015 and 0 RBCs, a protein of 590 mg/dL and glucose of zero again. CSF cultures were again sent for routine stains and cultures plus HSV PCR, acid-fast bacterial and fungal stains and cultures, as well as culture on selective media for Ureaplasma and Mycoplasma hominis media.
Due to the severity, imaging was done with an MRI revealing subgaleal, subdural, subarachnoid and intraventricular blood with multiple, scattered deep white matter infarcts (Figure 3).
Case Discussion
Hydrocephalus is not an unusual consequence with intraventricular bleeding, but the other damage to the infant’s brain shown in the MRI figures would not be explained by intracranial hemorrhage alone. Clearly, the infant had meningitis, and there was no way to know the cause with certainty until the Ureaplasma culture came back positive for (D) U. urealyticum. So, don’t feel bad if you missed it. However, using good test-taking skills, there are some hints that should lead one to suspect something a bit unusual.
When I was a fellow in 1982, Jim Bass, my fellowship director, taught us about U. urealyticum and M. hominis and their capability for causing a variety of infectious problems in newborns (especially U. urealyticum, particularly in premature neonates), from pneumonia to sepsis and meningitis to stillbirths. However, this is only the second case I can recall seeing since then. I suspect there have been others that went unrecognized because I either did not think about it or our cultures missed it. As noted in the question section, these organisms require special culture media, but can be recovered within a few days if properly handled.
As pointed out in the Red Book, sometimes it is difficult to tell what is caused by the Ureaplasma and what is part of the infant’s underlying problems related to prematurity. However, in a case like this, with the positive CSF culture and the temporal improvement with appropriate anti-Ureaplasma therapy, it would seem to leave little doubt as to the role of Ureaplasma in the infant’s meningitis. These organisms lack cell walls, therefore, penicillin and cephalosporin will not work. Macrolides work by binding to the 50S subunit of the ribosome, inhibiting protein synthesis, resulting in a bacteriostatic effect. Tetracyclines also inhibit protein synthesis by binding to the 30S subunit. Ureaplasma is a prokaryote, having both subunits. Therefore, either should be effective if able to reach the organism; however, there are no clear guidelines, only advice from recognized experts and case reports.
The treatment given to this infant, azithromycin plus doxycycline, was based on similar reports, and the CSF rapidly began to normalize on subsequent CSF analyses. Since this case occurred, an excellent review was published in The Pediatric Infectious Disease Journal by Clifford and colleagues in Australia. If you are faced with a similar case, I would review this paper before making any therapy decisions. The infant in our case had significant hydrocephalus on repeat MRI (Figure 4), requiring a ventriculoperitoneal shunt, but was doing remarkably well 6 years later.
Regarding the other choices:
A. Partially treated E. coli meningitis would have likely had a positive Gram stain or culture, and should have improved on the empiric antimicrobial therapy given, unless resistant to both antimicrobials used. But you can never be sure until a definitive cause is found.
B. Inflammation due to intraventricular blood is expected, including a lower-than-normal glucose, but a persistent CSF glucose of zero is unheard of without some other disease process.
C. Tuberculous meningitis in newborns can certainly cause similar derangements in the CSF, but is very rare in this country. And, for similar reasons as above, the stains and cultures usually reveal the answer, certainly by the time a second lumbar puncture is done.
I know that a rare case with nothing but radiographic images to show is probably not of much interest to those of you not involved in the care of premature neonates. But we should be aware of this possibility and understand the potentially devastating consequences of a central nervous system infection caused by one of these peculiar organisms in these fragile neonates. Even if you do not see newborns at all, you may be taking care of one of these babies when they seek a primary provider, and the aforementioned paper will give you good information on that infant’s infectious disease background. Also, the parents may come, already well-read on this subject.
The infant in our case had significant hydrocephalus on repeat MRI (Figure 4).
Some years back, CSF from neonatal cases of meningitis was routinely plated on these selective media, but (at least at our facility) this was discontinued in favor of sending the samples to a reference lab due to the extra work involved in having the selective media locally available. However, according to a chapter written by Samir S. Shah, MD, in Long, Pickering and Prober’s Principles and Practice of Pediatric Infectious Diseases, 4th edition, U. urealyticum can be found on the mucosal surfaces of 40% to 80% of asymptomatic women. Dr. Shah goes on to point out that up to 50% of these women transmit the organism to their newborns, particularly if there has been prolonged rupture of membranes, and can be responsible for invasive infectious disease problems, including meningitis, congenital pneumonia and chronic lung disease.
Dr. Shah also writes the chapters on Mycoplasma diseases, and points out the similar risks to the newborn. Clearly, these organisms can be the unrecognized cause for culture-negative meningitis in newborns, and considering the numbers possible, perhaps we should go back to routinely plating CSF from neonates on these selective media, especially if the Gram stain is negative. For excellent reviews of these organisms, I would suggest reading the brief and easy-to-read chapters by Samir Shah referenced above.
- For more information:
- James H. Brien, DO, is with the department of infectious diseases at McLane Children’s Hospital, Baylor Scott & White Health, Texas A&M College of Medicine in Temple, Texas. He also is a member of the Infectious Diseases in Children Editorial Board. Brien can be reached at: jhbrien@aol.com.
Disclosure: Brien reports no relevant financial disclosures.