Mycoplasma hominis and Ureaplasma species brain abscess in a neonate

We report an infant with a mixed Mycoplasma and Ureaplasma brain abscess who was successfully treated with intravenous doxycycline and erythromycin. Therapeutic concentrations of doxycycline were demonstrated in cerebrospinal fluid. This report is evidence of the potential for Mycoplasma and Ureaplasma to produce focal central nervous system infection, as well as meningitis, in neonates and that use of doxycycline can be efficacious in the therapy of such infections. Brain abscesses in neonates are rare and Mycoplasma and Ureaplasma brain abscesses are even less common. There is little information about the effective treatment of Mycoplasma and Ureaplasma brain abscesses, and many of the antibiotic agents used in the older population are considered toxic to infants. We recently cared for an infant with a Mycoplasma and Ureaplasma brain abscess who was successfully treated with intravenous doxycycline, and we demonstrated therapeutic concentrations of the drug in the infant’s cerebrospinal fluid. Case report. A 3-week-old male infant was transferred to Children’s Hospital of Wisconsin for evaluation and treatment of a brain abscess. The child was the product of a full term uncomplicated pregnancy by a 17-year-old mother and weighed 3070 g at birth. The child went home at 2 days of age and did well until the day of admission when he developed fever to 102°F, decreased oral intake and increased irritability. The child was admitted, and after evaluation for source of fever he was treated with intravenous ampicillin and gentamicin. A peripheral white blood cell count was 17 500/mm3 with a differential of 23% polymorphonuclear forms (PMNs), 25% band forms, 36% lymphocytes and 12% monocytes. Hemoglobin was 14.2 g/dl and platelets were 102 000/mm3. A lumbar puncture was performed that showed 19 700/mm3 red blood cells, 32 000/mm3 white blood cells (79% PMNs), an undetectable glucose concentration and protein 1156 mg/dl. Cerebrospinal fluid (CSF), blood and urine cultures were sterile. A CSF Gram-stained smear revealed no organisms. Fever persisted despite antibiotic therapy, and the child manifested increasing somnolence. After the development of focal seizure activity, a computerized tomograph of the head demonstrated a 4- by 5-cm ring-enhancing lesion in the right frontal area of the brain with associated midline shift of cerebral structures consistent with an intraparenchymal brain abscess (Fig. 1).Fig. 1: Computerized axial tomograph with enhancement demonstrating a 4- by 5-cm right frontal brain abscess in a 3-month-old infant.Aspiration of the lesion 50 ml of purulent yellow fluid. Gram stain of the material demonstrated many PMNs, but no organisms were seen. Therapy was changed to intravenously administered vancomycin, cefotaxime and metronidazole, pending culture results. The fluid was cultured to blood agar, chocolate agar and an enriched thioglycollate broth (Becton Dickinson Microbiology Systems, Cockeysville, MD), which were incubated in 5% CO2 at 35°C. A CDC anaerobic blood agar (Becton Dickinson) was also inoculated and incubated anaerobically at 35°C. After 24 h of incubation, the thioglycollate broth was subcultured to a blood agar incubated in 5% CO2 at 35°C, and a CDC anaerobic blood agar was incubated anaerobically at 35°C. After 96 h of incubation, tiny pinpoint colonies were observed. Gram-stain smears demonstrated faint pleomorphic, tiny Gram-variable coccoid organisms. These organisms were subcultured to a variety of media including Lowenstein-Jensen slants, chocolate agar, Mueller-Hinton blood agar and buffered charcoal yeast extract agar. After 72 h, Gram-variable coccoid organisms were again seen on the CDC anaerobic blood agar. They could not be identified by routine means or propagated for routine susceptibility testing and were sent to two reference laboratories for identification. The child defervesced, but fluid reaccumulated in the abscess cavity. Five days after the first aspiration, the lesion was reaspirated, yielding 16 ml of pus. This material was tested with the use of a latex agglutination panel and was negative for Haemophilus influenzae type b, Streptococcus pneumoniae and Neisseria meningitidis antigens. Cultures of the material again grew an unidentifiable Gram-variable coccobacillus. During the next 2 weeks the patient deteriorated clinically, with the development of a new parietal lobe abscess and increasing dilatation of the ventricular collecting system. He also developed a loculated fluid collection in the left middle cranial fossa with radiologic demonstration of an infarct in the area of distribution of the left middle cerebral artery. Eventually a report was received from the microbiology laboratory at the Wisconsin State Laboratory of Hygiene that the organism had been identified as Mycoplasma hominis according to 16S ribosomal RNA sequencing. Based on this report, therapy was altered to intravenous doxycycline (4 mg/kg/day; 19 mg loading dose followed by 9.5 mg every 12 h) and clindamycin (40 mg/kg/day, 45 mg every 6 h). Doxycycline ventricular fluid concentrations were measured at 1.0 μg/ml. MICs were performed by nonstandardized methods with E test streps (AB Biodisk, Solna, Sweden). 1 MICs of doxycycline and clindamycin were determined to be 0.063 μg/ml and 0.031 μg/ml, respectively. Thus the microorganism was considered susceptible to the therapy being given. CSF cultures taken 5 days after the start of doxycycline and clindamycin were again positive for Mycoplasma but became negative by Day 8 of therapy. Once culture medium appropriate for the cultivation of mycoplasms and ureaplasms was used, 2Ureaplasma species also grew from the CSF, and intravenous erythromycin therapy was added. Ureaplasma species could no longer be cultured after 2 days of erythromycin therapy. Erythromycin and doxycycline therapy was continued until all CSF cultures were negative for 6 weeks. A follow-up computerized tomograph of the head showed resolution of the abscesses and decrease of hydrocephalus. The child has been left with moderate developmental delay and has required placement of a ventriculoperitoneal shunt for CSF diversion. Discussion. Brain abscesses can occur in children at any age but are most common between 4 and 6 years of age. They are very rare in neonates. The bacteria most commonly associated with brain abscess in infants include Staphylococcus aureus, streptococci (viridans, S. pneumoniae, microaerophilic), anaerobes (Bacteroides spp., Fusobacterium spp., Prevotella spp., Actinomyces spp. and Clostridium spp.) and Gram-negative aerobic bacilli (enteric bacilli, Proteus spp., Pseudomonas aeruginosa, Citrobacter diversus and Haemophilus spp.). A single organism is cultured from the majority of abscesses (70%), two organisms from 20% and three or more from 10% of cases. 3, 4 Mycoplasma hominis and Ureaplasma spp. are rare causes of brain abscess in neonates, and there are no detailed reports in the medical literature of previous cases. We suspect that our patient developed a mycoplasmal and ureaplasmal CSF infection that evolved into a brain abscess. Meningitis caused by Mycoplasma infection is more common than neonatal brain abscess, although isolation of Mycoplasma hominis in neonates is usually not routinely sought. 4, 5Ureaplasma urealyticum has rarely been reported as a cause of meningitis in infants. 6 However, this organism has been shown to be an important cause of perinatal morbidity and mortality in certain high risk populations, primarily because of the frequency with which it is transmitted in utero and its association with premature birth. 6, 7 The mycoplasmas or mollicutes are small pleomorphic bacteria 0.3–0.8 μm in diameter which lack cell wall and are bounded by a cell membrane with significant sterol content. At least 15 species have been isolated from humans. M. hominis and Ureaplasma spp. frequently colonize the genital tracts of adult women. 6, 8 Both have been isolated from the amniotic fluid as early as 16 weeks of gestation and can be transmitted to the fetus in utero or during delivery. 6–8 Increased prevalence of these organisms is linked to younger women, lower socioeconomic status, sexual activity with multiple partners and oral contraceptive use. 7, 8 The clinical findings in infants with U. urealyticum and M. hominis meningitis vary from symptomless self-limited conditions to more extensive and permanent neurologic damage. 6M. hominis and U. urealyticum may produce CSF pleocytosis with either polymorphonuclear or mononuclear cells predominating, or inflammatory cells may be minimal or absent. In some infants these organisms are eradicated spontaneously from the CSF, but in other cases may persist in the CSF for weeks to even months. 7 Such persistence was evident in our patient who had an active infection for over 6 weeks. Clinical experience reveals that in many patients M. hominis infections resolve either spontaneously or by drainage and debridement of an infected focus without specific antibiotic therapy. However, appropriate antimicrobial therapy is recommended for invasive infection of blood and the central nervous system. 8 Mycoplasmas are generally not susceptible to antibiotics used to treat neonatal infections. They do not possess a cell wall nor do they produce folic acid, so they are not susceptible to antibiotics that inhibit cell wall synthesis or interfere with folic acid synthesis. M. hominis is usually resistant to erythromycin, and tetracyclines are the drug of choice for this organism. 9 Doxycycline use has been rarely reported in neonates. 5 The doxycycline concentrations in the CSF of our patient demonstrate that this agent can reach therapeutic values in the neonatal CSF. Our current report demonstrates the potential for genital mycoplasmas and Ureaplasma species to produce a focal central nervous system infection as well as meningitis. It is also evident that use of doxycycline can be efficacious in treatment of central nervous system infections in neonates infected with Mycoplasma.