Background

Mycoplasma Hominis is a frequent habitant of the human genital tract [1]. Neonates are susceptible to acquire this microbe either in utero or through the colonized birth canal during the process of parturition [2]. However, invasive infections with this organism in preterm and term infants are noted to be rare [1, 3]. We describe an unusual case of neonatal meningitis due to Mycoplasma Hominis, confirmed in bacterial cultures and 16s ribosomal RNA Polymerase chain reaction (PCR) analysis. This case report demonstrates the importance of suspecting Mycoplasma Hominis as a cause of central nervous system (CNS) infection in neonates who present with clinical signs but no growth on conventional bacterial cultures and no improvement on empirical antibiotic treatment.

Case presentation

A male neonate was born at 25 + 6 weeks of gestation to 35 years old gravida 2, parity1 (G2P1) mother with antiphospholipid antibody syndrome. This was an in vitro fertilization (IVF) pregnancy. She was Rubella immune and negative for Hepatitis B surface antigen, Venereal disease research laboratory test (VDRL), Human immunodeficiency virus, Gonococci and Chlamydia. Her urine culture and vaginal swab for Group B Streptococci was also negative. Her anatomy scan at 18 weeks of gestation showed normal fetal anatomy and a short cervix for which she underwent cervical cerclage placement at 20 + 5 weeks gestation. There was no history of chorioamnionitis or prolonged rupture of membranes. She presented at 25 + 5 weeks with preterm labor and had a spontaneous vaginal delivery the next day. She had received two doses of Betamethasone and Magnesium sulphate prior to delivery. He initially required positive pressure ventilation followed by intubation and surfactant administration with an acceptable response. Umbilical arterial and venous catheters were placed and the neonate was transferred to neonatal intensive care unit (NICU) in a stable condition.

A blood culture was drawn at birth and empiric treatment with Ampicillin and Gentamicin was intiated. The antibiotics were discontinued after 36 hours as the blood culture showed no growth and he remained clinically stable. In addition, as per unit protocol, an endotracheal aspirate was sent for a Ureaplasma culture on admission. This grew a Mycoplasma species for which 3 days of intravenous Azithromycin was given.

A head ultrasound on day 4 of life showed bilateral intraventricular hemorrhage (grade 3) with mild ventriculomegaly and a hemorrhagic venous infarct in the right frontoparietal region. Umbilical arterial catheter was removed on day 4 of life whereas umbilical venous catheter was removed on day 6 of life. He remained stable till day 7 of life when he presented with an acute clinical deterioration presenting as lethargy and tachycardia; a septic work up including blood and cerebrospinal fluid (CSF) cultures was completed and Vancomycin and Cefotaxime were started as empiric antibiotics. His CSF analysis showed red blood cells (RBC) count of 4887 × 10E6/l, pleocytosis of white blood cells (WBC) 9620 × 10E6/l with neutrophilic predominance, hyperproteorachia - protein 8300 mg/l and hypoglycorachia - glucose 0.1 mmoles/liter. Gram staining in CSF was negative. Within hours he developed generalized clinical seizures which was treated with Lorazepam and Phenobarbital followed by a Levetiracetam load and maintenance dose. He had further episodes of clinical and subclinical seizures which resolved after further adjustment of levetiracetam doses.

His first CSF culture was suspected to grow Mycoplasma Hominis and was sent to the regional national diagnostic laboratory for confirmation. The CSF culture was repeated along with a sample for 16s ribosomal RNA PCR analyses. The 16s ribosomal RNA PCR analysis confirmed Mycoplasma hominis in both CSF samples. Antibiotics were then changed to Moxifloxacin and Doxycycline. The repeat CSF culture after 48 hours of Moxifloxacin and Doxycycline showed no growth. CSF analysis done seven days after starting this treatment showed RBC 5650 × 10E6/l, WBC 449 × 10E6/l, protein 3763 mg/l and glucose 0.2 mmoles/liter. Subsequent CSF findings repeated after another three days showed further improvement, RBC 7187 × 10E6/l, WBC 93 × 10E6/l, protein 3691 mg/l and glucose 0.5 mmoles/liter. No growth was seen on any of these CSF cultures. Serial head ultrasound scans were done which showed progressive ventriculomegaly, post hemorrhagic ventricular dilatation and cystic evolution of hemorrhagic/venous infarct in the right frontoparietal region. Although these findings could be attributed to extreme prematurity; there is a likelihood that infection with Mycoplasma Hominis may have played a role in its progression. There have been documented cases of Mycoplasma Hominis meningitis resulting in CNS complications including intraventricular and periventricular hemorrhage, hydrocephalus, and infarction [3]. Lumbar taps were repeated as therapeutic measure to reduce ventriculomegaly, but optimal volumes of CSF were not obtained. Because of the ventriculomegaly, he was evaluated for a possible shunt on day 31; this was deferred because of the stable ventriculomegaly.

Antibiotics were given for a total duration of six weeks after which he was discharged home with a normal neurological examination. He was referred to Neurodevelopmental clinic for follow-up.

Discussion and conclusion

Neonatal CNS infections with Mycoplasma Hominis, although rare, can cause severe morbidity and mortality in neonates [1]. In a case series of 29 neonates with Mycoplasma Hominis infection (age of presentation day1-32 of life), Hata and colleagues reported complications such as brain abscess, hydrocephalus, infarction, cerebritis and periventricular/intraventricular hemorrhage in 34 % cases, death in 28 % and sequelae mostly hemiparesis in another 28 % cases [3]. This may be attributed to a delay in diagnosis, ineffective antibiotic treatment, or suboptimal treatment regimens for neonatal CNS disease [3, 4]. Therefore, prompt diagnosis, early initiation and optimal duration of appropriate antibiotic therapy is necessary for a favorable prognosis.

The clinical presentation may include apnea, temperature instability, lethargy, vomiting, irritability poor tone, twitching or seizures [7]. Detection of Mycoplasma Hominis can be challenging since they lack peptidoglycan cell wall which renders them unidentifiable by gram staining [5]. In addition to this, they grow very slowly on routine culture media and require a specific blood agar medium for their detection [1, 6]. Due to this, there is a likelihood that Mycoplasma Hominis infections may remain undiagnosed or diagnosed late in infants presenting with clinical signs and symptoms [2]. Hence, it is important to consider the possibility of Mycoplasma Hominis infection in cases where the CSF shows pleocytosis and no growth of organism on routine culture media. A 16S ribosomal RNA PCR analysis has proven useful for detection of Mycoplasma Hominis in blood and CSF which are difficult to grow on standard culture media [7, 8]. The microbe is identified by direct sequencing analysis after amplification by PCR. Use of pathogen-specific primers in 16S RNA analysis results in rapid detection of the specific organism [4]. Prematurity, low birth weight, and neural tube defects are recognized to be the most common risk factors for neonatal meningitis with Mycoplasma Hominis [6]. However, it has also been seen in term neonates with no neurological birth malformations [1, 3].

The treatment options for Mycoplasma Hominis meningitis and its duration remains unclear [1, 9]. Due to the rarity of this infection in neonates, the current recommendations are based on clinical experiences and in-vitro susceptibility test results [9, 10].

Mycoplasma Hominis has shown susceptibility to Chloramphenicol, Tetracyclines, Lincosamide and Fluoroquinolones in in-vitro testing [11]. Fluoroquinolones have been used in past to treat neonatal mycoplasma hominins meningitis successfully [1, 3, 5]. Moxifloxacin, a fourth-generation fluoroquinolone is preferred because of its ability to concentrate in CSF and its bactericidal effects in CNS infections [1, 12, 13]. Although there are cases which were successfully treated with Moxifloxacin monotherapy [1, 6], there is a risk of development of resistance with fluoroquinolones during treatment [9, 14, 15]. Our literature search revealed only two reported cases of neonates who were given Moxifloxacin in combination with another antibiotic for Mycoplasma Hominis meningitis [3, 5] (Table 1).

Table 1 Characteristics of neonates treated with combination therapy (Moxifloxacin and another antibiotic) for Mycoplasma Hominis CNS infection
Full size table

Evidence from previously published cases show that 28 % neonates with Mycoplasma Hominis CNS infection died whereas 34 % had some CNS complications and 28 % cases developed some neurological sequelae mostly hemiparesis [3]. In our case, treatment with a combination therapy of Moxifloxacin and Doxycycline resulted in significant clinical and laboratory improvement in terms of negative CSF culture after 48 hours of initiating this regimen and decreased WBC count and protein in subsequent CSF analysis. Our patient was discharged home with a normal neurological examination and stable ventriculomegaly. His neurodevelopmental follow up at six months of age showed normal neurological findings.

It is important to consider Mycoplasma Hominis as a potential cause for neonatal meningitis in infants particularly those with previous colonization with this rare but devastating species. Our case demonstrates effective eradication of Mycoplasma Hominis with a combination therapy of Moxifloxacin and Doxycycline. However, further research is required to understand the pharmacokinetics of these antibiotics to establish optimal dosing and duration for effective treatment of CNS infections with Mycoplasma Hominis.