The study compared brain abscess and meningitis with regard to pre-existing comorbidities, clinical characteristics, laboratory results and microbiology, to explore potentially risk factors of brain abscess.
We found more neonates in brain abscess group had sPDA. In previous studies, the physiological right-to-left shunt of neonatal circulation had been validated to be one of risk factors of brain abscess [7]. But in our study, there were only 16 neonates with meningitis develop brain abscess during the past ten years, and only two of them had sPDA. So we couldn’t validate this finding due to limited data.
No significant difference was found in symptoms and nerve system signs, so it is difficult to predict the occurrence of brain abscess by symptoms and signs.
As regard to complications, more cases in brain abscess group developed respiratory failure, circulatory failure, and NEC. A possible explanation is that onset of brain abscess is usually associated with life-threatening infection, which may lead to organ dysfunction, thus affect the respiratory, cardiovascular systems and gastrointestinal tract [8–9]. These could also explain the larger proportion of mechanical ventilation and TPN dependent in brain abscess group as well as the longer antibiotic use. We also found more neonates develop encephalomalacia in brain abscess group. According to reports dealing with pathology of brain abscess, cerebral necrosis which is characteristic feature of brain abscess could lead to encephalomalacia, so the incidence of encephalomalacia is significantly higher in brain abscess group [10].
As for laboratory results, a significantly elevation in CRP was also found in brain abscess group compared with meningitis group, and we confirmed that CRP༞50mg/L is a risk factor of brain abscess. As CRP is an inflammatory marker, a high level of CRP not only indicate severe infection, but also support creation of a pro-inflammatory microenvironment which cause ongoing damage to neurocyte, and lead to necrocytosis which play an important role of abscess formation. [11–12].
CSF culture is the most important means to identify pathogenic bacteria. However, rate of positive CSF culture in our study was only 15%. Because lumbar puncture is not widely accepted in China, many parents agreed to perform lumbar puncture after initiation of antibiotics, which possibly biasing CSF culture results.
In neonates with positive bacteriology culture, we found a larger proportion of neonates were infected with gram negative bacteria, and Escherichia coli was the most encountered pathogen in brain abscess group, which is not consistent with previous studies that Citrobacter and Proteus were most commonly associated with brain abscesses. This probably due to difference in distribution of bacterial communities in different region that Escherichia coli were the most common pathogen of neonatal infection in China [13–15].
β-hemolytic streptococcus and MRSA were found in brain abscess group merely. A possible reason is that characteristics of these two bacteria make it easier for them to form abscess in neonates’ brain tissue. According to literatures,β-hemolytic streptococcus expresses hydrolytic enzymes which may lead to tissue liquefaction and pus formation[16–17]. MRSA will produce PVL (pore-forming leucotoxin), which produce several toxins that damage the membranes of host cells, and ultimately lead to cell lysis, which also play an important role in abscess formation [18–19].
We also found three species of pathogens (Escherichia coli, MRSA and Enterococcus faecium) in brain abscess group were common MDR bacteria in nosocomial infections, and multidrug resistant bacterial infection had been identified as risk factors of brain abscess. In China, empirical antibiotic treatment was used for anti-infection when neonates had increased inflammatory indexes or severe clinical manifestations before result of bacteriology culture and drug susceptibility test were achieved. However, MDR bacteria we found in brain abscess group were resistant to commonly prescribed antibiotics of CNS infection such as ampicillin or cefotaxime [20]. So infection in their brain tissue could not be effectively controlled, and the persistent neutrophil activation and accumulation in brain tissue caused inflammation spread, which may lead to abscess formation [21].
The results of this study indicate that MDR bacterial infection or the level of CRP on admission to the hospital represents the most important risk factors of brain abscess in neonates with bacterial meningitis. It is essential for a neonatologist to perform prevention, early identification, and provide appropriate interventions for these neonates.
There are also several limitations to this study and its results. First, as a retrospective study, lost of some clinical data may have caused an underestimation of their potential value. Second, it is a single-center study, which may affect the universality of our findings. Finally, the development in neonatal medical care during the past decades may act as a confounder when we explore risk factors of brain abscess.