Brevibacterium species are catalase-positive, nonspore forming, immotile, aerobic Gram-positive rods, closely related to corynebacteria. They are environmental bacteria often found in dairy products but also on human skin surfaces and have been considered as nonpathogenic commensals.1,2 However, they can cause severe infections like peritonitis, meningitis, cholangitis and sepsis, mainly in immunocompromised patients.1 Over the last decade, Brevibacterium catheter-related infection are also reported,3–6 probably related to the biofilm-forming capacity of this bacterium.7 There is conflicting evidence whether surgical removal of the catheter is needed to clear a Brevibacterium catheter-related infection,1,3–6 and no management guidelines are established. To date, we are only aware of 1 published case of a Brevibacterium infection in a child younger than 1 year of age, a neonate with an osteomyelitis.8 We here describe a newborn infant with hydrocephalus and a ventriculoperitoneal (VP) shunt infection caused by Brevibacterium casei. We discuss current literature and conservative management of device-related Brevibacterium infections.
A term born infant was prenatally diagnosed with an obstructive hydrocephalus due to intrauterine germinal matrix bleeding and secondary aqueduct stenosis. A VP shunt was inserted on day 5 of life due to rapidly increasing head circumference. He had a complicated postoperative course, initially with over-drainage and rebleeding. Later there were signs of underdrainage with mild subcutaneous cerebrospinal fluid (CSF) swelling along the shunt, and valve adjustments were done. He was discharged home after 1 month in hospital. Five days after discharge, he was readmitted presenting with increasing head circumference, bulging fontanel, irritability, CSF leakage from the surgical wound. There was also a subcutaneous lesion in the neck, stretching from the VP shunt valve to a surgical incision point behind the right ear. Some days later, he also developed fever (38.9 °C). Except for the CSF leakage and irritability, there was no other focus for infection. Laboratory tests did not show any signs of a systemic infection (Table 1). Ultrasound of the brain showed no signs of ventriculitis or meningitis.
TABLE 1. -
Summary of Laboratory Investigations, CSF Studies and Therapy
||Day 0* (33 Days)
||Day 4 (37 Days)
||Day 16 (49 Days)
||Day 41 (74 Days)
||Day 160 (193 Days)
| WBC (109/L)
| CRP (mg/L)
| PCT (μg/L)
| CSF culture
||Gram-positive rods, morphologically identical to those identified on day 4
||Vancomycin (15 mg/kg × 3) and rifampicin (10 mg/kg × 2) from days 8 to 23
*Onset of clinical signs.
†CSF sample obtained after disinfection of the skin overlying the subcutaneous fluid-filled lesion in the neck.
‡CSF sample obtained after disinfection of the skin overlying the valve of the VP shunt system.
CRP indicates C-reactive protein; NA, not available; PCT, procalcitonin; WBC, white blood cells.
MICROBIOLOGIC INVESTIGATIONS OF CSF
By aseptic procedure, a CSF sample for cell count and culture was obtained from the fluid-filled subcutaneous lesion in the neck. This sample showed markedly increased cell count (Table 1) and CSF protein of 4.8 g/L, but the Gram stain showed no visible bacteria. However, there was pure growth of Gram-positive rods, initially characterized as coryneform bacteria and interpreted as possible skin contamination. Due to persistent symptoms, a second CSF sample was obtained from the VP shunt valve 4 days later. This CSF sample also revealed growth of Gram-positive rods, morphologically identical to the first CSF sample. Species identification was now performed by using matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (Bruker Daltonics, Billerica, MA) and suggested B. casei. The antibiotic susceptibility pattern of both bacterial isolates obtained from the CSF was identical. Minimum inhibitory concentration (MIC) analysis using gradient strips (Liofilchem, Roseto degli Abruzzi, Italy) revealed the following values (mg/L): rifampicin 0.016, vancomycin 0.25, gentamicin 0.25, clindamycin 0.5, ciprofloxacin 0.5, cefotaxime 1.0, penicillin 1.5, chloramphenicol 4.0. There are no defined susceptibility breakpoints for Brevibacterium spp. We interpreted MIC values according to the suggested clinical breakpoints for Corynebacterium spp. in Europe (European Committee on Antimicrobial Susceptibility Testing; https://eucast.org/ast_of_bacteria).
TREATMENT, OUTCOME AND FOLLOW-UP
Following interdisciplinary discussion and a literature review, we decided to attempt intravenous antibiotic therapy with vancomycin (15 mg/kg every 8 hours) and rifampicin (10 mg/kg every 12 hours) and to leave the VP shunt system in situ. Intrathecal therapy was also considered but not started to avoid potential neurotoxicity and because of challenges with nonlinear clearance of antibiotics from CSF and finding an appropriate dose. The body temperature returned to normal 4–5 days after the initiation of antibiotic therapy, and his irritability gradually resolved. He received 15 days antibiotic therapy. Repeated CSF cultures, concurrent with clinical outpatient follow-up, were collected up to 4 months after onset of therapy and remained negative (Table 1).
Brevibacterium spp. are rarely causing infections in infants,8 and this is, to our knowledge, the first report of a Brevibacterium VP shunt infection in this age group. Our case demonstrates some salient features of a central nervous system shunt infection caused by opportunistic bacteria. Clinically the child had mainly low-grade fever and no biochemical signs of a systemic infection, in line with a previous report showing that Brevibacterium spp. only mildly activate the host immune response.1 However, cerebral irritability pointed to the VP shunt as a focus of infection. Proper identification of commensal bacteria that are possible skin contaminants, but also can cause infection, is challenging and require a high degree of suspicion. Current use of matrix-assisted laser desorption/ionization time-of-flight for species typing is very useful for the clinician and may lead to a more rapid diagnosis and appropriate therapy.1,3,5,7 It was only after the second CSF sample was identified as B. casei, we felt confident that this was a VP shunt infection.
Previous reports on Brevibacterium catheter-related infections demonstrate a range of therapeutic approaches. Often antibiotic therapy has been administered in combination with removal or replacement of catheters.1,3,5,7 In only a few case reports, catheter removal was not done, and the infection was cleared with antibiotics alone.4,6 In our case, the patient had a hydrocephalus that was clearly shunt-dependent. Removing the entire VP shunt system would have required general anesthesia and neurosurgery and temporal CSF diversion before a new operation with insertion of a VP shunt. We therefore decided first to attempt a more conservative approach. Our empiric choice of antibiotics included 2 antibiotics with low MICs for B. casei, bioavailability relevant for the site of the infection and a combination that often is used for device-related infections. Coryneform bacteria and B. casei are usually susceptible to glycopeptides.2 However, catheter-related infections are often associated with a biofilm on catheter surfaces, and vancomycin has poor activity in eradicating biofilm-embedded bacteria.9 In corynebacterial catheter-related bloodstream infection, catheter removal is not always necessary, in particular when therapy also consists of nonglycopeptide antibiotics.10 In our case, the combination of vancomycin plus rifampicin proved to be successful without the need for further surgical intervention. We decided to continue therapy at least 1 week after having a negative CSF culture (Table 1), administering a total of 15 days systemic antibiotic therapy. To ascertain that there was no regrowth after stopping antibiotics, we obtained 2 CSF cultures up to 4 months after end of therapy, and both remained negative.
The diagnosis of central nervous system shunt infection caused by opportunistic bacteria such as Brevibacterium spp. is challenging. Rapid species identification may guide appropriate management. Several reports indicate that catheter-related infections by coryneform bacteria, including Brevibacterium spp., may be treated with systemic antibiotics and without catheter removal. As demonstrated in our case, this can save a child from further surgical interventions. However, as this is a single-case report, cautious interpretation of data is needed, and our findings may not be generalizable. Similar infections need to be treated individually, based on both expert opinion and available scientific and clinical evidence.
1. Asai N, Suematsu H, Yamada A, et al. Brevibacterium paucivorans bacteremia: case report and review of the literature. BMC Infect Dis. 2019;19:344.
2. Troxler R, Funke G, Von Graevenitz A, et al. Natural antibiotic susceptibility of recently established coryneform bacteria. Eur J Clin Microbiol Infect Dis. 2001;20:315–323.
3. Althaf MM, Abdelsalam MS, Alsunaid MS, et al. Brevibacterium casei
isolated as a cause of relapsing peritonitis. BMJ Case Rep. 2014;2014:bcr2014203611.
4. Bal ZS, Sen S, Karapinar DY, et al. The first reported catheter-related Brevibacterium casei
in a child with acute leukemia and review of the literature. Braz J Infect Dis. 2015;19:213–215.
5. Joshi S, Misra R, Kirolikar S, et al. Catheter-related Brevibacterium casei
in a child with aplastic anaemia. Indian J Med Microbiol. 2020;38:226–228.
6. Piccinelli G, Morello E, Cancelli V, et al. Central venous catheter-related bloodstream infection
caused by Brevibacterium casei
in a hematology patient. Clin Microbiol Newsl. 2018;40.
7. Jermakow K, Fraczkiewicz F, Nowicka J, et al. Brevibacterium casei
- a new/rare pathogen of catheter-related bloodstream infections (CR-BSI) in immunocompromised children. In: ECCMID, April 25, 2017. Vienna, Austria; 2017.
8. Neumeister B, Mandel T, Gruner E, et al. Brevibacterium species as a cause of osteomyelitis in a neonate. Infection
9. Olson ME, Ceri H, Morck DW, et al. Biofilm bacteria: formation and comparative susceptibility to antibiotics. Can J Vet Res. 2002;66:86–92.
10. Ghide S, Jiang Y, Hachem R, et al. Catheter-related Corynebacterium bacteremia: should the catheter be removed and vancomycin administered? Eur J Clin Microbiol Infect Dis. 2010;29:153–156.