Secondary Logo

Journal Logo



Michelow, Ian C. M.B.B.Ch., F.C.Paed.(SA); McCracken, George H. Jr. M.D.; Luckett, Peter M. M.D.; Krisher, Karen Ph.D.

Author Information
The Pediatric Infectious Disease Journal: August 2000 - Volume 19 - Issue 8 - p 760
  • Free


The genus Abiotrophia refers to fastidious Gram-positive bacteria previously classified as nutritionally variant streptococci. 1 Since Frenkel and Hirsch 2 first described these organisms in 1961, knowledge has accumulated regarding their unique nutritional requirements, variable morphologic appearance, biochemical characteristics and antimicrobial susceptibility. 3 Current taxonomy distinguishes three species of nutritionally variant streptococci:Abiotrophia defectiva, Abiotrophia adiacens and Abiotrophia elegans. 1, 4Abiotrophia para-adiacens, a potentially new species that is biochemically similar to A. adiacens, has been recently characterized. 5

The recovery of Abiotrophia spp. from the normal flora of the oral cavity and from the intestinal and urogenital tracts is well-described. 5–7 Recently these relatively avirulent microbes have attracted attention because they have caused invasive disease in immunocompetent hosts in a variety of clinical settings. Abiotrophia spp. have been implicated in endocarditis in 5 to 6% of microbiologically proved cases and may be a cause of culture-negative endocarditis. Numerous reports have implicated Abiotrophia spp. in adults with a variety of systemic infections including endophthalmitis, vertebral osteomyelitis and discitis, iatrogenic meningitis and brain abscess after neurosurgery. 3, 8–11 Little is known about the pathogenesis and clinical presentations of disease caused by these organisms in children.

We describe the clinical manifestations, management and outcome of a child with a cerebral abscess caused by Abiotrophia spp.

Case report.

A 30-month-old previously healthy girl with Down’s syndrome was transferred to Children’s Medical Center of Dallas (CMC) 1 day after presenting to a community hospital in Oklahoma with fever, progressive lethargy and hypotonia. Her past medical history was significant for mildly delayed neuromotor and speech development consistent with trisomy 21 disorder. She had received only the first in the series of routine immunizations in early infancy. There were no previous experiences of sinopulmonary, middle ear or cutaneous infections or symptoms attributable to cardiac disease. In addition we learned that her diet was regularly supplemented with unpasteurized cows’ milk from a local dairy.

Two weeks before admission she developed severe tongue lacerations likely associated with teething and relative macroglossia. Forty-eight hours before admission she developed low grade intermittent fever, emesis and nondysenteric diarrhea. The referring physician’s initial evaluation revealed a pale, dehydrated and febrile young girl with an axillary temperature of 40°C. There was no clinical evidence of a neurologic infectious process. The salient laboratory findings at the referring hospital included the following: white blood cell count of 28 000/μl with 69% polymorphonuclear cells, 20% band forms, 6% lymphocytes and 5% monocytes. Additional findings included: a hemoglobin value of 11 g/dl, a platelet count of 185 000/μl and a sodium concentration of 132 mmol/l. Empiric antibiotic therapy was commenced with iv cefotaxime (100 mg/kg/day). Twelve hours later the patient became unresponsive to stimulation and had fixed, dilated pupils. She was promptly intubated, hyperventilated and given iv dexamethasone and mannitol. A cranial computed tomography (CT) scan revealed a large right-sided cerebral mass.

The patient was transferred to the CMC intensive care unit where physical examination revealed an afebrile, hemodynamically stable young girl with morphologic features of Down’s syndrome, asymmetric dilated pupils, multiple dental caries and ulcerations of the tongue and buccal mucosa with overlying plaques. A soft precordial ejection systolic murmur was audible but there were no stigmata of cyanotic congenital heart disease. No carotid artery bruit or thrill was detected.

A cranial CT scan with contrast demonstrated panventriculomegaly, with significant asymmetric dilatation of the right lateral ventricle, right to left midline shift of 1.1 cm and right uncal herniation. Massive vasogenic edema surrounding the occipital and temporal horns of the right lateral ventricle was evident, and debris formed a fluid level in the right lateral ventricle. These findings as well as the ependymal enhancement throughout the right lateral, third and part of the fourth ventricle were consistent with ventriculitis. A focus of presumed infection, identified as a ring-enhancing lesion in the right centrum semiovale, likely represented the primary process that had ruptured into the lateral ventricle. The paranasal and mastoid sinuses were unaffected.

Surgical decompression of the ventricles was accomplished with bilateral ventriculostomies. The aspirated fluid resembled pus, and analysis of the fluid was as follows: protein 505 mg/dl, glucose <20 mg/dl, white blood cell count 6940/mm 3 (81% polymorphonuclear cells, 15% band forms, 4% lymphocytes), red blood cell count 250/mm 3. Microscopy of the cerebrospinal fluid (CSF) obtained from both lateral ventricles revealed pleomorphic Gram-positive cocci in pairs and chains. Anaerobic, fungal and mycobacterial CSF cultures and serial blood cultures were sterile.

Cerebrospinal fluid from each ventricle was streaked on tryptic soy blood agar plates with 5% sheep blood (BAP) and chocolate agar plates (CAP) (Remel, Inc., Lenexa, KS). The plates were incubated for 24 h in 5% CO2 at 35°C. Numerous small colonies of a single morphotype were isolated on both CAP. One BAP plate yielded a single colony of coagulase-negative staphylococci. Several well-isolated colonies from the CAP were inoculated on a BAP, which was then cross-streaked with Staphylococcus aureus. Within 24 h tiny alpha-hemolytic colonies appeared on the BAP that grew only in areas adjacent to the colonies of S. aureus. Additional testing confirmed that the isolate was catalase-negative, leucine aminopeptidase-positive and pyrrolidonylarylamidase (PYR)-positive. These findings fulfilled the criteria for a presumptive identification of Abiotrophia spp. Biochemical reactions obtained using the RapID ANA II system (Remel, Inc.) included positive reactions for beta-glucosidase and leucine arylamidase production and negative reactions for alkaline phosphatase and alpha- and beta-galactosidases. These biochemical reactions most closely resembled those patterns obtained for A. adiacens. The isolate was differentiated from various species of streptococci based on the positive PYR reaction and ability to grow on BAP only when exhibiting satellitism with S. aureus.

A standardized testing method and interpretive criteria for performance of susceptibility testing of Abiotrophia spp. has not yet been approved by the National Committee for Clinical Laboratory Standards (NCCLS). MIC testing using the Epsilometer (Etest) system (AB Biodisk, Solna, Sweden) and chocolate agar as the growth medium was obtained for several antimicrobial agents. MIC values of 0.06, 0.125, 0.064 and 2 μg/ml were obtained for penicillin, cefotaxime, clindamycin and vancomycin, respectively.

At the time of admission to CMC empiric therapy included iv nafcillin, cefotaxime and metronidazole as well as an 8-day tapering course of dexamethasone for cerebral edema. After the presumptive microbiologic diagnosis was made, the antibiotic regimen was changed to iv vancomycin, rifampin and gentamicin for possible synergistic activity. Subsequent repeat CSF cultures were sterile although samples from the ventricular drain on the third and fourth days of hospitalization revealed Gram-positive cocci in pairs and chains. When the susceptibility data were available, ampicillin (400 mg/kg/day divided 6-hourly) was substituted for vancomycin. Gentamicin was discontinued after 7 days of therapy. Duration of combination therapy with iv ampicillin (60 days) and sequential iv and enteral rifampin (32 days) was guided by clinical response, serial laboratory indices of inflammation and cranial CT scans with contrast. The erythrocyte sedimentation rate and C-reactive protein values improved from a maximum of 131 mm/h and 18.7 mg/dl, respectively, to 9 mm/h and <0.7 mg/dl, respectively, after a total of 64 days of antimicrobial therapy.

Transthoracic echocardiography documented the presence of an incomplete atrioventricular septal defect with aneurysmal valve tissue attached to the ventricular septum. There were mild tricuspid and mitral valve regurgitation and a left to right intracardiac shunt. Cardiac function was normal, and no valvar or mural vegetations could be identified. Chest roentgenogram and urinalysis were normal.

The patient’s clinical progress was gradual because of a number of complications. Recurrent generalized tonic-clonic seizures that started 3 days after admission were difficult to control partly because fosphenytoin metabolism was enhanced by rifampin. Transient hyponatremia (lowest sodium concentration, 127 mmol/l) that developed 4 days after admission was explained by inappropriate secretion of antidiuretic hormone (4.4 pg/ml; normal range for age, 0.7 to 3.8 pg/ml). Serial cranial CT scans demonstrated asymmetric loculated hydrocephalus that fluctuated in size. Multiple ventriculostomy tube revisions were required to achieve optimal CSF drainage. The neurologic deficits could be attributed to the severe necrosis at the primary focus of infection, as well as to an evolving infarction in the distribution of the right posterior cerebral artery that became apparent during the second week of hospitalization. The distribution and severity of inflammation and edema suggested that either a prior embolic phenomenon had preceded the intracranial infectious process or that cerebrovascular thrombosis had developed. During the fifth week of hospitalization, once the CSF laboratory indices of inflammation had improved, a ventriculoperitoneal shunt was inserted with good effect. The patient was discharged from hospital at the end of the seventh week of therapy and continued to receive supervised iv therapy at home.

After completion of antibiotic therapy, persistent neurologic sequelae were observed. There were asymmetric pupillary responses, strabismus, regression of motor and speech developmental milestones, hypotonia and left hemiplegia. Gross motor function improved moderately with a home physical therapy program. The visual evoked responses were normal. Hearing was clinically intact but early auditory brainstem response evaluation revealed moderate right peripheral auditory impairment.

Underlying immunodeficiency was considered unlikely because of normal findings for immunoglobulin G, A and M isotypes; total hemolytic complement; chemiluminescence; and T cell subset numbers. A transient moderate deficiency of IgG was noted with proportionate decreases in the IgG subclasses. A total IgG concentration of 285 mg/dl that was obtained during Week 11 postdiagnosis increased to 413 mg/dl 2 weeks later. Evaluation of erythrocyte morphology reflected normal splenic function.


Intracranial infection caused by Abiotrophia spp. has not been previously reported in children. In the present case ulceration of the oral mucosa and the presence of dental caries likely permitted invasion of commensal Abiotrophia organisms. Ingestion of unpasteurized cow’s milk by the patient was another possible contributor to the presence of Abiotrophia in the patient. A recent report documented the isolation of Abiotrophia spp. from a case of bovine mastitis. 12 However, we were unsuccessful in two attempts to recover this organism from milk obtained from the original source used by the patient. Immune dysfunction, as described in Down’s syndrome, was not conclusively proved, but subtle defects could not be ruled out. 13 It is likely that transient bacteremia predisposed the patient to infective endocarditis. Although intracardiac vegetation was not visualized, the CT finding of a cerebral abscess in the territory of a large infarction suggested the possibility of a prior infected thromboembolic event, likely from a cardiac source. Alternatively invasive Abiotrophia organisms could have gained access to the systemic circulation via the atrioventricular septal defect, although echocardiography did not document a right-to-left intracardiac shunt.

Possible explanations for the low recovery rate of Abiotrophia spp. in clinical specimens include the paucity of organisms, delay in transportation to the laboratory, relatively slow growth rate and variable efficacy of microbiologic media to support the growth of these fastidious streptococci. 8, 14 Many commonly used laboratory media, such as unsupplemented tryptic soy agar with 5% sheep blood, are incapable of supporting their growth. 14 Culture media supplemented with 0.001% pyridoxal hydrochloride or 0.01% l-cysteine is required to facilitate isolation. Alternatively Abiotrophia spp. can be presumptively identified by their propensity to derive nutrients from adjacent “helper” colonies that are capable of supporting a halo of satellite growth, a phenomenon called microbial commensalism. 3

Because no standardized susceptibility testing method is currently recommended by the NCCLS, direct comparison of antimicrobial susceptibility data between laboratories is difficult, especially if different types of culture medium and incubation conditions are used. 15 Low MIC values were obtained with antimicrobial agents tested against the patient’s isolate; however, recent data using the current in vitro NCCLS interpretive susceptibility criteria for viridans streptococci demonstrate a trend toward increased MIC values to beta-lactam agents for Abiotrophia spp. 16, 17 Among 41 clinical isolates of A. adiacens and A. defectiva, the proportions of strains with intermediate and high level resistance to beta-lactam antibiotics were 43 and 85%, respectively, for penicillin (MIC >0.12 μg/ml) and 36 and 15%, respectively, for ceftriaxone (MIC >0.5 μg/ml). 18 Both species were fully susceptible to clindamycin (MIC <0.25 μg/ml), rifampin (MIC <1 μg/ml) and vancomycin (MIC <1 μg/ml). In addition meropenem susceptibility (MIC <0.5 μg/ml) was found in 96% of A. adiacens isolates and 100% of A. defectiva isolates.

Monotherapy that is usually effective for other streptococci has been associated with higher rates of treatment failure and mortality when used for Abiotrophia spp. Empiric treatment using combinations of two or more synergistic antimicrobial agents such as vancomycin and rifampin or vancomycin and an aminoglycoside 17 may be advantageous for managing central nervous system infections involving Abiotrophia spp. In addition clindamycin has favorable in vitro activity and may be efficacious for treating certain types of infections. Conversely clinical experience suggests that in vitro susceptibility data do not correlate well with clinical outcome in endocarditis caused by Abiotrophia spp. 19

1. Kawamura Y, Hou XG, Sultana F, Liu H, Yamamoto H, Ezaki T. Transfer of Streptococcus adjacens and Streptococcus defectivus to Abiotrophia gen. nov. as Abiotrophia adiacens comb. nov. and Abiotrophia defectiva comb. nov., respectively. Int J Syst Bacteriol 1995; 45:798–803.
2. Frenkel A, Hirsch W. Spontaneous development of L forms of streptococci requiring secretions of other bacteria or sulphydryl compounds for normal growth. Nature 1961; 191:728–30.
3. Ruoff KL. Nutritionally variant streptococci. Clin Microbiol Rev 1991; 4:184–90.
4. Roggenkamp A, Abele-Horn M, Trebesius K, Tretter U, Autenrieth IB, Heesemann J. Abiotrophia elegans sp. nov., a possible pathogen in patients with culture-negative endocarditis. J Clin Microbiol 1998; 36:100–4.
5. Kanamoto T, Sato S, Inoue M. Genetic heterogeneities and phenotypic characteristics of strains of the genus Abiotrophia and proposal of Abiotrophia para-adiacens sp. nov. J Clin Microbiol 2000; 38:492–8.
6. Roberts RB, Krieger AG, Schiller NL, Gross KC. Viridans streptococcal endocarditis: the role of various species, including pyridoxal-dependent streptococci. Rev Infect Dis 1979; 1:955–66.
7. George RH. The isolation of symbiotic streptococci. J Med Microbiol 1973; 7:77–83.
8. Namdari H, Kintner K, Jackson BA, et al. Abiotrophia species as a cause of endophthalmitis following cataract extraction. J Clin Microbiol 1999; 37:1564–6.
9. Heath CH, Bowen SF, McCarthy JS, Dwyer B. Vertebral osteomyelitis and discitis associated with Abiotrophia adiacens (nutritionally variant Streptococcus) infection [Letter]. Aust N Z J Med 1998; 28:663.
10. Schlegel L, Merlet C, Laroche JM, Fremaux A, Geslin P. Iatrogenic meningitis due to Abiotrophia defectiva after myelography. Clin Infect Dis 1999; 28:155–6.
11. Bierman C, Fries G, Jehnichen P, Bhakdi S, Husmann M. Isolation of Abiotrophia adiacens from a brain abscess which developed in a patient after neurosurgery. J Clin Microbiol 1999; 37:769–71.
12. Higgins R, Messier S, Désilets A. Isolation of Abiotrophia spp. with Staphylococcus aureus in a case of mastitis in a cow. Can Vet J 1997; 38:114.
13. Loh RK, Harth SC, Thong YH, Ferrante A. Immunoglobulin G subclass deficiency and predisposition to infection in Down’s syndrome. Pediatr Infect Dis J 1990; 9:547–51.
14. Reimer LG, Reller LB. Growth of nutritionally variant streptococci on common laboratory and 10 commercial blood culture media. J Clin Microbiol 1981; 14:329–32.
15. Clark RB, Gordon RE, Bottone EJ, Reitano M. Morphological aberrations of nutritionally deficient streptococci: association with pyridoxal (vitamin B6) concentration and potential role in antibiotic resistance. Infect Immun 1983; 42:414–7.
16. Bosley GS, Facklam RR. Biochemical and antimicrobic testing of “nutritionally variant streptococci” [Abstract C-307]. In: Abstracts of 90th Annual Meeting of the American Society for Microbiology. Washington, DC: American Society for Microbiology, 1990: 395.
17. Douglas CP, Siarakas S, Gottlieb T. Evaluation of E Test as a rapid method for determining MICs for nutritionally variant streptococci. J Clin Microbiol 1994; 32:2318–20.
18. Tuohy M, Procop GW, Washington JA. Antimicrobial susceptibility of Abiotrophia adiacens and Abiotrophia defectiva [Abstract 2277]. In: Proceedings of the 39th Interscience Conference on Antimicrobial Agents and Chemotherapy, San Francisco, September 26 to 29, 1999. Washington, DC: American Society for Microbiology, 1999: 271.
19. Johnson CC, Tunkel AR. Viridans streptococci and groups C and G streptococci. In: Mandell GL, Bennet JE, Dolin R, eds. Principles and practice of infectious diseases. New York: Churchill Livingstone, 1995: 1845–61.

Brain abscess; Abiotrophia; nutritionally variant streptococci; child

© 2000 Lippincott Williams & Wilkins, Inc.