Enterococcus avium (EA) is a rare human pathogen, and only a few case series and reports exist. For example, Patel et al reported 9 cases of EA bacteremia at their institution from 1986 to 1991.1 Patel et al also noted that only 12 cases had been reported, 10 of which had no clinical context.1 According to these few reports, EA has caused a variety of different types of clinical illness. These cases of EA infection involved vancomycin-susceptible EA (VSEA). Rosato et al2 described the mechanism of resistance to vancomycin of a vancomycin-resistant EA (VREA) clinical isolate but failed to provide any clinical information. Similarly, other reports describe VREA as clinical isolates but then fail to provide additional clinical information.3,4 As a result, little is known about the epidemiology, clinical manifestations, laboratory features, and clinical course of VREA infections. This report includes 2 cases of EA infections with resistance to vancomycin. This is the first report that we are aware of that describes both mechanism of resistance and clinical manifestations of VREA infections.
Case Definitions and Current Cases
Cases of infections were defined as a systemic inflammatory response due to the isolation of clinically significant EA from sterile body fluid in a patient at least 18 years. A review of the medical records at the Johnson City Medical Center from 1993 to June of 2002 yielded 2 cases of VREA infections, both of which occurred in 2002.
A PubMed search was conducted to identify all adult cases of infections due to EA. Cases were considered for inclusion in this review only if patients were 15 years or older. Relevant articles were identified from 1993 to October 2003 using the key words Enterococcus and streptococci cross-referenced with avium. Secondary references were also reviewed.
The following materials and methods were used for recovery of and in vitro antimicrobial susceptibility testing of EA.
Blood samples were obtained aseptically and sent to the Department of Microbiology, Johnson City Medical Center, Johnson City, Tennessee. Ten milliliters of blood were inoculated into each of the following: a BacT/Alert standard aerobic and a BacT/Alert standard anaerobic culture bottle and incubated in the BacT/Alert 3D Microbial Detection System. Bottles were thereafter monitored continuously for signs of bacterial growth. Once growth was detected, blood was removed from the bottle with microbial growth and plated to 5% sheep blood agar/blood agar plates and chocolate agar obtained from BBL Microbiology Systems. The agar plates were then incubated at 35°C in 5% CO2 for about 24 hours. Characteristic morphologic colonial facultative anaerobic growth was analyzed under light microscopy, and gram-positive organisms in pairs and chains were then selected for further characterization. Enterococci were identified by way of growth in 6.5% NaCl broth and hydrolysis of esculin on bile-esculin agar. EA was initially identified by the Vitek system using a gram-positive identification card. EA identification was also thereafter confirmed using a manual identification system performed by the Division of Microbiology, James H. Quillen College of Medicine, East Tennessee State University, using the API 20 Strep system. Susceptibilities were initially derived by the Vitek system. Resistance to vancomycin was thereafter confirmed by a vancomycin E test (AB Biodisk, Sweden).
Sterile body fluid samples were obtained aseptically and sent to the Department of Microbiology, Johnson City Medical Center. Body fluid samples were plated to 5% blood agar plates, chocolate agar, and brain heart infusion broth. The blood agar plates and chocolate agar were incubated at 35°C in 5% CO2 for about 24 hours. The brain heart infusion broth was incubated at 35°C in ambient air. Characteristic morphologic colonial facultative anaerobic growth was then processed and examined by the methods described above under blood cultures. Similarly, standard procedures and routine microbiologic techniques such as the semiquantitative culture technique as described by Maki et al5 were used for prosthetic devices such as central venous catheter tips.
Polymerase Chain Reaction Amplification
For all bacterial strains used, amplification was performed on genomic DNA purified from an overnight brain heart infusion culture using the GenElute Bacterial Genomic DNA Kit (Sigma) according to manufacturer's directions. One hundred nanograms of genomic DNA were used in a polymerase chain reaction (PCR) containing 50 mM KCl, 10 mM Tris-HCl (pH 9.0), 2.5 mM MgCl2, 0.2 mM each Enterococcus-specific primer (either the EntVanA pair or the EntTuf pair), 200 mM each deoxynucleoside triphosphate (Invitrogen), 1 U Taq polymerase (Pierce). The sequences for the Enterococcus vancomycin A (vanA) gene specific primers are EntVanA forward: GGT GGC AGC TAC GTT TAC CTA T and EntVanA reverse: TTT CCC AAT ACC GCA CAA C. The sequences for the Enterococcus Tuf gene specific primers are EntTuf forward: TAC TGA CAA ACC ATT CAT GAT G and EntTuf reverse: ACC TTC GTC ACC AAC GCG AAC. The thermocycling parameters used a 2-step system: 68°C for 30 seconds and 94°C for 15 seconds for 35 cycles. PCR products were examined on a 3% NuSieve agarose gel. The EntTuf product is 112 bp and the EntVanA is 189 bp. The EntTuf reaction confirmed that the isolated DNA was from an Enterococcus sp. and that the DNA had no inhibitors of PCR reactions.6
The patient was a 63-year-old Caucasian female who was admitted for recurrent abdominal pain. Her medical history was significant for gallstone pancreatitis with pseudocyst formation. A percutaneously placed catheter into the pseudocyst initially yielded sterile and nonmalignant fluid. She was maintained on bowel rest and total parenteral nutrition provided through a Hickman catheter. She also had an allergy to penicillin.
Her physical examination was notable for a tender abdomen. Her paraclinical data were significant for 18,400 white blood cells per microliter of blood. An abdominal-pelvic computed tomographic scan showed a residual 4.0 × 3.0 cm pseudocyst and a malpositioned percutaneous catheter.
The patient was initiated on empiric intravenous vancomycin and levofloxacin. Blood and urine cultures were negative. She underwent a computed tomography-guided percutaneous aspiration of the residual pancreatic pseudocyst, which yielded the growth of VREA, pansusceptible Escherichia coli, and Klebsiella pneumoniae. The VREA was susceptible to linezolid and penicillin. The vancomycin was thereafter replaced by linezolid and metronidazole. A rectal swab collected thereafter confirmed that her gastrointestinal tract was also colonized with VREA.
The patient defervesced. Her abdominal pain and leukocytosis resolved. A repeat abdominal-pelvic computed tomographic scan about 2 weeks after demonstrated resolution of the pseudocyst. She was discharged home to complete a 4-week course of linezolid, levofloxacin, and metronidazole.
This case demonstrates VREA as a contributory cause of an infected pancreatic pseudocyst. Of note, the patient was colonized with VREA. She also had many risk factors for the acquisition and colonization of her gastrointestinal tract with vancomycin-resistant enterococci (VRE) such as critical illness, multiple comorbidities, and an extended length of hospital stay. We surmise that the source of her infected pancreatic pseudocyst was her gastrointestinal tract. We suspect that VREA and other gastrointestinal flora seeded the pancreatic pseudocyst by translocation. Although this particular strain of VREA was susceptible to penicillin, the patient had reported an allergy to penicillin. We did not perform allergy tests to confirm this history or try to desensitize her to penicillin. Her infected pancreatic pseudocyst responded to percutaneous drainage and antimicrobial therapy, which included linezolid. On her discharge, her fever, abdominal pain, and leukocytosis had resolved.
The patient was an 85-year-old Caucasian female admitted for abdominal pain. She underwent an exploratory laparotomy, which revealed incarcerated small bowel secondary to adhesions. She underwent small bowel resection for ischemic colitis. Despite this intervention, she became progressively more debilitated. Specifically, her postoperative course was very complicated and notable for recurrent ischemic colitis, anemia, anasarca, deep venous thrombosis, renal insufficiency, depression, and a nosocomial methicillin-resistant Staphylococcus aureus pneumonia. A Foley catheter had been inserted because of her debilitated state. A urine culture was obtained to evaluate one of her many fevers and grew out at least 100,000 colony forming units of VREA/mm urine. Her blood cultures were without growth. Because the patient had complained of suprapubic discomfort, the patient was treated with intravenous linezolid 600 mg/d for 12 hours for 1 week for a VREA catheter-associated urinary tract infection (UTI). The Foley catheter was exchanged and a repeat urine culture was without growth. A rectal swab was subsequently found to be positive for both VREA and vancomycin-resistant Enterococcus faecium (VRE). The patient eventually expired due to a myocardial infarction.
This case demonstrates that VREA is a cause of nosocomial UTI. Of note, the patient was colonized with both VREA and vancomycin-resistant E. faecium. She also had many of the aforementioned risk factors for the acquisition and colonization of her gastrointestinal tract with VRE. We suspect that fecal flora such as VREA ascended the indwelling catheter to her bladder and then caused cystitis. This particular strain of VREA was resistant to both vancomycin and penicillin. This strain of VREA did, however, remain susceptible to nitrofurantoin. We chose to avoid the use of nitrofurantoin secondary to her renal insufficiency. Treatment with linezolid and exchange of the Foley catheter were successful. She remained afebrile and without suprapubic discomfort for several weeks before she had expired.
Table 1 describes the biochemical profiles of the 2 strains of VREA isolated at our institution.
Table 2 summarizes the features of the 2 cases of VREA infections. Patient 1 was diagnosed with a polymicrobial (with VREA)-infected pancreatic pseudocyst. Patient 2 was diagnosed with a VREA catheter-associated UTI.
Table 3 summarized the features of the 17 cases of VSEA infections reported previously. Two cases of bacterial meningoencephalitis probably involved EA as a colonizer of a decubitus ulcer and thus were not included.7 One case of bacteremia secondary to thrombosis of the lateral sinus and the internal jugular vein was reported in Spanish and thus was not included.8 Three reports describe various EA infections in the pediatric population, which were not included in this review.9-11
Table 4 describes the antimicrobial susceptibility profile of the 2 strains of VREA from our institution.
Table 5 displays the VRE colonization rates for the years 2001 to 2003. The rate essentially doubled to about 3.3% from the year 2001 to 2003.
Figure 1 displays the electrophoretic gel of the PCR analysis for 1 strain of VREA from our institution (case 1).
Over the last 20 years, enterococci have become significant nosocomial pathogens. Enterococci are the third most commonly isolated nosocomial pathogen and overall cause about 12% of all nosocomial infections.20 Mortality rates associated with an enterococcal infection have been reported to be as high as 50%.20 Although the enterococci lack specific virulence factors, several factors are believed to have facilitated their emergence as significant nosocomial pathogens. These factors include antimicrobial abuse, their ability to survive under rather harsh conditions, their intrinsic resistance to a variety of different antimicrobials, and, perhaps most importantly, their unusual capacity to exchange genetic information among themselves and with other genera.21 Indeed, resistance to antimicrobials such as vancomycin has evolved and undoubtedly contributed to the increasing prevalence of enterococcal nosocomial infections. Since the first reports of their existence in the United States in the late 1980s, there has been a dramatic increase in isolation of VRE identified by the Centers for Disease Control and Prevention's National Nosocomial Infection Surveillance System. The percentage of nosocomial enterococci resistant to vancomycin increased from 0.3% to 21% among non-intensive care unit patients, while the percentage among intensive care unit patients increased form 0.4% to 23% between 1989 and 1998.21
Mackowiak22 in a review of all 295 enterococcal isolates at his institution found 12 EA isolates but concluded that the results of the investigation could not establish EA as a pathogen in human diseases. In this report, Mackowiak reported that EA was more resistant to penicillin G and ampicillin and less resistant to other antimicrobial agents. The minimal inhibitory concentration (MIC90) for EA to penicillin G and ampicillin in the 12 EA isolates was 16 and 14.5 μg/mL, respectively. In contrast, the MIC90 to penicillin G and ampicillin for the 329 Enterococcus faecalis isolates was 3.4 and 1.2 μg/mL, while the MIC90 to penicillin G and ampicillin in the 22 E. faecium isolates was 8 and 4 μg/mL, respectively. Of note, the MIC90 to vancomycin of the 12 EA isolates was <2 μg/mL.
Patel et al1 described 9 cases of EA infections in humans after a review of blood cultures at their institution from 1986 to 1991. These 9 cases are included in Table 3. All 9 of the EA isolates were susceptible to vancomycin. Patel et al1 concluded that EA not only was a human pathogen but also was often associated with gastrointestinal abnormalities.
Table 3 also provides the 8 other cases identified after a review of the English literature. VSEA usually caused gastrointestinal or genitourinary tract infections (Table 3). VSEA also was less commonly the cause of an endovascular infection such as endocarditis and catheter-related sepsis (Table 3). Isolated reports of breast prosthesis infection, cellulitis, pneumonia, osteomyelitis, and splenic abscess were also described (Table 3). All of the aforementioned EA isolates were susceptible to vancomycin.
The first report of enterococci resistant to high concentrations of glycopeptide antibiotics was published in 1988.23 Various types of VRE have been characterized on phenotypic and genotypic bases.23 Several mechanisms of vancomycin resistance exist.23 VRE with vanA-mediated resistance is resistant to high levels of vancomycin (>64 μg/mL).23 The vanA glycopeptide resistance gene clusters are carried on transposons.23 The mechanism of the acquired vancomycin resistance mediated by vanA VRE involves the encoding of an alternate biosynthetic pathway for the production of a mutated cell wall d-Ala-d-Lac precursor that bind vancomycin poorly.23 Risk factor for the acquisition of vancomycin-resistant E. faecium have been described and include critical illness, multiple comorbidities, and extended length of hospital stay.24 It is quite likely the same risk factors exist for the acquisition of VREA.
VREA appears to cause infections of the gastrointestinal and genitourinary tract. Although species identification error has been described for EA, Table 1 highlights the confirmatory biochemical properties of the 2 VREA isolates from our institution.25 Table 2 describes the 2 cases of VREA infections from our institution. Case 1 describes an elderly woman who developed a polymicrobial with VREA-infected pancreatic pseudocyst. Case 2 describes an elderly woman who developed a VREA catheter-associated UTI. Table 4 describes the antimicrobial susceptibilities of both EA isolates from our institution. Both EA isolates were found to be resistant to vancomycin (MIC > 128 μg/mL). These are the first VREAs that we are aware of for which their role as a human pathogen is clinically described. To elaborate the mechanism of resistance to vancomycin, PCR studies were conducted. Figure 1 displays the electrophoretic study of the VREA isolate from case 1. The electrophoretic pattern of the strain of VREA from case 1 is consistent with the presence of the vanA gene. This mechanism of resistance has been described in a vancomycin-dependent strain of EA.26
Table 4 displays the antimicrobial susceptibilities of both strains of VREA. Both VREA infections responded to drainage of pus as well as a course of linezolid. The most recently approved drug with good activity against enterococci is linezolid.23 Linezolid has nearly uniform activity against enterococci although it has only bacteriostatic activity, and reports of resistance in vancomycin-resistant E. faecium have been recently described.23,27 One strain retained activity to penicillin. We did not perform synergy studies. A previous report suggests possible but not uniform synergy when vancomycin and gentamicin were used for VSEA. However, another report described VSEA with high-level gentamicin resistance. Indeed, aminoglycoside-modifying enzymes encoding genes have been identified in EA by multiplex PCR studies. Unfortunately, combinations of linezolid with ampicillin, gentamicin, or streptomycin are indifferent.27
Table 5 displays the VRE colonization rates for the years 2001 to 2003. Of interest, VREA infections were first noted at our institution with VRE colonization rates of about 3% to 4%. Likely, VREA infections will increase if VRE colonization rates continue to increase.
VREA, like its vancomycin-susceptible counterpart, is a human pathogen. It can cause UTI and infections of pancreatic pseudocyst. Likely, VREA colonization of the gastrointestinal tract is a risk factor for VREA infections. An increase in the VRE colonization rates may thus be a risk factor for VREA infections. An assessment of VREA's virulence will require future research, but both of our cases involved hosts with several comorbidities such as pancreatitis and ischemic colitis. Likely, resistance to vancomycin contributed to VREA's pathogenicity, however. Further investigations should be performed to assess as to how VREA has acquired resistance to vancomycin. We surmise that other enterococcal species with resistance to vancomycin such as vancomycin-resistant E. faecium are capable of transferring the necessary genetic information that when expressed in EA lead to resistance to vancomycin. Resistance to vancomycin is mediated at least in part by vanA-mediated resistance. Infection control measures should thus be considered whenever VREA is isolated in a nosocomial setting. We know of no community-acquired cases of VREA infections, and like other strains of VRE, it is likely to be a part of nosocomial flora. Therapy with linezolid was effective.
1. Patel R, Keating MR, Cockerill FR, et al. Bacteremia due to Enterococcus avium
. Clin Infect Dis
2. Rosato A, Pierre J, Billot-Klein D, et al. Inducible and constitutive expression of resistance to glycopeptides in glycopeptide-resistant Enterococcus avium
. Antimicrob Agents Chemother
3. Pearce CL, Evans MK, Peters SM, et al. Clonal diversity of vancomycin-resistant enterococci from an outbreak in a tertiary care university hospital. Am J Infect Control
4. van den Braak N, Power E, Anthony R, et al. Random amplification of polymorphic DNA versus pulsed field gel electrophoresis of Sma
I macrorestriction fragments for typing strains of vancomycin-resistant enterococci. FEMS Microbiol Lett
5. Maki DG, Weisse CE, Sarafin HW. A semiquantitative culture method for identifying intravenous-catheter-related infection. N Engl J Med
6. Ke D, Picard FJ, Martineau F, et al. Development of a PCR assay for rapid detection of enterococci. J Clin Microbiol
7. Fujimoto C, Yazawa S, Matsuoka F, et al. Bacterial meningoencephalitis in patients undergoing chronic hemodialysis: two case reports. Clin Neurol Neurosurg
8. Lozano F, Esteban F, Florez C, et al. Bacteremia caused by Enterococcus avium
secondary to thrombosis of the lateral sinus and the internal jugular vein. Enferm Infecc Microbiol Clin
9. Das I, Gray J. Enterococcal bacteremia in children: a review of seventy-five episodes in a pediatric hospital. Pediatr Infect Dis J
10. Green PA. Transient, asymptomatic bacteremia due to Enterococcus avium
in a 33 month old. Clin Infect Dis
11. Rosales AM, Bolivar J, Burke RP, et al. Enterococcus avium
endocarditis in an infant with tetralogy of Fallot. Pediatr Cardiol
12. Perez-Castrillon JL, Martin-Luquero M, Martin-Escudero JC, et al. Endocarditis caused by Enterococcus avium
. Scand J Infect Dis
13. Verhaegen J, Pattyn P, Hinnekens P, et al. Isolation of Enterococcus avium
from bile and blood in a patient with acute cholecystitis. J Infect
14. Ablaza VJ, LaTrenta GS. Late infection of a breast prosthesis with Enterococcus avium
. Plast Reconstr Surg
. 1998 July;227-230.
15. Dealler SF, Grace RJ, Norfolk DR. Enterococcus avium
septicemia in an immunocompromised patient. Eur J Clin Microbiol Infect Dis
16. Mellman RL, Spisak GM, Burakoff R. Enterococcus avium
bacteremia in association with ulcerative colitis. Am J Gastroenterol
17. Farnsworth TA. Enterococcus avium
splenic abscess: a rare bird. Lancet Infect Dis
18. Suzuki A, Matsunaga T, Aoki S, et al. A pancreatic abscess 7 years after a pancreatojejunostomy for calcifying chronic pancreatitis. J Gastroenterol
19. Cottagnoud P, Rossi M. Enterococcus avium
osteomyelitis. Clin Microbiol Infect
20. Dolin GL, Bennett J, Mandell R. Principles and practice of infectious diseases. In: Bennett J, Mandell GL, Dolin R, eds. Philadelphia: Churchill Livingstone; 2000:2147-2152.
21. Xia L, Murray B. Trends in treatment of antimicrobial-resistant enterococcal infections. Curr Clin Top Infect Dis
22. Mackowiak PA. The enterococci: evidence of species-specific clinical and microbiologic heterogeneity. Am J Med Sci
23. Gold HS. Vancomycin-resistant enterococci: mechanisms and clinical observations. Clin Infect Dis
24. Pai MP, Rodvold KA, Schreckenberger PC, et al. Risk factors associated with the development of infection with linezolid- and vancomycin-resistant Enterococcus faecium
. Clin Infect Dis
25. Wilke WW, Marshall SA, Coffman SL, et al. Vancomycin-resistant Enterococcus raffinosus
: molecular epidemiology, species identification error, and frequency of occurrence in a national resistance surveillance program. Diagn Microbiol Infect Dis
26. Sifaoui F, Gutmann L. Vancomycin dependence in a VanA-producing Enterococcus avium
strain with a nonsense mutation in the natural d-Ala-d-Ala ligase gene. Antimicrob Agents Chemother
27. Eliopoulos GM. Quinupristin-dalfopristin and linezolid: evidence and opinion. Clin Infect Dis