Streptococcus bovis group and biliary tract infections: an analysis of 51 cases

Transcription

1 ORIGINAL ARTICLE BACTERIOLOGY Streptococcus bovis group and biliary tract infections: an analysis of 51 cases J. Corredoira 1, M. P. Alonso 2, F. Garcıa-Garrote 2, M. J. Garcıa-Pais 1, A. Coira 2, R. Rabu~nal 3, A. Gonzalez-Ramirez 4, J. Pita 2, M. Matesanz 3, D. Velasco 2,M.J.Lopez- Alvarez 1 and J. Varela 1 1) Infectious Disease Unit, 2) Department of Clinical Microbiology, 3) Department of Internal Medicine and 4) Gastroenterology Unit, Hospital Lucus Augusti, Lugo, Spain Abstract Streptococcus bovis is a well-known cause of endocarditis, but its role in other infections has not been well described. We analysed prospectively all patients with biliary tract infections caused by S. bovis group during the period We selected those cases associated with cholangitis and cholecystitis, defined according to Tokyo guidelines. Identification of the strains was performed using the API 20 Strep and the GP card of the Vitek 2 system, and was confirmed by molecular methods. Our series included 51 cases (30 cholangitis and 21 cholecystitis). The associated microorganisms were: Streptococcus infantarius (biotype II/1) 29 cases (57%), Streptococcus gallolyticus subsp. pasteurianus (biotype II/2) 20 cases (39%) and Streptococcus gallolyticus subsp. gallolyticus (biotype I) two cases (4%). The only difference found between S. infantarius and S. gallolyticus subsp. pasteurianus was a greater association of the first with malignant strictures of the bile ducts: 48% (14/29) versus 5% (1/20), p < Thirty-seven of the cases also had bacteraemia, causing 20% (37/185) of all S. bovis bacteraemia, with differences between S. gallolyticus subsp. gallolyticus (2/112; 2%) and the other two microorganisms: S. infantarius and S. gallolyticus subsp. pasteurianus (35/73; 48%; p <0.001). The vast majority of biliary tract infections due to S. bovis group are caused by S. infantarius and S. gallolyticus subsp. pasteurianus (S. bovis biotype II), and nearly half of the bacteraemia due to these two species has a biliary source (43% of the S. infantarius and 56% of S. gallolyticus subsp. pasteurianus). Keywords: Biliary tract infection, Streptococcus bovis, Streptococcus gallolyticus, Streptococcus infantarius, Streptococcus pasteurianus Original Submission: 12 May 2013; Revised Submission: 10 July 2013; Accepted: 11 July 2013 Editor F. Allerberger Article published online: 30 August 2013 Clin Microbiol Infect 2014; 20: / Corresponding author: Dr J. Corredoira, Infectious Disease Unit, Hospital Lucus Augusti, San Cibrao s/n, Lugo, Spain juan.corredoira.sanchez@sergas.es Introduction The Streptococcus bovis group (SBG) includes several species and subspecies of microorganisms that can cause humans infections. This group was previously classified in two biotypes, I and II (this latest divided in two subtypes (II/1 and II/2); the current classification comprises different species: Streptococcus gallolyticus subsp. gallolyticus (formerly Streptococcus bovis biotype I), S. gallolyticus subsp. pasteurianus (formerly S. bovis biotype II/2) and Streptococcus infantarius, subsp. coli and subsp. infantarius (formerly S. bovis biotype II/1). The SBG has been primarily associated with endocarditis and colorectal neoplasia [1,2], This association with colorectal neoplasms is higher than in controls [3], and this association is stronger with advanced neoplasms [4], especially with S. gallolyticus subsp. gallolyticus [5]. The SBG has also been linked to osteoarticular infections, urinary tract infections, meningitis, peritonitis and spontaneous bacteraemia in cirrhotic, neonatal infections and non-colorectal cancer [6 11]. Over 20 years ago, Ruoff et al. [12] reported the association between S. bovis biotype II and biliary tract infections. Since then, this association has received little attention [13 16]; the percentage of SBG bacteraemia with biliary origin has been estimated in several reports at between 0 and 38% [17 27]. However, the features associated with this infection, risk factors, underlying Clinical Microbiology and Infection ª2013 European Society of Clinical Microbiology and Infectious Diseases

2 406 Clinical Microbiology and Infection, Volume 20 Number 5, May 2014 CMI diseases, and the relationship with the new taxonomic species of SBG has not been well determined. For this reason, the aim of this study is to describe clinical, microbiological features and risk factors of the biliary tract infections associated with SBG in our institution. Materials and Methods diagnosed were: 30 cholangitis and 21 cholecystitis (Table 1). The causes of cholangitis were: malignant biliary stricture in 17 cases [pancreatic cancer (eight cases), cholangiocarcinoma (four cases), ampulloma (two cases), gastric cancer (two cases), duodenal cancer (one case)]; choledocholithiasis in nine cases, and benign strictures or fistulas caused by previous surgery in four cases. The causes of the cholecystitis in all 21 patients was cholelithiasis. The study was performed at the Hospital Lucus Augusti (formerly Hospital Xeral-Calde, Lugo), a large community teaching hospital with 740 beds, serving a mixed urban and rural area of inhabitants, and is the reference centre for two regional hospitals that provide health care to an area of inhabitants. Between 1 January 1988 and 31 December 2011, all adult patients with positive blood cultures were followed prospectively. During this period we detected bacteraemias, of which, 185 were caused by SBG (1.6%). Of these 185 cases, 112 were identified as S. gallolyticus subsp. gallolyticus,27ass. gallolyticus subsp. pasteurianus and 46 as S. infantarius, subsp. coli and subsp. infantarius. Furthermore, since 2003 we have also followed all patients with SBG isolated in any other type of sample other than haemoculture. We isolated SBG in bile in 17 patients, which was 3.4% of the microorganisms isolated in bile (495 cases). All patients with biliary tract infection and SBG isolated in blood cultures or bile were included in the study. The diagnosis of acute cholangitis and acute cholecystitis was made, according to Tokyo guidelines [28,29], on the basis of the clinical findings in combination with laboratory data and imaging findings. Isolates were stored in skimmed milk Difco TM (Becton-Dickinson, Sparks, MD, USA) at 70 C. Identification of the strains was performed using the API 20 Strep gallery and card Vitek 2 system GP (both from biomerieux, Marcy l Etoile, France). For molecular identification, the complete 16S rrna gene sequence was determined as previously described by Beck et al. [23]. Antimicrobial susceptibility testing was performed using the disc diffusion technique according to CLSI guidelines [30] and the E-test method (AB Biodisk, Solna, Sweden), following the manufacturer s recommendations. Results General characteristics of the series During the study period, SBG was isolated in 51 patients diagnosed with biliary tract infection according to Tokyo guidelines. The mean age of the patients was 73.6 years (range years), and 65% were male. The biliary tract infections Clinical manifestations and complications The most common clinical manifestations were fever (84%), abdominal pain (63%) and jaundice (29%). Seventy-nine percent of patients presented with leucocytosis. Major complications were: abdominal or liver abscesses (11 cases), pancreatitis (five cases), surgical wound infections (four cases), gallbladder perforation with peritonitis (three cases) and septic shock (three cases). The most frequent associated co-morbidities were: cancer (22 cases, 17 of them digestive non-colorectal, two prostate, two breast, one leg sarcoma), diabetes mellitus (20 cases), ischaemic heart disease (11 cases) and chronic obstructive pulmonary disease (nine cases). Colorectal adenomas were detected in four patients. Five patients with biliary infection by SBG died (10%): two of them from sepsis and the other three due to their underlying disease (pancreatic cancer). Treatment Empirical antimicrobial treatment was appropriate in 87% of patients. All isolates were susceptible to penicillin, ceftriaxone and vancomycin, and 69% and 75% of them were susceptible to clindamycin and levofloxacin, respectively. Surgery was performed in 15 cases (29%); any other drainage was carried out in 17 cases (33%), and a biliary stent was placed or replaced in five cases (10%). Microbiological characteristics Streptococcus bovis group organisms were isolated in 51 patients with biliary infection: 34 from blood cultures, 14 from bile and three in both samples. The SBG accounted for 3.3% of biliary infections documented microbiologically. SBG was isolated from polymicrobial infections in 30 patients (59%): in 18 cases with two microorganisms and in 12 cases with three or more microorganisms. SBG was isolated with other microorganisms in 71% (10/14) of bile samples and in 54% (20/37) of haemocultures. Most frequently associated microorganisms were: Escherichia coli (24 cases) and Enterococccus spp. (11 cases). The only statistically significant difference between the polymicrobial biliary infections due to SBG compared with SBG monomicrobial biliary infections was the lower percentage of

3 CMI Corredoira et al. Streptococcus bovis and biliary tract infections 407 TABLE 1. Characteristics of the biliary tract infections caused by Streptococcus bovis group Cholecystitis (n = 21) (%) Cholangitis (n = 30) (%) p Total (n = 51) (%) Age (mean in years) NS 73.6 Sex (male) 14 (67) 19 (63) NS 33 (65) Underlying diseases Diabetes mellitus 6 (29) 14 (47) NS 20 (39) Cancer 3 (14) 19 (63) (43) Ischaemic heart disease 4 (19) 7 (23) NS 11 (22) Chronic obstructive pulmonary disease 5 (24) 4 (13) NS 9 (18) Biliary disease Benign Lithiasis 21 (100) 9 (30) (59) Other 0 4 (13) NS 4 (8) Malignant Biliary pancreatic 0 14 (47) (27) Gastroduodenal 0 3 (10) NS 3 (6) Nosocomial infections 3 (14) 7 (23) NS 10 (20) Type of infection Monomicrobial 8 (38) 14 (47) NS 22 (43) Polymicrobial 13 (62) 16 (53) NS 29 (57) Clinical manifestations Fever 15 (71) 28 (93) NS 43 (84) Abdominal pain 20 (95) 12 (40) (63) Jaundice 3 (14) 12 (40) NS 15 (29) Complications Gangrenous cholecystitis 6 (29) (12) Abscess 9 (43) 2 (7) (22) Peritonitis 3 (14) 0 NS 3 (6) Shock 3 (14) 0 NS 3 (6) Pancreatitis 3 (14) 2 (7) NS 5 (10) Surgical wound infection 4 (19) (8) Mortality 1 (5) 4 (13) NS 5 (10) Isolation site Haemocultures 6 (29) 28 (93) (67) Bile 12 (57) 2 (7) (27) Haemocultures and bile 3 (14) 0 NS 3 (6) Microorganism S. gallolyticus subsp. gallolyticus 0 2 (7) NS 2 (4) S. infantarius 12 (57) 17 (57) NS 29 (57) S. gallolyticus subsp pasteurianus 9 (43) 11 (37) NS 20 (39) abscesses found in monomicrobial biliary infections (1/21: 5% versus 10/30: 33%; p <0.01). After molecular identification, the isolates were taxonomically classified as follows: 29 as S. infantarius (57%), 20 as S. gallolyticus subsp. pasteurianus (39%) and two as S. gallolyticus subsp. gallolyticus (4%). When comparing the clinical characteristics of the infections caused by S. infantarius with those caused by S. gallolyticus subsp. pasteurianus, the only difference found was the stronger association of the former with malignant strictures of the bile duct (14 of 29, 48% versus 1 of 20, 5%, p <0.001). Biliary bacteraemia caused by S. bovis group During the period there were 1109 cases of bacteraemias of biliary origin in our hospital. Of these, 37 were caused by SBG (3.3%). We compared 1072 biliary bacteraemias caused by microorganisms other than SBG with the 37 cases of SBG bacteraemias of biliary origin, to analyse the risk factors associated with these infections. The demographic characteristics, clinical manifestation, complications, co-morbidity and mortality were not different between the two groups, except for a higher percentage of polymicrobial bacteraemia in the SBG with respect to the general group (51% (19/37) vs 20% (219/1072); p <0.0001) and a higher association with malignant pathology of the biliary tract in the subgroup of patients with bacteraemia caused by S. infantarius. Hence, biliary bacteraemia in our hospital was associated with neoplasia in 279 of 1091 cases (26%) and S. infantarius bacteraemia was associated with 13 of 20 cases (65%, p <0.0001). The 37 cases of biliary bacteraemia caused by SBG accounted for 20% (37/185) of all S. bovis bacteraemia. We compared 148 non-biliary SBG bacteraemia with the 37 cases of SBG bacteraemia with biliary origin and the main differences between the two groups are presented in Table 2. In all the bacteraemia due to SBG, when we analysed the percentage with a biliary source, we found significant differences among the three subspecies. Only two (2%) of the 112 bacteraemias caused by S. gallolyticus subsp. gallolyticus were of biliary origin, whereas higher frequencies were found both in S. infantarius, 20 of 46 cases (43%), and S. gallolyticus subsp. pasteurianus, 15 of 27 cases (56%); p < in both cases. Seventeen of the 37 patients with SBG bacteraemia (one with S. gallolyticus subsp. gallolyticus, four with S. infantarius and 12 with S. gallolyticus subsp. pasteurianus) underwent colonoscopy. In the other 20 patients a colonoscopy was not performed either because of non-colorectal cancer (16 cases), or because the patient was dead or had a bad prognosis from co-morbidities (four cases). Colorectal adenomas were

4 408 Clinical Microbiology and Infection, Volume 20 Number 5, May 2014 CMI TABLE 2. Differences between biliary bacteraemia and non-biliary bacteraemia caused by Streptococcus bovis group detected in four patients (one with S. gallolyticus subsp. gallolyticus, one with S. infantarius and two with S. gallolyticus subsp. pasteurianus). Discussion Biliary SBG bacteraemia (n = 37) (%) Non-biliary SBG bacteraemia (n = 148) (%) p Age (mean in years) Sex (male) 24 (65) 117 (79) 0.07 Co-morbidities Diabetes mellitus 18 (49) 23 (15.5) Digestive non-colorectal 17 (46) 5 (3) cancer Colorectal neoplasia 4 (11) 89 (60) Neutropenia 2 (5) 7 (5) NS Inmunosuppression 1 (3) 10 (7) NS Nosocomial infections 5 (13) 8 (5) 0.08 Polymicrobial bacteraemia 19 (51) 22 (15) Microorganism S. gallolyticus subsp. gallolyticus 2 (5) 110 (74) S. infantarius 20 (54) 26 (18) S. gallolyticus subsp. 15 (41) 12 (8) pasteurianus Shock 2 (5) 7 (5) NS Mortality 5 (13) 17 (11) NS There are few reported cases of biliary tract infection caused by SBG [13 16]. A possible explanation is that in previous publications about a-haemolytic streptococci, the bacteria were rarely identified to the species level in intrabdominal samples and, in studies about group D streptococci, both enterococci and SBG were often grouped together [31 35]. However, even when performing an accurate identification, the percentage of SBG in biliary tract infections is <1% [36]. In the few reports that analysed SBG in samples other than blood cultures, the isolations in bile were scarce, ranging from 0 to 2% [37,38]. We have identified all SBG isolates in different clinical specimens over the study period and, among the patients with biliary tract infections, SBG represents a small percentage of the isolates from blood cultures and bile. However, in our institution biliary origin is the second commonest cause of SBG bacteraemia after endocarditis. This could be explained because SBG is able to grow in bile [39,40], in contrast to most of the a-haemolytic streptococci. If we analyse only S. infantarius and S. gallolyticus subsp. pasteurianus (formerly S. bovis biotype II) bacteraemia, 48% are of biliary origin; however, bacteraemia caused by S. gallolyticus subsp. gallolyticus is very rarely of biliary origin [5,12,21,24,27]. The different association of SBG species with biliary tract infections could explain the differences found in several studies of SBG bacteraemia, in which the bacteraemia of biliary origin ranges between 0 and 38% depending on the percentage of cases involving S. gallolyticus subsp. gallolyticus or the other species [12,20 27]. Clinical features and complications of these infections do not appear to be different from those caused by other microorganisms that are frequently isolated in bile, we have not found differences between SBG monomicrobial and polymicrobial infections (Table 1), except in the lower tendency of SBG to cause abscesses. In fact, there are few reports of abscesses caused by SBG [41,42]. Little is known about the characteristics of biliary infections caused by SBG. To our knowledge there is only one article reported from Hong Kong that analysed this subject [20]; in this report the authors comment that the association of biliary tract disease with S. bovis bacteraemia was due to a high incidence of biliary sepsis in their geographical area due to the endemic parasitism of the biliary tract by Clonorchis sinensis; in their report most of the isolates were S. gallolyticus subsp. pasteurianus. In our patients, gallstones and malignant strictures of the biliary tract were the most frequent predisposing factors, with strong association with S. infantarius and S. gallolyticus subsp. pasteurianus. We also found a strong association between S. infantarius and malignant neoplasm of the bile duct and pancreas, as has been referred to in a previous report [11]. We conclude that SBG is a rare cause of biliary tract infection, although if we consider all infections caused by this group, the biliary source is one of the most frequent, with a different percentage depending on the species. Streptococcus gallolyticus subsp. gallolyticus bacteraemia rarely has a biliary source, whereas almost half of the former S. bovis biotype II bacteraemia (both S. infantarius and S. gallolyticus subsp. pasteurianus) have this origin. Streptococcus infantarius seems to be associated with biliary pancreatic cancer whereas S. gallolyticus subsp. pasteurianus is associated with benign biliary tract disease. Transparency Declaration The authors have nothing to disclose. References 1. Klein R, Catalano M, Edberg S, Casey J, Steigbigel N. Streptococcus bovis septicemia and carcinoma of the colon. Ann Intern Med 1979; 91: Beeching N, Christmas T, Ellis-Pegler R, Nicholson G. Streptococcus bovis bacteremia requires rigorous exclusion of colonic neoplasia and endocarditis. Q J Med 1985; 56:

5 CMI Corredoira et al. Streptococcus bovis and biliary tract infections Hoen B, Briacon S, Delahaye F et al. Tumors of the colon increase the risk of developing Streptococcus bovis endocarditis: case control study. Clin Infect Dis 1994; 19: Sharara A, Abou Havolan T, Melli A et al. Association of Streptococcus bovis endocarditis and advanced colorectal neoplasia. A case control study. J Dig Dis 2013; 14: Corredoira J, Garcıa-Garrote F, Rabu~nal R et al. Association between bacteremia due to Streptococcus gallolyticus subsp. gallolyticus (S. bovis I) and colorectal neoplasia: a case control study. Clin Infect Dis 2012; 55: Emerton M, Crook D, Cooke P. Streptococcus bovis-infected total hip arthroplasty. J Arthroplasty 1995; 10: Roncoroni A, Bianchi L, Garcia Damiano C et al. Identification of Streptococcus viridans in clinical specimens. Rev Argent Microbiol 1987; 19: Villarrasa N, Prats A, Pujol M, Gason A, Viladrich P. Streptococcus bovis meningitis in a healthy adult patient. Scand J Infect Dis 2002; 34: Vilachoine R, Mahachai V, Kullavanijaya P, Nunthapisud P. Spontaneous bacterial peritonitis caused by Streptococcus bovis: case series and review of the literature. Am J Gastroenterol 2002; 97: Floret N, Bailly P, Thouverez N et al. A cluster of bloodstream infections caused by Streptococcus gallolyticus subspecies pasteurianus that involved 5 preterm neonates in a university hospital during a 2-month period. Infect Control Hosp Epidemiol 2010; 31: Corredoira J, Alonso P, Coira A et al. Association between Streptococcus infantarius (formerly S. bovis II/1) bacteraemia and noncolonic cancer. J Clin Microbiol 2008; 46: Ruoff K, Miller S, Garner C, Ferraro M, Calderwood S. Bacteremia with Streptococcus bovis and Streptococcus salivarius: clinical correlates of more accurate identification of isolates. J Clin Microbiol 1989; 27: Sams R, Haltiner A, Botse-Baidoo E, Coomer C, Moosariparambil M, Catto B. First report of Streptococcus bovis-associated acute cholecystitis in North America. J Clin Gastroenterol 2008; 42: Landau D, Blendis L, Lurie Y. Streptococcus bovis bacteremia associated with acute cholecystitis. J Clin Gastroenterol 2006; 40: Medina L, Mora L, Garcıa V, Santos J. Acute cholecystitis and bacteremia due to Streptococcus bovis biotype II. Enferm Infecc Microbiol Clin 2011; 29: Deering EM, Muravec Z, Castineira CF, O Donoghue G. Streptococcus bovis-related cholecystitis. BMJ Case Rep 2013; doi: /bcr Reynolds J, Silva E, McCormack W. Association of Streptococcus bovis bacteremia with bowel disease. J Clin Microbiol 1983; 17: Pigrau C, Loente A, Pahissa A, Martinez-Vazquez J. Streptococcus bovis bacteremia and digestive system neoplasms. Scand J Infect Dis 1988; 20: Clardrige J, Attori S, Zhang Q, Bartell J. 16 S ribosomal DNA sequence analysis distinguishes biotypes of Streptococcus bovis: Streptococcus bovis biotype II/2 is a separate genospecies and the predominant clinical isolate in adult males. J Clin Microbiol 2001; 39: Lee R, Woo P, To A, Lau S, Wong S, Yuen K. Geographical difference of disease association in Streptococcus bovis bacteraemia. J Med Microbiol 2003; 52: Jean S, Teng L, Hsueh P, Ho S, Luh K. Bacteremic Streptococcus bovis infections at a university hospital, J Formos Med Assoc 2004; 103: Corredoira J, Alonso P, Garcıa J et al. Clinical characteristics and significance of Streptococcus salivarius bacteremia and Streptococcus bovis bacteremia: a prospective 16-year study. Eur J Clin Microbiol Infect Dis 2005; 24: Beck M, Frodl R, Funke G. Comprehensive study of strains previously designated Streptococcus bovis consecutively isolated from human blood cultures and emended description of Streptococcus gallolyticus and Streptococcus infantarius subsp. coli. J Clin Microbiol 2008; 46: Vaska V, Faogagali J. Streptococcus bovis bacteraemia: identification within organism complex and association with endocarditis and colonic malignancy. Pathology 2009; 41: Fernandez-Ruiz M, Vilar J, Llenas J et al. Streptococcus bovis bacteraemia revisited: clinical and microbiological correlates in a contemporary series of 59 patients. J Infect 2010; 61: Romero B, Morosini M, Loza E et al. Reidentification of Streptococcus bovis isolates causing bacteremia according to the new taxonomy criteria: still an issue? J Clin Microbiol 2011; 49: Gomez-Garces J, Gil Y, Burillo A, Wilhelmi I, Palomo M. Diseases associated with bloodstream infections caused by the new species included in the old Streptococcus bovis group. Enferm Infecc Microbiol Clin 2012; 30: Wada K, Takada T, Kawarada Y et al. Diagnostic criteria and severity assessment of acute cholangitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg 2007; 14: Hirota M, Takada T, Kawarada Y et al. Diagnostic criteria and severity assessment of acute cholecystitis: Tokyo Guidelines. J Hepatobiliary Pancreat Surg 2007; 14: Clinical and Laboratory Standards Institute. Performance standards for antimicrobial susceptibility testing: 17th informational supplement. Document M100-S17, Wayne, PA: CLSI, Suzuki Y, Kobayashi A, Ohto M et al. Bacteriological study of the transhepatically aspirated bile. Relation to cholangiographic findings in 295 patients. Dig Dis Sci 1984; 29: Hanau L, Seitbigel N. Acute (ascending) cholangitis. Infect Dis Clin North Am 2000; 14: Reiss R, Eliashiv A, Deutsch A. Septic complications and bile cultures in 800 consecutive cholecystitis. World J Surg 1982; 6: Pitt H, Postier R, Cameron J. Biliary bacteria. Significance after antibiotic therapy. Arch Surg 1982; 117: Brook I. Aerobic and anaerobic microbiology of biliary tract disease. J Clin Microbiol 1989; 27: Claesson B, Holmlund D, M atzsch T. Microflora of the gallbladder related to duration of acute cholecystitis. Surg Gynecol Obstet 1986; 162: Facklan R. Recognition of group D streptococcal species of human origin by biochemical and physiological tests. Appl Microbiol 1972; 23: Uh Y, Kwon O, Yoon K et al. Underlying diseases associated with Streptococcus bovis bacteremia and antimicrobial susceptibility of the organism. Korean J Clin Microbiol 2006; 9: Chuard C, Reller B. Bile esculin test for presumptive identification of enterococci and streptococci: effects of the bile concentration, inoculation technique, and incubation time. J Clin Microbiol 1998; 36: Rusniok C, Couve E, Da Cunha V et al. Genoma sequence of Streptococcus gallolyticus: insights into its adaptation to the bovine rumen and its ability to cause endocarditis. J Bacteriol 2010; 193: Steerman P, Steerman S. Unretrieved gallstones presenting as a Streptococcus bovis liver abscess. JSLS 2000; 4: Montejo M, Aguirrebengoa K, Testillano M, Aguirre C. Brain abscess caused by Streptococcus bovis. Eur J Clin Microbiol 1985; 4: