Streptococcus bovis Group Bacteremia in the 21st Century: Review of 42 Episodes Over a 12-Year Period (2006–2017) at a Large Community Teaching Hospital

Introduction Advanced phenotypic, genomic, and proteomic laboratory techniques have recently modified Streptococcus bovis group (SBG) nomenclature. We wished to determine if physicians continue to recognize the importance of SBG and its association with gastrointestinal (GI) tract abnormalities and infective endocarditis amid the changes in microbiologic identification and nomenclature of these organisms. Methods We reviewed the medical records of adult patients (≥18 years of age) with positive blood cultures for SBG organisms admitted to our 510-bed teaching hospital from January 1, 2006, to December 31, 2017. We report the epidemiology, sources of bacteremia, comorbid conditions, courses of treatment, and the mortality for these patients. We also assess the hospital treatment team's (HTT's) knowledge of SBG nomenclature and of the associations of SBG bacteremia and underlying GI disease and infective endocarditis amid the changes in nomenclature of these organisms. Results There were 42 cases of SBG bacteremia during the 12-year study period: 22 in women (52.4%) and 20 in men (47.6%). Patient ages ranged from 51 to 96 years (mean age, 74.3 years; median age, 72.0 years). All but 2 patients had multiple comorbid conditions. Diabetes mellitus was the most common comorbidity. Colonoscopy was performed during hospitalization in 22 (52.5%) of 42 patients. The identifiable sources of bacteremia were as follows: lower GI tract in 19 patients (45.2%), upper GI tract in 5 patients (11.9%), Laennec cirrhosis in 3 patients (7.1%), and pancreatic disorders in 2 patients (4.6%). Eleven patients (26.2%) had primary bacteremia. Two patients with primary bacteremia had prior splenectomy. The historic association between SBG bacteremia and underlying GI tract disease was recognized by 37 (88.1%) of 42 HTTs, but all available provider progress notes mention only "colon carcinoma" as the possibly associated GI tract pathology. The historic association of SBG bacteremia with infective endocarditis was recognized in writing by 32 (76.2%) of 42 HTTs. Endocarditis was diagnosed in 12 patients (28.6%): 9 definite endocarditis and 3 possible endocarditis. The mitral valve was the most commonly involved valve. Four SBG isolates were intermediately susceptible to penicillin G with minimum inhibitory concentrations of 0.125 μg/mL or greater. Twenty-three (54.8%) of 42 SBG strains were resistant or intermediately susceptible to clindamycin. Twenty-four (57.1%) of 42 strains were resistant or intermediately susceptible to erythromycin. All strains were tested for susceptibility to ceftriaxone and vancomycin and retained susceptibility to both antimicrobial agents throughout the study period. Six of 42 patients died, for a mortality rate of 11.9%. Infectious disease consultation was obtained in 35 (80.0%) of 42 patients. Infectious disease consultation was positively associated with survival (P = 0.0041, Fisher exact test). The new nomenclature schemes for prior members of the SBG were recognized by all HTTs because our microbiology laboratory reported each member of the group, regardless of new name, with "bovis group" added to the identification on all positive culture reports. Conclusions Streptococcus bovis group bacteremia is a disease of older adults with all but 3 patients 60 years or older and a mean age at onset of 73.4 years. Most HTTs considered colon carcinoma as a possible source for and infective endocarditis as a potential complication of SBG bacteremia. However, most HTTs were not aware that SBG bacteremia could be associated with nonmalignant colonic lesions especially polyps, Laennec cirrhosis, or with pancreatic, biliary, and upper GI tract anatomic abnormalities. Of our SBG isolates, 54.8% were not sensitive to clindamycin. Clindamycin should not be used for empiric treatment of SBG bacteremia. The ID service should be consulted on all patients with SBG bacteremia because such consultation had a positive correlation with patient survival (P = 0.0041). Correspondence to: Joseph P. Myers, MD, Department of Medicine, Summa Akron City Hospital, 55 Arch St, Suite 1A, Akron, OH 44304. E-mail: myersj@summahealth.org. The authors have no funding or conflicts of interest to disclose. Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.

https://ift.tt/2CJ5BG5