Group B Streptococcus Bacteremia in Nonpregnant Adults

Publisher Website
Google Scholar
Abstract
Background: We report the largest series of group B streptococcal (GBS) bacteremia cases reported at a single institution. Methods: During a 10-year period (1985-1994), 90 GBS bacteremia cases (0.95% of significant bacteremic episodes) were detected. We describe the 51 episodes that occurred in nonpregnant adults for which enough clinical and microbiological information is available. Results: Incidence of GBS has significantly increased during the study period (from 0.08 per 1000 admissions in 1985 to 0.3 per 1000 in 1994). Mean age of patients was 63.3 years (range, 21-88 years) and 53% were men. The most common underlying conditions were liver diseases (35.3%), malignancies (33.3%), and diabetes mellitus (27.5%). Only 2 patients did not have any underlying condition and no patient with the human immunodeficiency virus had GBS bacteremia in our series. The origins of the episodes of bacteremia were as follows: primary bacteremia (39.2%), skin and soft tissue infections (15.7%), urinary tract infections (11.8%), pneumonia (9.8%), peritonitis (9.8%), catheter infection (5.9%), postendoscopic bacteremia (5.9%), and endocarditis (2%). All isolates were susceptible to penicillin G potassium, ampicillin sodium, cephalothin sodium, cefotaxime sodium, and vancomycin hydrochloride. One ciprofloxacin hydrochloride—resistant strain was discovered and resistance to erythromycin stearate increased from 8% in 1992 to 18% in 1994. The overall mortality rate was 33.3% and deaths were considered related to the GBS bacteremia in 25.5% of the cases. Factors for poor prognosis were central nervous system diseases, alcoholism, shock, renal failure, and consciousness impairment. Conclusions: Group B streptococcus is a rising cause of bacteremia in elderly patients with severe underlying conditions. It conveys high morbidity and mortality rates. Macrolides should not be used empirically for treatment of patients with penicillin allergies. Arch Intern Med. 1997;157:213-216