Treatment should include removing the nidus if present, finding and administering the appropriate antimicrobial therapy, evaluating for possible complications, and consulting with ID
by Andrea H. Kim, MD, Jennifer N. Stojan, MD, and Denege A. Ward, MD
An 82-year-old man with non-Hodgkin’s lymphoma in remission and a history of congestive heart failure and hypertension presents with one week of generalized malaise and intermittent fevers. Vitals show a temperature of 101oF, blood pressure of 130/60 mmHg, and heart rate of 100. His exam is notable for an erythematous and tender chest port site, with no murmurs. Blood cultures drawn upon presentation show gram-positive cocci speciated to Staphylococcus aureus. What are the next steps in management of this patient?
S. aureus bacteremia (SAB) is a common infectious cause of morbidity and mortality worldwide, causing both community-acquired and hospital-acquired bacteremia. In the U.S. alone, it accounts for 23% of all bloodstream infections and is the bacterial pathogen most strongly associated with death.1 Mortality rates are approximately 42% in those with methicillin-resistant S. aureus (MRSA) bacteremia and 28% in those with methicillin-sensitive S. aureus (MSSA) bacteremia.2
Recognizing the severity of SAB, the Infectious Disease Society of America (IDSA) published treatment guidelines in 2011 to help direct the clinical care of this disease process.3 However, the majority of the recommendations are based on observational studies and expert opinion, as less than 1,500 patients have been enrolled in randomized controlled trials specifically targeted to investigate the treatment of SAB.4
A clinically significant SAB usually is defined as the isolation of S. aureus from a venous blood culture with associated symptoms and signs of systemic infection.5 As SAB contamination is rare and can be associated with multiple complications, including metastatic infections, embolic stroke, recurrent infection, and death, any finding of a positive blood culture must be taken seriously.4
SAB treatment is multifaceted and should focus on the removal of any nidus of infection, such as a catheter or a prosthetic device, the use of prolonged antimicrobial therapy, and the evaluation of potential complications. In a retrospective study, Johnson et al showed that failure to remove the source is one of the strongest independent predictors of relapse in patients with SAB.6 However, 10% to 40% of patients have no identifiable focus, which increases the impetus to evaluate for complications.7-8 Overall, approximately one-third of patients with SAB develop metastatic complications, either from hematogenous seeding or local extension of infection.9
In addition to advanced age and such comorbid conditions as cirrhosis, the strongest predictor of complications is a positive blood culture at 48 to 96 hours after an initial positive blood culture, as shown in a large prospective cohort study by Fowler et al.7,10-11 Additional independent risk factors (see Table 1) include community acquisition (likely due to prolonged duration of bacteremia), skin examination suggesting the presence of acute systemic infection, and persistent fever at 72 hours after the first positive blood culture. Patients with even one of these risk factors are at high risk for a complicated course (which occurs in about 35%). In a case-control study, Chihara et al showed that S. aureus bacteruria in the absence of urinary tract pathology or recent urinary tract instrumentation might be associated with threefold increased mortality compared with those without bacteriuria, even after adjustment for comorbid conditions.12
The initial choice of antibiotic therapy for SAB must take into account the MRSA prevalence in the community and hospital. If suspicion is high enough for MRSA, the IDSA’s 2011 guidelines suggest treatment with vancomycin or daptomycin.3 Although there are no published RCTs to support a particular antibiotic regimen, there are trials to suggest that a delay in treatment could be harmful. One study, by Lordis et al, showed that a delay in treatment, as defined by treatment after 44.75 hours, was associated with a longer hospital stay, with the delayed treatment group being hospitalized for 20.2 days and the early treatment group being hospitalized for 14.3 days.13 A delay in treatment was also found to be an independent predictor of mortality.13
Once susceptibilities are known, it is important to appropriately tailor antibiotics, as studies have shown lower treatment failure rates with the use of beta-lactam antibiotics when compared with empiric MRSA coverage.14-15 In one prospective study of 123 hemodialysis patients with MSSA bacteremia, Stryjewski et al showed that those treated with vancomycin were at higher risk of experiencing treatment failure than those treated with cefazolin.15 In another prospective observational study of 505 patients with SAB, Chang et al found that treatment with nafcillin was superior to vancomycin in preventing persistent bacteremia or relapse for MSSA bacteremia.14 These studies highlight the benefits of adjusting the empirically selected antibiotics, as narrowing the spectrum can result in less treatment failure.
If susceptibilities confirm MRSA, the IDSA recommends continued treatment with vancomycin or daptomycin.3 Although vancomycin is most commonly used, partly because of low cost and familiarity, Fowler et al published a study of 246 patients with SAB with or without endocarditis, assigning them to treatment with daptomycin, initial low-dose gentamicin plus vancomycin or an antistaphylococcal penicillin.16 The study found that daptomycin was not inferior to the other therapies, confirming that daptomycin is a reasonable choice in the treatment of MRSA infections.
Oral antibiotics are an option to treat SAB when necessary. A RCT by Heldman et al of 85 intravenous drug users with SAB (and suspected right-sided endocarditis, 65% of which had HIV) showed similar efficacy of ciprofloxacin plus rifampin versus standard intravenous therapy.17 A subsequent randomized trial of 104 patients with SAB comparing oral fleroxacin plus rifampin against conventional intravenous therapy also showed similar cure rates, with the added benefit of earlier discharge.18 Furthermore, in a meta-analysis of five randomized studies by Shorr et al (see Table 2), linezolid was found to have outcomes that were not inferior to vancomycin (clinical cure/microbiological success of 56%/69% in the linezolid group and 46%/73% in the vancomycin group).19
Recommendations for the duration of antibiotic treatment for SAB are mainly based on observational studies, which show mixed results. In one study done in the 1950s, about two-thirds of cases of SAB were associated with endocarditis, and longer courses of intravenous therapy (greater than four weeks) were recommended.20
More recently, with the increasing rates of catheter-related SAB and its relatively high rate of expeditious blood culture clearance, a shorter duration has been evaluated in several studies. In 1992, an analysis of published data and a retrospective case series concluded that fewer than 10 days of intravenous antibiotics might be associated with an increased risk of recurrence, but 10 to 14 days of intravenous therapy was effective for most cases of catheter-associated SAB.5 In another prospective study, Fowler et al found that a seven-day course of intravenous antibiotic therapy may be sufficient for simple, catheter-related infections.21 A subsequent prospective study by Jensen et al reported that a course of antibiotic therapy of less than 14 days might be associated with higher mortality compared to a longer course.9 A prospective study of 276 patients by Thomas et al found there was no relationship between relapse and duration of treatment (seven to 15 days) in catheter-related SAB, concluding that more than 14 days of antibiotic therapy was unnecessary.22
Per IDSA guidelines, uncomplicated SAB (no implanted prosthesis, negative blood cultures within two to four days, defervescence within 72 hours of initiating therapy, and lack of metastatic complication) can be treated with a two-week course of antibiotics, while complicated bacteremia (any of above criteria) should be treated within four to six weeks.3
Based on the IDSA guidelines, echocardiography is recommended in all patients with bacteremia, with a preference of transesophageal echocardiography (TEE) over transthoracic echocardiography (TTE).3 More recently, Kaasch et al developed simple criteria to identify patients with nosocomial SAB at low risk for infective endocarditis based on two prospective cohort studies.23 Lack of any of these criteria, which include prolonged bacteremia of more than four days’ duration, presence of a permanent intracardiac device, hemodialysis dependency, spinal infection and nonvertebral osteomyelitis, along with a negative TTE indicates that a TEE is not necessary (see Table 3). However, these patients need close follow-up to ensure that bacteremia clears and no new signs or symptoms concerning for metastatic infection develop.
Several studies have shown that ID consultation not only improves adherence to evidence-based management of SAB, but it also reduces mortality.24-27 In a recent prospective cohort study in a tertiary-care center, even after adjusting for pre-existing comorbidities and severity of disease, an ID consult was associated with a 56% reduction in 28-day mortality.24 The patients who were followed by an ID consult service were more likely to receive appropriate duration of antibiotics (81% vs. 29%, respectively) and undergo appropriate workup for the evaluation of metastatic infections (34% and 8%, respectively). This study concluded that routine ID consult should be considered in patients with SAB, especially those with severe illness and multiple comorbid conditions.
The patient was started on empiric therapy with vancomycin and serial blood cultures were obtained. He remained hemodynamically stable but febrile, with persistently positive blood and urine cultures. Given concern for the port being the source of his infection, his chest port was removed. A high-quality TTE was performed and was unremarkable.
ID was consulted. Blood cultures subsequently grew MSSA and vancomycin was switched to cefazolin 2g every eight hours. On hospital Day 5, his fever resolved and blood cultures turned negative. There were no clinical signs or symptoms for metastatic infections. A PICC line was placed after blood cultures remained negative for 48 hours. The decision was made to treat him with four weeks of antibiotics from his last positive blood culture, with follow-up in ID clinic.
SAB is a common worldwide cause of morbidity and mortality. Treatment should include removing the nidus if present, finding and administering the appropriate antimicrobial therapy, evaluating for possible complications, and consulting with ID.
Dr. Ward is an assistant professor, Dr. Kim a clinical instructor, and Dr. Stojan a clinical lecturer at the University of Michigan Health System in Ann Arbor.
The authors would like to thank Dr. Jeffrey Rohde for reviewing the manuscript.
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.