New IDSA/ESCMID S aureus Bacteremia Consensus Statements: a Risk Stratification Framework

The IDSA and ESCMID published the first set of consensus statements from their joint S aureus bacteremia (SAB) guideline project in September, addressing 7 clinical questions on risk stratification, diagnostic evaluation, and treatment duration.¹ The panel suggests a risk stratification framework in place of the traditional "complicated" or "uncomplicated" classification, citing the heterogeneity of SAB and its evolving clinical course.¹ Low-risk adults without deep-seated or metastatic foci should receive 14 days of therapy, and increased-risk adults may also qualify when a tailored evaluation is negative and symptoms resolve.²

Data remain insufficient to define a low-risk subgroup of children, and the panel states the framework itself requires validation.^2,3 Henry F. Chambers, MD, professor of medicine emeritus at the University of California San Francisco and a member of the guideline panel, has published extensively on staphylococcal disease, endocarditis, and antimicrobial resistance. Chambers explains how the SAB statements are akin to the stages of cancer and how they guide treatment associated with this bacteremia. For example, if you have stage 4 cancer, patients are treated with chemotherapy and have surgery to remove tumors. The guidelines are designed to provide clinicians with a pathway for treatment. In a conversation with Chambers, he explains why the panel moved away from the "complicated" label, how the framework plays out at the bedside, and which questions the next round of guidance must answer.

Contagion: The new guidelines move away from the traditional classification toward risk stratification. Why did the older model fall short, and what gaps did it miss?

Chambers: I don't know if it quite reached the designation of a model before. I think it was an approach or conceptualization of Staphylococcus bacteremia that focused more on complicated as a term rather than risk. So to get into the nuance of that difference, I can identify factors that would put a patient at risk, but their disease may or may not prove to be complicated, if that makes sense.

There was a lot of emphasis on complicated, which we felt is a retrospectively defined, post hoc conclusion. I may not know if you have complicated disease or not when you present, but I will make myself aware of risk factors that would predict such a complication, which you may or may not prove to have. Then I try to address the approach to diagnosis and the workup based upon how those risk factors pile up, if you will, which helps determine, at the end of the day, things like duration of therapy. Do I need additional studies? What is the role of imaging, for example?

So it's a risk stratification framework, and I think formalizing that is the major contribution of this first tranche of consensus statements as part of this guidance, if you see what I'm getting at.

Contagion: How does the framework work at the bedside, and how does it change follow-up blood cultures and testing in the first 48 to 72 hours?

Chambers: There's a figure we created, and we hope we accomplished the goal of showing how the flow of thinking should be. There's the initial encounter with the patient: an appropriate, complete history and a physical exam, certainly focusing on findings that are evident but may merit further investigation. Let's say, for example, that you show up and you have signs of a knee infection. Then you would tap the knee to get fluid for analysis, for example. And of course, you would get blood cultures, and your therapy would be driven by what those tests yielded.

It may be that, in addition to your knee being a problem, you give me a history of having had an abnormal heart. In the guidance, we say that essentially all patients... In medicine, nothing is always; it's "strongly consider" or whatever. But we came down on this: if you have Staph aureus in your bloodstream, isolated on blood culture, you should get a transthoracic echo. With the scenario that I went through, depending on the quality of the transthoracic echo, you might proceed to a transesophageal echo, and so we try to articulate those risk factors that would help in making your decision to do that or not, as an example.

Contagion: How should clinicians weigh risk factors that may be inapparent at first when deciding whether to shorten or extend therapy?

Chambers: We didn't really address longer courses of therapy in this initial listing of guidance. We focused on those who might be appropriate for a shorter course. I can't remember the numbers, but I think it's questions 6 and 7, or 5 and 6, that address the duration of therapy and in whom 2 weeks would be appropriate.

Essentially, there are some patients who are clearly at low risk when you initially identify them. I'll just give you an example: a patient who's in the hospital for, let's say, workup of a gastrointestinal bleed, and they have an IV in place. That IV becomes inflamed, and the patient develops a fever. Blood cultures are obtained, and they grow Staphylococcus, and you know it's from the IV site because that is inflamed, and maybe they have signs of a local phlebitis. So you're pretty confident what the source is and the duration of the bacteremia, because they were there under your eyesight. Those patients are very low risk unless they have concomitant risk factors. Let's give them a prosthetic heart valve, for example. But our patient is there for a GI bleed. They don't have a prosthetic heart valve. So that is an inherently low-risk situation, and there are good data, not randomized controlled trials, but nevertheless good experiential data, that those patients will do well with this short course of therapy.

Now let's contrast that with somebody who comes into the hospital with a skin abscess, but they are also an injection drug user. Those 2 factors make a diagnosis of a more complicated infection more likely. First of all, it's community acquired. As a risk factor, our initial patient had hospital-acquired bacteremia. We were there on the spot and noticed when the blood cultures were drawn in relation to the infection. We know the source of the infection; it's pretty clear what the source is. Well, with the second patient, we don't know how long that blood culture has been positive. It's possible, although not necessarily likely, that it even preceded the abscess, because they have a risk factor for abscesses, which is the injection drug use. People who have community-onset bacteremia are at higher risk for endocarditis. People who use injection drugs are at a higher risk for endocarditis. So now I have 2 risk factors that would make me concerned about endocarditis, and I'm going to be much more diligent with that set of circumstances in ruling out endocarditis and doing the appropriate workup, maybe looking for other sources and things like that.

So in the first case, we say in low-risk infections you can treat for 2 weeks. The next question was, well, what about people who have high-risk factors, but you do a workup and it's all negative? So in the second case, I get an echocardiogram. Maybe I get a transesophageal echo. I've repeated the blood cultures, and those are negative. I'm pretty sure it's just a skin abscess. So they have risk factors for more complicated disease, but I've done an appropriate workup to address the likelihood of those and reduce the probability of complicated disease. In that situation, 2 weeks of therapy should be considered and is probably appropriate. It's that kind of approach, if that makes sense.

Contagion: What key unanswered questions or research priorities did the panel identify, and what would you want studied to validate the framework's real-world impact?

Chambers: Two good questions. I'll take on the first one: What work is to be done? If you look at the questions, we have not told you anything about what drug to use. We've not articulated longer courses of therapy. We've not addressed MRSA or methicillin-susceptible Staphylococcus aureus infections, or how you deal with more complicated cases. There are 6 or 7 additional questions that we've proposed to the IDSA Guidelines Committee, and we hope to take those on in the next bit of work that we think should be done. So I think there's a lot of work left to do. I told you 2 weeks, but I haven't told you whether you can take pills. What pills should you take? How long should the IV therapy be? Do you even need IV therapy? Those kinds of questions in terms of management. This mainly focused on the framework itself, of how you go about thinking about a patient with Staph aureus.

With respect to the framework, the framework has not been rigorously tested, just to be totally transparent. We sat down and tried to identify factors that we know carry risk probabilities for a complication, and how to go about thinking about those. We've proposed a set of criteria that we think are a rational starting point, and I think it will be a work in progress as the imaging changes and moves on and further data become available. They'll be tested as a foundation for understanding: Do they improve the approach to the patient and the care of the patient?

Editor's Note: This transcript has been edited for grammar and clarity using artificial intelligence tools.

REFERENCES

1.IDSA/ESCMID 2026 consensus statements on Staphylococcus aureus bacteremia: risk stratification, diagnostic evaluation, and management of adults and children. Infectious Diseases Society of America; European Society of Clinical Microbiology and Infectious Diseases. Published September 9, 2026. Accessed October 1, 2026. https://www.idsociety.org/practice-guideline/staphylococcus-aureus-bacteremia/

2. Staph aureus bacteremia: new framework aims to personalize care. Epocrates. Published September 24, 2026. Accessed October 1, 2026. https://www.epocrates.com/online/article/staph-aureus-bacteremia-new-framework-aims-to-personalize-care

3.Consensus statement 1 on the risk stratification of patients with Staphylococcus aureus bacteremia. Infectious Diseases Society of America; European Society of Clinical Microbiology and Infectious Diseases. Accessed October 1, 2026. https://www.idsociety.org/globalassets/idsa/practice-guidelines/sab/consensus-statement-1-manuscript.pdf