Bacterial Infections Sepsis & Critical Care Antimicrobial Resistance & Stewardship Clinical Care & Diagnostics

STSS found to be strongly associated with GBS mortality

Share

Streptococcal toxic shock syndrome was the factor most strongly associated with 30-day death among adults with group B Streptococcus bacteremia, according to a multicenter retrospective cohort study published in Open Forum Infectious Diseases.

Overall, 9% of patients died within 30 days, with 39% of patients who developed streptococcal toxic shock syndrome dying during that period. Immunosuppressive therapy, cirrhosis, respiratory infection, advanced renal dysfunction and polymicrobial bacteremia were also independently associated with increased odds of death.

Study design and patient characteristics

The investigators conducted a retrospective observational study of 442 adults hospitalized with group B Streptococcus bacteremia at seven hospitals in western Japan between January 2015 and December 2024. Eligible patients were aged 18 years or older and had at least one blood culture positive for GBS during the study period. Patients who were not hospitalized, had recurrent GBS bacteremia, had insufficient data to ascertain clinical outcomes because of discharge or transfer within 2 days of admission, or lacked detailed clinical information were excluded.

The primary outcome was all-cause death within 30 days of the index blood culture positive for GBS. Investigators collected demographic, clinical, microbiological and treatment data from electronic medical records and used univariable and multivariable logistic regression analyses to identify factors associated with 30-day death.

The median patient age was 75 years, and 54% were male. Most patients (83%) had at least one underlying comorbidity, and 47% had two or more. The most common conditions included diabetes mellitus (38%), a history of solid tumors (29%), heart failure (18%), cirrhosis (10%) and advanced renal dysfunction (10%).

Bacteremia without an identifiable focus was the most common clinical presentation, occurring in 27% of patients, followed by skin and soft tissue infections (26%), urinary tract infections (22%) and musculoskeletal infections (18%). Multiple infection sites were identified in 10% of patients.

Streptococcal toxic shock syndrome occurred in 28 patients (6%), while 66 patients (15%) had polymicrobial bacteremia. Approximately 22% required intensive care unit admission.

Mortality and associated risk factors

Overall, 41 of 442 patients (9%) died within 30 days, and 50 patients (11%) died during hospitalization. In multivariable analysis, STSS was the factor most strongly associated with 30-day death (adjusted odds ratio, 9.16; 95% CI, 3.28–25.58; P < .001).

Other factors independently associated with increased mortality included:

  • Immunosuppressive therapy: 4.6 times the odds of death

  • Cirrhosis: 3.4 times the odds

  • Respiratory infection: 3.3 times the odds

  • Advanced renal dysfunction: 2.7 times the odds

  • Polymicrobial bacteremia: 2.4 times the odds

In contrast, skin and soft tissue infection was independently associated with lower odds of death, with an adjusted odds ratio of 0.07.

Among patients with STSS, 11 of 28 (39%) died within 30 days, compared with 30 of 414 (7%) without STSS. The proportion who died within 30 days was 26% among patients with respiratory infections, 14% among those with bacteremia without an identifiable focus and 1% among those with skin and soft tissue infections.

Among patients with four or more comorbidities, 36% died within 30 days. Hospital-acquired infections accounted for 13% of cases and were associated with a higher proportion of deaths than community-acquired infections in univariable analysis.

Subgroup analyses

Among patients with STSS, the proportion who died within 30 days remained high across all adult age groups, ranging from 29% to 50%, with no clear age-related trend.

Compared with patients without STSS, those with STSS had higher rates of necrotizing fasciitis (14% vs. 1%), polymicrobial bacteremia (46% vs. 13%) and intensive care unit admission (86% vs. 18%).

Baseline comorbidities were generally similar between patients with and without STSS.

In a supplementary analysis, patients with pneumonia were more likely to be male, live in long-term care facilities and have dementia, and they had a lower body mass index than those without pneumonia.

A sensitivity analysis excluding polymicrobial bacteremia from the multivariable model produced materially unchanged associations between the remaining factors and 30-day death.

Antimicrobial susceptibility and treatment

GBS isolates remained highly susceptible to beta-lactam antibiotics. Susceptibility rates were 99% for penicillin G, 99.5% for ampicillin, 99.7% for ceftriaxone, 100% for meropenem and 98% for cefepime. All tested isolates were also susceptible to vancomycin and linezolid (100% each).

Susceptibility was lower for erythromycin (67%), clindamycin (86%) and levofloxacin (66%). Susceptibility rates did not differ significantly between survivors and nonsurvivors for any antimicrobial agent tested.

The most frequently administered definitive antimicrobial therapies were ampicillin (32%), ampicillin/sulbactam (19%), ceftriaxone (11%) and penicillin G (7%).

In unadjusted comparisons, 4% of patients receiving ampicillin monotherapy died within 30 days, a lower proportion than among patients receiving other treatments.

However, the investigators cautioned that the association may have reflected treatment selection among patients with less severe illness rather than a therapeutic advantage. The observational design did not permit conclusions about the comparative effectiveness of antimicrobial regimens.

Study limitations

The investigators acknowledged several limitations, including the retrospective design, potential residual confounding, and incomplete measures of illness severity and comorbidity burden.

The findings may not fully represent all cases of invasive GBS disease because infections identified exclusively through cultures from sterile sites other than blood were not captured.

Infection sites were classified retrospectively without centralized adjudication, and all participating hospitals were located in western Japan, potentially limiting generalizability.

Additionally, the absence of serotype and molecular data precluded a detailed assessment of strain-specific virulence factors. Antimicrobial susceptibility testing varied across institutions, and inducible clindamycin resistance was not systematically evaluated.

The relatively small number of deaths may have constrained model stability, and data on the timeliness of appropriate therapy and adequacy of source control were not systematically available.

The authors reported having no relevant conflicts and received no funding for this study.

Author insight

Kohei Oguni, MD, PhD, of the Department of Infectious Diseases at Okayama University Graduate School of Medicine, Dentistry and Pharmaceutical Sciences in Japan, and a study author, discussed the findings, their clinical implications and directions for future research in comments to IDSA Reporter.

What are the key takeaways?

Dr. Oguni: Our multicenter study of 442 adults with GBS bacteremia in Japan demonstrated an overall 30-day mortality rate of 9.3%. However, mortality varied substantially according to the clinical presentation and underlying patient characteristics.

The most striking finding was the particularly poor prognosis associated with GBS-related streptococcal toxic shock syndrome (STSS), which had a 30-day mortality rate of 39.3% and was the strongest independent predictor of mortality. Immunosuppressive therapy, cirrhosis, advanced renal dysfunction, respiratory infection and polymicrobial bacteremia were also independently associated with increased mortality.

Importantly, mortality among patients with STSS remained high across all adult age groups, emphasizing that this severe manifestation should not be regarded as a concern limited to older patients.

What are the implications of the study findings?

Dr. Oguni: Our findings highlight the importance of recognizing that GBS bacteremia is a clinically heterogeneous condition, with prognosis varying markedly according to infection site, disease severity and underlying comorbidities.

In clinical practice, the identification of STSS and other mortality-associated factors may help clinicians recognize patients who require particularly close monitoring and prompt management. Although GBS remains largely susceptible to beta-lactam antibiotics, favorable antimicrobial susceptibility does not necessarily translate into favorable clinical outcomes, particularly in patients with severe systemic manifestations.

We believe that careful assessment of both the clinical presentation and host-related risk factors is essential when managing adults with GBS bacteremia.

What cautions or next steps for research would you highlight?

Dr. Oguni: Future prospective, multicenter studies with larger populations are needed to validate these prognostic factors and clarify their clinical utility. Furthermore, incorporating bacterial serotyping and genomic analyses may provide important insights into the pathogen-related mechanisms underlying severe manifestations, particularly GBS-associated STSS.

Is there anything else that you would like to discuss?

Dr. Oguni: GBS has traditionally been recognized as an important pathogen in neonates and pregnant women. However, invasive GBS disease is also an important clinical concern among nonpregnant adults, particularly older individuals and those with underlying medical conditions.

One important message from our study is that GBS bacteremia encompasses a broad spectrum of clinical presentations, ranging from relatively favorable outcomes in patients with skin and soft tissue infections to life-threatening disease in those with STSS or respiratory infections.

We hope our findings will contribute to greater awareness of invasive GBS disease in adults and encourage further research into improving risk stratification and clinical outcomes.

Independent expert perspective

For an independent assessment of the findings and their implications for clinical practice, IDSA Reporter invited David M. Aronoff, MD, FIDSA, FAAM, FRCP, the John B. Hickam Professor of Medicine and chair of the Department of Medicine at Indiana University School of Medicine, to provide his perspective on the study.

What makes this study important for infectious diseases specialists?

David M. Aronoff, MD, FIDSA, FAAM, FRCP

Dr. Aronoff: When physicians and other health care professionals think about the bacterium group B strep or GBS, which is an abbreviation for a bacterium called Streptococcus agalactiae, we usually associate that bacterium with infections that complicate pregnancy or the early life of a newborn. GBS is a common cause of severe infection of the newborn that we call neonatal sepsis. It can also cause infection inside the pregnant uterus that leads to complications including the early onset of labor and delivery prior to term (37 weeks of gestation) and the premature rupture of the membranes that surround the amniotic fluid and the fetus, which is what is meant when a person's water breaks before term.

Having a group B strep infection in the uterus can also infect the developing fetus and if that happens the fetus might not survive. That's why GBS is also an important cause of stillbirth. So, many of us in medicine, and particularly infectious diseases, have developed a real appreciation for the impact of GBS in newborns and pregnant patients.

This study, focused on GBS infections of adults outside of pregnancy, reminds us that group B strep can cause serious and even life-threatening infections in nonpregnant persons. That in and of itself is not news; we have known that GBS poses a threat to nonpregnant adults and that there are important risk factors for such infections, which this study continues to underscore. For example, the presence of diabetes mellitus is a major, well-characterized risk factor for GBS infections. This new, large, multicenter investigation of group B strep bloodstream infections provides ID professionals with information that will help guide the care of hospitalized patients with suspected or confirmed GBS infections.

What do the findings reveal about antimicrobial resistance?

Dr. Aronoff: This study is a modern look at invasive group B strep infections, specifically limited to those infections that involve the bloodstream, which is also called bacteremia. The team studied 442 adult patients with GBS bacteremia over a 10-year period and across multiple hospitals in Japan. Having a fresh look at a pathogen we've known about for a long time and asking questions about one of the more severe presentations of that infection — which is a patient who has group B strep in the bloodstream, and paying attention to the clinical aspects of that illness, the risk of death from that illness and some of the risk factors — is important. Partially it's important because like so many things in infectious diseases, the clinical presentations and the way we treat infections can change over time. One of the areas that GBS has us a little bit worried about is resistance to antibiotics. These investigators were very careful to study the vulnerability or susceptibility of GBS in the bloodstream of these patients to common antibiotics that we often use in sick patients.

Fortunately, the investigators found that group B strep maintains susceptibility or vulnerability to penicillins and cephalosporins and related medications, which is great, but there continues to be problems with drug resistance to macrolide antibiotics — things like erythromycin and azithromycin and fluoroquinolone antibiotics like levofloxacin. Macrolides and fluoroquinolones are very commonly used medications. In this study, about a third of the isolates of group B strep were resistant to those types of drugs, even though close to 100% of the GBS isolates were sensitive to penicillins and cephalosporins and related antibiotics. So, one of the important things of this study is that it helps reinforce the idea that selection of antibiotics for group B strep infections is important.

What are the most common clinical presentations of group B strep?

Dr. Aronoff: The other thing that this study really underscores for clinicians is that it reminds us that in the nonpregnant adult, there are some common presentations for group B strep infection, even those that get complicated by a bloodstream infection.

Urinary tract infections are relatively common among the infections caused by GBS. Skin and soft tissue infections, including cellulitis, are among the illnesses that group B strep causes. Sometimes deeper musculoskeletal infections like infected arthritis can be caused by GBS. Sometimes patients show up with a bloodstream infection with group B strep and we never really know what the source was or how it got into the bloodstream. We call that a bacteremia with an unknown focus. About one in four patients in this study had bacteremia with an unknown focus. About one in 20 patients in this study also had pneumonia caused by group B strep. Although pneumonia is not a particularly common presentation of GBS, it is important because approximately one in four patients with GBS pneumonia died within 30 days.

Another important aspect of this study, which reinforces findings from previous studies, is that patients hospitalized with group B strep bacteremia tend to be older. Of the 442 adults in this study who had GBS bloodstream infection, the median age was 75 years and 75% of the patients were either 64 years of age or older. So these are older adults and the study again found that diabetes mellitus was a risk factor for invasive GBS infection. And when the investigators looked at risk factors for dying, they found a few things that I think are important for people to take home. One is that 30-day mortality was quite high (39.3%) if somebody presented with a clinical presentation consistent with a “toxic shock syndrome,” defined by the authors as shock accompanied by at least two of the following: liver failure, renal failure, acute respiratory distress syndrome, disseminated intravascular coagulation, soft tissue inflammation including necrotizing fasciitis, generalized erythematous rash, or central nervous system abnormalities. And while only about 6% of the patients in this study had a toxic shock syndrome-like presentation, that's a pretty high mortality.

Why is streptococcal toxic shock syndrome particularly concerning?

Dr. Aronoff: When people think about streptococcal toxic shock syndrome, we usually associate that with group A Streptococcus, which is caused by Streptococcus pyogenes, which is the common cause of strep throat. And GAS causes a toxic shock syndrome through mechanisms that are similar to how Staphylococcus aureus causes toxic shock. It's through a process mediated by something called superantigens that cause massive inflammation in the infected person and cause their blood pressure to drop precipitously, which is hard to treat. With GBS, there can also be severe low blood pressure or shock and a clinical picture that looks similar to GAS toxic shock syndrome but, to our knowledge, we don't actually think that GBS makes a superantigen and it's not entirely clear how GBS causes toxic shock.

And I found it interesting in this study that almost half of the patients who were diagnosed with toxic shock syndrome had polymicrobial infections of the bloodstream — they had two or more different types of bacteria in their blood, including GBS. That might have contributed to why those patients got so sick. Overall, this study does a good job of highlighting that people can get very sick and have a relatively high mortality with GBS, and we need to be aware of that.

Which patient characteristics are associated with an increased risk of death?

Dr. Aronoff: The other thing that I would say is that in patients with pneumonia caused by GBS that those patients tended to be frail, thinner (have a lower body mass index) than other patients in the study and they tended to be living in long-term care settings. The investigators were concerned that these are patients who may be aspirating. In other words, they may have difficulty swallowing and they may be getting gastric secretions or oral secretions into their lungs. And when somebody who's frail and older gets pneumonia of any cause, mortality tends to be higher than when somebody young gets pneumonia. And that may be why group B strep pneumonia in this study was associated with an increased risk for death and had a survival of about three out of four patients survived with pneumonia, but about one out of four did not.

The last thing I'll say that the study really points out is that people on immunosuppressive medications, including corticosteroids and other immune-depressing medications, also had a higher mortality with GBS bacteremia.

What are the key clinical takeaways for infectious diseases specialists?

Dr. Aronoff: In summary, this study underscores a number of things clinicians need to know about GBS infections in nonpregnant adults. When people are hospitalized with GBS bacteremia, nearly one out of 10 people don't survive 30 days.

The health of the person who's infected matters. If somebody is frail or has advanced liver disease or is on immunosuppressive therapy or has a syndrome consistent with a toxic shock-like illness, they are really at increased risk for doing poorly. And that may be important when we talk to families about prognosis or we are thinking about the setting in which we want to care for those patients — like inside an intensive care unit versus on a regular hospital bed or keeping someone in the emergency department. When we're worried about GBS, it's okay to use beta-lactam antibiotics like penicillins and cephalosporins. There are also alternatives for people who can't take those medications. GBS remains very sensitive to vancomycin and linezolid, but we want to avoid fluoroquinolones and macrolides because resistance is a problem. I would say that resistance is increasing in clindamycin, which is another drug that people often use in the setting of streptococcal toxic shock syndrome.