While sepsis is rare, it's nearly 80% more likely after treatment with certain "high-risk" antibiotics, compared with no antibiotic therapy, according to the Centers for Disease Control and Prevention. From MedPage Today.
This article first appeared October 27, 2016 on MedPage Today.
NEW ORLEANS -- Exposure to antibiotics during a hospital stay can sharply increase the risk of sepsis or septic shock after discharge, a researcher said here.
While sepsis is rare, it's nearly 80% more likely after treatment with certain "high-risk" antibiotics, compared with no antibiotic therapy, according to James Baggs, PhD, of the Centers for Disease Control and Prevention (CDC).
But the risk is elevated regardless of what antibiotics are used and rises with the duration of treatment, Baggs reported at the annual IDWeek meeting. The meeting brings together four organizations that focus on infectious disease -- the Infectious Diseases Society of America (IDSA), the Society for Healthcare Epidemiology of America(SHEA), the HIV Medicine Association (HIVMA), and the Pediatric Infectious Diseases Society (PIDS).
The study is "interesting because it brings together a number of really hot topics -- antibiotic use, sepsis, and this whole concept of the good bacteria in our bodies, the microbiome," commented Arjun Srinivasan, MD, of the CDC, and the conference program chair for SHEA.
Srinivasan told reporters the study builds on intriguing data from animals that has suggested, "if you mess up their microbiome" with antibiotics, they are much more susceptible to serious infection and sepsis.
Baggs and colleagues, he said, have shown that in humans, the same is true: "There really is a tremendous disturbance in the healthy bacteria when we get an antibiotic that can have a significant downside."
It's "one more call to focus our efforts on using antibiotics only when we need them," he said.
Baggs said there has been "emerging evidence that environmental exposures, such as antibiotics, can lead to a shift in the microbiome with great pathogenic potential."
To investigate the issue, he and colleagues turned to the Truven Health MarketScan Hospital Drug Database, which includes demographic, treatment, and hospital admission and discharge data for millions of patients. The database also includes pharmacy data, making it possible to track antibiotic use.
For the years 2006 through 2010, Baggs said, the researchers obtained data on 12.7 million hospital stays in 516 facilities, including 18,307 that were associated with a sepsis admission within 90 days of discharge following a previous admission.
For the retrospective cohort study, the team defined a list of antibiotics with a high risk of disturbing the microbiome: third- or fourth-generation cephalosporins, fluoroquinolones, lincosamides, beta-lactam/beta-lactamase inhibitor combinations, oral vancomycin, and carbapenems.
Another set of drugs, including earlier cephalosporins, tetracycline, and sulfa drugs, was regarded as low-risk, while control antibiotics, such as penicillin, were thought to have limited potential to disrupt the microbiome.
The primary endpoint of the analysis was the risk of sepsis following use of those drugs compared with no antibiotic use, Baggs said, but they also looked at the risk associated with other antibiotics and the risk associated with longer versus shorter treatment lengths.
In 43% of the hospital stays, patients had not been given any antibiotics, Baggs reported, while patients got high-, low-, and no-risk drugs in 28%, 24%, and 5% of the stays, respectively.
A multivariate analysis showed that, compared with no antibiotics, the odds ratio for sepsis was:
- 1.78 (with a 95% confidence interval from 1.72 to 1.85) after high-risk drugs
- 1.10 (1.04 to 1.16) after low-risk drugs
- 1.22 (1.11-1.35) after no-risk drugs
Baggs added that duration of therapy also played a role: Regardless of drug type, patients treated for more than 14 days had twice the risk of later sepsis as those given shorter therapy, with an odds ratio of 2.4.
"The exact mechanisms remain under investigation," Baggs said, but the study "contributes to the growing body of observational evidence and animal data [that antibiotics can have] broad detrimental effects."
But he noted that physicians and health authorities have one preventive measure that's currently available: "improved antibiotic stewardship."
Baggs cautioned that, while the study had a large population and many years of data, it is based on an administrative database with a potential for misclassification. He noted that the study could not account for antibiotic exposures outside the hospital setting and captured only sepsis patients admitted to the same hospital they had recently left.