Two Symptoms Linked to Antibiotic Delays in Presumed Sepsis

Patients with community-onset presumed bacterial sepsis were less likely to get timely antibiotics if they had gastrointestinal (GI) or respiratory symptoms at the time of presentation, an observational cohort study found.

In the analysis of nearly 30,000 patients, multivariable analysis showed that GI symptoms were associated with a lower odds of timely antibiotic receipt both in patients who had hypotension (adjusted OR [aOR] 0.84, 95% CI 0.75-0.94) and those who did not (aOR 0.80, 95% CI 0.75-0.86), reported Elizabeth Munroe, MD, of Intermountain Medical Center in Murray, Utah, and colleagues.

Respiratory symptoms, meanwhile, were linked to a lower likelihood only in the subset without hypotension (aOR 0.91, 95% CI 0.85-0.99), according to findings detailed in JAMA Network Open.

And those delays were linked to hard outcomes. When accounting for symptoms at presentation, the patients who received timely antibiotics — within 3 hours for patients with hypotension and 5 hours for those without hypotension — had a lower 30-day mortality rate (18.9% vs 21%, P<0.001).

“These findings highlight the importance of a patient’s clinical history in facilitating sepsis identification and treatment, and suggest that paying particular attention to patients with symptoms that may mask sepsis, such as GI and respiratory symptoms, may be helpful in early sepsis diagnosis,” the research team wrote.

Symptoms associated with better odds of timely antibiotic receipt included prehospital fever, regardless of hypotension status (with: aOR 1.34, 95% CI 1.19-1.52; without: aOR 1.26, 95% CI 1.17-1.35), and urinary symptoms in patients without hypotension (aOR 1.13, 95% CI 1.05-1.22).

A life-threatening over-response to infection, sepsis is a common cause of hospitalization that plays a role in about half of U.S. hospital deaths.

“Early administration of antibiotics is associated with improved survival in sepsis, and the 2026 Surviving Sepsis Campaign guidelines recommend antibiotic therapy within 1 hour in patients in shock and within 3 hours in hemodynamically stable patients,” noted Kristin Sheehan, MD, and Ryan Maves, MD, in an accompanying editorial.

Recognizing sepsis may be straightforward in patients whose illness at presentation goes by the script, according to the editorialists, but patients’ complex and ambiguous presentations make diagnosis challenging and delays in antibiotic treatment remain common in clinical practice.

“The variability of sepsis can confound our ability to see it when it lies in front of our eyes,” wrote Sheehan and Maves, both of Wake Forest University School of Medicine in Winston-Salem, North Carolina.

Delays in antibiotics for patients with GI symptoms “may reflect the long list of common alternative diagnoses that may be seen with GI concerns,” they suggested, noting that just fewer than 1 in 20 adults who present to the emergency department for diarrhea need to be hospitalized or treated for sepsis.

While it may be tempting to respond to the study findings “with more aggressive protocols for earlier antibiotic administration for more patients, we must remember it is also the responsibility of the clinician to minimize overdiagnosis and practice good antibiotic stewardship,” Sheehan and Maves cautioned. “This careful balance is precisely what makes sepsis management so challenging.”

Munroe’s team analyzed 29,647 adult patients hospitalized for community-onset presumed bacterial sepsis from November 2020 to May 2024 at 67 hospitals in the Michigan Hospital Medicine Safety Consortium Sepsis Initiative. Given timely antibiotics’ differing benefits in patients with and without hypotension, the researchers analyzed hypotension status. The primary outcome of timely systemic antibiotic receipt was defined as within 3 hours in patients who presented with hypotension and within 5 hours in those without hypotension.

Median patient age was 71 years, 49.5% were men, and the median Charlson Comorbidity Index was 3. Just over 1 in 5 patients were hypotensive (22.6%), and the most common presenting symptoms were respiratory (69.5%), prehospital fever (53.7%), and altered mental status (49.4%).

While a majority of patients overall got timely antibiotics, it was more common in patients who presented without hypotension compared with those with hypotension (76.5% vs 65.8%).

Thirty-day mortality rates with timely antibiotics were significantly lower in patients who weren’t hypotensive (15.2% vs 17.3%, P<0.001), but not in those with hypotension (32.1% vs 32.9%, P=0.45). However, Munroe and colleagues cautioned that “this finding should not be taken to mean that timely antibiotics are not important for patients with hypotension,” given that other research has shown an association between faster antibiotics and lower mortality in those patients.

Limitations included the observational study’s potential for residual confounding, and the researchers said the secondary association results should be considered exploratory.