Antibiotics did not improve outcomes in adult patients who were hospitalized for an asthma exacerbation and received systemic corticosteroids, a study found.
The retrospective cohort study included 19,811 adults hospitalized for an asthma exacerbation and treated with systemic corticosteroids in any of 542 U.S. acute care hospitals from Jan. 1, 2015, through Dec. 31, 2016. Almost half of the patients (44.4%) had antibiotics started during the first two days of hospitalization and prescribed for a minimum of two days. Outcomes were compared by whether patients did or didn't receive antibiotics, and results were published by JAMA Internal Medicine on Jan. 28.
Patients who received antibiotics were older (median age, 48 vs. 45 years), were more likely to be white (48.6% vs. 40.9%), and were more likely to smoke (6.6% vs. 5.3%) than those who didn't, as well as have a higher rate of comorbidities (e.g., 6.2% with congestive heart failure vs 5.8%). The study's primary outcome, hospital length of stay, was significantly longer in the antibiotic group (median, 4 days vs. 3 days; ratio in a propensity score-matched analysis, 1.29 [95% CI, 1.27 to 1.31]). The groups had similar rates of treatment failure within 30 days (defined as initiation of mechanical ventilation, transfer to the ICU after hospital day 2, in-hospital mortality, or readmission for asthma): 5.4% with antibiotics versus 5.8% without. Hospital costs were higher in the patients treated with antibiotics.
“These findings are novel, reflect the experience of unselected patients cared for in routine settings, and lend strong support to current guidelines that recommend against the use of antibiotics in the absence of concomitant infection. In addition, the findings highlight the need for future research to improve antimicrobial stewardship in the setting of asthma,” the authors said. They noted that biomarkers for guiding antibiotic therapy, such as procalcitonin, may be helpful in this area.
Limitations of the study include the possibility of residual confounding by indication, although the authors said they conducted robust statistical and sensitivity analyses to try to avoid this problem. Because the study was observational, the results do not demonstrate causality. The data set also lacked physiological measures of disease severity.