Corticosteroid use as adjunctive therapy for treating influenza was associated with increased mortality but the existing evidence was of very low quality, researchers concluded in a study published in the Cochrane Library.

Treatments for influenza are typically limited to neuraminidase inhibitors and adamantanes. Corticosteroids have demonstrated benefit in sepsis and related conditions that may be due to their anti-inflammatory and immunomodulatory properties. There lies uncertainty, however, over their possible benefit or harm as they are commonly prescribed for severe influenza.

RELATED: CDC Urges Greater Treatment for Suspected Influenza

Study authors searched randomized controlled trials, quasi-randomized controlled trials, and observational studies to systematically assess the efficacy and potential adverse effects of corticosteroids as adjunctive therapy in treating influenza, accounting for the varied timing and doses of corticosteroids. They included 19 eligible studies (n=3,459)—all observational studies—that compared corticosteroid therapy vs. no corticosteroid therapy for influenza or influenza-like illness. Quality of the evidence was established via the Grading of Recommendations Assessment, Development and Evaluation (GRADE) framework.

Of the 19 identified studies, 13 were appropriate for inclusion in a mortality meta-analysis. Data regarding mortality, however, were of “very low quality” according to researchers. Corticosteroid doses that were used were high and indications for use were not well reported.

Data from the meta-analysis showed that corticosteroid therapy was linked to increased mortality (odds ratio [OR] 3.6, 95% CI: 1.61-4.92). Higher chances of hospital-acquired infection related to corticosteroid therapy were reported in 3 studies but all were unadjusted estimates and the evidence was assessed as “very low quality.”

Overall, study authors were not able to identify any completed randomized controlled trials of adjunctive corticosteroid use in the treatment of influenza. Data analyzed from the observational studies was of very low quality. They noted that confounding by indication posed a major potential concern. They concluded insufficient evidence for this review to determine the efficacy of corticosteroids for patients with influenza.

For more information visit