Should all adjunctive corticosteroid therapy be avoided in the management of hemodynamically stabile Staphylococcus aureus bacteremia?

The purpose of this study was to examine the prognostic impact of corticosteroids in hemodynamically stabile Staphylococcus aureus bacteremia (SAB). There were 361 hemodynamically stabile methicillin-sensitive SAB patients with prospective follow-up and grouping according to time-point, dose and indication for corticosteroid therapy. To enable analyses without external interfering corticosteroid therapy all patients with corticosteroid therapy equivalent to prednisone > 10 mg/day for >= 1 month prior to positive blood culture results were excluded. Twenty-five percent (92) of patients received corticosteroid therapy of which 11 % (40) had therapy initiated within 1 week (early initiation) and 9 % (31) had therapy initiated 2-4 weeks after (delayed initiation) positive blood culture. Twenty-one patients (6 %) had corticosteroid initiated after 4 weeks and were not included in the analyses. A total of 55 % (51/92) received a weekly prednisone dose > 100 mg. Patients with early initiated corticosteroid therapy had higher mortality compared to patients treated without corticosteroid therapy at 28 days (20 % vs. 7 %) (OR, 3.11; 95% CI, 1.27-7.65; p < 0.05) and at 90 days (30 % vs. 10 %) (OR, 4.01; 95% CI, 1.82-8.81; p < 0.001). Considering all prognostic markers, early initiated corticosteroid therapy predicted 28-day (HR, 3.75; 95% CI, 1.60-8.79; p = 0.002) and 90-day (HR, 3.10; 95% CI, 1.50-6.39; p = 0.002) mortality in Cox proportional hazards regression analysis. When including only patients receiving early initiated corticosteroid therapy with prednisone >= 100 mg/week the negative prognostic impact on 28-day mortality was accentuated (HR 4.8, p = 0.001). Corticosteroid therapy initiation after 1 week of positive blood cultures had no independent prognostic impact. Early initiation of corticosteroid therapy may be associate to increased mortality in hemodynamically stabile SAB.