Introduction: Timely and appropriate therapy is critical in patients with Gram-negative bloodstream infections (GNBSI). Most bacteriology laboratories process blood specimen in the daytime, during laboratory operating hours, and use conventional culture for antimicrobial susceptibility testing (AST). We simulated the potential impact of real-time processing and rapid AST (7 hours) on early adaptation of the antibiotic regimen in intensive care unit (ICU) patients with GNBSI. Methods: All GNBSI episodes occurring in the ICUs of 2 hospitals in Paris were included. Data were collected. For each episode of bacteremia, we simulated the impact of three strategies: (1) Real-time processing coupled with conventional techniques (Gram stain and standard AST); (2) Standard processing coupled with rapid AST; and (3) Real-time processing coupled with rapid AST. Results: We included 109 episodes in 98 patients. Forty-two patients (48%) died during ICU stay. AST results led to a change of the antibiotic regimen in 66 (61%) episodes, mainly de-escalation (54/109, 55%). In standard care, median time from sample collection to definitive AST result was 65.9 hours (±26.7). The three strategies would have reduced time-to-result by 9.2 hours (±7.1), 30.8 hours (±19.7) and 40.0 hours (±20.6) respectively. Compared to standard care, strategies 1, 2 and 3 would have avoided 20, 69 and 90 patient-days of broad-spectrum antibiotics respectively. Conclusion: In addition to real-time processing of blood samples, rapid AST would be the most effective strategy to shorten time-to-result in critical patients with GNBSI.