Rapid Empiric Treatment With Oseltamivir Study (RETOS)

Overview

Current guidelines recommend early initiation of empiric antibiotic therapy to cover typical and atypical bacteria that may cause community-acquired pneumonia (CAP). Influenza antiviral therapy in patients with suspected or confirmed influenza. However, many clinicians do not suspect influenza among patients with CAP or other acute lower respiratory tract illness (LRTI) and often do not test for influenza. Additionally, results from currently available diagnostic tests for influenza may be delayed and several tests have low sensitivity and will give false negative results. Thus, anti-influenza treatment for patients with hospitalized influenza CAP and LRTI is frequently initiated late if at all. There is an association between delayed time to administration of empiric antibiotic therapy with increased clinical failure and mortality. As a result, empiric antibiotic therapy for patients with suspect CAP is begun within 4 - 6 hours of hospitalization. This has recently been demonstrated for delayed antiviral treatment as well. We hypothesize that, as happens with early empiric antibiotics for bacterial CAP, a standardized approach of adding early empiric anti-influenza therapy during the influenza season to hospitalized patients with suspect CAP and LRTI will improve clinical outcomes of patients with influenza associated CAP and LRTI. To test our hypothesis we plan a prospective, randomized, multicenter clinical trial of hospitalized patients with acute LRTI, including suspect CAP, during . If early anti-influenza medications were not included on the patients admission orders, patients will be randomized to standard care, including empiric antibacterial therapy as recommended by ATS/IDSA guidelines plus standard influenza diagnostics and treatment (Standard of care) versus early initiation of empiric antiinfluenza therapy plus standard care, e.g. empiric antibacterial (oseltamivir group). The primary study outcome will be development of clinical failure and selected clinical outcomes during the 30 days after enrollment. Other clinical outcomes that will be compared between study groups include time to clinical stability, duration of hospitalization, development of cardiovascular events, re-hospitalization, short-term mortality (30 days), and long-term mortality (1 year). The secondary study outcome will be the cost-effectiveness of the intervention.

This will be both a prospective, randomized, unblinded clinical study of hospitalized patients with acute LRTI admitted in one of four institutions in Louisville, KY (rapid empiric treatment with oseltamivir study\[RETOS\]) and a prospective observation study to describe influenza LRTI (Flu LRTI study). All hospitalized patients with acute LRTI will be invited to participate in one of the arms of study. If the admitting clinician does not order oseltamivir or zanamivir at the time of hospital admission, the patient is eligible for randomization into Group A (standard clinical care, including empiric antibiotics and anti-influenza drugs at the clinician discretion) or Group B (oseltamivir administered to the patient within 24 hours of admission, ideally within 8-12 hours of admission, plus empiric antibiotics). Patients will be enrolled from one of four hospitals, the University of Louisville Hospital, Veterans Affairs Medical Center of Louisville, Norton Hospital of Louisville, and Jewish Hospital of Louisville. Eligible patients will be identified primarily in the Emergency Departments of all four hospitals and evaluated for inclusion/exclusion criteria after hospital admission orders are written. Patients will be enrolled only during the influenza season. For this study, the influenza season is defined as December 1st until May 1st, unless surveillance data suggests that influenza viruses are circulating earlier or have stopped circulating. For all three study groups, diagnosis of influenza will be based on nucleic acid amplification through polymerase chain reaction (PCR). At the time of enrollment into the study, a nasopharyngeal swab will be obtained for PCR. The University of Louisville Infectious Diseases Reference Laboratory has extensive experience using molecular techniques for the diagnosis of respiratory pathogens and will test batched specimens at monthly intervals. In addition, we will collect the results from tests done for routine care and bacterial or virus isolates identified during routine care for further characterization. The management of patients in Group A and Group B will be different only in regard to early empiric anti-influenza therapy. All other aspects of the management of these patients will be in compliance with national guideline recommendations from IDSA/ATS (2). Patients in Group A may have antiviral therapy started later in hospitalization or not treated at all. The study will not interfere with Group A patient care. A 1:1 randomization ratio within the two study arms is planned for EOS. A pre-defined randomization chart will be designed in order to have the randomization process Internet-based. The randomization table will be accessible by the project manager as a back up in the event that any problem occurs with the Internet or the computerized system. We will attempt to begin oseltamivir within 8 - 12 hours after hospital admission, and no later than 24 hours. The study nurse will facilitate receipt of early oseltamivir treatment for the consented patient in collaboration with the hospital pharmacies. The time of oseltamivir administration will be recorded for all enrolled patients.