A prospective randomized study of inpatient iv. antibiotics for community-acquired pneumonia. The optimal duration of therapy.
Academic Article
Overview
abstract
STUDY OBJECTIVE: To compare therapeutic outcome and perform a cost-benefit analysis of inpatients with community-acquired pneumonia (CAP) treated with a shortened course of i.v. antibiotic therapy. DESIGN: A prospective, randomized, parallel group study with a follow-up period of 28 days. SETTING: Bronx Veterans Affairs Medical Center (VAMC) and the Castle Point VAMC; university-affiliated VAMC general medical wards from September 1993 to March 1995. PATIENTS: Seventy-two male veterans and 1 female veteran with 75 episodes of CAP defined by a new infiltrate on chest radiograph and either history or physical findings consistent with pneumonia. Study population was 42%(31) black, 33%(24) white, and 25%(18) Hispanic. INTERVENTIONS: Patients were randomized (1:1:1) to 1 of 3 treatment groups: group 1 received 2 days of i.v. and 8 days of oral therapy; group 2 received 5 days of i.v. and 5 days of oral therapy; and group 3 received 10 days of i.v. therapy. Antibiotics consisted of cefuroxime, 750 mg every 8 h for the i.v. course, and cefuroxime axetil, 500 mg every 12 h for the oral therapy. MEASUREMENTS AND RESULTS: No differences were found in the clinical course, cure rates, or resolution of chest radiograph abnormalities among the three groups. A significant difference was found in the length of stay (LOS) among the three groups. The mean +/- SD LOS was 6 +/- 3 days in group 1, 8 +/- 2 days in group 2, and 11 +/- 1 days in group 3. The shortened LOS could potentially save $95.5 million for the Department of Veterans Affairs and $2.9 billion for the US private sector. CONCLUSIONS: Adult patients hospitalized for CAP who are not severely ill can be successfully treated with an abbreviated (2-day) course of i.v. antibiotics and then switched to oral therapy. A longer course of i.v. therapy prolongs hospital stay and cost, without improving the therapeutic cure rate.
