Association of early antibiotic therapy and in-hospital mortality in adult mild-to-moderate acute aspiration pneumonitis: a cohort study

• Published in: Internal and emergency medicine 2021-10, Vol.16 (7), p.1841-1848
• Main Authors: Aga, Masaharu, Naganuma, Toru, Ohashi, Yusuke, Matsuzawa, Hiroki, Matsuzaka, Suguru, Cammack, Ivor, Yamada, Gen, Serizawa, Yoshimoto
• Format: Article
• Language: English
• Subjects: Antibiotics; Antimicrobial agents; Cohort analysis; Diagnosis; Electronic medical records; Emergency medical care; Fever; Im - Original; Inflammation; Internal Medicine; Length of stay; Medicine; Medicine & Public Health; Mortality; Patients; Pneumonitis; Statistical analysis
• ISSN: 1828-0447; 1970-9366
• DOI: 10.1007/s11739-021-02695-y

Background Patients with aspiration pneumonitis often receive empiric antibiotic therapy despite it being due to a non-infectious, inflammatory response. Objective To study the benefits of early antibiotic therapy in patients with suspected aspiration pneumonitis in an acute care hospital. Design Retrospective cohort study using electronic medical records from Teine Keijinkai Hospital. Participants Adults aged over 18 years admitted with a diagnosis of aspiration pneumonitis to the Department of General Internal Medicine or Emergency Department between January 1, 2008, and May 31, 2019. A diagnosis of aspiration pneumonitis was defined as a documented macro-aspiration event and a chest radiograph demonstrating new radiographic infiltrates. Main measures Patients were classified into the “early antibiotic treatment” group and the “no or late treatment” group depending on whether they received antibiotic therapy for respiratory bacterial pathogens within 8 h of arrival. The primary outcome was in-hospital all-cause mortality. Secondary outcomes included length of hospital stay, antibiotic-free days, duration of fever, readmission within one month, and incidence of complications. Key results Of the 146 patients enrolled, 52 (35.6%) did not receive early antibiotic therapy, while the remaining 94 (64.4%) did. There was no difference in in-hospital mortality rates between the groups after adjustment for potential confounding variables using Cox proportional hazards analysis (hazard ratio 2.78; 95% confidence interval, 0.57–13.50, p  = 0.20). Patients in the no or late treatment group had more antibiotic-free days ( p