1467. Clinical Significance of Microbiologic Treatment Failure Following Clinical Cure of Pneumonia

Abstract

BackgroundMicrobiologic cure is a frequent outcome in pneumonia trials, but its clinical relevance remains poorly understood. We aimed to evaluate the association between microbiologic failure in pneumonia in the setting of clinical cure and recurrent pneumonia and mortality.MethodsRetrospective, single-center cohort study of adult patients hospitalized between January 1, 2008 and January 1, 2017. Patients with index pneumonia (defined as a positive respiratory culture and continuous receipt of ≥5 days of antibiotics) who demonstrated clinical cure (defined as cessation of all antibiotics for ≥48 hours and survival for 7 days following antibiotic completion) were included. All included patients had to have follow-up respiratory cultures obtained between 3 days before and 7 days after completion of antibiotic therapy. Patients with persistence of the inciting pathogen were classified as microbiologic failure and all others as microbiologic cure. Primary outcomes were 30-day mortality and a 30-day composite of mortality and/or recurrent pneumonia.ResultsOf 376 included patients, 61% had microbiologic cure compared with 39% with microbiologic failure. Mean age was 55.4 years, 62% of patients were male and 79% White. Mean antibiotic duration was 14.8 days. Sixty-one percent of patients were mechanically ventilated at the time of index pneumonia. The most common pathogens were Enterobacteriaceae (35%), Staphylococcus aureus (31%) and Pseudomonas aeruginosa (22%). In the microbiologic failure group, the primary composite outcome occurred in 18.9% of patients compared with 11.8% in the microbiologic cure group (OR 1.73, 95% CI 0.98–3.09) (Figure 1). All-cause 30-day mortality was greater in patients with microbiologic failure (16.2%) than microbiologic cure (8.3%) (OR 2.13, 95% CI 1.12–4.04). These associations were particularly strong among nonventilated patients (Figure 2). Rates of recurrent pneumonia were similar between groups.ConclusionPatients with clinical cure but microbiologic failure were more than twofold more likely to die 30 days after their index pneumonia than those with microbiologic cure. Patients with microbiologic failure who were not mechanically ventilated at the time of index pneumonia had a greater than fourfold increased mortality risk. Disclosures K. Kaye, Zavante Therapeutics, Inc.: Scientific Advisor, Consulting fee