Switch from parenteral to oral antibiotics for brain abscesses: a retrospective cohort study of 109 patients Antoine Asquier-Khati 1 *, Colin Deschanvres 1 , David Boutoille 1 , Maeva Lefebvre 1 , Paul Le Turnier 1 , Benjamin Gaborit 1 , Karim Lakhal 2 , Kevin Buffenoir 3 , Lydie Khatchatourian 4 and Nathalie Asseray 1 on behalf of the Nantes Brain Abscesses study group† 1 Infectious Diseases, CHU Hotel Dieu, Nantes, France; 2 Intensive Care Unit, CHU Lae ¨ nnec, Nantes, France; 3 Neurosurgery, CHU Hotel Dieu, Nantes, France; 4 Infectious Diseases, CH Cornouaille, Quimper, France *Corresponding author. E-mail: antoine.asquierkhati@chu-nantes.fr †Members are listed in the Acknowledgements section. Received 31 January 2020; accepted 2 June 2020 Objectives: Brain abscess is one of the most serious diseases of the CNS and is associated with high morbidity and mortality. With regard to the lack of data supporting an optimal therapeutic strategy, this study aimed to ex- plore the prognostic factors of brain abscess, putting emphasis on the impact of therapeutic decisions. Methods: We retrospectively included patients hospitalized for brain abscess during a period of 13years. Comorbidities (Charlson scale), clinical presentation, microbiology culture, radiological features and therapeutic management were collected. Glasgow Outcome Scale (GOS) at 3months and length of hospital stay were, re- spectively, the main and the secondary outcomes. Logistic regression was used to determine factors associated with outcome independently. Results: Initial Glasgow Coma Scale (GCS) 14 and comorbidities (Charlson scale 2) were associated with poor neurological outcome while oral antibiotic switch was associated with better neurological outcome. Oral switch did not appear to be associated with an unfavourable evolution in the subset of patients without initial neuro- logical severity (GCS >14) on admission. Duration of IV regimen and time to oral switch were associated with the length of inpatient stay. Conclusions: This study confirms the role of GCS and comorbidities as prognostic factors and presents reassuring data regarding the safety of oral switch for the antibiotic treatment of brain abscesses. Oral switch could prevent catheter-induced iatrogenic complications and allow a higher quality of life for patients. Introduction Pyogenic brain abscesses are focal bacterial infections of brain par- enchyma and remain one of the most serious diseases of the CNS. 1,2 Even though prognosis has improved over recent decades, 3 morbidity and mortality are still high. 2,4 Alteration of conscious- ness at initial presentation is clearly identified as a poor prognostic factor, 5–8 as well as underlying conditions. Nevertheless, no medic- al decision, including antibiotic choice, has proven its impact on prognosis. Currently no international guidelines are available for the treat- ment of this serious infection. The data available are based on clin- ical practices published several years ago by two learned societies, 9,10 with a reminder that due to the paucity of data, the possibility and modalities of oral switch could not be the subject of strong recommendations. Local pharmacological conditions remain an issue for antibiotic treatment since brain penetration of drugs is known to be uncertain. This study aimed to explore the prognostic factors of brain abscess, putting an emphasis on oral switch. Methods We retrospectively reviewed all adult patients hospitalized in Nantes University Hospital for pyogenic brain abscess from March 2003 to December 2016. Inclusion criteria required: (i) a characteristic collection vis- ible on cranial imaging; and (ii) a bacterial causality of infection confirmed by appropriate microbiological testing or clinically diagnosed and treated with antibiotic agents. We excluded infections caused by mycobacteria, fungi or parasites. The clinical course of each patient was collected from the hospital computerized charts until 1 year of follow-up. Oral switch was defined as an exclusive oral therapy after initial IV regimen. The main out- come measure was the Glasgow Outcome Scale (GOS) at 3 months post- V C The Author(s) 2020. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. 3062 J Antimicrob Chemother 2020; 75: 3062–3066 doi:10.1093/jac/dkaa285 Advance Access publication 22 July 2020 Downloaded from https://academic.oup.com/jac/article/75/10/3062/5875229 by guest on 25 October 2022