A MacGowan1, S Grier2, M Stoddart2, R Reynolds3, C Rogers4, K Pike4, H Smartt4, M Wilcox5, P Wilson6, M Kelsey7, J Steer8, F K Gould9, J D Perry9, R Howe10, M Wootton10. 1. Department of Pathology Sciences, North Bristol NHS Trust, Southmead Hospital, Bristol, UK. Electronic address: alasdair.macgowan@nbt.nhs.uk. 2. Department of Pathology Sciences, North Bristol NHS Trust, Southmead Hospital, Bristol, UK. 3. Population Health Sciences, Bristol Medical School, Bristol, UK. 4. Bristol Royal Infirmary, Clinical Support Unit, Bristol, UK. 5. Department of Microbiology, Leeds General Infirmary, Leeds Teaching Hospitals NHS Trust, Leeds, UK. 6. Department of Clinical Microbiology, UCLH NHS Foundation Trust, London, UK. 7. Department of Medical Microbiology, Whittington Hospital, Whittington NHS Trust, London, UK. 8. Department of Microbiology, Derriford Hospital, University Hospitals Plymouth NHS Trust, Plymouth, UK. 9. Department of Medical Microbiology, Freeman Hospital, Newcastle Upon Tyne Hospitals NHS Foundation Trust, Newcastle Upon Tyne, England, UK. 10. Department of Microbiology, Public Health Wales, Cardiff University Hospital of Wales, Cardiff, Wales, UK.
Abstract
OBJECTIVES: Bloodstream infection has a high mortality rate. It is not clear whether laboratory-based rapid identification of the organisms involved would improve outcome. METHODS: The RAPIDO trial was an open parallel-group multicentre randomized controlled trial. We tested all positive blood cultures from hospitalized adults by conventional methods of microbial identification and those from patients randomized (1:1) to rapid diagnosis in addition to matrix-assisted desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) performed directly on positive blood cultures. The only primary outcome was 28-day mortality. Clinical advice on patient management was provided to members of both groups by infection specialists. RESULTS: First positive blood culture samples from 8628 patients were randomized, 4312 into rapid diagnosis and 4136 into conventional diagnosis. After prespecified postrandomization exclusions, 2740 in the rapid diagnosis arm and 2810 in the conventional arm were included in the mortality analysis. There was no significant difference in 28-day survival (81.5% 2233/2740 rapid vs. 82.3% 2313/2810 conventional; hazard ratio 1.05, 95% confidence interval 0.93-1.19, p 0.42). Microbial identification was quicker in the rapid diagnosis group (median (interquartile range) 38.5 (26.7-50.3) hours after blood sampling vs. 50.3 (47.1-72.9) hours after blood sampling, p < 0.01), but times to effective antimicrobial therapy were no shorter (respectively median (interquartile range) 24 (2-78) hours vs. 13 (2-69) hours). There were no significant differences in 7-day mortality or total antibiotic consumption; times to resolution of fever, discharge from hospital or de-escalation of broad-spectrum therapy or 28-day Clostridioides difficile incidence. CONCLUSIONS: Rapid identification of bloodstream pathogens by MALDI-TOF MS in this trial did not reduce patient mortality despite delivering laboratory data to clinicians sooner.
RCT Entities:
OBJECTIVES: Bloodstream infection has a high mortality rate. It is not clear whether laboratory-based rapid identification of the organisms involved would improve outcome. METHODS: The RAPIDO trial was an open parallel-group multicentre randomized controlled trial. We tested all positive blood cultures from hospitalized adults by conventional methods of microbial identification and those from patients randomized (1:1) to rapid diagnosis in addition to matrix-assisted desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) performed directly on positive blood cultures. The only primary outcome was 28-day mortality. Clinical advice on patient management was provided to members of both groups by infection specialists. RESULTS: First positive blood culture samples from 8628 patients were randomized, 4312 into rapid diagnosis and 4136 into conventional diagnosis. After prespecified postrandomization exclusions, 2740 in the rapid diagnosis arm and 2810 in the conventional arm were included in the mortality analysis. There was no significant difference in 28-day survival (81.5% 2233/2740 rapid vs. 82.3% 2313/2810 conventional; hazard ratio 1.05, 95% confidence interval 0.93-1.19, p 0.42). Microbial identification was quicker in the rapid diagnosis group (median (interquartile range) 38.5 (26.7-50.3) hours after blood sampling vs. 50.3 (47.1-72.9) hours after blood sampling, p < 0.01), but times to effective antimicrobial therapy were no shorter (respectively median (interquartile range) 24 (2-78) hours vs. 13 (2-69) hours). There were no significant differences in 7-day mortality or total antibiotic consumption; times to resolution of fever, discharge from hospital or de-escalation of broad-spectrum therapy or 28-day Clostridioides difficile incidence. CONCLUSIONS: Rapid identification of bloodstream pathogens by MALDI-TOF MS in this trial did not reduce patient mortality despite delivering laboratory data to clinicians sooner.