Background:
There is limited evidence to guide the management of patients with life-threatening invasive gram-positive infections, such as toxic shock syndrome and necrotizing fasciitis, caused by Streptococcus pyogenes and Staphylococcus aureus. While adjunctive clindamycin is commonly used to inhibit bacterial toxin synthesis, emerging observational data suggest linezolid as a promising alternative with much lower resistance rates and reduced adverse effects. Seemingly insurmountable design challenges have prevented definitive head-to-head comparisons. Recent clinical trial innovations, such as waiver of consent for electronic health record (EHR)-based randomisation, may overcome key barriers to timely enrolment.
Methods:
We are undertaking a retrospective analysis using 5 years of real-world data from 2 tertiary hospitals in Melbourne, Australia, to inform a planned pragmatic multicentre randomised controlled trial (RCT) comparing adjunctive clindamycin and linezolid. Potential EHR-based structured data triggers for randomisation will be evaluated for their ability to identify paediatric and adult patients with suspected invasive gram-positive toxin-mediated infections. Sensitivity, specificity, and predictive values of these triggers will be assessed at various time points within 48 hours of emergency department presentation.
Results:
This study will optimise EHR-based randomisation processes and assess their feasibility for real-time patient identification and recruitment. Insights from this analysis will guide the design of a pilot RCT, refine recruitment strategies, and improve treatment allocation.
Conclusion:
The preliminary retrospective analysis will assess EHR-based randomisation processes to inform design of a pilot RCT, paving the way for a full-scale trial to address enduring knowledge gaps in the management of difficult-to-study and challenging-to-treat life-threatening toxin-mediated infections.