A study looked at the utilization of the technology with this antibiotic class in hospital-acquired pneumonia (HAP) and ventilator-associated pneumonia (VAP).
In the intensive care unit, HAP and VAP are the most common infections.
An investigative team at the UF Health Shands Hospital performed a retrospective cohort study, utilizing machine learning to define the impact of beta-lactam early and cumulative PK/PD target attainment on ICU patients with these 2 forms of pneumonia.
For the study, adult ICU patients who received cefepime, meropenem, or piperacillin/tazobactam for HAP or VAP and had its concentration measured were included. Using PK models, beta-lactam administration was given for every patient.
According to the investigators, calculated exposure were time the free concentration remained above the minimum inhibitory concentration (fT>MIC) and four multiples of MIC (fT>4xMIC) during the first 24 hours, 0-10 days, and 0-end of therapy.
The investigators evaluated clinical cure, mechanical ventilation (MV)-free days and the composite outcome of 28-day survival and clinical cure. According to the investigators, regression analyses and machine learning methods, including clinical important covariates, was performed to evaluate the exposure-outcome relationship.
“Early and cumulative target attainment have great impact on pneumonia outcomes,” the investigators wrote. “Beta-lactam exposure should be optimized early and maintained through therapy duration.”
The study, “Using machine learning to define the impact of beta-lactam early and cumulative PK/PD target attainment on outcomes in ICU patients with hospital-acquired and ventilator-associated pneumonia,” was presented at the 24th Annual Making a Difference in Infectious Disease (MAD-ID) Meeting 2022, in Orlando, Florida from May 18-21.
Contagion spoke to study coauthor Mohammad Alshaer PharmD, PhD, research assistant professor, Infectious Disease Pharmacokinetics Laboratory College of Pharmacy, and Emerging Pathogens Institute University of Florida, at the meeting who provided insights on the study, and the bigger role of machine learning in the clinical setting.