Patients in the intensive care unit (ICU) are often very ill and require more care than patients in other hospital wards. Antibiotics are prescribed often to these patients to treat infections and the most important subgroup are beta-lactam antibiotics and fluoroquinolones. However, when we and other researchers measure these antibiotic concentrations in plasma, 40% of the patients do not achieve the required concentrations. Not achieving these concentrations is linked to less clinical cure, less bacteriological eradication, and increased antimicrobial resistance formation. Therefore, we need to find ways to optimize the dosing of these antibiotics in ICU patients.
An often-proposed method is using therapeutic drug monitoring (TDM). TDM is personalizing the dosing of drugs based on measurements, such as the drug concentrations in the plasma. Since there are limitations to TDM, such as having to wait for steady-state concentrations of the antibiotic levels, models are often used to estimate if a patient will reach the required target concentrations. This is called model-informed precision dosing (MIPD).
Using this information, our research team conducted the DOLPHIN trial. After a patient or their family gave informed consent to participate in the trial, we analyzed their antibiotic concentrations in the plasma. In the patients that were randomized to the MIPD treatment, we used MIPD-software to estimate if they would reach the target and advice a dose adjustment to the treating physician. The first time we conducted this intervention was at 24-36 hours after the first antibiotic dose was given. We repeated this intervention every 48 hours until the 5th day of treatment.
Conducting this trial required high levels of cooperation from the eight study centers: After a patient was included, samples were drawn in the early morning, and these samples needed to be transported before noon to our laboratory for analysis. The analysts had to analyze these samples that afternoon, using a relatively slow, but precise, method. After receiving the antibiotic concentrations, patient characteristics, dosing data, and the measured concentrations needed to be entered into the MIPD-software to provide the appropriate advice. All links had to be connected, and due to the excellent planning and motivation of all research centers, it all worked out.
We did not observe a difference in the duration of ICU admission between the group that received the intervention and those that received standard therapy, nor in any other clinical outcome. We also did not find a difference in target attainment between the two groups. Some interesting observations were made from the data we gathered.
We were limited to the maximum doses under which the antibiotics were registered. Even though patients did not reach the target with these maximum doses, we could not increase the dose. This leads to the conclusion that higher doses are needed to treat patients in the ICU.
Another interesting observation was that traditional dose adjustments were not sufficient to reach the required targets in this population. Traditionally, doses are increased or decreased conservatively which means that a maximum of 25% to 50% change in dosing was advised. More vigorous dose adjustments seem to be required to get patients on target.
ICU patients are very heterogeneous, which means that patients are admitted with a wide variety of diagnoses, organ failure, and treatment requirements. Finding groups of patients that might benefit from MIPD may be required. We are still investigating the subgroups, such as those at risk for not reaching the target, patients with a certain organ failure status, or those that had a short amount of time between the first antibiotic dose and receiving the intervention.
This trial might not have shown a result in favor of MIPD in all ICU patients. However, very important lessons can be drawn, and interesting post-hoc analyses await. The classic approach to dose personalization is not fit for all patients. More questions need to be answered to personalize the dosing of antibiotics in the ICU: who to TDM, when to TDM, and how to TDM.
Please sign in or register for FREE
If you are a registered user on Research Communities by Springer Nature, please sign in