By Stan Deresinski, MD, FACP, FIDSA
Clinical Professor of Medicine, Stanford University
SYNOPSIS: Pharmacodynamic/pharmacokinetic analysis of patients in the POET study provides understanding of the efficacy of intravenous-to-oral stepdown antibiotic therapy in patients with endocarditis.
SOURCE: Bock M, Theut AM, van Hasselt JGC, et al. Attainment of target antibiotic levels by oral treatment of left-sided infective endocarditis: A POET Substudy. Clin Infect Dis 2023;77:242-251.
The POET study, published five years ago, compared intravenous (IV) to oral administration of antibiotics in 400 patients with left-sided endocarditis due to streptococci, Enterococcus faecalis, Staphylococcus aureus, or coagulase-negative staphylococci (CNS) in a randomly assigned trial.1,2 All patients assigned to oral therapy received two antibiotics, each with different mechanisms of action and different “metabolization processes.” All patients received at least 10 days of IV antibiotics prior to randomization. There was no significant difference in the composite primary outcomes at six months.
To better understand the use of oral antibiotic therapy in endocarditis, Bock and colleagues performed a subset analysis of the pharmacokinetic/pharmacodynamic attainment (PTA) achieved with oral antibiotic administration in these patients, supplementing the much more limited data presented in the original paper. This subset involved 236 patients with targeted sampling on days 1 and 5, and it included 175 who had received two oral antibiotics. In each case, the probability of target attainment in relation to both the EUCAST breakpoint and the organism minimum inhibitory concentration (MIC) determined by Etest was calculated. This involved examination of the time the free (non-protein bound) antibiotic concentration was above the MIC or breakpoint for β-lactams and the total drug above the target area under the curve (AUC)/MIC for linezolid and rifampin.
For simplicity, this discussion will review only the results relative to the organism MIC, not the breakpoint. The amoxicillin PTA for E. faecalis on day 1 was 97%, while that for streptococci was 100% and, at day 5, it was 100% for both. Linezolid achieved day 1 PTAs of 88%, 90%, and 100% for E. faecalis, staphylococci, and streptococci, respectively, and 90%, 100%, and 92%, respectively, at day 5. Data from a very small number of patients given linezolid intravenously did not demonstrate an advantage of this route.
The one- and five-day PTAs of moxifloxacin were 81% and 79% for enterococci, 100% and 100% for staphylococci, and 75% and 81% for streptococci. As with linezolid, data from a very small number of patients found that intravenous administration did not appear to improve these results. With rifampin, the one- and five-day PTAs for staphylococci were each 100%, while for streptococci, they were 78% and 71%, respectively. The result with dicloxacillin against staphylococci was disappointing: 9% and 17% at one- and five-day sampling and, in nine patients given dicloxacillin IV, the PTA was 56%.
Data were available for each antibiotic of a combination in a subset of the patients. Of these, 13/74 (18%) reached the target PTA for only one of the antibiotics at day 1 and 14/50 (22%) did so at day 5. Only one patient failed to reach target for each of the antibiotics they were receiving.
COMMENTARY
The POET study provided strong evidence for the efficacy of the use of oral antibiotics in the follow-on treatment (after at least 10 days of IV administration) of endocarditis due to four groups of susceptible organisms. This substudy of POET provides the pharmacokinetic/pharmacodynamic rationale for this approach.
In POET, patients assigned to oral antibiotic therapy received combinations of two agents. While the PTA data discussed here suggest that the use of two antibiotics in combination may be preferable to monotherapy, I suspect that monotherapy with highly active, high-bioavailability agents would be just as effective. It should be noted that, although the data were extremely limited, IV antibiotics did not provide an obvious pharmacodynamic advantage over orally administered ones.
The low PTA achievement with dicloxacillin is the result of its protein binding, which is reported to be as high as 98%. Whether this affects outcomes is uncertain, but its use should give pause.
To overcome concern about the use of oral antibiotics in this context, therapeutic drug monitoring with consequent dosing adjustment, if indicated, would be ideal, but this is not readily available in a timely fashion. As the investigators pointed out, the dosing in POET was fixed and there was no adjustment for body weight. They also pointed out that administration of a loading dose might achieve PTA more rapidly.
Although study entry required having received at least 10 days of antibiotic therapy, the median time to randomization in each study arm was 17 days. I believe this duration prior to the switch to oral antibiotic administration is likely to be excessive in many patients. It is likely that many cases of infection with highly susceptible organisms may have been cured at the time of randomization.
REFERENCES
- Iversen K, Ihlemann N, Gill SU, et al. Partial oral versus intravenous antibiotic treatment of endocarditis. N Engl J Med 2019;380:415-424.
- Deresinski S. IV to oral antibiotic switch for selected cases of endocarditis. Infectious Disease Alert 2018;38:1.
