MUNICH – Patients with left-sided endocarditis who are clinically stable after a couple weeks of inpatient intravenous antibiotics may at that point become candidates for discharge on oral antibiotics for the remainder of their treatment course, according to the findings of the randomized, multicenter, Danish POET trial.
“Shifting to oral antibiotic treatment in stabilized patients with endocarditis was as effective and safe as continued intravenous antibiotic treatment and was given during half the antibiotic treatment period. These novel findings may have a significant impact on future clinical practice for the management of patients who are stable,” Henning Bundgård, MD, said at the annual congress of the European Society of Cardiology.
Both ESC and American Heart Association/American College of Cardiology guidelines now recommend treatment of infective endocarditis with intravenous antibiotics for up to 6 weeks. Safely cutting the duration of in-hospital intravenous antibiotics in half is likely to generate major cost savings while improving patient quality of life and avoiding prolonged exposure to the iatrogenic risks inherent to the hospital environment, noted Dr. Bundgård, a cardiologist at Copenhagen University Hospital.
The rationale for the POET (Partial Oral Treatment of Endocarditis) trial was the investigators’ recognition that, even though infectious endocarditis is a feared disease with an in-hospital mortality of 15% or more, the great majority of serious complications occur in the early critical phase of therapy; that is, during the first 10 days or so of inpatient intravenous antibiotic therapy.
“After stabilization, the main reason for staying in the hospital is just to receive IV antibiotics,” Dr. Bundgård noted.
POET included 400 patients with left-sided endocarditis hospitalized at multiple cardiac centers across Denmark, 35% of whom had at least one major comorbid condition. When this reporter observed that this was the smallest study he’d ever seen reported from Denmark, where researchers famously like to utilize interconnected national databases to conduct nationwide observational studies incorporating the country’s entire population, the cardiologist replied, “Denmark is a small country, but we like to make big trials. And this is actually the largest-ever clinical trial in endocarditis, so we are still going big.”
Important to the generalizability of the POET results was the requirement that all 400 participants had to be infected with streptococcus, Staphylococcus aureus, Enterococcus faecalis, or coagulase-negative staphylococci – the major pathogens responsible for three-quarters of all cases of infectious endocarditis.
Once participants were clinically stable after a median of 17 days of intravenous antibiotics, they were randomized to continued in-hospital intravenous antibiotic therapy for a median of another 19 days or to discharge on two oral antibiotics from different classes with different mechanisms of action administered for a median of 17 days, with selection of the oral agents being guided by the results of bacterial susceptibility testing.
The primary outcome was a composite of all-cause mortality, embolic events, unplanned cardiac surgery, and relapse of bacteremia from randomization through 6 months after completion of antibiotic therapy. This occurred in 9.0% of the orally treated group and 12.1% of patients on full-course intravenous therapy for a 28% relative risk reduction, which statistically established the noninferioritiy of the partial oral regimen. The results were similar in patients with native as compared with prosthetic valves, with or without major comorbidities, and in surgically as opposed to conservatively treated patients.
Rates of three of the four components of the composite endpoint were similar in both groups. However, all-cause mortality occurred in 3.5% of the oral therapy group, compared with 6.5% of those on intravenous therapy. Dr. Bundgård said he and his coinvestigators think the disparity in mortality was probably caused by play of chance, although he added that they were struck that four sudden deaths occurred in the intravenous group and none in patients who got oral antibiotics.
Side effects were similarly mild and low frequency in both study arms.
Audience members were eager for details on how the Danish investigators decided patients were clinically stable on intravenous antibiotics and thus ready for randomization, as well as the outpatient follow-up procedures employed in those discharged on oral therapy.
Dr. Bundgård explained that clinical stability required that a patient be afebrile, have C-reactive protein and leukocyte levels less than 35% of their peaks, and needed to have been on intravenous antibiotics for a minimum of 10 days. Moreover, patients who underwent valve surgery during their hospitalization, as did 38% of POET participants, had to wait a minimum of 7 days afterwards before they could be declared clinically stable. Lastly, just prior to randomization all participants underwent transesophageal echocardiography to rule out abscess formation or other valve abnormalities requiring surgery.
Outpatient follow-up required that patients drop in two or three times per week to be checked by a familiar physician or nurse at the hospital ward where they had stayed. Compliance was very good, although it should be noted that only five patients in the POET study were intravenous drug abusers.
Asked why investigators discharged patients on oral therapy rather than on home intravenous antibiotics, Dr. Bundgård explained that home intravenous antibiotic therapy isn’t utilized in Denmark because of the expense and logistic complexity.
Discussant Chris P. Gale, MD, urged care in generalizing the study findings.
“The ‘O’ in POET does not stand for ‘outpatient.’ Outpatients were only selected for oral therapy if they had no heart failure, no emboli, no arrhythmia, no complicating comorbidities, and they were strictly monitored – and frequently. Should we elect to adopt POET into practice, I would recommend strict adherence to the study’s patient selection and monitoring criteria,” said Dr. Gale, a cardiologist at the University of Leeds (England).
The POET results clearly swayed the full-house audience attending the late-breaking Hot Line session in the conference main arena. Immediately before Dr. Bundgård’s presentation, 66% of the audience indicated electronically that they would continue intravenous antibiotics for another 2-4 weeks in a patient with infectious endocarditis who had responded well to 2 weeks of such therapy. After seeing the study results, however, only 19% would still follow that course of action, while 59% of the audience would switch to oral antibiotics and discharge the patient.
Dr. Bundgård reported having no financial conflicts regarding the POET study, which was funded by the Danish Heart Foundation and other research foundations.
Simultaneous with Dr. Bundgård’s presentation in Munich, the POET results were published online by the New England Journal of Medicine (2018 Aug 28. doi: 10.1056/NEJMoa1808312).