Switching from IV to oral antibiotics effective, safe for infective endocarditis

Roshini Claire Anthony
21 Mar 2019
Professor Henning Bundgaard

Long-term results of the POET* trial have suggested that patients with left-sided infective endocarditis (IE) could safely switch from conventional intravenous (IV) antibiotics to oral antibiotics, with oral antibiotics associated with a reduced mortality risk.

The results demonstrated that oral antibiotics may be as effective as IV antibiotics in certain stabilized patients, previously in a short term and now with longer-term follow-up, said lead author Professor Henning Bundgaard from Copenhagen University Hospital, Copenhagen, Denmark.

The Danish population-based POET study included 400 adult patients (mean age 67 years, 77 percent male) with left-sided IE caused by Staphylococcus aureus (n=87), Enterococcus faecalis (n=97), streptococci (n=196), or coagulase-negative staphylococci (n=23) who were deemed stabilized after receiving IV antibiotics for 10 days (or ≥7 days after valve surgery). They were then randomized to either continue receiving IV antibiotics (n=199) or swap to oral antibiotics (n=201). The oral regimen comprised two antibiotics of different classes, mechanisms, and metabolizations, and with moderate-to-high bioavailability.

A total of 107 and 293 patients had prosthetic and native valves, respectively, and 152 patients underwent valve surgery compared with 248 who received conservative treatment.

The initial follow-up at 6 months showed that partial treatment with oral antibiotics was noninferior to IV antibiotics in terms of a composite of all-cause mortality, unplanned cardiac surgery, embolic events, and relapse of bacteraemia with the primary pathogen (n=18 vs 24; difference, 3.1 percent, 95 percent confidence interval [CI], -3.4 to 9.6 percent; p=0.40 [noninferiority met]). [N Engl J Med 2019;380:415-424]

In this update taking place after a median 3.5-year follow-up period, 53 and 76 patients on oral and IV antibiotics, respectively, experienced a primary outcome event, with a lower risk among those who received oral antibiotics (26.4 percent vs 38.2 percent, hazard ratio [HR], 0.64, 95 percent CI, 0.45–0.91; p=0.01). [ACC.19, abstract 405-12]

This result was primarily driven by the lower risk of all-cause mortality in patients on oral vs IV antibiotics (16.4 percent [n=33] vs 27.1 percent [n=54], HR, 0.57, 95 percent CI, 0.37–0.87; p=0.009), as the incidences of embolic events (2.5 percent in each group; p=0.8), unplanned surgery (6.0 percent vs 9.5 percent; p=0.2), or bacteraemia relapse (4.0 percent vs 5.5 percent; p=0.4) were comparable between oral and IV antibiotic recipients.

These results were consistent among patients with native or prosthetic valves, and regardless of whether they were surgically or conservatively treated, said Bundgaard.

Cardiovascular causes and infections were the most common causes of death among IV antibiotic recipients (10.6 and 7.0 percent, respectively), while infection was the most common among oral antibiotic recipients (5.0 percent).

Patients with IE are usually treated in-hospital with IV antibiotics for up to 6 weeks, said Bundgaard. While the primary reason patients remain in hospital is to receive IV antibiotics, hospitalization is linked to further complications, he said.

In this study, patients who continued to receive IV antibiotics remained hospitalized for the duration of their treatment (median, 19 days post-randomization), while those who received oral antibiotics were discharged from hospital a median 3 days post-randomization (p<0.001).

“This new treatment may halve the hospital stay for patients with a heart valve infection,” said Bundgaard. “Oral antibiotics may be safely administered during approximately half of the recommended antibiotic treatment period and potentially as outpatient treatment … These findings clearly support a change in the standard of care for this condition,” he said.

However, he added that these results may not pertain to IE caused by bacteria species different from those included in the trial. Furthermore, none of the patients had antibiotic-resistant bacteria, a factor which could affect the findings.


Editor's Recommendations