MUNICH — Challenging a standard that has been in place since the 1950s, a randomized controlled trial demonstrated that tailored antibiotic therapy for infective endocarditis on the left side of the heart substantially and significantly reduced the duration of antibiotic treatment with acceptable safety and without altering survival. 

“The results from our landmark trial have the potential to change clinical practice,” said principal investigator Henning Bundgaard, MD, DMSc, a clinical professor of cardiology at the Rigshospitalet in Copenhagen, Denmark.

The tailored approach reduced the median duration of treatment to 26 days compared with 41 days in the standard treatment arm, a difference that was statistically (P < .001) and clinically significant, Bundgaard said. 

For the primary endpoint of days alive without antibiotic therapy, the median times were 183 and 169 days (P < .001) for the tailored and standard protocols, respectively.

Tailored Therapy Also Favored for Safety 

At 6 months of follow-up, a primary safety event, defined as an unplanned cardiac surgery or symptomatic embolic event, was tested for noninferiority and was easily met. An event occurred in 8.2% of patients in the tailored therapy group and 10.7% of patients on standard therapy, an absolute reduction of 2.4% (P < .001 for noninferiority), Bundgaard reported.

The results of this phase 3 trial, called POET II, were presented at the European Society of Cardiology Congress 2026 and simultaneously published in The New England Journal of Medicine. 

In POET II, 508 patients with stable infective endocarditis associated with Staphylococcus aureus, Enterococcus faecalis, or Streptococcus species were randomized to receive a response-driven antibiotic therapy after meeting stabilization criteria on at least 2 weeks of antibiotic therapy. 

These stabilization criteria were developed in the first POET trial, published in 2019, which also challenged longstanding infective endocarditis therapy standards. The first POET trial demonstrated that patients could be safely switched from intravenous antibiotics, which were typically maintained for at least 6 weeks at the time of the trial, to oral antibiotics once stabilization criteria were met. 

The main stabilization criterion was a satisfactory clinical response to initial treatment, but other criteria, such as absence of abscess formation or valve abnormalities on transesophageal echocardiography, also had to be met.

Building on the positive results of the first trial, for POET II, “we hypothesized that tailoring antibiotic therapy in those who have an initial positive response could safely reduce the duration of antibiotics,” said Bundgaard, who participated in both studies.

In POET II, treatment courses for uncomplicated S aureus were reduced from the 4 weeks in the standard regimens created by historical observation to 2 weeks after patients met stabilization criteria. Treatment courses for complicated S aureus were reduced from 6 weeks to 4 weeks. The same reductions, respectively, were applied to uncomplicated and complicated Streptococcus species infections. Treatment courses for E faecalis infections, whether uncomplicated or complicated, were reduced from 6 weeks to 4 weeks.

Stabilization Criteria

As in the first POET trial, the stabilization criteria applied in POET II appear to be key to safely reducing the time on intravenous antibiotics, Bundgaard said.

“The post-stabilization criteria appear to identify a turning point beyond which antibiotic treatment can be safely de-escalated,” he specified.

Based on these studies, Bundgaard estimated that approximately 50% of patients with infective endocarditis may be candidates for a tailored antibiotic regimen.

Although there was a safety advantage for the tailored therapy overall, no significant advantage was seen for any of the three major components of the safety endpoint, which were unplanned cardiac surgery, embolic events, or all-cause mortality. 

Relapse of bacteremia or endocarditis was significantly higher in the tailored therapy group (5.1% vs 1.6%; P = .04). When this outcome was added to the safety analysis, a slight numerical advantage of tailored therapy was maintained (-0.1%) and noninferiority remained significant (P = .02).

Invited discussant Gilbert Habib, MD, PhD, chair of the cardiology department at La Timone Hospital in Marseille, France, agreed that the protocol can be considered a new standard with some limitations. 

He cautioned that the results seen in POET II are probably dependent on strict adherence to stabilization criteria and “a highly structured outpatient antibiotic service to manage these patients.” 

He also noted that only 20% of those enrolled in the study had an E faecalis infection, which is relatively challenging, and he called for further study of whether the shorter duration of antibiotic treatment is safe for this infection. He also called for further study of the higher relapse rate even though the clinical consequences in this study appeared to be relatively “minor,” he said.

Despite these limitations, he agreed with the premise that POET II defines a new standard for shorter courses of therapy that can be expected to lead to reduced length of hospital stay, hospital-related complications, and costs when used appropriately. 

“Antibiotic therapy for infective endocarditis should move from fixed-duration therapy to tailored treatment duration,” he said, but emphasized the approach is relevant only when employed at experienced centers with a dedicated endocarditis team.

Bundgaard has reported financial relationships with Amgen, Bristol Myers Squibb, General Electric, MSD, Novo Nordisk, and Pfizer. Habib has reported no relevant financial relationships.

Ted Bosworth, a career medical writer based in New York City, has been covering advances in clinical medicine, including cardiology, for several decades.