POET-IIEndocarditisNew England Journal of Medicine
Response-Tailored Versus Standard Antibiotic Therapy for Left-Sided Infective Endocarditis
Response-Tailored or Standard-Duration Antibiotic Treatment for Infective Endocarditis
View Full PaperWhy the trial?
Left-sided infective endocarditis is treated with fixed 4-6 week antibiotic courses set by convention rather than individual response. Whether tailoring treatment duration to each patient's clinical and microbiological response can reduce antibiotic exposure without compromising safety had not been tested in a randomised trial.
Does response-tailored antibiotic therapy improve days alive without antibiotics and maintain safety compared to standard-duration therapy in patients with left-sided infective endocarditis?
Population
508 stabilized patients with left-sided infective endocarditis after 2-4 weeks of therapy
Comparison
Response-tailored antibiotic discontinuation vs standard 4-6 weeks of therapy
Design
Randomized trial: superiority for efficacy, noninferiority for safety (7.5 pp margin)
Follow-up
6 months after randomization
Key result
Response-tailored antibiotic therapy increased days alive without antibiotics (difference 13 days) and was noninferior to standard therapy for safety (P<0.001), but increased relapse (5.1% vs 1.6%).
Authors
Discussion
Key questions
Member takes
Where experts stand
Experts see POET-II as a credible signal that shorter, response-tailored antibiotic therapy can work in stable left-sided endocarditis, but the higher relapse rate tempers enthusiasm and makes careful patient selection the central question.
Most clinicians welcome the finding that tailored shorter antibiotic courses met noninferiority for the main safety composite, and the trial investigator calls it potentially practice-changing. However, the higher relapse rate in the shorter-treatment arm is a consistent concern, and experts stress that only carefully selected, stable patients should be considered. The live question is whether guidelines will adopt this approach and how clinicians should weigh the trade-off between reduced antibiotic burden and increased relapse risk.
Agreement
Multiple clinicians agree that while the tailored strategy met noninferiority for the safety composite, the higher relapse rate is a real concern that demands careful patient selection and follow-up.
2 clinicians say this directly
What they’re arguing about
supportiveneutralcautiouscritical
Counts are expert takes we classified by axis. Tap a row to see the takes behind its count.
Still unclear
Whether the higher relapse rate will limit guideline uptake or be accepted as a manageable trade-off in selected patients remains unresolved. Experts have not yet addressed which subgroups (by organism or clinical stability) might benefit most or face the greatest relapse risk. How this strategy integrates with the oral step-down approach from the original POET trial is also undefined.
Key expert perspectives
HBHenning BundgaardCardiologist, Rigshospitalet - Copenhagen University HospitalPractice takeAug 30
Results could change practice for roughly half of left-sided endocarditis patients
As the trial investigator, argues the results have the potential to change clinical practice, estimating that about half of patients with left-sided infective endocarditis could be eligible for tailored reduced-duration therapy. Notes that shorter courses may also improve quality of life by reducing prolonged treatment and hospitalisations.
Shorter antibiotic courses are not free: relapse was higher
Highlights that while the tailored strategy beat standard therapy on days alive off antibiotics and met noninferiority for safety, the relapse rate was notably higher at 5.1% vs 1.6%. Concludes that "shorter isn't free."
Distilled from their postX post
Current 6-week recommendation rests on expert opinion, not randomized evidence
Notes that the longstanding recommendation of up to 6 weeks of antibiotics for left-sided endocarditis is based largely on expert opinion and historical observation, and that POET II tested a shorter, response-tailored alternative.
Overview
Supports response-tailored shorter antibiotics in left-sided IE with noninferior safety; extends RCT evidence but requires relapse monitoring.
Key Points
- To determine whether a clinical response-tailored antibiotic regimen can safely shorten treatment duration compared with standard 4-to-6-week therapy in patients with left-sided infective endocarditis.
- Randomized trial (NCT03851575) of 508 stabilized patients with left-sided infective endocarditis who completed 2 to 4 weeks of initial antibiotic treatment.
- Participants were randomized to either discontinue antibiotics immediately (tailored therapy, n=255) or continue standard antibiotic therapy for a total duration of 4 to 6 weeks (n=253), with 6 months of follow-up.
- The primary efficacy endpoint was days alive without antibiotics (tested for superiority); the primary safety composite endpoint was all-cause mortality, unplanned cardiac surgery, or embolic events (tested for noninferiority, 7.5 percentage-point margin).
- The median days alive without antibiotics was 183 days (IQR, 181 to 183) in the tailored group versus 169 days (IQR, 166 to 171) in the standard group (Hodges-Lehmann difference, 13 days; 95% CI, 12 to 13; P<0.001 for superiority).
- Primary safety endpoint events occurred in 21 of 255 patients (8.2%) assigned to tailored therapy versus 27 of 253 patients (10.7%) assigned to standard therapy (absolute difference, -2.4 percentage points; 95% CI, -7.7 to 2.7; P<0.001 for noninferiority).
- Relapse of bacteremia or infective endocarditis occurred in 13 patients (5.1%) in the tailored-therapy group compared with 4 patients (1.6%) in the standard-therapy group (P=0.04).
Evidence details
What drove the result?
| Outcome | Tailored | Standard |
|---|---|---|
| Median days alive without antibiotics | 183 | 169 |
| Difference 13 days (95% CI 12-13); P<0.001 for superiority | ||
| Relapse of bacteremia or endocarditis | 13 (5.1%) | 4 (1.6%) |
| Key secondary - higher with tailored therapy (P=0.04); opposite direction from the efficacy gain | ||
Limitations & tradeoffs
Safety
Death, unplanned cardiac surgery, or embolic events (primary safety) 21/255 (8.2%) vs 27/253 (10.7%), noninferior (difference -2.4 pp, 95% CI -7.7 to 2.7; P<0.001) — but relapse of bacteremia or endocarditis was more frequent when therapy was shortened (5.1% vs 1.6%).
Representation
applies only to patients already stabilized after 2-4 weeks of antibiotics, not to acute presentations.
Design limitations
the 6-month follow-up may miss late relapses.
Structured PICO
Does response-tailored antibiotic therapy improve days alive without antibiotics and maintain safety compared to standard-duration therapy in patients with left-sided infective endocarditis?
P
Population
508 patients with left-sided infective endocarditis who had received 2 to 4 weeks of therapy and were clinically stable, followed for 6 months.
I
Intervention
Response-tailored antibiotic therapy (discontinuation of antibiotics after initial 2 to 4 weeks of therapy)
C
Comparator
Standard-duration antibiotic therapy (continued standard treatment for a total duration of 4 to 6 weeks)
O
Outcome
Primary efficacy end point: days alive without antibiotic treatment for infective endocarditis or bacteremia within 6 months after randomization. Primary safety end point: composite of death from any cause, unplanned cardiac surgery, or symptomatic embolic events within 6 months after randomization.composite
In stable patients with left-sided infective endocarditis, a response-tailored antibiotic strategy shortened treatment duration and was noninferior for safety, but increased the risk of relapse.
Main Result
Mean Difference: 13 (95% CI 12–13)
Absolute Event Rate: 183% vs 169%
p-value: p=<0.001 for superiority
Coverage & sources
Journal, society, and media accounts. Useful signal, not independent expert judgment.
Cite This Study
Bundgaard et al. (2026) conducted an RCT in Infective endocarditis on the left side of the heart (n=508). Response-tailored antibiotic therapy vs. Standard-duration antibiotic therapy was evaluated on Days alive without antibiotic treatment for infective endocarditis or bacteremia within 6 months (Hodges-Lehmann estimated difference 13 days, 95% CI 12 to 13, p=<0.001 for superiority). Response-tailored antibiotic therapy increased days alive without antibiotics (difference 13 days) and was noninferior to standard therapy for safety (P<0.001), but increased relapse (5.1% vs 1.6%).