Mortality from invasive aspergillosis remains high, and triazole resistance is rising against the first-line agents. The randomised trial of voriconazole plus anidulafungin versus voriconazole alone showed a mortality reduction that was clinically meaningful and narrowly missed conventional statistical significance — and has been read ever since as a negative trial.
This Personal View argues that reading is wrong. A result that misses P = 0.05 by a small margin is not evidence of no effect; it is an imprecise estimate of an effect that was probably real. Combined with the laboratory evidence for triazole-echinocandin synergy, the authors argue the totality favours combination therapy for immunocompromised patients with proven or probable invasive aspergillosis, and that further combination approaches deserve study as a way of preserving the agents that remain.
This is an argument rather than new evidence, and it is published in a journal that will also print the rebuttal. But it is the kind of argument worth acting on when the alternative is a monotherapy with high mortality and falling susceptibility. If your unit treats proven or probable invasive aspergillosis in immunocompromised patients, this is a conversation to have with your antimicrobial stewardship team rather than a paper to file.
- Raise combination triazole-echinocandin therapy with your stewardship team for proven or probable invasive aspergillosis in immunocompromised patients.
- Send isolates for azole susceptibility testing wherever possible — rising resistance is what makes this question urgent.
- Check triazole levels; voriconazole exposure varies enormously between patients and underdosing is a common reason monotherapy fails.
- Watch for the drug interactions an echinocandin avoids and a triazole does not, particularly with calcineurin inhibitors.
- Environmental azole resistance in Aspergillus is documented in India, which strengthens rather than weakens the case for reconsidering monotherapy here.
Why it matters
A trial widely filed as negative may have been showing a real mortality benefit all along.
Don't overread it
This is a reinterpretation of an existing trial, not new randomised evidence, and it does not change any guideline.
The statistics, in plain English
A trial that "narrowly failed to reach significance" has not shown that a treatment does not work — it has produced a confidence interval that includes both a worthwhile benefit and no benefit. Treating P = 0.06 as a negative result and P = 0.04 as a positive one is the error this article is about. What it cannot do is replace the larger trial that would settle the question.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for infectious diseases, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free