🚨🚨 Critical result presented at #ESCMIDGlobal2026 🚨🚨

Suspect we are dealing with the strain first described in Swaziland, which is also found in Mozambique (link in next message)

Big implications for clinical care and diagnostic algorithms

Terrifying results from Dr Nomonde Mvelase (OS081)

I491F, rifampicin-R NOT detected by Xpert assay, is common in South Africa - 47% of isonizid-R isolates have this occult MDR

Likely to be strain initially observed in Swaziland, which was commonly BDQ resistant (testing of RSA isolates planned)

Looks great! Will a recording be available afterwards?

We are so far off nuance in elderly inpatients in the NHS

Terrible sample taking, failure to consider alternative diagnoses, thoughtless co-amoxiclav prescriptions

I don't doubt it! I'm jaded from spending too long signing out contaminated/uninterpretable urine sample results on microbiology rotations

Tell me about it

Sounds like all the quality papers are going elsewhere these days @onisillos.bsky.social, and I hear the peer review is pleasingly light touch!

a referee wearing a microphone and a fifa patch ALT: a referee wearing a microphone and a fifa patch

Flurry of papers at the end

Final score MDPI 4, Lancet Microbe 1 !!

She's running the clock down, limiting your opportunities for a come back

Oh, MDPI 2, Lancet Microbe 0

Slightly leftfield choice of papers, so far

Is this urine or perineum?

So far MDPI 1, Lancet Microbe 0 ... Not looking so good for @onisillos.bsky.social

It is going to be very large, isn't it? Also, you don't need a full menu of options to massively expand our evidence base

NeoSep is going to randomise 3000 infants, and will provide a head to head comparison of a good number of agents using a PRACTical design

Suspect empiric antibiotic choice has most impact on outcome, and important to enroll early so that can be randomised (perhaps with deferred consent)

Agree antibiotic choices later in disease less likely to impact mortality, but that's ok if using NI design (low event rate doesn't impact power)

To guide clinical practice, I would like to see an RCT of indivual drugs

E.g. carbapenem vs temocillin vs ...

In UK context, would like to see evaluation of narrow-spectrum beta-lactam plus aminoglycoside (to cover potential resistance) strategy

Don't think these agents are interchangeable

Regards PRACTical design, I do think there are meaningful differences between beta lactams, both PKPD and toxicity (see SNAP, etc)

Potentially much bigger differences between non-beta lactam options. I am not sure I am happy to average over e.g. cipro and co-trim (clearly different!)

Agree not easy

One advantage of MAMS-ROCI is that, even if 4 days fails (too short or you under-recruit), top of CI may allow you to reduce to e.g. 5 days

Clearly much more efficient than failed trial

@matteoq21.bsky.social will know if similar approach available in Bayesian framework

Talk of the conference for me

Covers both TB and NTM - highly recommend watching back (session SY076)

#ESCMIDGlobal2026

here in Germany ENT physicians like to give Cefuroxime axetil for this viral condition and it works fine too!

Is it a 'trial'? Looks like an observational study to me

I can think of scenarios where you could omit an efficacy outcome entirely, e.g. in palliative OPAT, your aim is usually to improve comfort

Not yet

If that's the case, I think you are fine with e.g. death/infection-related death/cure as a non inferiority endpoint, and tolerability or similar as a superiority endpoint

Primary efficacy endpoints in phase 3 non-inferiority trials to establish new tuberculosis treatment regimens should only include microbiological outcomes - PubMed Primary efficacy endpoints in phase 3 non-inferiority trials to establish new tuberculosis treatment regimens should only include microbiological outcomes

I'm not into composite outcomes

Have thought about this most in context of TB treatment trials, where purely microbiological efficacy outcomes work best (most deaths NOT due to TB)

100%

Personalised randomised controlled trial designs-a new paradigm to define optimal treatments for carbapenem-resistant infections - PubMed Antimicrobial resistance is impacting treatment decisions for, and patient outcomes from, bacterial infections worldwide, with particular threats from infections with carbapenem-resistant Enterobacter...

Do we think all 'de-escalation', all oral BL, all oral non BL antibiotics are equivalent?

Why not compare individual antibiotics using PRACTical design (developed for gram neg antibiotic RCTs)

Could randomise empiric therapy, again when DST available then oral switch, stratified by AmpC risk

ENDS

Rethinking non-inferiority: a practical trial design for optimising treatment duration - PubMed Background Trials to identify the minimal effective treatment duration are needed in different therapeutic areas, including bacterial infections, tuberculosis and hepatitis C. However, standard non-in...

They plan a 7 vs 4 days randomisation

What if the optimum duration of antibiotics is 5 or 3 days?

Why not use a MAMS-ROCI (DURATIONS) design?