ECCMID Day 3 – Oh no, not more MDROs!

**Standard disclaimer: these are my notes taken during the sessions, as accurate as I can make them, please let me know if any inaccuracies. I’m putting them here as others have said they found them useful. However worth re-checking anything before you re-quote it. If this continues to be useful, I’ll continue. If it’s clear i’m not reaching the standards of accuracy I’m aiming for, I’ll review future posting plans.**

Take home message for me today – screening for carriage of Carbapenem-Resistant Enterobacteriaceae isn’t as easy as you might think/I thought/I hoped…

Some data presented suggesting CRE screening using culture alone may miss a lot of carbapenemase carriers that PCR can detect. Truly worrying for those using culture-based detection to stop transmission of resistance elements.

John Otter from St Marys, London  who has previously written about it: had a poster where they  screened 816  admissions to a UK teaching hospitals with rectal swabs – found a very low carriage rate of CRE identified by culture (0.1%) but a considerably higher rate by PCR (2%)

There was also a very interesting study quoted on respiratory microbiome – where depleting the microbiome of mice with multiple antibiotics (poor mice!) appeared to be bad when they subsequently gave them Strep. pneumoniae pneumonia (it really is rubbish to be a mouse!) . More died, more lung inflammation  and the alveolar macrophages were bad at phagocytosing.  The respiratory microbiome appeared protective either at maintaining immune function or colonisation resistance or something else. Needs validation in clinical setting, but good message for the ‘don’t take antibiotics for viral infections just in case’ interventions if true.

The talk on the vaginal microbiome was hilarious. Was with one of my male colleagues who admitted  it made him Rather Uncomfortable. Especially when the presenter said that the best way to resolve recurrent bacterial vaginosis is circumcision of the partner. Drastic.

More below:

Expert session on CRE detection.Mazzariol:

This was basically an experienced Microbiologist from Verona talking about their method.  I note disagreement with her method from other people, but for a resource-limited setting starting out and no money for PCR it seems a decent place to start. Hopefully talk will be up and do her method more justice than I can in text.

A good note from the chair: RECTAL MUST BE RECTAL not peri-anal – must be colored/stained swab, or they send it back. they feel that enrichment for faec samples doesn’t add much if the rectal swab and testing are done properly.

I commented about this method on twitter and it immediately got shot down by two UK researchers who say that whilst it might select for phenotypic resistance  it’s not sensitive for detecting carbapenemase genes – see the John Otter poster above.


ESBL session

A mixed session, but with two useful studies of ESBL transmission:

ESBL transmission rates following cessation of isolation of ESBL-positive patients. A Prospective cohort study at acute medical ward and geri/rehab ward in Basel, Switzerland.  Inclusion: pt  with ESBL clinical isolate. Rectal and wound swabs of patient and anyone in same room >24hrs.  Contacts were screened at 4 days (acute) and 15 days (geri).  211 index pts. 231 contact patients. They think 11/231 contacts acquired ESBLS from contact with an ESBL-clinical-isolate-patient. With transmission rates 2.6% acute wards and 8.8% at geri/rehab wards. Small numbers so difficult to see associations but contact >5 days associated OR 10. You do note that they didn’t have a control group (not in same room as ESBL patient) which would also have been interesting, as an alternative hypothesis would be that these patients are just all picking up the endemic strains around the unit.

Jorgensen studied 101pts with ESBL UTI,  mostly women, and took faecal samples from the patients and household contacts every 3 months. in a mathematical model, they estimated household contacts had a 67% chance of transmission from patient. Pickup in the community without ESBL contact 36x less. Usefully they defined clearance as time until 2 negative samples (as no subsequent re-colonisations with same strain if participant had 2 negative samples in a row) and took on average 1 year for patient to become ESBL negative whilst contacts took 3 months.



Clinical Microbiome session brief summary:

Wade gave a nice, and animated summarisation of the literature – take home message- gum disease probably involves the shift from a self-buffering system to overcoming the buffering which leads to a propagation of of acidic anaerobic pro-inflammatory microenvironment.

Noted association of protozoa levels with gum disease. But bad gum disease = more plaque = more food = more protozoa, assoc not cause? Propsed Saliva contains a 10^8 /ml bacterial ‘soup’ – mouth covered with  structured bacterial biofilm Made everyone feel vaguely yuk I suspect.

Some very good quotes which I paraphrase as accurately as I can:

“It is fairly easy to run a microbiome study these days, if you have a body site, you can get some easy bar charts but there is a gap between what is given and what is a clinically important question”

“The human body expects a microbiome, it develops with a microbiome, the immune system expects to see a colonising microbiome”.

Key point where you overcome the natural buffering ability of saliva for prolonged period of time that selects for caries-specific bacteria.

Wiersinga presented the data on the protective role of the respiratory microbiome as discussed above.

Edit: found the references, now I can see the slides on ECCMID live!! Looks like the Strep. pneumoniae data is WIP, but the some of the work is published

A good quote on the effect of antibiotics – “Each generation could be beginning life with a smaller endowment of ancient microbes than the last”

Not much from the vaginal microbiome study, very well presented talk if you’re interested in the vaginal microbiome.

Finally interesting talk by a Dr Luna from Texas.  Gave loads of examples of the potentials for microbiome research to revolutionise clinical practice. I hope they publish a lot of their very promising findings as many points made with unpublished data it seemed (may have just been incompletely presented), and without peer reviewed results, remains open to the suggestion that it could be anecdote/pattern rather than scientific study set up to robustly answer the question of whether it really can do what we think it can do, but hope very much this is just preliminary data, and it is as good as it looks.  Nevertheless was a stunningly interesting talk about the possibilities of microbiome testing to be a companion, as she said, to traditional diagnostics.

Edit 8am 28/4/15   update (finishing off!) : Understanding the Microbiota: recent breakthroughs

A mixed session, which perhaps highlighted many of the issues with the nascent topic of microbiome research. The issue with Rhodanobacter DNA contaminating meconium samples highlighted by Paalaane, sometimes in high levels. Correctly identified as “kit-ome” rather than “sample-ome”… Probably reflecting the low biomass of their samples – see excellent work of Birmingham and Nick Loman:  Re-iterating what they say, but unfortunately means that without rigorous negative controls and ensuring samples are processed with recorded extraction kit, machine etc and mixed,  often very hard to find what is true sample positive, and what is kit/bench/machine, especially when you’re running highly similar samples. Lesson – go into the extraction/prep process with decent amounts of sample producing decent amounts of DNA, and assume your sample will be contaminated – it will. Honestly. It will.

There was also some highly promising work from a Biomerieux group on proof-of-concept for direct-from-sample sequencing from sputum samples. Sensible considered approach including spiking to establish limits of detection, and consideration of whether human-DNA depletion  biased community analysis (not massively from the eyeball I saw). They found 70ish percent of DNA in  their extract was human without human depletion, and around 15% with depletion = more useful DNA per run. Sequenced to depth of 1 million reads. Sequenced to depth of 10 million reads+ some suggestion they can assemble the Staph. aureus gene denovo. Promising.

Over and out!



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s