A “farewell” to my blog readers

In May this year I wrote my first blog, and I have come to realize that it gives me a lot of fun. Professionally, it’s a major step back in time: as a medical student I earned some money with writing columns in the university magazine (for instance on students’ night life) and reporting on soccer games in the local journal. I was paid by “the number of words printed” and the head editor weekly  taught me that you could easily reduce and improve a 3000 word piece (as submitted) to 300 words (as printed). Taking advantage of that exercise, I hope you found my blogs not too long and boring.

Blogging sharpens your reasoning, makes you read manuscripts carefully and – when blogging on others’ work – makes you think twice on how to word conclusions (of the blog and your own future work). Naturally, I accept (and hope) that others will be equally critical (or laudable) on my own research.

So, why a farewell? I have decided to join the “professional” bloggers Jon Otter, Martin Kiernan and Andreas Voss in “Reflections on Infection Prevention and Control”, which may be the most well-known and read blogsite in the field. I sincerely hope that will follow me there.


Risk factors Cipro-resistance in E. coli

Just out in JAC a nice study on risk factors for ciprofloxacin resistance (CIP-R) in community-acquired E. coli UTI in elderly, demonstrating the power of linking a human cohort to a microbiological database. The results: risk factors for CIP-R are age and prior CIP use, as were high intake of pork (OR=3.68) and chicken (OR=2.72), and prescription of calcium supplements (OR=2.51) and proton pump inhibitors (PPI) (OR=2.04), while receiving CIP. The conclusion: Modification of antibiotic use in animals, and temporary discontinuation of calcium supplements and PPI as prevention strategies to prevent CIP-R need further evaluation.

Hypothesis: PPI reduces absorption of CIP, leading to sub-inhibitory CIP levels, leading to CIP resistance. Sounds reasonable, but how to bridge the gap between association and causality? A randomized approach? Let’s look at the absolute risks: they had 15,000 subjects, studied for 180,000 personyears (“back of cigar box calculus”), yielding 1,080 with E. coli UTI (110 CIP-R, of which 31% without prior CIP exposure), that’s 1 UTI per 167 person years (1 CIP-R UTI per 1,636 person years)…. The PPI/CIP risk was based on 34 CIP-R UTI in PPI/CIP users and 92 in CIP only users. The odss may double, but absolute risks seem very low.

What about the meat: diet information was available for 507 (of 1,080) subjects and ORs were based on comparisons (1st to 3rd tertiles of use in gram/day) of 7 versus 19 (for pork) and 11 versus 16 CIP-R events (for chicken). According to the conclusion: no doubt that these CIP-R must have originated from the slaughtered and cooked animals.

So, do we need further evaluation of these risks? If we ask the authors to give us the absolute risks first (it must be in their database), we might conclude that further study attempts should be proportional to the public health risk of concomittant use of PPI and CIP or eating pork and/or meat.

I met-a-analysis I did not like*

Donald is a dedicated clinical microbiologist making the best of his job, every day. Yesterday, his labs’ Bactec alarm went off: growth in a blood culture. He was on his way to chair the hospitals’ Antibiotic Committee meeting during lunch. After 1.89 hours he was in his lab to do the molecular rapid diagnostic test and reported the coagulase-negative staph (with recommendation “check line status”) to the intensivist. Proudly, he walked to his desk: “modern microbiology in optima forma”.

On his desk a new CID study: A meta-analysis on the effects of molecular rapid diagnostic testing (mRDT) on clinical outcomes in BSI. 31 studies (25 USA) with 5,920 patients. Techniques used PCR (n=20), PNA-FISH (n=6), MALDI-TOF (n=4). In 20 studies testing was accompanied by introduction of an antibiotic stewardship program (ASP).

Increasingly frightened he read: mRDT reduced the time to effective therapy with 1.89 hours (he rightfully discarded the 5 hours estimated because of an I2 of 75%), see. This  1.89 hour delay was associated with an increased 30-day mortality (OR 0.66, 95% CI 0.54-0.8 for testing) and with prolonged length of hospital stay (8 days). He realized that his lunch meeting might well have killed a patient, and that from now one he would need to work 24×7 to do his job appropriately. Donald felt like a loser, and called a good-looking epidemiologist, he knew from one of his hobbys, for help.

She reassured him. More strange conclusions: “mRDT reduces mortality only when combined with ASP.” Would be nice, but the ORs were 0.64 (95% CI 0.51-0.79) with and 0.72 (95% CI 0.46-1.12) without ASP. Yet, the first analysis had 3,803 patients, the second only 1,332. Pretty bold statements, she said, and gave him another mRDT study, that for unknown reasons was not included. Donald said: “Meta-analysts should question whether their findings are biologically plausible.”

* Quote from Robert A. Weinstein

The humanization of LA-MRSA

Ahead of print in CID the long-awaited “evidence for human adaptation and foodborne transmission of LA-MRSA”. As a long-time member of the NYC-LAMTH society (“Not-Yet-Convinced of the LA-Mrsa Threat to Humans”) I am longing to end my membership.

As described before, see blog May 27, there is a subpopulation of LA-MRSA, called CC9/CC398 which is the most appealing candidate to provide the promised evidence. This new study started with the recognition of 12 patients with CC9/CC398, some without apparent exposure to livestock. WGS was performed and compared to WGS of 110 CC9/CC398 (101 MRSA) from other European countries, and all these isolates clustered within the larger CC398 family.

These 12 isolates represented 4 clusters (linked through hospital or family connection) and 3 singletons. So, in fact, there were 7 distinct isolates (confirmed by WGS).

10 of 12 isolates (1 epi-cluster excluded) belonged to a clade of 49 isolates, all harboring a Sa3 phage, and derived from humans, animals and retail food. Moreover, 18 of 19 CC9/CC398 isolates derived from poultry and poultry meat were in this clade (as compared to 4 of 56 CC9/CC398 isolates from other animal species and retail foods). So, this clade was labelled “poultry-associated subclade”. The origin of the 3 Danish meat isolates was traced back to French production facilities!

So what was concluded from this brief report of 12 (or 7) isolates?

  1. Danes have been exposed tot this pathogen (!, only 1 associated with infection) with some frequency through consumption or handling of contaminated food products.
  2. Persons who handle raw meat professionally are probably at higher risk for expsoure.
  3. CC9/CC398 may have arrived in urban Denmark through person-to-person transmission, with a colonized/infected livestock worker as first link in each chain. But this was considered less likely.

There is a lot of “might”, “probably” and “suggests” in the discussion, which ends with the disclaimer that “the cases described here are not sufficient to change the generally accepted view that foodborne transmission plays only a minor role in the epidemiology of LA-MRSA in humans.”

The NYC-LAMTH society is alive & kicking and welcomes new memberships.

Did VRE protect our patients?

Today in Science a great paper. In short, Enterococcus faecium produces SagA (secreted antigen A), and this appears to block invasiveness of Salmonella in the gut (of C. elegans and mice). SagA introduced in Lactobaccillus plantarum had similar effects. There is even a mechanism: “The protective activity of E. faecium and SagA in mice requires the TLR signaling adaptor MyD88, the peptidoglycan pattern recognition receptor NOD2, and the C-type lectin RegIIIγ.”

Why is this of interest for infection preventionists?

E. faecium has caused a pandemic of intestinal carriage and infections in hospitalized patients. The pandemic was first noticed in the US with the increase of VRE in the nineties, followed by the emergence of VRE in Europe and the rest of the world 10 years later. The VRE population consists for >90% of E. faecium, which species took over an epidemiological niche previously occupied by E. faecalis. The precursor of VRE is ampicilline-resistant E. faecium (ARE), belonging to the same genetic clade. SagA is an essential protein for E. faecium and contributes to biofilm formation.

The authors conclude: “These results together suggest that E. faecium and SagA may function through evolutionarily conserved pathways to enhance epithelial barrier integrity and protect animals from enteric pathogens.”

Up till now we have seen VRE as a threat to our patients, although I think that the studies claiming attributable mortality due to VRE are not that convincing. In light of these new findings, though, massive gut colonization with E. faecium (all expressing SagA) may protect patients from invasive gut-derived infections caused by Enterobacteriaceae. Actually, the emergence of ARE in Dutch ICU coincided – in time – with our studies on SDD. May be, it was not the antibiotics in the gut that created the benefit ….. That’s probably too much speculation, but this study significantly reduces my skepticism related to microbiome studies.

The spin of AMR

In 1988 a student went from door to door. Instead of a religious pamphlet he had a small container with a spoon in the top. The next day he returned for collecting his – what he thought – golden brown. He plated the stuff and found that almost all fecal samples contained resistant bacteria, see, and concluded: “The results of this study underscore the presence of a human reservoir of antibiotic resistant microorganisms.” The scientific breakthrough of that year was cited once per year since then.

In September 2016, on a Friday, the former student found himself preparing a European clinical trial network for combatting AMR (Sept 16). Something, important people consider relevant, as he learnt from the statement of the G7 Health summit:“

“We encourage governments to consider the need for establishing a global clinical studies network on drug resistance that provides access to a large clinical research infrastructure for the design, coordination and conducting of clinical trials and studies in cooperation with the existing global experts networks to ensure the common benefit of the outcomes.” (Sept 12)

On the Tuesday, thereafter, (Sept 20), he received an email: Your recently submitted working group proposal: “Consensus group on the design, analysis and reporting of antibiotic stewardship trials”, to the JPIAMR call launched in April 2016, was evaluated by a Peer Review Panel (PRP) composed of scientists/experts with an internationally recognised expertise with regard to the aim of the call and your working group was ranked in the list of proposals recommended for funding.”

The next day someone pointed him to this statement: “Leading Pharmaceutical Companies Present Industry Roadmap to Combat Antimicrobial Resistance” (Sept 20), stating – amongst other things – we will explore new opportunities for open collaborations between industry and the public sector to address challenges in the research and development of new antibiotics, vaccines, and diagnostics, recognizing the value these bring to society.

And when he woke up on Tuesday his iPhone told him this: “Dutch health minister Schippers invests millions in new antibiotics and alternatives” (Sept 22), she announced at the United Nations General Assembly (UNGA) High-Level Meeting on Antimicrobial Resistance. The platform will be part of the Netherlands Centre for One Health.

So, what does this say:

  1. A good week for – at least – 1 AMR researcher.
  2. Don’t worry if nobody pays attention to your students project.
  3. If you currently study the prevalence of Brachyspira pilosicoli in the human gut, you may well have a similar good week in September 2044.

AMR & the beauty of co-existence

As we all know, by 2050 about 10M may die from AMR each year, according to the O’Neill report. One of the assumptions is that “all infections (=100%) of a certain bacterium will be caused by a fully resistant variant.” How likely is that scenario? Yesterday, Marc Lipsitch shared his thoughts on this question with us.

He’s already thinking on this for years. At night he lie awake and wondered “if serotypes from pneumococci differ in their transmissibility and duration of carriage  why do we not see a clear winner that occupies the ecological niche”. Like a firm that can produce candy for a price of 11 cents, where others need to spend at least 13 cents, will drive the market to a price of 12 cents, outcompeting all others while still making profit. His team delivered some fundamental insights on this, see, for the pneumococcus (read it before you drink your beer).

Now, can we address a similar question to AMR? We all think to know from our data that AMR is never 100% or 0%. Do we really? At least from surveilance data: MRSA increased (and decreased) but I never saw a country with 100% of S. aureus infections being MRSA (or it is because EARSS doesn’t have a colour fort hat). In all ICU studies I was involved in, AMR of a certain pathogen never reached 100% of all isolates or all patients carrying that bug. And once colonized in the gut with VRE are all enterococci VRE?

These are simple questions, probably with simple answers that can be used by simple people in (not so-)simple mathematical models to provide basic insights in AMR epidemiology. There are at least 2 Marcs that think that we urgently need such data for better prediction of what we can expect from AMR in the (near) future.

May be, the economists in the O’Neill group did know, but considered such knowledge as highly classified.