Blog

Gordon West from United Kingdom wrote 07-16-2013 12:01:51 pm
There is no doubt in my mind the major reason this has not taken off is the concerns over resistance generation. I can't see how this question can be answered even in a big RCT as the time to develop resistance might be longer than the duration of the study. I'd love to hear from the investigators how they plan to overcome this limitation.

Hella from Denmark wrote 07-19-2013 03:26:13 pm
I don't know that antibiotic resistance has never been associated with SDD. These claims are more scare tactics than real life. It has been practiced in Netherlands for decades without concerns

Ian Seppelt from Australia wrote 07-22-2013 01:13:55 pm
I'll declare my COI up front, as one of the SuDDICU Investigators. We are a multinational collaboration of intensivists, infectious diseases physicians, microbiologists, nurses, ethicists, statisticians, health economists, clinical trial methodologists .... Our goal is to do not just the next (37th) RCT for the next Cochrane metaanalysis, but to do the DEFINITIVE trial which answers the question, one way or the other.

Our intention is to do a multicentre cluster RCT of SDD plus usual practice, vs usual practice alone, combined with a parallel ecology study which will continue beyond the end of the RCT, and a formal heath economic analysis. Current plans are for a trial involving 100 ICUs in four countries (Australia, New Zealand, Canada and the UK) recruiting 30,000 patients - that will give us the power to answer the primary question (Does SDD save lives?). We will be recruiting sick ICU patients, expected to be ventilated at least until "the day after tomorrow" [and also patients where we got it wrong but they have in fact remained intubated for 2 days or more]. For the longitudinal ecology study [of every patient in the ICU, whether or not treated with SDD] we will need up to 50,000 patients over a number of years.

Terrifying numbers? Yes, but feasible because of the nature of a cluster trial where every eligible patient in a participating ICU is by definition in the trial (so a busy ICU can potentially recruit 300 - 500 patents per year and even a smaller unit can recruit 100 patients). As SDD (or not) will be "the unit policy" it will happen in the background, without individual patient consent. Patients and families will however be informed about the trial, and will give permission for followup. A (properly funded) research coordinator will do most of the hard work making it run in your ICU, if you choose to participate.

The intervention will be similar to classic 'Dutch' SDD. It is now almost impossible to get good old-fashioned amphotericin, so we will be using a mixture of nystatin, colistin and tobramycin orally and enterally [manufactured especially for the trial], with an initial 4 day course of IV antibiotics [ceftriaxone or ciprofloxacin, unless the patient is already on an appropriate antibiotic for therapeutic purposes].

The trial is likely to start early 2015, and we will be spending the next 12 months getting funding together - this trial has to be properly funded or it won't happen.

This is a really exciting initiative with the potential to significantly change practice, if in fact SDD saves lives as a lot of European data suggests it does.

If you or your ICU are interested in participating, please get in touch with any of the following:

Australia: Ian Seppelt, ian.seppelt@sydney.edu.au
New Zealand: Paul Young, Paul.Young@ccdhb.org.nz
UK: Anthony Gordon, anthony.gordon@imperial.ac.uk
Canada: Brian Cuthbertson, Brian.Cuthbertson@sunnybrook.ca [Brian is the international CI]

We look forward to any discussion or comment, whether positive or negative - the more this question is discussed the better. Blog away .....

Best wishes, Ian

Matbailey from New Zealand wrote 07-22-2013 07:07:36 pm
I'll be really interested to hear what the responses from most centres is to this trial. For everybody's sake, I hope that the collective opinion of our senior group ("I want this question answered, I just want somebody else to get it for me") is not too pervasive! I for one would not be too worried if I lost colistin and nystatin from my armamentarium. I was much more concerned by the use of lost 3rd gen cephalosporins and quinolones - losing them would be scary!

Con from Greece wrote 07-24-2013 10:44:49 am
Sounds interesting. Will you be doing surveillance cultures as part of SDD or only if clinical indication?

David from Australia wrote 07-24-2013 11:04:06 am
Interesting that you say that the research coordinator will be doing all the work - not entirely true. You're also going to have a job on your hands convincing the nursing staff in the unit to do it. My understanding is that this can be laborious, unpleasant and time consuming, so winning the "hearts and minds" of the RNs might be difficult.

Paul from New Zealand wrote 07-24-2013 01:06:24 pm
Okay, so I have the same conflict as Ian. I am one of the investigators. I also have a confession - I wish SDD was sexy. People think SDD is boring. This is the largest challenge in designing a trial in SDD. David is right. Winning hearts and minds is crucial.

Here is my short summary of what we know:
The evidence base for SDD includes 36 published randomised controlled trials enrolling over 7000 patients. On average, an absolute risk reduction of 3-6% with use of SDD is present in the most recent randomised studies. Obtaining the appropriate balance between antibiotic use whilst avoiding driving antibiotic resistance is crucial. However, existing data suggest SDD does not increase antibiotic resistance and may actually reduce it [see, for example: Lancet Infectious Diseases (2013) 13: 328-341.]. Further data on the effect of SDD on antimicrobial resistance are needed. However, resistance issues aside, people are reluctant to implement SDD without further evidence of its effectiveness outside the Netherlands.

In the SuDDICU study, we will be powered to detect a realistic clinical effect. The study will be conducted in four countries providing greater external validity than the current data. We will evaluate antimicrobial resistance patterns through surveillance cultures as well as routine cultures taken for clinical purposes.

This is the definitive trial of a boring therapy that has been around for decades. However, this trial needs to be done now. The current situation means that patients across the world are being denied a potentially life saving treatment whilst we await further evidence for fear of potential longer-term effects. Don't try and think of reasons why your unit can't do this trial; try and think of how you can convince your colleagues to take part. We need your support to establish whether or not this works.

Freddo from United Kingdom wrote 07-28-2013 08:05:14 pm
Is it really boring? I wouldn't have thought so.

Is it really more boring than Insulin? You lot got quite excited about that for a while.

Or enoxaparin? Snore... but you did a study on that too.

Is it really that its boring? Or is it that no-one wants to swim against the tide and say they are for it when everyone else seems to have been "toeing the political line" on the issue for the past 20 years.

As you say, there are few interventions with this degree of literature support, and the fear mongers' worst prophecies are completely unsupported by available evidence.

So if, as Ian suggests, you can power a trial to show safety as well as efficacy, then Viva la Antipodes!

Todd Fraser from Australia wrote 07-29-2013 10:11:04 am
That's a very interesting point Freddo. It would be great to know what the issues are that have prevented SDD from becoming more widely adopted. I would imagine someone would have looked into this in more detail.


This would be fairly important I guess as even if the SuDDICU trial demonstrates a benefit, it might still struggle to be implemented if these issues are not dealt with. Does that modify the trial design at all? Perhaps the study arm needs to be adjusted so that it is "acceptable" to the wider community. I wonder if Ian or Paul could comment.

The other points to pick up on are Gordon's point about the time-scale to detect resistance generation, and Con's about surveillance cultures. Perhaps Ian or Paul could comment on this also?

Ian Seppelt from Australia wrote 07-29-2013 02:16:56 pm
Thanks for all the comments to date.

The SuDDICU group has been trying to systematically address a lot of these things. We have done a formal qualitative study [currently in press] which specifically addresses the question of 'why we don't believe the evidence'. The short summary is (1) we all believe it is an important question which needs answering (2) we have equipoise but (3) we re concerned about resistance [no surprises in any of that].

What was really interesting was the difference in attitudes between senior doctors and senior nurses in the study. The statement is often made, and David has mentioned it in this forum, that SDD is hard work and messy and the nurses are going to hate it. The interesting bit is that while doctors say that, the nurses themselves do not. Oral hygiene is a very important part of ICU nursing practice, and for a nurse who is already regularly cleaning patients' teeth, administering mouthwash, applying chlorhexidine paste (in 50% of our units) etc, the addition of SDD paste is just more of the same and they are not concerned. Most of the nurses we interviewed were actually quite excited by the concept, not because they knew anything about the evidence base for SDD, but because they believed in good oral hygiene.

We also did case studies of the 12 ICUs in the UK that have already implemented SDD [likewise in press] and the nurses there were perfectly comfortable with the process.

So, as David highlighted, the challenge will be not so much convincing nurses to do something nasty, but rather in convincing doctors that in fact it is perfectly okay as far as the nurses are concerned!

Con asked the question about surveillance. Surveillance cultures are an essential component of SDD and will be required in both the intervention and control ICUs. Gordon and Todd ask about the timescale to detect resistance generation. While the ID physicians in our group argue that six months is plenty to see the emergence of resistance, the ecology study and followup will go substantially longer (up to 5 years) as we don't need to so much convince the scientists but convince all the sceptics out there [or alternatively prove there is a danger and then the therapy is dead]

Hopefully we have at least piqued your interest, and very happy to discuss further. Ultimately we need to prove feasibility to the NHMRC to fund the study which means enough non-binding expressions of interest from ICUs, so please contact us if you are interested.

We promise that it will be properly funded [as without adequate funding it ain't gonna happen]

Ian

Floris vbh from Netherlands wrote 08-01-2013 07:28:23 pm
Dear Colleagues,

Here's a view from the Netherlands... I work in a 10 bed general ICU, teaching hospital in Hilversum, NL.

I am very thrilled by the fact that you are planning such a big important study outside Europe. Not boring at all!!

I would like to add to the discussion that it would be quite important to try to define "usual care", i.e. the control group of the study. Of course the intervention arm will be well defined. By the way really cool that you will be using nystatin because Amfo B indeed is getting expensive and hard to obtain.

However, we all know that there are huge differences in how we deal with antimicrobial policies around the globe. You lose a lot of discriminative power if you are going to compare a well defined intervention arm with a very diffuse control group....

I wish you all the best with this enormous task. Greetings
Floris van Braam Houckgeest

Todd Fraser from Australia wrote 08-02-2013 11:34:07 am
Hi Floris,

Welcome to the blog! Great to hear from you, and say hi to Rutger for me!

That's an excellent point about the variance in "usual practice". I'd be interested in hearing from the investigators on this.