Exam question: Which intervention targeted at preventing infection and mortality in critical illness has been evaluated positively in 71 randomised controlled trials over 35 years?

Well, of course you know the answer. Indeed, it is the wrong question to ask. The real question is why, given this body of research, has selective digestive decontamination (SDD) not been widely adopted, and why was it excluded from consideration by the most recent iteration of the international Surviving Sepsis Campaign (SSC) guidelines? We will try to unpackage this ‘riddle wrapped in a mystery inside an enigma’ [1].

SDD was excluded from the first iteration of the SSC guidance in 2004. Reasons offered at the time included the view that SDD was not a treatment for sepsis. The illogicality of this approach was immediately evident given the inclusion in the guidelines of prophylaxis for thromboembolic disease and for peptic ulcer disease. Following representations, SDD was included in 2008 and 2012. In 2008 the SSC adopted private polling to minimise bias, developing the ‘GRADE grid’ for that purpose [2]. This revealed two opinion groups: nine respondents weakly in favour and eight weakly opposed. Of the remaining participants, four were neutral, and one strongly opposed, giving SDD a 2B rating (consider using the intervention). This rating was retained for the 2012 guidelines. However, in the most recent iteration in 2016 [3], SDD has been excluded without explanation, presumably being classed amongst a group of interventions of ‘low importance and of low priority to clinicians’. If SDD is indeed of ‘low priority’, is this because of new and important scientific evidence demonstrating that it does not work, or because of deeply held but unfounded anxieties or antipathies? To assist with this analysis, we have summarised current evidence on the effect of SDD on mortality, infection, resistance and the microbiome in Table 1 (references in electronic supplementary material). This includes data from 71 randomised controlled trials, and confirms that SDD reduces mortality, reduces infections, reduces Gram-negative antimicrobial resistance, and does not promote Gram-positive infections. What other intervention in critical care medicine can match this record over 35 years of research?

Table 1 Summary of the impact of selective decontamination of the digestive tract on morbidity, mortality, resistance, and gut microbiome in critically ill patients
Full size table

Attitudes to SDD were explored in an international survey in 2013 of 141 participants which demonstrated substantial equipoise on risks and benefits of SDD, and expressed concerns that SDD might contribute to antimicrobial resistance [4]. A commonly expressed view is that it is wrong to give prophylactic enteral antimicrobials for long periods to critically ill patients to prevent infection; and yet the current approach of treating septic episodes with repeated and prolonged courses of parenteral antimicrobials can hardly be described as a success from the perspective of antimicrobial resistance. By contrast, a systematic review in 2013 of resistance related to SDD found no relation between the use of SDD and antimicrobial resistance in ICU patients, and concluded that ‘the perceived risk of long-term harm related to selective decontamination cannot be justified by available data’ [5]. This conclusion was confirmed by a 4-year ecological study showing that the introduction of SDD in an ICU is associated with a significant reduction in resistance rates [6], and is reinforced by our summary of current research evidence (Table 1). Those who regard SDD as a resistance risk need to explain why ICUs in the Netherlands, where SDD is widely adopted, use fewer non-SDD antimicrobials and have a much reduced problem with antimicrobial resistance [7, 8] than those countries which rely predominantly on parenteral antimicrobials to prevent or treat nosocomial infections in critically ill patients.

As the traditional SDD regimen does not target MRSA and vancomycin-resistant enterococci, adjustment of the SDD medication by adding vancomycin should be considered in ICUs with a high endemicity of these Gram-positive infections [9, 10]. However, it should be noted that while some studies show an increase in Gram-positive isolates, a meta-analysis did not demonstrate an increase in Gram-positive infections [11].

The argument against SDD has now shifted to the microbiome. A recent study [12] reported the presence of numerous resistance genes in the gut microbiota of ten intensive care patients receiving SDD, compared with healthy individuals. But again, this is the wrong exam question. The real question is how does the gut microbiota of critically ill patients receiving SDD differ from those receiving only parenteral antimicrobials for nosocomial infection, and how do these two groups compare with hospitalised but not critically ill patients, and with a healthy population? Finally, how does the gut microbiota of surviving patients in the first two groups evolve following return to the community?

We hope that the answers to these focused questions might emerge from current research [13]. Researchers have struggled to obtain funding for further international collaborations evaluating SDD, perhaps because funders have noted the large evidence base which already exists in favour of the intervention. However, both the Australian and New Zealand Intensive Care Society (ANZICS) and the Canadian components of this study are now funded, and it is hoped that the UK will also be able to participate. If the SuDDICU study confirms (yet again) the benefits of SDD, then we should stop denying critically ill patients one of the few interventions in intensive care medicine which has been proven to be effective in saving lives and reducing harm. To those of our colleagues who regard SDD as being of ‘low importance and of low priority to clinicians’ [3], we feel justified to ask who it is they think they are treating—the clinicians, or our patients?