In 1983 Chris Stoutenbeek et al. [1] introduced selective digestive decontamination (SDD) as a means of preventing infection in clinical intensive care medicine. The first SDD studies consisted of a series of observations on several different antimicrobial interventions to prevent and treat pathological colonization of the oropharynx, rectum, and organ sites such as the respiratory tract and urinary tract. Stoutenbeek and colleagues made two original observations in these trauma trials: (a) Regarding the pathogenesis of infections, in particular lower airway infections: practically all infections are endogenous, i.e., preceded by throat and/or rectal carriage. Primary endogenous infections due to micro-organisms present in the admission flora are distinguished from secondary endogenous infections due to micro-organisms acquired on the ICU and subsequently carried in throat and gut, i.e., secondary carriage. Approximately 20% of infections are exogenous, i.e., due to micro-organisms not previously carried. (b) Regarding eradication of aerobic Gram-negative bacilli (AGNB) from the oropharynx. This knowledge allowed them to design an antibiotic protocol, a four-component strategy, termed SDD consisting of: (a) parenteral antimicrobials to control primary endogenous infections, (b) enteral antimicrobials to control secondary carriage and subsequent endogenous infections, (c) a high level of hygiene to control exogenous infections, and (d) regular surveillance samples to monitor the efficacy of the SDD protocol.

Initial results were compared with data obtained from a cohort of previously treated polytrauma patients (historical controls). Impressive results were found: urinary tract infections were virtually absent, bloodstream infections were reduced by 50%, and lower airway were reduced from over 80% to less then 10% [2, 3]. However, even more important than the reported results was the proposed pathogenesis of the ventilator-associated pneumonia (VAP), consisting of three main routes: the primary endogenous route, the secondary endogenous route, and the exogenous. All routes need specific measures to prevent VAP, ignoring one or more of these routes would lead to suboptimal results [2].

The proposed new insight into the pathogenesis of VAP was only reluctantly accepted. This was because the general consensus at that time considered breakdown of hygiene as the pivotal step for development of infectious complications in the ICU patient. The endogenous route was merely discarded as failure of hygiene ignoring the crucial role of the abnormal carrier state. For ultimate scientific proof of the SDD concept, studies with robust design were needed. A single-center, double-blind, placebo-controlled trial was conducted by Stoutenbeek et al. [3] and finished in 1986. The results were similar to those of their first study. The preparations for the study which Stoutenbeek et al. [4] now report in Intensive Care Medicine began in 1989. The study, a multicenter, double-blind, randomized controlled study would allow optimal scientific evaluation of the SDD intervention in a defined cohort of polytraumatized patients. Shortly after his appointment as Professor in Intensive Care Medicine at the University of Amsterdam in 1994, Stoutenbeek was diagnosed with non-Hodgkin lymphoma, which prevented him finishing the study. He died in 1998. The study was in the end finished by van Saene and coworkers but not until van Saene himself had recovered from an acute brain bleeding.

One of the most important contributions in the field of intensive care medicine made by Stoutenbeek and colleagues is their understanding of the role of the abnormal carrier state of throat and gut as the most important step in the pathogenesis of VAP. The second major contribution is the development of an adequate oral decontamination regimen using a sticky paste (Orabase®) containing 2% polymyxin E, tobramycin, and amphotericin B, to eliminate the Gram-negative carrier state and to prevent acquisition of ICU acquired AGNB, after attempts using clorhexidine rinsing had failed to do so. The impact of the researchers' initial work on the scientific intensive care community was huge. SDD became one of the most studied therapeutic topics in intensive care medicine [5]. Several SDD meta-analyses followed, the first one in 1991, showing a significant reduction in the incidence of VAP [6]. In 1994 the first meta-analysis that showed a significant mortality reduction of 20% was published [7]; a decade later, in 2004, a Cochrane review confirmed the previous results [8]. In the present study, well designed by Stoutenbeek et al. and published nearly a quarter of a century after their first study, VAP is reported to be significantly reduced, as is the occurrence of AGNB septicemia [4]. Mortality, an endpoint in this study, is reduced but not statistically significantly due sample size limitations. The impact on mortality, however, is in line with the meta-analyses on this topic.

Is everything said and done now with regard to SDD? Many are still reluctant to implement SDD. However, in the hands of the experienced intensivists, trained by Stoutenbeek and Zandstra using SDD, mortality can be reduced as can antimicrobial resistance [9]. Appropriate perception and execution of the SDD concept reduces mortality in middle-range APACHE II scores without the induction of antimicrobial resistance [10, 11]. Still the enormous potential of SDD is not fully appreciated based on the unfounded fear for induction of antimicrobial resistance or the perceived costs. For some, SDD is still emotion-based medicine (ebm); others consider it evidence-based medicine (EBM).

Chris Stoutenbeek, the pioneer, deserves to be posthumously honored for his research in the field of intensive care medicine and his solution which minimizes the problem of AGNB infections. The outcome of ICU patients can significantly be improved as a result of his innovative contributions to intensive care medicine.