Dear Editor,

We read the review by Nair and Niederman [1] on ventilator-associated pneumonia (VAP) covering new data that have emerged about epidemiology, diagnosis, treatment and prevention in the past decade. It is puzzling to us why the authors failed to discuss selective digestive decontamination (SDD) under the topic of prevention. It was just in the last 10 years that SDD was confirmed as an evidence-based method to control severe infections of lower airways and blood, and to provide survival benefit in critically ill patients, with antimicrobial resistance being controlled [2].

SDD is based on the evidence that critical illness profoundly impacts the body flora, both qualitatively and quantitatively. Critical illness changes the flora from normal (Streptococcus pneumoniae, Haemophilus influenzae and Moraxella catarrhalis in the oropharynx, Escherichia coli in the gut, and methicillin-sensitive Staphylococcus aureus and fluconazole-sensitive Candida albicans in throat and gut) into abnormal [aerobic Gram-negative bacilli (AGNB) such as Citrobacter, Klebsiella, Enterobacter, Proteus, Morganella, Serratia, Acinetobacter and Pseudomonas species, and methicillin-resistant Staphylococcus aureus (MRSA) in throat and gut], and from low carriage [≤105 colony forming units (CFU) per millilitre of saliva or gram of faeces] into high grade carriage (≥105 CFU). This abnormal condition is termed overgrowth. Overgrowth of both normal and abnormal potential pathogens invariably precedes severe infections of lower airways and blood.

Traditionally, only parenteral antimicrobials are given to the critically ill, and practically all of them are excreted via bile. Although faecal concentrations may be low, they are bactericidal against sensitive clones. However, overgrowth is associated with hypermutation and polyclonality resulting in both sensitive and resistant clones. The resistant clones survive and readily reach overgrowth concentrations, causing new endogenous infections. This mechanism might explain why resistance may occur within a few years of general use of a newly launched antimicrobial agent.

The essential part of SDD is the administration of parenteral and enteral antimicrobials to eradicate overgrowth, i.e. 4 days of parenteral cefotaxime to eradicate throat overgrowth of normal flora, and enteral antimicrobials in both throat and gut to control overgrowth of abnormal flora, e.g. polymyxin/tobramycin against extended-spectrum beta-lactamase (ESBL)-producing AGNB. Enteral polyenes, either amphotericin B or nystatin, are required in throat and gut to control normal Candida spp. SDD, being a technique using broad-spectrum antimicrobials, is not a correct term [3] and should perhaps be modified into “oropharyngeal and gut overgrowth control”.

SDD using parenteral and enteral antimicrobials reduces lower airway infections by 72 %, bloodstream infections by 37 % and mortality by 29 %, reaching 42 % in particular when successful decontamination is obtained, i.e. if surveillance samples are negative for AGNB; resistance is effectively controlled (Table 1) [2, 4, 5].

Table 1 Summary of the effectiveness of selective decontamination of the digestive tract assessed in meta-analyses of randomized controlled trials published in the past decade
Full size table

We are aware that, in their review, the authors only discussed the major issues of uncertainty in VAP prevention, and, therefore, we are confident that SDD was not included among them as the evidence of its effectiveness is currently robust. However, all the manoeuvres that the authors cited as pneumonia prevention do not reduce mortality and do not control resistance.