Ventilator-associated pneumonia prevention: one good turn does not always deserve another

Abstract

Ventilator-associated pneumonia (VAP) is a frequent complication of patients undergoing invasive mechanical ventilation and its occurrence is associated with considerable morbidity and mortality [1][2][3]. The key factor in the pathogenesis of VAP is the endotracheal tube (ETT) [4]. Its presence compromises the natural anatomical barriers (larynx) and functional mechanisms (mucus clearance and cough), and it expedites the formation of biofilm as well as macro-and microaspiration of oropharyngeal secretions. As a result, several strategies have been successfully implemented, namely subglottic secretion drainage or cuff pressure control [3,4].One of the current recommendations for VAP prevention is elevation of the head of bed to 30-45° to prevent the reflux of colonized gastric contents that has a potential role in VAP pathogenesis. This strategy has been evaluated in three randomized controlled trials (RCT) enrolling 337 patients, one positive and two negatives [5][6][7], and in a meta-analysis pooling these three studies that found a significant impact on prevention [8]. However, the quality of evidence is low, being considered a basic practice because of its simplicity, ubiquity, low cost, and potential benefit [3].Simultaneously, we can conceive that in a patient with an ETT, any position above horizontal increases the gravitational forces exerted by subglottic secretions above the cuff and, consequently, the risk of aspiration [9]. The results of several experimental studies challenged the recommendation of the semirecumbent position in VAP prevention. In animal models, the Trendelenburg position showed an increase in mucus clearance, decreasing aspiration and consequently VAP rates [9,10]. With this rationale an RCT was designed to evaluate this hypothesis.In the GRAVITY-VAP trial, recently published in Intensive Care Medicine, Li Bassi, Panigada, and co-workers performed a multicenter, multinational RCT to assess the impact of the lateral Trendelenburg position (LTP) vs. semi-recumbent position (SRP) in VAP incidence [11]. A total of 395 patients (N = 194 LTP; N = 201 SRP) were included. The trial was prematurely ended because of low incidence of VAP, lack of benefit in secondary outcomes, and serious adverse events (SAE). The incidence of microbiologically confirmed VAP was 0.5% in LTP and 4.0% in the SRP patients (P = 0.04); however, the rate of SAE was significantly higher in the LTP group, namely transient oxygen desaturation and hemodynamic instability. In addition, mortality (ICU, hospital, and day 28) was non-significantly higher in the LTP group, ranging from an absolute difference of 6.5% in ICU mortality to 4.5% in day 28 mortality. Moreover, in spite of the risk difference of −7.6% (P = 0.02) in microbiologically confirmed VAP in the LTP group, antibiotic consumption was similar in both groups. This may be due to the higher incidence of clinically suspected VAP in LTP patients with pulmonary infiltrates at the time of enrollment, encouraging BAL assessments, and empirical antibiotic prescription...