Dear Editor,

Selective decontamination of the digestive tract (SDD) has been shown to decrease mortality to a greater extent than selective oropharyngeal decontamination (SOD) in critically ill patients [1, 2]. This effect may be caused by a difference in the occurrence of ventilator-associated pneumonia (VAP), but this has never been compared head-to-head. We aimed to compare the occurrence of VAP in two clusters of patients: two half-year clusters of SDD versus a 1-year cluster of SOD.

This was a prospective observational cohort study in a single ICU. For this study we had data on VAP occurrence during 6 months of SDD, 12 months of SOD, and 6 months of SDD, consecutively. The occurrence of VAP, the primary endpoint, was prospectively scored by trained MARS researchers on a daily basis [3]. We performed three sensitivity analyses. First we broadened the criteria for VAP to all patients that were treated for VAP. Second, the analysis was restricted to patients who developed VAP after 4 days in the ICU, as this marked the end of systemic antibiotic therapy in the SDD regimen. Third, ICU-related nosocomial pneumonia was used as the endpoint. Data analysis was restricted to patients admitted for longer than 48 h. Cause-specific and subdistribution hazard ratios were estimated using competing risk analysis. Additional details can be found in the electronic supplementary material.

VAP was diagnosed in 5.6% of patients who received SOD and in 2.5% of patients who received SDD (P = 0.002). The VAP specific hazard ratio of SOD was 2.2 (95% CI 1.3–3.7), with a subdistribution hazard ratio (SDHR) of 2.1 (95% CI 1.2–3.5; Fig. 1). According to the sensitivity analysis with all patients in whom the attending physicians started antimicrobial treatment for VAP, the VAP specific hazard ratio for SOD treatment was 2.1 (95% CI 1.4–3.0) and the SDHR for SOD treatment was 1.9 (95% CI 1.4–2.8). A similar association was found for VAP that developed after day 4 (VAP specific hazard ratio 2.1, 95% CI 1.1–3.9; SDHR 2.0, 95% CI 1.1–3.7; Fig. S3). SOD was also associated with an increase in the rate of all ICU-related nosocomial pneumonias (specific hazard 1.6, 95% CI 1.1–2.4; SDHR of 1.7, 95% CI 1.1–2.4; Fig. S4). Cephalosporin exposure was higher in the SDD group (P < 0.001) as cefotaxime was part of SDD treatment, while quinolone treatment was higher in the SOD group (P = 0.006; Table S4).

Fig. 1
figure 1

Hazard ratios from competing risk analysis for multiple risk factors and all patients with possible, probable, and proven VAP. The figures display the cause-specific and subdistribution hazard ratios (HR) for VAP of multiple risk factors. The mean hazard ratio (rectangle) with confidence interval (horizontal bar) is displayed. Statistically significant associations do not cross the vertical dashed line. All values above 1.0, on the right side of the dashed line indicate an increased risk for VAP associated with the risk factor

Full size image

In this single-center large cohort of patients who either received SDD or SOD, the regime of SOD was associated with a twice as high occurrence of VAP compared to the SDD regime. SOD had a cause-specific hazard ratio of 2, suggesting that patients are at double the risk for VAP during a SOD regime when controlling for all risk factors and competing risks. This is the first study comparing SDD to SOD using likelihood scoring for VAP. The results were insensitive to subgroup analyses with broader or narrower definitions for the outcome, suggesting a reliable association. The major limitation is that this did not have a major impact on antibiotic exposure, which is in line with previous studies [4].