Are risk factors and impact of prevention strategies the same for early- and late-onset VAP?

EJRC Article Review

The risk factors and impact of prevention strategies for early-onset (EOP) versus late-onset (LOP) ventilator-associated pneumonia (VAP) are still under debate.

Saied and colleagues conducted a retrospective analysis of data prospectively entered from 01/1997 to 03/2016 into the OUTCOMEREA, a multicentre French database of 22 ICUs, which aimed to assess:

1. a) specific risk factors for EOP and LOP, and
2. b) the impact of the progressive implementation of recommended prevention strategies on the rates of EOP and of LOP.

All patients under mechanical ventilation (MV) for >48 hours were included in the study and censored after 28 days of MV or death. EOP was defined as VAP occurring between day 2 and 6, while the rest was defined as LOP. Regarding prevention strategies implementation, three periods were defined: before 2001, 2001-2006, after 2006, and the bundle for each period was recorded using questionnaires. EOP and LOP tendency was described taking account of the modification of bundles over these periods. From the 7,784 included in the analysis, the 1,234 developed VAP [EOP: 445 (36%)].

Risk factors specifically associated with EOP were admission for respiratory distress, chest tube, enteral feeding within the first two days of MV and previous MDR-P.Aeruginosa colonisation. Antimicrobials administered within the first 2 days of MV were all protective of EOP. ICU admission for COPD exacerbation or pneumonia were early risk factors for LOP, while imidazole and vancomycin use within the first two days of MV were protective. Late risk factors (3^rd-6th MV-day) were the intra-hospital transport, PAO2-FIO2<200 mmHg, vasopressor use, and known MRSA colonisation. Among the antimicrobials administered between the 3rd and 6th day, fluoroquinolones were the only protective ones. Male gender was the only common risk factor for EOP and LOP. Noteworthy, varying MV cut-off to five days for the definition of EOP and LOP had no impact on the final multivariate model. Regarding VAP incidence, only EOP risk decreased across the study time periods and this decrease was concomitant with an increase of compliance to bundle prevention measures.

Major study strengths

• Use of the large, multicentric OUTCOMEREA database that is comprised of both university and non-university ICUs that use a consistent VAP definition, which facilitates the generalisation of the findings.
• Use of Fine and Gray model for the first VAP episode identification that allows a simultaneous estimation of two independent competing events, i.e. being extubated for >48 hours and death before VAP occurrence.

Limitations of the study

• Bronchoscopic techniques or quantitative culture of endotracheal aspirates for bacteriological confirmation of VAP might have led to a lower VAP rate compared to studies based only on clinical evaluations; however, this should had impacted equally EOP and LOP.
• Compliance to prevention measures was not individually monitored, therefore it cannot be excluded that unmeasured confounders had contributed to the significant decrease of EOP over the study time periods.
• The impact of MDR and antimicrobial therapy might not be generalisable as it refers to MDR risk observed in France, where carbapenemase-resistant Enterobacteriaceae strains are rare.

Take home message

VAP risk factors were different according to the time of VAP onset: EOP risk was mainly associated with a quantitative increase of aspiration of the oropharyngeal content, while for LOP the persistence of organ system failures and shock, and colonisation with MRSA were the predominant risk factors. The above findings lead to the conclusion that prevention of EOP and LOP might need different strategies, and highlight the need for further research.

Article review for the ESICM Journal Review Club on behalf of the Pneumonia-WG provided by Dr. Despoina Koulenti and Dr Evdoxia Tsigou

Reference

Ibn Saied W, Souweine B, Garrouste-Orgeas M, Ruckly S, Darmon M, Bailly S, Cohen Y, Azoulay E, Schwebel C, Radjou A, Kallel H, Adrie C, Dumenil AS, Argaud L, Marcotte G, Jamali S, Papazian L, Goldgran-Toledano D, Bouadma L, Timsit JF; OUTCOMEREA study group. Respective impact of implementation of prevention strategies, colonisation with multiresistant bacteria and antimicrobial use on the risk of early- and late-onset VAP: An analysis of the OUTCOMEREA network. PLoS One 2017; 12(11):e0187791. doi: 10.1371/journal.pone.0187791.