How accurate are predictive outcomes for ventilator-associated events?
How accurate are predictive outcomes for ventilator-associated events?
Results from the EUVAE study
In 2013, the Centre for Disease Control and Prevention (CDC) proposed (and in 2015 updated) a new diagnostic algorithm for ventilator-associated events (VAEs) based on more objective criteria and easily measurable clinical features [1,2]. The algorithm divides the causes of respiratory worsening in critically ill patients into tiers: ventilator associated complication (VAC), infection-related ventilator-associated complication (IVAC), possible ventilator-associated pneumonia (PVAC), and infection-related ventilator-associated complication-plus (IVAC-plus).
EUVAE was a prospective, observational, international, multicentre study conducted in 13 ICUs from eight countries (Australia, France, Greece, Iran, Italy, Slovenia, Spain, and Turkey) [3]. The aim of the project was to analyse the impact on patient outcome of VAEs as defined by the CDC and explore the correlation with ventilator-associated pneumonia (VAP) and tracheobronchitis (VAT). Eligibility criteria included age >18 years and mechanical ventilation (MV) > 48 hours; the first 30 subjects from each participating centre were followed up daily for 30 days and the first VAE episodes were recorded. Outcome end-points were extubation, ICU discharge, or ICU death.
The study cohort comprised of 244 patients (of 1078 eligible ones) with 2856 ventilator-days of complete follow-up. The ventilator-associated infections (VARIs) were 84, corresponding to 29.3 episodes/1000 ventilator-days; based on 2008 CDC criteria the 39.3% of episodes was classified as VAP (12.2/1000 ventilator-days) and the 60.7% as VAT (17.8/1000 ventilator-days). VAP had higher mortality than VAT (42.8 vs 19.6%, p<0.05). VARI patients had a significantly longer MV duration by 6 days compared to non-VARI ones. The recorded VAE episodes were 117 (40.8/1000 ventilator-days) developing after a median of 4 MV-days. VAE patients compared to non-VAE ones had significant increases in MV-duration, ICU- and hospital-LOS (by 5, 11 and 12 days, respectively). The main cause of VAE was IVAC-plus (39.6/1000 ventilator-days), accounting for 113 (96%) episodes, 64 (56.6%) of which met the CDC criteria for PVAP (22.3/1000 ventilator-days). In patients classified as IVAC-plus without PVAP criteria, most episodes were due to atelectasis (26, 55%). Both VAE and IVAC-plus definitions had a negative predictive value of 98%. However, it is noteworthy that only 24 (72.7%) of VAP and 15 (29.4%) of VAT episodes met IVAC-plus criteria.
Strengths
- Large, multicentre, international project with prospectively collected data of a mixed ICU population, and with predefined outcomes.
- Identification of both VAP and VAT episodes.
Limitations
- A potential selection bias of the participating ICUs, the high proportion of participating university hospitals, and the exclusion of children, chronically ill patients, and patients ventilated outside the ICU might limit the extrapolation of the results to other settings.
- The study might have been underpowered to demonstrate statistically significant differences of events with low incidence.
Take home messages
- Respiratory infections were the most common complications in mechanically ventilated adults, with VAT being the most frequent one.
- VAEs were associated with significantly worse outcomes compared to non-VAEs.
- VAE algorithm identified only the most severe cases with surrogates of severe oxygenation deterioration. Consequently, IVAC definitions missed ¼ and ¾ of VAP and VAT episodes, respectively.
- The current study drives new research questions and highlights the need for identification of better breakpoints and for exploration of the effects of new management algorithms for ventilated adults, using more subtle definitions of VAE.
Review provided by Dr Savino Spadaro on behalf of the WG on Pneumonia.
References
- Magill SS, Klompas M, Balk R, Burns SM, Deutschman CS, Diekema D et al. (2013) Developing a new national approach to surveillance for ventilator-associated events: executive summary. Am J Infect Control 41:1096–1099
- Magill SS, Li Q, Gross C, Dudeck M, Allen-Bridson K, Edwards JR (2016) Incidence and characteristics of ventilator-associated events reported to the national healthcare safety network in 2014. Crit Care Med 44:2154–2162
- Ramirez‑Estrada SS, Lagunes L, Pena‑Lopez Y, Vahedian‑Azimi A, Nseir S, Arvaniti K, et al., and the EU-VAE Study Investigators Group. (2018) Assessing predictive accuracy for outcomes of ventilator‑associated events in an international cohort: the EUVAE study. Intensive Care Med. 2018 Jul 12. doi:10.1007/s00134-018-5269-7. [Epub ahead of print]