Ventilator-associated events prevention, learning lessons from the past: A systematic review

Heart & Lung xxx (2015) 1e9

Contents lists available at ScienceDirect

Heart & Lung journal homepage: www.heartandlung.org

Ventilator-associated events prevention, learning lessons from the past: A systematic review Jad Chahoud, MD a, Adele Semaan, MPH b, Khalid F. Almoosa, MD, MS, FCCP a, c, * a

Department of Internal Medicine, School of Medicine, University of Texas Health Science Center, Houston, TX, USA Department of Management, Policy and Community Health, School of Public Health, University of Texas Health Science Center, Houston, TX, USA c Transplant Surgery ICU, Memorial Hermann Hospital TMC, Houston, TX, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 28 August 2014 Received in revised form 16 January 2015 Accepted 20 January 2015 Available online xxx

Background: Preventing Ventilator-associated events (VAE) is a major challenge. Strictly monitoring for ventilator-associated pneumonia (VAP) is not sufficient to ensure positive outcomes. Therefore, the surveillance definition was updated and a change to the broader VAE was advocated. Objective: This paper summarizes the scientific efforts assessing VAP preventive bundles and the recent transition in surveillance methods. Methods: We conducted a systematic review to identify lessons from past clinical studies assessing VAP prevention bundles. We then performed a thorough literature review on the recent VAE surveillance algorithm, highlighting its advantages and limitations. Conclusion: VAP prevention bundles have historically proven their efficacy and the introduction of the new VAE definition aimed at refining and objectivizing surveillance methods. Randomized controlled trials remain vital to determine the effect of VAE prevention on patient outcomes. We recommend expanding beyond limited VAP prevention strategies towards VAE prevention bundles. Ó 2015 Elsevier Inc. All rights reserved.

Keywords: Ventilator-associated pneumonia Ventilator-associated Events Prevention Bundles Surveillance Critical care

Introduction Ventilator-associated pneumonia (VAP) is prominent in intensive care units worldwide. In the United States, VAP accounts for approximately 300,000 cases of ICU acquired infections per year.1 VAP is not only the most common hospital-acquired infection in ICUs, it is also associated with increased mortality, morbidity, and economical burden on the health care system.2e8 The impact of VAP on mortality has always been controversial due to limitations in most of the previous studies. Small sample sizes, inability to perform relevant subgroup analyses and the

Abbreviations: VAE, ventilator-associated events; ICU, intensive care unit; VAP, ventilator-associated pneumonia; CDC, centers for disease control; LOS, length of stay; IHI, Institute of Healthcare Improvement; HOB, head of bed; PUD, peptic ulcer disease; DVT, deep vein thrombosis; CINAHL, Cumulative Index to Nursing and Allied Health; IVAC, infection-related ventilator-associated complications; NHSN, National Healthcare Safety Network; PEEP, positive end-expiratory pressure; FIO2, fraction of inspired oxygen; ATS, American Thoracic Society; IDSA, Infectious Diseases Society of America. * Corresponding author. Department of Internal Medicine, School of Medicine, University of Texas Health Science Center, 6431 Fannin Street, MSB 1.134, Houston, TX 77030, USA. Tel.: þ1 713 500 6839; fax: þ1 713 500 6829. E-mail address: [email protected] (K.F. Almoosa). 0147-9563/$ e see front matter Ó 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.hrtlng.2015.01.010

presence of several confounding factors, have hampered a reliable measurement of VAP-attributable mortality rates. In the literature, a wide range of estimates was reported for VAP-attributable mortality rates ranging between 20 and 55%. A recent meta-analysis, published by Melhsen et al (2013) used original individual patient data from published randomized trials on VAP prevention. With a total sample size of 6284 patients from 24 trials, the reported overall VAP-attributable mortality was 13%.9 Acquiring VAP is also associated with increased ICU length of stay (LOS). In fact, studies have shown an increase in ICU LOS ranging from 4.3 to 13 days, with an average of a 6-day increase attributable to VAP.10e13 This ultimately led to increasing the cost of each hospital admission associated with a diagnosis of VAP by more than $40,000.14e16 VAP is the hospital acquired infection with the highest economic impact per episode. With an added cost per episode of more than twice that of a central line-associated bloodstream infection and ten times that of an episode of catheter-associated urinary tract or Clostridium Difficile infection.17 During the first week of mechanical ventilation, patients are at highest risk of acquiring VAP with risk rates of approximately 3% per day.18 A potentially growing burden of VAP is to be anticipated in the future, as a consequence of population aging.19 Furthermore, an

2

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

increase in bacterial resistance is expected as a result of the rise in antimicrobial prescription accompanying VAP.20,21 Considering all the aforementioned reasons, VAP prevention in mechanically ventilated patients must be regarded as a critical mission. Therefore, a vital need arises to establish a multi-dimensional strategy for VAP prevention. The components of such strategy would combine continuous staff education, VAP prevention bundles and effective surveillance. VAP preventive measures are numerous and some remain controversial.5,17 This manuscript will provide a two-faceted overview on the preventive efforts aiming to improve the quality and safety of the mechanically ventilated population over the last decade. First, we will provide a summary of the studies that incorporated VAP preventive bundles and evaluated their effect on patient outcomes. To this purpose, we will present the results and highlight the most important lessons learned from these past preventive efforts. Second, we will dissect the recently defined 2013 Ventilator-associated Event (VAE) surveillance innovative algorithm and its effects on the way we have historically tried to handle VAP detection and prevention. A thorough review of the most recent literature will be presented to describe the development of the new surveillance definition. This manuscript will culminate by evaluating the current gaps and the possible opportunities for improvement in ventilated patients’ outcomes.

assessed each of the selected studies for eligibility following an unblinded standardized manner. The selection criteria were based on exposure, outcome, population and methodology. Excluded studies did not measure “preventive bundling” as exposure and “VAP rates or mortality rates” as primary outcome. On the other hand, included papers adopted a cohort, prospective, or pre-post observational study design and targeted adult ventilated patients. As such, abstracts, letters to the editor, case reports, reviews, and original studies with less than 20 subjects were excluded. A final list of 22 studies was compiled and included in the systematic review. Fig. 1 is a diagram illustrating the adopted selection process of articles. Data were retrieved by one reviewer (J.C.) and checked for accuracy and completeness by the second reviewer (A.S.), any disagreements were discussed and resolved to reach consensus. The extracted data consisted of three categories; (1) general study information: last name of first author, year of publication, location, scope (single or multicenter), study design, and sample size (number of subjects enrolled or alternatively number of ventilatordays included and cohort size); (2) exposure information: number and type of preventive measures included in the bundle; (3) outcome information (whenever applicable): pre- and postintervention VAP rates, percentage change in VAP rates postintervention, mortality rates, hospital LOS, ICU LOS, number of days on ventilator and/or compliance rates with the VAP bundle.

History of bundling Health care providers have at their disposal an arsenal of tools to prevent VAP, including (a) VAP prevention bundles, (b) health care providers’ education and (c) surveillance programs. These components are seldom organized into one strategic quality and patient safety improvement plan. Usually, bundled preventive measures are the cornerstone of every promising preventive strategy. The concept of bundling in medicine dates back to the day of the northern plains Indians. At that time, Indian medicine bundles composed of a multitude of herbs and other elements that were believed to provide their carrier with the needed strength to prevent disease.22 Nowadays, modern medicine defines bundles as the implementation of various grouped measures which when combined together achieve better outcomes than individually implemented interventions. In order to capture the entirety of the past literature on VAP prevention bundles and their impact on ventilated patients’ outcomes, we conducted a systematic review of relevant databases. The adopted process is described in the following paragraphs.

Results Review of selected original studies Extracted data from the 22 reviewed articles are summarized in Table 1.23e44 All articles were published after 2004, the year IHI had

Methods: search strategy and data extraction We performed a systematic search on Ovid MEDLINE, PubMed, and CINAHL for original studies examining the clinical outcome of VAP prevention bundle practice on mechanically ventilated patients. The search was limited to English language articles, published from January 2005 to January 2014. The choice of this timeframe was intended to retrieve articles subsequent to the “save a 100,000 lives campaign” launched by the Institute of Healthcare improvement (IHI) and which was the first to introduce the concept of VAP bundling. The subject headings “Pneumonia, Ventilatorassociated” and “Prevention” were entered and explored to retrieve an extensive research resource for review. After removing duplicates, a preliminary screening of the resource list was conducted using the title and/or abstract to identify relevant studies that reported or evaluated the implementation of VAP preventive measures. Additionally, in order to ensure comprehensiveness, the reference lists of identified articles were checked further for related published materials. Two reviewers (J.C. and A.S.) independently

Fig. 1. Strategic search of studies on VAP prevention bundles. This diagram explains the selection process of articles while highlighting the inclusion/exclusion criteria.

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

launched the “save a 100,000 lives campaign”. The campaign aimed at improving patient safety and outcomes by implementing a prevention bundle, known as the IHI ventilator bundle; and which included four preventive measures (1) head of bed (HOB) elevation above 30 , (2) Daily sedation vacation and readiness to wean assessment, (3) peptic ulcer disease (PUD) prophylaxis, and (4) deep vein thrombosis (DVT) prophylaxis.5,17 Of these reviewed articles, 13 (59%) were conducted in a multicenter setting and 9 studies (41%) in a single institution. Studies were geographically distributed between developed and developing countries, with a majority of the publications (n ¼ 8) in the USA. All the studies, except two, had a retrospective pre- and postintervention observational study design; the multicenter Spanish intervention36 adopted a cohort study design to assess the outcome of VAP prevention bundles and the study by Croce et al applied a prospective observational design, with no pre-intervention control sample.42 The exact number of mechanically ventilated patients or ventilator days was not mentioned in 7 of these articles. The highest number of reported ventilator days was by Berenholtz et al (550,800 ventilator-days) and the largest reported sample of ventilated patients was in the study by Lim et al (27,125 patients).32,39 When assessing the number of elements included in each VAP preventive bundle, our sample of studies had a mean of 5.27 preventive measures. Eleven studies (50%) applied strictly the IHI VAP preventive bundle with its four preventive elements. Two of the reviewed articles, implemented the highest number of preventive measures, with a total of eleven elements.35,40 The type of preventive elements established in each study is detailed in Table 1. All the summarized studies, except for three,24,25,42 monitored and compared the pre- and post-intervention VAP rates. Seventeen of the studies reporting VAP rates (17/19; 89.47%) demonstrated a significant decrease after initiating the VAP bundle independently of the number of preventive measures. Only two studies26,43 did not report a significant decrease in VAP rates post-intervention, and had respectively 5 and 4 preventive measures in their bundles. Overall compliance with the bundle elements was measured, and the conducted trials reported an improvement in all the centers with rates as high as 95e100%.23,28 Even the trial that reported the lowest compliance, with rates below 30%, still had a significant reduction of VAP rates in the five study centers.36 Interestingly, none of the studies showed a significant improvement in ventilated patients’ mortality rates, except for one small pre-post observational study reporting around 50% decrease in morality.24 A significant decrease in ICU LOS as well as the number of days on mechanical ventilator was noticed in all the studies reporting those outcomes with the exception of one study which demonstrated only a significant decrease in ICU LOS without reporting a significant decrease in ventilator days.29 Limitations of reviewed studies The appropriate interpretation of quality improvement strategies relies on the ability to measure their effects. VAP prevention efforts are restricted by many limitations, the first being the lack of sensitivity and specificity of the VAP diagnostic criteria. This is considered a significant challenge that was addressed by the Centers for Disease Control (CDC) working group, in an effort that led to the development of the new innovative VAE definition. The second limitation is the retrospective pre- and post-intervention observational design of all the studies except two. Additionally, the use of a non-randomized design to compare historical control increases selection bias. The third limitation is a low external validity of some results, as eight of the reviewed studies were single-centered.

3

Moreover, most of the studies did not include a power analysis for their sample size nor did they compare the basic patient characteristics in their pre- and post-intervention samples. The fourth and final limitation found in the majority of the studies, is that patient outcomes indicators (i.e. mortality, LOS in ICU, or number of days on the ventilator) were not the primary endpoints. Instead, all the efforts were targeted towards demonstrating a reduction of VAP rates as a primary endpoint of implementing the preventive bundle. Discussion The primary objective of VAP prevention efforts goes beyond the simple reduction of VAP cases, towards improving ventilated patients’ overall outcomes. In the current era of health care change, VAE surveillance algorithm was introduced to detect a broader spectrum of complications not limited solely to pneumonia. Thus, the discussion of this literature focuses on analyzing the current limitations of studies to optimize future clinical trials. In addition, it utilizes the evidence collected over the years on developing and applying VAP bundles to be incorporated in the creation of VAE bundles. The above-reviewed VAP prevention publications (Table 1) identify a need to establish multi-center randomized clinical trials aiming to measure patients’ outcomes at a primary endpoint and the surveillance process at a secondary endpoint. Furthermore, applying the new VAE definition for surveillance would eliminate the problematic VAP diagnostic criteria of low sensitivity and specificity. As for the creation and implementation of VAE-specific preventive bundles, it will be examined further in the following paragraphs. A number of important lessons may be learned from previous VAP bundle studies to try and optimize the VAE bundles: (a) using evidence-based bundle elements; (b) considering the local resources capacity when choosing the bundle elements; (c) adopting a team-based approach to help increase the adherence of the nursing and respiratory teams with the bundle; (d) keeping in mind that increasing the compliance with the elements is more efficacious than increasing the number of elements; (e) establishing a comprehensive implementation process combining a continuous education program, process and outcome evaluation and regular staff feedback and performance improvement sessions; and finally (f) setting realistic goals with scheduled reevaluations. Until recently VAP was the only quality indicator used for mechanically ventilated patients. However, this specific population is at jeopardy for multiple other preventable diseases, undetected by the National Healthcare Safety Network (NHSN) VAP surveillance. This was one of the reasons to consider the need for change in the ventilated patients’ surveillance strategies. Furthermore, three other key issues have always hindered VAP surveillance, rendering reported VAP data heterogenous among centers. First, a gold standard for VAP diagnosis was absent. Second, the NHSN pneumonia definition, the most commonly used in VAP surveillance, was not specifically designed for this purpose and was last updated more than ten years ago. Third, the three parameters used for the NHSN pneumonia diagnostic definition, lacked in sensitivity and specificity. These three parameters are (a) new or progressive chest radiographic consolidation, (b) clinical signs and symptoms suggestive of pneumonia and (c) laboratory results. More specifically, diagnostic criteria (a) and (b) could cause a delay in determining the exact diagnosis and produce an elevated number of both false-positive and false-negative VAP diagnoses. Moreover, since both criteria (a) and (b) are usually prone to interspecialist interpretational variability, they reflect in inter-facility heterogenous comparisons. Eventually, this non-homogeneity

Author (year) 1 Resar (2005)

Location 23

Design

Sample

Na Preventive measures

Resultsc b

Pre-post observational

NA

4 IHI ventilator bundle

Pre-post observational

NA

4 IHI ventilator bundleb

3 Crunden (2005)25

Single center UK

Pre-post observational

658 patients

4 IHI ventilator bundleb

4 Abbott (2006)26

Multicenter USA

Pre-post observational

106 patients

5 Hatler (2006)27

Single center USA

Pre-post observational

NA

6 Youngquist (2007)28

Multicenter USA

Pre-post observational

429 patients

5 1 HOB elevation 30 ; 2 Oral care; 3 Hand hygiene; 4 Glove use; 5 Change/empty condensation in tubing 6 (1e4) IHI ventilator bundleb; 5 Oral care Q2H; 6 Turning Q2H 6 (1e4) IHI ventilator bundleb 5 Oral care; 6 Hand hygiene

8 Al-Tawfiq (2010)30 Single center Saudi Arabia 9 Bird (2010)31 Single center USA 10 Berenholtz et al Multicenter 32 (2011) USA

Pre-post observational

NA

HOB elevation 30 ; Oral care with chlorhexidine; Hand hygiene; Glove use; Tracheal cuff pressure maintenance > 20 cm H2O; 6 Use of orogastric tubes; 7 Avoidance of gastric overdistension; 8 Elimination of nonessential tracheal suction 4 IHI ventilator bundleb

Pre-post observational

NA

4 IHI ventilator bundleb

Pre-post observational

550,800 ventilator-days

4 IHI ventilator bundleb

11 Morris (2011)33

Single center Scotland

Pre-post observational

NA

12 Gallagher (2012)34

Single center USA

Pre-post observational

83 patients

3 1 HOB elevation 30 ; 2 Oral care; 3 Daily sedation vacation and readiness to wean assessment 7 1 HOB elevation 30 ; 2 Oral care; 3 Hand hygiene; 4 Condensation in ventilator tubing checked; 5 Daily assessment of readiness to wean; 6 DVT prophylaxis; 7 PUD prophylaxis;

7 Bouadama (2010)29

Single center Pre-post observational France (20 bed ICU)

1649 ventilator-days

8 1 2 3 4 5

1 2 1 2 3 1 2 3 4 1 2

VAP rates significant Y 6.6e2.7 Compliance [ to 95% Average ventilation days significant Y 6.1e3.5 ICU LOS Y 4e3.4 days Mortality rate Y by >50% Mean ventilation days Y 10.8e6.1 ICU LOS Y 13.75e8.36 days Mortality rate unchanged Compliance rates of 79.1% VAP rates at center II Y 31e20 (non-significant) ICU LOS Y at both centers causing cost savings

1 2 3 1

54% significant Y in VAP rates 18% Y in mean LOS Estimated cost savings of 1.0e2.3 Million USD VAP rate: Center (1) significant Y 6.1e2.7; center (2) from 2.66e0 ICU LOS: Center (2) (p ¼ 0.02); center (1) not statistically significant Compliance rate: 100% VAP rates significant Y 22.6e13.1 (43% reduction) No significant difference between the baseline and the intervention period for the duration of mechanical ventilation or the ICU and hospital death rates ICU LOS was significantly shorter during the intervention compared with the baseline

2 3 1 2

3

VAP rates Y from a mean of 9.3e2.2 Overall compliance [ 20e82% VAP rates significant Y 10.2e3.4 Overall compliance [ 53e91% VAP rates significant Y from 6.9 baseline cases to 3.4 after 16e18 months and 2.4 after 28e30 months 2 Compliance [ from 32% at baseline to 75% after 16e18 months and 84% after 28e30 months 1 VAP rates significant Y 32e12 2 Compliance rates with overall bundle elements of 70% 1 2 1 2 1

VAP rates significant Y 25.5e0

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

Multicenter USA, Canada Single center USA

2 Hampton (2005)24

4

Table 1 Description of individual studies between 2005 and 2014 that implemented VAP preventive bundle and were included in the review.

518 patients

HOB elevation 30 Oral care Daily sedation vacation and readiness to wean assessment Noninvasive ventilation use and duration minimization Use of orotracheal intubation Maintenance of endotracheal cuff pressure of 20-cm H2O 7 Removal of the condensate from ventilator circuits 8 Change of the ventilator circuit only when visibly soiled or malfunctioning 9 Avoidance of gastric overdistension 10 Avoidance of histamine receptor 2, (H)eblocking agents and proton 2 pump inhibitors 11 Use of sterile water to rinse reusable respiratory equipment 5 1Oral care; 2 Hand hygiene; 3 No ventilator circuit changes unless clinically indicated; 4 Daily sedation vacation and readiness to wean assessment; 5 Intra-cuff pressure control 4 IHI ventilator bundleb

VAP rates significant Y 18.6e11.8

NA

4 IHI ventilator bundleb

VAP rates significant Y 17.43e10.81 (38% reduction)

Multicenter 14 developing countries

Pre-post observational

137,666 ventilator-days 11

14 Rello (2013)36

Multicenter Spain

Cohort (over 19 months)

1034 patients

15 Viana (2013)37

Single center Brazil Multicenter India Multicenter Thailand (surgical ICUs) Multicenter Turkey

Pre-post observational Pre-post observational

Pre-post observational

27,125 12,913 14,212 30,585

Multicenter Cuba Multicenter USA Multicenter USA Multicenter Korea

Pre-post observational

2364 vent-days

IHI ventilator bundle Oral care; Hand hygiene IHI ventilator bundleb Oral care Maintenance of endotracheal cuff pressure of 20-cm H2O 7 Removal of the condensate from ventilator circuits 8 Change of the ventilator circuit only when visibly soiled or malfunctioning 9 Avoidance of gastric overdistension 10 Avoidance of histamine receptor 2,(H)eblocking agents and proton 2 pump inhibitors 11 Use of sterile water to rinse reusable respiratory equipment 4 IHI ventilator bundleb

Prospective observational Pre-post observational

630 patients

4 IHI ventilator bundleb

684 patients

4 IHI ventilator bundleb

Pre-post observational

19,962 vent-days

4 1 2 3 4

16 Mehta (2013)38 17 Lim (2013)

39

18 Leblebicioglu (2013)40

19 Guanche-Garcell (2013)41 20 Croce (2013)42 21 Ding (2013)43 22 Eom (2014)44

Pre-post observational

patients: pre; post vent-days

1 2 3 4 5 6

6 (1e4) 5 6 11 (1e4) 5 6

HOB elevation 30 ; Oral care; PUD prophylaxis DVT prophylaxis

b

VAP rates significant Y 22.0e17.2 (22% reduction)

1 2 3 4

VAP rates significant Y 15.5e11.7% Median ICU LOS Y 10e6 days Duration of mechanical ventilation Y 8e4 days Compliance with all measures after intervention was <30%

VAP rates significant Y 3.3e1.4 cases

VAP rates significant Y 31.14e16.82 (46% reduction)

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

13 Rosenthal (2012)35

VAP rates significant Y 52.63e15.32 (70% reduction) VAP occurred in 36% with a 15% mortality (no pre-intervention rates) VAP rates [ 9.0e10.1; change was not significant 1 VAP rates significant Y 4.08e1.16 2 Compliance [ 41.1e71.8%

VAP: ventilator-associated pneumonia; IHI: Institute of Healthcare Improvement; ICU: intensive care unit; LOS: length of stay; HOB: head of bed; PUD: peptic ulcer disease; DVT: deep vein thrombosis. a Number of preventive measures included in the bundle. b IHI ventilator bundle: HOB elevation 30 ; daily sedation vacation and readiness to wean assessment; DVT prophylaxis; PUD prophylaxis. c All VAP rates are reported in per 1000 ventilator-days. 5

6

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

causes large inter-facilities discrepancies, which has detrimental effects under the current pay-for-performance and data driven incentive policies.

Development of the VAE surveillance program The CDC acknowledged the inadequacy in inter-facility comparison and the need for improving VAP surveillance strategies. As a result, a working group convened including experts from multiple societies involved in the prevention of hospital-acquired infections with the goal of addressing the above-mentioned limitations and advising on a new approach to empower homogenous and efficient inter-facility surveillance of mechanically ventilated patients. Table 2 summarizes the advantages offered by the new CDC VAE surveillance. The CDC working group developed an innovative definition for detecting VAE, as a solution for all the above-mentioned challenges. Research conducted lately, proved the cost-effectiveness of the implementation of such surveillance algorithm targeting VAE instead of VAP solely. These programs identify the broader spectrum of complications measured, which encompasses VAP among others, such as pulmonary edema and atelectasis. The new algorithm was designed to rely on objective measures that are available in every ICU and can be translated into easily automated computerized data. This new approach is divided, into three definition tiers: (i) VAE, (ii) Infection-related Ventilator-associated Complications (IVAC) and (iii) Possible VAP and Probable VAP.45 Fig. 2 provides the key definitions of the new three-tiered algorithm. To better understand the evidence supporting the new surveillance strategy, we conducted a second literature search to identify relevant scientific studies specific to VAE. We identified four peer-reviewed articles and one abstract accepted by the IDSA 2012 meeting that satisfied our inclusion criteria, and these are summarized in Table 3.46e50 Three studies were conducted in multi-center settings,46,47,50 while the other two were single-center studies.48,49 Only one study had a prospective design, but was constituted of a small sample of only 127 patients.48 The remaining four studies had a retrospective chart review design, which is a major limitation for a study’s generalizability and validity.46,47,49,50 One study50 presented at the ID week 2012, benefited from the largest sample with 1320 mechanically ventilated patients. It demonstrated that patients diagnosed with VAE had significantly higher hospital mortality, LOS, and antibiotic days. However, one of the important limitations to this study was the lack of comparison with the old VAP definition. We will review the short history of the new CDC VAE algorithm by discussing the results of these studies in Table 3. The first effort to test the new VAE surveillance algorithm was conducted by Klompas et al46 in three major US medical centers with 200 randomly selected patients from each institution. The study retrospectively applied both the old VAP and the new VAE diagnostic criteria. The objective was to compare the LOS, number of days on the ventilator, and hospital mortality. The results demonstrated that both definitions significantly predicted longer

Table 2 Advantages of the new CDC VAE surveillance. a b c d e f

Inclusive of all the mechanical ventilator-associated complications Automatable/easily computerized Chest radiograph not required Objective parameters defining diagnosis Minimal risk of gamesmanship Homogenous inter-facility VAE rates comparison

CDC: Center for Disease Control; VAE: ventilator-associated events.

Fig. 2. Three tiered algorithm. This figure highlights the key definitions of the new three-tiered algorithm.

LOS and number of days on the ventilator when compared to matched controls. Interestingly, the new VAE definition was significantly superior in predicting increased hospital mortality while the conventional definition was not. Moreover, the process analysis demonstrated that the innovative algorithm produced faster and more objective results. Further efforts47,51 were conducted to establish the parameters of the VAE surveillance paradigm. For instance, the CDC prevention epicenters program tested 32 different definitions of the new algorithm and concluded that the one based on worsening respiratory quantitative evidence after a period of stability, significantly predicts increased LOS and hospital mortality. The only prospective study that tried to validate the innovative VAE definition applied both definitions to 127 ventilated patients.48 The results showed worsening outcomes in patients diagnosed with VAE (n ¼ 19) compared to the remainder of the population. However, the findings’ validity was limited by the small sample. Another study49 retrospectively applied the VAE new definition to 543 patients, out of which 143 with VAE were associated with significantly increased ICU LOS, duration of mechanical ventilation, and consumption of broad-spectrum antibiotics (Meropenem and ciprofloxacin had respectively 1.9 and 2.2 times more than non-VAE patients). No association was identified with longer hospital stays or ICU mortality. This study assessed the etiology of VAE and found that among the 143 patients with VAE, 31% had definite microbiological evidence of VAP, 31% had no definite diagnosis, around 16% had atelectasis, and approximately 12% had acute pulmonary edema. Three major limitations were inherent in the study approach. First, this is a single-center retrospective study, which precludes generalizability and validity of the results. Second, the etiologies for VAE were grounded on retrospective chart review, with a substantial percentage (31%) of unexplainable cases. And third, the study did not compare the VAP conventional definition to the VAE definition. The largest study by Klompas et al retrospectively analyzed data from a multicenter study by the Canadian Critical Care Trials Group in their abstract presentation at the ID week 2012. The researchers applied the VAE definition to 1320 patients, and found that patients with VAE had significant higher mortality rates, antibiotic days and hospital days. They further analyzed the effects of the VAP preventive bundle, which significantly decreased the VAE rates 2 years post-implementation. More specifically, spontaneous awakening and breathing trials were the preventive elements with the most impact on VAE rates reduction.

Table 3 Studies assessing the VAE novel definition and/or comparing it to the VAP conventional definition. Location

Design

Population

Limitations

Resultsa

1 Klompas (2011)46

Multi-center USA (3 academic centers)

Retrospective chart review, comparing: (1) Surveillance time (2) Reproducibility (3) Outcomes for VAE vs. VAP surveillance

600 patients (Each patient was assessed for VAE and VAP)

2 Klompas (2012)47

Multi-center USA (3 academic centers)

Retrospective chart review, comparing: (1) Surveillance time (2) Reproducibility (3) Outcomes for VAE vs. VAP surveillance

600 patients

1 Limited to 3 medical centers 2 Observational, retrospective 3 VAP surveillance time should be treated as approximations since they measure time for retrospective chart review 1 Limited to 3 medical centers 2 Retrospective chart review design

3 Prospero (2012)48

Single center Italy

Prospective study. Applied the epicenters’ VAP and VAE definitions to the study population

127 patients

1 Limited to a single center 2 Small sample size (n: 127)

4 Hayashi (2013)49

Single center Australia

Retrospective observational study. Investigating the association between VAE and these parameters: clinical diagnoses; ICU LOS; duration of mechanical ventilation; antibiotic use; Mortality

543 patients: - 390 without VAE - 153 with VAE

1 Limited to a single center 2 Retrospective chart review with a significant proportion of unexplainable cases 3 Use of slightly lower thresholds for significant changes in PEEP and FIO2 compared to CDC definition 4 No comparison between VAE and VAP definitions 5 Impact of VAE on mortality was limited by small sample size

5 Klompas (2013)50

Multi-center North America

Retrospective chart review

1320 patients

1 Retrospective analysis 2 No comparison between VAE and VAP definition

1 VAP rates: 8.8; VAE rates: 21.2 2 Both VAP and VAE significantly [ LOS, number of ventilator-days and ICU LOS 3 The VAE definition was a superior predictor of hospital mortality 4 Surveillance using VAE was faster 1 Surveillance using VAE was faster 2 The VAE definition predicted [ in ventilator-days, ICU days and hospital mortality as effectively as VAP surveillance 3 The VAE definition showed a sensitivity: 30%; specificity: 98%; positive predictive value: 57%; negative predictive value: 93%; agreement (j): 0.35 1 VAP rates: 1.32; VAE rates: 12.5 (2 patients with VAP and 19 with VAE) 2 VAE Patients had significantly longer duration of mechanical ventilation, ICU LOS, and hospital mortality compared to non-VAE. 3 VAP patients: no apparent difference in duration of mechanical ventilation nor hospital LOS compared to non-VAP patients 1 VAE were associated with significantly [ ICU LOS, duration of mechanical ventilation, and consumption of broad-spectrum antibiotics but not with longer hospital stays or ICU mortality 2 Of the 153 patients with VAE: - 47 (30.7%) had positive microbiological results from respiratory samples - Other etiologies in descending order of prevalence: 1. atelectasis (16.3%); 2. acute pulmonary edema (11.8%); 3. acute respiratory distress syndrome (6.5%); 4. pleural effusion (3%); 5. pulmonary embolism (2.0%); 6. aspiration (2.0%); 7. abdominal distension (1.3%) N.B: 47 cases (31%), with no defined etiology for VAE. 1 139 patients (10.5%) developed VAE. These patients had significantly [ ventilator-days, hospital days, antibiotic days, and [ hospital mortality compared to non-VAE patients 2 VAE rate Y significantly from 45/330 (13.6%) patients at baseline to 32/330 (9.7%) patients after 24 months 3 Measures potentially protective against VAE: spontaneous awakening trials and spontaneous breathing trials

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

Author (year)

VAE: ventilator-associated events; VAP: ventilator-associated pneumonia; ICU: intensive care unit; LOS: length of stay; PEEP: positive end-expiratory pressure; FIO2: fraction of inspired oxygen; CDC: Center for Disease Control. a All VAP and VAE rates are reported in per 1000 ventilator-days.

7

8

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

Limitations of reviewed studies First, the search was confined to studies published in the English language, thus research in other languages was excluded from the review. Second, the review adopted the IDSA and ATS guidelines as they are the most prominent on the VAE topic, however, other societies such as the American College of Chest Physicians (ACCP) and the Society of Critical Care Medicine (SCCM) have not been included which could have caused some bias. Third, certain studies may have been missed due to publication bias; studies not reporting any significant improvement in VAP/VAE rates may be less likely to be accepted for publication. Finally, the absence of a quality score that is reliable and specific for VAP/VAE hindered the assessment of methodological quality for the reviewed studies rendering unfeasible the performance of a meta-analysis on postintervention VAP/VAE rates. Conclusion Surveillance is a critical component of every quality improvement effort; you cannot prevent it if you cannot measure it. VAP prevention bundles have proven their efficacy in decreasing VAP rates, regardless of the number of preventive elements. This was demonstrated by our current systematic review, which till this date, and to our knowledge, is the most extensive published review on the topic. Furthermore, the choice of preventive measures should be customized to each institution to ensure the highest rates of compliance and cost-effectiveness. The essence of surveillance, as portrayed by the father of observational sciences, Galileo Galilei: “It is important to measure what is measurable and make measurable what is not so”.48,52 The new VAE surveillance approach uses data that are objective, readily available and effortlessly transferable to computerized systems. As such, it strives to ensure reliable inter-facilities case monitoring, and to prevent gamesmanship in reporting pay-forperformance measurements. Multiple retrospective studies have demonstrated the validity and numerous advantages of the novel VAE algorithm in predicting worsening outcomes. However, randomized clinical trials are needed to assess the predictive value of the VAE definition and the effects of preventive measures on VAE rates and their correlation with improvement in the clinical outcomes of ventilated patients. A window of opportunity is now open to improve ventilated patients’ health, by the creation and implementation of preventive VAE bundles with added measures that mainly target fluid management and atelectasis. Thus, an important recommendation is to go beyond the limited VAP prevention strategies, and explore VAE prevention bundle development. References 1. McEachern R, Campbell GD. Hospital-acquired pneumonia: epidemiology, etiology, and treatment. Infect Dis Clin North Am. 1998;12(3):761e779. Available at: http://www.ncbi.nlm.nih.gov/pubmed/9779389. Accessed 10.02.14. 2. Richards MJ, Edwards JR, Culver DH, Gaynes RP. Nosocomial infections in medical intensive care units in the United States. National Nosocomial Infections Surveillance System. Crit Care Med. 1999;27(5):887e892. Available at: http://www.ncbi.nlm.nih.gov/pubmed/10362409. Accessed 10.02.14. 3. Chastre J, Fagon J-Y. Ventilator-associated pneumonia. Am J Respir Crit Care Med. 2002;165(7):867e903. http://dx.doi.org/10.1164/ajrccm.165.7.2105078. 4. Kollef MH, Hamilton CW, Ernst FR. Economic impact of ventilator-associated pneumonia in a large matched cohort. Infect Control Hosp Epidemiol. 2012;33(3):250e256. http://dx.doi.org/10.1086/664049. 5. Guidelines for the management of adults with hospital-acquired, ventilatorassociated, and healthcare-associated pneumonia. Am J Respir Crit Care Med. 2005;171(4):388e416. http://dx.doi.org/10.1164/rccm.200405-644ST. 6. Anand N, Kollef MH. The alphabet soup of pneumonia: CAP, HAP, HCAP, NHAP, and VAP. Semin Respir Crit Care Med. 2009;30(1):3e9. http://dx.doi.org/ 10.1055/s-0028-1119803.

7. Agrafiotis M, Siempos II, Ntaidou TK, Falagas ME. Attributable mortality of ventilator-associated pneumonia: a meta-analysis. Int J Tuberc Lung Dis. 2011;15(9):1154e1163. http://dx.doi.org/10.5588/ijtld.10.0498. iev. 8. Vincent J-L, Rello J, Marshall J, et al. International study of the prevalence and outcomes of infection in intensive care units. JAMA. 2009;302(21):2323e2329. http://dx.doi.org/10.1001/jama.2009.1754. 9. Melsen WG, Rovers MM, Groenwold RH, et al. Attributable mortality of ventilator-associated pneumonia: a meta-analysis of individual patient data from randomised prevention studies. Lancet Infect Dis. 2013;13(8):665e671. http://dx.doi.org/10.1016/S1473-3099(13)70081-1. 10. Chawla R. Epidemiology, etiology, and diagnosis of hospital-acquired pneumonia and ventilator-associated pneumonia in Asian countries. Am J Infect Control. 2008;36(4 suppl):S93eS100. http://dx.doi.org/10.1016/ j.ajic.2007.05.011. 11. Rello J, Ulldemolins M, Lisboa T, et al. Determinants of prescription and choice of empirical therapy for hospital-acquired and ventilator-associated pneumonia. Eur Respir J. 2011;37(6):1332e1339. http://dx.doi.org/10.1183/ 09031936.00093010. 12. Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C. The attributable morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. The Canadian Critical Trials Group. Am J Respir Crit Care Med. 1999;159(4 Pt 1):1249e1256. http://dx.doi.org/10.1164/ajrccm.159.4.9807050. 13. Fagon JY, Chastre J, Hance AJ, Montravers P, Novara A, Gibert C. Nosocomial pneumonia in ventilated patients: a cohort study evaluating attributable mortality and hospital stay. Am J Med. 1993;94(3):281e288. Available at: http://www.ncbi.nlm.nih.gov/pubmed/8452152. Accessed 10.02.14. 14. Tablan OC, Anderson LJ, Besser R, Bridges C, Hajjeh R. Guidelines for preventing health-careeassociated pneumonia, 2003: recommendations of CDC and the Healthcare Infection Control Practices Advisory Committee. MMWR Recomm Rep(RR-3):1e36. Available at: http://www.ncbi.nlm.nih.gov/pubmed/ 15048056, 2004;53. Accessed 10.02.14. 15. Warren DK, Shukla SJ, Olsen MA, et al. Outcome and attributable cost of ventilator-associated pneumonia among intensive care unit patients in a suburban medical center. Crit Care Med. 2003;31(5):1312e1317. http:// dx.doi.org/10.1097/01.CCM.0000063087.93157.06. 16. Rello J, Ollendorf DA, Oster G, et al. Epidemiology and outcomes of ventilator-associated pneumonia in a large US database. Chest. 2002;122(6): 2115e2121. Available at: http://www.ncbi.nlm.nih.gov/pubmed/12475855. Accessed 10.02.14. 17. Gandra S, Ellison RT. Modern trends in infection control practices in intensive care units. J Intensive Care Med. 2014 Nov-Dec;29(6):311e326. http:// dx.doi.org/10.1177/0885066613485215. 18. Cook DJ, Walter SD, Cook RJ, et al. Incidence of and risk factors for ventilator-associated pneumonia in critically ill patients. Ann Intern Med. 1998;129(6):433e440. Available at: http://www.ncbi.nlm.nih.gov/pubmed/ 9735080. Accessed 10.02.14. 19. Arvanitis M, Anagnostou T, Kourkoumpetis TK, Ziakas PD, Desalermos A, Mylonakis E. The impact of antimicrobial resistance and aging in VAP outcomes: experience from a large tertiary care center. PLoS One. 2014;9(2): e89984. http://dx.doi.org/10.1371/journal.pone.0089984. 20. Ferrer M, Liapikou A, Valencia M, et al. Validation of the American Thoracic Society-Infectious Diseases Society of America guidelines for hospital-acquired pneumonia in the intensive care unit. Clin Infect Dis. 2010;50(7):945e952. http://dx.doi.org/10.1086/651075. 21. Jones RN. Microbial etiologies of hospital-acquired bacterial pneumonia and ventilator-associated bacterial pneumonia. Clin Infect Dis. 2010;51(suppl 1): S81eS87. http://dx.doi.org/10.1086/653053. 22. Northern Plains Indian Medicine BundlesjNative American Netroots. Available at: http://nativeamericannetroots.net/diary/859; Accessed 14.02.14. 23. Resar R, Pronovost P, Haraden C, Simmonds T, Rainey T, Nolan T. Using a bundle approach to improve ventilator care processes and reduce ventilator-associated pneumonia. Jt Comm J Qual Patient Saf. 2005;31(5):243e248. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15960014. Accessed 23.01.14. 24. Hampton DC, Griffith D, Howard A. Evidence-based clinical improvement for mechanically ventilated patients. Rehabil Nurs. 2005;30(4):160e165. Available at: http://www.ncbi.nlm.nih.gov/pubmed/15999861. Accessed 01.02.14. 25. Crunden E, Boyce C, Woodman H, Bray B. An evaluation of the impact of the ventilator care bundle. Nurs Crit Care. 2005;10(5):242e246. http://dx.doi.org/ 10.1111/j.1362-1017.2005.00134.x. 26. Abbott CA, Dremsa T, Stewart DW, Mark DD, Swift CC. Adoption of a ventilatorassociated pneumonia clinical practice guideline. Worldviews Evid Based Nurs. 2006;3(4):139e152. http://dx.doi.org/10.1111/j.1741-6787.2006.00066.x. 27. Hatler CW, Mast D, Corderella J, et al. Using evidence and process improvement strategies to enhance healthcare outcomes for the critically ill: a pilot project. Am J Crit Care. 2006;15(6):549e555. Available at: http://www.ncbi. nlm.nih.gov/pubmed/17053262. Accessed 01.02.14. 28. Youngquist P, Carroll M, Farber M, et al. Implementing a ventilator bundle in a community hospital. Jt Comm J Qual Patient Saf. 2007;33(4):219e225. Available at: http://www.ncbi.nlm.nih.gov/pubmed/17441560. Accessed 01.02.14. 29. Bouadma L, Deslandes E, Lolom I, et al. Long-term impact of a multifaceted prevention program on ventilator-associated pneumonia in a medical intensive care unit. Clin Infect Dis. 2010;51(10):1115e1122. http://dx.doi.org/ 10.1086/656737. 30. Al-Tawfiq JA, Abed MS. Decreasing ventilator-associated pneumonia in adult intensive care units using the Institute for Healthcare Improvement bundle.

J. Chahoud et al. / Heart & Lung xxx (2015) 1e9

31.

32.

33.

34.

35.

36.

37.

38.

39.

40.

Am J Infect Control. 2010;38(7):552e556. http://dx.doi.org/10.1016/j.ajic. 2010.01.008. Bird D, Zambuto A, O’Donnell C, et al. Adherence to ventilator-associated pneumonia bundle and incidence of ventilator-associated pneumonia in the surgical intensive care unit. Arch Surg. 2010;145(5):465e470. http://dx.doi.org/ 10.1001/archsurg.2010.69. Berenholtz SM, Pham JC, Thompson DA, et al. Collaborative cohort study of an intervention to reduce ventilator-associated pneumonia in the intensive care unit. Infect Control Hosp Epidemiol. 2011;32(4):305e314. http://dx.doi.org/ 10.1086/658938. Morris AC, Hay AW, Swann DG, et al. Reducing ventilator-associated pneumonia in intensive care: impact of implementing a care bundle. Crit Care Med. 2011;39(10):2218e2224. http://dx.doi.org/10.1097/CCM.0b013e3182227d52. Gallagher JA. Implementation of ventilator-associated pneumonia clinical guideline (Bundle). J Nurs Pract. 2012;8(5):377e382. http://dx.doi.org/10.1016/ j.nurpra.2012.02.017. Rosenthal VD, Álvarez-Moreno C, Villamil-Gómez W, et al. Effectiveness of a multidimensional approach to reduce ventilator-associated pneumonia in pediatric intensive care units of 5 developing countries: International Nosocomial Infection Control Consortium findings. Am J Infect Control. 2012;40(6):497e 501. http://dx.doi.org/10.1016/j.ajic.2011.08.005. Rello J, Afonso E, Lisboa T, et al. A care bundle approach for prevention of ventilator-associated pneumonia. Clin Microbiol Infect. 2013;19(4):363e369. http://dx.doi.org/10.1111/j.1469-0691.2012.03808.x. Viana WN, Bragazzi C, Couto de Castro JE, Alves MB, Rocco JR. Ventilatorassociated pneumonia prevention by education and two combined bedside strategies. Int J Qual Health Care. 2013;25(3):308e313. http://dx.doi.org/ 10.1093/intqhc/mzt025. Mehta Y, Jaggi N, Rosenthal VD, et al. Effectiveness of a multidimensional approach for prevention of ventilator-associated pneumonia in 21 adult intensive-care units from 10 cities in India: findings of the International Nosocomial Infection Control Consortium (INICC). Epidemiol Infect. 2013;141(12):2483e2491. http://dx.doi.org/10.1017/S0950268813000381. Lim K-P, Kuo S-W, Ko W-J, et al. Efficacy of ventilator-associated pneumonia care bundle for prevention of ventilator-associated pneumonia in the surgical intensive care units of a medical center. J Microbiol Immunol Infect; 2013:1e6. http://dx.doi.org/10.1016/j.jmii.2013.09.007. Leblebicioglu H, Yalcin AN, Rosenthal VD, et al. Effectiveness of a multidimensional approach for prevention of ventilator-associated pneumonia in 11 adult intensive care units from 10 cities of Turkey: findings of the International Nosocomial Infection Control Consortium (INICC). Infection. 2013;41(2):447e 456. http://dx.doi.org/10.1007/s15010-013-0407-1.

9

41. Guanche-Garcell H, Morales-Pérez C, Rosenthal VD. Effectiveness of a multidimensional approach for the prevention of ventilator-associated pneumonia in an adult intensive care unit in Cuba: findings of the International Nosocomial Infection Control Consortium (INICC). J Infect Public Health. 2013;6(2):98e 107. http://dx.doi.org/10.1016/j.jiph.2012.11.009. 42. Croce MA, Brasel KJ, Coimbra R, et al. National Trauma Institute prospective evaluation of the ventilator bundle in trauma patients: does it really work? J Trauma Acute Care Surg. 2013;74(2):354e360. http://dx.doi.org/10.1097/ TA.0b013e31827a0c65. discussion 360e2. 43. Ding S, Kilickaya O, Senkal S, Gajic O, Hubmayr RD, Li G. Temporal trends of ventilator-associated pneumonia incidence and the effect of implementing health-care bundles in a suburban community. Chest. 2013;144(5):1461e1468. http://dx.doi.org/10.1378/chest.12-1675. 44. Eom JS, Lee M-S, Chun H-K, et al. The impact of a ventilator bundle on preventing ventilator-associated pneumonia: a multicenter study. Am J Infect Control. 2014;42(1):34e37. http://dx.doi.org/10.1016/j.ajic.2013.06.023. 45. Centers for Disease Control and Prevention e National Healthcare Safety Network. Ventilator-associated Event Protocol:1e46. Available at: http://www. cdc.gov/nhsn/pdfs/pscManual/10-VAE_FINAL.pdf; 2014. 46. Klompas M, Khan Y, Kleinman K, et al. Multicenter evaluation of a novel surveillance paradigm for complications of mechanical ventilation. PLoS One. 2011;6(3):e18062. http://dx.doi.org/10.1371/journal.pone.0018062. 47. Klompas M, Kleinman K, Khan Y, et al. Rapid and reproducible surveillance for ventilator-associated pneumonia. Clin Infect Dis. 2012;54(3):370e377. http:// dx.doi.org/10.1093/cid/cir832. 48. Prospero E, Illuminati D, Marigliano A, et al. Learning from Galileo: ventilatorassociated pneumonia surveillance. Am J Respir Crit Care Med. 2012;186(12): 1308e1309. http://dx.doi.org/10.1164/ajrccm.186.12.1308. 49. Hayashi Y, Morisawa K, Klompas M, et al. Toward improved surveillance: the impact of ventilator-associated complications on length of stay and antibiotic use in patients in intensive care units. Clin Infect Dis. 2013;56(4):471e477. http://dx.doi.org/10.1093/cid/cis926. 50. Klompas M, Sinuff T, Jiang X, Day A, Muscedere J. Abstract: The Morbidity and Preventability of Ventilator-associated Conditions (ID Week 2012 Meeting). Available at: https://idsa.confex.com/idsa/2012/webprogram/Paper36985. html; 2012. Accessed 07.02.14. 51. Klompas M, Magill S, Robicsek A, et al. Objective surveillance definitions for ventilator-associated pneumonia. Crit Care Med. 2012;40(12):3154e3161. http://dx.doi.org/10.1097/CCM.0b013e318260c6d9. 52. Moores B. Monitoring: making use of the tools which are available. Health Soc Serv J. 1980;90(4691):578e582. Available at: http://www.ncbi.nlm.nih.gov/ pubmed/10246720. Accessed 11.08.14.