Impact of evidence-based bundles on ventilator-associated pneumonia prevention: A systematic review

Introduction: This review aimed at investigating the impact of bundle components on the prevention of ventilator-associated pneumonia (VAP) in adults and the elderly. Methodology: The databases consulted were PubMed, EBSCO, and Scielo. The terms Bundle and Pneumonia were searched in combination. The original articles were selected in Spanish and English; published between January 2008 and December 2017. After eliminating the duplicate papers, an analysis of the titles and the abstracts was performed in order to select the assessed articles. A total of 18 articles were included in this review that were evaluated according to the following criteria: research reference, country of data collection, type of study, characteristics of the studied patients, analysis and intervention performed, bundle items investigated and their results, and research outcome. Results: Four bundle items were presented in all the investigated papers. 61% of those works were considered from seven to eight bundle items. Daily evaluation of sedation interruption and daily assessment for verifying extubation condition, head-of-bed elevation at 30 degrees, cuff pressure monitoring, coagulation prophylaxis, and oral hygiene were the most reported bundle items. One study described the increased mortality of patients under mechanical ventilation when omitted the bundle items of oral hygiene and stress ulcer prophylaxis. Head-of-bed elevation at 30 degrees was the item reported in 100% of the studied papers. Conclusions: Existing research demonstrated that VAP reduction occurred when bundle items were performed for adults and the elderly. Four works showed the relevance of team education as a central approach to the event reduction related to the ventilator


Introduction
In the context of healthcare-related infections (HCRIs), pneumonia is the second most frequent and severe infection in hospital patients.It occurs from the inflammatory response of the pulmonary parenchyma and the uncontrolled penetration of infectious agents as multidrug-resistant microorganisms which lead to severe respiratory signs and symptoms [1,2].
Endotracheal intubation and mechanical ventilation are widely used for ensuring suitable oxygen supply and can save patient lives in the intensive care unit (ICU).However, these interventions can be deleterious since ventilator-associated pneumonia (VAP) can occur after 48 hours of intubation [3,4].
VAP has a high prevalence in ICU, representing one of the main infectious complications in severe patients, and is related to high mortality rates [5,6].VAP incidence can rise when the days of mechanical ventilation increase.Rates in ICU ranging from 30 to 70% and from 8% to 33% for incidence and mortality associated with VAP are recorded, respectively.Previous studies demonstrated that VAP can delay the mechanical ventilation time by 7.6 to 11.5 days and can stay in the hospital for extended time intervals by 11.5 to 13.1 days.Moreover, the costs can increase by 40,000 U.S. dollars per episode [6,7].
In this scenario, improving care for ventilated patients has become a key concern for the national health systems of many countries.Prevention and control approaches are essential and priorities in order to avoid adverse events such as VAP.A basic set of interventions must be performed for preventing VAP and these strategies are widely recognized in literature as a care bundle [7,8].In brief, the aim of care bundles is to improve health outcomes by facilitating and promoting changes in patient care and encouraging guideline compliance [6].VAP bundle involves different care procedures such as the head-of-bed elevation, the daily sedation interruption protocols, the daily spontaneous breathing test, oral hygiene using mainly chlorhexidine, the aspiration of subglottic secretion, the cuff pressure verification, the coagulation prophylaxis, and the stress ulcer prophylaxis [9,10,11].
However, VAP represents a burden to health care, and previous studies have reported changes in their protocols and sets of interventions.In spite of the guidelines of VAP, bundles reveal some differences, these strategies substantially reduce morbidity and mortality rates when suitably implemented [6,7].Taking all these into account, the present paper aimed to review the previous reports concerning the evidencebased bundles on ventilator-associated pneumonia prevention in adults and elderly patients admitted to the hospital ICUs.

Methodology
The present study was carried out according to the PRISMA guideline (Preferred Reporting Items for Systematic Reviews and Meta-Analyzes) [12].
The PICOS approach was used in order to formulate the guiding question [13,14], where "P" refers to the population or patient; "I" to the intervention; "C" to the comparison; "O" to outcome; and "S" to study design.Considering these parameters, this systematic review was based on: P: adult and elderly patients; I: to have received the bundle care package on PAV prevention; C: have not received the care bundle on PAV prevention; O: intervention efficacy; and S: observational studies and clinical trials.

Literature search strategy
The studies were identified using search strategies provided by the given databases (PubMed, EBSCO and SciELO) as described in Table 1.

Inclusion/exclusion criteria
Inclusion criteria for the studies were: clinical trials based on observational and interventional studies which were performed involving adult and/or elderly populations and were published in English and in Spanish between January 2008 and December 2017.After identifying the studies from these databases, two researchers carefully assessed if they fulfilled the inclusion criterion for this study.Duplicate papers were excluded.Titles and abstracts were then examined.The papers that did not fit to the PICOS parameters and those that did not describe the patients' comorbidities were also excluded.

Literature identification
The data collected were summarized according to the following information: population, country, type of study, patients' characterization as age and comorbidities, analysis and intervention performed, recorded bundle items, main results, and outcome (Table 2).

Results
Figure 1 shows the flowchart of the methodology used for identifying research work for this study.Considering the inclusion and exclusion criteria, a total of 18 papers were included for an in-depth review.Regarding the clinical trials investigated, ten papers were cohort studies, six articles were devoted to reporting case-control studies, and two works were experimental studies.Considering the comorbidities related to the samples evaluated, cardiovascular diseases were the most prevalent ones followed by respiratory disorders, bone fractures, trauma, and renal insufficiency.
The bundle items described in these papers for VAP prevention are shown in Table 3.Four bundle items were reported in all the investigated papers.A maximum of nine bundle items were recorded and 61% of the studies considered seven to eight bundle items.The five main items assessed were: (a) daily evaluation of sedation interruption and daily assessment for verifying extubation condition, (b) head-of-bed elevation at 30 degrees, (c) cuff pressure monitoring, (d) coagulation prophylaxis, and (e) oral hygiene.One study described the increased mortality of patients  [5] Daily evaluation of sedation interruption Daily assessment for verifying extubation condition Head-of-bed elevation at 30 degrees  under mechanical ventilation when omitted the bundle items of oral hygiene and stress ulcer prophylaxis.In particular, head-of-bed elevation at 30 degrees was the item reported in 100% of the studied papers.

Discussion
The care bundle to prevent VAP is a priority in hospital ICU services due to these patients show high morbidity and mortality rates [15,16].Risk factors that lead to the VAP evolution are commonly classified as non-modifiable, including age, patient severity score at ICU admission, and presence of comorbidities.In contrast, the modifiable factors are associated with mechanical ventilation, previous use of antimicrobial agents, colonization and biofilm formation in the buccal cavity and endotracheal tube, patient position in bed, and other invasive procedures [17,18].
Regarding the main non-modifiable risk factors, 15 studies (83.33%) investigated patients with a mean age of 47 to 66 years, two papers (11.11%) examined patients over 65 years (77 to 78.8 years), and one work evaluated adults with a mean age of 35 years.These clinical trials demonstrated that elderly patients are more vulnerable to acquiring HCRIs compared to young adults because there are physiological changes concerning aging, the declining immune response, the number of invasive procedures, and a higher predisposition to chronic diseases [19,20,21].Comorbidities were remarkable elements for patients' admission to ICU services due to complications related to cardiovascular, respiratory, renal, and hepatic disorders [15,22,23], and these findings were similar among the 18 papers here investigated.
In this perspective, the care bundle was described as a suitable approach for ICU patients to reduce the incidence of VAP considering the 18 articles studied.Among the bundle items for VAP prevention, five of them demonstrated the highest level of evidence for supporting a prevention action plan against events associated with the ventilator: (1) head-of-bed elevation at 30 degrees, (2) daily evaluation of sedation interruption, and daily assessment for verifying extubation condition, (3) oral hygiene, (4) cuff pressure monitoring, and (5) coagulation prophylaxis.The headof-bed elevation between 30 and 45 degrees is one of the main recommended care procedures to decrease the gastroesophageal pressure and the succeeding bronchial aspiration.This simple procedure ensures increased inspired tidal volume and muscular effort reduction.In addition, the findings from randomized clinical trials showed that head-of-bed elevation is related to the reduction of days spent in mechanical ventilation, which reinforces that this no-cost procedure has a positive effect on mechanically ventilated patients [10].Two studies involving VAP prevalence in mechanically ventilated patients revealed appropriate results concerning the reduction of VAP rates by 60% when the head-of-bed elevation was associated with other care procedures [24,25].
It is well known that suitable sedation is essential in the intensive care of critically ill patients with respiratory problems and under mechanical ventilation.The obtained results from a retrospective review of the VAP incidence in mechanically ventilated patients were lower in the group of patients with daily sedation interruption when compared to the continuous sedative infusion [26,27,28].Another study devoted to evaluating the association among bundle items for VAP prevention reported a high team adherence to daily sedative infusion interruptions (94%).This care procedure led to a decreased extubation time [hazard ratio (HR), 1.81; 95% confidence interval (CI), 1.54-2.12,p < 0.001) and to a low risk for mortality in ventilated patients (HR, 0.51; 95% CI, 0.38-0.68;p < 0.001) [29,30].
A cohort study involving seven bundle components for VAP prevention presented that the daily sedation interruption and the management of breathing patterns in order to attempt to discourage mechanical ventilation support showed a reduction in VAP rates (from 8.6 per 1,000 days to 2.0 per 1,000 days (p < 0.0001).However, there was no change in the mean mechanical ventilation time (6.8 ± 9.0 days vs. 6.9 ± 14.1 days, p = 0.78) [31].A case-control study with 535 patients achieved suitable results after the team education as a decreased PAV incidence rate from 21.6 to 11.6 per 1,000 days of ventilation, a lower mean ICU stay from 36 to 27 days, and a lower value of mechanically ventilated days from 26 to 21 days [32].As well, the PAV incidence before and after educational workshops was 22.0 vs. 17.2.The linear regression model demonstrated a reduction of 55.83% in the VAP rate at the end of the study time interval [33].
Concerning oral hygiene, biofilm formation in critical patients can worsen their health status.Several studies demonstrated intense colonization in the oropharynx after 48 hours of patient admission, mainly in those with endotracheal tube [34].Previous papers reported the presence of Pseudomonas aeruginosa, Staphylococcus aureus, Klebsiella pneumoniae, Escherichia coli, and Streptococcus pneumoniae in hospital patients, which differed from normal buccal microbiota [35].
A retrospective cohort study was performed for investigating four bundle components for VAP prevention in mechanically ventilated adults and found that oral care using chlorhexidine decreased the ventilator-associated events (HR, 0.44; 95% CI, 0.26-0.77,p < 0.05) [36,37].However, only one paper demonstrated dissimilar results, in which oral care using chlorhexidine was associated with an increased risk of mortality in ventilated patients (HR, 1.63; 95% CI, 1.15-2.31;p = 0.006) [11].A cohort study including nine bundle components for VAP prevention revealed that the highest team adherence was observed for oral hygiene.This procedure when added to other performed care as daily sedation interruption achieved a significant reduction in VAP (χ 2 = 9.16, p = 0.0103) [38,39].
Endotracheal tube (ETT) cuff pressure is one of the main components of the VAP prevention package due to this procedure avoids the secretion spread from the upper airways to the lower airways and lungs [40,41].Therefore, it is recommended that the inflated cuff pressure remains between 20 and 25 mmHg and should be checked every 4 hours.Excessive pressure can compromise the tracheal mucosa microcirculation and can cause ischemic lesions.However, if the pressure is lacking, a ventilation problem with positive pressure can occur and leakage of subglottic secretion between the tube and the trachea can arise [42,43].According to a non-controlled clinical trial, the VAP incidence rate reduced from 18.6 ± 7.8 per 1,000 days of ventilation (95% CI 8.7-14.9) to 11.8 ± 7.8 per 1,000 days of ventilation (95% CI 15.5-21.7)(p = 0.002) before and after an educational module for VAP prevention, respectively.In brief, this intervention consisted of a daily bundle checklist, daily oral care, and measurements at least twice a day of the ETT cuff pressure that was kept between 20 and 30 cm H 2O [44].
Critically ill patients with respiratory problems have a high risk of presenting venous thromboembolism (VTE) and consequently account for more days spent in mechanical ventilation, ICU stays, and hospitalization [45].A retrospective study based on prediction scores found that 65.2% of patients did not receive suitable coagulation prophylaxis (p < 0.001).Hence, it was concluded that coagulation prophylaxis is underused in hospital patients with pulmonary diseases by taking into consideration the scoring system used [46,47].Severe patients, mechanical ventilation, and medications such as sedatives were the main risk factors that contributed to the high risk of VTE.

Conclusions
In summary, this systematic review identified that the care bundle implementation was suitable for preventing VAP incidence in adults and the elderly.The success of the care bundle on VAP prevention demanded three key points: (a) regular training of health professionals, (b) teamwork, and (c) continuous feedback.Moreover, the VAP bundle implementation and maintenance required the regular effort of all ICU staff.In this context, the results demonstrated that the training ensured high adherence to the ICU care protocols and suitable prevention of the VAP incidence as evidenced by the decrease in both the ICU days and the mechanical ventilation days.

Figure 1 .
Figure 1.Flow through the different phases of systematic review.

Table 2 .
Data extraction from included articles.

Table 1 .
Search strategy used in the chosen databases.Turkey Case-control 133 patients, 37 cases and 96 controls; The mean age was 60.32 ± 21.6 yr. for cases and 61.34 ± 19.8 yr. for control patients (p = 0.7); Comorbidities: respiratory disease, cardiac disease, and trauma.Control group was carried out during the first 6 mos; patients intubated with endotracheal tube with subglottic secretion drainage were included as cases.Health professionals have had training.VAP per 1,000 days was significantly higher for controls(40.8)than for cases (22.1) (p < 0.05).There was a significant increase in adherence of cuff pressure measurement, use of subglottic drainage, and oral care with chlorhexidine for cases compared to controls.

Table 3 .
Description of PAV bundle items extracted from included papers.