Ain-Shams Journal of Anaesthesiology

: 2016  |  Volume : 9  |  Issue : 2  |  Page : 170--177

Root cause analysis of ventilator-associated pneumonia and the effect of analysis of expanded ventilator bundle of care

Hossam Ibrahim Eldesuky Ali, Ayman Ali Rayan, Tamer Hussein Ibrahim 
 Department of Anesthesia, Zagazig University, Zagazig; Department of Intensive Care, Monoufia University, Monoufia, Egypt

Correspondence Address:
Hossam Ibrahim Eldesuky Ali
100 Tabouk, Saudi Arabia


Objectives The study had the following aims: (i) assess the risk factors and causes of ventilator-associated pneumonia (VAP) by means of the root cause analysis (RCA) module; (ii) compare the compliance with expanded ventilator bundle (EVB) and the effect of EVB on VAP rate, length of ICU stay (LOS), and mortality rate before and after bundle implementation; and (iii) suggest an action plan to reduce VAP. Patients and methods An 18-month study was conducted on all ventilated patients in our ICU. The preinterventional period was 9 months before implementing the bundle. We analyzed the causes and risk factors of VAP by using the RCA module. The postinterventional period was 9 months after implementing the bundle. Compliance with EVB was compared before and after implementing the bundle on a quarterly basis. We analyzed the effect of the bundle on VAP rate, LOS, and mortality rate. We suggested an action plan to reduce the VAP rate. Results There was lower compliance with the bundle in the preinterventional period than in the postinterventional period (P < 0.001). There were highly statistically significant reductions in VAP rate, LOS, and mortality rate after implementation of EVB (P < 0.001). Conclusion EVB was associated with effective and significant reductions in VAP rate, LOS, and mortality rate. Implementation of the RCA module was helpful in suggesting a new action plan that would improve bundle of care and its compliance.

How to cite this article:
Eldesuky Ali HI, Rayan AA, Ibrahim TH. Root cause analysis of ventilator-associated pneumonia and the effect of analysis of expanded ventilator bundle of care.Ain-Shams J Anaesthesiol 2016;9:170-177

How to cite this URL:
Eldesuky Ali HI, Rayan AA, Ibrahim TH. Root cause analysis of ventilator-associated pneumonia and the effect of analysis of expanded ventilator bundle of care. Ain-Shams J Anaesthesiol [serial online] 2016 [cited 2020 Jul 15 ];9:170-177
Available from:

Full Text


Ventilator-associated pneumonia (VAP) is the second most common healthcare-associated infection. It is responsible for 25% of the infections that occur in the ICU. VAP is associated with increasing death rates, length of hospital stay, and healthcare costs. The death rate from VAP exceeds the rate of death due to infections associated with central venous catheter, severe sepsis, and respiratory tract infections in nonintubated patients. VAP patients had an ICU stay that was 4-19 days longer than that of intubated patients who did not acquire VAP [1]. Among patients treated with mechanical ventilation (MV), the mortality rate was 46% in patients with VAP and 32% in those without VAP [2]. The VAP bundle has been incorporated into a set of treatment modalities implemented simultaneously to reduce VAP incidence. There are limited data on the exact method of functioning of the bundle in preventing VAP [3]. The current study aimed to evaluate and assess the causes and risk factors of VAP using the root cause analysis (RCA) module. This study used a modification of ventilator bundle [expanded ventilator bundle (EVB)] as a trial to reduce the rate of VAP, mortality rate, and length of ICU stay (LOS). A new action plan was concluded from the RCA of VAP.

 Patients and methods

The study was conducted in a 12-bed ICU for adults of both sexes at King Abdul-Aziz Central Hospital in Saudi Arabia, which is a 550-bed tertiary hospital. It has about 600 annual ICU admissions. The patients admitted to other care units were not included in the study. The study period was from April 2012 to September 2013. Approval of the hospital ethical committee and informed consent from patients were obtained. All patients were aged above 18 years and from both sexes. All patients undergoing invasive MV were included in the study. The study period was divided into preinterventional period and postinterventional period, of a duration of 9 months each, on a quarterly basis. The staff responsible for implementing the EVB were not made aware that data would be evaluated in the current study. Infection control strategies were strictly enforced throughout the period of study. Centers for Diseases Control and Prevention defines VAP as a condition in which a patient receiving MV has new or progressive pulmonary infiltrates along with fever, leukocytosis, and purulent tracheobronchial secretions. Pneumonia was considered ventilator-associated if the patient was intubated and receiving MV at the time of or within 48 h before the onset of infection [4]. VAP was diagnosed by two-stage definition: first, clinically suspected VAP based on clinical criteria and further investigations such as chest radiograph (new infiltrates, cavity, or consolidation); second, on the basis of culture and sensitivity tests of microbiological tracheal aspirates. Patients were screened for infection on admission and monitored for signs of developing infection after admission - for example, temperature more than 38°C or less than 36°C, heart rate 90 beats/min or more, respiratory rate 20 breaths/min or more, white blood cells 12 000 cells/mm 3 or more, or more than 10% bands. The decision to start broad-spectrum antibiotics was based on the fulfillment of two or more criteria until the pathogen was isolated. All patients were screened for methicillin-resistant Staphylococcus aureus (MRSA). During the preinterventional period, data were collected and analyzed retrospectively by completing the RCA forum and before implementing the expanded bundle. The RCA forum was similar to the forum of the Strong Memorial Hospital with new modifications [5]. Analysis of causes and risk factors are summarized in the RCA forum. The RCA module started by organizing a team consisting of physicians, infection control staff, staff nurses, and respiratory therapists. Thereafter, the problem was defined (increasing rate of VAP), studied (causes and risk factors as per the RCA forum), and measured, and data were collected. We identified a strategy (EVB), improved our bundle compliance, and suggested an action plan. Finally we followed up our compliance with the bundle, documented results, and obtained feedback on our results. All causes and risk factors were recorded according to the RCA forum as basal data of VAP before implementation of EVB. During the preinterventional period we applied the unit protocol and guidelines to reduce VAP - such as infection control guidelines, weaning protocols, sedation vacation, spontaneous breathing trial (SBT), and peptic ulcer disease (PUD) and deep vein thrombosis (DVT) prophylaxis - but EVB as a whole was not implemented. During the postinterventional period, data were collected and analyzed prospectively. All components of EVB were implemented on all patients and at the same time. The ventilator bundle included the following:

Elevation of the head of the bed (HOB) to 30-45°;Daily interruption of sedation and readiness for extubation;Peptic ulcer (PUD) and DVT prophylaxis;Oral care and hygiene (mouth care with chlorhexidine ≥ 1-2% as gel or liquid every 6 h and brushing of teeth every 12 h after 2 h from mouth care to avoid inactivation of chlorhexidine by standard toothpaste);Subglottic aspiration (suction) by means of a special endotracheal tube (ETT) with sideway tube for continuous drainage of secretion every 2 h; and Tracheal tube cuff pressure monitoring using a bedside cuff manometer every 4 h (20-30 cmH 2 O).Bundle care from (1) to (3) was the ventilator bundle. Bundle care from (4) to (6) was the expanded or modified bundle [6,7]. Compliance with the bundle was assessed by completing the compliance audit daily (yes or no). The compliance percentage was calculated and compared quarterly. Compliance with the bundle was audited daily at unannounced and variable times by an independent, nonclinically trained audit clerk who audited charts from the previous day to minimize observation bias. Compliance was defined as clear documentation of all elements of the bundle together at the same time for all patients. Failure to complete the bundle was noted as lack of compliance. The bundle was considered compliant if all items were compliant. The bundle was considered noncompliant if any item was contraindicated. The only exception was contraindication to DVT prophylaxis in patients with head injury; in such cases the bundle was still considered compliant. VAP was diagnosed on the basis of clinical criteria as before or by microbiological confirmation and was recorded and compared before and after implementation of the bundle. VAP rate was calculated as the number of patients suspected to have VAP divided by the total days on MV multiplied by a factor of 1000. LOS was calculated as the total number of ICU days divided by the total number of ICU admissions. Mortality rate of VAP was calculated as total number of deaths due to VAP divided by the total number of ICU discharges, included deaths, multiplied by 100. The total ICU death rate was also reported. Total number of ventilator days, total number of free days from ventilation, total number of ICU days, and total number of ICU admissions and discharges were recorded.

Statistical analysis

Demographic data such as age, sex, and Acute Physiology and Chronic Health Evaluation II (APACHE II) score were obtained for all patients. The number of ICU admissions and discharges, number of ventilated patients, ventilation days, ICU days, and number of VAP cases were obtained separately before and after implementation of the bundle from our database. The Student t-test, the Mann-Whitney test, and the Z-test were performed to compare mean and proportions before and after the bundle. Changes were also assessed from statistical process control charts containing monthly data obtained from observation periods. Total compliance rate and percentage were calculated manually from audit sheets. P value less than 0.05 was considered statistically significant; P value more than 0.05 was considered statistically nonsignificant; and P value less than 0.001 was considered statistically highly significant.


There was no statistically significant difference as regards demographic data between preinterventional and postinterventional periods (P > 0.05) [Table 1]. During the preintervention period, 458 patients were admitted to the ICU for 3840 days; 49.3% were ventilated for different reasons and were in the ICU for 1944 days on MV. During the postinterventional period, after applying the EVB, 454 patients were admitted to the same unit but with no statistically significant difference (P > 0. 05); however, there was statistically significant reduction in total ICU days, number of ventilated days, total ventilation days, and total free days from ventilation (P < 0.001). LOS was statistically reduced after bundle implementation, from 7.94 to 5.02 days (P < 0.001). There were no significant differences between interventional periods as regards total number of discharged patients and total number of antibiotic days (P > 0.05) [Table 2].{Table 1}{Table 2}

During the preintervention period, elements of the bundle such as sedation vacation and prophylaxis of PUD and DVT were compliant (91-92% and 90-93%, respectively), whereas other elements were not compliant (50-65%); overall, elements of the bundle were not compliant (68.8-72%). After bundle implementation, compliance with the bundle increased from 92.1 to 95.8%; also, all individual elements showed greater improvement compared with the preinterventional period (P < 0.001). Elevation of the HOB was a little improved compared with other elements of the bundle [Table 3].{Table 3}

During the preinterventional period the total number of patients with suspected VAP was 50; VAP % was high (22.3%) and VAP rate was 25.7 patients for 1000 ventilation days. After implementation of the bundle, there was marked statistically significant reduction in the number of patients with suspected VAP and in the VAP rate (P < 0.001). There was reduction in the number and percentage of VAP (P = 0.050) [Table 4]. When VAP rate was compared with total ventilation days on a quarterly basis, there were statistically significant reductions in VAP numbers, rate, and total ventilation days in the postinterventional period compared with the preinterventional one (P < 0.001) [Figure 1] [Figure 2] [Figure 3].{Figure 1}{Figure 2}{Figure 3}{Table 4}

There were no statistically significant differences before or after implementation of the bundle as regards clinical diagnosis of suspected VAP and its rate or microbiological diagnosis (P > 0.05) [Table 5]. However, there were reductions in the rate of VAP after bundle implementation.{Table 5}

There are many risks and precipitating factors affecting VAP according to the RCA forum [Table 6]. There were no statistically significant differences between the causes and risk factors of VAP before and after bundle implementation (P > 0.05) except in prolonged ventilation, weaning trials, and circuit and suction tube changing, which were statistically significant (P < 0.05). There was a statistically significant reduction in MRSA infection after bundle implementation (31-8.3%). There were no significant differences as regards other isolated organisms (P > 0.05) but the prevalence of late and early isolated organisms was reduced after bundle implementation [Table 7].{Table 6}{Table 7}

The mortality rate related to VAP was significantly reduced after bundle implementation (22-8.3%) and ICU mortality related to other causes reduced from 5.9 to 3.4% [Table 8].{Table 8}

New preventive strategies were concluded from the RCA forum. Each item from the RCA forum was covered by one or more suggestions as action plans after RCA [Table 9].{Table 9}


Evidence-based clinical practice guidelines including dozens of clear preventive strategies did not result in implementation in the clinical setting, nor in VAP prevention [8]. In our study, during the preinterventional period we did not have a systematic approach for controlling VAP, but we followed ICU policy, infection control guidelines, and different bundle elements, all of which were insufficient to control VAP, as seen in the study by Marra et al. [9]. The preinterventional period was an optimal time for brainstorming of all causes and related risk factors of VAP summarized in the RCA forum. It was different from the forum of Strong Memorial Hospital, which included only intubation, ventilator use, treatment-related factors, and the bundle itself. Our forum can predict direct causes or risk factors of VAP. Our RCA forum was based on clinical evidence based on previous studies [10,11]. But the difference compared with their studies was that we formulated an action plan for each factor in detail, with positive feedback results on VAP, together with implementation of the expanded bundle and improving compliance with the bundle at the same time. Effective auditing of bundle compliance facilitated rapid feedback to our staff on whether their performance was in line with bundle care and its impact on the quality of patient care [12]. The concept of the VAP bundle has grown from live campaigns that involved only the first three elements of our bundle [13]. Ventilator bundle of care originated from and was modified at the Institute for Healthcare Improvements [14]. The key components of ventilator care bundles were the first three elements of the bundle. Youngquist et al. [15] described that adherence to ventilator bundles was followed by a reduction in VAP rates. Consequently, the VAP bundle was created out of the ventilator bundle and they were similar because they incorporated a series of interventions related to ventilator care. The difference between the two bundles lay in their components. All components of the VAP bundle directly reduced VAP, except DVT and PUD prophylaxis, which improved the outcome of mechanically ventilated patients and did not affect the VAP. Therefore, in our study there were modifications that helped in reducing the VAP rate and improving ventilation outcomes at the same time; this was called EVB of care, in agreement with Hatler et al. [16]. Our ventilation bundle of care started with HOB elevation, which was neglected by our staff, and therefore compliance was 75% or less; this improved after the bundle program to 85% or more. Our action plan improved that element to a large extent by introducing rotating beds or bed alarms and daily checkup of our documentation and audit. We suggested an alternative method to overcome the problem of bed elevation in semi-recumbent position that did not prevent aspiration of subglottic secretion across tracheal cuff, and secretion of lower respiratory tract cannot be cleared. The suggestion of laying the patients in the lateral horizontal position solved the problem, as seen in the study by Mauri and colleagues. They found more ventilation-free days and a trend toward a lower incidence of VAP in the lateral horizontal position. Their study found that, when the end tracheal tube was in the horizontal plane of the bed or slightly below, the mucus flow reversed toward the lung in semirecumbent position and moved away from the lung in the horizontal position [17]. The second element of ventilation bundle (daily sedation vacation) was implemented before as routine ICU policy with high compliant rate 92% or less and there was more improvement after our bundle program. It provided more reduction in ventilation days, length of stay, and VAP rate because it allowed more time to perform SBTs, achieve planned extubation, and encouraged the patients to resume his or her cough reflex to clear the airway. The recommendation allowed daily vacation or complete holding off of sedation to perform SBT, or performing breathing trials that would help the patient to breath without any assistance from a ventilator, depending only on the patient's efforts, in agreement with the study by Girard et al. [18]. The third element was prophylaxis against peptic ulcer and DVT, which was not a specific strategy for VAP prevention but was included in the ventilator bundle as a strategy to prevent stress-related mucosal disease, as MV was a significant risk factor. It remained part of the ventilator bundle to prevent other serious complications. The compliance rate was high and improved after the bundle and we recommended use of proper treatment or prophylactic treatment for PUD with proton pump blockers and by avoiding gastric alkalinization. Thus, it defends the airway against acidic contents. The first three elements of the bundle are called ventilator bundle or modified VAP bundle, and most of the studies demonstrated the positive impact on reduction of VAP but they did not report the bundle compliance rates and did not control for other specific VAP risk factors. Implementation of the expanded bundle has been associated with a reduction in VAP in published studies [19-21]. We advocated three additions to the ventilator bundle: the fourth element was oral care with chlorhexidine antiseptic. We reported that the compliance rate was very low during the preinterventional period by 54% or less because it was not routine practice in our ICU. Thus, we started to implement it as routine oral use every 6 h daily, with regular brushing of teeth twice daily for all patients except for those with allergy to it. The compliance rate increased to 96%. We added a new recommendation for severe head injury patients to use povidone-iodione instead of it. There was positive feedback by way of reduction in VAP rate and ventilation days. Our results matched those of a previous study that used oral antiseptic chlorhexidine in 2144 patients [22]. The fifth element was subglottic secretion drainage. The compliance rate was low (≤65%) because continuous tracheal suction caused mucosal injuries. The rate improved after bundle implementation to 99% when we began to use a subglottic secretion drainage system as routine practice in our ICU, which consisted of an accessory aspiration conduit opening above the ETT cuff and a vacuum source. This element was the only protective factor against VAP and played an important role in reduction of VAP and reduction of ventilation days and LOS. Our results matched those of a randomized study that involved 712 patients who had undergone cardiac surgery [23]. The last element of the expanded bundle was tracheal cuff care and monitoring. It was a new modification to EVB. The rate of compliance was also low, but it improved after implementation of the bundle (55-99%). The maintenance of cuff pressure is vital to reduce draining of oropharyngeal and gastric secretions around the ETT. An inflating cuff pressure less than 20 cmH 2 O may favor secretion aspiration, whereas greater than 30 cmH 2 O may result in mucosal ischemia. We recommend the use of a cuff manometer at the bedside every 4 h daily to control ETT cuff pressure to between 20 and 30 cmH 2 O. The previous randomized trial on 122 patients confirmed its effectiveness in reducing microaspiration and lowering the VAP rate with no evident adverse effects [24]. From our results we found that the total compliance rate was 52% because we did not apply the EVB; instead, we applied only infection control and ICU policy, which were not sufficient. The higher VAP rate of 25.7/1000 ventilation days in the preinterventional period was alarming, driving us to use the EVB. We started to analyze all risk factors related to VAP by using the RCA module through brainstorming sessions using previous data. The modified EVB and our new recommendations were helpful to cover most of risk factors and previous bundle deficiencies. After the bundle implementation, the compliance rate of the bundle increased to our target 95% or more. The VAP rate reduced dramatically to 7.79/1000 ventilation days. The ventilation days, LOS, and MRSA were all reduced. The total number of free days also reduced because the total number of ICU days and total ventilation days before the bundle were more than that after the bundle implementation. The total mortality rate in the ICU reduced from 5.9 to 3.4% and mortality rate related to VAP reduced from 22 to 8.3%. The suggestions from the RCA were helpful in reducing the duration of antibiotic usage from a mean 7 days to a mean 4.5 days; this may be helpful in reducing the colonization of Candida, which was not a cause of VAP but was considered a risk factor. The current study used a different trial to reduce the VAP rate by applying the RCA forum for the risk factors and causes and then analyzing the risk factors, which were statistically nonsignificant before and after implementation of the bundle in most of the cases due to the difference between numbers of suspected VAP, but it was effective in reducing the risk factors or causes of VAP. The RCA module was helpful in organizing a team, defining the problem (increasing VAP rate), studying the problem (brainstorming), determining the causes and risk factors (RCA forum), measuring and collecting data, and identifying the reduction strategies (EVB and suggestions), which covered all causes and risk factors. We then followed up the effect of EVB on the VAP rate, LOS, and mortality rate. We improved bundle compliance, documentation of the improvement of VAP, and performance of the staff.


The RCA forum was helpful in detecting the causes and risk factors of VAP. The EVB succeeded in reducing the incidence of VAP, the number of ventilation days, LOS, and the morality rate, simultaneously maintaining excellent compliance. Compliance was achieved by incorporating the bundle and new suggestions into daily multidisciplinary rounds and through regular documentation, audit, and feedback.

Limitation of the study

Among the weakness of this study, we did not account for the contraindications to bundle items when scoring compliance. This may have led to underestimation of actual guideline compliance. We did not calculate the incidence of VAP among noninvasive mechanically ventilated patients. We did not consider the seasonal changes affecting VAP. The strengths of our study included the creation of a systematic approach for detection of the causes and risk factors for VAP using the RCA and reducing the incidence of VAP with EVB. The compliance rate of the bundle increased to the target needed and our staff became more motivated to improve their performance as well as documentation of audits.


Conflicts of interest

None declared.


1Restrepo MI, Anzueto A, Arroliga AC, Afessa B, Atkinson MJ, Ho NJ, et al. Economic burden of ventilator-associated pneumonia based on total resource utilization. Infect Control Hosp Epidemiol 2010; 31:509-515.
2Bekaert M, Timsit JF, Vansteelandt S, Depuydt P, Vésin A, Garrouste-Orgeas M, et al. Attributable mortality of ventilator-associated pneumonia: a reappraisal using causal analysis. Am J Respir Crit Care Med 2011; 184:1133-1139.
3Morris AC, Hay AW, Swann DG, Everingham K, McCulloch C, McNulty J, et al. Reducing ventilator-associated pneumonia in intensive care: impact of implementing a care bundle. Crit Care Med 2011; 39:2218-2224.
4Division of Healthcare Quality Promotion, National Center for Preparedness, Detection and Control of Infectious Diseases. National Health and Safety Network (NHSN) manual: patient safety component protocol. Available from: http// [Accessed 3 May 2012].
5Thomsen GE, Snow GL, Rodriguez L, Hopkins RO. Ventilator associated pneumonia root cause investigation/analysis. Crit Care Med 2008; 36:1119-1124.
6Cason C, Tyner T, Sunder S. Center for Diseases Control and Prevention. Nurses' implementation of guidelines for ventilator-associated pneumonia from the Center for Diseases Control and Prevention. Am J Crit Care 2007; 16:28-34.
7Tolentino-Delos Reyes AF, Ruppert SD, Shiao SY. Evidence-based practice: use of the ventilator bundle to prevent ventilator-associated pneumonia. Am J Crit Care 2007; 16:20-27.
8Safdar N, Dezfulian C, Collard HR, Saint S. Clinical and economic consequences of ventilator-associated pneumonia: a systemic review. Crit Care Med 2005; 33:2184-2193.
9Marra AR, Cal RG, Silva CV, Caserta RA, Paes AT, Moura DF Jr, et al. Successful prevention of ventilator-associated pneumonia in an intensive care setting. Am J Infect Control 2009; 37:619-625.
10Flanders SA, Collard HR, Saint S. Nosocomial pneumonia: state of the science. Am J Infect Control 2006; 34:84-93.
11Million Lives C. Getting started kit. How-to guide: prevention ventilator-associated pneumonia. Cambridge, MA: Institute for Healthcare Improvement; 2010.
12Cocanour CS, Peninger M, Domonoske BD, Li T, Wright B, Valdivia A, Luther KM. Decreasing ventilator-associated pneumonia in a trauma ICU. J Trauma 2006; 61:122-129; discussion 129-130.
13Berenholtz SM, Dorman T, Ngo K, Pronovost PJ. Qualitative review of intensive care unit quality indicators. J Crit Care 2002; 17:1-12.
14Resar 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; 31:243-248.
15Youngquist P, Carroll M, Farber M, Macy D, Madrid P, Ronning J, Susag A. Implementing a ventilator bundle in a community hospital. Jt Comm J Qual Patient Saf 2007; 33:219-225.
16Hatler CW, Mast D, Corderella J, Mitchell G, Howard K, Aragon J, Bedker D. Using evidence and process improvement strategies to enhance healthcare outcomes for the critically ill: a pilot project. Am J Crit Care 2006; 15:549-555.
17Mauri T, Berra L, Kumwilaisak K, Pivi S, Ufberg JW, Kueppers F, et al. Lateral-horizontal patient position and horizontal orientation of the endotracheal tube to prevent aspiration in adult surgical intensive care unit patients: a feasibility study. Respir Care 2010; 55:294-302.
18Girard T, Kress J, Fuchs B, Thomason JW, Schweickert WD, Pun BT, et al. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care unit (Awaking and Breathing Controlled Trial): a randomized controlled trial. Lancet 2008; 371:126-134.
19Blamoun J, Alfakir M, Rella ME, Wojcik JM, Solis RA, Anees Khan M, DeBari VA. Efficacy of an expanded ventilator bundle for the reduction of ventilator-associated pneumonia in the medical intensive care unit. Am J Infect Control 2009; 37:172-175.
20Krein SL, Kowalski CP, Damschroder L, Forman J, Kaufman SR, Saint S. Preventing ventilator-associated pneumonia in the United States: a multicenter mixed-methods study. Infect Control Hosp Epidemiol 2008; 29:933-940.
21Lansford T, Moncure M, Carlton E, Endress R, Shik N, Udobi K, et al. Efficacy of a pneumonia prevention protocol in the reduction of ventilator-associated pneumonia in trauma patients. Surg Infect (Larchmt) 2007; 8:505-510.
22Chan EY, Ruest A, Meade MO, Cook DJ. Oral decontamination for prevention of pneumonia in mechanically ventilated adults: systematic review and meta-analysis. BMJ 2007; 334:889.
23Bouza E, Pérez MJ, Muñoz P, Rincón C, Barrio JM, Hortal J. Continuous aspiration of subglottic secretions in the prevention of ventilator-associated pneumonia in the postoperative period of major heart surgery. Chest 2008; 134:938-946.
24Nseir S, Zerimech F, Fournier C, Lubret R, Ramon P, Durocher A, et al. Continuous control of tracheal cuff pressure and micro aspiration of gastric contents in critically ill patients. Am J Respir Crit Care Med 2011; 184:1041-1047.