Decreasing ventilator-associated pneumonia in adult intensive care units using the Institute for Healthcare Improvement bundle
←
→
Page content transcription
If your browser does not render page correctly, please read the page content below
Decreasing ventilator-associated
pneumonia in adult intensive care units
using the Institute for Healthcare
Improvement bundle
Jaffar A. Al-Tawfiq, MD, FACP,a and Mahmoud S. Abed, BS, RNb
Dhahran, Saudi Arabia
Background: Ventilator-associated pneumonia (VAP) increases in-hospital mortality of ventilated patients to 46%, compared with
32% for ventilated patients who do not develop VAP. In addition, VAP prolongs time spent on the ventilator, length of intensive care
unit (ICU) stay, and length of hospital stay.
Methods: In this study, we implemented VAP bundle to decrease the rate of VAP infection. This is a pre- and postintervention trial
beginning in 2006 to decrease the rate of VAP in adult ICUs after initiation of the Institute for Healthcare Improvement (IHI) VAP
bundle compared with the VAP rate for the preceding 12 months. The study was conducted at a private general hospital in Saudi
Arabia. The study included all adult patients who were on mechanical ventilation from 2006 to 2008. An interdisciplinary perfor-
mance improvement team was formed. The team implemented an evidence-based VAP bundle adopted from the IHI.
Results: The implementation of the VAP prevention bundle resulted in the reduction of VAP rates from a mean of 9.3 cases per
1000 ventilator-days in fiscal year 2006 to 2.3 cases per 1000 ventilator-days in 2007 and to 2.2 in 2008 (P , .001). It is estimated
that each VAP case increases the hospital length of stay attributable by 10 days and the mean hospital cost by $40,000. Thus, the
potential decrease in hospital cost is $780,000 annually.
Conclusion: Implementing the IHI VAP bundle significantly resulted in the reduction of the VAP rate with potential great cost
avoidance.
Key Words: Ventilator-associated pneumonia, health care-associated infection; VAP; IHI; Institute for Healthcare Improvement;
bundle; quality improvement; device-related infections.
Copyright ª 2010 by the Association for Professionals in Infection Control and Epidemiology, Inc. Published by Elsevier Inc. All rights
reserved. (Am J Infect Control 2010;38:552-6.)
Ventilator-associated pneumonia (VAP) is a serious surveillance study of VAP in a pediatric intensive care
health care-associated infection, resulting in high mor- unit (ICU) in Saudi Arabia, the mean VAP rate was 8.87/
bidity, high mortality, and high costs of treatment.1,2 1000 ventilation-days with a ventilation utilization rate
The incidence of VAP ranges from 10% to 25%, with of 47%.4 The mean pediatric ICU stay was 34 days for
mortality of 10% to 40%.3 VAP also prolongs hospital VAP patients and 15 days for non-VAP patients.4 In an-
stay and drives up hospital costs.3 In a prospective other study, the incidence density was 16.8/1000
person-days of ventilation.5 The Centers for Disease
Control and Prevention (CDC) National Healthcare
From the Internal Medicine Divisiona and Infection Control,b Dhahran Safety Network hospitals report a mean VAP rate of 3.6
Health Center, Saudi Aramco Medical Services Organization, Saudi per 1000 ventilator-days in medical-surgical ICUs.6 In
Aramco, Dhahran, Saudi Arabia.
developing countries, the rates of VAP vary from 10 to
Address correspondence to Jaffar A. Al-Tawfiq, MD, FACP, PO Box 76,
Room A-428-2, Building 61, Dhahran Health Center, Saudi Aramco, Dhah-
41.7 per 1000 ventilator-days.7
ran 31311, Saudi Arabia. E-mail: jaffar.tawfiq@aramco.com or jaltawfi@yahoo. In December 2004, the Institute for Healthcare Im-
com. provement (IHI) started a campaign to save 100,000
The authors thank the Saudi Aramco Medical Services Organization lives by implementing bundles. The concept of the
(SAMSO) facilities for the research data utilized in this paper. Opinions care ‘‘bundle’’ works to facilitate the application of
expressed in this article are those of the authors and not necessarily of best practices and evidence-based care. A bundle is
SAMSO.
‘‘a structured way of improving the processes of care
Conflicts of interest: None to report. and patient outcomes that, when performed collec-
0196-6553/$36.00 tively and reliably, are proven to improve patient out-
Copyright ª 2010 by the Association for Professionals in Infection comes.’’8 Therefore, we undertook this study in the
Control and Epidemiology, Inc. Published by Elsevier Inc. All rights beginning of 2007 to decrease the rate of VAP in adult
reserved.
ICUs after initiation of the IHI VAP bundle compared
doi:10.1016/j.ajic.2010.01.008
with the VAP rate for the preceding 12 months.
552www.ajicjournal.org Al-Tawfiq and Abed 553
Vol. 38 No. 7
METHODS
This study was conducted in the 18-bed adult ICU in-
cluding the medical, surgical, and coronary care units
at Dhahran Health Center, a 350-bed hospital in East-
ern Saudi Arabia. Dhahran Health Center has 5 ICUs
(cardiac, medical, surgical, pediatric, and neonatal).
The study was conducted at the adult ICUs.
Design
In this preinterventional and postinterventional
study, we compared the rates of VAP, on a monthly ba-
sis, from January to December 2006, for a 1-year period
before the initiation of the VAP bundles. This rate was
compared with the VAP rates after intervention from
January 2007 to December 2008 (a 2-year period).
Fig 1. Box-and-whisker plot of VAP rates pre- and
Intervention postintervention.
In addition to routine infection control protocols, the
VAP bundle was adapted from the IHI bundle of care.9 of an audit tool, data collection, and tracking of the
The IHI bundle is composed of (1) head-of-bed elevation, measures. A team approach to drive and maintain
(2) a daily ‘‘sedation vacation,’’ (3) a readiness-to-wean the initiative was developed and included the follow-
assessment, (4) peptic ulcer disease prophylaxis, and ing: infection control professional, critical care nurs-
(5) deep vein thrombosis prophylaxis.9 A sedation vaca- ing, respiratory therapist, intensivists, and chairman
tion is implemented by stopping a continuous sedation of the infection control committee. A protocol was de-
infusion until a patient is able to follow commands or be- veloped for sedation vacation and assessment to
comes fully awake. If continued sedation is required after extubate.
the vacation period, clinicians administer a bolus seda- Staff education was accomplished by multiple pre-
tion dose to the patient before restarting the sedative in- sentations on VAP, the importance of the VAP bundle,
fusion at half of the previous rate. We included all adult discussion on the VAP elements, and encouraging the
patients in the adult ICUs undergoing mechanical intensivist to adopt a protocol for the sedation vaca-
ventilation. tion. After the initial educational session, a working
group worked on the development of the VAP bundle
Measures checklist. The compliance with the bundle elements
was recorded on daily basis using a checklist during a
Two sets of measures were monitored: process and multidisciplinary round. Feedback was provided on
outcome measures. The process measure reflected compliance with these processes to the ICU team (doc-
the rate of adherence to the ventilator bundle. The out- tors, nurses, and respiratory therapists). If noncompli-
come measure included the device utilization ratio and ance with an element of the bundle was detected, the
the rate of VAP per 1000 ventilator-days. The device uti- nurse intervened in this process at the time of the mon-
lization ratio was calculated by dividing the ventilator- itoring process. A quarterly report of the compliance
days by the total ICU patient-days. The definition of rate and the VAP rates were communicated to each
VAP was based on the CDC’s National Nosocomial In- ICU staff and was shared with the quality improvement
fections Surveillance System definitions.10 We also es- and the executive staff committees.
timated the potential cost avoidance based on the
average attributable cost estimates of a single case of VAP definitions
VAP as $40,000.11 The estimation of additional hospital
stay was based on the report that patients with VAP had Based on the infection control committee guidelines
a significantly longer duration of hospital length of stay in our hospital, VAP surveillance was performed by a
of about 10 days.12 trained infection control professional using the US
CDC definition.10 A single infection control profes-
Implementation process sional, who has more than 20 years of experience in in-
fection control, collected the data on the development
Multiple steps were taken to implement the ventila- of the VAP using a standardized CDC criteria-based
tor bundle and include staff education, development checklist. The infection control professional and the554 Al-Tawfiq and Abed American Journal of Infection Control
September 2010
Fig 2. A run chart showing the monthly rate of VAP before and after the implementation of the VAP bundle. Dotted
lines indicate the upper control limits, and the solid straight lines indicate the annual mean rate per 1000 ventilator-days.
The x-axis shows the months from January 2006 to December 2009.
ICU team were not blinded to the study because the It is estimated that each VAP case increases the hos-
study was done as a part of the quality improvement pital length of stay attributable by 10 days and the
projects. The incidence of VAP was expressed as cases mean hospital cost by $40,000.11 Thus, the potential
of VAP per 1000 ventilator-days. reduction in hospital days was 115 days (290 vs 95
days, respectively) (Table 1). The attributed VAP costs
RESULTS were $380,000 after implementation of the VAP pre-
vention bundle with a potential avoidable cost of
Adherence with all the elements of the VAP bundle $780,000 annually (Table 1).
improved from 20% in the first 3 months to 82% in
the final 3 months of the intervention. There were 29 DISCUSSION
VAP events (9.3 events/1000 ventilator-days) during
the 12-month period from January 2006 through De- In the current study, we adopted the IHI VAP bundle
cember 2006 compared with 10 VAP events (3 events/ that resulted in a significant decrease in the VAP rate
1000 ventilator-days) in 2007 and 9 VAP events (2.1 and maintained such a reduction over 24 months. In
events/1000 ventilator-days) in 2008 (Fig 1). Thus, the Saudi Arabia, 2 previous studies were conducted to de-
annual average VAP rate in the postintervention period crease the VAP rate.5,13 In a prospective randomized
was 2.5/1000 ventilator-days. This improvement ap- study in a tertiary care ICU, authors compared the inci-
pears to be sustained over the study period (Fig 2). dence of VAP using the heat and moisture exchanger
The steepest decrease in VAP rate was observed after versus the heated humidifying system. They found
the initial full implementation of the bundle and after- no significant difference in the incidence of VAP
wards reached a plateau. As a measure of the acuity between the 2 systems in which VAP rates were 15.7
and invasiveness in the ICUs, the device utilization ratio and 13.3 per 1000 ventilator-days, respectively.5 In
was 0.55. Thus, the device utilization ratio did not the same ICU, Arabi et al showed significant reductions
change over the study period and was comparable be- in the use of analgesics and sedatives after 3 months
tween pre- and postintervention. of implementing a multifaceted multidisciplinarywww.ajicjournal.org Al-Tawfiq and Abed 555
Vol. 38 No. 7
Table 1. A comparison between pre- and postintervention parameters
No. of VAP rates per 1000 Total Added LOS Costs in $ 1 (added)$ 2
Period infections ventilator-days ventilator-days (10 days/VAP) (avoided) $40,000/inf
January-December 2006 (before VAP 29 9.3 3126 290 ($1,160,000)*
bundle application)
Average annual (January 2007- 9.5 2.5 3740 95 1($380,000)2($780,000)*
December 2008) (after VAP
bundle application)
inf, infection; LOS, length of stay.
*Avoidable cost.
approach and teamwork. This was associated with a between nursing staff and the critical care pharmacist
decrease in the VAP rate from 28 to 11 per 100 was important to develop a protocol for these
patients.13 elements.23
We observed an 80% reduction in the rate of VAP; In conclusion, we adopted an aggressive VAP bundle
however, a lower reduction rate of 44.5% reduction resulting in a marked reduction of the VAP rate. The
was observed by other investigators.14 The application rate was kept in the lower range even after 2 years of
of the VAP bundle in Minnesota resulted in reduction of implementation. Thus, such measures are effective
the VAP rate from 6.1 to 2.70 per 1000 ventilator-days and require staff training and a multidisciplinary pro-
in one unit and from 2.66 to 0 per 1000 ventilator- gram.24 A well-developed and supported program will
days in another unit.15 In another study, the use of a enhance the success rate of such interventions, espe-
collaborative approach resulted in a decrease of the cially if coupled with data feedback.
VAP rate from 13.3 to 8.3 per 1000 ventilator-days.16
In another study, the application of the VAP bundle re-
sulted in a 41% reduction in VAP rate.17 In a study from References
the United Kingdom, the VAP incidence fell signifi- 1. Bercault N, Boulain T. Mortality rate attributable to ventilator-
cantly from 19.2 to 7.5 per 1000 ventilator-days.18 Sim- associated nosocomial pneumonia in an adult intensive care unit: a
prospective case control study. Crit Care Med 2001;29:2303-9.
ilar to the finding in the current study, a near zero VAP 2. Leroy O, Sanders V, Girardie P, Devos P, Yazdanpanah Y, Georges H, et al.
rate was accomplished after the implementation of an Mortality due to ventilator-associated pneumonia: impact of medical
expanded respiratory therapist-driven set of protocols versus surgical ICU admittance status. J Crit Care 2001;16:90-7.
for decision regarding extubation.19 In a recent review, 3. Cocanour CS, Peninger M, Domonoske BD, Li T, Wright B, Valdivia A,
the VAP bundle was found to be an effective method to et al. Decreasing ventilator-associated pneumonia in a trauma ICU.
J Trauma 2006;61:122-30.
reduce VAP rates in ICUs. The authors suggested that 4. Almuneef M, Memish ZA, Balkhy HH, Alalem H, Abutaleb A. Ventila-
the VAP bundle should be modified and expanded to in- tor-associated pneumonia in a pediatric intensive care unit in Saudi
clude oral care and hygiene; chlorhexidine in the pos- Arabia: a 30-month prospective surveillance. Infect Control Hosp
terior pharynx; and specialized, silver-coated Epidemiol 2004;25:753-8.
endotracheal tubes (continuous aspiration of subglottic 5. Memish ZA, Cunningham G, Oni GA, Djazmati W. The incidence and
risk factors of ventilator-associated pneumonia in a Riyadh hospital. In-
secretions).20 Daily and weekly auditing of compliance fect Control Hosp Epidemiol 2000;21:271-3.
with any intervention seems to also improve the VAP 6. Edwards JR, Peterson KD, Andrus ML, Tolson JS, Goulding JS, Dudeck
rates. In one study, VAP rates significantly decrease MA, et al. National Healthcare Safety Network, (NHSN) report, data
with intervention, from a range of 22.3-32.7 to a range summary for 2006, issued June 2007. Am J Infect Control 2007;35:
of 0-12.8 per 1000 ventilator-days.3 290-301.
7. Arabi Y, Al-Shirawi N, Memish Z, Anzueto A. Ventilator-associated
During the planning and implementation process, a pneumonia in adults in developing countries: a systematic review. Int
few obstacles arose. One of the obstacles was related to J Infect Dis 2008;12:505-12.
the perception that such changes of the VAP bundle 8. Institute for Healthcare Improvement: What is a bundle? Available
may compromise patient care and presents a risk for from: http://www.ihi.org/IHI/Topics/CriticalCare/IntensiveCare/Impro
adverse events. This finding was more evident for nurs- vementStories/WhatIsaBundle.htm. Accessed October 11, 2009.
9. Institute for Healthcare Improvement: Implement the ventilator bundle.
ing staff as reported previously.21 However, physicians Available from: http://www.ihi.org/IHI/Topics/CriticalCare/Intensive
were more resistant to change related to the implemen- Care/Changes/ImplementtheVentilatorBundle.htm. Accessed October
tation of the sedation vacation. The overall nonadher- 11, 2009.
ence rate to VAP bundle in a previous study was 10. Horan TC, Andrus M, Dudeck MA. CDC/NHSN surveillance definition
37.0%.22 We gave the physician autonomy to develop of health care-associated infection and criteria for specific types of infec-
tions in the acute care setting. Am J Infect Control 2008;36:309-32.
the protocol for sedation vacation and the readiness 11. Tablan OC, Anderson LJ, Besser R, Bridges C, Hajjeh R, CDC; Health-
to extubate and thus were able to implement the VAP care Infection Control Practices Advisory Committee. Guidelines for
bundle. In a previous report, a collaborative effort preventing health care-associated pneumonia, 2003: recommendations556 Al-Tawfiq and Abed American Journal of Infection Control
September 2010
of CDC and the Healthcare Infection Control Practices Advisory Com- pneumonia and catheter-related bloodstream infection rates. Jt Comm J
mittee. MMWR Recomm Rep 2004;53(RR-3):1-36. Qual Patient Saf 2008;34:639-45.
12. Rello J, Ollendorf DA, Oster G, Vera-Llonch M, Bellm L, Redman R, 18. Hawe CS, Ellis KS, Cairns CJ, Longmate A. Reduction of ventilator-
et al. VAP Outcomes Scientific Advisory Group. Epidemiology and associated pneumonia: active versus passive guideline implementation.
outcomes of ventilator-associated pneumonia in a large US database. Intensive Care Med 2009;35:1180-6.
Chest 2002;122:2115-21. 19. Blamoun J, Alfakir M, Rella ME, Wojcik JM, Solis RA, Anees Khan M,
13. Arabi Y, Haddad S, Hawes R, Moore T, Pillay M, Naidu B, et al. Changing et al. Efficacy of an expanded ventilator bundle for the reduction of
sedation practices in the intensive care unit: protocol implementation, ventilator-associated pneumonia in the medical intensive care unit.
multifaceted multidisciplinary approach and teamwork. Middle East J Am J Infect Control 2009;37:172-5.
Anesthesiol 2007;19:429-47. 20. Wip C, Napolitano L. Bundles to prevent ventilator-associated
14. Resar R, Pronovost P, Haraden C, Simmonds T, Rainey T, Nolan T. Us- pneumonia: how valuable are they? Curr Opin Infect Dis 2009;22:
ing a bundle approach to improve ventilator care processes and re- 159-66.
duce ventilator-associated pneumonia. Jt Comm J Qual Patient Saf 21. Ricart M, Lorente C, Diaz E, Kollef MH, Rello J. Nursing adherence
2005;31:243-8. with evidence-based guidelines for preventing ventilator-associated
15. Youngquist P, Carroll M, Farber M, Macy D, Madrid P, Ronning J, et al. Im- pneumonia. Crit Care Med 2003;31:2693-6.
plementing a ventilator bundle in a community hospital. Jt Comm J Qual 22. Rello J, Lorente C, Bodı́ M, Diaz E, Ricart M, Kollef MH. Why do
Patient Saf 2007;33:219-25. physicians not follow evidence-based guidelines for preventing
16. Unahalekhaka A, Jamulitrat S, Chongsuvivatwong V, Øvretveit J. Using ventilator-associated pneumonia: a survey based on the opinions of
a collaborative to reduce ventilator-associated pneumonia in Thailand. an international panel of intensivists. Chest 2002;122:656-61.
Jt Comm J Qual Patient Saf 2007;33:387-94. 23. Evans B. Best-practice protocols: VAP prevention. Nurs Manage 2005;
17. Bonello RS, Fletcher CE, Becker WK, Clutter KL, Arjes SL, Cook JJ, et al. 36:10, 12, 14 passim.
An intensive care unit quality improvement collaborative in nine Depart- 24. Kollef MH. Prevention of hospital-associated pneumonia and ventilator-
ment of Veterans Affairs hospitals: reducing ventilator-associated associated pneumonia. Crit Care Med 2004;32:1396-405.You can also read
