1. A Retrospective Case-Control Study From A Level I Trauma Center
Risk Factors For Ventilator-Associated Pneumonia In The Adult Trauma Patient:
A Retrospective Case-Control Study From A Level I Trauma Center
Summer Chapman
DMPNA Candidate

2. Research Committee Dr. Alberto Coustasse, DrPH, MD, MBA, MPH
Committee Chair, College of Business, Marshall University
Dr. Cassy Taylor, CRNA, DMP
CAMC School of Nurse Anesthesia
Dr. Donna Slayton, MD
CAMC Hospital Anesthesiologist

3. Introduction
Ventilator-Associated Pneumonia (VAP) is defined as a hospital-acquired pneumonia occurring at the time of or within 48 hours of an intubation with no minimum time of mechanical ventilation required
VAP has been linked to increased: ventilator days, ICU length of stay, hospital stay, and medical costs
VAP has been associated with increased morbidity and mortality in trauma patients
Nearly every study conducted about VAP has reported: increased ventilator time, ICU LOS, Hospital LOS, and medical costs
Morbidity/Mortality: Most studies have linked VAP with increased mortality in the trauma patient, a few studies have shown no significant increase in mortality in the patients with VAP
NHSN = National Healthcare Safety Network
Since 2001, pneumonia has been one of Joint Commission “core measures”
CDC established the NHSN for surveillance of healthcare-associated infections with new mandatory reporting guidelines to begin January 2013

4. Introduction
VAP is one of the most common complications among the adult trauma population
Trauma patients show an increased incidence of VAP compared to other mechanically ventilated patients despite they are generally younger with fewer co-existing disease
CMS and other third party payers believe VAP is preventable and no longer reimburse for costs associated with VAP
2010, CMS changed reimbursement
Based on quality assurance measures
Hospitals mandated to report incidence of complications and readmissions of certain quality measures
Pneumonia is one of these measures
CMS will no longer pay for costs associated with hospital-acquired pneumonias
**Preventing VAP would improve patient outcomes and save hospitals considerable financial resources

5. Literature Review
Previous studies researching VAP have named the following potential risk factors:
Duration of Intubation and Mechanical Ventilation
Aspiration
Injury Severity Score
Glasgow Coma Score
Supine Position
Advanced Age
Reintubation
Emergent Intubation
Low SBP In Emergency Room
Prehospital Intubation
Aspiration- in 2011 Wahl, Zalewski, & Hemmila reported 59% of early BAL specimens and 47% of late BAL specimens grew aspiration-type organisms
Emergent intubation- Several studies that have looked at intubation as a risk factor have grouped prehospital and emergency room intubations together as “emergent intubations” and found it a significant risk factor
Prehospital intubation- a few studies have found association between prehospital intubation and VAP, except… 2010, (Evans, Zonies, Warner, et al):
retrospectively evaluated for an association between the timing of intubation and VAP.
572 patients included
412 intubated prehospital with well-established RSI protocol
101 developed VAP
70 of the 101 were intubated prehospital but no association made between prehospital intubation and VAP
This is the only study that conflicts with an association between prehospital intubation and VAP

6. Statement of Problem
VAP continues to be the most common complication reported in adult trauma patients
Further research is needed to discover what makes specific trauma patients more at risk
No research exists comparing prehospital intubation of the trauma patient to intubation by anesthesia personnel within the trauma center
Ventilator-associated pneumonia (VAP) is a common complication in the adult trauma population causing increasing ventilator days, increasing length of intensive care unit (ICU) days, and increased hospital costs. Further research in the area needs to focus on this population to discover what makes this population more at risk for developing VAP to help develop a plan of care to prevent VAP from occurring to improve patient outcomes
Little advancement has been made to prevent VAP in the trauma patient
Not all EMS providers are trained in RSI therefore there are no standardized RSI protocols in place in the prehospital setting to prevent aspiration of stomach contents at the time of tracheal intubation

7. Research Purpose
To evaluate for an associated link between prehospital intubation and the development of VAP
Determine other risk factors that make specific trauma patients at higher risk for developing VAP.
The purpose of this research was primarily …..

8. Methodology
RESEARCH HYPOTHESIS: The adult trauma patient who is intubated in the prehospital setting has a higher risk of developing VAP than those intubated by anesthesia personnel after arrival to the trauma center.

9. Methodology RESEARCH DESIGN Retrospective Case-control
RESEARCH SETTING:
CAMC General Hospital
Level I trauma center
non-profit, academic medical center

10. Methodology INCLUSION CRITERIA Priority 1 or 2 trauma
Adult age years old
Required endotracheal intubation with a minimum of 24 hours with mechanical ventilation
EXCLUSION CRITERIA
Age less than 18 or greater than 64 years
Transferred from another facility
Incomplete data records
Death or documented brain death within 48 hours of admission
Patients suffering burns, asphyxiation, or drowning

11. Methodology SAMPLE DESCRIPTION
A convenience sample of 494 patients from the CAMC trauma registry divided into 2 groups:
VAP group; n = 247,(cases)
No-VAP group; n = 247, (controls)
Admitted to CAMC General between January 1, and May 1, 2012

12. Patient Selection 894 patients from trauma registry
255 diagnosed with pneumonia
4 excluded- pneumonia prior to intubation
2 excluded- ventilation time < 24 hours
1 excluded- documented symptoms in community
1 excluded- outlier with 98 days ventilation
VAP group
(n= 247)
639 patients remained
258 patients excluded for ventilation time
< 24 hours
30 patients excluded for positive cultures without diagnosis
2 excluded for documented brain death
2 excluded for incomplete records
NO-VAP group
347 patients remained
computer randomization
Patient Selection
*ALL patients fitting inclusion criteria with a ICD-9 codes: (VAP); 486 (pneumonia- organism unspecified); (bacterial pneumonia) were extracted for VAP group. EMR confirmed timing of pneumonia to be associated with time on the ventilator
*Patients for the control group were randomly selected using computer randomization after ensuring fit with inclusion criteria

13. Methodology DEPENDENT VARIABLE: INDEPENDENT VARIABLES:
Ventilator-Associated Pneumonia
INDEPENDENT VARIABLES:
Age
Gender
Body Mass Index (BMI)
Injury Severity Score (ISS)
Glasgow Coma Score (GCS)
Location of first endotracheal intubation (prehospital, Emergency Department (ED), ICU/floor, Operating Room (OR)
Lowest recorded Systolic Blood Pressure (SBP) in the ED
Number of days on the ventilator
ICU length of stay
Hospital length of stay
Mechanism of Injury
Injury Type (Blunt or penetrating)

14. Statistical Analysis Binary regression (forward conditional method)
Independent t-test
Pearson Chi-squared analysis
Pearson Correlations (p < .01 was significant)
Linear regression (Enter method)
A p-value < .05 was statistically significant
The hypothesis was evaluated utilizing logistic regression with VAP as dependent variable and the main independent variable as location of first intubation
T-test evaluated for statistical differences between the groups for: age, BMI, lowest recorded SBP in ED, hospital LOS, ICU LOS, ventilator days, GCS, ISS
Pearson’s Chi-square analysis evaluated for statistical differences between the groups for: gender, location of first intubation, mechanism of injury, and injury type
Correlations tested for associations between VAP, ventilator time, ICU LOS, and hospital LOS, and intubation location
Linear regression tested for factors associated with increased length of ICU stay and hospital LOS

15. Patient Characteristics
Variable
Study Groups
Statistical Values
VAP
N = 247
Mean ± Standard Deviation
No-VAP
p-value
(2-tailed t-test)
Age
40.6 ± 13.8
39.5 ± 13.3
.406
ISS
28.5 ± 10.9
20.5 ± 10.8
.001*
GCS
7.3 ± 5.2
9.2 ± 5.3
001*
Vent
(Days)
12.7 ± 8.9
2.9 ± 2.3
ICU stay
(days)
13.7 ± 7.1
4.5 ± 3.2
Hospital
23.9 ± 12.3
10.7 ± 8.1
BMI
28.2 ± 6.1
28.9 ± 6.8
.231
SBP
110.9 ± 27.6
± 24.4
.032*
Gender: n(%)
Male
Female
N= 184 (74.5)
N= 63 (25.5)
N= 188 (76.1)
N = 59 (23.9) NS
Injury type: n(%)
Blunt
Penetrating
N = 232 (93.9)
N = 15 (6.1)
N = 211 (85.4)
N = 36 (14.6)
.002**
Location of first intubation: n(%)
Prehospital ED
ICU/floor OR
N = 108 (43.7)
N = 80 (32.4)
N = 36 (16.2)
N = 19 (7.7)
N = 70 (28.3)
N = 136 (55.1)
N = 20 (8.1)
N = 21 (8.5)
.001**
In this sample, Variables that showed statistically significant differences between the two study groups using t-test were…
Chi-squared analysis showed…
No difference for GENDER Injury type: chi-squared analysis showed significance for penetrating trauma in the No-VAP group

16. Comparison of Prehospital Intubation to Intubation after Arrival to Trauma Center in Adult Trauma Patients
Study Groups
Total
No-VAP
VAP
Prehospital Intubation no
Count
177
139
316
Std. Residual
1.5
-1.5
yes 70
108
178
-2.0
2.0
247
494
Pearson Chi-Squared: .001*
Odds Ratio for prehospital intubation (no/yes):

17. Comparison of Intubation Location between VAP and Control Groups
Study Groups
Total
No -VAP
VAP
Location of first intubation
prehospital
Count 70
108
178
Std. Residual
-2.0
2.0 ED
136 80
216
2.7
-2.7
ICU/floor 20 40 60
-1.8
1.8 OR 21 19 .2
-.2
247
494
Pearson Chi-Squared: *

18. Binary Regression Analysis for VAP
S.E.
Wald df
Sig.
Exp(B)
AGE
-.005
.011
.268 1
.605
.995
GENDER
.098
.319
.094
.759
1.103
Injury Severity Score
.003
.013
.040
.842
1.003
Glasgow Coma Score
-.053
.039
1.807
.179
.949
INJURY type
-.570
.554
1.059
.303
.566
Mechanism of injury
-.029
.047
.396
.529
.971
BMI
-.038
.021
3.104
.078
.963
SBP
.006
.005
1.330
.249
1.006
Ventilator DAYS
.210
.059
12.849
.001*
1.234
ICU LOS
.185
.057
10.574
1.203
Hospital LOS
.027
.019
1.983
.159
1.027
Emergent intubation
.212
.721
.087
.768
1.237
Intubation LOCATION
.219
1.502
.220
1.307
Constant
-2.623
1.297
4.089
.043
.073

19. Representation of Ventilator Time by Study Group
Almost linear association between VAP and length of time on the ventilator. It should be noted that only 6 patients in the NO-VAP group required ventilation > 10 days and 156 patients in VAP group required intubation > 10 days. ALL patients who were intubated > 14 days acquired pneumonia

20. Correlations between VAP, Intubation Location, Ventilator Time, ICU LOS, Hospital LOS
ICU/floor
Intubation
.124
(.006)*
.038
(.403)
.118
(.009)*
.098
(.029) OR
-.015
(.742)
-.031
(.767)
-.013
.057
(.203) ER
-.229
(.001)**
-.132
-.196
-.136
(.002)**
Prehospital
.160
(.001)*
.128
.130
(.004)*
.041
(.363)
Ventilator time
.592
---
.846
.722
.640
___
Expressed as:
Pearson correlation coefficient (p value)
P value < .01 significant
*Indicates statistical significance for positive correlation
** indicates statistical significance of negative correlation

21. Logistic Regression after Correcting for Ventilator Time, ICU LOS, Hospital LOSB
S.E.
Wald df
Sig.
Exp(B)
Step 1
ISS
.063
.009
48.513 1
.001
1.065
Constant
-1.539
.239
41.373
.215
Step 2
.060
43.491
.001*
1.062
Prehospital vs anesthesia intubation
.495
.201
6.044
.014*
1.641
-1.649
.246
44.868
.192
* Indicates statistical significance
Other variables entered but not significant: age, gender, GCS, Injury type, BMI

22. Regression Analysis for Intubations Performed Inside the Trauma Center by Anesthesia Personnel on the Adult Trauma Patient B
S.E.
Wald df
Sig.
Exp(B)
Step 1a
ICUfloor
1.154
.302
14.566 1
.001*
3.172
Constant
-.461
.128
12.910
.000
.631
*Indicates statistical significance
Other variables entered but not significant:
ED intubation, OR intubations
Prehospital intubation patients excluded
316 patients analyzed

23. Linear Regression for ICU LOS in Adult Trauma Patients
Model
Unstandardized Coefficients
Standardized Coefficients t
Sig. B
Std. Error
Beta
(Constant)
2.316
1.450
1.597
.111
OR intubation
.596
.661
.023
.901
.368
prehospital intubation
.630
.477
.042
1.319
.188
ICU floor intubation
1.710
.603
.078
2.835
.005*
AGE
.013
.024
.997
.319
GENDER
-.215
.380
-.013
-.566
.572
ISS
.032
.016
.051
1.994
.047*
GCS
.015
.050
.011
.294
.769
Injury type
-.730
.557
-.031
-1.311
.191
BMI
.026
.020
.879
pneumonia
2.710
.414
6.540
.001*
lowest SBP
-.016
.007
-.059
-2.445
.015*
Ventilator DAYS
.616
.696
24.006
Entered but excluded: ED intubation

24. Linear Regression for Hospital LOS in Adult Trauma Patients
Model
Unstandardized Coefficients
Standardized Coefficients t
Sig. B
Std. Error
Beta
(Constant)
-2.041
3.096
-.659
.510
AGE
.041
.027
.046
1.537
.125
GENDER
-.713
.810
-.025
-.880
.379
ISS
.100
.034
.095
2.971
.003*
GCS
.121
.106
.053
1.141
.254
Injury type
1.291
1.188
.032
1.087
.278
Ventilator DAYS
.386
.081
.257
4.767
.001*
ICU Length of Stay (days)
.816
.097
.480
8.404
BMI
.026
.055
.014
.472
.637
pneumonia
1.948
.921
.080
2.116
.035*
lowest SBP
.020
.043
1.418
.157
ICU floor intubation
-.312
1.295
-.008
-.241
prehospital intubation
-.887
1.018
-.035
-.871
.384
OR intubation
2.101
1.409
.047
1.492
.136
Entered but excluded: ED intubation

25. Discussion
Initial logistic regression failed to support the hypothesis, but subsequent regressions did once ventilator time, ICU LOS, and hospital LOS were removed from the equation due to high correlations
Prehospital intubation was associated with a 1.6 times greater risk of developing VAP
Trauma patients intubated in ED were more likely NOT to develop VAP
This finding are consistent with similar studies
FIRST Restate hypothesis – “Adult Trauma Patients undergoing endotracheal intubation in the prehospital setting will be at higher risk of VAP than those intubated by anesthesia personnel after arrival to trauma center”
Bochicchio et al 2003 reported prehospital intubation provided a 1.5 times increased risk for VAP
Eckert et al in 2004 noted prehospital intubation as a risk factor but not ED intubations
No previous research has looked at an association between ICU/floor intubations and VAP

26. Discussion Other risk factors identified:
Higher ISS was associated with increased risk of VAP
These findings are consistent with findings in published research
Patients in the VAP group had statistically lower GCS, but it did not show significance in regression as in other research
Patients in VAP group had statistically lower SBP in ED but this difference was not clinically significant (110 vs 115)

27. Discussion
Patients requiring emergent intubation in ICU or medical floor were 3.17 times more likely to develop VAP than others intubated by anesthesia
No previous research has looked at all intubation locations within the trauma center
In the current study, the VAP group presented with a maximum ISS score of 59 and 82% of that group presented with an ISS greater than 17 with 24.5% being greater than or equal to 35. The No-VAP group included a maximum ISS of 55 with 55.1% presenting with an ISS greater than or equal to 17 with only 8.5% above 35. However, as previously noted, the gold standard for the designation of a major trauma is an ISS greater than 15. The mean ISS of both the groups in this study was greater than 20, indicating both groups suffered similar numbers of major traumatic events so this variable alone does not explain the increased rate of VAP in the trauma patient.
ED intubation - ____ et al also reported that prehospital intubation but not ED intubation was associated with increased risk of VAP

28. Discussion
Increased ICU length of stay in the intubated trauma patient was associated with:
Presence of pneumonia
Ventilator days
ISS
ICU/floor intubation
ISS, pneumonia, and ventilator time has been associated with increased ICU stay in numerous published studies
ICU/floor intubations have not been specifically researched before now

29. Discussion
Increased Hospital length of stay in the intubated trauma patient was associated with:
Higher ISS
Ventilator days
Pneumonia
ICU length of stay
These findings supports previous published research

30. Discussion Other possible explanation of results:
Trauma patients who require prehospital intubation are often more severely injured with lower GCS, possibly leading to longer ventilator time and ICU LOS simply because of injury
Trauma patients with lower GCS (head injury) are prone to vomiting and may aspirate prior to arrival of EMS personnel
Anesthesia personnel routinely use RSI technique for intubation of trauma patients, EMS personnel vary in training so RSI is not standardized among all EMS providers
The means of both groups was greater than 20 (gold standard to define major trauma) so both groups were severely injury therefor this can not me the only explanaition.
In the VAP group, 132 patients had a GCS of three representing 53.4% of this group. The no-VAP group included 91 patients with a GCS of three, representing 36.8% of that group. However, when comparing only those intubated in the prehospital setting, the mean GCS was 3.1 with no difference between the ones that acquired VAP and those who did not.

31. Discussion Study limitations Retrospective study
Potential for bias
Findings are based on correlations and associations and can not indicate causality
Lack of standardization of VAP diagnosis criteria
Sample may include more VAP cases than other studies due to lack of diagnostic criteria
Some cases of VAP may not have been included

32. Discussion Study limitations
Only first intubation location is represented
Did not address multiple intubations
Use of antimicrobial therapy was not addressed
Some may have received antibiotics for surgical prophylaxis, open wounds, other infections

33. Contribution to the Field
No other research exists comparing anesthesia intubations with prehospital intubations
No other research exists that differentiate the varying locations of intubation within the trauma center (ICU/medical floor, OR)
First study to show a relationship between emergent intubations by anesthesia personnel in the ICU or on the medical floor and VAP

34. Implications/Recommendations
All EMS providers and anesthesia personnel should be aware of the potential risk of developing VAP with strict vigilance to prevent aspiration and tracheal contamination during intubation.
Establishment of a standardized method for diagnosis and appropriate treatment of VAP or suspected VAP at CAMC
Prospective studies for risk factors in trauma patients at CAMC after diagnostic criteria established
prevention protocols should include education to staff, nursing care, diagnostic parameters, treatment, weaning, quality assurance and feedback

35. Implications/Recommendations
Initiation of active pneumonia prevention protocols on all trauma patients requiring mechanical ventilation should be employed
Follow-up prospective studies after the implementation of prevention protocol should be warranted
Further evaluation of ICU/floor intubations performed by anesthesia personnel, assessing for contributing factors for the increased risk
prevention protocols should include education to staff, nursing care, diagnostic parameters, treatment, weaning, quality assurance and feedback

36. Implications/Recommendations
Standardization of RSI protocol across all EMS personnel who respond to trauma calls that have the ability to provide advanced airway management may be warranted
A prospective, randomized clinical trial of prehospital RSI protocols may provide definitive data in preventing VAP
The current setting design did not employ standardized RSI protocols throughout the prehospital setting. The findings of this study support other research where RSI protocol has not been well-established and adds to the understanding of the potential role of prehospital care in the trauma patients’ eventual outcomes.

37. Conclusion
The goal of this study was to evaluate for an associated link between the location of intubation and the development of ventilator-associated pneumonia.
The high correlation between prehospital intubation and ventilator-associated pneumonia demonstrated in this study suggests that prehospital care may influence subsequent development of VAP.
All emergency care providers and anesthesia personnel should be aware of the potential risks involved at the time of intubation and practice extreme vigilance to prevent aspiration or contamination of the airway.

38. QUESTIONS??