ORIGINAL ARTICLE
Year : 2022 | Volume
: 10 | Issue : 2 | Page : 33--37
Evaluation of characteristics, clinical relevance and outcome of ventilator associated pneumonia and ventilator associated events algorithm
Neeru Sahni1, Manisha Biswal2, Kulbeer Kaur2, Anjishnujit Bandyopadhyay1, Vikas Saini1, Lakshmi Narayana Yaddanapudi1,
1 Department of Anaesthesia and Intensive Care, Post Graduate Institute of Medical Education and Research, Chandigarh, India
2 Department of Medical Microbiology, Post Graduate Institute of Medical Education and Research, Chandigarh, India
Correspondence Address:
Dr. Kulbeer Kaur
Department of Medical Microbiology, Post Graduate Institute of Medical Education and Research, Chandigarh - 160 012
India
Abstract
Background: Surveillance for ventilator-associated pneumonia (VAP) has long been a challenge because of the lack of objective, reliable definitions. Hence, National Healthcare Safety Network (NHSN) has introduced ventilator-associated events (VAEs) as a surveillance definition. While VAE identifies all the possible complications in a mechanically ventilated patient, VAP recognises only the infective complications. There are several retrospective studies reporting no concordance between the occurrence of VAE and VAP. A prospective, observational study in medical intensive care unit (ICU) was conducted over 1 year with the objective of comparing all three tiers of VAE, which are ventilator-associated condition (VAC), infection-related ventilator-associated complication (iVAC) and possible VAP (PVAP), along with VAP in terms of predictive value, characteristics and clinical relevance.
Materials and Methods: A prospective cohort study was conducted from July 2018 to June 2019 at PGIMER, Chandigarh, in a 12-bedded medical ICU. All patients with more than 48 h of mechanical ventilation (MV) were included. The demographic data, Acute Physiology and Chronic Health Evaluation-II at 24 h of admission, days of MV, length of ICU stay and outcome of patients were recorded. The patients were screened for both VAP and VAE.
Results: Out of a total of 405 patients, 274 patients were included with 3945 patient days and 3330 MV days. The incidence of VAP, VAC, iVAC and PVAP was 6.91, 8.41, 5.41 and 1.50/1000 ventilator days, respectively. Kendall's W-test showed that there was no concordance between VAP and VAE.
Conclusion: The study concluded no concordance between cases identified as VAE and VAP.
How to cite this article:
Sahni N, Biswal M, Kaur K, Bandyopadhyay A, Saini V, Yaddanapudi LN. Evaluation of characteristics, clinical relevance and outcome of ventilator associated pneumonia and ventilator associated events algorithm.J Patient Saf Infect Control 2022;10:33-37
|
How to cite this URL:
Sahni N, Biswal M, Kaur K, Bandyopadhyay A, Saini V, Yaddanapudi LN. Evaluation of characteristics, clinical relevance and outcome of ventilator associated pneumonia and ventilator associated events algorithm. J Patient Saf Infect Control [serial online] 2022 [cited 2023 Jun 7 ];10:33-37
Available from: https://www.jpsiconline.com/text.asp?2022/10/2/33/370888 |
Full Text
Introduction
Ventilator-associated pneumonia (VAP) and other healthcare-associated pneumonias are important, common healthcare-associated infections, but the concerns of low accuracy, inter-observer variability and subjectivity with traditional VAP surveillance led to the proposal of objective ventilator-associated event (VAE) surveillance algorithm by the Centers of Disease Control and Prevention (CDC) to streamline the surveillance of complications occurring in mechanically ventilated patients.[1] As per CDC, VAP is identified using a combination of imaging, clinical and laboratory criteria. However, VAEs are identified using a combination of objective criteria: deterioration in respiratory status after a period of stability or improvement on the ventilator, evidence of infection or inflammation and laboratory evidence of respiratory infection. There are three definition tiers within the VAE algorithm: ventilator-associated condition (VAC), infection-related ventilator-associated complication (iVAC) and possible VAP (PVAP). VAE surveillance is based on objective, streamlined and potentially automatable criteria that identify a broad range of conditions and complications occurring in mechanically ventilated adult patients. The VAE definition algorithm is for use in surveillance; it is not a clinical definition algorithm and is not intended for use in the clinical management of patients.
A systematic review and meta-analysis concluded that VAE surveillance does not accurately detect cases of VAP and there is a high probability that actual cases of traditional VAP are missed.[2] However, conflicting results of another study reported a good correlation between cases identified as VAE and VAP.[3] As any change in surveillance paradigm (from VAP to VAE) will guide the bundles for prevention, it is imperative to confirm that the same data are generated using two definitions addressing complications occurring in patients receiving mechanical ventilation (MV).
With the availability of data from mainly retrospective studies and conflicting evidence regarding the correlation between the two surveillance algorithms, we conducted this prospective cohort study over 1 year in adult patients requiring MV in the medical intensive care unit (ICU) of a tertiary care academic centre.
Materials and Methods
Aim
The aim of the study was to compare VAE and VAP cases identified using the CDC definition in terms of predictive value, characteristics and clinical relevance. An attempt was made to explore the sensitivity, specificity and predictability of VAE (VAC, iVAC and PVAP)-defined cases for VAP-defined cases.
The objectives of the study were:
To determine the VAC, iVAC, PVAP and VAP per 1000 ventilator daysTo compare the sensitivity, specificity and predictive values of VAE-defined cases (VAC, iVAC and PVAP) for VAP-defined casesTo explore the relationship between VAE (VAC, iVAC and PVAP)/VAP to Acute Physiology and Chronic Health Evaluation-II (APACHE-II) scores, length of stay (LOS) in ICU, age, ventilator days, pre-ICU stay in hospital and survival status of patients.
Study design and methods
After approval from the Institute Ethics Committee (IEC/NK/4614/RS/340), a prospective cohort study was conducted from July 2018 to June 2019 in the medical ICU of a tertiary care academic centre, which is a 12-bedded medical ICU. All patients with more than 48 h of MV were included. The demographic data, " APACHE II" at 24 h of admission, days of MV, length of ICU stay and survival status of patients were recorded. The patients were screened for VAP as well as three tiers of VAE algorithm, as per CDC definitions of VAC, iVAC and PVAP. Statistical analysis was performed using the Chi-square test. SPSS version 26 (IBM, Released 2019, IBM SPSS statistics, version 26.0, Armonk, NY: IBM Corp) was used for the analysis.
Results
Out of a total of 405 patients admitted to ICU throughout the study period, 131 patients were excluded due to <2 days of ICU stay or <2 days of MV. Two hundred and seventy-four patients were included, with a total of 3945 patient days and 3330 MV days. A total of 238 patients neither had VAP or VAE, whereas 36 patients had VAP and/or VAE. Twenty-three patients had only VAP and 28 patients had only VAE (any one of the three tiers) [Figure 1]. The demographic and baseline parameters such as APACHE-II score, days of MV, LOS in ICU and clinical outcome of all patients (including 36 who had VAE and/or VAP) are shown in [Table 1].{Figure 1}{Table 1}
The overall rate of VAP, VAC, iVAC and PVAP was calculated to be 6.91, 8.41, 5.41 and 1.50/1000 ventilator days, respectively. The monthly VAP rate is shown in [Figure 2]. The sensitivity, specificity, positive predictive value and negative predictive value of diagnosing VAP in relation to VAE algorithm tiers are shown in [Table 2]. Kendall's W-test showed that there was no concordance between cases detected as VAP and VAE [Table 2]. However, 15 of 28 VAC cases (53.57%), 12 of 18 iVAC cases (66.66%) and 3 of 5 PVAP cases (60%) developed into VAP.{Figure 2}{Table 2}
The Chi-square test was applied to check the relation of VAE/VAP with APACHE-II scores, LOS in ICU, age, MV days, pre-ICU stay in hospital and survival status. It was found that age (P = 0.014), MV days (P < 0.001) and LOS in ICU (P = 0.019) were significantly related to VAE/VAP. VAE/VAP was significantly more in patients in a higher age group (>60 years) and had more MV days (>30 days) and more LOS in ICU (>10 days). However, APACHE-II scores (P = 0.003), age (P = 0.006) and MV days (P = 0.020) were related to the survival status of patients. Patients who expired had more MV days (>30 days) and higher APACHE scores (>24).
The mortality rate of patients who had only VAC, only iVAC and those who had both VAP and iVAC was 91%–94%. The overall mortality rate of patients with VAP and/or VAE was found to be much higher than those without VAP/VAE [Figure 3]. APACHE-II scores were available for 222 patients. It was found that 24 (10.81%) of these patients scored >24 APACHE score. Three out of 33 (9.09%) patients with VAP and/VAE had >24 scores, while 21 out of 189 (11.11%) without VAP and/VAE patients had APACHE scores >24.{Figure 3}
Discussion
This prospective cohort study shows that 66.7% of iVAC cases, 60% of PVAP cases and 53.6% of VAC cases were also found to have developed VAP clinically, while 3 out of 5 (60%) PVAP and 15 out of 28 (53.57%) VAC cases developed into VAP. The positive cases of VAP showed symptoms including high-grade fever, leucopenia, leucocytosis, increased secretions, change in the character of sputum, worsening gas exchange, bronchial breath sound, chest X-ray changes including new/progressive infiltrates/consolidation. At the same time, only 15 out of 23 (65.2%) VAP cases showed VAE. Out of remaining eight VAP cases that didn't develop VAE, seven patients did not have a increase in positive end expiratory pressure (PEEP) or fraction of inspired oxygen (FiO2 ) as per criteria. However, one patient had a PEEP rise of 2.
Apart from clinical reasons, a main surveillance-related reason for it could be the definition. The definition of VAE is strictly objective and easier to apply, but it does need adherence to the criteria of the window period. For example, clinicians may not send the tracheal aspirate (TA) cultures or do PVAP-related investigations within the window period or the culture report may come sterile. In the present study, out of four 'iVAC-only' patient categories, for one iVAC patient, the TA sample was sent on day 8 of ICU stay and the date of event was on day 11 of ICU. Two of iVAC patients had mixed growth in the TA culture. One patient had sterile culture reports in the window period. Nine patients were classified in 'iVAC + VAP' patient category, of which TA samples of 4 patients were sent outside the window period, one patient's date of the event was day 3 in ICU and TA was sent on day 2 of ICU admission and could not be included, no sample was sent for three patients and one patient had sterile TA culture. On the other side, some of the VAE cases did not progress to VAP, as evidenced by lack of infiltrates/consolidation. In the present study, 15 VAEs progressed to VAP, while 13 did not [Figure 1].
In a systematic review and meta-analysis, Fan et al. concluded that the application of VAE surveillance misses out on cases of actual VAP. Hence, it is not possible to detect cases of VAP if the VAE algorithm is used routinely. However, out of 18 studies included in this meta-analysis, 10 were retrospective.[2]
In a large retrospective cohort study by Lilly et al. in 8408 patients requiring MV during 2009–2012, the authors reported low sensitivity of VAE definitions to identify patients with VAP. The various issues described were the non-inclusion of chest radiograph findings in VAE surveillance, acute respiratory distress syndrome patients without lower airway infection getting diagnosed as VAE and the inability of VAE surveillance to compare hospital-acquired complications between various health-care settings. Furthermore, even VAP surveillance had poor sensitivity for diagnosing the PVAP tier of the VAE algorithm.[4]
In a prospective study on 1209 patients, Boyer et al. identified non-concordance in cases identified as VAC and VAP and also commented that most of the VAC events were not preventable, which is one of the major goals of surveillance algorithms.[5] Thus, without having a reliable, comprehensive VAE prevention bundle, it may not be possible to use VAE algorithm solely as a tool for comparison between different health-care settings. Furthermore, similar to our data, the authors observed higher mortality in patients having VAE as compared to patients having VAP. The higher mortality rate in patients with VAC can be due to reasons such as cardiopulmonary failure, cardiogenic and septic shock and pulmonary embolism.[5]
A retrospective study by Piriyapatsom et al. in patients admitted to trauma ICU revealed that the algorithm for the detection of iVAC is unable to diagnose as much as 75% of cases of actual VAP. Instead, it identified patients who had deterioration of ventilatory parameters due to other reasons. The authors expressed concern over the routine use of the iVAC algorithm to detect infections of the respiratory tract in patients requiring MV.[6] The results of our prospective study in medical ICU are similar to theirs, with low sensitivity and poor positive predictive value of the VAE algorithm to detect VAP. However, our mortality rate for iVAC, VAP and VAC and VAP and iVAC was higher than VAP which is not in line with Piriyapatsom et al.'s study where they found iVAC mortality rate lesser than VAP or VAP combination with VAC and iVAC.
Klompas and Berra published a review discussing the pros and cons of recognising the VAE algorithm as a quality indicator for the performance of ICUs. On the one hand, VAE definitions are not liable for manipulation by the clinician and have the potential to objectively track complications and outcomes of patients receiving MV. On the other hand, the biggest issue is poor sensitivity and specificity to identify their actual progression to VAP. The other major concern is the dependence of the VAE algorithm on the ventilator setting, which makes it prone to manipulation and gaming to give a false impression of good quality care.[7] In the present study, it was noted that in some patients, the PEEP/FiO2 setting was changed many times a day, including a change in the mode of ventilation, and as per definition, the minimum PEEP/FiO2 maintained by a patient for at least 1 h is to be noted. Hence, the actual VAC cases might be missed.
Conclusion
In the present study, it was found that 53.5% of VAE cases progressed to VAP and 46.4% did not. This prospective study concludes that VAP and VAE are two separate definitions which cannot be used as the replacement of each other. Finding the concordance between two definitions is not a wise decision to make. Although the sensitivity and positive predictive value of VAE for VAP were found to be lesser than specificity and negative predictive value, it should not affect the importance of both definitions as both definitions are different entities and cannot be compared. Hence, it would be prudent to use both definitions in tandem and not replace VAP surveillance with VAE algorithm.
Financial support and sponsorship
Nil.
Conflicts of interest
There are no conflicts of interest.
References
| 1 | Klompas M. Barriers to the adoption of ventilator-associated events surveillance and prevention. Clin Microbiol Infect 2019;25:1180-5. |
| 2 | Fan Y, Gao F, Wu Y, Zhang J, Zhu M, Xiong L. Does ventilator-associated event surveillance detect ventilator-associated pneumonia in intensive care units? A systematic review and meta-analysis. Crit Care 2016;20:338. |
| 3 | Havaldar AA, Rajagopalan N, Gupta V, Manjunath PH. Do ventilator associated events (VAE) predict ventilator associated pneumonia (VAP)? Intensive Care Med 2014;40:S54. |
| 4 | Lilly CM, Landry KE, Sood RN, Dunnington CH, Ellison RT 3rd, Bagley PH, et al. Prevalence and test characteristics of national health safety network ventilator-associated events. Crit Care Med 2014;42:2019-28. |
| 5 | Boyer AF, Schoenberg N, Babcock H, McMullen KM, Micek ST, Kollef MH. A prospective evaluation of ventilator-associated conditions and infection-related ventilator-associated conditions. Chest 2015;147:68-81. |
| 6 | Piriyapatsom A, Lin H, Pirrone M, De Pascale G, Corona De Lapuerta J, Bittner EA, et al. Evaluation of the infection-related ventilator-associated events algorithm for ventilator-associated pneumonia surveillance in a trauma population. Respir Care 2016;61:269-76. |
| 7 | Klompas M, Berra L. Should ventilator-associated events become a quality indicator for ICUs? Respir Care 2016;61:723-36. |
|
