Infection Control Hotline

Management of ventilator-associated pneumonia (VAP) requires instituting preventive strategies, rapid and correct diagnosis and treatment, appropriate dosing and duration of antibiotics, and accurate assessment of response, i.e., identifying super-infection, resistance, toxicity and microbiologic failure. Taken together, these steps can improve mortality.

Q: What does a clinician versus an infection preventionist look at when diagnosing ventilator-associated pneumonia?

One of the ways we diagnose VAP is the Clinical Pneumonia Infection Score (CPIS), introduced by Pugin in 1994. CPIS looks at basic clinical variables like temperature, blood leukocytes, tracheal secretions, oxygenation and chest radiography. When a patient has a CPIS of 6 or more, there's a 92 percent chance that pneumonia will be found on an open lung biopsy.

The problem with CPIS is that it's difficult to assess the character and volume of tracheal aspirate production retrospectively. To use the CPIS, I feel that it must be determined in real-time assessment for the presence of VAP.

A recent study found that the sensitivity and specificity of tracheal aspirates was in the mid-60 percent range, not as accurate as we would like. A reliable quantitative culture of tracheal aspirates may accurately distinguish between colonization and infection, but is difficult to do. Relying only on semi-quantitative endotracheal aspirates for VAP diagnosis may lead to needless antibiotic use in up to 31 percent of patients. The strength of the endotracheal aspirates is that if they are negative, there's a 94 percent chance that patient does not have pneumonia. However, CPIS can assist the clinician in getting a diagnosis that is in the ballpark.

CPIS is not a tool that infection preventionists would use, though. They would use the Centers for Disease Control and Prevention's (CDC) surveillance criteria for diagnosis of VAP. The problem with the CDC surveillance diagnosis is that the infection preventionist must rely on the chest radiograph report and determine whether there is a new or progressive pulmonary infiltrate suggestive of pneumonia. Many of the reports may not contain this information to make this determination. Therefore, the infection preventionist is dependent on someone else to read the X-ray.

Q: Which is the most definitive of the techniques available for diagnosing VAP?

The controversies surrounding the diagnosis of VAP have been intense. We're never really going to have a gold standard that can be compared with respiratory sampling. When we introduce a test, we should ask, "What will this test do for outcomes?"

With the use of techniques like mini-BAL (bronchoalveolar lavage) or bronchoscopy-guided BAL, the major objective is to identify the pathogen causing the pneumonia, allowing us to specifically target the pathogen with antibiotics and thereby avoiding the overuse of antibiotics.

For example, a patient on mechanical ventilation may show all the signs of pneumonia, but a subsequent examination may show a pulmonary embolism or a pulmonary infarct and no need for antibiotic therapy. You want to identify the patient who truly has pneumonia.

VAP occurs primarily in the lower lobes of the lung when seen on open biopsy or post-mortem examination. We have used BAL, bronchoscopy-guided BAL, or blind mini-BAL because the instillation of sterile fluid with its retrieval will cover a larger area of the lung.

Fagon and Chastre, in their classic study using bronchoscopically BAL in diagnostic testing, showed a reduction in antimicrobial use in patients suspected of VAP. Antimicrobial-free days for the invasive diagnosis BAL patients were 11 days versus 7.5 days for the noninvasive endotracheal sample-driven diagnosis group. Furthermore, this study was the first to show a mortality advantage for the invasive diagnosis strategy BAL, with a 16.2 percent mortality rate for patients tested using the invasive diagnosis and 25.8 percent for the noninvasive diagnosis at day 14. (Fagon JY, et al. Invasive and Noninvasive Strategies for Management of Suspected Ventilator-Associated Pneumonia: A randomized trial. Annals of Internal Medicine. 2000:132;621-30.)

The primary reason we use mini-BAL at Emory University Hospital Midtown is because it can be done at any time. We never have to wait. Further, mini-BAL is cheaper to do than bronchscopically guided BAL, with mini-BAL having a base cost of $44 compared to bronchoscopy with a base cost of $215. A very interesting study by Ost found in a decision analysis model that if you want to minimize cost, minimize antibiotic use and maximize survival then perform a mini-BAL and start three antibiotics that will cover the pathogens that reside in your unit that cause VAP. (Ost D, et al. American Journal of Respiratory Critical Care Medicine. 2003:168;1060-1067.)

Q: What important preventive measures have you observed that have been useful in VAP prevention?

Recently, I have been very impressed by picking the low-hanging fruit of infection prevention outlined by the Institute for Healthcare Improvement and enumerated in the ventilator bundle. This simple implementation has led to marked reductions in VAP in every hospital I have looked at.

Intubation can increase the risk for VAP at least tenfold. Oral care is extremely important in trying to prevent VAP.

I was fortunate to travel to a hospital in Barcelona several years ago, where a colleague showed me intubated patients in their intensive care unit. Though they secure their patients' airways much like we do, the nursing staff cleaned and brushed the patients' teeth regularly. I noted that the patients' teeth were clean and their was no foul oral odor.

My colleague had residents scrape off the plaque and microbiologically plate it. For any patient who developed VAP, if the cultures grew out, he would go back and review the plaque cultures. They observed that the organisms embedded in plaque were the same organism found in VAP.

In our intensive care unit at Emory University Hospital Midtown, we strive to be compliant with the ventilator bundle with an emphasis on keeping the head of the bed greater than 30 degrees and on aggressive oral hygiene.

Q: What other prevention measures can infection preventionists convey to clinicians?

Interventions to prevent aspiration can include semi-recumbent bed positioning and draining the ventilator circuit of subglottic condensate. Care should be taken not to drain ventilator condensate too frequently as this can increase the incidence of VAP. Avoiding unnecessary patient transports and re-intubation is another way to reduce the incidence of VAP. If you send a patient to get a CT scan of the head, then later send them to get a CT scan of the chest, you're going to increase the incidence of VAP.

Limiting sedation and increasing the use of "sedation vacations" helps extubate these patients quicker, thereby reducing the incidence of VAP. The use of noninvasive ventilation is an interesting story. Bouchard and co-workers came out with a nice study showing that if you use noninvasive ventilation early enough, you can reduce the incidence of VAP.

The ventilator bundle is familiar to everyone and includes common-sense approaches that have been tested to see if they result in a lower incidence of VAP. Elevating the head of the bed to 30 degrees is part of the ventilator bundle. Data suggests if you elevate the head of the bed to 30 degrees, there is some incidence of reduction in VAP. Regularly checking the patient's readiness to be weaned off sedation is another element of the ventilator bundle.

We have a goal sheet in our ICU that we check every day to see if a patient is ready to come off a ventilator. Sedation vacation is still another part of the ventilator bundle, as is deep venous thrombosis prophylaxis, stress ulcer prophylaxis and oral care. What we are trying to get at here is the low-hanging fruit of preventive elements.

Q: Beyond the low-hanging fruit of VAP prevention, are there any ways to reduce the incidence of VAP?

Another way to try to prevent VAP is through the use of modified endotracheal tubes. The ideal tube would allow continuous aspiration of subglottic secretions (CASS). Data from a meta-analysis of five studies showed that improved outcomes with CASS in patients expected to be ventilated longer than 72 hours. CASS reduced the risk of VAP by nearly half, decreased ICU stays by three days, reduced mechanical ventilation by two days, and delayed the onset of VAP by nearly seven days. (Dezfulian C. American Journal of Medicine. 2005:118;11-18.) There are several modifiable endotracheal tubes on the market and they all show to varying degrees a reduction in VAP rates.

Q: In your experience, how has the culture of prevention changed for the clinician and the infection preventionist?

Routine hand disinfection prior to patient contact has been a very interesting story. The public is very aware of this. I was in a clinical setting recently, getting ready to see a patient, and the patient's family was there. After washing my hands and turning to the patient, the patient's family told me, "We were just waiting to see if you washed your hands." In the intensive care unit, we have signs that say, "Please remind your caregivers to wash their hands."

This has really been a change in the culture that I did not think would happen for another 10 years. We have definitely been very successful with getting that message out, along with using hand disinfectants with alcohol, and keeping track of the number of times you wash your hands.

The infection preventionist and the clinician can complement each other well. Through prevention there's definitely a cost savings. With improved diagnosis, there are improved survival rates, decreased length of stay in the hospital and a reduction in hospital costs.

This article first appeared in the January 2010 issue of Materials Management in Health Care.

To respond to this article, please click here.