TARRAGONA, SPAIN—Despite substantial gains in our knowledge about ventilator-associated pneumonia (VAP), misconceptions and controversy about its treatment abound. However, a new report from an international consensus conference significantly clarifies VAP management.[1]

“Naturally, we were not able to agree on everything,” acknowledged Jordi Rello, MD, conference chairman, in an interview with PULMONARY REVIEWS. Surprisingly, though, the group was able to reach consensus on perhaps the most controversial issue—the role of microbiologic tests. Such tests, they concluded, should not be used to diagnose VAP but are important for modifying empiric antibiotic therapy. Diagnosis based on clinical grounds may be as sensitive as other methods, they stressed.

The report “goes a long way toward dispelling many of the myths associated with the management of VAP,” stated Paul E. Marik, MD, and Joseph Varon, MD, in an editorial.[2] Such myths include the beliefs that antifungals are always indicated when Candida spp turns up in respiratory samples from non-neutropenic patients, that antibiotic therapy should cover anaerobes, and that vancomycin should be a part of initial therapy for patients who have not previously received antibiotics.

THE CONSENSUS CONFERENCE

The 12 conference participants responded to 21 commonly asked questions about VAP diagnosis and treatment. They gave their initial responses independently and anonymously without discussion. Abstentions were permitted.

After the results were tabulated and reported, the participants shared, debated, and, if they wished, changed their responses. A report was then drafted, reviewed, revised, and approved.

INVASIVE TESTING

Most participants said they always use invasive diagnostic procedures (eg, bronchoscopy) because of the increased specificity and positive predictive value of the test results. The goal, however, is not pneumonia diagnosis but rather improved antibiotic selection. Almost all of the participants initiate invasive testing as soon as they suspect VAP; results may not be available soon enough to affect survival if the procedure is not performed within 12 hours of fever onset.

Drs. Marik and Varon concur with the importance of improving antibiotic selection. However, they do not believe that this always requires an invasive procedure. A more cost-effective approach would be to substitute blind protected-specimen brush sampling, said Dr. Marik, Director of the Trauma Life Support Center at Mercy Hospital of Pittsburgh, and Dr. Varon, an Associate Professor of Pulmonary and Critical Care Medicine at Baylor College of Medicine in Houston.

FOLLOWING GUIDELINES

All of the conference participants were from Europe and South America, said Dr. Rello, Chief of the Critical Care Department at the University Hospital Joan XXIII in Tarragona, Spain. During their discussions, these physicians discovered that the VAP pathogens and multiresistance patterns at their institutions are substantially different not only from those reported in studies from North America, but also from one another. Thus, they decided not to recommend following general VAP guidelines, such as those from the American Thoracic Society (ATS), but rather adhering to guidelines customized to local epidemiology, microbiology, and resistance patterns. They also opted for customization because they felt that the risk stratification method used in the ATS guidelines ignores important variables, such as previous antibiotic administration.

SELECTING THERAPY

Clearly, early antibiotic therapy is the most important influence on VAP outcome. Most participants said they would not wait longer than 12 hours to start empiric antibiotics when they suspected the condition. However, there was a clear lack of consensus on the optimum duration of antibiotic treatment. Nearly all participants chose a seven- to 14-day range, though.

The time to clinical response, not the infecting pathogen, should be the main factor in deciding the duration of antibiotic therapy, the participants agreed. Such therapy should continue for at least 72 hours after a clinical response is seen, they advised.

Monotherapy is adequate for early-onset VAP in patients who have not previously received antibiotics, most participants said. They also agreed that the number of days of hospitalization is more important than the number of days of ventilation in the decision to use monotherapy. Other important factors that must be considered in this decision include the patient’s immune status, antibiotic administration in the past three months, recent prolonged corticosteroid use or malnutrition, and the presence of chronic obstructive pulmonary disease.

Combination therapy is required for late-onset VAP because of the risk of infection with Pseudomonas aeruginosa or multiresistant Acinetobacter, Enterobacter, or Klebsiella spp. There was no consensus about the best regimen, however. Most participants agreed, though, that the definitions of early-onset and late-onset VAP cannot be rigid. Seven days after intubation is the cutoff generally used, but the specific day on which the shift from a primary endogenous pattern to a secondary exogenous pattern occurs can vary from institution to institution.

For VAP caused by methicillin-resistant Staphylococcus aureus (MRSA), most participants advocated a loading dose of vancomycin followed by continuous infusion and then lower doses once serum levels stabilize. Because of the low incidence of MRSA, vancomycin should not be part of the initial antibiotic regimen for patients with VAP who have not previously been given antibiotics.

All but one of the participants agreed that antibiotic therapy should not cover anaerobes, since those organisms are not important pathogens in VAP. Neither is Candida spp. Therefore, none of the participants said they would prescribe antifungals for non-neutropenic patients with Candida spp in their respiratory samples, even if those samples were obtained bronchoscopically. Most participants felt that initial use (up to the first 48 hours) of broad-spectrum antibiotics does not significantly raise the risk of multiresistance in VAP. However, they all advocated de-escalation therapy—the practice of switching to less costly and less toxic narrower-spectrum antibiotics once microbiologic data are available.

Piperacillin/tazobactam was the antibiotic of choice for empiric therapy, except in patients with head injury. Head-injured patients do not require an agent with antipseudomonal activity and thus should get adequate treatment with amoxicillin/clavulanate, cefuroxime, or a third-generation cephalosporin, some participants argued.

In order of preference, alternatives for empiric therapy in most VAP patients included a cephalosporin with antipseudomonal activity and carbapenems. “None of the participants chose vancomycin,” Dr. Rello told PULMONARY REVIEWS. Vancomycin is not often used in the empiric treatment of VAP at the participants’ institutions because of its poor efficacy against methicillin-sensitive S aureus in patients with pulmonary infection due to its poor penetration, he explained.

—Timothy Begany

References
1. Rello J, Paiva JA, Baraibar J, et al. International conference for the development of consensus on the diagnosis and treatment of ventilator-associated pneumonia. Chest. 2001;120:955-970.
2. Marik PE, Varon J. Ventilator-associated pneumonia: science and hocus-pocus. Chest. 2001; 120:702-704.

Return to table of contents