Ventilator-Associated Pneumonia May Not Increase Mortality
Abstract & Commentary
The mortality due to the development of ventilator-associated pneumonia (VAP) in critically ill adult patients has been variously estimated to be between 10% and 50%. Due to the difficulty in precise diagnosis of VAP and the high mortality of patients requiring prolonged mechanical ventilation, the validity and relevance of these estimates remains in doubt. In this study, Bregeon and associates have tried to separate the attributable mortality from VAP by comparing the incidence of VAP in patients who died in the ICU, with case-matched patients who survived. Control cases were matched by diagnosis category, age (within 10 years), sex, date of ICU admission (within 1 year), APACHE II Score (within 7 points), and duration of mechanical ventilation (control at least as long as cases).
Patients were selected from the 475 patients who were not immunocompromised, who required at least 48 hours of mechanical ventilation, and who were admitted to the ICU between 1993 and 1996. Of the 135 consecutive patients who died, 108 could be adequately matched with contemporaneous controls. These pairs served as the study populations. During the study period, VAP was prospectively identified using strict and consistent clinical and bacteriological diagnostic criteria. Treatment with appropriate antibiotics was begun empirically at the completion of cultures and terminated if cultures were not diagnostic. Matching was performed with no knowledge of the presence or absence of VAP. More than 85% of patients were identically matched in all of the selected variables. Groups differed in several important unmatched variables: controls tended to have previously been admitted to another ICU (P = NS), had a slightly lower organ system failure number (P < .01), and were less likely to have received steroids prior to ICU admission (P < .01). The groups were not different in the frequency of emergency surgery, presence of cancer, requirement for hemofiltration, incidence of the acute respiratory distress syndrome (ARDS), or use of stress ulcer prophylaxis.
The incidence of VAP in the cases was 36.1% compared to an identical percentage of the controls, who received mechanical ventilation for the same duration. When the individual pairs were considered, 23 were concordant (both VAP or both no VAP) and 32 were discordant, half of which were one way or the other. There was no contribution to mortality from development of VAP. Multivariate analysis identified only renal failure, bone marrow failure and treatment with steroids, and not the development of VAP, as independent predictors of mortality. The only risk factor identified in this study for developing VAP was the prior use of antibiotics. Other factors such as duration of ventilation, age, APACHE II score, organ system failure score, use of sucralfate, prior steroid treatment, immunosuppressive agents, and the presence of cancer were not apparent risk factors for VAP in this study (Bregeon F, et al. Anesthesiology. 2001;94[4]:554-560).
Comment by Charles G. Durbin, Jr., MD, FCCM
There are 2 schools of thought regarding the effect of VAP on outcome of care. The more popular concept is that VAP increases morbidity, cost, and mortality. Most poorly controlled studies show that patients who develop VAP experience a worse outcome than those who do not. This usually includes higher mortality. However, when other risk factors for mortality are controlled, the differences in mortality from VAP are less apparent. This study lends support to the idea that VAP does not actually increase mortality. Its strength is in the success of matching cases with controls. The powerful technique of multivariate analysis suggests that the only independent mortality predictors of death are renal failure, prior use of steroids, and bone marrow failure.
The results of this study should not change the aggressive approach to VAP diagnosis, prevention, and treatment. What was not examined in the study was the effect of the development of VAP on the cost of care. This effect has been adequately delineated in other studies and has been shown to be a large increase in costs due to increased duration of mechanical ventilation and hospital length of stay. This alone justifies the preventive measures suggested.
It should be noted that the diagnosis of VAP is controversial. The criteria used in this study were broad, and the results were unchanged even using different (more or less stringent) definitions. Antibiotic treatment was also aggressive, empiric, and initially resulted in the appropriate choice in more than 75% of cases in both groups.
Caution should be used in applying these conclusions to other patient groups. Very few postsurgical patients were in this study cohort, partly due to the low mortality of surgical as compared to medical patients requiring mechanical ventilation. Cardiac surgery patients, as well as other patients following surgery, may experience increased mortality from VAP, as VAP is a contributor to organ system failure.1
Reference
1. Kollef MH. The impact of nosocomial infections on patient outcomes following cardiac surgery. Chest. 1997;112:666-675.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content