| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Guest Access | Sign In via User Name/Password
|
|
|
|||||||
Guest Access | Sign In via User Name/Password |
|||||||||
Toronto, ON, Canada
Dr. Nopmaneejumruslers is a senior Pulmonary Fellow, University of Toronto. Dr. Chan is Associate Professor of Medicine, Head, Joint Division of Respirology, University Health Network & Mt. Sinai Hospital.
Correspondence to: Charles K.N. Chan, MD, FCCP, Toronto General Hospital, 585 University Ave, Toronto, ON, Canada M5G2N2; e-mail: charles.chan{at}uhn.on.ca
Ventilator-associated pneumonia (VAP) is the major cause of infection in critically ill patients who are receiving mechanical ventilation, with a prevalence of 8 to 28%,12 and is one of the leading causes of death from hospital-acquired infections in critical care units.3 VAP is also associated with prolonged hospitalization and increased health-care costs.4 There is no doubt that the diagnosis of VAP remains one of the most controversial and challenging topics in the management of patients receiving mechanical ventilation.5 In general, the most acceptable standards for the diagnosis of VAP require quantitative cultures of BAL fluid or protected specimen brush (PSB) samples.67 Using 106 cfu/mL as the interpretative cutoff point for respiratory secretion cultures from endotracheal aspirates (EAs) has a comparable accuracy compared to that with the PSB technique, with a higher sensitivity (82%) and a lower specificity (83%).89 However, as soon as a lower threshold is used, specificity declined significantly. Irrespective of which standard is chosen, there is still an inherent delay from the time the BAL fluid, PSB, or EA specimen was obtained to the availability of the culture reports. In practice, patients who are suspected to have VAP would be started on empiric therapy pending culture results.
The main problem in dealing with patients who have a high clinical suspicion of VAP is the striking of a balance between avoiding a delay in initiating appropriate antibiotic therapy and reducing the inappropriate use of broad-spectrum antibiotics. Patients who are initially treated inadequately had poorer outcomes than those who received adequate antibiotic coverage at the beginning.101112 Thus, in clinical practice, initial broad-spectrum antibiotic therapy in patients in whom there is a high suspicion of VAP is the usual approach. However, it has been obvious that the emergence of resistance pathogens, especially in critical care units, is linked to the overuse of antibiotics.13
In this issue of CHEST (see page 589), Michel et al proposed twice weekly surveillance quantitative cultures of EAs in all intubated patients who were receiving mechanical ventilation as a means of assisting in the choice of antibiotic therapies when the presence of VAP was subsequently suspected. The authors reported that pre-VAP EA cultures identified the same pathogens with similar antibiotic susceptibility patterns compared to the results of BAL fluid cultures obtained when VAP was suspected in 34 of 41 cases (83%). Antibiotic selection based on the available results of pre-VAP EA cultures was adequate in 38 of 40 patients (95%). In contrast, had the American Thoracic Society guidelines and Trouillet guidelines been used, the empiric antibiotic treatment would have been adequate in 68% and 83% of patients, respectively. The main reason for the inadequate coverage using the published guidelines was the failure of the guidelines to inform the empiric treatment selection in the coverage of highly resistant pathogens. In addition to the better coverage, antibiotic selection based on the results of pre-VAP EA cultures also reduced the unnecessary use of some antibiotics like the ß-lactams compared to strategies based on the American Thoracic Society and Trouillet guidelines. The results of this study suggest that twice-weekly quantitative surveillance cultures of EAs may assist in the early prescription of appropriate antibiotic treatments for patients who develop VAP. This strategy may improve clinical outcomes, potentially may reduce antibiotic resistance in patients in critical care units and related complications, and may reduce hospitalization costs.
However, before the large-scale adoption of the surveillance EA culture strategy, we should address the following important issues related to surveillance EA cultures. First, this is an exciting investigation, but it is still a small study that includes only 41 cases of VAP confirmed by BAL fluid cultures. Second, this study used 103 cfu/mL as an interpretative cutoff point for the surveillance culture EA specimens, which is substantially lower than the cutoff points used for diagnosing VAP (ie, 106 to 107 cfu/mL). Despite the low cutoff points used for the surveillance EA cultures, the study by Michel et al reported a surprisingly high concordance (83%) between pre-VAP BAL fluid and EA cultures at a cutoff point of 103 cfu/mL. Clearly, a larger scale study using the same strategy with similar success is needed. Third, there was still a 5% rate of false-negative results in pre-VAP EA cultures, which stresses that negative results of surveillance EA cultures cannot exclude a diagnosis of VAP. Fourth, this study did not focus on clinical outcomes. Thus, there were no available data to indicate whether a strategy of using surveillance EA quantitative cultures leads to better patient survival or to the reduction of critical care unit or hospital stays. Last, the cost of routine surveillance EA quantitative cultures in all patients receiving mechanical ventilation is clearly an important consideration.
In summary, before we can recommendation the use of surveillance EA quantitative cultures as routine practice, a randomized controlled trial of this strategy compared to the current practice is needed in order to assess whether the new strategy can improve survival, control bacterial pathogen resistance, and finally reduce hospitalization costs.
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
