| 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 |
|||||||||
Université de Lille Centre Hospitalier Tourcoing, France
Correspondence to: Olivier Leroy, MD, Service de Réanimation Médicale et Maladies Infectieuses, Hôpital G. Chatilliez, 135 rue du Président Coty, 59208 Tourcoing, France
To the Editor:
In a recent study, Rello et al1 (January 2003) described their use of microbiological testing in 204 patients with severe community-acquired pneumonia (SCAP). Furthermore, they evaluated the impact of bacteriologic data on the management and prognosis of such patients, and compared etiologic agents according to whether patients underwent intubation or not.
The main results of this study were the following: a microbiological diagnosis of pneumonia was made in 71 intubated patients and in 46 nonintubated patients. Infections due to Legionella pneumophila and Pseudomonas aeruginosa infections were significantly more frequent in intubated than in nonintubated patients (15.1% vs 7.1% and 6.6% vs.1.0%, respectively). Positive microbiological test results led to antimicrobial treatment modifications in 85 patients; in 11 cases, modifications were justified because causative organism was resistant to initial empiric therapy. In 65 patients, modification was a simplification of the initial therapy. Finally, mortality rates were similar in patients with (26.4%) and without (19.5%) a positive etiologic diagnosis.
According to these results, the authors conclude that "microbiological testing is fully justified in patients with SCAP, because identifying the causative agent and adjusting treatment both impact on patient outcome." Moreover, they suggest that "intubated patients should be empirically treated for Pseudomonas and Legionella while awaiting bacteriology results." Moreover, they emphasized that "this is the first study to evaluate the impact of diagnostic testing on the outcome of SCAP."
Despite the interest of this study, we wonder if such an analysis and such results allow such conclusions. First of all, this study was a retrospective analysis of prospectively collected data. Such a study design could indicate that microbiological investigations were not exhaustive and different in the two groups of patients, intubated vs nonintubated. According to the reported microbiological investigations, samples for initial and follow-up serologic studies were collected from most patients. Protected brush cultures were performed for patients who required mechanical ventilation. In fact, in the overall population of patients, serologic studies and urine enzyme-linked immunosorbent assays for detection of antigens from L pneumophila serogroup 1 were performed in 35.7% and 31.3% of cases, respectively. Moreover, differences existed between intubated and nonintubated patients, since serologic studies were performed in 40.5% and 30.6% of cases, respectively, and urine tests in 45.2% and 16.3% of patients, respectively. Similarly, in intubated patients, bronchoscopy for protected brush specimens was performed in 43.3% of cases, compared to 16.3% in nonintubated patients. These data clearly demonstrate that microbiological investigations were quite different from the reported methods, as well as nonexhaustive and nonhomogenous in the two groups. Consequently, we think that it could be wrong to assess that incidences of L pneumophila and P aeruginosa infections in intubated patients were significantly higher than in nonintubated patients. Furthermore, considering the very low frequency of P aeruginosa isolated in this study (6.6%), we strongly disagree with the authors’ suggestion to systematically target this organism in the initial empirical antimicrobial therapy for intubated patients. Some details about patients exhibiting P aeruginosa infection (ie, COPD, prior antimicrobial treatment, etc.) and about antimicrobial sensitivity could have been useful to really identify risk factors for such an infection and, thus, help physicians in selecting an antimicrobial therapy including in its spectrum both Streptococcus pneumoniae and P aeruginosa.
One of the most important point suggested by Rello et al1 is the usefulness of bacteriologic tests (when results are positive) to adapt and, most often, to simplify the initial empiric antimicrobial therapy. This concept of de-escalation therapy is now proposed in the management of nosocomial pneumonia,2 and we agree with such a concept; however, we wonder if it could be applied to the management of SCAP. It has been demonstrated that common typical and atypical pathogens could be associated in community-acquired pneumonia and, even, in SCAP.3 Two studies4 5 have demonstrated that antimicrobial regimens covering both typical and atypical organisms are associated with a lower rate of death than regimens only covering typical organisms. Consequently, a therapeutic simplification based on a positive bacteriologic result could have a deleterious impact if bacteriologic studies are not exhaustive, if results of tests are delayed, and if the antimicrobial spectrum becomes too narrow. For example, let us assume a patient exhibiting an initial favorable evolution with an antibiotic regimen including a nonpseudomonal third-generation cephalosporin combined with a macrolide or quinolone. We think the ß-lactam agent should be adapted to the antimicrobial susceptibility of the causative agent (ie, amoxicillin for penicillin-susceptible S pneumoniae), but we also think the macrolide or quinolone must not always be withdrawn, as there is no 100% effective test to rule out atypical bacteria. In the present study, Rello et al1 did not report what kind of simplification they performed and what was the ultimate outcome. Such data would have been interesting to validate the de-escalation therapy or to determine the limits of this concept.
Moreover, we do not understand the statement appearing in the abstract conclusion stating that microbiological testing impacts patient outcome. Data shown in the text indicates similar outcome for patients with or without an etiologic diagnosis.
Finally, we do not completely agree with the authors when they underlined that their study is the first to assess the impact of diagnostic testing on clinical outcomes in patients with severe SCAP. In 1990, Pachon et al,6 studying 67 patients with SCAP, reported that the "diagnosis of the causative agent during the course of disease did not significantly influence the outcome." In 1995, we demonstrated in 299 patients with SCAP admitted in our unit that the etiology of pneumonia was not significantly related to prognosis.7 Finally, it is interesting to note that Moine et al8 reported in 1994 that presence of microbiologic diagnosis was significantly associated with death.
References
This article has been cited by other articles:
|
|
 |
|
  S. R. Bapoje, J. F. Whitaker, T. Schulz, E. S. Chu, and R. K. Albert Preoperative Evaluation of the Patient With Pulmonary Disease Chest, November 1, 2007; 132(5): 1637 - 1645. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Brunelli, F. Xiume, M. Al Refai, M. Salati, R. Marasco, and A. Sabbatini Air leaks after lobectomy increase the risk of empyema but not of cardiopulmonary complications: a case-matched analysis. Chest, October 1, 2006; 130(4): 1150 - 1156. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 F. Paganin, F. Lilienthal, A. Bourdin, N. Lugagne, F. Tixier, R. Genin, and J-L. Yvin Severe community-acquired pneumonia: assessment of microbial aetiology as mortality factor Eur. Respir. J., November 1, 2004; 24(5): 779 - 785. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. Brunelli, M. Monteverde, A. Borri, M. Salati, R. D. Marasco, and A. Fianchini Predictors of prolonged air leak after pulmonary lobectomy Ann. Thorac. Surg., April 1, 2004; 77(4): 1205 - 1210. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
