| 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 |
|||||||||
Buenos Aires, Argentina
Dr. Luna is Associate Professor of Internal Medicine, University of Buenos Aires.
Correspondence to: Carlos M. Luna, MD, FCCP, Acevedo 1070, Banfield (CP 1828), Buenos Aires, Argentina; e-mail: cymluna{at}giga.com.ar
The traditional diagnostic approach in community-acquired pneumonia (CAP) includes the performance of two sets of blood cultures (BCs).1 In different recent clinical practice guidelines for CAP, the performance of BCs has been systematically recommended for the hospitalized patients, preferably before antibiotic treatment is commenced.2 3 4 5 Nevertheless, the category of evidence supporting this recommendation ranges from moderate to low degree (expert opinion and informal consensus).
Studies of large series of patients with CAP usually report that positive BC results are present in approximately 10% of those admitted to the hospital and that most of these bacteremias are due to pneumococcus.6 7 8 Positive BC results are more common among the severely ill patients and in patients who did not receive antibiotics before hospitalization.
In this issue of CHEST (see page 1142), Campbell et al report a prospective observational study to investigate the contribution of BCs to the management of patients with CAP. They concluded that in spite of the fact that a positive BC result affords reassurance to clinicians, the chance of changing the therapy directed by the result of the culture is very low and there is little evidence that it is cost-effective. They also observed that the severity of illness measured by the pneumonia severity index (PSI) correlated poorly with the yield of BC. And so, what is the reason that, in spite of the background given by such weak evidences supporting the performance of BCs in every patient with CAP admitted to the hospital, this diagnostic tool is still universally recommended?
Bryan,1 in a previous editorial published in CHEST about the value of BCs in CAP, argued that the cost is not so high, it may be invaluable for decision making, data derived from susceptibility testing are crucially important, and it is debatable whether simplifying therapy would actually save money. This author concluded that the time has not yet come to abandon BCs in CAP requiring hospitalization.1 In contrast, Theerthakarai et al,9 in another study, concluded that BCs have no value in the management of hospitalized patients with nonsevere CAP without comorbid factors.
Several issues should be considered together when trying to answer why this diagnostic tool is still recommended in every patient admitted with CAP. Among these reasons, I believe that the usefulness of detecting bacteremia for the diagnosis, therapy, and risk evaluation of CAP is an important topic that should be discussed.
In relation to the usefulness of BCs in the diagnosis of disease, this test is not useful most of the time, as positivity becomes evident no earlier than 24 h after obtaining the specimen; however, late confirmation of bacteremia, especially due to some specific pathogens (ie, Streptococcus pneumoniae or Haemophilus influenzae), may occasionally be of help to confirm retrospectively the diagnosis in patients presenting an acute respiratory illness. Concerning the etiologic diagnosis, the value of BC is limited, as only some of the pathogens can be detected by BCs, and bacterial pathogens produce bacteremia occasionally (pneumococcus being the most frequently isolated organism). Although BCs are highly specific to confirm the etiology, its sensitivity in CAP is too low. As a consequence, the usefulness of BCs as a diagnostic test does not justify its generalized use.
Regarding the impact of BCs on the prescription of antibiotics in CAP, it is widely accepted that it is advisable to initiate the administration of antibiotics as soon as possible, preferably during the first 8 h after hospital admission, to improve patient survival.10 This fact mandates to prescribe antimicrobials empirically in nearly all the cases; however, the potential benefit of further modifications on the therapy enhancing the coverage of pathogens according to the BC result has no practical utility. Therapy must be appropriate from the start, because inadequate initial antibiotic therapy has been found to be associated with higher mortality rate.11 12 Moreover, Waterer and Wunderink,13 in a study on the usefulness of BCs in CAP, found that 4 of 16 patients (25%) with culture-guided change in antibiotic therapy died, compared to 5 of 31 patients (16.1%) who had an empiric change. The result of BCs could be of value in proceeding to step-down the initial empiric therapy, even though physicians seldom simplify antimicrobial therapy, even when penicillin-sensitive S pneumoniae has been isolated from BC. This was the case in 12 of 29 patients with CAP in the study by Campbell et al published in this issue of CHEST. Similarly, in a previously published study on bacteremic pneumococcal pneumonia, physicians rotated to penicillin in only one of five cases due to penicillin-sensitive pathogens and abandoned coverage for atypicals in only one of three cases.14
Finally, linked to risk evaluation, univariate analysis suggested that bacteremia could be associated with higher mortality in severe CAP.11 15 In contrast, Moroney et al,16 in a case-control study, demonstrated that invasive infection due to pneumococcus is more common in the more severely ill patients, but when cases were compared with matched control subjects, admission to the ICU and mortality rate were not significantly different, suggesting that bacteremia is more the consequence than the cause of severity. Another study14 found that nearly half of the patients hospitalized with CAP were in grades I, II, and III according to the PSI score, suggesting that pneumonia severity should not be considered to be a predictor of the presence of bacteremia. This lack of correlation between the PSI score and positive BC result was present also in the study by Campbell et al. Other investigations suggested that the higher mortality rate observed in patients with bacteremia could reflect residual confounding due to severity of illness, since increasing pneumonia severity on hospital admission has been associated with blood culture performance (p = 0.009).17
In conclusion, although several authors have argued to the contrary, there are enough data on the reduced usefulness of BC for the diagnosis, therapy, and risk evaluation in CAP. Based on this data, it is time to overcome the hesitation. In my opinion, chest physicians are ready to limit the recommendation of BCs to patients with severe CAP and those not responding to therapy.
If it be now, ’tis not to come; if it be not to come, it will be now; if it be not now, yet it will come. The readiness is all.
Hamlet: Act v; scene ii.
William Shakespeare
References
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
