American Journal of Respiratory and Critical Care Medicine

Good research is supposed to clear up controversy and move us forward with greater understanding. Just occasionally, however, good research does exactly the opposite.

Community-acquired pneumonia (CAP) has always been a disease characterized by educated guesswork. As pathological confirmation of pneumonia is rarely obtained, the diagnosis of CAP is a presumptive one based on the history, clinical signs, and a chest X-ray deemed consistent with acute consolidation (1). Because the pathogen is almost never known at the time treatment commences, educated guesswork (based on the most locally applicable etiological studies) is further employed to select the most appropriate antibiotic therapy (1).

Implicit in our whole approach to CAP is that sometimes we do get the diagnosis wrong. There are a variety of rarer diseases that may present like CAP. An initial chest X-ray may not show infiltrates easily apparent 24–72 hours later. Occasionally, chronic changes on chest X-ray may be misinterpreted as acute pathology. Despite these limitations, the general perception of clinicians is that we get the diagnosis of CAP right most of the time, and not only have we become comfortable with our empiric approach to therapy but also, health payers are comfortable with higher reimbursements for pneumonia than other nonspecific respiratory infections, and CAP is one of the most stringently monitored conditions with respect to quality-of-care measures.

In this issue of the Journal, Claessens and colleagues (pp. 974–982) present their study of 319 patients with a clinical diagnosis of CAP who had both a chest X-ray and a thoracic computed tomography (CT) scan at the time of admission (2). Disturbingly, 30% of patients who were felt to have CAP based on the presentation and chest X-ray had no evidence of pneumonia on CT scan. Furthermore, one third of patients who had no change on chest X-ray had CT changes consistent with pneumonia. Overall, the CT scan results showed that the combination of clinical features and chest X-ray lead to a misdiagnosis of the absence or presence of CAP in nearly one-third of all patients studied, and clinicians adjusted their perception of the likelihood of CAP being the diagnosis in more than 50%. If confirmed by further studies, this shifts the assessment of the diagnosis of CAP from “we might occasionally get it wrong” to “Houston, we have a problem.” The implications for everything from empiric therapy to reimbursement and quality of care measures are enormous.

Is a CT scan is an adequate gold standard for the diagnosis of pneumonia? There are no studies correlating CT scan results with pathology in the setting of CAP, nor are there likely to be in anything other than severe, fatal disease. In the absence of data to the contrary, then, it seems reasonable to accept CT evidence of consolidation as the gold standard. Although advances in CT scanning have dramatically reduced the acquisition time, cost, and radiation exposure such that modern generations of machines could perceivable replace plain radiography during the next decade, for the time being, a plain chest X-ray will remain the primary diagnostic tool.

How, then, should clinicians respond to the challenge from Claessens and colleagues that we are getting the diagnosis wrong in a third of our patients?

First, we should be reassured that selection of antibiotic therapy is still based on etiological data from studies on patients that other clinicians were comfortable calling CAP. If we are getting it wrong now, there are no data to suggest key etiological studies were getting it more or less wrong before, so the studies are still valid and there is no urgent need to change our approach. Moving forward, we are going to need to know whether CT-positive, chest X-ray (CXR)-negative CAP has the same prevalence of pathogens and the same clinical outcomes as CT-positive, CXR-positive disease. Equally we will need to know whether patients deemed to have CAP but with a subsequent negative CT scan have different pathogens or outcomes. These will be critical questions to not just clinicians but also health-payers and those focused on quality-of-care metrics.

Are there any patients in whom a CT scan should be currently performed at admission, based on the data from Claessens and colleagues (2)? It was notable that the only real predictors of pneumonia being present on a CT scan in the setting of a negative chest X-ray were the presence of unilateral crackles and a very high C-reactive protein (CRP). Examining patients thankfully remains important. Conversely, lower CRP levels were really the only helpful marker of a potential false-positive chest X-ray diagnosis. Interestingly, in the setting of both false-positive and false-negative chest X-rays, procalcitonin was not a discriminator, and although there were small differences in the white cell count, these differences were unlikely to be clinically helpful.

Finally, it is worth reflecting on why clinicians may be overcalling the diagnosis of CAP based on chest X-rays. In the right clinical context, convincing ourselves that pneumonia might be present based on soft radiological changes allows us to go down a well-validated clinical pathway. If we decide soft radiological changes are not pneumonia, then we have no clear clinical pathways, no metrics to tell us quality of care is being met, and in some settings, possibly reimbursement issues. Although radiologists and research studies like to talk in terms of probability of pneumonia (definite, probable, possible) for clinicians, health-payers, and quality-of-care metrics, the patient either does or does not have CAP. What we need is as clear data on what to do if we have a CAP-like syndrome but decide pneumonia is not present on CXR as we have for when it is.

If we really are getting the diagnosis wrong as often as Claessens and colleagues suggest we are (2), then we need to start back at the beginning and redefine the etiology and outcomes of patients with CAP-like presentations based on a CT gold standard overriding the chest X-ray interpretation. In this respect, I cannot help but wonder whether older categorizations we have largely abandoned, such as lobar pneumonia and bronchopneumonia, will come back as we try to better define subgroups of patients, especially those in whom the chest X-ray is sufficient and a CT scan not required. I am, however, quite sure that questioning the fundamental diagnosis of CAP is now a major consideration when studies produce differing results, and much of what we think is true now needs to be reevaluated in the light of how new technology can better inform us of what is really going on in individual patients.

1. Mandell LA, Wunderink RG, Anzueto A, Bartlett JG, Campbell GD, Dean NC, Dowell SF, File TM Jr, Musher DM, Niederman MS, et al.; Infectious Diseases Society of America; American Thoracic Society. Infectious Diseases Society of America/American Thoracic Society consensus guidelines on the management of community-acquired pneumonia in adults. Clin Infect Dis 2007;44:S27S72.
2. Claessens YE, Debray MP, Tubach F, Brun AL, Rammaert B, Hausfater P, Naccache JM, Ray P, Choquet C, Carette MF, et al. Early chest computed tomography scan to assist diagnosis and guide treatment decision for suspected community-acquired pneumonia. Am J Respir Crit Care Med 2015;192:974982.

Author disclosures are available with the text of this article at

Comments Post a Comment

New User Registration

Not Yet Registered?
Benefits of Registration Include:
 •  A Unique User Profile that will allow you to manage your current subscriptions (including online access)
 •  The ability to create favorites lists down to the article level
 •  The ability to customize email alerts to receive specific notifications about the topics you care most about and special offers
American Journal of Respiratory and Critical Care Medicine

Click to see any corrections or updates and to confirm this is the authentic version of record