Proposed cut-off for harmless parapneumonic effusion: 2.5 cm on CT (Eur Resp J)
Richard Light and friends established that parapneumonic effusions (PPE) associated with community-acquired pneumonia very rarely progress if the effusion is freely layering and less than 1 cm in height on a lateral decubitus chest film. But who orders those anymore?
Chest CT use has risen 20-fold since Light's seminal 1980 paper. Often the CT has been done already in the emergency department, and its rich data combined with the potentially harmful delay involved in getting a seemingly redundant lateral decubitus X-ray have discouraged most physicians from ordering the old-school chest films, despite their proven track record.
Trying to update Light's work for a new technological era, B.K. Moffett et al propose that a 2.5 cm thickness of freely layering fluid on chest CT (at the fluid's thickest point) is equivalent to the 1 cm lateral decubitus cut-off, in predicting the safety of forgoing drainage of parapneumonic effusions.
They filtered retrospective data from a single VA center in Lousiville Kentucky, to obtain a cohort of 419 patients with parapneumonic effusions after either CAP or hospital-associated pneumonia (2005-2009), very few of whom got thoracentesis on initial presentation:
Among 79 patients who had both a CT scan and a lateral decubitus film, the 2.5 cm CT cut-off captured the vast majority of those with a >1 cm PPE on lateral decubitus film. (But not all; r-squared was ~0.7 and there were several outliers.)
Among 95 patients who had a PPE < 2.5 cm on CT, all but 3 were managed without thoracentesis. Only one patient had a bad outcome attributable to the PPE, and that person's effusion was < 1 cm on lateral decubitus film initially (but rapidly progressed). Four patients had unobtainable data, mostly due to hospice.
This was retrospective, and so there was baked-in bias, as the decision not to perform thoracentesis had already been made (and certainly, that decision would have included patients' clinical stability). Ultrasound was not incorporated here, even though it's arguably the standard of care in the U.S. today. These authors' argument is that based on this data, for PPEs of < 2.5 cm on a CT scan (done by the emergency department, for instance), no further imaging or drainage procedure is necessary (unless the patient's condition warrants it).
Moffett BK et al. Computed tomography measurements of parapneumonic effusion indicative of thoracentesis. Eur Resp J 2011;38:1406-1411.