The Role of Diagnostic Thoracentesis in the Emergency Department
What is the role of diagnostic thoracentesis in the emergency department? This is a question that I recently found myself asking.
The Undifferentiated Pleural Effusion
The space between the visceral and parietal pleurae is often described as a potential space but in reality –even in a healthy individual—is a space that is occupied by fluid. In a non-smoking, otherwise healthy individual, there is about 0.13-mL/kg of pleural fluid in each space. Pleural effusions can occur when there is an increase in fluid entry into the pleural space or there is a decrease in the fluid removal but most commonly occurs because of both. Pleural effusions have traditionally been classified as transudative versus exudative and thoracentesis with application of Light’s Criteria is used to differentiate the two with 98% sensitivity and 83% specificity for exudative process. Transudative effusions include CHF, nephrotic syndrome, malnutrition, cirrhosis and exudative effusions include infection/empyema, malignancy, infarction, rheumatological etiologies, and pancreatitis among others. The most common causes of pleural effusion in the United States are congestive heart failure, pneumonia, and cancer with annual incidences of about 500,000, 300,000, and 200,000 respectively. 
|Differentiation of Exudative and Transudative Pleural Effusions (Light’s Criteria)|
|Ratio of Pleural:Serum Protein||>0.5||<0.5|
|Pleural LDH||>2/3 of the upper limit of normal||<2/3 of the upper limit of normal|
|Ratio Pleural:Serum LDH||>0.6||<0.6|
|Only one criteria in favor of exudative is needed
Adapted from Light et al.
Often times with patients with pleural effusions, there is high pretest probability for a certain etiology and there is probably not a need for emergent diagnostic thoracentesis. For instance, an 89 year-old female with known cancer presenting with recurrent pleural effusion and symptoms consistent with previous pleural effusions may not require a thoracentesis in the emergency department for diagnostic purposes. Even a patient with new onset pleural effusions may not require diagnostic thoracentesis; for instance, a 58 year-old male with congestive heart failure and bilateral effusions of similar size, is afebrile, and has no chest pain may just need 48 hours of diuresis. In fact, in a series of 33 cases, clinicians were able to correctly identify all 17 transudates with clinical assessment alone (later confirmed by Light’s Criteria).
However, it would seem there is a subset of pleural effusions that remain undifferentiated even after taking in the patient’s clinical scenario; especially if there is concern for exudative processes. Consider, for instance, the patient who presents with mild dyspnea, right pleuritic pain, temperature of 37.7 who has ‘atelectasis’ on x-ray and new 1.0-cm mass on CT without PE. Is this pleural effusion secondary to this mass or an infectious process? Knowing this information would greatly change the physician’s management of this patient. It is simply not appropriate to delegate this care to an inpatient team (at least not without a discussion). At many facilities, the patient with an undifferentiated pleural effusion who is admitted overnight may not have a thoracocentesis done until late morning for various reasons. Does this mean 14-hours of unnecessary antibiotic coverage? Does this mean no necessary antibiotic coverage? It is a little bit of comparing apples and oranges, but one would never subject the patient with concern for septic arthritis or spontaneous bacterial peritonitis to these conditions. Delays in treatment of parapneumonic effusions are not benign. What starts out as a small to moderate free flowing effusion that could be managed with antibiotics alone transforms into a more loculated picture with increasing risk of sepsis, poor outcome, need for drainage and possibly even surgical intervention.
Diagnostic Thoracentesis: Indications and Contraindications
Indications for a diagnostic thoracentesis is the presence of a clinically significant pleural effusion (more than 10-mm thick on ultrasound or lateral decubitus radiography) with no known cause.  There are no absolute contraindications to diagnostic thoracentesis. Recently there have been studies that even those with coagulation abnormalities can be safely done under the guidance of ultrasound. 
Diagnostic Thoracentesis: Well within the emergency medicine physician’s scope of practice
There are several known complications of thoracentesis including pneumothorax (as high as 6%), cough, infection and less common complications including hemothorax, splenic rupture, reexpansion pulmonary edema (uncommon in general but especially so in diagnostic thoracentesis), and air embolism.  Pneumothoraces are known complications and can occur with even an experienced physician and are three times more likely in a patient with BMI<18, nearly four times more likely in large volume (>1500-mL), and three times more likely if a second pass is required. There have been numerous studies demonstrating that the complications are reduced with the help of ultrasound.[9-11]
Barriers that currently exist to emergency medicine physicians preforming diagnostic thoracenteses are discomfort with the procedure and a need to facilitate departmental flow. Again, the emergency medicine physician need not be intimated by the thoracentesis. Especially as we are called to far more invasive procedures on a regular basis such as central venous catheters and lumbar punctures and we can emergently manage all complications of thoracentesis. While flow is important to a department, so is timely administration of appropriate antibiotics. So if you are an emergency medicine physician and not actively preforming diagnostic thoracenteses on undifferentiated pleural effusions, consider adding it your practice.