Diagnostic and therapeutic thoracentesis may be complicated by iatrogenic pneumothorax, with a reported incidence of 3 to 20 percent. It is standard procedure to order a posteroanterior chest radiograph in patients who have undergone thoracentesis to rule out this complication, but it is unclear whether this is a cost-effective or useful test. Alemán and colleagues conducted a prospective study to assess the yield of the follow-up chest radiographs and to determine whether the clinical symptoms were associated with the presence of pneumothorax in these patients.

Consecutive patients requiring thoracentesis for therapeutic or diagnostic reasons were included in the study. Patient information (e.g., age, gender, size of the pleural effusion and presence of bullae) and the procedure characteristics (e.g., number of needle passes before fluid was obtained, evidence of air aspiration and symptoms during and after the procedure) were obtained. Each patient had a chest radiograph within 12 hours after the thoracentesis. The radiologist who interpreted the films was blinded to the patient and procedure characteristics.

In 370 patients, 506 thoracenteses were evaluated. Air aspiration during three procedures caused the clinician to suspect a pneumothorax. Symptoms during 18 other procedures (including cough, dyspnea, pleuritic chest pain and a combination of these symptoms) raised the suspicion of pneumothorax. Of the 488 asymptomatic patients, five developed a pneumothorax. Only one of these five patients required a chest tube. Of the 18 symptomatic patients, 13 developed a pneumothorax and eight patients required chest tubes. Thus, eight of the nine patients who required chest tubes could have had their pneumothorax predicted. If chest radiographs were restricted to symptomatic patients, 488 radiographic studies would have been avoided and only one serious pneumothorax would have been missed in the 12 hours following the thoracentesis.

The authors conclude that patients who are post-thoracentesis and are symptomatic are the only ones who should have a post-procedure chest radiograph. Using this guideline, there would be essentially no change in patient management, and a significant savings would be realized.