Although a pleural effusion (PE) is a common condition in clinical practice (with a prevalence of slightly > 400 patients/100,000), a definitive diagnosis is not established in approximately 20% of cases. In the classic publications, thoracentesis provided a definitive or presumptive diagnosis in 73% of cases, although it is estimated that today this percentage could reach 95%. An undiagnosed PE is defined as one that remains of unknown origin after performing complete pleural fluid analysis (nucleated cell counts with differential diagnosis, biochemistry, culture, cytology, and flow cytometry). The aim of this paper is to review the difficulties that may be encountered when attempting to establish the cause of a PE. The reason for this diagnostic dilemma may be an atypical clinical or radiologic presentation or falsepositive or false-negative results. Furthermore, the limited frequency of some PEs is often not considered when determining the differential diagnosis.