Purpose: Surgical complications during sinus floor elevation are frequently caused by septa in the maxillary sinus. In this study, the prevalence of septa was retrospectively determined via three-dimensional imaging to assess the necessity for appropriate imaging prior to sinus floor elevation. Materials and Methods: The cone-beam computed tomography scans of 1,029 consecutive patients were evaluated to assess the prevalence of septa, taking into account patient age and sex as well as the number and orientation of septa and their locations. If septa were present, the height was measured and the orientation was determined. Results: Septa were found in 47% of patients and 33.2% of sinuses. There was no statistical difference in prevalence with respect to age, sex, or side. Most patients with septa showed one septum in one sinus (24.6%); 13.7% showed one septum in each sinus. Other combinations (up to three septa per sinus) were found in 8.7% of patients. Septa were most commonly located in the first molar region (256 septa), followed by the second molar region (225 septa), the third molar region (144), second premolar region (136), the first premolar region (44), and the canine region (5). The mean septal height was 11.7 ± 6.08 mm for septa in a sagittal orientation (n = 206, 25.3%; maximum height of 37 mm) and 7.3 ± 5.08 mm for those oriented transversely (n = 608, 74.7%; maximum height of 36 mm). The variance was larger for the sagittal orientation group (37.03 mm) than for transversely oriented septa (25.9 mm). Conclusion: Septa were found in 47% of the patients in this sample. Appropriate imaging prior to performing sinus surgery seems justified, since complications and the success rate of sinus floor elevation are clearly related to the presence of septa. Palabras clave: antral septa, cone-beam computed tomography, dental implants, dental radiography, sinus floor elevation, sinus membrane