Correlation Between Facial Index and Maxillary Sinus Dimensions in CBCT Scans

Background The maxillary sinus, a key component of the midfacial skeleton, exhibits significant anatomical variability. The facial index, a craniometric ratio, classifies facial morphology and is influenced by underlying skeletal structure. Cone-beam computed tomography (CBCT) provides a precise three-dimensional platform for evaluating these anatomical relationships. Methods A retrospective study was conducted on 180 CBCT scans of adult patients (90 males, 90 females), categorized into three groups based on facial index: euryprosopic (n=60), mesoprosopic (n=60), and leptoprosopic (n=60). Facial index was calculated from 3D surface renderings. Maxillary sinus dimensions (width, height, length, and volume) were measured on axial, coronal, and sagittal reconstructions. Statistical analysis included ANOVA, post-hoc Tukey tests, and Pearson correlation. Results A significant correlation was observed between facial index and maxillary sinus morphology. Euryprosopic individuals (wide faces) had significantly wider sinuses (mean width: 30.2 ± 2.5 mm) compared to mesoprosopic (27.8 ± 2.2 mm) and leptoprosopic (25.1 ± 2.1 mm) individuals (p < 0.001). Conversely, leptoprosopic individuals (long faces) had significantly longer sinuses (mean length: 38.5 ± 3.1 mm) compared to other groups (p < 0.001). A strong negative correlation was found between facial index and sinus width (r = -0.68, p < 0.001), and a strong positive correlation with sinus length (r = 0.62, p < 0.001). Conclusion The facial index is a significant predictor of maxillary sinus dimensions. Individuals with wider faces tend to have wider, shorter sinuses, while those with longer faces have narrower, longer sinuses. This correlation provides valuable anthropological data and has practical clinical applications in maxillofacial surgery, ENT, and implantology for pre-operative planning and anatomical assessment. Bangladesh Journal of Medical Science Vol. 25. Supplementary Issue 2026, Page : S124-S128