Pre‐Surgical Risk Assessment for Sinus Membrane Perforation: A Retrospective Cohort Study

ABSTRACT Aim This study aims to validate an existing risk assessment (RA) tool for predicting sinus membrane perforation (SMP) during the lateral‐window sinus floor elevation (LSFE). Materials and Methods A multiple‐center study was conducted in a hospital and university setting. Cone‐beam computed tomography (CBCT) images and health records of patients who received LSFE from 2005 to 2021 were assessed. The effectiveness of the RA tool was evaluated on our dataset and additional analyses were performed to identify the key risk factors using logistic regression and the classification and regression trees (CART) methods. Results A total of 126 sinuses in 113 patients were assessed, including 49 females and 64 males (60.1 ± 11.65 y/o). There were 24 SMP events in the study population. Patient‐level and sinus‐level analyses were consistent. The reduced logistic regression model suggested that the odds ratio (OR) of having an SMP event decreases by 23.7% with a 1 mm increase in residual bone height (RBH). Using a 3‐mm RBH as the cut‐off between “low risk” and “high risk,” rather than 4 mm as suggested in the original RA tool, provided better separation of SMP rates. CART analysis revealed that age may also be related to SMP risk. Conclusion In this exploratory study, RBH was the factor most correlated with SMP risk during LSFE, followed by age at the time of surgery. The critical thresholds for predicting SMP events were identified as 3.02 mm for RBH and 55.5 years of age. Within the inherent limitations, the current risk assessment tool could not be fully validated.