Potential of computational models in personalized treatment of obstructive sleep apnea: a patient-specific partial 3D finite element study

Abstract

The upper airway experiences mechanical loads during breathing. Obstructive sleep apnea is a very common sleep disorder, in which the normal function of the airway is compromised, enabling its collapse. Its treatment remains unsatisfactory with variable efficacy in the case of many surgeries. Finite element models of the upper airway to simulate the effects of various anatomic and physiologic manipulations on its mechanics could be helpful in predicting surgical success. Partial 3D finite element models based on patient-specific CT-scans were undertaken in a pilot study of 5 OSA patients. Upper airway soft tissues including the soft palate, hard palate, tongue, and pharyngeal wall were segmented around the midsagittal plane up to a width of 2.5 cm in the lateral direction. Simulations of surgical interventions such as Uvulopalatopharyngoplasty (UPPP), maxillo-mandibular advancement (MMA), palatal implants, and tongue implants have been performed. Our results showed that maxillo-mandibular advancement (MMA) surgery of 1 cm improved the critical closing pressure by at least 212.2%. Following MMA, the best improvement was seen via uvulopalatopharyngoplasty (UPPP), with an improvement of at least 19.12%. Palatal and tongue implants also offered a certain degree of improvement. Further, we observed possible interacting mechanisms that suggested simultaneous implementation of UPPP and tongue stiffening; and palatal and tongue stiffening could be beneficial. Our results suggest that computational modeling is a useful tool for analyzing the influence of anatomic and physiological manipulations on upper airway mechanics. The goal of personalized treatment in the case of OSA could be achieved with the use of computational modeling.