The prognostic role of ultrasound and magnetic resonance imaging in obstructive sleep apnoea based on lateral oropharyngeal wall obstruction

Abstract

Purpose This study examined the prognostic value of the lateral pharyngeal wall (LPW)-based obstruction and obstructive sleep apnoea (OSA) prediction using ultrasound (US) and MRI (magnetic resonance imaging). Methods One hundred patients with and without OSA were enrolled, according to overnight polysomnography. The LPW thickness (LPWT) was measured using a Philips Ingenia 1.5 T MRI device, and US measurements were carried out at rest and during Müller’s manoeuvre (MM) with a Samsung RS85 device. The obstruction was localised under drug-induced sleep endoscopy. Results Significantly greater LPWT using MRI was observed in the OSA group compared to the control group, while US results showed a significant difference only in the case of LPWT during MM on the left side. Obese patients presented significantly higher LPWT values. A significant correlation between BMI and LPWT was observed. Men presented significantly higher LPWT MRI values and left-sided LPWT using US compared to women. LPWT and AHI parameters were significantly correlated. The severity of LPW obstruction correlated with LPWT, while the LPW collapse significantly correlated with AHI. The severity of LPW collapse differed depending on the AHI values. Using US LPWT values and anthropometric parameters, a 93% effectiveness in OSA prognostication and 89% in LPWT-based obstruction were detected. MRI detected OSA in 90% and LPW-based obstruction in 84%. US successfully detected LPW-based collapse severity in 67%. Conclusion US LPWT measurements were helpful in detecting OSA and LPWT-based obstruction. These examinations may be useful for surgical planning.

Obstructive sleep apnea detection during wakefulness: a comprehensive methodological review

“…Contrary to previous work [ 36 ], both the lateral and anterior–posterior diameters were found smaller in OSA subjects in [ 37 ]; this could be due to the small sample size in the study reported in [ 36 ], noting that, although the difference in lateral diameter was significant, it was smaller than the difference in anterior–posterior diameter. Furthermore, 2D soft tissue measurements showed that the lateral pharyngeal wall was larger in OSA subjects and was associated with an increased risk of OSA [ 32 ]; this is congruent with the findings in [ 32 ] using 1.5-T MRI that showed larger LPW thickness in OSA group, while the associated risk of the volumetric measurements was substantially greater [ 32 ].…”

“…Another research examined the prognostic value of the lateral pharyngeal wall’ (LPW) thickness for the detection of OSA using two both US and MRI [ 32 ]. One hundred individuals with and without OSA (36 healthy controls and 64 OSA subjects) were enrolled in the study and performed an overnight PSG, and then the LPW thickness was measured using 1.5-T MRI and ultrasound during wakefulness.…”

“…However, the DISE succeeded in identifying the incidence of multilevel collapses ( p = 0.001), while the incidence of oropharyngeal obstruction in patients with Mallampati scores I and II was significantly higher in DISE compared to that measured by Müller’s maneuver ( p = 0.021). Some other studies have focused on the ability to use ultrasound or MRI imaging of the tongue for OSA prediction [ 32 , 34 ]. One research [ 34 ] showed the ability of OSA prediction using AI applied to ultrasound imaging of the UA and subcutaneous adipose tissues (SAT) in the regions of the neck, chest, and abdomen measurements [ 35 ].…”

“…These results suggest a potential application for dynamic OSA imaging during wakefulness. In the surveyed literature, three studies used MRI imaging to observe the UA regions during wakefulness [ 31 , 32 , 36 , 37 ]. The findings in [ 36 ] were consistent with those using CT imaging [ 29 , 30 ].…”

“…Some other studies have focused on the ability to use ultrasound or MRI imaging of the tongue for OSA prediction [ 32 , 34 ]. One research [ 34 ] showed the ability of OSA prediction using AI applied to ultrasound imaging of the UA and subcutaneous adipose tissues (SAT) in the regions of the neck, chest, and abdomen measurements [ 35 ].…”