藤井 直子

Objective: The Eustachian tube is difficult to evaluate because it is located deep in the head. However, the introduction of 320-row area detector CT has made it possible to evaluate this region. In the present study, movement of the Eustachian tube during sniffing was visualized using area detector CT in patients with patulous Eustachian tube. Methods: Four patients with patulous Eustachian tube were examined using an area detector CT scanner (Aquilion ONE, Toshiba). This scanner supports 320-row scanning of 0.5-mm slices at up to 0.275 s/rot., eliminating temporal mismatch between various parts of the acquired images and permitting 4-dimensional CT (4DCT) images to be obtained by continuous scanning. The scan conditions were 120 kV, 120 to 150 mA, 0.5 mm x 280 to 320 slices, and 0.35 seconds per rotation x 9 rotations. The patient was seated on a reclining chair tilted to 45 degrees and was instructed to sniff during continuous scanning. Images of the Eustachian tube were generated at 0.1-second intervals. Conclusion: At the start of sniffing, the cartilaginous portion of the Eustachian tube closed from the isthmus toward the pharynx. The starting point differed from patient to patient. In patients with patulous Eustachian tube, sniffing (an unconscious habit that helps to relieve ear discomfort) is an important factor in the development of middle ear diseases. We have successfully depicted this event for the first time, demonstrating various patterns of Eustachian tube closure during sniffing in patients with patulous Eustachian tube. This method may be useful for evaluating Eustachian tube dysfunction.

In patients with unilateral pharyngeal paresis and dysphagia, the head is rotated to the paretic side to prevent food flow to the rotated side during swallowing. Only a few studies to date have reported on pyriform sinus morphology upon head rotation. The purpose of this study was to measure the volume, depth, and cross-sectional area of the pyriform sinus during head rotation using 320-row area detector computed tomography. We imaged the neck during head rotation at 0A degrees and at 30A degrees, 45A degrees, and 60A degrees to the left or right in nine healthy young adults and determined the volume, depth, and cross-sectional area of the pyriform sinus in each position. On the rotated side, volume and cross-sectional area were significantly decreased at 60A degrees. In contrast, volume, cross-sectional area, and depth were all significantly increased on the opposite side at 60A degrees. These results suggest that head rotation at 60A degrees significantly increases the volume, cross-sectional area, and depth of the opposite side, and significantly decreases the volume and depth of the rotated side of the pyriform sinus.

The present study examined the effect of bolus viscosity on the onset of laryngeal closure (relative to hyoid elevation), the duration of laryngeal closure, and other key events of swallowing in ten healthy volunteers. All volunteers underwent 320-row area detector computed tomography swallow studies while swallowing 10 ml of honey-thick barium (5 % v/w) and thin barium (5 % v/w) in a 45A degrees reclining position. Three-dimensional images of both consistencies were created in 29 phases at an interval of 0.10 s (100 ms) over a 2.90-s duration. The timing of the motions of the hyoid bone, soft palate, and epiglottis; the opening and closing of the laryngeal vestibule, true vocal cords (TVC), and pharyngoesophageal segment; and the bolus movement were measured and compared between the two consistencies. The result showed differing patterns of bolus movement for thin and thick liquids. With thin liquids, the bolus reached the hypopharynx earlier and stayed in the hypopharynx longer than with thick liquids. Among events of laryngeal closure, only the timing of TVC closure differed significantly between the two consistencies. With thin liquids, TVC closure started earlier and lasted longer than with thick liquids. This TVC movement could reflect a response to the faster flow of thin liquids. The results suggest that bolus viscosity alters the temporal characteristics of swallowing, especially closure of the TVC.

The purpose of this study was to (1) depict normal dynamic swallowing and (2) measure (a) the temporal characteristics of three components of laryngeal closure, i.e., true vocal cord (TVC) closure, closure of the laryngeal vestibule at the arytenoid to epiglottic base, and epiglottic inversion, and (b) the temporal relationship between these levels of laryngeal closure and other swallowing events, hyoid elevation, and the pharyngoesophageal segment (PES) using 320-detector-row multislice computed tomography (320-MSCT). The swallowing of a 10-ml portion of honey-thick liquid (5% w/v) was examined in six healthy volunteers placed in a 45A degrees reclining position. Three-dimensional CT images were created in 29 phases at an interval of 0.10 s over a 2.90-s duration. Dynamic swallowing and TVC movement were depicted clearly. The sequence for laryngeal closure was the following: (1) the hyoid started to elevate, (2) the PES opened, (3) TVC closure and closure at the arytenoid to epiglottic base occurred almost simultaneously during the hyoid elevation, and (4) the epiglottic maximum inversion occurred after the hyoid maximum displacement. Those results indicated that the onset of hyoid elevation and the early opening of the PES occurring before three levels of laryngeal closure are critical components for airway protection. 320-MSCT allowed the 3D depiction and kinematic analysis of target structures, which will increase our knowledge of airway protection mechanisms during swallowing.

A 320-detector-row multislice computed tomography (320-MSCT) scanner can acquire a volume data set covering a maximum range of 16 cm and can generate axial images 0.5-mm thick at 0.5-mm intervals. Three-dimensional (3D) images reconstructed from the thin axial slices include multiplanar reconstruction and 3D-CT. Single-phase 3D images are reconstructed from 0.175-s data, and multiphase 3D images are created in 29 phases at intervals of 0.1 s. Continuous replay of these 3D images produces four-dimensional moving images. In order to determine the feasibility of the morphologic and kinematic analyses of swallowing using 320-MSCT, single-phase volume scanning was performed on three patients and multiphase volume scanning was performed on one healthy volunteer. The single-phase 3D images clearly and accurately showed the structures involved in swallowing, and the multiphase 3D images were able to show the oral stage to the early esophageal stage of swallowing, allowing a kinematic analysis of swallowing. We developed a reclining chair that allows scanning to be performed with the subject in a semisitting position, which makes swallowing evaluation by 320-MSCT applicable not only to research on healthy swallowing but also to the clinical examination of dysphagia patients.