Cone beam computed tomography and obstructive sleep apnoea

Authors


Professor Robert Strauss
VCU Medical Center
PO Box 980566
Richmond
Virginia 23233
USA
Email: rstrauss@mcvh-vcu.edu

Abstract

Sleep apnoea is characterized by an intermittent cessation or diminution of airflow during sleep that may result in significant pulmonary and cardiac consequences, and is associated with significant morbidity and mortality. It is a common disease that affects approximately 20% of patients who snore, equating to a total prevalence in the population of about 4% of all middle aged men and 2% of women. In reality, due to lack of recognition and the difficulty with obtaining an accurate diagnosis, the actual incidence is likely much higher.

Abbreviations and acronyms:
A-P

anterior-posterior

BMI

body mass index

CBCT

cone beam computed tomography

CSA

central sleep apnoea

FOV

field of view

GORD

gastro-oesophageal reflux disease

GTA

genial tubercle advancement

LAT

lateral

MMA

maxillomandibular advancement

OSA

obstructive sleep apnoea

PSG

polysomnogram

Introduction

Sleep apnoea is characterized by an intermittent cessation or diminution of airflow during sleep that may result in significant pulmonary and cardiac consequences, and is associated with significant morbidity and mortality. It is a common disease that affects approximately 20% of patients who snore, equating to a total prevalence in the population of about 4% of all middle aged men and 2% of women.1 In reality, due to lack of recognition and the difficulty with obtaining an accurate diagnosis, the actual incidence is likely much higher.

Many dentists are now involved in the diagnosis and management of sleep apnoea. General dentists, prosthodontists, orthodontists and periodontists have become primarily involved in the construction and placement of oral appliances, which have recently been shown to be reasonably effective for mild to moderate sleep apnoea and have been determined to be an acceptable first line treatment for these conditions.2 Paediatric dentists in particular can be instrumental in the diagnosis of paediatric sleep apnoea. They often see children on a more regular basis than their physicians and are intimately involved in examination of the mouth and oropharynx where such conditions as tonsillar enlargement have been shown to be a prime cause of apnoea in this age group.3 Oral and maxillofacial surgeons, as a result of their medical and surgical training, have long been involved in the surgical treatment of sleep apnoea and perform a wide variety of surgical procedures for its management. Importantly, all dental providers have the opportunity to screen their patients for sleep apnoea and this should be an integral part of any new patient questionnaire and exam.

Because it has many medical sequelae and is associated with a high degree of morbidity and mortality, dentists must show great care in ensuring proper diagnosis and workup of these patients prior to performing any non-surgical or surgical intervention. Referral to, or involvement of, the patient’s primary care physician, or even better, a practitioner in a sleep-related specialty (usually pulmonology, neurology or psychiatry) who specializes in sleep apnoea, is critical to good comprehensive care. Complete workup should include a variety of blood tests (e.g. thyroid function tests, CBC, etc.), a complete physical examination, an ECG, and diagnostic flexible endoscopic nasopharyngoscopy (this is usually done by an oral and maxillofacial surgeon or otolaryngologist). Nasopharyngoscopy is utilized to both eliminate neoplastic or other lesions as a possible causative factor, and to aid in determining the likely site, or sites, of obstruction. There should also be a multi-channel polysomnogram (PSG). This is an overnight sleep study that documents any respiratory-related sleep disturbances and any cardiac or pulmonary sequela to them. As there are many patients with sleep apnoea that do not have significant symptoms other than snoring, the PSG is the only objective testing mechanism to determine the presence and degree of apnoea present. Since there are now many possible non-surgical and surgical treatment options, the degree of sleep apnoea seen on the PSG, as well as the patient positioning during apnoeic events and the extent of oxygen desaturation, can help the clinician in determining which procedures pose the best risk-to-benefit ratio.

In addition to the laboratory studies, it is imperative to have diagnostic imaging as part of the workup of the apnoeic patient. Imaging provides an insight into the possible sources of the apnoeic events, as well as potential sites for non-surgical and surgical intervention. For many years, the only imaging available were simple two-dimensional (2-D) studies, such as plain films or a lateral cephalometric film. Single plane tomograms, such as panoramic radiographs have also been utilized. While these were useful, the 2-D nature rendered them somewhat limited. In addition, these lateral and anterior-posterior 2-D images were rendered from transmission through a large three-dimensional (3-D) mass, leading to significant overlap of structures and difficulty in interpreting anatomical structures located within the field. The advent of small slice, single plane (e.g. a true midline section) and 3-D imaging has opened up new vistas and the potential for much improved diagnosis and guidance for therapeutic options.

Pathophysiology of sleep apnoea

Sleep apnoea is characterized by intermittent arousal from sleep (although this can occur as many as 100 times per hour) that is associated with a complete or partial lack of breathing. This can lead to excessive daytime somnolence, morning headaches, and loss of mental acuity, as well as a wide variety of serious physiologic abnormalities such as an increased incidence of stroke, heart attacks and sudden death. There are two major forms of sleep apnoea: central sleep apnoea (CSA) and obstructive sleep apnoea (OSA). CSA results from a centrally mediated decrease, or complete lack of, respiratory drive and is manifested by an absence of diaphragmatic and chest wall movement during sleep, i.e. no attempt is actually made to inspire. Although the aetiology of CSA is not well known, it has been associated with a variety of neurologic diseases and intracranial lesions such as Arnold-Chiari malformation, as well as a number of medical conditions including gastro-oesophageal reflux disease (GORD), obesity and hypothyroidism.4 Conversely, obstructive apnoea is associated with normal inspiratory effort against a partially or totally occluded airway. The pathophysiology of OSA is somewhat better understood and is known to be a result of either an anatomical abnormality within the airway leading to occlusion or increased elasticity and compliance of the airway that allows for collapse during inspiration.

During inspiration, negative intraluminal pressure is created through the Bernoulli effect (a rapidly moving gas or liquid creates decreased pressure on surrounding objects – the same principle that explains airplane wings). This negative pressure has the effect of constricting the airway during inspiration. Under normal conditions there is adequate stability created by the neck musculature to maintain at least a moderate airway cross-section and airway integrity. However, when the airway is compromised by an enlarged structure within it, the amount of cross-sectional airway can decrease significantly and limitation of airflow can occur. Depending on the degree of limitation, this may be expressed as snoring (which is essentially a vibration caused by the negative pressure), hypopnoea or complete apnoea. Anatomical structural abnormalities that have been associated with OSA include a variety of neoplasms, nasal septal deformity, a long soft palate, enlarged tonsils and adenoids, hyperplastic pharyngeal tissues, a floppy epiglottis, a large tongue base, mandibular and maxillary hypoplasia, and even vocal cord paralysis.5 Conversely, even a normal structure can lead to obstruction if it demonstrates excessive movement and prolapse into the airway during sleep (when tissues are more relaxed). The increased compliance of these structures allows for collapse when subjected to even normal amounts of negative intraluminal pressure. Thus, the purpose of diagnostic imaging is to attempt to isolate the potential source, or sources, of anatomical obstruction that can lead to OSA, whether this is from an enlarged structure or an area of increased compliance.

2-D evaluation

Using radiographic methods for evaluation of the posterior airspace is an important diagnostic tool to evaluate for OSA. Some of the traditional methods include cephalometric measurements and lateral neck radiographs to assess the skull base, position of the hyoid bone, mandible position and configuration, posterior airspace of the oropharynx, length and width of the uvula and tongue dimensions. Schwab and colleagues reviewed the cephalometric literature and reported that the most common skeletal abnormalities were: (1) mandibular and maxillary deficiency; (2) reduced dimension of the posterior airway space (measured at the base of the tongue); (3) enlarged tongue; (4) enlarged soft palate; and (5) caudally displaced hyoid.6 While CBCT has the primary advantage of being a 3-D imaging modality (Fig. 1), it should also be remembered that it can be used to provide excellent 2-D images as well. Unlike the traditional cephalometric radiograph, which compresses 3-D anatomy into a 2-D image, CBCT provides a true single plane midline image. This essentially eliminates the traditional amount of overlap of head and neck structures that makes identification and measurement of specific anatomic entities difficult. Also, since CBCT data is not magnified and is measured directly on a 1:1 scale, it allows for much more precise measurement when compared to other 2-D modalities where image magnification must be identified and accounted for during analysis (Fig. 2).

Figure 1.

 A 3-D construct showing a patient’s bony landmarks and airway.

Figure 2.

 An axial slice of CBCT image demonstrating linear measurement with 1:1 precise 2-D A-P and lateral view. Some software packages also allow for 1:1 3-D measurements.

Chung et al. compared angular cephalometric values from traditional lateral cephalograms that were digitally scanned versus CBCT-generated images and found no significant difference in the angular measurements between the two modalities.7 Thus, a lateral slice from CBCT imaging can be used to perform traditional cephalometric analysis with levels of precision and accuracy comparable to traditional lateral cephalometrics (Fig. 3). This is not surprising since magnification should have no effect on angular measurements. Lenza et al. showed that the upper airway cannot be accurately expressed by single linear measurements as performed on cephalograms.8 They scanned 34 patients with CBCT and performed a 3-D evaluation of the upper airway. Linear sagittal, transverse and cross-sectional measurements reproducing those usually performed on lateral cephalograms were generated. The final analysis of their study showed a weak correlation between most of the linear measurements except for the lower part of the nasopharynx which was highly correlated with sagittal measurement and with area.8

Figure 3.

 A true midline sagittal slice taken from a CBCT dataset.

Two-dimensional measurements of the skeletal structures have significant correlation to the presence and degree of apnoea. Approximately 58% of apnoea patients present with mandibular micrognathia and retrognathia in relation to the maxilla.9 Under normal conditions, inspiration produces negative intraluminal pressure and there are more than 20 muscles that act as pharyngeal dilators to prevent airway narrowing. Out of the 20 muscles, the genioglossus has been the most investigated structure as it plays an important role in pharyngeal maintenance and dilatation.10 Because the genioglossus originates on the posterior surface of the mandible and inserts into the tongue base, it is a direct determinant of tongue base position and it would be reasonably expected that there would be a reduction of the posterior airway space with a retrognathic mandible. This has been generally measured as the distance between the posterior pharyngeal wall and the base of the tongue, either along the line measured from B point to gonion or at the level of smallest A-P dimension (Fig. 4). Rivlin et al., using cephalometric analysis, demonstrated that OSA patients have smaller mandibular lengths (by a mean of 5.4 +/− 6.6 mm).11 In the same study, he also found that the overall posterior displacement of the mandibular symphysis, which is representative of the skeletal support of the anterior pharyngeal wall and is dependent on both mandibular size and position, was highly significant (6.4 +/− 4.7 mm).

Figure 4.

 Cephalometric analysis of airway: MP-H (mandibular plane to hyoid); PAS (posterior airway space); MPAS (minimal posterior airway space); PNS-P (posterior nasal spine to palate).

This region can be further divided into three different anatomic levels: (1) the upper posterior air space (between the pharyngeal posterior wall and the most distal soft palate); (2) the middle (between the pharyngeal posterior wall and the inferior limit of the uvula); and (3) the inferior (between the pharyngeal posterior wall and the base of tongue). It has also been shown that the length of the soft palate averages 48 mm in individuals with OSA and 35 mm in healthy individuals. From an anatomical standpoint, this marked increase in soft palate length leads to reduction of the nasopharyngeal airspace and increased contact between the soft palate and tongue, leading to collapse of the airway in this region.12

Transverse airspace reduction is also evident among OSA patients, especially at the level of the uvula and the mandibular plane.13 One of the most important findings gathered from cephalometric analysis in OSA patients is that there is a reduction of the velopharyngeal space in 86% of the cases.14 However, 75% of the OSA patients presented with more than one obstructive site.15 In fact it has been shown that, in the majority of cases, the obstructive area actually changes position throughout the course of the night.16 Tangugsorn et al. investigated the cephalometric features of 100 males with OSA and compared them to 36 healthy control subjects and found that the hyoid bone was inferiorly placed at the level of C4–C6 instead of the usual position, at the level of C4.17 They hypothesized that this was due to either a high mandibular plane angle or excessive fat pads in the submental, submandibular and pharyngeal regions. Young et al. evaluated the severity of OSA in relation to the vertical position of the hyoid bone and found that a sella-hyoid distance greater than 120 mm correlated with severe OSA, and less than or equal to 120 mm with mild to moderate OSA.18 Another study conducted by Gulleminault et al. also demonstrated the same findings. The distance between the hyoid bone and the mandibular plane tends to be greater (27.8 mm) in patients with OSA than in healthy people (12 mm).19 It is evident from these multiple studies that the longer the hyoid vertical dimension to any superior anatomical landmark will relate to the severity of the disease. Although cephalometric analyses for obstructive sleep apnoea are useful, the studies and findings from quoted authors suffer from limitations inherent in examining a 3-D object by using a 2-D image: cephalometrics provides no information regarding the transverse relationship of the oropharyngeal region.

3-D evaluation

Three-dimensional evaluation and measurement of the airway have become more common as technological developments in both imaging and computer analysis have advanced and converged during the past few years. These advances have been especially advantageous for the ability to understand and diagnose OSA and its relationship to the airway and craniofacial anatomy. The improved availability of CBCT, 3-D imaging and computer simulation in dentofacial analysis and treatment planning has facilitated the use of these methods not only for the workup of orthognathic surgery patients but for evaluation of the airway for OSA as well. Previously used diagnosis and treatment planning methods that relied on standard cephalometric analsysis for OSA had limitations despite inclusion of the patient’s sleep history, nasopharyngoscopy, polysomnography and conventional imaging data. It would seem obvious that the ability to have cross- sectional and 3-D imaging in addition to standard lateral 2-D cephalometrics would provide greater and more accurate diagnostic and treatment planning data.

A precise anatomic analysis of the airway that could be correlated with the presence and severity of OSA and be easily obtainable would be a very valuable asset for diagnosis and treatment planning. Medical grade CT provides excellent and superior soft tissue contrast when compared to CBCT (especially for soft tissue pathology with or without intravenous contrast), but the ability to get CBCTs done in the office and the inherent ability of the end user to manipulate the data in any plane desired, to do direct measurements, and import the standardized DICOM dataset into readily available and OSA-specific software, makes CBCT the clear choice for the diagnosis and evaluation of the airway for most patients.

Until recently, airway calculations from medical CT or CBCT data required time-consuming manual data segmentation and measurement, the accuracy of which has been questioned. Automatic data segmentation has the ability to provide rapid and reliable airway analysis results (Fig. 5). In addition, while helical CT scans are popular and widely accepted for evaluation of the upper airway, the high radiation dose and longer exposure times may not represent the best risk-to-benefit ratio for some patients. For example, using a measure of the effective absorbed dose, a traditional medical CT exposes the patient to a radiation dose of 124.9 to 528.4 μSv for the mandible and 17.6 to 656.9 μSv for the maxilla (depending on the volume imaged and operational settings of the CT).20 In contrast, the CBCT (NewTom 9000, Verona, Italy) used in the Ogawa et al. study only required 36 μSv to 50.3 μSv to provide a similar amount of information when compared to the traditional CT scan exposure.20 While many CBCT machines are now available, they all provide radiation doses far less than traditional medical grade CT scans. Thus, a CBCT represents a far better risk-to-benefit ratio in the evaluation of the OSA patient.

Figure 5.

 A 3-D construct showing minimal axial area and airway volume.

Ogawa et al. were also able to show that 3-D airway imaging with CBCT in awake patients in the supine position (using the single supine CBCT machine currently available) was able to provide the data points necessary to distinguish OSA cases from non-OSA cases.20 They demonstrated that the A-P dimension of the airway of OSA patients and the cross-sectional area of the narrowest airway slice in the oropharynx were significantly smaller than non-OSA subjects. Other non-invasive 3-D imaging methods such as MRI can also provide excellent evaluation of the airway, but the longer examination time inherent in the technique may result in decreased airway image quality due to motion artefacts. CBCT also has an advantage when compared with MRI that it is accessible and financially affordable.

Using various commercially available automatic or manual segmentation software packages, it is possible to evaluate both the cross-sectional area and the volumetric 3-D representation of the entire airway using a lower radiation method with a rapid scan, which is generally within the range of 10–30 seconds.21 By taking advantage of the short scan time, Osorio et al. were able to perform a Muller’s manoeuvres test on a 45-year-old male in the sitting position where volumetric airway changes were obtained under different amounts of negative inhalation pressure, a process that cannot be provided by traditional CT scan and MRI due to the longer scanning times.22 From their study, CBCT provided far more information than nasopharyngoscopy as not only airway changes were detected, but also changes in the soft tissue surrounding it. In fact, they were able to use image processing software to generate a virtual laryngoscopy reconstruction, resulting in video clips of high quality, similar to fibre optic imaging but without the invasiveness of the actual procedure.22 This is a very valuable tool in that CBCT can be used to not only evaluate airway obstruction but to also predict difficult intubation by using volumetric reconstruction methods.

The most common CBCT measurements used to compare the static morphology of the upper airway between OSA patients and non-OSA patients are the minimum surface area of the oropharyngeal region and the anterior-posterior and lateral dimensions of this area.23 The airway of a patient with OSA is smaller and is narrowed laterally. Using 3-D volumetric software, the upper airway volume was noted to vary mildly between 9.2 cm3 and 11.56 cm3 during expiration and inspiration phases in a normal individual while sleeping. It fluctuated moderately between 3.74 cm3 and 9.91 cm3 in a habitual snorer, and varied greatly between 2.73 cm3 and 16.01 cm3 in an OSA patient.24

Grauer et al. studied and assessed the differences in airway shape and volume among subjects with various facial patterns by using CBCT. The airway was divided into the superior and inferior compartments by a plane perpendicular to the sagittal plane that included the posterior nasal spine and the lower medial border of the first cervical vertebra. In their study, there was a statistically significant correlation between the volume of the inferior component of the airway and the anterior-posterior jaw relationship (i.e. skeletal Class II patients had inferior compartment airway volumes that were smaller), and between airway volume and both the size of the face and gender (larger faces correlated with larger airway volumes). Skeletal Class II patients also often had forward inclination of the airway whereas skeletal Class III patients had a more vertically oriented airway.25

Other authors have correlated OSA with body mass index (BMI) and found a positive relationship between the two factors.18 In addition to BMI, it is also possible that the shape of the airway is a predictor of collapse. With higher BMI patients, the airway shape tends to be more spherical secondary to a decrease in the lateral dimension of the pharynx.26 Several researchers have speculated that the collapse of the airway is caused by tissue hypertrophy; swelling and inflammation of the oropharynx, uvula, and tongue; narrowing of the lateral wall of the airway during inspiration; or mandible retrusion causing tongue relapse.27 On the other hand, Shigeta et al. evaluated the retroglossal airway configuration on an axial slice at the level of the anterior-inferior corner of the second cervical vertebra using CBCT and found no differences in the retroglossal airway area between OSA and non-OSA subjects after adjusting for age, gender and BMI. In addition, the correlation between BMI and age with normalized airway area was not statistically significant.28 Donnelly et al. reported significant differences in the patterns of airway dynamic motion when they compared young OSA patients with subjects without OSA using 3-D imaging.29 CBCT are now being routinely taken from a cohort of patients receiving orthodontic treatments, TMJ surgeries and dental implant evaluations. Even though CBCT does not have the highest resolution compared to traditional CT scans, its low radiation exposure allows utilization on a variety of clinical patients to collect data for more longitudinal studies. As the data is collected and analysed, a clearer picture of the normal and abnormal airway as seen on CBCT will become evident and standardized dataset measurements will emerge.

CBCT evaluation of surgical intervention

One of the major advantages of CBCT is that its relatively low radiation dose and wide availability make it an ideal tool to evaluate the effects of therapeutic interventions. The only current method to do this is postoperative PSG, but that is an expensive and labour-intensive process. CBCT allows the clinician the opportunity to visualize and measure changes in airway size and configuration after both non-surgical and surgical therapy.

A review of the literature regarding the most common site of obstruction during sleep indicates that it is at the level of the oropharynx, with extension to the hypopharynx.27 Pharyngeal airways oriented in more of the anterior-posterior (A-P) than lateral (LAT) dimension in obese patients are associated with greater risk of obstructive sleep apnoea. Airways with lower LAT/A-P ratios demonstrated more compliance.30 These airways showed greater reduction in their residual volume versus total lung capacity and functional residual capacity, suggesting that these airways are less effective at maintaining their patency. Patients with higher degrees of OSA versus patients with only mild snoring and controls exhibited airways with more narrow LAT dimensions. Li and colleagues recently noted no differences in A-P dimension but significant decrease in the LAT dimension of the oropharynx comparing normal controls to those with OSA.31 As mentioned before, a low LAT/A-P ratio of the oropharyngeal airway likely explains the presence of obstructive events which Fairburn and colleagues demonstrated with the geometric enhancement of airways following orthognathic advancement.32 From their study, the A-P and LAT dimensions have shown substantial post-surgical improvement and significant enlargement statistically.

There are many non-surgical and surgical treatments for OSA. All are designed to increase the dimension of the airway. Most Phase I treatments are target-specific, have limited success at higher levels of OSA, and are best used for mild to moderate OSA. Maxillomandibular advancement (MMA), a Phase II therapy, enlarges the entire posterior airway and increases the tension on the suprahyoid and airway musculatures by elevating the tissues and musculatures attached to the maxilla, mandible and hyoid.

Maxillomandibular advancement combined with genial tubercle advancement (GTA) procedures have also been extensively researched and accepted as the single surgery that results in successful outcomes for patients with severe OSA.33 As such, it is generally considered the ‘gold standard’ of treatment against which all other therapy is measured against. Large surgical advancement in patients with OSA is known to result in relatively stable repositioning of the maxilla and mandible over the long term.34 The application of MMA surgery to patients without pre-existing maxillomandibular deficiencies has proven to be highly effective and to result in excellent patient satisfaction.35 According to Riley et al., 10 mm advancement of the maxilla and mandible provides sufficient oropharyngeal opening (Fig. 6a and 6b).36 However, even at slightly less advancement, postoperative CBCT examination revealed that the oropharyngeal volume was doubled in size, and the surface area of minimal axial slice was more than tripled in size compared with the pretreatment records.

Figure 6.

 (a) Preoperative cephalometric measurement for an OSA patient. (b) Postoperative cephalometric measurement after 10 mm maxillomandibular advancement.

It should be noted that airway size and shape are extremely variable depending on head posture and the breathing stage. However, CT comparison of the initial and postoperative airway volumes exhibited a dramatic change in volume after maxillomandibular advancement and genial tubercle advancement surgeries.37 Another interesting finding was that all measured dimensions of the airway increased, especially the LAT dimensions. In a study by Fairburn et al., pre- and postoperative CT evaluations of 20 consecutive patients with OSA treated with maxillomandibular advancement revealed enhancements at all levels in the A-P and LAT dimensions.32 They also stated that LAT dimensions were enhanced greater than the A-P dimensions in the retroglossal region. Abramson et al. evaluated changes in airway size and shape in patients with OSA after MMA and GTA using CT scans and showed a significant increase in lateral and anteroposterior airway diameters, volume, surface area and cross-sectional areas at multiple sites. Airway length decreased and airway shape became more uniform with a decreased mean change in RDI of 60%.37 In a study by Muto et al., a set back of the mandible of 1 cm resulted in a 0.4 mm decrease in the airway in an A-P dimension only.38 Comparing the results of two different case studies conducted by Schendel et al. where the smallest preoperative airway space of one patient went from 6.62 mm2 to 112.39 mm2 postoperatively with 1 cm MMA and 59.5 mm2 to 72 mm2 with 1.2 cm MMA, shows the significance of using 3-D imaging to assess accuracy of actual volumetric change.39,40

Horizontal versus upright CBCT

Most of the studies conducted by different authors regarding modalities to evaluate OSA patients were done using cephalometrics in an upright position, which has a major disadvantage because the measurements do not reflect patients’ natural sleeping position. Traditional supine CT scans are useful in obtaining 3-D volumetric data of the airway; however, patients are at risk of high radiation exposure. The use of multidetector medical CT scanners may mitigate the increase in radiation exposure.

With the advent of CBCT, some of the disadvantages can be eliminated. Ogawa et al. were able to show the utility of diagnosis of anatomy with 3-D airway imaging using CBCT in awake patients in the supine position and showed the characteristics of OSA airway that may contribute to distinguishing OSA cases from non-OSA cases.20 Enciso et al. conducted their study using the Newtom CBCT (Imageworks, New York, USA), which is the only commercially available CBCT that images the patient in supine position.41 They were able to show that the OSA cases had a slightly more spherical airway shape than snorers that was consistent with Mayer et al.’s study,26 a narrower LAT dimension of the airway compared with snorers that is consistent with Hora et al.’s study using MRI in awake patients.42 A study by Pevernagie et al. showed the effects of changes in body position on upper airway size and shape using ‘fast’ CT scan. They were able to demonstrate significant differences when subjects turned from prone to the right side or supine position, tending to decrease upper airway size by decreasing the lateral distance in non-position-dependent OSA patients, while the opposite effect was found in the position-dependent OSA group.43 Schwab et al. studied 15 subjects in the supine position during awake nasal breathing and found that the upper airway was significantly smaller in apnoeic than normal subjects, especially at the retropalatal and retroglossal anatomic level; little airway narrowing occurred in inspiration, suggesting that the action of the upper airway dilator muscles balanced the effects of negative intraluminal pressure.44

It is clear that the use of supine CBCT has advantages over sitting or standing CBCT. However, as there is only one commercially available unit presently on the market, which is more expensive and requires a greater footprint than other machines, it remains to be seen whether the increase in diagnostic accuracy will be considered vital in the future.

Sedated scans

Sedation is a common adjunct to accommodate patients who are not able to tolerate or keep still for diagnostic scanning processes. It is commonly provided in MRI suites for patients that are claustrophobic. Most of the diagnostic imaging to evaluate airway for OSA patients are done while the patient is awake and upright. Although accurate measurements of the airway can be obtained, they do not reflect the true dynamic changes as compared to when the patients are asleep. It is difficult to achieve the natural state of sleep required to evaluate apnoea without induced sedation, given that the patient is in a new environment and the associated intimidation at the imaging centre. Additionally, airway compromise is a major concern in any sedated patient, especially in the OSA population because OSA has been associated with increased perioperative morbidity and mortality.45

Common side effects of sedative medications include respiratory depression, obtunded reflexes and muscle relaxation, all of which can worsen apnoeic events with patients that already have OSA. Kheterpal et al. assessed the difficulty of mask ventilation and intubation of 22 660 patients and found that 2% of the attempts were either inadequate or unstable, required two providers, or impossible to mask ventilate. They concluded that abnormal neck anatomy, sleep apnoea, snoring and BMI of 30 kg/m or more were independent predictors of inadequate mask ventilation and difficult intubation.46 The literature relating to adult patients who have OSA receiving moderate or deep sedation for diagnostic imaging is lacking. It is very valuable to include procedural sedation in evaluating airway obstruction to mimic patient’s natural state of sleep. However, it must be used with strict caution in adults with OSA or a known history of difficult airway.

Medico-legal aspects of CBCT

While the use of CBCT in the diagnosis and management of OSA cannot be questioned, and has arguably led to a paradigm shift in the way we evaluate the airway both pre- and postoperatively, its use has generated significant controversy.47 As with any therapeutic modality, a risk-to-benefit ratio must be calculated and evaluated in the context of any single patient. For example, while CBCT radiation doses are significantly lower than medical-grade CT, they are significantly higher than doses for panoramic or plain radiographs. This has resulted in issues of standard of care and whether the use of a CT is necessary in all cases, and whether a CBCT is adequate (or perhaps better) than a medical-grade CT. A review of the literature would indicate that standard of care is a regional issue and varies from locality to locality. However, traditionally the use of new technology becomes standard of care when its use is validated in the scientific literature and when it becomes affordable to a critical mass of practitioners in the community (as opposed to larger academic centres only). As is the case with dental implants, both of these are rapidly approaching for OSA and the use of CT (and CBCT in particular due to its low radiation dose, easy manipulation by the end user and increasing availability) may well become necessary in the majority of cases.

Another area of great controversy is interpretation of the CBCT study. Since many CBCT machines are now located in individual practitioner’s offices rather than in hospital radiology suites or outpatient radiology facilities, the question of who is reading the study becomes important. It is obvious that, as with any radiographic modality, the usefulness of the information is more related to the diagnostic acumen of the clinician interpreting the study than to the mechanics of the study itself. Previous radiographic imaging in dental offices, such as panoramic and periapical radiographs, were limited to anatomy very familiar to most dental practitioners. Thus, the interpretation of normal vs. pathology was relatively simple. It must be remembered that CBCT datasets represent (more or less) a full 360 degree volume, which is then reformatted, recalculated, and cropped or truncated to represent isolated cross-sectional, axial and 3-D images. Regardless of what the clinician may wish to focus on, the dataset always includes the entire volume. Many of the currently available CBCT machines available to dental practitioners have a field of view (FOV) that can include the paranasal sinuses, pharyngeal air spaces, skull base, cervical spine and upper neck in addition to the oral and perioral structures.48 Therefore, it is incumbent on the clinician to decide whether they are fluent in interpreting the data from the entire volume and not just their anatomical area of interest (Fig. 7). It is now generally accepted that the standard of care is that the entire volume of data must be evaluated and reported.49 To fill this gap, many medical and dental radiologists have made themselves available to interpret CBCT studies taken by non-radiologist practitioners. This is accomplished by sending a copy of the study on digital media (usually a CD or DVD-ROM) or it can be done online. In addition, some machines now allow multiple different FOVs so that the clinician can generate a volume dataset that only includes the necessary areas that are within the ability of the clinician to evaluate adequately.

Figure 7.

 A finding of a left lateral septal mass (red arrows) taken from CBCT initially taken for evaluation of OSA.

Conclusions

The use of CBCT imaging for diagnosis and treatment planning in the management of OSA is gaining in popularity as the machines become more available to the practitioner. There are many advantages to its use, the primary ones being the ability to look at 3-D volumes rather than the standard 2-D images used to date and the ability of the clinician to manipulate the data. However, the normal values for these volumetric datasets in both the OSA and normal populations is still being determined. As these data are measured, the use of CBCT for both preoperative planning and for postoperative evaluation of therapeutic interventions will become increasingly important.

Ancillary