RPS 1408 - Maxillofacial and sinonasal imaging

Purpose:

Computed tomography (CT) or cone-beam computed tomography (CBCT) are the gold standards in clinical traumatology imaging when structures of the head and neck region are involved. However, CT and CBCT harbour inherent limitations in regards to soft tissue visualization. In mandible fractures, the inferior alveolar nerve (IAN) is at particular risk for damage. The aim of this study was to assess the IAN pre- and postoperatively and correlate elevated signal-to-noise-ratios (SNR) with potential sensitivity impairment.

Methods and materials:

18 patients suffering from a mandible fracture were examined on a 3T scanner. The conducted sequence protocol consisted of a 3D STIR, 3D WATS, and a 3D T1 FFE "black bone"-sequence (15 min scan time). Nerve function was assessed as normal, hypesthesia, anesthesia, or allodynia.

Results:

The 3D T1 TFE sequences allowed for fracture detection in all patients. The 3D STIR sequence showed the highest SNR with regard to potential nerve damage (p<0.05). Besides some diagnostic accuracy in fracture detection like with CT, the direct visualisation of nerve structures was possible. Positive clinical nerve testing correlated with physiological SNR and hypoesthesia correlated with increased SNR (p<0.05).

Conclusion:

The study highlights the feasibility of MRI for fracture detection and affection of the IAN. The applied sequence protocol showed comparable diagnostic power in detecting mandible fractures as with CT. Additionally, a reliable assessment of potential IAN damage with a good correlation to clinical sensory testing could be achieved.

Applying the described sequence protocol allowed for precise fracture detection as well as reliable nerve damage assessment.

Limitations:

Motion artefacts due to a scan time of approximately 10 min.

Ethics committee approval

Ethics committee approval obtained.

Funding:

No funding was received for this work.

Purpose:

Depth of invasion (DOI) has been introduced into the latest TNM classification of oral squamous cell carcinoma(OSCC) as a required parameter to correctly define T staging and prognosis. Despite its primarily pathologic definition (pDOI), there is increasing evidence that radiological DOI (rDOI) can preoperatively predict its final value even though a standard and practical definition is lacking. We wanted to assess the clinical impact and utility of computed tomography-derived (CT) rDOI in a cohort of OSCC patients and to better define its measurement.

Methods and materials:

We retrieved 58 cases of OSCC operated on at our institution from 2016-2019. After accounting for plane-specific shrinkage factors and different oral subsites, we compared pDOI and rDOI for each spatial plane by a paired difference test and correlation coefficient. Radiological accuracy and survival analysis were also determined to identify the rDOI’s clinical value.

Results:

For the lateral tongue, pDOI was more strongly related with axial rDOI (ρ=0.923, P<0.01). For the hard palate, the best plane was the sagittal one (P<0.01). In floor-of-mouth (FOM) lesions and cheek buccal mucosa, the strongest correlation was with coronal rDOI (P<0.01). Sagittal scans seem to be the best to evaluate the dorsum of the tongue and retromolar trigone. Gengiva (P<0.01) was more correctly evaluated in the coronal plane. The overall accuracy of rDOI restaging was 75.41%. Disease-free survival seems to be worse as rDOI increases, although differences are not statistically significant (log-rank, P=0.204).

Conclusion:

CT-derived rDOI can be of great value in the preoperative assessment of OSCC. We have identified specific spatial planes where its measurement is more strongly related to final pathological report and we hope that a standardisation of imaging and pathological analysis is needed.

Limitations:

n/a

Ethics committee approval

Informed consent obtained.

Funding:

No funding was received for this work.

Purpose:

Magnetic resonance imaging (MRI) is the main imaging modality in the diagnosis of temporomandibular disorders. Furthermore, sagittal cuts are the chief plane in diagnosing these disorders. This study aimed to investigate the capability of parasagittal cuts taken parallel to the lateral pterygoid muscle in depicting the structures of the temporomandibular joint.

Methods and materials:

20 symptomatic temporomandibular joint (TMJ) patients performed MRI bilaterally using traditionally used parasagittal cuts and another plane parallel to the lateral pterygoid muscle. Four-point scale was used to assess the clarity of articular disc position, disc morphology, cortical bone, and overall quality of the images of both investigated techniques in closed and open positions. Moreover, the final diagnosis of the cases was obtained in each technique.

Results:

The mean and standard deviation of muscular aligned cuts in the clarity of revealing disc position, disc position, cortical bone, and overall image quality in closed position were 2.83± 0.39, 2.63± 0.49, 2.53± 0.51, and 2.83± 0.39, respectively, while those for traditional plane were 2.55± 0.50, 2.43± 0.50, 2.80± 0.41, and 2.53± 0.51, with the same order. In the open position, the same investigated points got 2.80± 0.41, 2.73± 0.45, 2.73± 0.45, and 2.80± 0.41 in muscular aligned images, and 2.78± 0.42, 2.65± 0.48, 2.82± 0.39, and 2.78± 0.42 in the traditional plane. Both techniques provided the same diagnosis in 97.5% of cases.

Conclusion:

Parasagittal magnetic resonance images taken parallel to the lateral pterygoid muscle have higher image quality than a traditional plane in revealing temporomandibular structures.

Limitations:

Small number of cases (40 temporomandibular joints).

Ethics committee approval

The study was approved by the research ethics committee of Faculty of Dentistry, Cairo University (19-4-18).

Funding:

No funding was received for this work.

Purpose:

To calculate the accuracy and reliability of a DSR system when in vitro changes were made on anterior maxillary incisor teeth, which simulate post-traumatic complications (e.g. continued root development and pulp canal obliteration).

Methods and materials:

Subtle RC density changes were made through filing 15 teeth which were then placed in customised acrylic sockets in a skull. Baseline and subsequent digital radiographs were taken after the use of 40, 60, and 90 K file size, and after no filing. During the image acquisition stage, standardised conditions, in terms of image geometry and density, were achieved by using a cephalostat method of head stabilisation (cephalostat headset and a customized film holder) and software for image processing.
DSR was applied and the subtraction images represented small, medium, large, or no RC density change (gain or loss).
Radiographic assessment slides were done by 3 observers. Observer A assessed the slides at two different times.

Results:

DSR sensitivity (0.93-0.96 at a possible cut-off point, 0.78-0.82 at a probable cut-off point, and 0.44-0.62 at a definite cut-off point), specificity (perfect: 1.00), Youden’s J index (range between 0.44 and 0.96 at all the cut-off points), intraexaminer agreement (wk: 0.96, 95% CI: 0.93-0.99) and interexaminer agreement (wk range: 0.90-0.95, 95% CI: 0.83-0.98) were calculated. The results were analysed in relation to the site, size, and kind (gain or loss) of the density change.

Conclusion:

DSR led to a more accurate and reproducible diagnosis. Highly accurate dignoses of small lesions at a low degree of confidence (sensitivity: 0.80-0.87) proved DSR to be a diagnostically useful tool, particularly when early detection of tooth density change is needed.

Limitations:

n/a

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To evaluate the usefulness of radiological density measurements of mandibular condyle (MC) and axis vertebra (AV) in patients with a risk of osteoporosis according to mental index (MI) values.

Methods and materials:

Cone-beam computed tomography (CBCT) images of 96 patients were retrospectively analysed. Radiological density values of AV and MC (left and right), and measurements of MI (left and right), were performed from the areas formed by reference lines in the coronal section. Demographic data of patients such as age and gender were recorded. Patients were divided into two groups according to the mean age and into two categories as high and low risk according to the risk of osteoporosis by taking MI values ​​(cut-off value: 3.0 mm) into the consideration. ROC curve analysis was used to determine the power of MC and AV’s radiological densities to explain the risk of osteoporosis.

Results:

There was a statistically significant difference between gender and MI values, and age groups and MC, AV, and MI values. In addition, MC and AV mean values ​​were significantly lower in patients with a high risk of osteoporosis than in low-risk patients. There was a positive correlation between MC, AV, and MI variables in different strengths. Finally, the results of the ROC curve analysis showed that MC was more powerful with a lower difference than AV in determining the risk of osteoporosis.

Conclusion:

MC and AV measurements are considered to be important predictors in determining the osteoporotic risk.

Limitations:

The cut-off value of MI was considered as 3.0 mm, however, this value might be related to race or ethnicity.

Ethics committee approval

Ethics committee approval obtained.

Funding:

No funding was received for this work.

Purpose:

To specify normal and pathologic values and intervals of main numeric CT signs of lacrimal sac dilatation to compare the diagnostic value of those CT signs and to compare the diagnostic value of virtual endoscopy (VE) and conventional text reports for LS position description for intraoperative navigation purpose.

Methods and materials:

102 adult patients with dacryocystitis were included in this study. 204 CT volumes including lacrimal sacs (LS) and nasolacrimal ducts (NLD) images were obtained.

The main group included CT images of 123 LS and NLD on the affected side.

The control group included CT images of 81 LS and NLD of the same patients on the intact side, excluding patients with bilateral dacryocystitis.

Pearson's Chi-square test and receiver operating characteristic curve (ROC) analysis were applied to the collected data.

Results:

CT signs of normal LS volume were specified: volume ≤0,29±0.1 cm³ (sensitivity 73.17%, specificity 90.12%), height ≤13.9±1.3 mm (sensitivity 68.29%, specificity 74.07%), width ≤5.9±0.2 mm (sensitivity 65.85%, specificity 91.36%),
and depth ≤6.55±0.4 mm (sensitivity 89.43%, specificity 74.07%).

The diagnostic value of CT signs of LS dilatation was found regressing in the following sequence: volume (area under the curve, AUC=0.886), depth (AUC=0.865), width (AUC=0.841), height (AUC=0.734), upper dislocation (AUC=0.586), and forward dislocation (AUC=0.564).

Significant differences in the diagnostic accuracy of text reports and VE were found: χ^₂=52.61 (P<0.0001) and AUC difference 0.418±0.03 (VE AUC=0.992±0.01, text reports AUC=0,574±0,1).

Conclusion:

LS volume is the most accurate CT sign of LS dilatation in patients with dacryocystitis. VE is better for LS position description than text reports and thus VE is recommended for routine clinical use.

Limitations:

The numeric criteria of LS dilatation in children and other races will probably be different.

Ethics committee approval

n/a

Funding:

No funding was received for this work.

Purpose:

To test a new CT method for the internal nasal valve (INV) angle measurement and compare it with current techniques and endoscopy.

Methods and materials:

40 INV angles in 20 adult patients (mean age 38.25±10.9) were measured on CT images using a "nasal base view (NBV)" method and a new method and compared by Student’s t-test. All patients were candidates for functional endoscopic sinus surgery (FESS) and preoperative CT examination. The exclusion criteria were a history of external nose trauma, septoplasty, esthetic nasal reconstructions, and diseases that may distort the INV angle. Angles were also measured on endoscopic images and correlated statistically with CT techniques. Measurements were made by two radiologists independently, and the interclass correlation coefficient (ICC) was calculated.

Results:

INV angles were not significantly different between NBV and the new method (t=-1.256, p=0.216), although the new method provided slightly higher angles (NBV 11.19 ± 3.62 vs. the new method 11.86 ± 3.90). Both methods correlated significantly with angles measured on endoscopy images, but a much stronger correlation was detected with the new CT method (R=0.387, p=0.014 vs. R=0.808, p<0.001). The new CT method showed a stronger level of agreement (ICC=0.930) between radiologists in contrast to the NBV method (ICC=0.658).

Conclusion:

A stronger correlation with nasal endoscopy and a better reproducibility gives priority to the new CT method for the INV angle measurement in daily radiological practice.

Limitations:

A relatively small number of patients might limit the study. The INV narrowing was not diagnosed in all patients who participated in the study.

Ethics committee approval

The study was approved by the Ethics Committee of the Faculty of Medicine, No. 29/V-1.

Funding:

No funding was received for this work.

Purpose:

Chronic rhinosinusitis (CRS) contributes to the disease burden of patients with cystic fibrosis (CF). However, its onset and progression in infants and preschool children with CF remain poorly understood. Our aim was to determine the prevalence and extent of CRS in early CF using magnetic resonance imaging (MRI).

Methods and materials:

MRI was performed in 67 infants and pre-school children with CF (mean age 2.3±2.1y, range 0-6y), and 30 non-CF controls (3.5±2.0y). Paranasal sinus dimensions and structural abnormalities including mucosal swelling, mucopyoceles, and nasal polyps of the maxillary, frontal, sphenoid, and ethmoid sinuses, and in addition medial maxillary sinus wall deformation, were assessed using a novel CRS MRI scoring system.

Results:

Pneumatisation and dimensions of paranasal sinuses did not differ between the two groups. MRI detected an increased prevalence of mucosal swelling (83% vs 17%, P<0.001), mucopyoceles (75% vs 2%, P<0.001), polyps (26% vs 7%, P<0.001), and maxillary sinus wall deformation (68% vs 2%, P<0.001) in CF compared to controls. Furthermore, the extent of these abnormalities was also increased with an MRI sum score of 22.9±10.9 in CF compared to 4.5±7.6 in non-CF controls (P<0.001).

Conclusion:

MRI detected normal dimensions of paranasal sinuses and a high prevalence and severity of paranasal sinus abnormalities due to CRS in infants and preschool children with CF without radiation exposure. Our results support the development of MRI for sensitive non-invasive diagnosis and the monitoring of CRS in young children with CF, and as an outcome measure for clinical trials.

Limitations:

We did not assess the relationship between structural abnormalities and disease burden obtained by a questionnaire.

Ethics committee approval

Approved by the institutional ethics committee and informed written consent was obtained from parents or legal guardians.

Funding:

Supported by the German BMBF, the Mukoviszidose e.V., and the Einstein Foundation.