Oral squamous cell carcinoma (OSCC) of the tongue is the most common type of oral cavity cancer, and tumor depth of invasion (DOI) is an important prognostic factor. In this study, we investigated the accuracy of intraoral ultrasound and magnetic resonance imaging (MRI) for assessing DOI in patients with OSCC. Histopathological measurement of DOI was used as a reference standard. We conducted a prospective study including patients planned for surgical treatment of OSCC in the tongue. The DOI was measured in an outpatient setting by intraoral ultrasound and MRI, and was compared to the histopathological DOI measurements. Bland–Altman analysis compared the mean difference and 95% limits of agreement (LOA) for ultrasound and MRI, and the Wilcoxon signed-rank test was used to test for significance. The correlation was evaluated using Pearson’s correlation coefficient. We included 30 patients: 26 with T1 or T2 tumors, and 4 with T3 tumors. The mean difference from histopathology DOI was significantly lower for ultrasound compared to MRI (0.95 mm [95% LOA −4.15 mm to 6.06 mm] vs. 1.90 mm [95% LOA −9.02 mm and 12.81 mm], p = 0.023). Ultrasound also led to significantly more correct T-stage classifications in 86.7% (26) of patients compared to 56.7% (17) for MRI, p = 0.015. The Pearson correlation between MRI and histopathology was 0.57 (p < 0.001) and the correlation between ultrasound and histopathology was 0.86 (p < 0.001). This prospective study found that intraoral ultrasound is more accurate than MRI in assessing DOI and for the T-staging of oral tongue cancers. Clinical practice and guidelines should implement intraoral ultrasound accordingly.

Our aim was to investigate whether dynamic contrast-enhanced ultrasound (DCE-US) can detect early changes in perfusion of colorectal liver metastases after initiation of chemotherapy. Newly diagnosed patients with colorectal cancer with liver metastases were enrolled in this explorative prospective study. Patients were treated with capecitabine or 5-fluorouracil-based chemotherapy with or without bevacizumab. DCE-US was performed before therapy (baseline) and again 10 days after initiation of treatment. Change in contrast-enhancement in one liver metastasis (indicator lesion) was measured. Treatment response was evaluated with a computed tomography (CT) scan after three cycles of treatment and the initially observed DCE-US change of the indicator lesion was related to the observed CT response. Eighteen patients were included. Six did not complete three series of chemotherapy and the evaluation CT scan, leaving twelve patients for analysis. Early changes in perfusion parameters using DCE-US did not correlate well with subsequent CT changes. A subgroup analysis of eight patients receiving bevacizumab, however, demonstrated a statistically significant correlation (p = 0.045) between early changes in perfusion measures of peak enhancement at DCE-US and tumor shrinkage at CT scan. The study indicates that early changes in DCE-US perfusion measures may predict subsequent treatment response of colorectal liver metastases in patients receiving bevacizumab.

Background and objectives The precise site of dural leaks in patients with spontaneous intracranial hypotension (SIH) is challenging to identify. We investigated for the first time the ability of the dural displacement sign seen on 3D T2 FIESTA (Fast Imaging Employing Steady-state Acquisition) weighted MR imaging to depict epidural collections and show the leakage site in type 1 leaks in patients with SIH. Materials and methods In a retrospective study, 23 patients with confirmed type 1 leakage received an MR of the spine with 3D T2 FIESTA-weighted imaging and ultrafast dynamic CT myelography. The two modalities were compared in 22 patients in relation to the leakage site. In one patient, conventional CT myelography (CTM) was used at our site, but an ultrafast dynamic CT myelography was later conducted at another site and confirmed the leakage location. Two neuroradiologists performed the evaluations while blinded. Results Dural displacement sign in the upper thoracic spine was identified in 12 of 13 patients (92%) in the upper thoracic spine, whereas it was not identified in the remaining 10 patients with leaks in the cervical, lower thoracic or lumbar regions. A Type 1 leakage in the upper thoracic region was at the exact same level as the dural displacement sign in 9 (72%) patients and in the remaining 3 the leakage level was located at one level below the dural displacement sign. In one of the 3 patients where the leakage was located at the level below the visible dural displacement sign, a dural tear at the leakage level in the form of discontinuity of the dura was identified. Conclusions 3D FIESTA sequences and the dural displacement sign show the potential site of type 1 leakage in the upper thoracic spine. We find that the 3D FIESTA sequence and the dural displacement sign can be valuable when planning ultrafast dynamic CT myelography. This may reduce the investigation time and the radiation doses.