Search

Background and Purpose The diagnostic work-up in patients with spontaneous intracranial hypotension (SIH) and spinal longitudinal extradural CSF collection (SLEC) on magnetic resonance imaging (MRI) comprises dynamic digital subtraction myelography (dDSM) in prone position for leak detection. Dynamic computed tomography (CT) myelography (dCT-M) in prone position follows if the leak is not unequivocally located. A drawback of dCT‑M is a high radiation dose. This study evaluates the diagnostic needs of dCT-M examinations and measures to reduce radiation doses. Methods Frequency, leak sites, length and number of spiral acquisitions, DLP and effective doses of dCT‑M were retrospectively recorded in patients with ventral dural tears. Results Of 42 patients with ventral dural tears, 8 underwent 11 dCT‑M when the leak was not unequivocally shown on digital subtraction myelography. The median number of spiral acquisitions was 4 (range 3–7) and the mean effective radiation dose 30.6 mSv (range 13.1–62.16 mSv) mSv. Five of eight leaks were located in the upper thoracic spine (range C7/Th1–Th2/3). Bolus tracking of intrathecal contrast agent in dCT‑M was used to limit the number and length of spiral acquisitions. Discussion A dCT‑M in prone position to localize a ventral dural tear is needed in every fifth patient with a SLEC on MRI. It is typically needed when the leak is located in the upper thoracic spine and when patients have broad shoulders. Measures to reduce the radiation dose include bolus tracking or to repeat the DSM with adjusted positioning of patient.

Background and Purpose To develop a fully automatic algorithm for the magnetic resonance imaging (MRI) identification of patients with spontaneous intracranial hypotension (SIH). Material and Methods A support vector machine (SVM) was trained with structured reports of 140 patients with clinically suspected SIH. Venous sinuses and basal cisterns were segmented on contrast-enhanced T1-weighted MPRAGE (Magnetization Prepared-Rapid Gradient Echo) sequences using a convolutional neural network (CNN). For the segmented sinuses and cisterns, 56 radiomic features were extracted, which served as input data for the SVM. The algorithm was validated with an independent cohort of 34 patients with proven cerebrospinal fluid (CSF) leaks and 27 patients who had MPRAGE scans for unrelated reasons. Results The venous sinuses and the suprasellar cistern had the best discriminative power to separate SIH and non-SIH patients. On a combined score with 2 points, mean SVM score was 1.41 (±0.60) for the SIH and 0.30 (±0.53) for the non-SIH patients (p Conclusion A fully automatic algorithm analyzing a single MRI sequence separates SIH and non-SIH patients with a high diagnostic accuracy. It may help to consider the need of invasive diagnostics and transfer to a SIH center.