PCCT improved CVF detection sensitivity without a significant specificity penalty. Across all three readers, sensitivity was significantly higher for PCCT myelography than for EID-CT myelography, while specificity did not differ significantly.
The authors attribute the diagnostic gain mainly to much higher spatial resolution. PCCT was acquired in UHR mode with 0.2 mm slices versus 0.625 mm on EID-CT, and readers rated nerve root sleeve discernibility and overall image quality higher on PCCT.
Higher image noise did not prevent better clinical performance. Although both objective and subjective noise were higher on PCCT, the authors concluded that the spatial-resolution advantage outweighed this tradeoff for detecting subtle CVFs.
You will find an updated version of this protocol/publication in our community database at:
Updated CSF protocol from 2025
Figure 2. 59-year-old woman with symptoms of SIH for 1.5 years, with prior temporary symptom relief after fibrin blood patch. (A,B) Images from EID-CT myelogram, performed to assess for CSF leak. (A) Axial image with slice thickness of 0.625 mm. (B) Coronal maximum intensity projection (MIP) image with slice thickness of 6 mm. Nerve root sleeve diverticulum is visualized at right T12 level (arrow, A and B). (C,D) Images from PCD-CT myelogram performed 5 months later; during this interval, patient received no interventions, and symptoms remained similar. (C) Axial image with slice thickness of 0.2 mm. (D) Coronal MIP image with slice thickness of 6 mm. Nerve root sleeve diverticulum is visualized at right T12 level (white arrow, C and D). Course of CVF is also visualized at this level (red arrows, C and D). All images for both examinations were acquired in right lateral decubitus position and are presented in figure with window center of 460 and window width of 1530. CVF was classified as definite based on study criteria. CVF was detected by one reader by EID-CT myelogram and by all three readers by PCD-CT myelogram. Coronal MIP images are shown for illustrative purposes and were not assessed by readers. After PCD-CT myelogram, patient was treated by embolization, leading to symptom relief. SIH = spontaneous intracranial hypotension; EID = energy-integrating detector; PCD = photon-counting detector; CVF = CSF-venous fistula
Source: Schwartz et al., AJR American Journal of Roentgenology, 2024.
Protocol
Detail | Protocol |
|---|
Scanner | NAEOTOM Alpha |
Scan area | Whole spine CT myelography |
Scan mode | Quantum HD |
Scan direction | Right lateral decubitus, left lateral decubitus, and prone acquisitions |
Tube voltage [kV] | 120 |
Effective mAs [mAs] | n.r. |
IQ level | 80 (CARE keV) |
Dose modulation | CARE Dose4D, CARE keV |
CTDIvol [mGy] | 20.1 (mean) |
DLP [mGy*cm] | 1213.87 (mean) |
Rotation time [s] | |
Pitch | |
Slice collimation [mm] | 120 × 0.2 |
Slice width [mm] | 0.2 |
Reconstruction kernel | Br48 |
Iterative reconstruction | QIR 4 |
Contrast agent | Isovue-M 300 |
Contrast volume | 10 mL intrathecal |
Flow rate | n.r. (fixed injection duration ~8 s) |
Saline chaser | 40 mL |
Acquisition phase | Pulmonary arterial (test bolus + peak + 7 s) |
Schwartz FR, Kranz PG, Malinzak MD, et al. Myelography Using Energy-Integrating Detector CT Versus PCCT for Detection of CSF-Venous Fistulas in Patients With Spontaneous Intracranial Hypotension. AJR Am J Roentgenol. 2024;222(4):e2330673. doi:10.2214/AJR.23.30673
