Objective: To evaluate and compare the prognostic utility of the Marshall and Rotterdam computed tomography (CT) scoring systems for predicting mortality in patients with traumatic brain injury (TBI).
Methods: A systematic review was conducted in accordance with PRISMA guidelines. The protocol was registered with PROSPERO (registration number: CRD42024610218). Studies published between 2012 and 2024 involving hospitalized TBI patients that reported both Marshall and Rotterdam CT scores from initial head CT imaging were included. Extracted data included patient demographics, injury severity, mortality outcomes, and measures of predictive accuracy, including area under the receiver operating characteristic curve (AUC).
Results: Twenty-two studies met inclusion criteria, including prospective and retrospective designs. Several studies identified a threshold score of ≥ 4 for both scales that was associated with significantly higher mortality. Across studies, both scoring systems demonstrated statistically significant discrimination for mortality prediction. Reported AUC values ranged from 0.58 to 0.92 for the Marshall score and from 0.63 to 0.93 for the Rotterdam score.
Conclusions: Both Marshall and Rotterdam CT scoring systems effectively differentiate survivors from non-survivors following TBI. While Rotterdam occasionally demonstrated marginally higher AUC values, overall prognostic performance was similar between the two systems.
Purpose: Robot-assisted, monoplane C-arm angiography systems designed for hybrid operating rooms represent an alternative approach to neuroendovascular procedures. Conversely, non-computer-assisted monoplane systems or conventional biplane systems in angiography suites are widely established. This study aims to evaluate the effectiveness, safety and efficiency of a robotic C-arm angiography system in performing neuroendovascular interventions, including cerebrovascular digital subtraction angiography (DSA), embolization of the middle meningeal artery (MMA) in subdural hematoma, and mechanical thrombectomy for vessel occlusion.
Methods: All patients undergoing DSA, MMA embolization, or mechanical thrombectomy between July 2020 and December 2024 were retrospectively included. Procedures were performed using a monoplane robotic C-arm system (ARTIS pheno, Siemens Healthineers, Munich, Germany) in a hybrid operating room. Clinical data, procedural details, and imaging outcomes were analyzed. Radiation exposure was assessed by fluoroscopy time, air kerma, and dose-area product (DAP).
Results: A total of 49 procedures were analyzed, including 28 DSAs, 6 MMA embolizations, and 15 mechanical thrombectomies. DSA and MMA embolization (EMMA grade ≥2) achieved 100% procedural success, while mechanical thrombectomies achieved successful reperfusion (mTICI ≥2b) in 93.3% of cases. Median procedure durations were 34.0 (IQR 18.0-45.0) minutes for DSA, 70.0 (IQR 28.0-126.0) minutes for MMA embolization, and 84.0 (IQR 67.0-106.0) minutes for mechanical thrombectomy. Median fluoroscopy times were 5.2 (IQR 2.9-11.5) minutes (DSA), 21.3 (IQR 8.8-36.5) minutes (MMA embolization), and 21.2 (IQR 18.5-42.9) minutes (mechanical thrombectomy). Median DAPs were 7262.5 (IQR 3867.8-11570.8) µGy·m² (DSA), 16135.5 (IQR 8244.2-18216.2) µGy·m² (MMA embolization), and 9875.2 (IQR 6524.3-18455.5) µGy·m² (mechanical thrombectomy). Additional 3D-angiography or cone-beam CT (CBCT) was associated with higher radiation exposure.
Conclusion: Basic neuroendovascular procedures can be safely and efficiently performed using a monoplane robotic C-arm in a hybrid operating room, achieving procedural success and radiation exposure levels comparable to conventional biplane systems.

