Background: Highly accurate detection of intracranial hemorrhages (ICH) on head computed tomography (HCT) scans can prove challenging at high-volume centers. This study aimed to determine the number of additional ICHs detected by an artificial intelligence (AI) algorithm and to evaluate reasons for erroneous results at a level I trauma center with teleradiology services. Methods: In a retrospective multi-center cohort study, consecutive emergency non-contrast HCT scans were analyzed by a commercially available ICH detection software (AIDOC, Tel Aviv, Israel). Discrepancies between AI analysis and initial radiology report (RR) were reviewed by a blinded neuroradiologist to determine the number of additional ICHs detected and evaluate reasons leading to errors. Results: 4946 HCT (05/2020-09/2020) from 18 hospitals were included in the analysis. 205 reports (4.1%) were classified as hemorrhages by both radiology report and AI. Out of a total of 162 (3.3%) discrepant reports, 62 were confirmed as hemorrhages by the reference neuroradiologist. 33 ICHs were identified exclusively via RRs. The AI algorithm detected an additional 29 instances of ICH, missed 12.4% of ICH and overcalled 1.9%; RRs missed 10.9% of ICHs and overcalled 0.2%. Many of the ICHs missed by the AI algorithm were located in the subarachnoid space (42.4%) and under the calvaria (48.5%). 85% of ICHs missed by RRs occurred outside of regular working-hours. Calcifications (39.3%), beam-hardening artifacts (18%), tumors (15.7%), and blood vessels (7.9%) were the most common reasons for AI overcalls. ICH size, image quality, and primary examiner experience were not found to be significantly associated with likelihood of incorrect AI results. Conclusion: Complementing human expertise with AI resulted in a 12.2% increase in ICH detection. The AI algorithm overcalled 1.9% HCT. Trial registration: German Clinical Trials Register (DRKS-ID: DRKS00023593).

In hydrocephalus treatment, ventriculo-peritoneal shunts (VPS) have become the most relevant therapy for seven decades among other treatment options. Due to the hydrostatic pressure in vertical pos...

Background: Chronic subdural hematoma (cSDH) is the most common complication of mild traumatic brain injury demanding neurosurgery in high-income countries. If undetected and untreated, cSDH may increase intracranial pressure and cause neurological deficiencies. The first-line intervention of choice is burr hole trepanation and hematoma evacuation. However, any third patient may experience rebleeding, demanding craniotomy with excess morbidity. Adjunct endovascular embolization of the frontal and parietal branches of the middle meningeal artery (MMA) is a promising approach to avoid relapse and revision but was hitherto not studied in a randomized trial. Methods: MEMBRANE is an investigator-initiated, single-center, randomized controlled trial. Male, female, and diverse patients older than 18 years scheduled for surgical evacuation of a first cSDH will be assigned in a 1:1 fashion by block randomization to the intervention (surgery plus endovascular MMA embolization) or the control group (surgery alone). The primary trial endpoint is cSDH recurrence within 3 months of follow-up after surgery. Secondary endpoints comprise neurological deficits assessed by the modified Rankin Scale (mRS) and recurrence- or intervention-associated complications during 3 months of follow-up. Assuming a risk difference of 20% of rebleeding and surgical revision, a power of 80%, and a drop-out rate of 10%, 154 patients will be enrolled onto this trial, employing an adaptive O’Brien-Fleming approach with a planned interim analysis halfway. Discussion: The MEMBRANE trial will provide first clinical experimental evidence on the effectiveness of endovascular embolization of the MMA as an adjunct to surgery to reduce the risk of recurrence after the evacuation of cSDH. Trial registration: German Clinical Trials Registry (Deutsches Register Klinischer Studien [DRKS]) DRKS00020465. Registered on 18 Nov 2021. ClinicalTrials.gov NCT05327933. Registered on 13 Apr 2022.

Modern ventriculoperitoneal shunts (VPS) are programmable, which enables clinicians to adjust valve-pressure according to their patients' individual needs. The aim of this retrospective analysis is to evaluate indications for valve-pressure adjustments in idiopathic normal pressure hydrocephalus (iNPH). Patients operated between 2004 and 2011 diagnosed with iNPH were included. Kiefer-Scale was used to classify each patient. Follow-up exams were conducted 3, 6, and 12 months after shunt implantation and yearly thereafter. Initial valve-pressure was 100 or 70 mmH2O. Planned reductions of the valve-pressure to 70 and 50 mmH2O, respectively, were carried out and reactive adjustment of the valve-pressure to avoid over- and under-drainage were indicated. A total of 52 patients were provided with a Medos-Hakim valve(Codman®) with a Miethke shunt-assistant(Aesculap®) and 111 patients with a Miethke-proGAV(Aesculap®). 180 reductions of the valve-pressure took place (65% reactive, 35% planned). Most patients (89%) needed one or two adjustments of their valve-pressures for optimal results. In 41%, an improvement of the symptoms was observed. Gait disorder was improved most often after valve-pressure adjustments (32%). 18 times an elevation of valve-pressure was necessary because of headaches, vertigo, or the development of subdural hygroma. Optimal valve-pressure for most patients was around 50 mmH2O (36%). The goal of shunt therapy in iNPH should usually be valve-pressure settings between 30 and 70 mmH2O. Reactive adjustments of the valve-pressure are useful for therapy of over- and underdrainage symptoms. Planned reductions of the valve opening pressure are effective even if postoperative results are already satisfactory.

Background: Cone beam CT (CBCT) is a widely available technique with possible indications in carpal ligament injuries. The accuracy of CBCT arthrography in diagnosing traumatic tears of the scapholunate ligament has not been reported. Questions/purposes: (1) What is the diagnostic accuracy of CBCT and how does it relate to the accuracy of multislice CT arthrography and conventional arthrography in diagnosing scapholunate ligament tears? (2) What is the estimated magnitude of skin radiation doses of each method? Methods: This secondary analysis of a previous prospective study included 71 men and women with suspected scapholunate ligament tears and indications for arthroscopy. Preoperative imaging was conventional arthrography and either MSCT arthrography for the first half of patients to be included (n = 36) or flat-panel CBCT arthrography for the remaining patients (n = 35). Index tests identified therapy-relevant SLL tears with dorsal or complete SLL ruptures, and these tears were compared with relevant SLL tears which were determined through arthroscopy as Geissler Stadium III and IV by probing the instable SL joint with a microhook or arthroscope. These injuries were treated by open ligament repair and Kirschner wire fixation. Accuracy values and 95% confidence intervals were calculated. Additional estimates of the radiation skin doses of each CBCT exam and two MSCT protocols were subsequently calculated using dose area products, dose length products, and CT dose indices. Results: The diagnostic accuracy was high for all imaging methods. 95% CIs were broadly overlapping and therefore did not indicate differences between the diagnostic groups: Sensitivity of CBCT arthrography was 100% (95% CI 77 to 100), specificity was 95% (95% CI 76 to 99.9), positive predictive value was 93% (95% CI 68 to 99.8), and negative predictive value was 100% (95% CI 83 to 100). For MSCT arthrography, the sensitivity was 92% (95% CI 64 to 99.8), specificity was 96% (95% CI 78 to 99.9), positive predictive value was 92% (95% CI 64 to 99.8), and negative predictive value was 96% (95% CI 78 to 99.9). For conventional arthrography, the sensitivity was comparably high: 96% (95% CI 81 to 99.9). Specificity was (81% [95% CI 67 to 92]); the positive predictive value was 77% (95% CI 59 to 89) and negative predictive value was 97% (95% CI 86 to 99.9). Estimated mean (range) radiation skin doses were reported in a descriptive fashion and were 12.9 mSv (4.5 to 24.9) for conventional arthrography, and 3.2 mSv (2.0 to 4.8) for CBCT arthrography. Estimated radiation skin doses were 0.2 mSv and 12 mSv for MSCT arthrography, depending on the protocol. Conclusion: Flat-panel CBCT arthrography can be recommended as an accurate technique to diagnose scapholunate ligament injuries after wrist trauma. Estimated skin doses are low for CBCT arthrography and adapted MSCT arthrography protocols. Level of Evidence: Level III, diagnostic study.

Background Timely treatment of scapholunate instability depends on early identification, but current imaging methods are either intricate or fail to demonstrate the dynamic stages. Purpose To calculate the diagnostic accuracy of four-dimensional (4D) CT for diagnosing instable scapholunate ligament (SLL) tears. Materials and Methods This prospective study enrolled consecutive participants with clinically suspected SLL tears who underwent 4D CT from July 2020 to May 2022. A historical study sample diagnosed at cineradiography served as a comparison, and wrist arthroscopy was the reference standard. Scapholunate joints greater than 3 mm were interpreted as instable at index 4D CT and cineradiography. Diagnostic accuracy was expressed as sensitivity and specificity. Areas under the receiver operating characteristic curve and cutoff values for both index tests were calculated. Intraclass correlation coefficients (ICCs) were computed to compare interrater reliability. Effective radiation doses at 4D CT were measured with thermoluminescent dosimeters. Results The study included 40 participants (mean age, 43 years ± 14 [SD]; 24 male) evaluated at 4D CT and 78 patients (mean age, 45 years ± 11; 50 male) historically evaluated at cineradiography. Four-dimensional CT helped detect instable tears in 26 of 35 participants (sensitivity, 74.3% [95% CI: 56.7, 87.5]. Cineradiography revealed instable tears in 52 of 63 patients (sensitivity, 82.5% [95% CI: 70.9, 91]). Four of five participants with stable scapholunate joints were identified at 4D CT (specificity, 80.0% [95% CI: 28.4, 99.5]), and 12 of 15 patients with stable SLLs were identified at cineradiography (specificity, 80.0% [95% CI: 51.9, 95.7]). Interrater agreement of radiologic measurements on 4D CT scans was good to excellent (ICC range, 0.89-0.96). The effective radiation dose ranged from 67 to 72 mSv at the wrist and was less than 1 mSv at the head. Conclusion Four-dimensional CT results are highly reproducible. Instable scapholunate joints greater than 3 mm were detected with a sensitivity of 74.3% and a specificity of 80% in an exploratory trial. Further evidence from larger randomized trials is warranted. German Register for Clinical Trials no. DRKS00021110 (Universal Trial Number U1111-1249-7884) Published under a CC BY 4.0 license. Supplemental material is available for this article. See also the editorial by Demehri and Ibad in this issue.