Search

BACKGROUND AND PURPOSE: Supervised deep learning is the state-of-the-art method for stroke lesion segmentation on NCCT. Supervised methods require manual lesion annotations for model development, while unsupervised deep learning methods such as generative adversarial networks do not. The aim of this study was to develop and evaluate a generative adversarial network to segment infarct and hemorrhagic stroke lesions on follow-up NCCT scans.MATERIALS AND METHODS: Training data consisted of 820 patients with baseline and follow-up NCCT from 3 Dutch acute ischemic stroke trials. A generative adversarial network was optimized to transform a follow-up scan with a lesion to a generated baseline scan without a lesion by generating a difference map that was subtracted from the follow-up scan. The generated difference map was used to automatically extract lesion segmentations. Segmentation of primary hemorrhagic lesions, hemorrhagic transformation of ischemic stroke, and 24-hour and 1-week follow-up infarct lesions were evaluated relative to expert annotations with the Dice similarity coefficient, Bland-Altman analysis, and intraclass correlation coefficient.RESULTS: The median Dice similarity coefficient was 0.31 (interquartile range, 0.08-0.59) and 0.59 (interquartile range, 0.29-0.74) for the 24-hour and 1-week infarct lesions, respectively. A much lower Dice similarity coefficient was measured for hemorrhagic transformation (median, 0.02; interquartile range, 0-0.14) and primary hemorrhage lesions (median, 0.08; interquartile range, 0.01-0.35). Predicted lesion volume and the intraclass correlation coefficient were good for the 24-hour (bias, 3 mL; limits of agreement, -64-59 mL; intraclass correlation coefficient, 0.83; 95% CI, 0.78-0.88) and excellent for the 1-week (bias, -4 m; limits of agreement,-66-58 mL; intraclass correlation coefficient, 0.90; 95% CI, 0.83-0.93) follow-up infarct lesions.CONCLUSIONS: An unsupervised generative adversarial network can be used to obtain automated infarct lesion segmentations with a moderate Dice similarity coefficient and good volumetric correspondence.

Background and purpose Flow diverters are increasingly used to treat intracranial aneurysms. We report the safety and efficacy of the p64 flow diverter, a resheathable and detachable device for intracranial aneurysms. Materials and methods We retrospectively reviewed 108 patients with 109 aneurysms treated with the p64 between March 2014 and July 2019. There were 87 women and 21 men, mean age 57 years. Of 109 aneurysms, 74 were discovered incidentally, 12 were symptomatic, 18 were previously treated, and five were ruptured dissection aneurysms. A total of 10 aneurysms were located in the posterior circulation. The mean aneurysm or remnant size was 8.1 mm. Results Hemorrhage by perforation with the distal guidewire occurred in two patients with permanent neurological deficits in one. In one patient, acute in-stent occlusion caused infarction with a permanent deficit. Permanent morbidity was 1.9% (2 of 108, 95%CI 0.1–6.9%); there was no mortality. During follow-up, three in-stent occlusions occurred, all asymptomatic. There were no delayed hemorrhagic complications. At six months, 77 of 96 aneurysms (80.2%) were completely occluded, and at last follow-up, this increased to 93 of 96 aneurysms (96.9%). In-stent stenosis at any degree occurred in 11 patients, progressing to asymptomatic complete occlusion in one. In the other patients, stenosis resolved or improved at further follow-up. Conclusion The p64 offers an effective and safe treatment option. Aneurysm occlusion rate was 97% at last follow-up, mostly achieved with a single device. There were no delayed hemorrhagic complications. Delayed in-stent stenosis infrequently progresses to occlusion but remains a matter of concern.

Background The novel low-profile p48 flow diverter has been designed to treat aneurysms on small vessels of 1.75–3mm. We report our first clinical experiences. Methods Between March 2018–January 2020, 22 patients with 25 aneurysms were treated with the p48 in 3 centers. One patient had 3 aneurysms covered by one p48 and one patient had 2 aneurysms. There were 5 men, 17 women, with a mean age of 55 years (median 59, range 29–73 years). Results In 25 aneurysms, 24 p48 flow diverters were placed. In 1 patient additional coils were placed in the aneurysm. Procedural vessel rupture by the micro guidewire occurred in 2 patients and vessel rupture during p48 balloon dilatation occurred in 1 patient. Overall, the permanent morbidity rate was 13.6% (3 of 22, 95%CI 3.9–34.2%) and mortality was 4.5% (1 of 22, 95%CI Conclusions Treatment of aneurysms in small-caliber vessels with the p48 is feasible and effective but is not without complications. More data is needed to establish indications, safety, and efficacy more accurately.

Summary Background Endovascular renal denervation reduces blood pressure in patients with mild-to-moderate hypertension, but its efficacy in patients with true resistant hypertension has not been shown. We aimed to assess the efficacy and safety of endovascular ultrasound renal denervation in patients with hypertension resistant to three or more antihypertensive medications. Methods In a randomised, international, multicentre, single-blind, sham-controlled trial done at 28 tertiary centres in the USA and 25 in Europe, we included patients aged 18–75 years with office blood pressure of at least 140/90 mm Hg despite three or more antihypertensive medications including a diuretic. Eligible patients were switched to a once daily, fixed-dose, single-pill combination of a calcium channel blocker, an angiotensin receptor blocker, and a thiazide diuretic. After 4 weeks of standardised therapy, patients with daytime ambulatory blood pressure of at least 135/85 mm Hg were randomly assigned (1:1) by computer (stratified by centres) to ultrasound renal denervation or a sham procedure. Patients and outcome assessors were masked to randomisation. Addition of antihypertensive medications was allowed if specified blood pressure thresholds were exceeded. The primary endpoint was the change in daytime ambulatory systolic blood pressure at 2 months in the intention-to-treat population. Safety was also assessed in the intention-to-treat population. This study is registered with ClinicalTrials.gov , NCT02649426 . Findings Between March 11, 2016, and March 13, 2020, 989 participants were enrolled and 136 were randomly assigned to renal denervation (n=69) or a sham procedure (n=67). Full adherence to the combination medications at 2 months among patients with urine samples was similar in both groups (42 [82%] of 51 in the renal denervation group vs 47 [82%] of 57 in the sham procedure group; p=0·99). Renal denervation reduced daytime ambulatory systolic blood pressure more than the sham procedure (−8·0 mm Hg [IQR –16·4 to 0·0] vs –3·0 mm Hg [–10·3 to 1·8]; median between-group difference –4·5 mm Hg [95% CI –8·5 to –0·3]; adjusted p=0·022); the median between-group difference was –5·8 mm Hg (95% CI –9·7 to –1·6; adjusted p=0·0051) among patients with complete ambulatory blood pressure data. There were no differences in safety outcomes between the two groups. Interpretation Compared with a sham procedure, ultrasound renal denervation reduced blood pressure at 2 months in patients with hypertension resistant to a standardised triple combination pill. If the blood pressure lowering effect and safety of renal denervation are maintained in the long term, renal denervation might be an alternative to the addition of further antihypertensive medications in patients with resistant hypertension. Funding ReCor Medical.

Background: The multicenter, international, randomized, blinded, sham-controlled RADIANCE-HTN SOLO trial (A Study of the ReCor Medical Paradise System in Clinical Hypertension) demonstrated a 6.3 mm Hg greater reduction in daytime ambulatory systolic blood pressure (BP) at 2 months by endovascular ultrasound renal denervation (RDN) compared with a sham procedure among patients not treated with antihypertensive medications. We report 6-month results after the addition of a recommended standardized stepped-care antihypertensive treatment to the randomized endovascular procedure under continued blinding to initial treatment. Methods: Patients with a daytime ambulatory BP ≥135/85 mm Hg and Results: A total of 69/74 RDN patients and 71/72 sham patients completed the 6-month ambulatory BP measurement. At 6 months, 65.2% of patients in the RDN group were treated with the standardized stepped-care antihypertensive treatment versus 84.5% in the sham group ( P =0.008), and the average number of antihypertensive medications and defined daily dose were less in the RDN group than in the sham group (0.9±0.9 versus 1.3±0.9, P =0.010 and 1.4±1.5 versus 2.0±1.8, P =0.018; respectively). Despite less intensive standardized stepped-care antihypertensive treatment, RDN reduced daytime ambulatory systolic BP to a greater extent than sham (−18.1±12.2 versus −15.6±13.2 mm Hg, respectively; difference adjusted for baseline BP and number of medications: −4.3 mm Hg, 95% confidence interval, −7.9 to −0.6, P =0.024). There were no major adverse events in either group through 6 months. Conclusions: The BP-lowering effect of endovascular ultrasound RDN was maintained at 6 months with less prescribed antihypertensive medications compared with a sham control. Clinical Trial Registration: URL: https://www.clinicaltrials.gov . Unique identifier: NCT02649426.