Asleep versus awake GPi DBS surgery for Parkinson's disease: A systematic review and meta-analysis.

BACKGROUND: Patients with Parkinson's Disease (PD) who receive either asleep image-guided subthalamic nucleus deep brain stimulation (DBS) or the traditional awake technique have comparable motor outcomes. However, there are fewer studies regarding which technique should be chosen for globus pallidus internus (GPi) DBS. This systematic review and meta-analysis aims to compare the accuracy of lead placement and motor outcomes of asleep versus awake GPi DBS PD population. METHODS: We systematically searched PubMed, Embase, and Cochrane for studies comparing asleep vs. awake GPi DBS lead placement in patients with PD. Outcomes were spatial accuracy of lead placement, measured by radial error between intended and actual location, motor improvement measured using (UPDRS III), and postoperative stimulation parameters. Statistical analysis was performed with Review Manager 5.1.7. and OpenMeta [Analyst]. RESULTS: Three studies met inclusion criteria with a total of 247 patients. Asleep DBS was used to treat 192 (77.7 %) patients. Follow-up ranged from 6 to 48 months. Radial error was not statistically different between groups (MD -0.49 mm; 95 % CI -1.0 to 0.02; I2 = 86 %; p = 0.06), with a tendency for higher target accuracy with the asleep technique. There was no significant difference between groups in change on motor function, as measured by UPDRS III, from pre- to postoperative (MD 8.30 %; 95 % CI -4.78 to 21.37; I2 = 67 %, p = 0.2). There was a significant difference in postoperative stimulation voltage, with the asleep group requiring less voltage than the awake group (MD -0.27 V; 95 % CI -0.46 to - 0.08; I2 = 0 %; p = 0.006). CONCLUSION: Our meta-analysis indicates that asleep image-guided GPi DBS presents a statistical tendency suggesting superior target accuracy when compared with the awake standard technique. Differences in change in motor function were not statistically significant between groups.

Replacement of traditional vagus nerve stimulation with cardiac-based device and seizure reduction: A systematic review and meta-analysis.

INTRODUCTION: For patients with drug-resistant epilepsy (DRE) who are not suitable for surgical resection, neuromodulation with vagus nerve stimulation (VNS) is an established approach. However, there is limited evidence of seizure reduction when replacing traditional VNS (tVNS) device with a cardiac-based one (cbVNS). This meta-analysis compares the seizure reduction achieved by replacing tVNS with cbVNS in a population with DRE. METHODS: We systematically searched PubMed, Embase, and Cochrane Central following PRISMA guidelines. The main outcomes were number of patients experiencing a ≥ 50 % and ≥80 % reduction in seizures, as defined by the McHugh scale. Additionally, we assessed the number of patients achieving freedom from seizures. RESULTS: We included 178 patients with DRE from 7 studies who were initially treated with tVNS and subsequently had it replaced by cbVNS. The follow-up for cbVNS ranged from 6 to 37.5 months. There was a statistically significant reduction in seizure frequency with the replacement of tVNS by cbVNS, using a ≥ 50 % (OR 1.79; 95 % CI 1.07 to 2.97; I²=0 %; p = 0.03) and a ≥ 80 % (OR 2.06; 95 % CI 1.17 to 3.62; I²=0 %; p = 0.01) reduction threshold. Nineteen (13 %) participants achieved freedom from seizures after switching to cbVNS. There was no difference in the rate of freedom from seizures between groups (OR 1.85; 95 % CI 0.81 to 4.21; I²=0 %; p = 0.14). CONCLUSION: In patients with DRE undergoing battery replacement, cbVNS might be associated with seizure reduction (≥50 % and ≥80 % threshold) after switching from tVNS. Randomised controlled trials are necessary to validate these findings.

Application of Machine Learning for Classification of Brain Tumors: A Systematic Review and Meta-Analysis.

BACKGROUND: Classifying brain tumors accurately is crucial for treatment and prognosis. Machine learning (ML) shows great promise in improving tumor classification accuracy. This study evaluates ML algorithms for differentiating various brain tumor types. METHODS: A systematic review and meta-analysis were conducted, searching PubMed, Embase, and Web of Science up to March 14, 2023. Studies that only investigated image segmentation accuracy or brain tumor detection instead of classification were excluded. We extracted binary diagnostic accuracy data, constructing contingency tables to derive sensitivity and specificity. RESULTS: Fifty-one studies were included. The pooled area under the curve for glioblastoma versus lymphoma and low-grade versus high-grade gliomas were 0.99 (95% confidence interval [CI]: 0.98-1.00) and 0.89, respectively. The pooled sensitivity and specificity for benign versus malignant tumors were 0.90 (95% CI: 0.85-0.93) and 0.93 (95% CI: 0.90-0.95), respectively. The pooled sensitivity and specificity for low-grade versus high-grade gliomas were 0.99 (95% CI: 0.97-1.00) and 0.94, (95% CI: 0.79-0.99), respectively. Primary versus metastatic tumor identification yields sensitivity and specificity of 0.89, (95% CI: 0.83-0.93) and 0.87 (95% CI: 0.82-0.91), correspondingly. The differentiation of gliomas from pituitary tumors yielded the highest results among primary brain tumor classifications: sensitivity of 0.99 (95% CI: 0.99-1.00) and specificity of 0.99 (95% CI: 0.98-1.00). CONCLUSIONS: ML demonstrated excellent performance in classifying brain tumor images, with near-maximum area under the curves, sensitivity, and specificity.