United States

Introduction
While delayed graft function (DGF) is less common after donation after brain death (DBD) than after circulatory death (DCD) in kidney transplantation (KT), this study investigated whether risk factors for DGF and post-transplant outcomes differ by donor type.
Methods
We analyzed all adult deceased donor kidney transplant recipients (DDKTR) at our center between 2005-2019, stratified by donor type (DBD vs. DCD). We assessed risk factors for DGF, acute rejection (AR), uncensored graft failure (UCGF), and death-censored graft failure (DCGF), within first year of transplant. 

Results
Among 2543 DDKTs, 804 (31%) were DCD. In both DBD and DCD recipients, older donor age, higher recipient BMI, and depleting induction agent increased the risk for DGF, while female recipient and preemptive transplant were protective. Additional risk factors in DBD but not in DCD recipients included higher terminal serum creatinine, higher KDPI, right donor kidney, and prolonged cold ischemia time (CIT). Female donors were protected against DGF in DCD only. DGF was linked to higher AR and UCGF risk, with no significant differences between DBD and DCD groups (AR: aHR 2.22 vs 2.37, p-interaction=0.65; UCGF: 3.04 vs 2.56; p-interaction=0.47). No adjusted DCGF data were available. 

Conclusion
DBD-KTs had more risk factors for DGF than DCD-KTs, including higher terminal creatinine, higher KDPI, right renal graft, and prolonged CIT. Female donors offered added protection against DGF in DCD-KTs. Despite higher DGF rates in DCD-KTs, the risk of adverse early outcomes after DGF development was similar between deceased donor types, suggesting DCD grafts remain a viable option despite DGF concerns.


AMA Research Challenge 2024 

Comparable Early Outcomes in Brain versus Circulatory Death Kidney Transplants After Delayed Graft Function

poster

Welcome to the **largest national, multi-specialty research** event featuring research from more than 750 medical students, residents, fellows, and international medical graduates.

Use the buttons below or navigation menu to access the event.
 

[![](https://assets.underline.io/markdown_image/1/image/20e1e77aa954efeac7a15d84c85eb383.jpg)](https://underline.io/events/468/sessions?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/068522d81aca0b68a369787c7dfccd1b.jpg)](https://underline.io/events/468/posters?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/d3971ece79d5e5575b178bdc2c8bbfae.jpg)](https://underline.io/events/468/help-desk)
 

#### Event Details

The 2024 American Medical Association Research Challenge includes a poster symposium hall and a semifinals hall with the top scored research. **Log in at any time throughout the event to:**

* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with participants** and provide feedback to help advance their research

At the conclusion of this event, the AMA will announce the top five finalists who will advance to the AMA Research Challenge finals, present their research to an elite panel of judges, and compete to win a $10,000 grand prize presented by Laurel Road. 

### Grand Prize Sponsor

AMA would like to thank Laurel Road for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road is not involved in winner selection process. The winner will be announced on February 20, 2025 during the AMA Research Challenge finals. Payment of the grand prize will be directly from Laurel Road. See the AMA Research Challenge Official Rules for terms and conditions.

[![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)](https://www.laurelroad.com/)

**There is no registration fee for this event! **

If you have used the AMA Research Challenge platform previously, please click the ‘Log-in’ button below. If you do not remember your password, please click “forgot password” and follow prompts to reset your password. After resetting your password, please refresh your webpage to access the platform.

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals. 

**Please use this link [here](https://www.ama-assn.org/member-benefits/events/ama-research-challenge) to register for free.**

Registration or Log-in Is Required

Join the AMA Research Challenge poster symposium and semifinals to explore top research posters and score posters in the semifinals (an AMA member exclusive opportunity).

Does CKD modify the effects of GLP-1RA and SGLT2i on weight loss?

Background:
It is unknown whether CKD modifies the weight loss effects of GLP1-RA or SGLT2i. 

Methods:
In an active comparator, new user study, we examined the effects of CKD status on bodyweight loss in veterans with T2D on metformin who began insulin glargine (IG), SGLT2i, or GLP1-RA from 1/1/18 to 3/31/21. Previous use of study medications between 1/1/08 and 12/31/17 was an exclusion. Index date was the date of study drug prescription fill. Outpatient bodyweight closest to the index date and within 1 year prior was defined as baseline bodyweight. Bodyweight change after drug initiation was defined as the difference between baseline bodyweight and an average of all outpatient weights across 6-month intervals after the index date, up to 24 months.

Follow up was until 24 months after index date or until loss to follow-up or death, whichever came earliest. Inverse probability weights (IPW) for each pairwise drug comparison resulted in balance of baseline variables across the drug classes. In IPW mixed effects models, study drug classes were related to weight change at 24 months in pairwise comparisons in non-CKD and CKD (eGFR<60) subgroups. Effect modification was tested by comparing the regression coefficients for each of the pairwise comparisons in non-CKD and CKD subgroups.

Results:
In the non-CKD group (N = 124,963), the percentages of female patients for GLP1-RA, SGLT2i, and IG respectively were 9.7%, 4.4%, and 5.9%. Mean ages were 62, 65, and 63; baseline eGFRs were 87.3, 85.6, and 87.3; and baseline BMIs were 35.8, 33.4, and 32.6, respectively. In the CKD group (N = 25,124), the percentages of female patients for GLP1-RA, SGLT2i, and IG respectively were 6.1%, 3.4%, and 4.3%. Mean ages were 71, 72, and 71; baseline eGFRs were 48.7, 51.0, and 48.6; and baseline BMIs were 34.2, 32.2, and 31.8, respectively. 

Absolute bodyweight difference at 24 months for IG compared to GLP1-RA was 6.77 ± 0.22 lbs in the non-CKD group and 7.15 ± 0.48 lbs in the CKD group (interaction p-value = 0.476). Corresponding numbers for IG compared to SGLT2i were 6.13 ± 0.18 lbs and 6.18 ± 0.37 lbs (interaction p-value = 0.906) and SGLT2i compared to GLP1-RA were 0.64 ± 0.20 and 0.97 ± 0.45 (interaction p-value = 0.507). Percent bodyweight difference at 24 months for IG compared to GLP1-RA was 3.01 ± 0.10 in the non-CKD group and 3.31 ± 0.21 in the CKD group (interaction p-value = 0.219). Corresponding numbers for IG compared to SGLT2i were 2.75 ± 0.08 and 2.92 ± 0.17 (interaction p-value = 0.365) and SGLT2i compared to GLP1-RA were 0.26 ± 0.09 and 0.39 ± 0.21 (interaction p-value = 0.575). 

Conclusion:
Patients lost significantly more weight on GLP1-RA or SGLT2i compared to insulin glargine, but there was minimal difference in weight change at 2 years between SGLT2i and GLP1-RA. CKD status did not modify weight loss on SGLT2i or GLP1-RA.

Does CKD modify the effects of GLP1-RA and SGLT2i on weight loss?

Background
Renal transplantation, a cornerstone in modern medicine, offers a crucial solution for those grappling with end-stage renal diseases. However, organ transportation challenges necessitate innovative preservation methods. This study aims to enhance transplant viability by optimizing kidney preservation through advanced ultrasound technology. It continues previous work with our hemoglobin-based oxygen carrier (HBOC) compound, addressing oxygen deprivation during transport with the OxyVita organ storage device. Enhanced with sugars and amino acids, HBOC creates an extracellular storage solution that maintains high oxygen delivery and metabolism ex vivo, offering greater clinical value than current practices.

Methods
Current market organ preservation solutions rely on dissolved oxygen, lacking active transport, which leads to ischemic necrosis and ROS build-up. Over 24-48 hours, control kidneys will be submerged in the University of Wisconsin (UW) solution, while study kidneys will be submerged and perfused with OxyVita. The NextGen LOGIQ e ultrasound system will be used to screen for healthy kidneys at removal and guide fine-needle biopsies every two hours. The Fine Needle Recognition Mode software allows for precise sectioning from the renal cortex, generating accurate data on tissue necrosis. Additionally, ultrasound-guided fluid collection will provide insights into critical metabolites. Post-experiment, periodic ultrasound assessments will monitor the transplanted kidney’s condition. The methodology incorporates color-flow Doppler and contrast-enhanced ultrasound for detailed structural and functional assessments, delineating perfusion patterns, detecting structural aberrations, and preserving renal echogenicity and corticomedullary differentiation, providing a reliable basis for clinical application.

Results
Preliminary research shows OXYVITA optimizing its storage solution, currently aiming to reach a storage time between 24-48 hours. Preliminary studies conducted at the UWM School of Medicine in Olsztyn, Poland have evaluated the role of OxyVita as a component of an organ storage solution in the preservation of renal tissue. Results demonstrated that more than 20% greater preservation of renal structure using OxyVita solution compared to control solution.

Conclusion
This study aims to confirm that the OxyVita solution deters pathophysiological processes. Data from renal tissue histology and spectrophotometry will be enhanced by high-quality imaging and contrast signals from the NextGen LOGIQ e system, strengthening the histopathological and biochemical findings. Leveraging advanced ultrasound technology, our investigation spans diverse dimensions, including assessing organ morphometrics, delineating architectural pathologies, understanding perfusion dynamics, and providing precise visualizations crucial for comprehensive research analyses to unravel rejection mechanisms within the application of the OxyVita product. Ultimately, our goal is to extend the shelf-life of kidneys to increase their transplantation viability for a broader patient population.

Utilizing Advanced Ultrasound Technology to Optimize OxyVita for Enhanced Preservation and Extended Viability of Kidney Transplants During Organ Transportation

Background
Chronic Traumatic Encephalopathy (CTE) is a progressive neurodegenerative disease associated with repetitive head trauma, predominantly affecting contact sports athletes and military personnel. Given the high prevalence of traumatic brain injuries (TBI) managed by neurosurgeons, it is crucial to understand CTE's pathophysiology, clinical presentation, and management strategies to enhance patient outcomes.
Methods
A systematic review and meta-analysis were conducted following PRISMA guidelines.
Searches were performed in PubMed, Cochrane Library, Embase, and Google Scholar for studies published from January 2010 to May 2024. Inclusion criteria encompassed studies on CTE epidemiology, pathophysiology, clinical presentation, diagnostics, and therapeutics. Data extraction and quality assessment were conducted independently by two reviewers, with discrepancies resolved by a third reviewer.

Results
Fourteen studies met the inclusion criteria, totaling 2729 observations and 799 events of CTE. The pooled proportion of CTE cases was 72.82% (95% CI: 48.17% to 88.53%) using a random effects model, with substantial heterogeneity (I2 = 98.6%). Subgroup analysis indicated high pooled proportions among athletes (71.43%) and military personnel (74.29%). Sensitivity analysis confirmed the robustness of the findings.
The pathophysiology of CTE involves the accumulation of hyperphosphorylated tau
protein and chronic neuroinflammation, leading to significant neuronal damage.
Clinically, CTE presents with cognitive impairment, mood disorders, and behavioral
changes, complicating its diagnosis due to overlapping symptoms with other
neurodegenerative diseases. Current treatments focus on symptomatic management, with emerging therapies targeting tau pathology and neuroinflammation showing promise.

Conclusion
This meta-analysis underscores the significant prevalence of CTE among individuals
exposed to repetitive head trauma, highlighting the need for standardized diagnostic
criteria and personalized management approaches. Neurosurgeons are pivotal in
advancing diagnostic protocols and therapeutic strategies to improve outcomes for
patients with CTE. Further research is needed to enhance understanding and
management of CTE, ultimately benefiting the broader TBI patient population frequently
encountered in neurosurgical

Chronic Traumatic Encephalopathy in Neurosurgical Practice: Insights from a Systematic Review and Meta-Analysis

Introduction: Proximal junctional kyphosis (PJK) is a common complication following correction of adult spinal deformity (ASD) at the transition of fused and mobile motion segments. PJF represents PJK with vertebral fracture, soft-tissue disruption, and/or screw pullout at the upper instrumented vertebra (UIV) accompanied with neurological deficit. Coronal decompensation following long-segment fusion may contribute to the development of proximal junctional pathology. This study reports on the natural history of coronal imbalance in ASD patients that develop PJF after fusion of the thoracolumbar spine to the pelvis. 
Methods: A single-institution retrospective review was performed for 1180 ASD patients who underwent posterior instrumented fusion to the pelvis (2009-2021). Patients that developed PJF were grouped by the UIV: upper thoracic (UT, T2-T6, 10 patients), lower thoracic (LT, T8-T11, 35 patients), or lumbar spine (L, L1-L3, 8 patients). Radiographic parameters, including the major scoliotic curve (MSC), coronal and sagittal proximal junctional angle (cPJA/sPJA), and the coronal and sagittal C7 vertical angles (CVA/SVA), were recorded before primary surgery, before revision surgery for failure, and at both 6-week and final follow-up following revision. 
Results: From 1180 ASD patients, 54 (4.6%) developed PJF (59% female, mean age 65.0 ± 6.8 and BMI 30.3 ± 5.9). Mean follow-up after revision was 3.1 ± 2.7 years. The ΔMSC (UT –17.6 ± 24.7°, LT –11.5 ± 15.97°, L –5.7 ± 7.6°) was calculated as the difference in MSC before revision surgery and before primary surgery. The mean ΔcPJA (UT 0.5 ± 3.6°, LT 0.6 ± 24.7°, L 2.5 ± 8.1°), ΔsPJA (UT: 19.1 ± 11.0°, LT: 14.8° ± 4.0°, L: 16.0° ± 9.0°), ΔCVA (UT: -13.6 ± 35.5 mm, LT –8.1 ± 21.1 mm, L 3.3 ± 29.4 mm), and ΔSVA (UT: -42.8 ± 86.8 mm, LT: 15.9 ± 38.4 mm, L: 48.6 ± 58.5 mm) were also calculated in that manner. 
Conclusion: Proximal junctional pathology presents differently radiographically following PJF after instrumented fusion from the UT, LT, and L spine. In this study, patients with a UIV in the L spine develop more coronal tilt and sagittal imbalance after PJF than patients with an UIV in the UT and LT spine.

Coronal Imbalance in Adult Spinal Deformity Patients That Develop Proximal Junctional Failure

Background:
The clinical use of routine CT imaging following intracranial tumor resection is controversial, with variable use in clinical practice. Socioeconomic studies investigating the clinical utility and cost-effectiveness of this resource are lacking.

Methods:
An institutional retrospective chart review was conducted on patients who underwent intracranial tumor resection from 2014-2024, performed by 17 individual surgeons. Data analyzed included incidence of postoperative CT imaging prior or in addition to post-resection MRI, imaging indications, and resulting changes in clinical management. A cost analysis was estimated based on compensation and cost-transparency datasets. Chi-squared and ANOVA tests were performed to identify variables associated with clinically meaningful routine imaging.

Results:
2,082 patients (2,177 resections) were included. 460 (22.1%) patients obtained CT imaging postoperatively, of which 305 (63.2%) were routine and 155 (37.5%) were targeted; no routine scans (0%) effected change in management, while targeted 50 scans (32.5%) did. Cost analysis showed that total costs for all post-operative CTs in this cohort was $398,008, with only $40,184 being CTs with acute findings that changed management. This produced excess costs of $348,257, which could have been prevented had a targeted approach been used across all patients in our hospital system. Only skull base and frontal tumors, intraoperative complications, and prolonged operative duration were associated with clinically beneficial CTs. Finally, no patient experienced delayed detection of postoperative deficits without routine imaging.

Discussion:
A targeted approach to postoperative CT imaging after intracranial tumor resection based on neurological exam monitoring is superior to routine imaging use in terms of cost and clinical benefit. Routine imaging may serve useful for patients undergoing skull base or frontal resections, particularly following longer surgeries or intraoperative complications.

Economic Analysis and Clinical Utility of Routine Versus Targeted Use Of Postoperative Computed Tomography Following Intracranial Tumor Resection: A Decade-Long Multi-Surgeon Institutional Study

Background:
Decompressive craniectomy (DC) is a crucial intervention for managing severe brain injury. The optimal timing for subsequent cranioplasty (CP) to maximize benefits and minimize complications is debated. This meta-analysis aims to synthesize current evidence to ascertain the impact of early versus delayed cranioplasty on neurological postoperative outcomes and complication rates.

Methods: 
An exhaustive literature search was conducted across PubMed, Scopus, and Embase up to January 2024, focusing on studies that addressed the timing of cranioplasty after craniectomy. The inclusion criteria were specifically designed to capture cross-sectional and retrospective studies, and case series of four or more adult patients, which directly examined the timing of cranioplasty. Studies not focused on timing, including those comparing materials, reporting techniques, or presenting cases with fewer than four patients, were excluded. Data extraction emphasized study design, patient demographics, timing intervals, and related outcomes, ensuring a rigorous quality assessment using the Newcastle-Ottawa Scale. Statistical analyses were aimed at evaluating study heterogeneity and synthesizing data regarding timing and its impact on outcomes.

Results: 
From the initial screening to the final selection, 7 studies with a total of 1009 patients were analyzed. The meta-analysis elucidates a significant association between the timing of cranioplasty and both neurological outcomes and complication rates. Notably, cranioplasty performed within 90 days post-craniectomy was linked to improved neurological recovery and reduced complication rates compared to delayed cranioplasty. Moreover, an ultra-early timing (within 30 days) was associated with lower infection rates but an increased risk of hydrocephalus, underscoring the complexity of timing decisions.

Conclusion: This meta-analysis provides compelling evidence that the timing of cranioplasty significantly influences neurological recovery and the incidence of postoperative complications. Early cranioplasty, preferably within 90 days of craniectomy, appears to offer advantages in terms of outcome and safety. However, the decision regarding timing should be carefully tailored to individual patient conditions, considering the nuanced balance between benefits and risks. Further high-quality, prospective studies are needed to refine these findings and support evidence-based clinical decision-making in cranioplasty timing.

Impact of Cranioplasty Timing on Postoperative Outcomes

Background: Opioid dependence is a major public health concern. Genetic factors likely contribute to opioid addiction vulnerability. A number of studies have examined associations between specific gene variants and opioid dependence, but results have been mixed.
Methods: A systematic literature search was conducted through January 2024 to identify studies evaluating relationships between single nucleotide polymorphisms (SNPs) and opioid dependence. Pooled odds ratios (OR) and 95% confidence intervals (CI) were calculated using random-effects meta-analysis models. Between-study heterogeneity was assessed with I2 statistics.
Results: 66 studies involving 21,485 participants were included in the meta-analysis. Significant associations were found between opioid dependence and variants in the mu-opioid receptor (OPRM1) gene (OR=0.466, 95% CI 0.346-0.628, p=5.456x10-7), prodynorphin (PDYN) gene (OR=0.296, 95% CI 0.221-0.398, p<0.001), catechol-O-methyltransferase (COMT) gene (OR=0.438, 95% CI 0.383-0.501, p<0.001), and other genes. Effects were detected across multiple genetic models and ancestral groups.
Conclusion: This meta-analysis identified multiple single nucleotide polymorphisms associated with heightened opioid dependence susceptibility. The findings advance knowledge of genetic contributions to opioid addiction vulnerability across diverse populations.

Single Nucleotide Polymorphisms Associated with Risk of Opioid Dependence

Introduction: Stroke patients often experience motor deficits in the limbs and do not reach satisfactory recovery. Deep brain stimulation (DBS) can target intrinsic pathways that project to vital cortical areas. One structure of interest is the ventro-oralis posterior (VOP)/ventro-intermedium (VIM) of the motor thalamus. The VOP/VIM is a thalamic relay with outgoing excitatory connectivity to the primary motor cortex (M1) and could be a potential target for DBS to facilitate motor function after stroke. We hypothesized that VOP/VIM stimulation may increase the excitability of M1 and strengthen corticospinal motor output and control.
Methods: Five non-human primates (NHP) were implanted with DBS electrodes into the ventral lateral motor thalamus, the internal capsule (IC), and intracortical microelectrode arrays over M1. This setup was replicated in six patients undergoing implantation of DBS electrodes targeting the VOP/VIM for essential tremor with the addition of a macroarray placed over M1. Intraoperatively, direct cortical stimulation (DCS) was paired with VOP/VIM stimulation. Additionally, a single patient with a traumatic brain injury (TBI) with motor deficits was implanted. When the TBI patient returned for DBS programming, he was asked to modulate their force along a trace and the error of deviation was measured with and without DBS while performing weighted and unweighted arm exercises.
Results: In the NHPs, a significantly increased amplitude of evoked potentials (EP) and single unit spike counts recorded in M1 with VOP/VIM stimulation was observed. IC stimulation with VOP/VIM stimulation increased the amplitude of motor evoked potentials (MEP) and kinematics of hand muscles. In the intraoperative human studies, when VOP/VIM was stimulated, similar amplification in M1 EPs and MEPs recorded in upper extremity muscles were observed. Stimulation also increased the height the TBI patient was able to reach in the arm exercises as well as reduced the error during force modulation tasks.
Conclusion: This study documented that direct electrical stimulation of the motor thalamus strengthens motor output, both in NHPs and humans. These outcomes support DBS of the motor thalamus as a potential therapeutic approach to treat post-stroke motor deficits.


Strengthening Motor Output with Direct Motor Thalamus Stimulation. Evidence For A Novel Deep Brain Neuromodulation Modality For Stroke.

Background: This systematic review and meta-analysis comprehensively compares
and analyzes the strengths and weaknesses of three neuromodulating treatment
options—electroconvulsive therapy (ECT), transcranial magnetic stimulation (rTMS),
and psychedelics—for pain management in individuals with Fibromyalgia.

Methods: A systematic literature search began May 31, 2023 and ended July 31,
2023 and was conducted across multiple databases, including PubMed, Google
Scholar, ScienceDirect, EMBASE, Wiley Online Library, Cochrane Library, and
EBSCO, using relevant keywords and controlled vocabulary terms. Key terms for
search included: “Fibromyalgia,” “Neuromodulation therapy,” “Electroconvulsive
Therapy,” “ECT,” “Transcranial Magnetic Stimulation.” Studies published between
2004-2023 were included in the review. The inclusion criteria encompassed
randomized controlled trials, qualitative studies, surveys, prospective cohort studies,
and case-control studies evaluating the effectiveness of ECT, TMS, and psychedelics
as neurocognitive treatments for Fibromyalgia pain management. The exclusion
criteria included animal trials, articles written in a language other than English, case
reports, and case studies.

Results: The search yielded ten studies with a total of 541 subjects, meeting the
inclusion criteria. There was an overall 2.7 mean pain decrease in pain rating scale
after therapy.

Conclusion: This review on neuromodulating Fibromyalgia pain treatments
recognizes marked improvement in pain perception in Fibromyalgia patients
receiving neuromodulating treatment.

A Systematic Review Comparing the Strengths and Weaknesses of Neuromodulating
Treatments for Pain Management in Fibromyalgia: ECT, TMS, and Psychedelics


Neuropathology holds the potential to advance our understanding of Alzheimer's disease and related dementias. Artificial intelligence (AI) models through well-labeled datasets can effectively perform complex tasks at or above human performance. Recent studies have analyzed neuropathological biomarkers like amyloid beta plaques and neurofibrillary tangles in whole slide images through AI models. However, appropriately labeled images are needed to detect which brain region the neuropathological biomarkers originate from for analytical purposes. The study aims to develop automated detection of gray matter regions in WSIs for downstream testing of established AI pipelines in neuropathology.The study involves manual annotations of gray matter regions on Bielschowsky silver stained WSIs from the Emory ADRC digital cohort. These annotations are used to train UNET semantic segmentation models for automated detection of gray matter regions on a larger cohort of non-annotated WSIs. Co-registration to immunohistochemically stained WSIs (amyloid beta, tau, TDP-43) is used to define gray matter annotations across stains. Established AI pipelines are then deployed in these regions to assess improvement on original results. 300 gray matter masks have been generated and an initial classifier has been trained using the YOLOv8 (You Only Look Once) model developed by Ultralytics. While the project is ongoing, our initial results have demonstrated the feasibility of developing an automated gray matter segmentation model to further understanding of biomarkers associated with clinical care.


Premium content

Comparable Early Outcomes in Brain versus Circulatory Death Kidney Transplants After Delayed Graft Function

Next from AMA Research Challenge 2024

Does CKD modify the effects of GLP1-RA and SGLT2i on weight loss?

Stay up to date with the latest Underline news!

PRESENTATIONS

CONFERENCES

COMPANY

RESOURCES