1
|
Yu Y, Chen Z, Yang Y, Zhang J, Wang Y. Development and validation of an interpretable machine learning model for predicting post-stroke epilepsy. Epilepsy Res 2024; 205:107397. [PMID: 38976953 DOI: 10.1016/j.eplepsyres.2024.107397] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/19/2024] [Revised: 06/19/2024] [Accepted: 06/20/2024] [Indexed: 07/10/2024]
Abstract
BACKGROUND Epilepsy is a serious complication after an ischemic stroke. Although two studies have developed prediction model for post-stroke epilepsy (PSE), their accuracy remains insufficient, and their applicability to different populations is uncertain. With the rapid advancement of computer technology, machine learning (ML) offers new opportunities for creating more accurate prediction models. However, the potential of ML in predicting PSE is still not well understood. The purpose of this study was to develop prediction models for PSE among ischemic stroke patients. METHODS Patients with ischemic stroke from two stroke centers were included in this retrospective cohort study. At the baseline level, 33 input variables were considered candidate features. The 2-year PSE prediction models in the derivation cohort were built using six ML algorithms. The predictive performance of these machine learning models required further appraisal and comparison with the reference model using the conventional triage classification information. The Shapley additive explanation (SHAP), based on fair profit allocation among many stakeholders according to their contributions, is used to interpret the predicted outcomes of the naive Bayes (NB) model. RESULTS A total of 1977 patients were included to build the predictive model for PSE. The Boruta method identified NIHSS score, hospital length of stay, D-dimer level, and cortical involvement as the optimal features, with the receiver operating characteristic curves ranging from 0.709 to 0.849. An additional 870 patients were used to validate the ML and reference models. The NB model achieved the best performance among the PSE prediction models with an area under the receiver operating curve of 0.757. At the 20 % absolute risk threshold, the NB model also provided a sensitivity of 0.739 and a specificity of 0.720. The reference model had poor sensitivities of only 0.15 despite achieving a helpful AUC of 0.732. Furthermore, the SHAP method analysis demonstrated that a higher NIHSS score, longer hospital length of stay, higher D-dimer level, and cortical involvement were positive predictors of epilepsy after ischemic stroke. CONCLUSIONS Our study confirmed the feasibility of applying the ML method to use easy-to-obtain variables for accurate prediction of PSE and provided improved strategies and effective resource allocation for high-risk patients. In addition, the SHAP method could improve model transparency and make it easier for clinicians to grasp the prediction model's reliability.
Collapse
Affiliation(s)
- Yue Yu
- Affiliated Hospital of Qingdao University, Qingdao, China; Qingdao Municipal Hospital, Qingdao, China
| | - Zhibin Chen
- Department of Neuroscience, Central Clinical School, Monash University, Melbourne, Victoria, Australia
| | - Yong Yang
- Affiliated Hospital of Qingdao University, Qingdao, China
| | | | - Yan Wang
- Affiliated Hospital of Qingdao University, Qingdao, China.
| |
Collapse
|
2
|
Fu CH, Chen CH, Lin YH, Lee CW, Tsai LK, Tang SC, Shun CT, Jeng JS. High fibrin and platelet clot predicts stroke recurrence or mortality after thrombectomy in patients with active cancer. J Neurointerv Surg 2024:jnis-2024-022033. [PMID: 39216988 DOI: 10.1136/jnis-2024-022033] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 06/15/2024] [Accepted: 08/20/2024] [Indexed: 09/04/2024]
Abstract
BACKGROUND Fibrin and platelet (FP)-rich clots have been shown to be associated with cancer-related stroke. This study aims to investigate the prognostic role of thrombus composition in clinical outcomes among cancer patients who experienced stroke and received endovascular thrombectomy (EVT). METHODS We included acute ischemic stroke patients who underwent EVT between March 2015 and November 2021. These patients were categorized into three groups: those with active cancer, those with non-active cancer, and those without cancer. The percentages of FP in clots were quantified under hematoxylin and eosin staining. The primary outcome was defined as any stroke recurrence or mortality within 90 days following the index stroke event. RESULTS A total of 420 patients with retrieved clots were included in the study. This cohort comprised 50 patients with active cancer, 23 patients with non-active cancer, and 347 patients without cancer. The percentage of FP was significantly higher in thrombi retrieved from patients with active cancer compared with the other two groups. Patients in the active cancer group exhibited a higher rate of the primary outcome compared with the other groups. After adjusting for clinical variables, a higher percentage of FP in thrombi remained significantly associated with the primary outcome in the active cancer group (adjusted odds ratio (aOR) =1.03 (1.00-1.06), P=0.028), but not in the other two groups. CONCLUSION Among stroke patients receiving EVT, thrombi with a higher percentage of FP not only identify individuals with active cancer but also predict stroke recurrence or mortality within 90 days.
Collapse
Affiliation(s)
- Chuan-Hsiu Fu
- Department of Neurology, National Taiwan University Hospital Hsin-Chu Branch, Hsinchu, Taiwan
| | - Chih-Hao Chen
- Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan
| | - Yen-Heng Lin
- Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan
| | - Chung-Wei Lee
- Department of Medical Imaging, National Taiwan University Hospital, Taipei, Taiwan
| | - Li-Kai Tsai
- Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan
| | - Sung-Chun Tang
- Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan
| | - Chia-Tung Shun
- Department of Pathology, National Taiwan University Hospital, Taipei, Taiwan
- Institute of Forensic Medicine, National Taiwan University College of Medicine, Taipei, Taiwan
| | - Jiann-Shing Jeng
- Department of Neurology, National Taiwan University Hospital, Taipei, Taiwan
| |
Collapse
|
3
|
Dou J, Zhao J, Lv Y, Jia H, Cao Y. Relationship between serum plasminogen activator and D-dimer levels and the severity of Kawasaki disease in children as well as their predictive value for coronary artery lesion. Am J Transl Res 2024; 16:3240-3247. [PMID: 39114687 PMCID: PMC11301476 DOI: 10.62347/fnhe1709] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 02/21/2024] [Accepted: 05/20/2024] [Indexed: 08/10/2024]
Abstract
OBJECTIVE To evaluate the relationships between serum plasminogen activator (PA) and D-dimer levels, the severity of Kawasaki disease (KD) in children, and their ability to predict coronary artery lesions (CAL). METHODS This retrospective study analyzed the clinical data of 102 children diagnosed with KD at the Affiliated Hospital of Jiangnan University from January 2020 to September 2023. The cohort was divided into two groups: 31 children with CAL in the CAL group and 71 without it in the non-CAL group. The study assessed the incidence of CAL and investigated the correlations between serum PA and D-dimer levels and various inflammatory markers (white blood cell (WBC) count, platelet count, and erythrocyte sedimentation rate (ESR)). Receiver operating characteristic (ROC) curves were used to evaluate the predictive value of these biomarkers for CAL. RESULTS CAL was present in 30.04% of the children. Pearson correlation analysis revealed that serum PA levels were inversely correlated with WBC count (P = 0.0187), platelet count (P = 0.0116), and ESR (P = 0.0041), while D-dimer levels were positively correlated with these markers (P < 0.001). A negative correlation between PA and D-dimer levels was also observed (P < 0.001). The combined use of PA and D-dimer levels to predict CAL achieved an area under the curve of 0.871. CONCLUSION Serum PA levels were negatively associated with the severity of KD, whereas D-dimer levels were positively associated. Together, these markers showed significant predictive value for CAL, highlighting their utility in assessing disease severity and guiding management in children with KD.
Collapse
Affiliation(s)
- Jijuan Dou
- Department of Pediatrics, Affiliated Hospital of Jiangnan UniversityNo. 1000 Hefeng Road, Binhu District, Wuxi 214122, Jiangsu, China
| | - Jinling Zhao
- Department of Pediatrics, Affiliated Hospital of Jiangnan UniversityNo. 1000 Hefeng Road, Binhu District, Wuxi 214122, Jiangsu, China
| | - Yuwen Lv
- Department of Pediatrics, Suzhou Municipal HospitalNo. 26 Daoqian Street, Gusu District, Suzhou 215000, Jiangsu, China
| | - Hongliang Jia
- Department of Pediatrics, Affiliated Hospital of Jiangnan UniversityNo. 1000 Hefeng Road, Binhu District, Wuxi 214122, Jiangsu, China
| | - Yachuan Cao
- Department of Pediatrics, Affiliated Hospital of Jiangnan UniversityNo. 1000 Hefeng Road, Binhu District, Wuxi 214122, Jiangsu, China
| |
Collapse
|
4
|
Hacialioglu RA, Kielkopf M, Branca M, Clenin L, Boronylo A, Silimon N, Göldlin MB, Scutelnic A, Kaesmacher J, Mujanovic A, Meinel TR, Seiffge DJ, Heldner MR, Liberman AL, Navi BB, Fischer U, Arnold M, Jung S, Bücke P, Beyeler M. Factors impacting D-dimer levels in patients with acute ischemic cerebrovascular events. J Stroke Cerebrovasc Dis 2024; 33:107834. [PMID: 38936311 DOI: 10.1016/j.jstrokecerebrovasdis.2024.107834] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 04/12/2024] [Revised: 06/12/2024] [Accepted: 06/23/2024] [Indexed: 06/29/2024] Open
Abstract
BACKGROUND AND OBJECTIVES A better understanding of the factors influencing D-dimer levels in code stroke patients is needed to guide further investigations of concomitant thrombotic conditions. This study aimed to investigate the impact of time from symptom onset and other factors on D-dimer levels in patients with acute ischemic stroke (AIS) or transient ischemic attack (TIA). METHODS Data on consecutive AIS and TIA patients treated at our tertiary-care stroke center between January 2015 and December 2020 were retrospectively assessed. Patients with available D-dimer levels were evaluated for eligibility. Multivariable non-linear regression analyses were performed. RESULTS In total, 2467 AIS patients and 708 TIA patients were included. The median D-dimer levels differed between the AIS and TIA groups (746 µg/L [interquartile range 381-1468] versus 442 µg/L [interquartile range 244-800], p<0.001). In AIS patients, an early increase in D-dimer levels was demonstrated within the first 6 h (standardized beta coefficient [β] 0.728; 95% confidence interval [CI] 0.324-1.121). This was followed by an immediate decrease (β -13.022; 95% CI -20.401 to -5.643) and then by a second, late increase after 35 h (β 11.750; 95% CI 4.71-18.791). No time-dependent fluctuation in D-dimer levels was observed in TIA patients. CONCLUSION The time from symptom onset may affect D-dimer levels in patients with AIS but not those with TIA. Further studies confirming these findings and validating time-specific variations are needed to enable D-dimer levels to be used efficiently as an acute stroke and thrombotic risk biomarker.
Collapse
Affiliation(s)
- Recep-Ali Hacialioglu
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Moritz Kielkopf
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Mattia Branca
- CTU Bern, Institute of Social and Preventive Medicine, University of Bern, Bern, Switzerland
| | - Leander Clenin
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Anna Boronylo
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Norbert Silimon
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Martina B Göldlin
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Adrian Scutelnic
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Johannes Kaesmacher
- Institute for Diagnostic and Interventional Neuroradiology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Adnan Mujanovic
- Institute for Diagnostic and Interventional Neuroradiology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Thomas R Meinel
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - David J Seiffge
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Mirjam R Heldner
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Ava L Liberman
- Clinical and Translational Neuroscience Unit, Feil Family Brain and Mind Research Institute and Department of Neurology, Weill Cornell Medicine, New York, NY, USA
| | - Babak B Navi
- Clinical and Translational Neuroscience Unit, Feil Family Brain and Mind Research Institute and Department of Neurology, Weill Cornell Medicine, New York, NY, USA
| | - Urs Fischer
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Marcel Arnold
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Simon Jung
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Philipp Bücke
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland
| | - Morin Beyeler
- Department of Neurology, Inselspital, Bern University Hospital, and University of Bern, Bern, Switzerland; Clinical and Translational Neuroscience Unit, Feil Family Brain and Mind Research Institute and Department of Neurology, Weill Cornell Medicine, New York, NY, USA; Graduate School for Health Sciences, University of Bern, Switzerland.
| |
Collapse
|
5
|
Li D, Zhou M, Zha F, Long J, Wang Y. Association between N-terminal pro-B-type natriuretic peptide and clinical outcomes in bedridden patients with stroke: a cross-sectional study. BMJ Open 2024; 14:e077083. [PMID: 38286702 PMCID: PMC10826584 DOI: 10.1136/bmjopen-2023-077083] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 06/27/2023] [Accepted: 11/29/2023] [Indexed: 01/31/2024] Open
Abstract
OBJECTIVES Patients with stroke often remain bedridden despite rehabilitation. Serum N-terminal pro-B-type natriuretic peptide (NT-pro-BNP) levels increase after stroke. Our study aimed to investigate the difference in NT-pro-BNP levels between bedridden and non-bedridden patients with stroke and to explore the factors influencing NT-pro-BNP levels in bedridden patients. DESIGN A single-centre, cross-sectional study. SETTING This study was conducted in a hospital, Shenzhen, China. PARTICIPANTS Between January 2019 and December 2022, 465 participants were included in this study. OUTCOME MEASURES The collected data included basic information, laboratory data and echocardiographic parameters. Binary logistic regression analysis and receiver operating characteristic curves were used to identify factors associated with high NT-pro-BNP levels. RESULTS Bedridden patients with stroke had higher levels of NT-pro-BNP, D-dimer, high-sensitivity C reactive protein (hs-CRP) and lower levels of creatinine, high-density lipoprotein cholesterol, albumin and haemoglobin, as well as lower left ventricular ejection fraction, fractional shortening and the ratio between the peak velocities of early and late diastolic filling than non-bedridden patients. In bedridden patients, age ≥75 years, high levels of hs-CRP and creatinine, and low levels of albumin were associated with high NT-pro-BNP levels. In non-bedridden patients, age ≥75 years and high creatinine levels were associated with high NT-pro-BNP levels. In bedridden patients with stroke, the area under the curve (AUC) of hs-CRP was 0.700 (p<0.001, 95% CI 0.638 to 0.762) with a cut-off value of 5.12 mg/L. The AUC of albumin was 0.671 (p<0.001, 95% CI 0.606 to 0.736) with a cut-off value of 37.15 g/L. CONCLUSIONS NT-pro-BNP levels were higher in bedridden patients with stroke than in non-bedridden patients. Decreased albumin and elevated hs-CRP levels were associated with high levels of NT-pro-BNP in bedridden patients. Further studies are needed to explore the risk stratification and potential treatments for elevated NT-pro-BNP in bedridden patients with stroke.
Collapse
Affiliation(s)
- Dongxia Li
- Shenzhen Second People's Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, China
| | - Mingchao Zhou
- Shenzhen Second People's Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, China
| | - Fubing Zha
- Shenzhen Second People's Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, China
| | - Jianjun Long
- Shenzhen Second People's Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, China
| | - Yulong Wang
- Shenzhen Second People's Hospital, The First Affiliated Hospital of Shenzhen University, Shenzhen, Guangdong, China
| |
Collapse
|
6
|
Mishra RK, Chavda VK, Moscote-Salazar LR, Atallah O, Das S, Janjua T, Maurya VP, Agrawal A. Systematic review and meta-analysis of studies comparing baseline D-dimer level in stroke patients with or without cancer: Strength of current evidence. J Neurosci Rural Pract 2024; 15:16-28. [PMID: 38476438 PMCID: PMC10927037 DOI: 10.25259/jnrp_379_2023] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Figures] [Journal Information] [Subscribe] [Scholar Register] [Received: 07/13/2023] [Accepted: 11/06/2023] [Indexed: 03/14/2024] Open
Abstract
Objectives D-dimer levels are increased in stroke and cancer. Cancer patients are at a higher risk of stroke. However, the evidence is unclear if high D-dimer in stroke patients can suggest the diagnosis of concomitant cancer or the development of stroke in a cancer patient. The objective is to assess the evidence available on the baseline D-dimer level in stroke patients with and without cancer. Materials and Methods We conducted the systematic review and meta-analysis using the Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines. We searched PUBMED, Cochrane, ScienceDirect, and Scopus for potentially eligible articles published till June 2023. All the review steps were iterative and done independently by two reviewers. The Newcastle-Ottawa scale tool was used to assess the quality of included studies for case control and cohort studies and the Agency for Healthcare Research and Quality tool for cross-sectional studies. The qualitative synthesis is presented narratively, and quantitative synthesis is shown in the forest plot using the random effects model. I2 of more than 60% was considered as high heterogeneity. Results The searches from all the databases yielded 495 articles. After the study selection process, six papers were found eligible for inclusion in the qualitative and quantitative synthesis. In the present systematic review, 2651 patients with ischemic infarcts are included of which 404 (13.97%) patients had active cancer while 2247 (86.02%) did not. The studies included were of high quality and low risk of bias. There were significantly higher baseline D-dimer levels in stroke patients with cancer than in non-cancer patients with a mean difference of 4.84 (3.07-6.60) P < 0.00001. Conclusion D-dimer is a simple and relatively non-expensive biomarker that is increased to significant levels in stroke patients, who have cancer and therefore may be a tool to predict through screening for active or occult cancer in stroke patients.
Collapse
Affiliation(s)
- Rakesh Kumar Mishra
- Department of Neurosurgery, Institute of Medical Sciences, Banaras Hindu University, Varanasi, Uttar Pradesh, India
| | - Vishal K. Chavda
- Department of Pathology, Stanford University School of Medicine, Stanford University Medical Center, CA-USA
| | | | - Oday Atallah
- Department of Neurosurgery, Hannover Medical School, Hannover, Germany
| | - Saikat Das
- Department of Radiation Oncology, All India Institute of Medical Sciences, Bhopal, Madhya Pradesh, India
| | - Tariq Janjua
- Department of Neurology, Regions Hospital, Saint Paul, Minnesota, United States
| | - Ved Prakash Maurya
- Department of Neurosurgery, Sanjay Gandhi Post Graduate Institute of Medical Sciences, Lucknow, Uttar Pradesh, India
| | - Amit Agrawal
- Department of Radiation Oncology, All India Institute of Medical Sciences, Bhopal, Madhya Pradesh, India
| |
Collapse
|
7
|
Chuang CY, Chou W, Chien TW, Jen TH. Trends and hotspots related to traditional and modern approaches on acupuncture for stroke: A bibliometric and visualization analysis. Medicine (Baltimore) 2023; 102:e35332. [PMID: 38050290 PMCID: PMC10695603 DOI: 10.1097/md.0000000000035332] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [MESH Headings] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 04/08/2023] [Accepted: 08/31/2023] [Indexed: 12/06/2023] Open
Abstract
BACKGROUND Acupuncture role in stroke treatment and post-stroke rehabilitation has garnered significant attention. However, there is a noticeable gap in bibliometric studies on this topic. Additionally, the precision and comprehensive methodology of cluster analysis remain underexplored. This research sought to introduce an innovative cluster analysis technique (called follower-leading clustering algorithm, FLCA) to evaluate global publications and trends related to acupuncture for stroke in the recent decade. METHODS Publications pertaining to acupuncture for stroke from 2013 to 2022 were sourced from the Web of Science Core Collection. For the assessment of publication attributes-including contributing countries/regions (e.g., US states, provinces, and major cities in China) in comparison to others, institutions, departments, authors, journals, and keywords-we employed bibliometric visualization tools combined with the FLCA algorithm. The analysis findings, inclusive of present research status, prospective trends, and 3 influential articles, were presented through bibliometrics with visualizations. RESULTS We identified 1050 publications from 92 countries/regions. An initial gradual rise in publication numbers was observed until 2019, marking a pivotal juncture. Prominent contributors in research, based on criteria such as regions, institutions, departments, and authors, were Beijing (China), Beijing Univ Chinese Med (China), the Department of Rehabilitation Medicine, and Lidian Chen (Fujian). The journal "Evid.-based Complement Altern" emerged as the most productive. The FLCA algorithm was effectively employed for co-word and author collaboration analyses. Furthermore, we detail the prevailing research status, anticipated trends, and 3 standout articles via bibliometrics. CONCLUSION Acupuncture for stroke presents a vast research avenue. It is imperative for scholars from various global regions and institutions to transcend academic boundaries to foster dialogue and cooperation. For forthcoming bibliometric investigations, the application of the FLCA algorithm for cluster analysis is advocated.
Collapse
Affiliation(s)
- Chao-Yu Chuang
- Department of Chinese Medicine, Chi Mei Medical Center, Tainan, Taiwan
| | - Willy Chou
- Department of Physical Medicine and Rehabilitation, Chiali Chi-Mei Hospital, Tainan 710, Taiwan
- Department of Physical Medicine and Rehabilitation, Chung San Medical University Hospital, Taichung 400, Taiwan
| | - Tsair-Wei Chien
- Department of Medical Research, Chi-Mei Medical Center, Tainan, Taiwan
| | - Tung-Hui Jen
- Department of Chinese Medicine, Chi Mei Medical Center, Tainan, Taiwan
- Department of Senior Welfare and Service, Southern Taiwan University of Science and Technology, Tainan, Taiwan
| |
Collapse
|