1
|
Hoffman H, Wood J, Cote JR, Jalal MS, Otite FO, Masoud HE, Gould GC. Development and Internal Validation of Machine Learning Models to Predict Mortality and Disability After Mechanical Thrombectomy for Acute Anterior Circulation Large Vessel Occlusion. World Neurosurg 2024; 182:e137-e154. [PMID: 38000670 DOI: 10.1016/j.wneu.2023.11.060] [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: 08/12/2023] [Revised: 11/14/2023] [Accepted: 11/15/2023] [Indexed: 11/26/2023]
Abstract
OBJECTIVE Mechanical thrombectomy (MT) improves outcomes in patients with LVO but many still experience mortality or severe disability. We sought to develop machine learning (ML) models that predict 90-day outcomes after MT for LVO. METHODS Consecutive patients who underwent MT for LVO between 2015-2021 at a Comprehensive Stroke Center were reviewed. Outcomes included 90-day favorable functional status (mRS 0-2), severe disability (mRS 4-6), and mortality. ML models were trained for each outcome using prethrombectomy data (pre) and with thrombectomy data (post). RESULTS Three hundred and fifty seven patients met the inclusion criteria. After model screening and hyperparameter tuning the top performing ML model for each outcome and timepoint was random forest (RF). Using only prethrombectomy features, the AUCs for the RFpre models were 0.73 (95% CI 0.62-0.85) for favorable functional status, 0.77 (95% CI 0.65-0.86) for severe disability, and 0.78 (95% CI 0.64-0.88) for mortality. All of these were better than a standard statistical model except for favorable functional status. Each RF model outperformed Pre, SPAN-100, THRIVE, and HIAT scores (P < 0.0001 for all). The most predictive features were premorbid mRS, age, and NIHSS. Incorporating MT data, the AUCs for the RFpost models were 0.80 (95% CI 0.67-0.90) for favorable functional status, 0.82 (95% CI 0.69-0.91) for severe disability, and 0.71 (95% CI 0.55-0.84) for mortality. CONCLUSIONS RF models accurately predicted 90-day outcomes after MT and performed better than standard statistical and clinical prediction models.
Collapse
Affiliation(s)
- Haydn Hoffman
- Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA.
| | - Jacob Wood
- Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA
| | - John R Cote
- Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA
| | - Muhammad S Jalal
- Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA
| | - Fadar O Otite
- Department of Neurology, State University of New York Upstate Medical University, Syracuse, New York, USA
| | - Hesham E Masoud
- Department of Neurology, State University of New York Upstate Medical University, Syracuse, New York, USA
| | - Grahame C Gould
- Department of Neurosurgery, State University of New York Upstate Medical University, Syracuse, New York, USA
| |
Collapse
|
2
|
Mitsuhashi T, Teranishi K, Tokugawa J, Mitsuhashi T, Hishii M, Oishi H. Prognostic Determinants of Anterior Large Vessel Occlusion in Acute Stroke in Elderly Patients. Geriatrics (Basel) 2024; 9:13. [PMID: 38247988 PMCID: PMC10801592 DOI: 10.3390/geriatrics9010013] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 09/25/2023] [Revised: 01/11/2024] [Accepted: 01/12/2024] [Indexed: 01/23/2024] Open
Abstract
This study investigated prognostic factors in elderly patients (80 years and older) undergoing mechanical thrombectomy (MT) for anterior circulation large vessel occlusion (LVO) in acute stroke treatment. Of 59 cases, 47.5% achieved a favorable outcome (mRS ≤ 3) at three months, with a mortality rate of 20.3%. Factors associated with better outcomes included younger age, lower admission National Institute of Health Stroke Scale (NIHSS) scores, lower N-terminal pro-brain natriuretic peptide (NT-proBNP) and D-dimer levels, the presence of the first pass effect (FPE), and successful recanalization. However, logistic regression showed that only lower admission NIHSS scores were significantly correlated with favorable outcomes. In addition, this study suggests that lower admission NT-proBNP and D-dimer levels could potentially serve as prognostic indicators for elderly LVO patients undergoing MT.
Collapse
Affiliation(s)
- Takashi Mitsuhashi
- Department of Neurosurgery, Juntendo University Nerima Hospital, Tokyo 177-8521, Japan
| | - Kohsuke Teranishi
- Department of Neurosurgery and Neuroendovascular Therapy, Juntendo University School of Medicine, Tokyo 113-8421, Japan
| | - Joji Tokugawa
- Department of Neurosurgery, Juntendo University Nerima Hospital, Tokyo 177-8521, Japan
| | - Takumi Mitsuhashi
- Department of Neurosurgery, Juntendo University Nerima Hospital, Tokyo 177-8521, Japan
| | - Makoto Hishii
- Department of Neurosurgery, Juntendo University Nerima Hospital, Tokyo 177-8521, Japan
| | - Hidenori Oishi
- Department of Neurosurgery and Neuroendovascular Therapy, Juntendo University School of Medicine, Tokyo 113-8421, Japan
| |
Collapse
|
3
|
Dunphy H, Garcia-Esperon C, Beom Hong J, Manoczki C, Wilson D, Lim Alvin Chew B, Beharry J, Bivard A, Hasnain MG, Krauss M, Collecutt W, Miteff F, Spratt N, Parsons MW, Alan Barber P, Ranta A, Fink JN, Wu TY. Endovascular thrombectomy for acute ischaemic stroke improves and maintains function in the very elderly: A multicentre propensity score matched analysis. Eur Stroke J 2022; 8:191-198. [PMID: 37021178 PMCID: PMC10069224 DOI: 10.1177/23969873221145778] [Citation(s) in RCA: 1] [Impact Index Per Article: 0.5] [Reference Citation Analysis] [Abstract] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 10/25/2022] [Accepted: 11/30/2022] [Indexed: 12/31/2022] Open
Abstract
Introduction: The very elderly (⩾80 years) are under-represented in randomised endovascular thrombectomy (EVT) clinical trials for acute ischaemic stroke. Rates of independent outcome in this group are generally lower than the less-old patients but the comparisons may be biased by an imbalance of non-age related baseline characteristics, treatment related metrics and medical risk factors. Patients and methods: We compared outcomes between very elderly (⩾80) and the less-old (<80 years) using retrospective data from consecutive patients receiving EVT from four comprehensive stroke centres in New Zealand and Australia. We used propensity score matching or multivariable logistic regression to account for confounders. Results: We included 600 patients (300 in each age cohort) after propensity score matching from an initial group of 1270 patients. The median baseline National Institutes of Health Stroke Scale was 16 (11–21), with 455 (75.8%) having symptom free pre-stroke independent function, and 268 (44.7%) receiving intravenous thrombolysis. Good functional outcome (90-day modified Rankin Scale 0–2) was achieved in 282 (46.8%), with very elderly patients having less proportion of good outcome compared to the less-old (118 (39.3%) vs 163 (54.3%), p < 0.01). There was no difference between the very elderly and the less-old in the proportion of patients who returned to baseline function at 90 days (56 (18.7%) vs 62 (20.7%), p = 0.54). All-cause 90-day mortality was higher in the very elderly (75 (25%) vs 49 (16.3%), p < 0.01), without a difference in symptomatic haemorrhage (very elderly 11 (3.7%) vs 6 (2.0%), p = 0.33). In the multivariable logistic regression models, the very elderly were significantly associated with reduced odds of good 90-day outcome (OR 0.49, 95% CI 0.34–0.69, p < 0.01) but not with return to baseline function (OR 0.85, 90% CI 0.54–1.29, p = 0.45) after adjusting for confounders. Conclusion: Endovascular thrombectomy can be successfully and safely performed in the very elderly. Despite an increase in all-cause 90-day mortality, selected very elderly patients are as likely as younger patients with similar baseline characteristics to return to baseline function following EVT.
Collapse
Affiliation(s)
- Harriette Dunphy
- Department of Neurology, Christchurch Hospital, Christchurch, New Zealand
| | - Carlos Garcia-Esperon
- Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia
- Heart and Stroke Program, Hunter Medical Research Institute, Newcastle, NSW, Australia
| | - Jae Beom Hong
- Department of Neurology, Auckland City Hospital, Auckland, New Zealand
| | - Csilla Manoczki
- Department of Neurology, Wellington Hospital, University of Otago, Wellington, New Zealand
| | - Duncan Wilson
- Department of Neurology, Christchurch Hospital, Christchurch, New Zealand
- New Zealand Brain Research Institute, Christchurch, New Zealand
| | | | - James Beharry
- Department of Neurology, Christchurch Hospital, Christchurch, New Zealand
| | - Andrew Bivard
- Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia
| | - Md Golam Hasnain
- Heart and Stroke Program, Hunter Medical Research Institute, Newcastle, NSW, Australia
| | - Martin Krauss
- Department of Radiology, Christchurch Hospital, New Zealand
| | | | - Ferdi Miteff
- Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia
| | - Neil Spratt
- Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia
- Heart and Stroke Program, Hunter Medical Research Institute, Newcastle, NSW, Australia
- The University of Newcastle, School of Biomedical Sciences and Pharmacy, Callaghan, NSW, Australia
| | - Mark W Parsons
- Department of Neurology, John Hunter Hospital, Newcastle, NSW, Australia
- University of New South Wales South Western Sydney Clinical Campus, The Ingham Institute for Applied Medical Research, Liverpool, NSW, Australia
| | - Peter Alan Barber
- Department of Neurology, Auckland City Hospital, Auckland, New Zealand
- Centre for Brain Research, University of Auckland, Auckland, New Zealand
| | - Annemarei Ranta
- Department of Neurology, Wellington Hospital, University of Otago, Wellington, New Zealand
- Department of Medicine, University of Otago, Wellington, New Zealand
| | - John N Fink
- Department of Neurology, Christchurch Hospital, Christchurch, New Zealand
| | - Teddy Y Wu
- Department of Neurology, Christchurch Hospital, Christchurch, New Zealand
- New Zealand Brain Research Institute, Christchurch, New Zealand
- Department of Medicine, University of Otago, Christchurch, New Zealand
| |
Collapse
|