Development and validation of a prognostic model predicting symptomatic hemorrhagic transformation in acute ischemic stroke at scale in the OHDSI network

English version

Autoři: Qiong Wang ^aff001; Jenna M. Reps ^aff003; Kristin Feeney Kostka ^aff003; Patrick B. Ryan ^aff003; Yuhui Zou ^aff007; Erica A. Voss ^aff003; Peter R. Rijnbeek ^aff003; RuiJun Chen ^aff003; Gowtham A. Rao ^aff003; Henry Morgan Stewart ^aff003; Andrew E. Williams ^aff003; Ross D. Williams ^aff003; Mui Van Zandt ^aff003; Thomas Falconer ^aff003; Margarita Fernandez-Chas ^aff003; Rohit Vashisht ^aff003; Stephen R. Pfohl ^aff003; Nigam H. Shah ^aff003; Suranga N. Kasthurirathne ^aff003; Seng Chan You ^aff003; Qing Jiang ^aff001; Christian Reich ^aff003; Yi Zhou ^aff015
Působiště autorů: Biomedical Engineering School, Sun Yat-Sen University, Guangzhou, China ^aff001; The Third Affiliated Hospital of Guangzhou Medical University, Guangzhou, China ^aff002; Observational Health Data Sciences and Informatics, New York, New York, United States of America ^aff003; Janssen Research and Development, Raritan, New Jersey, United States of America ^aff004; IQVIA, Durham, North Carolina, United States of America ^aff005; Department of Biomedical Informatics, Columbia University, New York, New York, United States of America ^aff006; Department of Neurosurgery, General Hospital of Southern Theatre Command, Guangzhou, China ^aff007; Department of Medical Informatics, Erasmus University Medical Center, Rotterdam, The Netherlands ^aff008; Department of Medicine, Weill Cornell Medical College, New York, New York, United States of America ^aff009; Tufts Medical Center, Institute for Clinical Research and Health Policy Studies, Boston, Massachusetts, United States of America ^aff010; Stanford Center for Biomedical Informatics Research, Stanford, California, United States of America ^aff011; Center for Biomedical Informatics, Regenstrief Institute, Indianapolis, Indiana, United States of America ^aff012; Department of Epidemiology, Indiana University Richard M. Fairbanks School of Public Health, Indianapolis, Indiana, United States of America ^aff013; Department of Biomedical informatics, Ajou University School of Medicine, Suwon, Korea ^aff014; Department of Biomedical Engineering, Zhongshan School of Medicine, Sun Yat-Sen University, Guangzhou, China ^aff015
Vyšlo v časopise: PLoS ONE 15(1)
Kategorie: Research Article
prolekare.web.journal.doi_sk: https://doi.org/10.1371/journal.pone.0226718

Souhrn

Background and purpose

Hemorrhagic transformation (HT) after cerebral infarction is a complex and multifactorial phenomenon in the acute stage of ischemic stroke, and often results in a poor prognosis. Thus, identifying risk factors and making an early prediction of HT in acute cerebral infarction contributes not only to the selections of therapeutic regimen but also, more importantly, to the improvement of prognosis of acute cerebral infarction. The purpose of this study was to develop and validate a model to predict a patient’s risk of HT within 30 days of initial ischemic stroke.

Methods

We utilized a retrospective multicenter observational cohort study design to develop a Lasso Logistic Regression prediction model with a large, US Electronic Health Record dataset which structured to the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM). To examine clinical transportability, the model was externally validated across 10 additional real-world healthcare datasets include EHR records for patients from America, Europe and Asia.

Results

In the database the model was developed, the target population cohort contained 621,178 patients with ischemic stroke, of which 5,624 patients had HT within 30 days following initial ischemic stroke. 612 risk predictors, including the distance a patient travels in an ambulance to get to care for a HT, were identified. An area under the receiver operating characteristic curve (AUC) of 0.75 was achieved in the internal validation of the risk model. External validation was performed across 10 databases totaling 5,515,508 patients with ischemic stroke, of which 86,401 patients had HT within 30 days following initial ischemic stroke. The mean external AUC was 0.71 and ranged between 0.60–0.78.

Conclusions

A HT prognostic predict model was developed with Lasso Logistic Regression based on routinely collected EMR data. This model can identify patients who have a higher risk of HT than the population average with an AUC of 0.78. It shows the OMOP CDM is an appropriate data standard for EMR secondary use in clinical multicenter research for prognostic prediction model development and validation. In the future, combining this model with clinical information systems will assist clinicians to make the right therapy decision for patients with acute ischemic stroke.

Klíčová slova:

Ischemic stroke – Hemorrhagic stroke – Health informatics – Medical risk factors – Hemorrhage – Electronic medical records – Infarction – Forecasting

Zdroje

1. Paciaroni M, Bandini F, Agnelli G, Tsivgoulis G, Yaghi S, Furie KL, et al. Hemorrhagic Transformation in Patients With Acute Ischemic Stroke and Atrial Fibrillation: Time to Initiation of Oral Anticoagulant Therapy and Outcomes. J Am Heart Assoc [Internet]. 2018 Nov 20 [cited 2019 Apr 30];7(22). Available from: https://www.ahajournals.org/doi/10.1161/JAHA.118.010133

2. Jaillard A, Cornu C, Durieux A, Moulin T, Boutitie F, Lees KR, et al. Hemorrhagic Transformation in Acute Ischemic Stroke. Stroke [Internet]. 1999 Jul [cited 2019 Apr 30];30(7):1326–32. Available from: https://www.ahajournals.org/doi/10.1161/01.STR.30.7.1326 10390303

3. Liu F, Zhao R, Feng X, Shi Y, Wu Y, reports XS-S, et al. Rivaroxaban does not influence hemorrhagic transformation in a diabetes ischemic stroke and endovascular thrombectomy model. nature.com [Internet]. [cited 2019 Apr 30]; https://www.nature.com/articles/s41598-018-25820-y

4. Xing Y, Jiang X, Yang Y, Xi G. Hemorrhagic Transformation Induced by Acute Hyperglycemia in a Rat Model of Transient Focal Ischemia. In 2011 [cited 2019 Apr 30]. p. 49–54. http://www.springerlink.com/index/10.1007/978-3-7091-0693-8_9

5. Lee S, Hong K, Lee J, Kim Y, … TS-C neurology and, 2018 undefined. Prediction of hemorrhagic transformation in patients with mild atrial fibrillation-associated stroke treated with early anticoagulation: post hoc analysis of the Triple AXEL. Elsevier [Internet]. [cited 2019 Apr 30]; https://www.sciencedirect.com/science/article/pii/S0303846718303457

6. Yang N, Lin M, Wang B, … WZ-ERM, 2016 undefined. Low level of low-density lipoprotein cholesterol is related with increased hemorrhagic transformation after acute ischemic cerebral infarction. europeanreview.org [Internet]. [cited 2019 May 1]; https://www.europeanreview.org/wp/wp-content/uploads/673-678.pdf

7. Nardi K, Leys D, Eusebi P, … CC-C, 2011 undefined. Influence of lipid profiles on the risk of hemorrhagic transformation after ischemic stroke: systematic review. karger.com [Internet]. [cited 2019 Apr 30]; https://www.karger.com/Article/Abstract/335014

8. Cordenier A, Smedt A De, … RB-AN, 2011 undefined. Pre-stroke use of statins on stroke outcome: a meta-analysis of observational studies. actaneurologica.be [Internet]. [cited 2019 Apr 30]; http://www.actaneurologica.be/pdfs/2011-4/02-Cordenier%20et%20al.pdf

9. Montaner J, Bustamante A, S G-M, 2016 undefined. Combination of thrombolysis and statins in acute stroke is safe: results of the STARS randomized trial (stroke treatment with acute reperfusion and simvastatin). Am Hear Assoc [Internet]. [cited 2019 Apr 30]; https://www.ahajournals.org/doi/abs/10.1161/strokeaha.116.014600

10. Bang O, Saver J, Liebeskind D, S S, 2007 undefined. Cholesterol level and symptomatic hemorrhagic transformation after ischemic stroke thrombolysis. AAN Enterp [Internet]. [cited 2019 Apr 30]; https://n.neurology.org/content/68/10/737.short

11. Levent ÖCEK, Derya GÜNER, İrem Fatma ULUDAĞ, Bedile İrem TİFTİKÇİOĞLU and YZ. Risk Factors for Hemorrhagic transformation in patients with acute middle cerebral artery infarction. ncbi.nlm.nih.gov [Internet]. [cited 2019 Apr 30]; https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5353105/

12. Xu X, Li C, Wan T, Gu X, Zhu W, Hao J, et al. Risk factors for hemorrhagic transformation after intravenous thrombolysis in acute cerebral infarction: a retrospective single-center study. Elsevier [Internet]. [cited 2019 Apr 30]; https://www.sciencedirect.com/science/article/pii/S1878875017301146

13. Landolfi A, Selvetella G, Cugino D, … G G journal of, 2016 undefined. Hemorrhagic transformation of acute ischemic stroke is limited in hypertensive patients with cardiac hypertrophy. Elsevier [Internet]. [cited 2019 Apr 30]; https://www.sciencedirect.com/science/article/pii/S0167527316310671

14. Kernan WN, Ovbiagele B, Black HR, Bravata DM, Chimowitz MI, Ezekowitz MD, et al. Guidelines for the prevention of stroke in patients with stroke and transient ischemic attack: a guideline for healthcare professionals from the American Heart Association/American Stroke Association. Stroke. 2014;45(7):2160–236. doi: 10.1161/STR.0000000000000024 24788967

15. Paciaroni M, Agnelli G, Caso V, Tsivgoulis G, Furie KL, Tadi P, et al. Prediction of Early Recurrent Thromboembolic Event and Major Bleeding in Patients With Acute Stroke and Atrial Fibrillation by a Risk Stratification Schema. Stroke [Internet]. 2017 Mar [cited 2019 Apr 30];48(3):726–32. Available from: https://www.ahajournals.org/doi/10.1161/STROKEAHA.116.015770 28183856

16. Gabay M. 21st century cures act. Hosp Pharm. 2017;52(4):264. doi: 10.1310/hpj5204-264 28515504

17. Observational Health Data Sciences and Informatics,OMOP Common Data Model [Internet]. [cited 2019 Apr 17]. https://github.com/OHDSI/CommonDataModel/

18. FitzHenry F, Resnic FS, Robbins SL, Denton J, Nookala L, Meeker D, et al. Creating a Common Data Model for Comparative Effectiveness with the Observational Medical Outcomes Partnership. Appl Clin Inform [Internet]. 2015 Dec 19 [cited 2019 May 1];06(03):536–47. Available from: http://www.thieme-connect.de/DOI/DOI?10.4338/ACI-2014-12-CR-0121

19. A H Jr, Bendixen B, Kappelle L, J B, 1993 undefined. Classification of subtype of acute ischemic stroke. Definitions for use in a multicenter clinical trial. TOAST. Trial of Org 10172 in Acute Stroke Treatment. Am Hear Assoc [Internet]. [cited 2019 May 1]; https://www.ahajournals.org/doi/pdf/10.1161/str.24.1.7678184

20. Schuemie M. WhiteRabbit [Internet]. github. [cited 2019 May 16]. https://github.com/OHDSI/WhiteRabbit

21. Clairblacketer. Definition and DDLs for the OMOP Common Data Model (CDM). 2019.

22. Reps J, Schuemie M, … MS-J of the, 2018 undefined. Design and implementation of a standardized framework to generate and evaluate patient-level prediction models using observational healthcare data. academic.oup.com [Internet]. [cited 2019 May 1]; https://academic.oup.com/jamia/article-abstract/25/8/969/4989437

23. Jie M, Collins G, Steyerberg E, … V J of clinical, 2019 undefined. A systematic review shows no performance benefit of machine learning over logistic regression for clinical prediction models. Elsevier [Internet]. [cited 2019 May 1]; https://www.sciencedirect.com/science/article/pii/S0895435618310813

24. Reps JM, Schuemie MJ, Suchard MA, Ryan PB RP. OHDSI Patient Level Prediction Package [Internet]. https://github.com/OHDSI/PatientLevelPrediction

25. Kassambara A. Machine Learning Essentials: Practical Guide in R [Internet]. 2018 [cited 2019 May 1]. https://books.google.com/books?hl=zh-CN&lr=&id=745QDwAAQBAJ&oi=fnd&pg=PP2&dq=Machine+Learning+Essentials:+Practical+Guide+in+R&ots=5DNtzSS2Pp&sig=vTw7xrJZfcAhBK-B-4Y6I8Mc1yAl

26. Rathmann W, Bongaerts B, Carius H-J, Kruppert S, Kostev K. Basic characteristics and representativeness of the German Disease Analyzer database. Int J Clin Pharmacol Ther. 2018;56(10):459–66. doi: 10.5414/CP203320  30168417

27. Tanaka S, Seto K, Kawakami K. Pharmacoepidemiology in Japan: medical databases and research achievements. J Pharm Heal care Sci. 2015;1(1):16.

28. Camm AJ, Fox KAA. Strengths and weaknesses of ‘real-world’studies involving non-vitamin K antagonist oral anticoagulants. Open Hear. 2018;5(1):e000788.

29. Jiang G, Kiefer RC, Sharma DK, Prud’hommeaux E, Solbrig HR. A consensus-based approach for harmonizing the OHDSI common data model with HL7 FHIR. Stud Health Technol Inform. 2017;245:887. 29295227