Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes

G. M. Podda, E. Grossi, T. Palmerini, M. Buscema, E. A. Femia, D. Della Riva, S. de Servi, P. Calabrò, F. Piscione, D. Maffeo, A. Toso, C. Palmieri, M. De Carlo, D. Capodanno, P. Genereux, M. Cattaneo

Research output: Contribution to journalArticle

3 Citations (Scopus)

Abstract

Background About 40% of clopidogrel-treated patients display high platelet reactivity (HPR). Alternative treatments of HPR patients, identified by platelet function tests, failed to improve their clinical outcomes in large randomized clinical trials. A more appealing alternative would be to identify HPR patients a priori, based on the presence/absence of demographic, clinical and genetic factors that affect PR. Due to the complexity and multiplicity of these factors, traditional statistical methods (TSMs) fail to identify a priori HPR patients accurately. The objective was to test whether Artificial Neural Networks (ANNs) or other Machine Learning Systems (MLSs), which use algorithms to extract model-like ‘structure’ information from a given set of data, accurately predict platelet reactivity (PR) in clopidogrel-treated patients. Methods A complete set of fifty-nine demographic, clinical, genetic data was available of 603 patients with acute coronary syndromes enrolled in the prospective GEPRESS study, which showed that HPR after 1 month of clopidogrel treatment independently predicted adverse cardiovascular events in patients with Syntax Score > 14. Data were analysed by MLSs and TSMs. ANNs identified more variables associated PR at 1 month, compared to TSMs. Results ANNs overall accuracy in predicting PR, although superior to other MLSs was 63% (95% CI 59–66). PR phenotype changed in both directions in 35% of patients across the 3 time points tested (before PCI, at hospital discharge and at 1 month). Conclusions Despite their ability to analyse very complex non-linear phenomena, ANNs or MLS were unable to predict PR accurately, likely because PR is a highly unstable phenotype.

Original languageEnglish (US)
Pages (from-to)60-65
Number of pages6
JournalInternational Journal of Cardiology
Volume240
DOIs
StatePublished - Aug 1 2017
Externally publishedYes

Fingerprint

clopidogrel
Acute Coronary Syndrome
Blood Platelets
Therapeutics
Platelet Function Tests
Demography

All Science Journal Classification (ASJC) codes

  • Cardiology and Cardiovascular Medicine

Cite this

Podda, G. M. ; Grossi, E. ; Palmerini, T. ; Buscema, M. ; Femia, E. A. ; Della Riva, D. ; de Servi, S. ; Calabrò, P. ; Piscione, F. ; Maffeo, D. ; Toso, A. ; Palmieri, C. ; De Carlo, M. ; Capodanno, D. ; Genereux, P. ; Cattaneo, M. / Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes. In: International Journal of Cardiology. 2017 ; Vol. 240. pp. 60-65.
@article{9ccb2bca16a6405ba8348af993729fff,
title = "Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes",
abstract = "Background About 40{\%} of clopidogrel-treated patients display high platelet reactivity (HPR). Alternative treatments of HPR patients, identified by platelet function tests, failed to improve their clinical outcomes in large randomized clinical trials. A more appealing alternative would be to identify HPR patients a priori, based on the presence/absence of demographic, clinical and genetic factors that affect PR. Due to the complexity and multiplicity of these factors, traditional statistical methods (TSMs) fail to identify a priori HPR patients accurately. The objective was to test whether Artificial Neural Networks (ANNs) or other Machine Learning Systems (MLSs), which use algorithms to extract model-like ‘structure’ information from a given set of data, accurately predict platelet reactivity (PR) in clopidogrel-treated patients. Methods A complete set of fifty-nine demographic, clinical, genetic data was available of 603 patients with acute coronary syndromes enrolled in the prospective GEPRESS study, which showed that HPR after 1 month of clopidogrel treatment independently predicted adverse cardiovascular events in patients with Syntax Score > 14. Data were analysed by MLSs and TSMs. ANNs identified more variables associated PR at 1 month, compared to TSMs. Results ANNs overall accuracy in predicting PR, although superior to other MLSs was 63{\%} (95{\%} CI 59–66). PR phenotype changed in both directions in 35{\%} of patients across the 3 time points tested (before PCI, at hospital discharge and at 1 month). Conclusions Despite their ability to analyse very complex non-linear phenomena, ANNs or MLS were unable to predict PR accurately, likely because PR is a highly unstable phenotype.",
author = "Podda, {G. M.} and E. Grossi and T. Palmerini and M. Buscema and Femia, {E. A.} and {Della Riva}, D. and {de Servi}, S. and P. Calabr{\`o} and F. Piscione and D. Maffeo and A. Toso and C. Palmieri and {De Carlo}, M. and D. Capodanno and P. Genereux and M. Cattaneo",
year = "2017",
month = "8",
day = "1",
doi = "10.1016/j.ijcard.2017.03.074",
language = "English (US)",
volume = "240",
pages = "60--65",
journal = "International Journal of Cardiology",
issn = "0167-5273",
publisher = "Elsevier Ireland Ltd",

}

Podda, GM, Grossi, E, Palmerini, T, Buscema, M, Femia, EA, Della Riva, D, de Servi, S, Calabrò, P, Piscione, F, Maffeo, D, Toso, A, Palmieri, C, De Carlo, M, Capodanno, D, Genereux, P & Cattaneo, M 2017, 'Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes', International Journal of Cardiology, vol. 240, pp. 60-65. https://doi.org/10.1016/j.ijcard.2017.03.074

Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes. / Podda, G. M.; Grossi, E.; Palmerini, T.; Buscema, M.; Femia, E. A.; Della Riva, D.; de Servi, S.; Calabrò, P.; Piscione, F.; Maffeo, D.; Toso, A.; Palmieri, C.; De Carlo, M.; Capodanno, D.; Genereux, P.; Cattaneo, M.

In: International Journal of Cardiology, Vol. 240, 01.08.2017, p. 60-65.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Prediction of high on-treatment platelet reactivity in clopidogrel-treated patients with acute coronary syndromes

AU - Podda, G. M.

AU - Grossi, E.

AU - Palmerini, T.

AU - Buscema, M.

AU - Femia, E. A.

AU - Della Riva, D.

AU - de Servi, S.

AU - Calabrò, P.

AU - Piscione, F.

AU - Maffeo, D.

AU - Toso, A.

AU - Palmieri, C.

AU - De Carlo, M.

AU - Capodanno, D.

AU - Genereux, P.

AU - Cattaneo, M.

PY - 2017/8/1

Y1 - 2017/8/1

N2 - Background About 40% of clopidogrel-treated patients display high platelet reactivity (HPR). Alternative treatments of HPR patients, identified by platelet function tests, failed to improve their clinical outcomes in large randomized clinical trials. A more appealing alternative would be to identify HPR patients a priori, based on the presence/absence of demographic, clinical and genetic factors that affect PR. Due to the complexity and multiplicity of these factors, traditional statistical methods (TSMs) fail to identify a priori HPR patients accurately. The objective was to test whether Artificial Neural Networks (ANNs) or other Machine Learning Systems (MLSs), which use algorithms to extract model-like ‘structure’ information from a given set of data, accurately predict platelet reactivity (PR) in clopidogrel-treated patients. Methods A complete set of fifty-nine demographic, clinical, genetic data was available of 603 patients with acute coronary syndromes enrolled in the prospective GEPRESS study, which showed that HPR after 1 month of clopidogrel treatment independently predicted adverse cardiovascular events in patients with Syntax Score > 14. Data were analysed by MLSs and TSMs. ANNs identified more variables associated PR at 1 month, compared to TSMs. Results ANNs overall accuracy in predicting PR, although superior to other MLSs was 63% (95% CI 59–66). PR phenotype changed in both directions in 35% of patients across the 3 time points tested (before PCI, at hospital discharge and at 1 month). Conclusions Despite their ability to analyse very complex non-linear phenomena, ANNs or MLS were unable to predict PR accurately, likely because PR is a highly unstable phenotype.

AB - Background About 40% of clopidogrel-treated patients display high platelet reactivity (HPR). Alternative treatments of HPR patients, identified by platelet function tests, failed to improve their clinical outcomes in large randomized clinical trials. A more appealing alternative would be to identify HPR patients a priori, based on the presence/absence of demographic, clinical and genetic factors that affect PR. Due to the complexity and multiplicity of these factors, traditional statistical methods (TSMs) fail to identify a priori HPR patients accurately. The objective was to test whether Artificial Neural Networks (ANNs) or other Machine Learning Systems (MLSs), which use algorithms to extract model-like ‘structure’ information from a given set of data, accurately predict platelet reactivity (PR) in clopidogrel-treated patients. Methods A complete set of fifty-nine demographic, clinical, genetic data was available of 603 patients with acute coronary syndromes enrolled in the prospective GEPRESS study, which showed that HPR after 1 month of clopidogrel treatment independently predicted adverse cardiovascular events in patients with Syntax Score > 14. Data were analysed by MLSs and TSMs. ANNs identified more variables associated PR at 1 month, compared to TSMs. Results ANNs overall accuracy in predicting PR, although superior to other MLSs was 63% (95% CI 59–66). PR phenotype changed in both directions in 35% of patients across the 3 time points tested (before PCI, at hospital discharge and at 1 month). Conclusions Despite their ability to analyse very complex non-linear phenomena, ANNs or MLS were unable to predict PR accurately, likely because PR is a highly unstable phenotype.

UR - http://www.scopus.com/inward/record.url?scp=85017181972&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85017181972&partnerID=8YFLogxK

U2 - 10.1016/j.ijcard.2017.03.074

DO - 10.1016/j.ijcard.2017.03.074

M3 - Article

C2 - 28343766

AN - SCOPUS:85017181972

VL - 240

SP - 60

EP - 65

JO - International Journal of Cardiology

JF - International Journal of Cardiology

SN - 0167-5273

ER -