Development of a Logistic Regression Model for Prediction of Conversion Risk from Mild Cognitive Impairment (MCI) to Alzheimer´s Disease (AD)

G.S. Gonçalves¹, M.A. Bicalho², M.T. Cintra², C. Moreira³, R.C. Duarte¹, E. Reis³, D.R. Rios⁴, L.M. Dusse¹, K.B. Borges¹, M.D.G. Carvalho¹

¹Universidade Federal de Minas Gerais, Faculdade de Farmácia, Belo Horizonte, Brazil, ²Universidade Federal de Minas Gerais, Faculdade de Medicina, Belo Horizonte, Brazil, ³Universidade Federal de Minas Gerais, Instituto de Ciências Exatas, Belo Horizonte, Brazil, ⁴Universidade Federal de São João Del Rey, Campus Centro-Oeste Dona Lindu, Divinópolis, Brazil

Abstract Number: PB0527

Meeting: ISTH 2020 Congress

Theme: Diagnostics and OMICs » Laboratory Diagnostics

Background: MCI is an intermediate picture between normal cognition and dementia. Although high, the rate of conversion from MCI to dementia, particularly AD, has been poorly studied. Thus, a tool for this purpose is desirable.

Aims: To develop a logistic regression model based on hemostatic and vascular parameters to investigate the risk of converting from MCI to AD based on a 3-year follow-up.

Methods: 38 patients diagnosed with MCI had their levels of hemostatic parameters (generation of thrombin [TG], protein S [PS], tissue factor inhibitor [TFPI], von Willebrand factor [FvW], plasminogen activator inhibitor 1 [PAI-1] , D-dimer [D-Di], factor VIII [FVIII], ABO classification) and vascular injury markers (intercellular adhesion molecule 1 [ICAM-1], homocysteine [Hcy], platelet microparticles [PMPs], conventional and nonconventional lipid profile) determined. These data were used to build a logistic regression model (“R” software) to estimate the probability of a patient evolving from MCI to AD after three years of follow-up, when the same patients had a new cognitive assessment.

Results: Of the 38 patients with MCI, 13 (34.2%) progressed to AD and 25 (65.8%) did not over 3 years. Through the logistic regression model, the covariables age, sex, increased both generation of thrombin (Peak) and homocysteine (Hcy) were independently associated with the conversion from MCI to AD. Through these covariables, an algorithm was built to predict the risk of conversion from MCI to AD (sensitivity = 100%; specificity = 84%), as fig. 1.

Conclusions: For elderly male patients, it is recommended to monitor and control serum homocysteine levels and hemostatic status, whose alterations are potentially modifiable in order to reduce the chances of progressing from MCI to AD.
Acknowledgments: FAPEMIG, CNPq and CAPES.

To cite this abstract in AMA style:

Gonçalves GS, Bicalho MA, Cintra MT, Moreira C, Duarte RC, Reis E, Rios DR, Dusse LM, Borges KB, Carvalho MDG. Development of a Logistic Regression Model for Prediction of Conversion Risk from Mild Cognitive Impairment (MCI) to Alzheimer´s Disease (AD) [abstract]. Res Pract Thromb Haemost. 2020; 4 (Suppl 1). https://abstracts.isth.org/abstract/development-of-a-logistic-regression-model-for-prediction-of-conversion-risk-from-mild-cognitive-impairment-mci-to-alzheimers-disease-ad/. Accessed October 1, 2023.

« Back to ISTH 2020 Congress

ISTH Congress Abstracts - https://abstracts.isth.org/abstract/development-of-a-logistic-regression-model-for-prediction-of-conversion-risk-from-mild-cognitive-impairment-mci-to-alzheimers-disease-ad/