Premium
New D ecision‐ T ree M odel for D efining the R isk of R eproductive F ailure
Author(s) -
RamosMedina Rocío,
GarcíaSegovia Áurea,
León Juan A.,
Alonso Bárbara,
TejeraAlhambra Marta,
Gil Juana,
Caputo Juan D.,
Seyfferth Ansgar,
Aguarón Ángel,
Vicente Ángeles,
Ordoñez Daniel,
Alonso Jorge,
de Albornoz Elena Carrillo,
Carbone Javier,
Caballero Pedro,
FernandezCruz Eduardo,
OrtizQuintana Luis,
SánchezRamón Silvia
Publication year - 2013
Publication title -
american journal of reproductive immunology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.071
H-Index - 97
eISSN - 1600-0897
pISSN - 1046-7408
DOI - 10.1111/aji.12098
Subject(s) - decision tree , multivariate statistics , flow cytometry , observational study , multivariate analysis , pregnancy , biology , tree (set theory) , gynecology , immunology , medicine , oncology , computer science , machine learning , genetics , mathematics , mathematical analysis
Problem Natural killer ( NK ) cells play a key role in embryo implantation and pregnancy success, whereas blood and uterine NK expansions have been involved in the pathophysiology of reproductive failure ( RF ). Our main goal was to design in a large observational study a tree‐model decision for interpretation of risk factors for RF . Methods of study A hierarchical multivariate decision model based on a classification and regression tree was developed. NK and NKT ‐like cell subsets were analyzed by flow cytometry. Results By multivariate analysis, blood NK cells expansion was an independent risk factor for RF (both recurrent miscarriages and implantation failures). We propose a new decision‐tree model for the risk interpretation of women with RF based on a combination of main risk factors. Conclusions Women with age above 35 years and >13% CD 56 + CD 16 + NK cells showed the highest risk of further pregnancy loss (100%).