Discriminant analysis for prognosis of thrombophilia development in patients with stage VD CKD
2018
OBJECTIVE: Introduction: Chronic kidney disease (CKD) is a critical medical and social problem. Every year 7& of such patients require renal replacement therapy associated with one of the most life threatening complications - disturbed hemostasis and development of thrombophilic conditions which are typical of complicated comorbid diseases. Prediction of potential thrombogenesis in that category of patients requires multidimensional approach, including comprehensive laboratory investigation of hemostasis on the one hand, and creation of mathematical model based on thrombophilia predictors on the other. The aim: To create the prognostic model of thrombophilia development in patients with stage VD CKD, treated by long term hemodialysis, using discriminant function analysis and biochemical variables of hemostasis. PATIENTS AND METHODS: Materials and methods: Of 12 hemostatic parameters (factors of influence), studied in 85 patients with stage VD CKD, two groups of variables were formed - pre-thrombosis (soluble fibrin (sF), D-dimer (D-d), protein C (PC)) and post-thrombosis (D-d, H (maximal end turbidity of plasma clot) and pC). RESULTS: Results: Linear discriminant functions (two models), based on two groups of variables, demonstrated rather high coincidence rate (92.2÷93.5&) in group distribution. The patients referred to those with thrombotic complications, 100& matched the prospective (for a year) study. CONCLUSION: Conclusions: By the results of correlation analysis, the most significant molecular biochemical hemostatic parameters (sF, D-d, pC, H) were determined. Using definite biochemical parameters, discriminant models complementing each other and having high prediction (р <0.0001) of thrombophilic states in 92.2-93.5& of cases, were developed for patients with stage VD CKD, treated by long term hemodialysis.
Keywords:
- Correction
- Source
- Cite
- Save
- Machine Reading By IdeaReader
0
References
0
Citations
NaN
KQI