open access
The performance of triglyceride to high-density lipoprotein cholesterol ratio in acute coronary syndromes using a diagnostic decision tree
open access
Abstract
Background. Modern modeling techniques, including decision trees, may potentially provide accurate prediction and classification of outcomes, and support the process of clinical decision making. The objective of our study was to evaluate the performance of triglyceride to high-density lipoprotein (TG:HDL-C) ratio in acute coronary syndromes (ACS) presented using a decision tree analysis.
Methods. The initial study group consisted of 220 consecutive patients admitted to hospital within the first 6 hours from the onset of chest pain. All these patients met clinical criteria of ACS and were compared with 116 healthy volunteers in a case-control study. Serum was assayed on admission for cardiac troponin I, C-reactive protein, apolipoproteins ApoAI and ApoB, and lipid parameters. Atherogenic lipid ratios: TC:HDL:C, LDL-C:HDL:C and TG:HDL-C were calculated.
Results. ACS patients showed almost twice as high median values of TG:HDL-C as controls [2.77 (1.88–4.08) vs. 1.47 (0.99–2.08); p < 0.0001]. The TG:HDL-C ratio was significantly related to the positive history of coronary artery disease, age and lipid parameters, except for LDL-C. The TG:HDL-C ratio was, after age, the most powerful independent predictor and classifier of the occurrence of ACS with the optimal cutoff being 2.28. The performance of TG:HDL-C was superior to other lipid parameters and ratios, and enabled identification of the additional 6 subjects with ACS.
Conclusion. The TG:HDL-C ratio is a useful risk marker of the ACS occurrence. Further prospective studies are needed to confirm our findings and clarify the interaction between TG and HDL-C concentrations in ACS patients.
Abstract
Background. Modern modeling techniques, including decision trees, may potentially provide accurate prediction and classification of outcomes, and support the process of clinical decision making. The objective of our study was to evaluate the performance of triglyceride to high-density lipoprotein (TG:HDL-C) ratio in acute coronary syndromes (ACS) presented using a decision tree analysis.
Methods. The initial study group consisted of 220 consecutive patients admitted to hospital within the first 6 hours from the onset of chest pain. All these patients met clinical criteria of ACS and were compared with 116 healthy volunteers in a case-control study. Serum was assayed on admission for cardiac troponin I, C-reactive protein, apolipoproteins ApoAI and ApoB, and lipid parameters. Atherogenic lipid ratios: TC:HDL:C, LDL-C:HDL:C and TG:HDL-C were calculated.
Results. ACS patients showed almost twice as high median values of TG:HDL-C as controls [2.77 (1.88–4.08) vs. 1.47 (0.99–2.08); p < 0.0001]. The TG:HDL-C ratio was significantly related to the positive history of coronary artery disease, age and lipid parameters, except for LDL-C. The TG:HDL-C ratio was, after age, the most powerful independent predictor and classifier of the occurrence of ACS with the optimal cutoff being 2.28. The performance of TG:HDL-C was superior to other lipid parameters and ratios, and enabled identification of the additional 6 subjects with ACS.
Conclusion. The TG:HDL-C ratio is a useful risk marker of the ACS occurrence. Further prospective studies are needed to confirm our findings and clarify the interaction between TG and HDL-C concentrations in ACS patients.
Keywords
acute coronary syndrome, triglyceride to high-density lipoprotein cholesterol ratio, classification matrix, decision tree
About this articleTitle
The performance of triglyceride to high-density lipoprotein cholesterol ratio in acute coronary syndromes using a diagnostic decision tree
Journal
Issue
Article type
Original article
Pages
13-19
Published online
2015-03-31
Page views
580
Article views/downloads
1395
Bibliographic record
Folia Medica Copernicana 2015;3(1):13-19.
Keywords
acute coronary syndrome
triglyceride to high-density lipoprotein cholesterol ratio
classification matrix
decision tree
Authors
Magdalena Krintus
Marek Koziński
Magdalena Kuligowska-Prusińska
Ewa Laskowska
Ewa Janiszewska
Jacek Kubica
Grażyna Odrowąż-Sypniewska