QT variability improves risk stratification in patients with dilated cardiomyopathy

Area of Science:

• Cardiology
• Biomedical Engineering
• Data Science

Background:

• Systolic and diastolic blood pressure variability (BPV) and segmented Poincare plot analysis (SPPA) aid risk stratification in dilated cardiomyopathy (DCM).
• Traditional heart rate variability analysis is insufficient for risk stratification in DCM patients.
• There is a need for improved risk stratification methods in DCM.

Purpose of the Study:

• To enhance risk stratification in DCM patients using a multivariate technique incorporating QT variability (QTV).
• To evaluate the effectiveness of combining BPV, SPPA, and QTV for predicting adverse outcomes in DCM.

Main Methods:

• Recruited 56 low-risk and 13 high-risk DCM patients.
• Applied various BPV and QTV methods, including symbolic dynamics and segmented Poincare plot analysis.
• Utilized multivariate discriminate analysis (DA) with different combinations of indices.

Main Results:

• A multivariate DA model using one BPV index (DBP(Shannon)) and one QTV index (QTV(log)) achieved an AUC of 92%, sensitivity of 92.3%, and specificity of 89.3%.
• An electrocardiogram-only analysis combining SPPA and QTV yielded an AUC of 88.3%.
• A three-index model (clinical, BPV, QTV) reached an AUC of 95.7%, with 100% sensitivity and 85.7% specificity.

Conclusions:

• QT variability (QTV) analysis in a multivariate approach significantly improves risk stratification for DCM patients.
• The combination of QTV with BPV and clinical data enhances the prediction of sudden cardiac death risk in DCM.
• Multivariate analysis incorporating QTV offers a promising tool for identifying high-risk DCM patients.