Age-Stratified Prognostic Comparison of Three Chronotropic Incompetence Definitions in Patients with Suspected Coronary Artery Disease Undergoing Exercise Stress SPECT-MPI

Figures

• 

  Figure 1. Flowchart of patient selection.

• 

  Figure 2. Overlap Patterns and Clinical Event Rates Across the Three CI Definitions. (A) UpSet plot showing overlap patterns across CI-APMHR, CI-HRR, and CI-Blunted. Horizontal bars indicate CI-positive patient counts for each definition, and vertical bars indicate intersection sizes. A total of 159 patients (14.0%) were negative by all three definitions. (B) Crude MACE and abnormal perfusion rates, defined as TPD ≥5%, across six overlap patterns.

• 

  Figure 3. Kaplan–Meier Curves for MACE-Free Survival According to CI Status and Age Group. Kaplan–Meier curves show MACE-free survival stratified by CI status and age group. (A) shows CI-APMHR, (B) shows CI-HRR, and (C) shows CI-Blunted. In each panel, patients were classified into four groups: age <60 years without CI, age ≥60 years without CI, age <60 years with CI, and age ≥60 years with CI. Numbers at risk are shown below each panel at 0, 24, 48, and 72 months. Vertical tick marks on the curves indicate censored observations. Log-rank P values indicate overall comparisons among the four groups within each panel.

• 

  Figure 4. Adjusted Hazard Ratios for MACE According to the Three CI Definitions, Overall and Stratified by Age Group. Forest plot showing the associations of CI-APMHR, CI-HRR, and CI-Blunted with MACE in the overall cohort and by age group. The left panel summarizes sample size, events, event rate, adjusted HRs, 95% CIs, and P values; the right panel displays the corresponding HR estimates, with horizontal lines representing 95% CIs and the vertical dashed line indicating HR = 1.0. Overall estimates were derived from Model 2, adjusted for age, sex, BMI, hypertension, dyslipidemia, diabetes, current smoking, family history of CAD, left ventricular ejection fraction, TPD, METs, and ST-segment depression, with study center as a stratification variable. Age-stratified estimates were derived from Model 1, adjusted for sex and BMI. Age-by-CI interactions were not statistically significant for any definition.

• 

  Figure 5. Global Model Chi-square Improvement After Adding the Three CI Definitions to MPI Findings for MACE Prediction. Bar charts show global chi-square (χ2) values from nested Cox models for MACE prediction in all patients (Panel A; n = 1,135, 144 events), patients aged <60 years (Panel B; n = 822, 77 events), and patients aged ≥60 years (Panel C; n = 313, 67 events). For each CI definition, the base model included TPD and LVEF, and the extended model additionally included the corresponding CI definition. Numbers above the bars indicate global χ2 values, brackets show likelihood ratio test P values, and Δ values denote the increase in global χ2 after adding CI. Study center was included as a stratification variable in all models.