Effects of Acute Exercise and 12‐Week High‐Intensity Interval Training on Inflammatory Biomarkers in Stable Coronary Artery Disease: A Randomized Controlled Trial

Figures

• 

  Figure 1. Flowchart and study design. The top panel outlines participant enrollment, randomization, and retention, including reasons for exclusion and discontinuation. The bottom panel presents a schematic overview of the study protocol, illustrating the acute exercise test, the subsequent 12‐week HIIT or standard care allocation, and the timing of assessments. CAD indicates coronary artery disease; HIIT, high‐intensity interval training; IFN‐γ, interferon‐γ; IL, interleukin; and TNF‐α, tumor necrosis factor‐α.

• 

  Figure 2. Short‐term changes in inflammatory cytokines following strenuous exercise. Panels show plasma concentrations of (A) interferon‐γ, (B) tumor necrosis factor‐α, (C) interleukin‐2, (D) interleukin‐6, and (E) interleukin‐10 at baseline (pre), immediately after exercise (≤5 minutes), and 2 hours after an incremental exercise test to exhaustion in patients with coronary artery disease (n=168 at each time point). All cytokine data were analyzed on the log₂‐transformed scale and back‐transformed to the original units for presentation; solid horizontal lines represent geometric means with 95% CIs. Points represent individual participants. Sidak‐adjusted pairwise comparisons are reported as exact P values.

• 

  Figure 3. Short‐term changes in C‐reactive protein and leukocyte counts following strenuous exercise. Panels show (A) plasma C‐reactive protein (n=167 at each time point) and (B) leukocyte counts (n=164 at each time point) at baseline (pre), immediately after exercise (≤5 minutes), and 2 hours after an incremental exercise test to exhaustion in patients with coronary artery disease. C‐reactive protein data were analyzed on the log₂‐transformed scale and back‐transformed to the original units for presentation; solid horizontal lines represent group‐specific geometric means with 95% CIs. Leukocyte counts met model assumptions without transformation and are therefore presented as arithmetic means with 95% CIs. Points represent individual participants. Sidak‐adjusted pairwise comparisons are reported as exact P values.

• 

  Figure 4. Inflammatory cytokine responses to 12‐week exercise training versus standard care. Panels show plasma concentrations of (A) interferon‐γ, (B) tumor necrosis factor‐α, (C) interleukin‐2, (D) interleukin‐6, and (E) interleukin‐10 at baseline (pre), midpoint (week 6), and after intervention (week 12) following a 12‐week high‐intensity interval training program or standard care in patients with coronary artery disease. Sample sizes (identical across panels; pre/week 6/week 12): exercise n=64/56/64; standard care n=78/69/77. All cytokine data were analyzed on the log₂‐transformed scale and back‐transformed to the original units for presentation; solid horizontal lines represent group‐specific geometric means with 95% CIs. Points represent individual participants. P values for the Time×Group interaction are shown in each panel, with Sidak‐adjusted pairwise comparisons reported as exact P values.

• 

  Figure 5. C‐reactive protein and leukocyte count responses to 12‐week exercise training versus standard care. Panels show (A) plasma C‐reactive protein and (B) leukocyte counts at baseline (pre), midpoint (week 6), and after intervention (week 12) following a 12‐week high‐intensity interval training program or standard care in patients with coronary artery disease. Sample sizes (pre/week 6/week 12): exercise n=63/64/63 for C‐reactive protein and leukocyte count; standard care n=78/78/78 for C‐reactive protein and n=77/78/78 for leukocyte count. C‐reactive protein data were analyzed on the log₂‐transformed scale and back‐transformed to the original units for presentation; solid horizontal lines represent group‐specific geometric means with 95% CIs. Leukocyte counts met model assumptions without transformation and are therefore presented as arithmetic means with 95% CIs. Points represent individual participants. P values for the Time×Group interaction are shown in each panel, and Sidak‐adjusted pairwise comparisons are reported as exact P values.