This study is a randomized, parallel-group clinical trial with assessor and statistical analyst blinding, conducted at a single center. Participants with stable ischemic heart disease enrolled in a cardiac rehabilitation program will be randomly allocated to one of two aerobic exercise interventions: long extensive continuous aerobic training (zone 2) or long extensive aerobic interval training (zone 3). Both groups will participate in supervised exercise sessions twice weekly for a period of 8 weeks, integrated into the standard cardiac rehabilitation program.

Assessments will be performed at baseline, immediately after the intervention period (8 weeks), and after a 4-week follow-up period without supervised training to evaluate potential detraining effects. The primary outcome is heart rate-power decoupling during a prolonged submaximal cycling exercise test, used as an indicator of cardiovascular efficiency. Secondary outcomes include heart rate response during submaximal exercise, exercise tolerance, heart rate recovery, blood pressure, functional capacity, quality of life, and adherence to the training program.

The objective of this study is to compare the effects of long extensive aerobic interval training versus long extensive continuous training on heart rate-power decoupling and related cardiovascular and functional outcomes in patients with ischemic heart disease undergoing cardiac rehabilitation.

BACKGROUND AND RATIONALE Cardiac rehabilitation (CR) based on supervised aerobic exercise is a well-established intervention that reduces cardiovascular morbidity and mortality, improves functional capacity, and enhances quality of life in patients with ischemic heart disease. Two aerobic training modalities are routinely used within CR programs and recommended by international guidelines: moderate-intensity continuous training (MICT, zone 2), characterized by prolonged steady-state exercise below the first ventilatory threshold (65-75% HRmax); and long extensive interval training (LEIT, zone 3), which alternates prolonged work intervals between the two ventilatory thresholds (75-85% HRmax) with active recovery periods. Both modalities are standard components of the Phase II CR program at Hospital Universitario 12 de Octubre (HU12O), Madrid, where this trial is conducted.

Despite the established benefits of both training modalities, no randomized controlled trial has directly compared their effects on heart rate-power decoupling during submaximal exercise - a physiological marker of cardiovascular efficiency that remains underexplored in the CR setting.

HEART RATE-POWER DECOUPLING: PHYSIOLOGICAL BASIS During constant-load cycling exercise, an efficient cardiovascular system maintains a relatively stable heart rate (HR) over time. When HR progressively increases despite unchanged mechanical power output, this phenomenon - termed HR-power decoupling or cardiovascular drift - reflects a declining stroke volume compensated by chronotropic response, resulting in greater cardiac work per unit of mechanical output. A higher decoupling index indicates lower cardiorespiratory efficiency during submaximal exercise.

HR-power decoupling can be assessed using standard CR equipment (calibrated cycle ergometer and validated chest-strap HR monitor), without maximal testing or specialized technology, making it a clinically applicable and safe measure in this population. Reducing HR-power decoupling after training would indicate that the heart operates with lower hemodynamic load for the same mechanical demand - a meaningful adaptation in terms of myocardial oxygen consumption, exercise tolerance, and long-term cardiovascular prognosis.

STUDY DESIGN RATIONALE A parallel-group design was chosen over a crossover design to avoid carry-over effects, as aerobic training adaptations may persist for weeks after program completion. Assessor and statistical analyst blinding is implemented to minimize detection and analysis bias; blinding of the treating physiotherapist is not feasible given the nature of the intervention.

A detraining follow-up assessment at week 12 (4 weeks after supervised training ends) is included to evaluate the durability of adaptations and whether potential between-group differences in central versus peripheral cardiovascular adaptations result in differential rates of detraining.

INTERVENTIONS

All participants follow the standard HU12O CR protocol, which includes an identical resistance and mobility training block (35-40 min, <50% 1RM) and warm-up and cool-down phases in every session. The sole differential component is the aerobic training block (approximately 40 min of active aerobic exercise):

• Continuous training group (Zone 2): 40 minutes of continuous cycling at 65-75% of individual HRmax, maintained at steady state. Target HR range is determined individually from the baseline incremental test and adjusted weekly to account for training-induced adaptations.
• Interval training group (Zone 3): 8 work intervals of 3 minutes at 75-85% HRmax, each followed by 2 minutes of active recovery below 65-70% HRmax (total block: 40 min), equating total training volume to the continuous group. Work intensity targets the zone between the two ventilatory thresholds.

Intensity is controlled by dual monitoring of HR and rating of perceived exertion (Borg scale 6-20; target RPE 11-13 for zone 2 and 13-16 for zone 3). In participants with blunted chronotropic response (e.g., beta-blocker therapy), RPE serves as the primary intensity reference. Power output (watts) is progressively adjusted across sessions to maintain target HR ranges as cardiorespiratory fitness improves.

PRIMARY OUTCOME ASSESSMENT The HR-power decoupling index is assessed via a 20-minute constant-load submaximal cycling test at an individualized fixed power output. HR is recorded continuously via validated chest-strap sensor. The decoupling index is calculated as the relative percentage change in mean HR between the first third (minutes 0-6) and the last third (minutes 14-20) of the test: [(HR₁₄-₂₀ - HR₀-₆) / HR₀-₆] × 100. Higher values indicate greater cardiovascular drift and lower efficiency. The minimal clinically important difference (MCID) is set at 0.12 units. Immediately following the 20-minute test, a 5-minute passive recovery period is used to assess post-exercise HR recovery (HRR₁, HRR₂, HRR₅).

SAMPLE SIZE Sample size was calculated using G*Power 3.1.9.7 for a two-factor repeated-measures ANOVA (2 groups × 3 time points), with 80% power, α = 0.05, correlation between repeated measures of 0.50, and an expected effect size of f = 0.51 (Cohen d = 0.83; MCID = 0.15 units; SD ≈ 0.18). This yielded a minimum of 34 participants. Accounting for an estimated 15% dropout rate, the target sample is 40 participants (20 per group).

STATISTICAL ANALYSIS The primary analysis will use a two-way repeated-measures ANOVA (group × time), with the group × time interaction as the primary estimator. If sphericity is violated, the Greenhouse-Geisser correction will be applied. If data are not missing completely at random, linear mixed-effects models will be used. The primary analysis follows the intention-to-treat principle; a per-protocol sensitivity analysis will include participants who completed ≥80% of sessions and all three assessments. Missing data will be handled by multiple imputation. Effect sizes will be reported as partial eta-squared (η²p) for ANOVA and Cohen's d for pairwise comparisons. Statistical significance is set at p < 0.05 (two-tailed), with 95% confidence intervals.

ETHICS The protocol has been submitted for approval to the Ethics Committee for Research with Medicinal Products (CEIm) of Hospital Universitario 12 de Octubre. The study will be conducted in accordance with the Declaration of Helsinki and applicable Spanish biomedical research legislation. All participants will provide written informed consent prior to enrollment.

Inclusion Criteria:

• Confirmed diagnosis of cardiovascular disease requiring cardiac rehabilitation, with medical indication for a supervised exercise training program.

• Prognostic risk classified as low or moderate according to the clinical risk stratification established in the unit, considering:

  • Killip class I or II (absence or mild heart failure)¹.
  • Absence of recent acute myocardial infarction (< 1 month) or presence of a stable previous infarction.
  • Stable and well-controlled cardiovascular symptoms.
  • Physical capacity demonstrated in functional tests ≥ 5 METs, preferably > 7 METs for low-risk patients.

• Age between 40 and 75 years.

• Ability to perform physical exercise without absolute contraindications.

• Written informed consent and willingness to attend the scheduled sessions twice per week for 8 weeks.

Exclusion Criteria:

• High prognostic risk, defined by:

  • Killip class > III (moderate to severe heart failure).
  • Recent acute myocardial infarction (< 1 month) with clinical instability.
  • Active anginal symptoms or decompensated heart failure.
  • Physical capacity < 5 METs on functional testing, indicating severely reduced exercise capacity.

• Presence of complex or unstable arrhythmias contraindicating vigorous physical exercise.

• Concomitant diseases that may limit participation or compromise safety (e.g., advanced pulmonary disease, severe renal failure, severe neuromuscular disorders).

• Absolute medical contraindications to maximal strength or interval training (e.g., aortic aneurysm, uncontrolled hypertension).

• Participation in another clinical trial or concurrent rehabilitation program.

• Inability to comply with attendance requirements or to follow the intervention protocol.

In addition to blinding the outcome assessor and the statistical analyst, no masking will be applied to other parties involved in the study. Due to the practical nature of the intervention, the personnel responsible for administering the training (physiotherapists) and the participants themselves will be aware of group allocation. Blinding of the intervention providers or the participants is not planned.

Thus, the only blinding implemented will be that of the assessor and the analyst, ensuring objectivity in outcome measurement and data analysis, and minimizing potential bias in these processes.