Do Heart Rate Reserve Zones Match Real Metabolic Thresholds?

Heart Rate Reserve (HRR) formulas, calculated via the Karvonen method, provide a rough population-level estimate of metabolic work rates, but they often fail to align precisely with individual ventilatory and blood lactate thresholds [2, 17]. While group averages show predictable patterns, individual threshold heart rates vary widely across fitness levels, sexes, and health statuses [8, 11, 19]. Relying strictly on standard HRR percentages frequently leads athletes and clinical populations to train in higher or lower metabolic domains than intended [8, 14, 17].

The Three-Domain Metabolic Model

Endurance exercise physiology divides work output into three primary metabolic intensity domains separated by two physiological boundary points [19]:

1. Moderate Domain: Work rates below the first ventilatory threshold (VT1) or first lactate threshold (LT1) [19]. In this domain, blood lactate remains near baseline (often defined as an increase of less than 0.5 mM above baseline), and fat oxidation is elevated [5, 17].
2. Heavy Domain: Work rates between the first metabolic threshold (LT1/VT1) and the second metabolic threshold (LT2/VT2, critical power, or critical velocity) [19]. Blood lactate rises above baseline (frequently heuristicized as an increase of $\ge$1.5 mM above baseline) but stabilizes over continuous efforts, approaching the maximal lactate steady state (MLSS) [3, 17].
3. Severe Domain: Work rates between the second metabolic threshold and maximal oxygen uptake ($\dot{V}{\text{O}}_2\text{max}$), characterized by an unsustainable rise in lactate and rapid fatigue [3, 19].

Elite endurance athletes commonly allocate roughly 80% of their training volume to the moderate domain (Zone 2 and below) to optimize fat metabolism, build mitochondrial efficiency, and spare muscle glycogen [5].

Average HRR Values vs. Individual Variation

The Karvonen formula calculates target heart rate by adding a fraction of the heart rate reserve to resting heart rate: $\text{Target HR}={\text{HR}}_{\text{rest}}+\text{intensity fraction}\times ({\text{HR}}_{\text{max}}-{\text{HR}}_{\text{rest}})$ [2]. Accounting for resting heart rate helps reflect baseline parasympathetic tone, yet standard percentage zones often diverge from direct physiological measurements [2].

In controlled cycling trials, group averages suggest reasonable mapping between HRR and metabolic markers. In a cohort of 33 aerobically trained male cyclists (mean $\dot{V}{\text{O}}_2\text{max}$: 62.1 $\pm$ 4.6 ml·kg⁻¹·min⁻¹), physiological landmarks corresponded to the following average values [3]:

• VT1: 184 $\pm$ 36 W (66 $\pm$ 9% HRR) [3]
• MLSS (subgroup of $n=14$): 256 $\pm$ 31 W [3]
• VT2: 298 $\pm$ 36 W (88 $\pm$ 6% HRR) [3]
• $\dot{V}{\text{O}}_2\text{max}$: 390 $\pm$ 34 W (100% HRR) [3]

Based on graded exercise tests (25 W·min⁻¹ ramps), five training zones emerged: R0 (53–62% HRR), R1 (62–71% HRR), R2 (74–86% HRR), R3 (86–99% HRR), and R3+ (100% HRR) [3].

However, broad meta-analytic data demonstrate that individual threshold locations scatter dramatically around these mean values [17]. In a meta-analysis covering 25 studies and 412 runners, individual heart rates at physiological thresholds spanned massive ranges [17]:

• LT1 Range: 65% to 90% HRR (or 69% to 94% ${\text{HR}}_{\text{max}}$) [17]
• LT2 Range: 73% to 95% HRR (or 80% to 98% ${\text{HR}}_{\text{max}}$) [17]

Because of this spread, prescribing an arbitrary target such as 75% HRR might place one runner safely below LT1 in the moderate domain, while placing another well above LT1 in the heavy domain [17, 19]. To achieve 95% certainty of exercising strictly below LT1 without laboratory testing, an athlete must remain below 65% HRR (or below 70% ${\text{HR}}_{\text{max}}$) [17].

Sources of Mismatch Between HR Formulas and Metabolic Thresholds

Several factors explain why formula-based HR reserve models fail to match real-time blood lactate and gas-exchange thresholds:

1. Inherent ${\text{HR}}_{\text{max}}$ Estimation Error

Standard age-predicted maximal heart rate equations carry estimation errors spanning $\pm$10 to 12 bpm [2]. In 1999, Meyer and colleagues observed that prescribing exercise at a fixed 85% of ${\text{HR}}_{\text{max}}$ resulted in individual exercise intensities scattering from 87% to 116% of athletes' individually measured anaerobic thresholds [2].

2. Aerobic Fitness and Non-Linear Threshold Shifts

The relationship between aerobic capacity and threshold position is not uniform across fitness categories [11]. In a pooled study of 863 participants, oxygen uptake at VT relative to oxygen uptake reserve (VT%$\dot{V}{\text{O}}_2\text{R}$) displayed a U-shaped curve [11]. The nadir occurred at approximately 43% $\dot{V}{\text{O}}_2\text{R}$ in individuals with a $\dot{V}{\text{O}}_2\text{peak}$ near 40 ml·kg⁻¹·min⁻¹, whereas average VT%$\dot{V}{\text{O}}_2\text{R}$ rose to roughly 75% in participants with the lowest and highest $\dot{V}{\text{O}}_2\text{peak}$ values [11].

3. Sex Differences

Fixed target thresholds do not behave identically between sexes [19]. In a cohort of 165 recreational runners, female runners exhibited significantly higher relative heart rates, $\dot{V}{\text{O}}_2$, and running speeds at LT1 compared to male runners ($p<0.001$), highlighting that uniform HR targets misrepresent metabolic strain between sexes [19].

4. Clinical Populations and Chronotropic Incompetence

The divergence between fixed HR target guidelines and physiological thresholds is especially pronounced in clinical cohorts [8, 14]. In a study of 350 individuals, standard guideline-based intensity domains misclassified VT1 as high intensity instead of moderate intensity in 67.1% (previous guidelines) and 79.6% (2020 ESC guidelines) of cardiac patients, whereas correspondence was 90% in healthy sedentary individuals and 93.9% in competitive athletes [8].

Similarly, in 18 chronic heart failure patients, target heart rates derived from 40–70% HRR were significantly higher than measured thresholds, risking the prescription of heavy or severe exercise above the ventilatory threshold or respiratory compensation point [14]. Measured ${\text{HR}}_{\text{max}}$ in these patients was 12% lower than age-predicted values, even after applying standard beta-blocker deductions [14].

Practical Alternatives and Anchors

When direct metabolic cart testing (measuring VT1/VT2) or blood lactate analysis is unavailable, several alternatives offer calibration against real physiological transitions:

• Perceived Exertion (RPE): Across diverse fitness levels ranging from low-fit cohorts to highly trained individuals, VT consistently occurs around a mean Borg RPE of 12.5 $\pm$ 0.93, despite wide differences in absolute work rate [11].
• Heart Rate Variability (HRV): Non-linear HRV analysis using detrended fluctuation analysis (DFA $\alpha 1$) tracks changes in fractal correlation properties, which decline from ~1.5 at low work rates down toward 0.75 near VT1/LT1 [22, 23]. In elite triathletes and male running cohorts, the DFA $\alpha 1=0.75$ threshold showed close agreement with LT1 and VT1, demonstrating mean differences of only 1.7 to 2.0 bpm [22, 23].
• Performance Anchors: In recreational runners, fixed speed-derived estimations yield lower error magnitudes for identifying LT1 and LT2 compared to heart rate or RPE alone, though estimating rather than directly measuring maximum speed increases threshold error [19].