Abstract
s study examined the influence of physiological parameters on peak velocity (Vpeak) and of kinematic variables on running economy (RE) during an outdoor incremental VAM-EVAL test completed by eleven national-level triathletes. Maximal oxygen uptake (VO2max), ventilatory thresholds, RE, and minimum muscle oxygen saturation (SmO2min) were obtained with a portable gas analyzer and near-infrared spectroscopy (NIRS), while cadence, stride length, vertical oscillation, and contact time were recorded with a foot- mounted inertial sensor. Multiple linear regression showed that VO2max and SmO2min to- gether accounted for 86% of the variance in Vpeak (VO2max: r = 0.76; SmO2min:r =−0.68 ), whereas RE at 16 km·h −1 displayed only a moderate association (r = 0.54). Links between RE and kinematic metrics were negligible to weak (r≤0.38). These findings confirm VO2max as the primary determinant of Vpeak and suggest that SmO2min can be used as a complementary, non-invasive marker of endurance capacity in triathletes, measur- able in the field with portable NIRS. Additionally, inter-individual differences in cadence, stride length, vertical oscillation, and contact time suggest that kinematic adjustments are not universally effective but rather highly
weak (r≤0.38). These findings confirm VO2max as the primary determinant of Vpeak and suggest that SmO2min can be used as a complementary, non-invasive marker of endurance capacity in triathletes, measur- able in the field with portable NIRS. Additionally, inter-individual differences in cadence, stride length, vertical oscillation, and contact time suggest that kinematic adjustments are not universally effective but rather highly individualized, with their impact on RE likely depending on each athlete’s specific characteristics. Keywords:VO2max; SmO2min; NIRS; Vpeak; incremental exercise testing 1. Introduction Running performance is influenced by physiological [1], biomechanical [2], and psy- chological factors [3]. From a physiological perspective, the determinants of performance are maximal oxygen consumption (VO2max), running economy (RE), and physiological thresholds [1]; this triad of performance determinants is related to the oxidative capacity of muscle tissue [4]. VO2max marks the upper limit of ATP resynthesis through oxidative phosphorylation metabolism [5], RE is defined as the volume of oxygen an athlete needs to cover a distance at a given speed [1], and metabolic thresholds influence the duration for which a % VO2max can be sustained [6]. The oxidative capacity of muscle fibers refers to the mitochondria’s ability to supply ATP using oxygen, which, in turn, provides the necessary energy to sustain activity [7]. VO2max is determined by the different components of the oxygen transport system [8,9] and is primarily limited by central factors such as cardiac output, blood volume, and hemoglobin mass [10]. However, it has also been shown to correlate with mitochondrial density [11] and the oxidative capacity of muscle tissue [12]. Sports2025,13, 316 https://doi.org/10.3390/sports13090316
Sports2025,13, 316 2 of 15 Physiological thresholds are influenced by the oxidative capacity of muscle tissue [1], associ- ated with the proliferation of certain enzymes at the mitochondrial level [13]. Regarding RE, numerous studies reference the factors that affect this performance determinant, suggesting that RE reflects the interaction of various physiological, biomechanical, anthropometric, and kinematic factors [14,15]. It has been shown that certain physiological aspects impact RE, such as the oxidative capacity [11], capillarization, and myoglobin [7] of skeletal muscle. Another aspect to consider that may affect RE is muscle fiber composition, with suggestions that better RE may be associated with a higher percentage of type 1 fibers [14], as these are more efficient at frequencies between 60 and 120 rpm [1,7]. This range encompasses the cadences used in training and competition in running, despite interindividual differences associated with this kinematic parameter [16]. Muscle oxygen saturation (SmO2) provides detailed insights into the balance between oxygen supply and demand by measuring changes in oxygenated and deoxygenated hemoglobin and myoglobin concentrations in muscle tissue [17]. While other metrics, such as deoxygenated hemoglobin and myoglobin (deoxy[heme]), are frequently em- ployed as proxies for oxygen extraction [18], SmO2is comparatively less influenced by blood volume changes [19,20]. It can also be measured non-invasively during sports ac- tivity using continuous-wave near-infrared spectroscopy (NIRS) devices [21], reported on a 0–100% scale. Feldmann et al. [ more recent studies have demonstrated the application and external validation of SmO2 dynamics in field protocols through comparisons with in-dependent physiological mark- ers such as lactate thresholds and EMG activity [23,24]. Nevertheless, NIRS signal can be affected by various factors, such as adipose tissue thickness (ATT) [25], muscle tissue heterogeneity [ or skin blood flow/volume [26], which tend to elevate SmO2values, since the measurement is influenced by the oxygenation status of hemoglobin in less metabol- ically active tissues, such as the skin or adipose tissue. Muscle oxidative capacity, which connects the three main performance determinants, has traditionally been limited to inva- sive or costly assessments such as biopsy or P MRS, but during the last 20 years, advances have been made. First,
the measurement is influenced by the oxygenation status of hemoglobin in less metabol- ically active tissues, such as the skin or adipose tissue. Muscle oxidative capacity, which connects the three main performance determinants, has traditionally been limited to inva- sive or costly assessments such as biopsy or P MRS, but during the last 20 years, advances have been made. First, Motobe et al. [27] developed a non-invasive approach using NIRS to infer muscle oxidative capacity based on the muscle oxygen consumption recovery rate constant (k). This approach was modified almost 10 years later [28], and lately, Pilotto et al. [29] have developed protocols for estimating muscle oxidative capacity and muscle diffusing capacity. The protocols of intermittent occlusions represent a significant advancement compared to the invasive methods implemented some time ago, but they are not efficient options for monitoring the evolution of the oxidative capacity of muscle tissue over time in the context of sports training. In this context, minimum muscle oxygen satu- ration (SmO2min) has been observed to remain consistent within a session upon reaching exhaustion and can predict task failure during high intensity efforts [30]. This marker has been associated with improved endurance [31] and maximal incremental performance [11]. Furthermore, the capacity to deoxygenate correlated significantly with VO2peak [32,33], and it appears that the capacity to reach a lower SmO2min correlates with performance [34–36]. Coaches commonly assess two key running intensities during track tests: peak velocity (Vpeak) and maximal aerobic speed (MAS) [37]. Vpeak represents the highest velocity achieved during a test, whereas MAS refers to the minimal speed required to elicit VO2max [38]. These two measures should not be viewed as interchangeable representations of a single concept [39]. Unfortunately, this distinction is often overlooked, and Vpeak is frequently used as a surrogate for MAS [40–42]. Both velocities result from the interaction of physiological performance determinants specific to each athlete [1,7]. However, Vpeak is likely to involve a greater contribution from the glycolytic energy system than MAS [39,43]. Despite their physiological differences, both Vpeak and MAS are closely
used as a surrogate for MAS [40–42]. Both velocities result from the interaction of physiological performance determinants specific to each athlete [1,7]. However, Vpeak is likely to involve a greater contribution from the glycolytic energy system than MAS [39,43]. Despite their physiological differences, both Vpeak and MAS are closely
Sports2025,13, 316 3 of 15 associated with running performance [39,44]. Specifically, regarding triathlon, Vpeak has shown a very strong correlation with overall triathlon performance [45], whereas, to our knowledge, the correlation between triathlon performance and MAS has not been directly assessed. Additionally, Vpeak may be considered a more practical performance marker, as it does not require the simultaneous measurement of oxygen consumption needed to determine MAS. Establishing a link between the mentioned physiological elements (VO2max, VT1, VT2, RE and SmO2min) and Vpeak can be beneficial for coaches and athletes to iden- tify the physiological qualities that may influence performance. Additionally, linking the kinematic aspects of running to RE may enhance our understanding of how movement variables influence the oxygen cost of running. Wiecha et al. [46] found that velocity at VT2 and VO2max were the strongest predictors of Vpeak. The sample consisted of 4001 recreational endurance athletes, and the derived equations predicted Vpeak accurately in this group. Rather than applying these equations to other populations, such as higher- level athletes or triathletes, we should study those cohorts to determine which variables most strongly predict Vpeak and develop or recalibrate models accordingly. In parallel, there is a need to investigate whether NIRS-derived metrics such as SmO2min predict performance and Vpeak in these higher performance cohorts. Evidence on the relationship between spatiotemporal parameters and RE is mixed: Pizzuto et al. [47] found no significant relationships between spatiotemporal parameters and RE in recreational runners, whereas Leite et al. [48] reported that higher cadence and greater vertical oscillation were associated with increased oxygen cost; for each additional 1 step·min −1 and 1 mm of vertical oscilla- tion, VO2rose by 0.09 and 0.10 mL·kg −1 · min −1 , respectively. Because both studies were conducted on treadmills, the relationship between RE and spatiotemporal variables should also be examined in real-world, overground contexts such as an athletics track. Taken together, these points underscore the need for ecologically valid, tightly controlled studies and for investigations in specific performance cohorts, such as triathletes, to generate findings that are directly actionable for coaches. To address these gaps and extend
on treadmills, the relationship between RE and spatiotemporal variables should also be examined in real-world, overground contexts such as an athletics track. Taken together, these points underscore the need for ecologically valid, tightly controlled studies and for investigations in specific performance cohorts, such as triathletes, to generate findings that are directly actionable for coaches. To address these gaps and extend evidence beyond treadmill-based studies in recreational runners, the primary objective of this study is to evaluate how VO2max, ventilatory thresholds, RE, and SmO2min influence Vpeak in national-level triathletes. We hypothesize that higher VO2max and lower SmO2min will be associated with higher Vpeak. The secondary objective is to analyze the impact of cadence, stride length, vertical oscillation, and contact time on RE in this population. We hypothesize that these kinematic variables will not show significant associations with RE. 2. Materials and Methods 2.1. Study Design Observational study evaluating the influence of physiological parameters (VO2max, ventilatory thresholds, SmO2min and RE) on Vpeak in runners. Additionally, the study examines the effect of kinematic parameters (cadence, stride length, vertical oscillation, and contact time) on RE. The physiological parameters were selected due to their estab- lished relationship with endurance performance, while the kinematic parameters provide insights into movement efficiency. Vpeak was chosen as the performance metric because it is a critical determinant of athletic performance in endurance sports. 2.2. Participants Eleven national-level triathletes voluntarily participated in the study. The inclusion criteria were as follows: participants had to be active triathletes with a federation license who competed in national triathlon events. The exclusion criteria included any participant with a cardiac condition or those who were currently injured or had suffered an injury
Sports2025,13, 316 4 of 15 in the last two months. Additionally, the exclusion of individuals with an ATT > 7 mm was necessary to minimize interference with NIRS signal quality [25]. ATT was calculated as 0.5×the mean skinfold thickness. Although the sample size was limited to eleven national- level triathletes, this homogeneous group was chosen to reduce inter-individual variability and focus on high-performance athletes. A priori sample size considerations indicated that detecting a medium effect (f 2 = 0.15) in a multiple regression with six predictors would require approximatelyn≈99, whereas a large effect (f 2 = 0.35) would requiren≈47. Given the restricted availability of national-level triathletes, we recruited the maximum feasible sample (n= 11). All participants signed an informed consent form, and the study was approved by the Clinical Research Ethics Committee of the Catalan Sports Administration (026/CEICGC/2023). 2.3. Instruments Muscle oxygenation was monitored using NIRS. The Moxy monitor (Fortiori Design LLC, Fort Collins, CO, USA) was used to measure SmO2, allowing an understanding of the relationship between oxygen supply and utilization in the analyzed muscle. The Moxy device uses four wavelengths of near-infrared light (680, 720, 760, and 800 nm), with the sensor equipped with a single LED and two detectors located at distances of 12.5 mm and 25 mm from the light source. The device was placed on the belly of the right vastus lateralis, precisely halfway between the greater trochanter and the lateral epicondyle of the femur [49]. To maintain the sensor’s position relative to the skin, it was secured with waterproof adhesive tape. Hair in the area was shaved, and participants were instructed to avoid applying moisturizers on the testing day. The Moxy sensor was secured to the leg using the Moxy Light Shield, a flexible polyurethane skirt that fits around the sensor. This accessory blocks ambient light, particularly sunlight passing through the tissue, which could interfere with the measurements. The Moxy Light Shield was fixed in place using waterproof adhesive tape to ensure stability during movement and to maintain effective light shielding. The Moxy device was operated in default mode, cycling through four wavelengths 80
polyurethane skirt that fits around the sensor. This accessory blocks ambient light, particularly sunlight passing through the tissue, which could interfere with the measurements. The Moxy Light Shield was fixed in place using waterproof adhesive tape to ensure stability during movement and to maintain effective light shielding. The Moxy device was operated in default mode, cycling through four wavelengths 80 times every 2 s and averaging the readings to produce an output rate of 0.5 Hz. SmO2data were averaged at ten second intervals and the SmO2min during the incremental running test was identified from a single data point of a ten-second mean value. Participants wore a dead-space mask (Hans Rudolph Inc., Shawnee, KS, USA) equipped with a bidirectional 28 mm digital turbine. Oxygen (O2) and carbon dioxide (CO2) concentrations were measured using a Galvanic fuel cell O2sensor and digital infrared CO2sensor, respectively, as part of the Cosmed K5 Wearable Metabolic System (Cosmed S.r.l, Albano Laziale, Rome, Italy). The gas analyzer was calibrated according to the COSMED instructions: a room air calibration, a flow meter calibration with a 3 L syringe, a scrubber calibration, a reference gas calibration using a known gas (16% O2, 5% CO2) and a delay calibration for the breath-by-breath mode. Breath-by-breath data for oxygen consumption (VO2) and carbon dioxide production (VCO2) were recorded and then averaged. VO2data were measured on a breath-by-breath basis and averaged at ten second intervals. Ventilatory thresholds were assessed by two independent researchers. VT1 was determined using the following criteria: increase in VE/VO2and end tidal partial pressure of O2(PETO2) without concomitant increase in VE/VCO2. VT2 was determined using the following criteria: increase in VE/VO2and VE/VCO2with a concomitant decrease in end tidal partial pressure of CO2(PETCO2) [50]. If the time values identified by the two researchers differed by 40 s or less, their values were averaged. In cases where the difference exceeded 40 s, a third independent researcher evaluated the ventilatory thresholds. The third researcher’s time value was then compared to those of the initial two researchers. If the third researcher’s value was within 40 s of either initial researcher’s
two researchers differed by 40 s or less, their values were averaged. In cases where the difference exceeded 40 s, a third independent researcher evaluated the ventilatory thresholds. The third researcher’s time value was then compared to those of the initial two researchers. If the third researcher’s value was within 40 s of either initial researcher’s
Sports2025,13, 316 5 of 15 value, the two closest time values were averaged to determine the final ventilatory threshold. This method is similar to the one applied by Okawara et al. [51]. VO2max was defined as the highest VO2value maintained for ten seconds. RE was calculated at two stages during the incremental test (at the 12 km·h −1 stage and at the 16 km·h −1 stage) by averaging VO2 during the last thirty seconds of the respective stages [52]. Cadence (CAD), vertical oscillation (VO), contact time (CT), and stride length (SL) were recorded using a Stryd inertial sensor (Stryd Inc., Boulder, CO, USA) placed on the right shoe of the athletes. Stryd is a carbon-fiber-reinforced power meter that attaches to the shoe, weighs 9.1 g, and uses a six-axis inertial motion sensor (three-axis gyroscope and three-axis accelerometer). This device has been considered suitable for evaluating kinematic parameters during running [53]. The following kinematic parameters, cadence, vertical oscillation, contact time, and stride length, were recorded throughout the run- ning incremental test and averaged over 10 s intervals. The kinematic parameters were calculated at two stages during the incremental test (at the 12 km·h −1 stage and at the 16 km·h −1 stage) by averaging data over the last thirty seconds of the respective stages to compare the results with RE data from the same intervals. 2.4. Experimental Procedure Participants were required to attend on one occasion to complete a single experimental session. The session for each one of the participants took place on a 400 m athletics track in Barcelona from February to March with a temperature and humidity of 21.5±5.8 ◦ C and 72.8±16.2%, respectively. In the 24 h prior to the test, participants were asked to refrain from engaging in high-volume and/or high-intensity sessions, whether swimming, cycling, or running. Athletes were advised to consume a carbohydrate-rich diet during the 48 h leading up to the test. Additionally, they were instructed to ensure that their last meal was consumed approximately 3 h before the test. Participants wore a loose-fitting running shirt and non-compressive shorts. The thickness of the skinfolds was
in high-volume and/or high-intensity sessions, whether swimming, cycling, or running. Athletes were advised to consume a carbohydrate-rich diet during the 48 h leading up to the test. Additionally, they were instructed to ensure that their last meal was consumed approximately 3 h before the test. Participants wore a loose-fitting running shirt and non-compressive shorts. The thickness of the skinfolds was measured on the vastus lateralis using a skinfold caliper (Baty Int., Sheffield, South Yorkshire, UK). A heart rate monitor (Polar H10, Polar Electro, Kempele, Finland) was placed on the participants. Finally, participants were fitted with a portable gas analyzer worn on their back, with the mask size selected to fit the individual’s face size. All participants underwent a VAM EVAL test, a commonly utilized method for assessing aerobic capacity in running, derived from the French term “vitesse aérobie maximale” (VAM), which translates to “maximal aerobic speed,” and EVALuation. This test starts at 8 km·h −1 for two minutes, and after that, speed increases by 0.5 km·h −1 per minute [44]. Cones were placed every 20 m for the participants to regulate their running pace to the audible signal. The researchers, distributed along the track, visually checked that the subjects maintained the imposed pace. The test was terminated when the participants voluntarily stopped due to exhaustion or when they failed to reach the marked cone twice in succession. 2.5. Statistical Analysis The VO2, heart rate (HR), SmO2, CAD, VO, CT, and SL data were filtered to remove outliers and averaged every 10 s. Data analysis was performed using Microsoft Excel (version 16.81 24011420), Cosmed Omnia (version 2.3), and Moxy Settings App (version 1.5.5). First, the normality of the data was assessed using the Shapiro–Wilk test at a significance level ofp< 0.05. To determine the association between Vpeak and the fol- lowing physiological parameters: VO2max, VT1, VT2, RE12, RE16, and SmO2min, Pearson correlation coefficients (r) were calculated. The procedure was repeated to assess the associ- ation between RE and the following kinematic parameters: CAD, VO, CT, and SL. Correla- tion was classified as negligible (0.00–0.30), weak (0.30–0.50), moderate (0.50–0.70), strong
Description
The study investigates physiological and kinematic factors affecting peak velocity in triathletes.