Abstract
ners achieve forward locomotion through diverse techniques. However, understanding the behavior of the involved kinematical variables remains incomplete, particularly when running overground and along an intensity spectrum. We aimed to characterize the biomechanical and physiological adaptations while running at low, moderate, heavy and severe intensities. Ten middle- and long-distance runners completed an incremental intermittent protocol of 800 m steps until exhaustion (1 km·h −1 velocity increments and 30 s intervals) on an outdoor track field. Biomechanical data were captured using two high-resolution video cameras, and linear and angular kinematic variables were analyzed. With the intensity rise, a decrease in stride, step and contact times ([0.70–0.65],[0.35–0.33]and [0.42–0.37] s) and an increase in stride length and frequency and flight time ([3.13–3.52] m, [1.43–1.52] Hz and [0.28–0.29] s;p< 0.05) were observed, together with an increase in oxygen uptake and blood lactate concentrations ([54.7–67.6] mL·kg −1 · min −1 and[3.1–10.2] mmol·L −1 ). A more flexed hip at initial contact and toe-off (152.02–149.36] and [165.70–163.64]) and knee at initial contact ([162.64–159.57];p< 0.05) were also observed. A consistent
and frequency and flight time ([3.13–3.52] m, [1.43–1.52] Hz and [0.28–0.29] s;p< 0.05) were observed, together with an increase in oxygen uptake and blood lactate concentrations ([54.7–67.6] mL·kg −1 · min −1 and[3.1–10.2] mmol·L −1 ). A more flexed hip at initial contact and toe-off (152.02–149.36] and [165.70–163.64]) and knee at initial contact ([162.64–159.57];p< 0.05) were also observed. A consistent gait pattern along each protocol step was exhibited, with minor changes without practical significance. Runners are constantly adapting their gait pattern, reflected in both biomechanical and physiological responses, both of which should be considered for better characterization. Keywords:gait analysis; linear kinematics; angular kinematics; intensity domains; overground running 1. Introduction Running is a popular physical activity characterized by continuous, cyclic and fairly unconstrained movements. A specific combination of stride length and frequency is adopted by runners (mainly subconsciously) for each velocity, resulting in a large inter- individual variation in stride patterns and lower-limb kinematics [1,2]. The changes in running patterns with the velocity rise seek to optimize movement at slower paces and maximize power output at sprinting velocities. These kinematical modifications of the segmental movements seek to improve several aspects, including metabolic energy expenditure, tissue stress, muscle power and fatigue [3]. The quantitative analysis of running kinematics is a powerful tool that has been evolv- ing over the last decades, aiming to better understand the influence of technique in perfor- mance, identify injury risk factors and/or facilitate recovery through more accurate and automated systems [4]. Despite the current feasibility of motion analysis using automatic, non-invasive and markerless methodologies [5], their utilization remains predominantly confined to laboratory conditions under controlled environments. Conversely, the indirect techniques of gait analysis based on video recording and planar analysis with manual Sensors2024,24, 7526.
Sensors2024,24, 7526 2 of 11 digitation have also progressed, allowing athlete monitoring in ecological conditions, such as overground running on a track field instead of running on a treadmill [4,6,7]. Notwithstanding the controversial and inconsistent biomechanical and physiological responses when comparing overground vs. treadmill running [8,9], the high association and applicability between field data and training and competing contexts are undeniable. In fact, during overground running, the body moves over the supporting lower limbs while when exercising on a treadmill, the running belt moves the supporting lower limbs at a constant velocity, leading to changes in propulsive forces between conditions [10]. Also, differences in surface stiffness, air resistance, comfort and altered velocity perception should be taken into account [8]. It is well known that running involves a complex interaction between both mechanical and physiological mechanisms, since there is a constant processing of different types of information coming from external and internal sources that are related to the movements involved and the corresponding consequences [3]. Changes in running velocity will alter the adopted biomechanical pattern that, consequently, will modify the pulmonary and ventilatory response [3,11]. The current study aimed to comprehensively characterize the biomechanical and physiological adaptations while running overground at low, moderate, heavy and severe exercise intensities. It was hypothesized that there would be a decrease in temporal and an increase in frequency- and distance-related variables, concurrently with a rise in the physiological variables, along the intensity spectrum. In addition, we sought to assess the consistency of the running gait pattern within each intensity domain, hypothesizing that the linear and angular kinematic responses would be maintained from the beginning to the end of each exertion. 2. Materials and Methods 2.1. Participants Ten middle- and long-distance male runners volunteered to participate (age: 26.8±5.7 years, body mass: 68.2±8.2 kg and body height: 180.0±6.5 cm). They were involved in running training practice for 9.9±3.9 years and in 12.9±2.2 h of weekly running training. Participants were recruited through direct contact and selected if there was no history of cardiorespiratory and physical diseases or injuries within the previous six months and if they had more
(age: 26.8±5.7 years, body mass: 68.2±8.2 kg and body height: 180.0±6.5 cm). They were involved in running training practice for 9.9±3.9 years and in 12.9±2.2 h of weekly running training. Participants were recruited through direct contact and selected if there was no history of cardiorespiratory and physical diseases or injuries within the previous six months and if they had more than two years of running training background. The current research was approved by the local ethics committee and the participants were informed about the purpose, benefits and any associated risks (providing their written individual consent for participation in accordance with the Helsinki Declaration). 2.2. Experimental Protocol After an individualized warm-up of ~20 min at low intensity, participants performed an incremental intermittent running protocol of 800 m steps until exhaustion (with 1 km·h −1 increments and 30 s rest intervals in-between) on a 400 m outdoor track field [6,12]. The velocities for each runner’s last step were established based on their best individual 3000 m performance at the time of the data collection or according to their own experience in previous tests. Afterward, six velocity increments were subtracted to define the subsequent step paces, with velocities being controlled for each step by audio feedback emitted at every 100 m where fluorescent cones were placed [12,13]. Before the evaluation moment, the acromion, iliac crest, greater trochanter, lateral femoral condyle, lateral malleolus and second metatarsal head of the runners’ left body side were manually marked [13] using black skin landmarks. The lower-limb kinematical data were recorded from the runners’ left sagittal plane using two high-resolution video cameras previously calibrated (1920×1080 pixels; GoPro HERO6 Black, CA, USA) at a sampling rate of 120 Hz, fixed on tripods and positioned 3 m from the middle of the 100–200 running track section and 4.5 m from the first lane. During the protocol, pulmonary gas exchange and ventilation were continuously measured breath-by-breath by a telemetric portable gas analyzer (K5, Cosmed, Rome, Italy)
of the 100–200 running track section and 4.5 m from the first lane. During the protocol, pulmonary gas exchange and ventilation were continuously measured breath-by-breath by a telemetric portable gas analyzer (K5, Cosmed, Rome, Italy)
Sensors2024,24, 7526 3 of 11 previously calibrated according to manufacturer instructions (using ambient air against known concentrations [16% O2and 5% CO2] and a 3 L syringe) and placed on runners’ backs, near the body center of mass. Heart rate was also continuously recorded using a Polar Vantage NV (Polar Electro Oy, Kemple, Finland) that telemetrically emitted to the portable gas analyzer unit [14]. Capillary blood samples for lactate concentration analysis were collected from the fingertip at rest, during the 30 s intervals and at 1, 3, 5 and/or 7 min at the end of the protocol (until obtaining maximal values) using a portable analyzer (Lactate Pro2; Arkay Inc., Kyoto, Japan) [15]. 2.3. Data Analysis For each running protocol step, a total of twelve strides were analyzed frame-by-frame using two-dimensional motion analysis software (Kinovea, version 0.8.27, Boston, MA, USA) to determine the stride time (time between the left foot touchdown and the next left foot touchdown), step time (time between the left foot touchdown and the right foot touchdown), stride length (mean velocity/stride frequency) and frequency (1/stride time) and contact and flight times (times from initial contact to toe-off and from toe-off to the initial contact of the same foot, respectively) as linear kinematic variables [1]. The flight and contact times normalized to the stride duration were also included (flight time/stride time and contact time/stride time). Based on the marked anatomic points described above, the hip, knee and ankle joint angles (both at initial contact and toe-off moments) were calculated as angular kinematical variables (Figure) [ 13]. Body marks were carefully detected, and the mean of three measurements was used. A 0.98 intraclass correlation coefficient was verified.Sensors 2024, 24, 7526 3 of 10 During the protocol, pulmonary gas exchange and ventilation were continuously measured breath-by-breath by a telemetric portable gas analyzer (K5, Cosmed, Rome, It- aly) previously calibrated according to manufacturer instructions (using ambient air against known concentrations [16% O2 and 5% CO2] and a 3 L syringe) and placed on runners’ backs, near the body center of mass. Heart rate was also continuously recorded using a Polar Vantage NV (Polar
were continuously measured breath-by-breath by a telemetric portable gas analyzer (K5, Cosmed, Rome, It- aly) previously calibrated according to manufacturer instructions (using ambient air against known concentrations [16% O2 and 5% CO2] and a 3 L syringe) and placed on runners’ backs, near the body center of mass. Heart rate was also continuously recorded using a Polar Vantage NV (Polar Electro Oy, Kemple, Finland) that telemetrically emi 4ed to the portable gas analyzer unit [14]. Capillary blood samples for lactate concentration analysis were collected from the fingertip at rest, during the 30 s intervals and at 1, 3, 5 and/or 7 min at the end of the protocol (until obtaining maximal values) using a portable analyzer (Lactate Pro2; Arkay Inc., Kyoto, Japan) [15]. 2.3. Data Analysis For each running protocol step, a total of twelve strides were analyzed frame-by- frame using two-dimensional motion analysis software (Kinovea, version 0.8.27, Boston, MA, USA) to determine the stride time (time between the left foot touchdown and the next left foot touchdown), step time (time between the left foot touchdown and the right foot touchdown), stride length (mean velocity/stride frequency) and frequency (1/stride time) and contact and flight times (times from initial contact to toe-off and from toe-off to the initial contact of the same foot, respectively) as linear kinematic variables [1]. The flight and contact times normalized to the stride duration were also included (flight time/stride time and contact time/stride time). Based on the marked anatomic points described above, the hip, knee and ankle joint angles (both at initial contact and toe-off moments) were calculated as angular kinematical variables (Figure 1) [13]. Body marks were carefully de- tected, and the mean of three measurements was used. A 0.98 intraclass correlation coef- ficient was verified. Figure 1. Hip, knee and ankle joint angle determination at initial contact and toe-off moments. The lactate-velocity curve modeling method, through the determination of the inter- ception point of the best fit of a combined linear and exponential pair of regressions, was used to determine the individual anaerobic threshold [6,16]. The mean oxygen uptake val- ues from the last
verified. Figure 1. Hip, knee and ankle joint angle determination at initial contact and toe-off moments. The lactate-velocity curve modeling method, through the determination of the inter- ception point of the best fit of a combined linear and exponential pair of regressions, was used to determine the individual anaerobic threshold [6,16]. The mean oxygen uptake val- ues from the last 30 s of each protocol step were selected, and conventional physiological criteria were applied to establish the maximal oxygen uptake, particularly the occurrence of a plateau in oxygen uptake despite a velocity increase (<2.1 mL·kg −1 ·min −1 ) and level of maximal blood lactate higher than 8 mmol·L −1 [14]. Using these two physiological indica- tors, the low, moderate, heavy (the steps below, at and above the anaerobic threshold, respectively) and severe (the step where maximal oxygen uptake was elicited) exercise intensity domains were identified. Figure 1.Hip, knee and ankle joint angle determination at initial contact and toe-off moments. The lactate-velocity curve modeling method, through the determination of the inter- ception point of the best fit of a combined linear and exponential pair of regressions, was used to determine the individual anaerobic threshold [6,16]. The mean oxygen uptake values from the last 30 s of each protocol step were selected, and conventional physiological criteria were applied to establish the maximal oxygen uptake, particularly the occurrence of a plateau in oxygen uptake despite a velocity increase (<2.1 mL·kg −1 · min −1 ) and level of maximal blood lactate higher than 8 mmol·L −1 [14]. Using these two physiological indicators, the low, moderate, heavy (the steps below, at and above the anaerobic threshold,
Sensors2024,24, 7526 4 of 11 respectively) and severe (the step where maximal oxygen uptake was elicited) exercise intensity domains were identified. 2.4. Statistical Analysis Statistical procedures were conducted using SPSS (version 29.0.0.0; IBM Corporation, Armonk, NY, USA) with a 5% significance level. The normal data distribution was checked for all variables with the Shapiro–Wilk test. The Wilcoxon signed-rank test was used to compare stride length and frequency, flight time, normalized flight time and respiratory frequency along the running intensity domains and between laps, with the respective median and interquartile range presented here. Their effect size estimation was conducted using eta-square (η 2 p) based on the Friedman test results and interpreted as small (0.01), medium (0.06) and large (0.14). For the remaining kinematical and physiological variables, presented as mean±standard deviation, a repeated-measures ANOVA was performed with Bonferroni post hoc analysis to identify the pairwise differences, andη 2 pwas calculated, with values of 0.04, 0.25 and above 0.64 considered as minimum, moderate and strong, respectively [17]. A repeated-measures Student’s T-test with effect size estimation (Cohen’sd: 0.20, 0.50 and 0.80 considered as small, medium and large, respectively [18]) was used to compare the kinematical responses between the first and second laps of the protocol steps corre- sponding to each intensity domain. The coefficient of variation (CV) with the respective 95% confidence interval (CI) was also determined. A sample size of 10 participants was deemed sufficient, assuming a statistical power of 80%, an effect size of 0.85 and anαer- ror probability of 5% (G*Power, version 3.1.9.7; Heinrich Heine Universität Düsseldorf, Düsseldorf, Germany). 3. Results The biomechanical and physiological responses when running at low, moderate, heavy and severe intensities are described in Table. With the increase in velocity, a decrease in stride time, step time, contact time and normalized contact time and an increase in stride frequency and normalized flight time were observed. Stride length increased from low to moderate (p= 0.005) and from low, moderate and heavy to severe (p= 0.001), whereas flight time was smaller at low (p< 0.003) and similar between moderate, heavy and severe domains. Hip and knee joint angles
time, contact time and normalized contact time and an increase in stride frequency and normalized flight time were observed. Stride length increased from low to moderate (p= 0.005) and from low, moderate and heavy to severe (p= 0.001), whereas flight time was smaller at low (p< 0.003) and similar between moderate, heavy and severe domains. Hip and knee joint angles at initial contact were similar at low and moderate and decreased at heavy (p= 0.008 and 0.002) and severe (p= 0.023 and 0.004), hip joint angle at toe-off only decreased at severe (p= 0.004), and knee joint angle at toe-off and ankle joint angle both at initial contact and toe-off remained unchanged along the running intensities. From low to severe, a progressive increase in oxygen uptake, minute ventilation, heart rate and blood lactate concentrations was also verified, while respiratory frequency only increased at heavy and severe intensity domains (p< 0.05). The variation between laps within each intensity domain of the assessed linear and an- gular kinematic variables are illustrated in Figures, respectively. From the first to the second laps, the mean velocity was maintained (4.45±0.57 and 4.38±0.51,4.75±0.52 and 4.67±0.48, 5.05±0.53 and 4.96±0.50 and 5.29±0.50 and5.32±0.57 m·s −1 ,d= 0.14–0.61, for low, moderate, heavy and severe intensities, respectively;p> 0.05). The assessed kine- matic variables were similar along each protocol step at the different intensities, except for flight time, normalized flight time and normalized contact time at low (d= 0.84, 0.74 and 0.74; CV [95% CI]: 1.70 [0.96–2.45], 1.82 [1.02–2.61] and 1.21 [0.68–1.74], respectively), hip joint angle at toe-off at moderate (d= 0.95; CV [95% CI]: 0.42 [0.24–0.60]) and stride time, step time, stride frequency, contact time and knee joint angle at toe-off at heavy intensity domains (d= 0.98, 0.97, 0.97, 0.97 and 0.82; CV [95% CI]: 1.18 [0.66–1.69], 1.17 [0.66–1.68], 1.17 [0.66–1.68], 2.04 [1.15–2.94] and 0.44 [0.25–0.63], respectively).
Sensors2024,24, 7526 5 of 11 Table 1.Kinematical and physiological responses at low, moderate, heavy and severe running intensity domains. Low Moderate Heavy Severe Eta-Square p Velocity (m·s −1 ) 4.44±0.55 m,h,s 4.74±0.53 h,s 5.05±0.51 s 5.34±0.54 0.925 <0.001 Stride time (s) 0.70±0.03 m,h,s 0.68±0.03 h,s 0.67±0.03 s 0.65±0.03 0.905 <0.001 Step time (s) 0.35±0.01 m,h,s 0.34±0.01 h,s 0.34±0.01 s 0.33±0.02 0.905 <0.001 Stride length (m) 3.13 (2.79–3.39) m,s 3.23±(2.95–3.52) s 3.28±(2.98–3.56) s 3.52 (3.22–3.71) 0.239 <0.001 Stride frequency (Hz) 1.43 (1.40–1.49) m,h,s 1.46 (1.41–1.52) h,s 1.51 (1.43–1.58) 1.52 (1.46–1.60) 0.248 <0.001 Flight time (s) 0.28 (0.27–0.28) m,h,s 0.29 (0.28–0.29) 0.29 (0.28–0.30) 0.29 (0.28–0.31) 0.168 0.017 Normalized flight time (%) 40.0 (36.8–42.3) m,h,s 41.4 (39.9–43.6) h,s 43.2 (40.6–45.3) s 44.6 (42.9–46.9) 0.246 <0.001 Contact time (s) 0.42±0.03 m,h,s 0.40±0.03 h,s 0.39±0.03 s 0.37±0.02 0.911 <0.001 Normalized contact time (%) 60.8±3.7 m,h,s 58.8±3.2 h,s 57.4±3.5 56.1 ±2.7 0.784 <0.001 Hip joint angle at initial contact ( ◦ ) 152.02±2.85 h,s 151.20±2.31 150.45 ±2.68 149.36 ±2.94 0.530 <0.001 Hip joint angle at toe-off ( ◦ ) 165.70 ±5.62 s 165.16±5.84 s 164.70±6.49 163.64 ±6.39 0.580 <0.001 Knee joint angle at initial contact ( ◦ ) 162.64±5.30 h,s 161.94±4.91 h,s 161.19±4.93 s 159.57±4.49 0.698 <0.001 Knee joint angle at toe-off ( ◦ ) 168.38 ±5.47 168.67 ±5.64 168.60 ±5.92 168.77 ±5.55 0.053 0.683 Ankle joint angle at initial contact ( ◦ ) 100.10 ±6.20 100.05 ±6.08 100.43 ±5.62 101.37 ±6.21 0.164 0.212 Ankle joint angle at toe-off ( ◦ ) 129.95 ±5.97 129.73 ±5.72 129.65 ±6.43 129.55 ±6.84 0.009 0.916 Oxygen uptake (mL·kg −1 ·min −1 ) 54.7±5.2 m,h,s 58.7±5.4 h,s 62.5±6.5 s 67.6±9.4 0.787 <0.001 Minute ventilation (L·min −1 ) 112.2±17.5 m,h,s 124.2±20.4 h,s 138.7±19.5 s 157.7±16.9 0.911 <0.001 Respiratory frequency (breaths·min −1 ) 40.0 (41.0–54.9) s 51.8 (41.5–57.9) h,s 56.1 (47.8–59.2) s 62.1 (56.1–68.6) 0.220 <0.001 Heart rate (bpm) 163±13 m,h,s 170±13 h,s 176±12 s 181±13 0.906 <0.001 Blood lactate concentration (mmol·L −1 ) 3.1±1.2 m,h,s 3.5±1.4 h,s 5.6±2.4 s 10.2±3.9 0.761 <0.001 m,h,s Different from moderate, heavy and severe intensity domains, respectively (p< 0.05).
) 40.0 (41.0–54.9) s 51.8 (41.5–57.9) h,s 56.1 (47.8–59.2) s 62.1 (56.1–68.6) 0.220 <0.001 Heart rate (bpm) 163±13 m,h,s 170±13 h,s 176±12 s 181±13 0.906 <0.001 Blood lactate concentration (mmol·L −1 ) 3.1±1.2 m,h,s 3.5±1.4 h,s 5.6±2.4 s 10.2±3.9 0.761 <0.001 m,h,s Different from moderate, heavy and severe intensity domains, respectively (p< 0.05).
Description
This study characterizes biomechanical and physiological adaptations in runners at varying intensities.