Skip to main content

Table 3 Characteristics and synthesis of the physiological validation studies using the PICO scheme

From: Metabolic Power in Team and Racquet Sports: A Systematic Review with Best-Evidence Synthesis

Study (Year)

Population

Intervention

Comparison

Outcome

Akubat et al. [28]

10 competitive amateur soccer players of unknown sex (20 ± 1 years)

2 test protocols at least 5 days apart: (1) lactate threshold test with 6 four-min stages (6, 8, 10, 12, 14, and 16 km/h) followed by a ramp test until exhaustion (increase of 0.2 km/h every 12 s); (2) modified version of Ball-Sport Endurance and Sprint Test for 30 min, performed twice with 2 days apart

Calculating iTRIMP and examine external/internal load ratios with GPS (5 Hz) derived metabolic power and PlayerLoad™; relationship between ratios and speed at lactate threshold and onset of blood lactate accumulation; influence of fatigue; use of modified equation for metabolic power analysis (Osgnach et al. [3])

Mean metabolic power ratio was largely correlated to speed at lactate threshold both in recovered (r = 0.59) and fatigued (r = 0.57) states; correlation to speed at onset of blood lactate accumulation was large (recovered, r = 0.61) and moderate (fatigued, r = 0.38); high metabolic power ratio was largely (recovered, r = 0.54) and small (fatigued, r = 0.27) correlated to speed at lactate threshold; correlation to speed at onset of blood lactate accumulation was large (recovered, r = 0.67) and very large (fatigued, r = 0.70)

Brown et al. [29]

27 team sport players (15 males, 12 females; 21 ± 2.7 years)

90 min exercise session on outdoor pitch divided into 6 bouts of 5 min of exercise (walking, jogging, running, 3 bouts of simulated team sport circuit) separated by 10 min of rest

GPS (5 Hz interpolated to 15 Hz) derived EE compared to VO2 (portable metabolic cart) derived EE; use of modified equation for metabolic power analysis (Osgnach et al. [3])

Moderate overall (complete 90 min session) underestimation of GPS derived EE (-19.0%); very large underestimation for team sport circuits (−44.0%); very large overestimation for walking (43.0%); no significant differences for jogging (7.8%) and running (4.8%)

Buchheit et al.[30]

14 French elite youth soccer players of unknown sex (15.4 ± 1.6 years)

4.5 min exercise circuit including technical actions with the ball (slaloms, pass and retrieve of a rebound wall, shot on goal) divided into 3 bouts of 1 min of exercise (at speeds of 6.5, 7.0, 7.5 km/h) separated by 30 s of rest; repetition of the circuit one week later

GPS (4 Hz) derived metabolic power compared to VO2 (portable metabolic cart) derived metabolic power; use of modified equation for metabolic power analysis (Osgnach et al. [3])

GPS derived metabolic power was 29 ± 10% lower during exercise and 85 ± 7% lower during recovery; correlation between GPS and VO2 derived metabolic power was small (r = 0.24, exercise and recovery phase) to moderate (r = 0.58, only exercise phase); reliability of GPS derived metabolic power was moderate (CV = 8.0%; ICC = 0.57)

Highton et al. [31]

16 male university rugby players (23.8 ± 4.8 years)

Repeated effort protocol to simulate physical contact including 3 sets of 6 rounds of: 8 m run at 14.4 km/h to collide with tackle bag to the ground, repositioning back, running backwards at 9 km/h to starting point

GPS (10 Hz) derived EE compared to VO2 (portable metabolic cart) derived EE; unknown equation for metabolic power analysis

GPS derived EE showed a systematic underestimation (−5.94 ± 0.67 kcal/min; ~ −45%); correlation between GPS and VO2 derived EE was moderate (r = 0.63)

Manzi et al. [32]

17 male professional Italian Serie A soccer players (28.2 ± 2.2 years)

Data from 19 championship matches and 2 aerobic fitness tests at least 24 h apart: (1) long-stage treadmill test for lactate profiling until exhaustion (1 km/h every 5 min until lactate of 4 mmol/l, then 0.5 km/h every 30 s); (2) short-stage running field test on a 400 m track for VO2max until exhaustion (start 8 km/h and increase of 0.5 km/h every min)

Comparison of aerobic fitness variables (VO2max, VO2VT, %VO2VT, maximal aerobic speed, VL4) and match data in metabolic power categories (distance at high power: > 20 W/kg; very high power: > 35 W/kg; max power: > 55 W/kg) using a video camera system (25 Hz); use of modified equation for metabolic power analysis (Osgnach et al. [3])

Correlations between metabolic power categories and aerobic fitness variables were: large for VO2max (r = 0.55–0.68) and %VO2VT (r = 0.62–0.65), very large for VO2VT (r = 0.72–0.83), and large to very large for maximal aerobic speed (r = 0.52–0.72) and VL4 (r = 0.56–0.73)

Oxendale et al. [33]

12 university team sport players (rugby, soccer, hockey, netball; males, 5 females; 20.8 ± 2.7 years)

3 different testing protocols: (1) 20 m shuttle fitness test with progressively increasing speed to determine VO2max; (2 & 3) multi-directional and linear circuit including 8 bouts of 60 s of intermittent activities (running, sprinting) followed by 120 s rest

GPS (4 Hz) derived EE compared to VO2 (portable metabolic cart) derived EE; use of modified equation for metabolic power analysis (Osgnach et al. [3])

GPS derived EE was lower during multi-directional (−52%) and linear (−34%) condition; GPS and VO2 derived EE was strongly correlated for multi-directional (r = 0.89) and linear (r = 0.95) condition

Stevens et al. [34]

14 amateur soccer players of unknown sex (23 ± 2 years)

2 sessions separated by 30 min on artificial turf: (1) 10 m continuous shuttle running and constant running following the same protocol of running with increasing speed (0.5 km/h every 3 min) from 7.5 to 10 km/h; (2) incremental protocol with an increase of 1 km/h per min until exhaustion followed immediately after the constant running

LPS (500 Hz filtered at 10 Hz) derived EC compared to VO2 (portable metabolic cart) derived EC; determining additional energy cost of 180° change of direction compared to constant running; use of original equation for metabolic power analysis (di Prampero et al. [15]) with adaptations concerning EC of running on flat terrain at constant speed and KT

LPS derived EC was significantly higher (6–11%, main effect: 0.34, p < 0.001) in constant running and lower (−13 to −16%, main effect: −0.94, p < 0.001) in shuttle running when compared to VO2 derived EC

  1. CV  coefficient of variation, EC  energy cost, EE  energy expenditure, GPS  global positioning system, Hz  hertz, ICC  intraclass coefficient correlation, iTRIMP  individualized training impulse, km/h  kilometers per hour, KT  terrain constant, LPS  local positioning system, m  meters, min  minutes, mmol/l  millimoles per liter, s  seconds, VL4  running speed at blood lactate concentration of 4 mmol/l, VO2  oxygen uptake, VO2max  maximum oxygen uptake, VO2VT  oxygen uptake at ventilatory threshold, %VO2VT  percentage of oxygen uptake at ventilatory threshold, W/kg  watts per kilogram