Table 1.
Overview of the methodological details and study outcomes of acute and chronic train-low studies according to the relevant train-low paradigm
| Reference | Subjects | Duration | Exercise protocol and glycogen status (mmol/kg dw) | Skeletal muscle adaptations | Exercise performance outcomes |
|---|---|---|---|---|---|
| Twice per day model | |||||
| Hansen et al. [9] | 7 Untrained men | 10 weeks 5 days × week |
Knee extensor exercise. One leg trained 50% of sessions with low glycogen (LOW), while the other trained all sessions with high glycogen (HIGH). Second session glycogen in LOW—pre: 200, post: 100 mmol/kg dw, respectively | Greater increase in CS activity in the LOW condition Increased β-HAD activity in the LOW condition only |
Improved TTE for knee extensor exercise |
| Yeo et al. [17] | 14 trained male cyclists/triathletes | 3 weeks 4 × week |
100 min steady-state cycling (63% PPO) followed by 8 × 5-min intervals at maximal pace either 2 h (LOW) or 24 h (HIGH) later. Pre-interval exercise glycogen—LOW: 256, HIGH: 390. Post-exercise glycogen—LOW: 124, HIGH: 229 | Increased CS and β-HAD activity in the LOW condition only Increased COXIV protein content in the LOW condition only |
Similar improvements (10%) in 60-min TT for both groups |
| Morton et al. [18] | 23 active men | 6 weeks 4 × week |
6 × 3-min running (90% VO2max). NORM trained once per day, while LOW + PLA and LOW + GLU trained twice per day (every other day). LOW + GLU ingested CHO before and during every second training session. Pre exercise glycogen—LOW: 232 and 253, HIGH: 412 and 387 in the gastrocnemius and vastus lateralis, respectively. Post-exercise glycogen—LOW: 107 and 176, HIGH: 240 and 262 in the gastrocnemius and vastus lateralis, respectively | Greater increase in SDH activity in LOW + PLA compared with LOW + GLU and NORM | Similar improvements in VO2max and YoYoIR2 for all groups |
| Yeo et al. [23] | 12 trained male cyclists/triathletes | Acute exercise | 100-min steady-state cycling (63% PPO) followed by 8 × 5-min intervals at maximal pace either 2 h (LOW) or 24 h (HIGH) later. Pre-interval exercise glycogen—LOW: 256, HIGH: 390. Post-exercise glycogen—LOW: 124, HIGH: 229 | Greater phosphorylation of AMPKThr172 in LOW | NA |
| Hulston et al. [19] | 14 trained male cyclists | 3 weeks 6 × week |
90-min cycling at 70% VO2max followed by (2 h apart) HIT (8 × 5 min) in the LOW group. The HIGH group performed alternate days of either steady state or HIT cycling. Acute glycogen status not measured | β-HAD protein content increased in LOW only Increased fat utilization from muscle triglycerides in LOW only |
Similar improvements in 60-min TT for both groups |
| Cochran et al. [22] | 10 Active men | Acute exercise | HIT cycling (5 × 4-min at 90–95% heart rate reserve) twice per day (separated by 3 h). One group consumed CHO (2.3 g.kg) between sessions (HIGH), whereas the other group restricted CHO intake (LOW). Pre-pm exercise glycogen—LOW: 256, HIGH: 390. Post-exercise glycogen—LOW: 124, HIGH: 229 | Greater phosphorylation of p38MAPK in LOW following pm exercise Similar increase in PGC-1α and COXIV gene expression |
NA |
| Cochran et al. [20] | 18 Active men | 2 weeks 3 days × week |
HIT cycling (5 × 4 min at 60% PPO) twice per day (separated by 3 h). One group consumed CHO (2.3 g kg) between sessions (HIGH), whereas the other group restricted CHO intake (LOW). Acute glycogen status not measured | Similar increase in maximal CS activity and protein content of both CS and COXIV | Greater improvement in 250-kJ TT performance in the LOW group |
| Fasted training model | |||||
| Akerstrom et al. [27] | 9 Active men | Acute exercise | 2 h one-legged knee extensor exercise (60% Wmax) in either a fasted (FAST) or fed (exogenous CHO during; FED) state. Pre-exercise glycogen: 500 mmol/kg dw in both groups. Post-exercise glycogen: 300 and 200 in the FED and FAST states, respectively | Reduced AMPKα2 activity in FED | NA |
| Lee-Young et al. [49] | 9 Active men | Acute exercise | 120-min cycling (65% VO2peak) exercise in either a fasted (FAST) or fed (exogenous CHO during; FED) state. Pre-exercise glycogen: 500 mmol/kg dw in both groups. Post-exercise glycogen: 150 and 100 in the FED and FAST states, respectively | Similar increases in AMPKα2 activity and AMPKα2Thr172 and ACC-βSer222 phosphorylation | NA |
| De Bock et al. [31] | 20 Active men | 6 weeks 3 × week |
1–2 h cycling (75% VO2peak). One group trained in the fasted state (FAST), with the other consuming CHO before and during exercise (FED). Acute glycogen status not measured | FABP increased in the FAST condition only | NA |
| Nybo et al. [32] | 15 untrained men | 8 weeks 3–4 × week |
3–6 min of high-intensity intervals (70–85% VO2max). Subjects received either CHO or PLA during exercise. Acute glycogen status not measured | Greater increases in β-HAD activity and basal muscle glycogen content in the PLA group only | Similar improvements in peak power, VO2max and 15-min TT performance |
| Van Proeyen et al. [30] | 20 active men | 6 weeks 4 × week |
1–1.5 h cycling (70% VO2max). One group trained in the fasted state (FAST), with the other consuming CHO before and during exercise (FED). Acute glycogen status not measured | CS and β-HAD maximal activity increased in the FAST condition only | Similar improvements in 1-h TT performance in both groups |
| Sleep-low model | |||||
| Pilegaard et al. [15] | Study A: 6 active men Study B: 6 active men |
Acute exercise Acute exercise |
Study A: 1-legged glycogen-depleting exercise followed by 2-legged cycling (2 h at 45% VO2max) on the subsequent day. Pre-exercise glycogen—LOW: 337, HIGH: 609. Post-exercise glycogen—LOW: 306, HIGH: 423 Study B: 3 h of 2-legged knee extensor exercise with either NORM or LOW glycogen. Pre-exercise glycogen—LOW: 240, HIGH: 398. Post-exercise glycogen—LOW: 101, HIGH: 153 |
Study A: Enhanced gene expression of PDK4, LPL and HKII at rest in LOW only Studies A and B: Enhanced gene expression of PDK4 and UCP3 post-exercise in LOW only |
NA NA |
| Wojtaszewski et al. [36] | 8 Trained men | Acute exercise | 60-min cycling at 70% VO2peak with either LOW or HIGH muscle glycogen (from exercise/diet manipulation the previous day). Pre-exercise glycogen—LOW: 163, HIGH: 909. Post-exercise glycogen—LOW: 150, HIGH: 400 | Increased AMPKα2 activity in LOW only Greater phosphorylation of ACCSer221 in LOW |
NA |
| Chan et al. [37] | 8 active men | Acute exercise | 60-min cycling (70% VO2peak) with either HIGH or LOW glycogen (achieved by exercise/diet manipulation the previous evening). Pre-exercise glycogen—LOW: 163, HIGH: 375. Post-exercise glycogen—LOW: 17, HIGH: 102 | Greater phosphorylation of p38MAPK in LOW Enhanced gene expression of IL-6 in LOW |
NA |
| Steinberg et al. [21] | 7 active men | Acute exercise | 60-min cycling at 70% VO2max with either LOW or NORM muscle glycogen. Pre-exercise glycoge—LOW: 150, HIGH: 390. Post-exercise glycogen—LOW: 17, HIGH: 111 | Greater AMPKα2 activity, phosphorylation of ACCSer221 and nuclear translocation of AMPKα2 in LOW only Enhanced gene expression of GLUT4 in LOW |
NA |
| Bartlett et al. [38] | 8 active men | Acute exercise | HIT running (6 × 3 min at 90% VO2max). The LOW group performed glycogen-depleting cycling the night before and restricted CHO overnight. The HIGH group consumed a high-CHO breakfast and CHO during exercise. Pre-exercise glycogen – LOW: 100, HIGH: 500. Post-exercise glycogen—LOW: 80, HIGH: 300 | Phosphorylation of ACCSer79 and p53Ser15 in LOW only Enhanced gene expression of PGC-1α, PDK4, Tfam and COXIV in LOW |
NA |
| Psilander et al. [24] | 10 trained male cyclists | Acute exercise | 6 × 10-min cycling (64% VO2max) with either HIGH or LOW glycogen (achieved by exercise/diet manipulation 14 h previously). Pre-exercise glycogen—LOW: 166, HIGH: 478. Post-exercise glycogen—LOW: 130, HIGH: 477 | Enhanced gene expression of PGC-1α in LOW Increased gene expression of PDK4 and COXIV in LOW only |
NA |
| Lane et al. [39] | 7 trained male cyclists | Acute exercise | Evening bout of high-intensity cycling (8 × 5 min at 82.5% PPO) followed by 120-min steady-state cycling (50% PPO) the subsequent morning. The LOW group restricted CHO overnight, whereas the HIGH group consumed a high-CHO diet (4 g.kg BM). Pre-exercise glycogen—LOW: 349, HIGH: 459. Post-exercise glycogen—LOW: 266, HIGH: 338 | Greater phosphorylation of ACCSer79 post-AM exercise in LOW Enhanced gene expression of CD36, FABP3 and PDK4 post-AM exercise in LOW |
NA |
| Marquet et al. [40] | 21 male triathletes | 3 weeks 6 × week |
HIT (8 × 5 min cycling at 85% MAP or 6 × 5-min running at individual 10-km intensity) in the evening followed by LIT (60-min cycling at 65% MAP) the subsequent morning. One group consumed CHO between training sessions (HIGH), whereas the other group restricted CHO intake (LOW). Acute glycogen status not measured | NA | Improved 10-km running TT performance and improved TTE cycling (150% peak aerobic power) in the LOW group only |
| Marquet et al. [41] | 11 trained male cyclists | 1 week 6 × week |
HIT (8 × 5-min cycling at 85% MAP) in the evening followed by LIT (60-min cycling at 65% MAP) the subsequent morning. One group consumed CHO between training sessions (HIGH), whereas the other group restricted CHO intake (LOW). Acute glycogen status not measured | NA | Improved 20-km cycling TT performance in the LOW group only |
| Recover-low model | |||||
| Pilegaard et al. [16] | 9 active men | Acute exercise | 75-min cycling (75% VO2max) followed by 24 h recovery with either HIGH or LOW CHO diet. Glycogen was restored to 576 and 348 with HIGH and LOW CHO diets, respectively, at 24 h | Gene expression of PDK4, UCP3, LPL and CPT1 remained elevated for 8–24 h with CHO restriction post-exercise | NA |
| Jensen et al. [54] | 15 male triathletes | Acute exercise | 4-h cycling (56% VO2max) followed by 4 h recovery feeding with either HIGH (1 g.kg.h) or LOW (water only) CHO. Post-exercise glycogen—LOW: 234, HIGH: 245. 4-h glycogen—LOW: 264, HIGH: 444 | Similar gene expression of PGC-1α, Tfam, NRF-1, COXIV, PDK4, LPL, PPAR, UCP3 and GLUT4 in both groups | NA |
| High-fat feeding | |||||
| Hammond et al. [43] | 10 active men | Acute exercise | High-intensity running (8 × 5 min at 85% VO2peak) followed by steady-state running (60 min at 70% VO2peak) 3.5 h later. Steady-state running was either commenced with high or low (but high fat) CHO availability. Muscle glycogen was similar in both groups (200 mmol/kg dw) post-steady-state running | p70S6K activity was suppressed with high-fat feeding Similar gene expression of PGC-1α, p53, CS, Tfam, PPAR and ERRα in both groups |
NA |
| Periodized model | |||||
| Impey et al. [48] | 11 amateur male cyclists | Acute exercise | Based on the principle of ‘fuel for the work required’. 4 × 30 s HIT cycling (150% PPO) and 45 min steady-state cycling (50% PPO) followed by 1 min efforts (80% PPO) until exhaustion with either HIGH or LOW glycogen (by previous exercise/diet manipulation for 36 h previously). The HIGH group consumed CHO before, during and after exercise, whereas the LOW group consumed leucine-enriched protein | 36 h of prior CHO restriction enhanced p53, SIRT1 and Tfam gene expression. CHO restriction before and during exercise induced work-efficient AMPK signalling. Post-exercise CHO restriction and keeping glycogen < 100 mmol/kg dw reduced p70S6K activity | Exercise capacity (1-min efforts at 80% PPO) enhanced in HIGH trial (158 vs. 100 min) |
| Burke et al. [45] | 22 international male race walkers | 3 weeks 7 × week |
3 weeks of intensified training (race walking, resistance training, cross training). Athletes consumed three different diets across the training period: (a) high CHO; (b) LCHF; (c) periodized CHO intake with periods of low CHO training. Acute glycogen status not measured | NA | Similar improvements in VO2peak between all groups Improved 10-km race times in the high CHO and periodized CHO groups (no change in LCHF) LCHF diet increased the O2 cost of race walking |
| Gejl et al. [55] | 26 elite male endurance athletes | 4 weeks 7 × week |
4 weeks of intensified training. Athletes either performed all sessions with high CHO availability or followed a periodized model, performing three sessions per week with reduced CHO availability. Glycogen content was 400 mmol/kg dw following LOW carbohydrate availability training session | Similar increase in maximal CS activity No increase in β-HAD activity in either group |
Similar improvement in VO2max and 30-min TT performance between groups |
Where possible, muscle glycogen status of the relevant experimental trials is also cited
β-HAD 3-hydroxyacyl-CoA dehydrogenase, ACC acetyl-CoA carboxylase, AMPK AMP-activated protein kinase, BM Body Mass, CHO carbohydrate, CD36 cluster of differentiation 36, CPT1 carnitine palmitoyltransferase 1, CS citrate synthase, COX cytochrome c oxidase, ERRα estrogen-related receptor α, FABP3 fatty acid binding protein, GLU glucose, GLUT4 glucose tr ansporter type 4, HIT high-intensity training, HKII hexokinase II, IL interleukin, LCHF low-carbohydrate, high-fat, LIT low intensity training, LPL lipoprotein lipase, MAP maximal aerobic power, NA not available, NORM normal, NRF-1 nuclear respiratory factor 1, p38MAPK p38 mitogen-activated protein kinase, p53 tumor protein 53, p70S6K ribosomal protein S6 kinase, PDK4 pyruvate dehydrogenase kinase 4, PGC-1α peroxisome proliferator-activated receptor gamma coactivator 1-α, PLA placebo, PM post meridian, PPAR peroxisome proliferator-activated receptor, PPO peak power output, SIRT1 NAD-dependent deacetylase sirtuin-1, SDH succinate dehydrogenase, Tfam transcription factor A, TT time trial, TTE time to exhaustion, UCP3 uncoupling protein 3, VO2max maximum rate of oxygen consumption, VO2peak peak rate of oxygen consumption, Wmax Watt maximum, YoYoIR2 Yo-Yo intermittent recovery test 2