The Effect of Consuming Carbohydrate With and Without Protein on the Rate of Muscle Glycogen Re-synthesis During Short-Term Post-exercise Recovery: a Systematic Review and Meta-analysis.

This meta-analysis examined two separate questions in a two-part design:

**Part 1:** Does consuming carbohydrate (CHO) after exercise increase the rate of muscle glycogen re-synthesis compared to consuming only water or a non-nutrient placebo?

**Part 2:** Does consuming carbohydrate plus protein (CHO+PRO) after exercise increase the rate of muscle glycogen re-synthesis compared to consuming the same amount of carbohydrate alone?

The primary outcome was the rate of muscle glycogen re-synthesis, measured in millimoles per kilogram of dry muscle mass per hour (mmol·kg dm⁻¹·h⁻¹). This was calculated from muscle biopsy samples taken immediately after exercise (within 30 minutes) and again within 8 hours of recovery.

The researchers also explored whether the timing of carbohydrate administration (hourly vs. less frequent intervals) influenced the rate of glycogen replenishment.

The meta-analysis included 29 trials drawn from 21 published studies, involving a total of 246 participants. All participants were healthy adults (≥18 years old) with no known medical conditions. The population consisted primarily of:

Trained and recreationally active individuals (specific training status varied across studies)

Mostly male participants (the review notes that few studies included female participants)

Athletes from various endurance and team sports backgrounds

Participants with no known metabolic or gastrointestinal disorders

The average age across studies was approximately 24 years (range roughly 19–30 years). Body mass averaged around 75 kg. All participants were non-smokers and not taking any medications known to affect metabolism.

Importantly, the review excluded studies involving people with medical conditions, so these findings apply specifically to healthy, active individuals — not to people with diabetes, metabolic disorders, or clinical conditions affecting glycogen metabolism.

The researchers used the gold-standard technique for measuring muscle glycogen: **needle biopsy of skeletal muscle**. This involves inserting a hollow needle into the vastus lateralis (thigh) muscle under local anaesthesia to extract a small sample of muscle tissue. The sample is then analysed chemically to determine glycogen concentration.

Key measurement details:

**First biopsy:** Taken within 30 minutes after exercise ended (this established the post-exercise baseline glycogen level)

**Second biopsy:** Taken between 1 and 8 hours after exercise (this established how much glycogen had been restored)

**Units:** Results were expressed in mmol·kg dm⁻¹ (millimoles per kilogram of dry muscle mass). Where studies reported values in wet mass, the authors converted using a standard factor of 4.35

**Rate calculation:** The rate of glycogen re-synthesis was calculated by dividing the change in glycogen concentration by the recovery time between biopsies

The researchers also extracted data on:

Carbohydrate dose (g·kg body mass⁻¹·h⁻¹)

Protein dose (g·kg body mass⁻¹·h⁻¹)

Timing of nutrient administration (hourly vs. less frequent)

Type of exercise used to deplete glycogen

Participant characteristics (age, sex, body mass, fitness level)

**Study design:** This is a systematic review and meta-analysis — a statistical synthesis of multiple existing randomised controlled trials. The authors followed the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) guidelines and pre-registered their protocol on PROSPERO (registration code CRD42020156841).

**Search strategy:** The authors searched Web of Science and Scopus from inception until September 2019, with an additional search in March 2020. They used the Boolean expression: (exercis* AND glycogen) OR (postexercis* AND glycogen). Two investigators independently screened studies, with a third resolving disagreements.

**Inclusion criteria:**

Controlled trials using repeated-measures (within-subject) designs

Adult participants (≥18 years) with no medical conditions

Muscle glycogen measured by needle biopsy (gold standard)

First biopsy within 30 minutes post-exercise, second biopsy within 8 hours

Full-text published in English

**Exclusion criteria:**

Between-subject designs

Studies administering other energy-containing nutrients (alcohol, fat, caffeine, creatine)

Non-standardised exercise protocols

Non-oral treatment administration

Studies using non-biopsy measurement methods (e.g., MRI, ultrasound)

**Quality assessment:** The authors used the Rosendal Scale to assess methodological quality. The average score was 61 ± 8% (mean ± standard deviation), which is considered "excellent" methodological quality (threshold ≥60%). No study scored below 50%, so none were excluded for poor quality.

**Statistical approach:** The authors used random-effects meta-analyses, which assumes that the true effect size may vary across studies (rather than assuming one fixed effect). They also conducted meta-regression analyses to explore whether factors like carbohydrate dose or timing influenced results. Heterogeneity was assessed using the I² statistic (which measures how much variation across studies is due to true differences rather than chance).

**What this design can and cannot prove:**

This design can prove:

Whether carbohydrate ingestion increases glycogen re-synthesis compared to water (causal relationship, because all included studies were controlled trials)

Whether adding protein to carbohydrate changes glycogen re-synthesis compared to carbohydrate alone

The average magnitude of these effects across multiple studies

Whether certain factors (dose, timing) modify the effect

This design cannot prove:

Mechanisms (why carbohydrate helps or why protein doesn't add benefit)

Effects on actual athletic performance (the outcome was glycogen, not performance)

Effects beyond 8 hours of recovery (the review specifically limited to short-term recovery)

Effects in special populations (diabetics, elderly, clinical populations)

Whether protein might help in other ways (e.g., muscle protein synthesis, reducing soreness) — the review only measured glycogen

**Methodological strengths:**

Pre-registered protocol reduces risk of selective reporting

Two independent screeners reduces selection bias

Gold-standard measurement (biopsy) rather than surrogate measures

Random-effects models account for between-study variation

Meta-regression explores moderating variables

**Methodological weaknesses:**

Only two databases searched (may miss some studies)

Only English-language publications included (language bias)

Small total sample (246 participants across 29 trials)

High heterogeneity in some analyses (I² = 66.8% for Part 1)

Most participants were young, healthy males — limited generalisability

The Rosendal Scale is not the most commonly used quality assessment tool (many meta-analyses use the Cochrane Risk of Bias tool)

**Part 1: Carbohydrate vs. Water/Placebo (10 trials, n = 86)**

**Primary outcome:** Ingesting carbohydrate during recovery (average dose 1.02 ± 0.4 g·kg BM⁻¹·h⁻¹) improved the rate of muscle glycogen re-synthesis by **23.5 mmol·kg dm⁻¹·h⁻¹** compared to water (95% CI: 19.0 to 27.9, p < 0.001)

**Heterogeneity:** I² = 66.8% (moderate-to-high heterogeneity, meaning there was real variation between studies beyond chance)

**Dose-response:** A significant positive correlation was found between the interval of carbohydrate administration and the effect size (R² = 0.44, p = 0.027). Specifically, consuming carbohydrate at least hourly produced a larger benefit compared to less frequent feeding

**Carbohydrate dose analysis:** Meta-regression showed that higher carbohydrate doses (up to about 1.2 g·kg⁻¹·h⁻¹) were associated with faster glycogen re-synthesis, but the relationship was not perfectly linear

**Part 2: Carbohydrate+Protein vs. Carbohydrate Alone (19 trials, n = 160)**

**Primary outcome:** Ingesting carbohydrate plus protein (CHO: 0.86 ± 0.2 g·kg BM⁻¹·h⁻¹; PRO: 0.27 ± 0.1 g·kg BM⁻¹·h⁻¹) did **not** improve the rate of muscle glycogen re-synthesis compared to carbohydrate alone (0.95 ± 0.3 g·kg BM⁻¹·h⁻¹)

**Effect size:** The mean difference was only **0.4 mmol·kg dm⁻¹·h⁻¹** (95% CI: -2.7 to 3.4, p = 0.805)

**Heterogeneity:** I² = 56.4% (moderate heterogeneity)

**Subgroup analyses:** The null effect held regardless of:

- Carbohydrate dose (low vs. high)

- Protein dose

- Training status of participants

- Type of exercise used for glycogen depletion

- Timing of nutrient administration

**Secondary analyses:**

The authors found no significant effect of training status, muscle contraction type, or degree of glycogen depletion on the rate of re-synthesis (though these analyses were limited by small numbers of studies)

Studies with higher methodological quality (Rosendal score >60%) showed similar results to lower-quality studies

**Part 1 (CHO vs. water):** The improvement of 23.5 mmol·kg dm⁻¹·h⁻¹ represents a substantial effect. To put this in perspective:

Typical resting muscle glycogen stores are around 350–500 mmol·kg dm⁻¹ in trained individuals

After exhaustive exercise, glycogen can drop to 50–100 mmol·kg dm⁻¹

A re-synthesis rate of ~23.5 mmol·kg dm⁻¹·h⁻¹ means that over 4 hours of recovery, you'd restore about 94 mmol·kg dm⁻¹ — roughly 20–30% of your total glycogen stores

Without any carbohydrate, glycogen re-synthesis is essentially negligible (near zero) during the first few hours of recovery

So the effect is not just statistically significant — it's practically enormous. Eating carbohydrate is the difference between meaningful glycogen restoration and virtually none

**Part 2 (CHO+PRO vs. CHO alone):** The effect of adding protein was essentially zero — 0.4 mmol·kg dm⁻¹·h⁻¹, with a confidence interval spanning from -2.7 to +3.4. This means:

Even if there were a real effect, it would be tiny (less than 2% of the carbohydrate effect)

The data are consistent with protein having no effect, a small positive effect, or even a small negative effect

For practical purposes, adding protein to adequate carbohydrate does nothing for glycogen replenishment

**Timing effect:** The meta-regression found that feeding interval explained 44% of the between-study variance in Part 1. This means that how often you eat carbohydrate matters almost as much as how much you eat. Hourly feeding produced faster glycogen restoration than less frequent feeding, even when total carbohydrate intake was similar.

**What the authors acknowledge:**

The small number of studies and participants limits statistical power, especially for subgroup analyses

High heterogeneity in Part 1 (I² = 66.8%) suggests that other factors beyond carbohydrate ingestion influence glycogen re-synthesis

Most studies used male participants, limiting generalisability to female athletes

The review only examined short-term recovery (≤8 hours), so findings don't apply to longer recovery periods

The Rosendal Scale is less commonly used than other quality assessment tools

**What a critical reader would note:**

**Population homogeneity:** Nearly all participants were young, healthy, trained males. Women may respond differently due to hormonal influences on glycogen metabolism. Older adults may have different glycogen storage capacities

**Biopsy limitations:** While the gold standard, needle biopsy samples only a tiny region of one muscle. Glycogen storage can vary within a muscle and between muscle groups. The procedure itself can cause local inflammation that might affect subsequent measurements

**Exercise standardisation:** Studies used different exercise protocols (cycling, running, resistance exercise) to deplete glycogen. The degree of depletion and muscle groups involved varied considerably

**Carbohydrate types:** Studies used different carbohydrate sources (glucose, sucrose, maltodextrin, mixed meals). Some carbohydrate types may be more effective than others, but the review didn't have enough data to analyse this

**Protein types:** The protein sources varied (whey, casein, soy, mixed amino acids, individual amino acids like leucine). Some proteins might theoretically enhance glycogen synthesis more than others, but the overall null effect suggests this doesn't matter much

**Publication bias:** The authors didn't formally test for publication bias (e.g., funnel plot analysis), though they did search for unpublished studies

**Industry funding:** Some included studies were funded by sports nutrition companies, which could introduce bias toward positive findings for their products

**Lack of blinding:** In many studies, participants knew whether they were consuming carbohydrate or water (it's hard to blind a carbohydrate drink vs. water). This could affect behaviour during recovery

**No performance outcome:** The review only measured glycogen, not actual athletic performance. Faster glycogen restoration doesn't automatically translate to better subsequent performance

For someone running their own n=1 experiment:

### What to test

**Primary test:** Compare your recovery between two conditions:

**Condition A (carbohydrate only):** Consume 1.0–1.2 g of carbohydrate per kg of body mass per hour during the first 4 hours after exercise

**Condition B (carbohydrate + protein):** Consume the same carbohydrate dose plus 0.3–0.4 g of protein per kg of body mass per hour

**Optional secondary test:** Compare hourly carbohydrate feeding vs. less frequent feeding (e.g., every 2 hours) with the same total carbohydrate dose.

### Minimum meaningful duration

**Recovery period:** 4 hours (the most common recovery window in the included studies)

**Number of sessions:** At least 3–5 trials per condition (to account for day-to-day variability)

**Total experiment:** 2–3 weeks if you alternate conditions every other training day

### What to measure (specific metrics)

**Primary metric:** Subjective recovery and performance in a subsequent exercise session:

Rate of perceived exertion (RPE, Borg 6–20 scale) during a standardised warm-up or time trial

Time to complete a fixed workload (e.g., 20-minute cycling time trial, 5 km run)

Power output or pace during a standardised interval session

**Secondary metrics:**

Subjective muscle soreness (0–10 scale, 0 = no soreness, 10 = worst imaginable)

Perceived recovery (0–10 scale, 0 = not recovered at all, 10 = fully recovered)

Heart rate during standardised exercise (lower heart rate at same workload = better recovery)

Blood glucose (if you have a glucometer) — though this doesn't directly measure glycogen

**Note:** You cannot directly measure muscle glycogen without a biopsy, so you'll need to rely on performance and subjective measures as proxies.

### Key confounds to control for

**Total calorie intake:** Keep total calories consumed during recovery the same between conditions. If you add protein, reduce fat or other calories to compensate

**Timing of last meal before exercise:** Standardise what you eat 2–4 hours before the depletion workout

**Hydration status:** Drink the same volume of fluid in both conditions

**Sleep:** Get the same amount of sleep the night before each trial (7–9 hours)

**Previous training:** Avoid heavy leg training in the 48 hours before each trial

**Time of day:** Conduct all trials at the same time of day

**Menstrual cycle (if female):** Conduct trials in the same phase of your cycle, as hormonal changes affect glycogen metabolism

**Supplement use:** Avoid creatine, caffeine, and other ergogenic aids during the experiment

**Exercise protocol:** Use the exact same depletion protocol each time (same duration,