Continuous Glucose Monitor (CGM): Training and Exercise Impact

What CGM Actually Measures During Exercise

A CGM measures glucose in interstitial fluid, not blood, using a subcutaneous filament sensor worn on the upper arm or abdomen. Readings update every 1 to 5 minutes depending on the device (Dexcom G7 updates every 5 minutes; Abbott Libre 3 every minute). The FDA device clearance database lists current approved sensors.

The Interstitial Lag Problem

Because the sensor samples interstitial fluid rather than capillary blood, there is an inherent 5 to 15-minute lag between actual blood glucose and the number on your receiver. Shah et al. (2019) in npj Digital Medicine confirmed a mean lag of approximately 9 minutes during rapid glucose changes. During exercise, when glucose is shifting quickly, that lag matters. A falling arrow on your CGM may mean your blood glucose is already 15 to 20 mg/dL lower than the displayed value.

Sweat, Pressure, and Sensor Accuracy

Intense training introduces two additional accuracy factors. Compression against a surface (during bench press, swimming, or contact sports) can temporarily compress the interstitial capillary bed and produce false-low readings. Heavy sweating changes skin conductance around the adhesive, which may cause signal noise. A 2020 study in Diabetes Technology and Therapeutics (N=30 athletes) found that compression artifact caused CGM readings to drop as much as 40 mg/dL below fingerstick values during resistance exercise.

How Different Exercise Types Affect Glucose

Exercise type is the single biggest predictor of your CGM pattern during a workout. Aerobic, anaerobic, and mixed-modality training produce distinct glucose signatures.

Aerobic Exercise: Glucose Falls

Sustained moderate-intensity aerobic work (cycling, running, rowing at 50 to 70% VO2max) increases glucose uptake in skeletal muscle via GLUT4 translocation, an insulin-independent pathway. Richter and Hargreaves (2013) in Physiological Reviews established the GLUT4 mechanism as the primary driver of exercise-induced glucose disposal. On a CGM, expect a steady 1 to 3 mg/dL per minute drop during the first 20 to 40 minutes of a moderate aerobic session. A one-hour run at an easy pace may lower glucose by 30 to 50 mg/dL from baseline.

For people with type 2 diabetes, this effect is clinically significant. The PREDIMED-Plus trial (N=6,874) and multiple sub-analyses demonstrated that structured aerobic exercise significantly reduced fasting glucose and glycated hemoglobin. CGM data captures this benefit in real time.

High-Intensity Interval Training: Glucose Spikes First

HIIT produces a different CGM pattern. A sprint or maximal effort triggers a catecholamine surge (primarily epinephrine and norepinephrine) that stimulates hepatic glycogenolysis, pushing glucose into the bloodstream faster than exercising muscle can consume it. Marliss and Vranic (2002) in Diabetes described this counterregulatory response in detail, showing blood glucose can rise 20 to 60 mg/dL during maximal-intensity bouts. On your CGM, expect an upward arrow during and immediately after an interval set, followed by a delayed fall as catecholamines clear.

The net glycemic benefit of HIIT is real. Meta-analyses show HIIT improves HbA1c and fasting glucose comparably to, or better than, moderate continuous exercise in people with type 2 diabetes. A Cochrane review (Jelleyman et al., 2015) of 50 trials found HIIT reduced HbA1c by 0.19% more than moderate-intensity continuous training. The spike is transient.

Resistance Training: A Mixed Signal

Weightlifting produces a hybrid pattern. Each set triggers a small catecholamine rise, causing brief glucose spikes. Between sets, glucose partially recovers. Over a full session, net glucose tends to fall modestly, though less dramatically than during aerobic work. A study by Feito et al. (2019) in Nutrients (N=24) found that a 45-minute resistance session lowered mean CGM glucose by approximately 12 mg/dL over 3 hours post-exercise.

Optimal CGM Ranges: Resting, Active, and Recovery

Resting and Fasting Targets

For metabolically healthy adults without diabetes, CGM studies consistently show mean glucose near 97 to 99 mg/dL with most readings falling between 70 to 120 mg/dL. Hall et al. (2018) in PLOS Biology (N=57 non-diabetic adults) found median CGM glucose of 99 mg/dL and that individuals spent 96% of time below 140 mg/dL.

The American Diabetes Association defines the clinical Time in Range (TIR) target as greater than 70% of readings between 70 to 180 mg/dL for people with diabetes, with fewer than 4% of readings below 70 mg/dL. The ADA Standards of Care (2024, Section 6) codify these TIR thresholds.

For non-diabetic athletes using CGMs for performance optimization, many longevity-medicine clinicians set tighter informal targets: mean glucose below 100 mg/dL and less than 1 hour per day above 140 mg/dL. These are not currently defined in formal guidelines.

During Exercise

A pre-exercise glucose of 90 to 140 mg/dL is a reasonable starting window for most non-diabetic individuals. Starting a long aerobic session below 90 mg/dL increases risk of symptomatic hypoglycemia mid-workout. Starting above 160 mg/dL may indicate inadequate pre-workout fuel timing or an unrecognized metabolic issue worth discussing with a clinician.

For people with type 1 diabetes, the American Diabetes Association and JDRF published joint consensus guidelines (Riddell et al., 2017, Lancet Diabetes and Endocrinology) recommending a pre-exercise glucose target of 126 to 180 mg/dL and specific carbohydrate strategies based on exercise type.

Post-Exercise Recovery Window

The post-exercise period carries underappreciated hypoglycemia risk. Muscle glycogen resynthesis continues for hours after training ends, maintaining elevated GLUT4 activity and insulin sensitivity. MacDonald (1987) in British Medical Journal first documented delayed post-exercise hypoglycemia occurring 6 to 15 hours after prolonged aerobic work. A CGM running overnight after an afternoon long run is one of the most clinically informative recordings you can collect.

Glycemic Variability: The CGM Metric That Fasting Glucose Misses

Single-point fasting glucose and even HbA1c miss glycemic variability, the amplitude and frequency of glucose swings throughout the day. CGM quantifies variability using the coefficient of variation (CV). A CV below 36% is considered acceptable by consensus; above 36% indicates excessive variability regardless of mean glucose. Battelino et al. (2019) in Diabetes Care defined these CV thresholds in the international CGM consensus report.

Why Variability Predicts Risk Independently

High glycemic variability drives oxidative stress through repeated glucose spikes. Monnier et al. (2006) in JAMA (N=83) showed that glucose excursions, not mean glucose alone, correlated significantly with 8-iso prostaglandin F2-alpha, a validated oxidative stress marker (P<0.001). For athletes, chronically elevated post-meal glucose variability may impair recovery and tissue repair.

How Exercise Reduces Variability

Regular aerobic training reduces CV by blunting post-meal glucose spikes. A randomized controlled trial by Little et al. (2011) in Journal of Physiology (N=16) found that three 30-second "all-out" sprint intervals performed before a meal reduced post-meal glucose area under the curve by 30% compared to no exercise. Walking 10 minutes after each meal produces a similar blunting effect with much lower intensity demand.

Fueling Strategies Guided by CGM Data

Using your CGM trace before, during, and after exercise gives you information that calorie counting alone cannot provide.

Pre-Workout Fueling

If your CGM shows glucose below 90 mg/dL thirty minutes before a session longer than 60 minutes, a small carbohydrate dose (15 to 25 g of fast-digesting carbs) may prevent mid-workout hypoglycemia. If glucose is above 130 mg/dL, additional carbohydrates before a moderate aerobic session are likely unnecessary. Your CGM trace from three to four previous similar workouts is the most reliable guide.

Intra-Workout Glucose Management

For sessions lasting over 90 minutes, checking CGM trend arrows (not just the number) every 20 to 30 minutes helps you decide whether to fuel. A downward double arrow below 100 mg/dL warrants immediate carbohydrate intake of 15 to 30 g. The Academy of Nutrition and Dietetics and American College of Sports Medicine joint position statement recommends 30 to 60 g of carbohydrate per hour for endurance exercise exceeding 60 minutes.

Post-Workout Recovery Nutrition

Consuming 0.3 to 0.4 g of protein per kilogram body weight combined with moderate carbohydrate within 30 minutes of finishing resistance or high-intensity training supports muscle glycogen resynthesis and limits the post-exercise glucose drop. Your CGM trace for the two hours after a session shows whether your recovery nutrition is working. A CGM that stays flat at 85 to 110 mg/dL through recovery indicates well-matched refueling.

Training-Specific CGM Patterns Worth Knowing

The following framework synthesizes published exercise physiology literature and clinical CGM use patterns across exercise modalities. Use it as a reference when interpreting your own traces.

| Exercise Type | Typical CGM Pattern (First 30 Min) | Net Effect at 2 Hours Post | Hypoglycemia Risk | |---|---|---|---| | Easy aerobic (50 to 65% VO2max) | Steady decline, 1 to 2 mg/dL per min | Glucose 20 to 40 mg/dL below baseline | Moderate | | Tempo run (70 to 80% VO2max) | Sharper decline, may flatten with catecholamine release | Glucose 15 to 30 mg/dL below baseline | Low-moderate | | HIIT / sprint intervals | Initial spike 20 to 40 mg/dL, then rapid fall | Glucose at or below baseline | Low during, moderate overnight | | Heavy resistance training | Small spikes per set, gradual net decline | Glucose 10 to 20 mg/dL below baseline | Low-moderate | | Mixed (CrossFit-style) | Variable, spike-then-fall pattern | Highly individual | Moderate | | Yoga / light mobility | Minimal change, may decline slightly | Near-baseline or mildly lower | Very low |

CGM in Non-Diabetic Athletes: What the Evidence Actually Shows

CGM use in people without diabetes is growing, but the evidence base for performance-specific interventions remains early-stage. Danne et al. (2017) in Diabetes Care noted that CGM accuracy studies have largely been conducted in diabetic populations, and performance in euglycemic non-diabetic individuals may differ.

What CGM Adds Beyond Standard Lab Work

Fasting glucose and HbA1c are snapshots. CGM provides continuous waveform data: how high your glucose goes after a specific meal, how quickly it returns to baseline, whether overnight glucose is stable, and how your body responds to specific workouts. Freckmann et al. (2007, Diabetes Technology and Therapeutics) showed that CGM identified post-meal excursions above 140 mg/dL in 73% of participants who had normal fasting glucose. That gap matters when evaluating early metabolic dysfunction.

Limitations to Acknowledge

CGM should not replace periodic HbA1c, fasting insulin, or OGTT in a full metabolic workup. A two-week CGM wear gives a richer picture of glycemic patterns than any single lab value, but sensor error rates of 10 to 15% MARD (mean absolute relative difference) still apply. Clinical decisions should never rest on a single CGM reading. The FDA's 2018 guidance on integrated CGM systems outlines these accuracy standards.

Practical Protocol for Using CGM Around Training

1. Start a new sensor 24 hours before a key training session to allow stabilization. First-day readings on Abbott Libre 3 and Dexcom G7 tend to be more variable.
2. Record your pre-exercise CGM value, trend arrow, and the time since your last meal for at least 4 to 6 sessions before drawing conclusions.
3. Compare the same workout type across multiple sessions. One CGM trace is anecdote. Four to six comparable sessions become a pattern.
4. Flag any pre-exercise glucose above 180 mg/dL for clinician review. Repeated readings above 180 mg/dL at rest warrant a formal diabetes screening with HbA1c and fasting insulin.
5. Use the time-in-range report, not the glucose graph alone. Most CGM apps generate a TIR summary automatically. Aim for greater than 90% of readings between 70 to 140 mg/dL if you are non-diabetic and using CGM for health optimization.

When to Contact a Clinician

A CGM is a monitoring tool, not a treatment. Contact your HealthRX provider if:

• Fasting CGM glucose consistently reads above 100 mg/dL on three or more separate mornings.
• You experience any reading below 60 mg/dL, with or without symptoms.
• Post-meal glucose exceeds 160 mg/dL more than twice per week.
• You see nocturnal dips below 70 mg/dL after exercise days.

The ADA 2024 Standards of Care define a fasting plasma glucose of 100 to 125 mg/dL as prediabetes and 126 mg/dL or above as diabetes on two separate occasions. A CGM trace showing repeated fasting values in that range should be confirmed with a venous fasting glucose.