High-Intensity Interval Exercise in Type 1 Diabetes: Stronger Protection Against Exercise-Related Glycaemic Decline in the Postabsorptive Than Postprandial State

Highlights

• A 2026 randomized crossover trial by Melin and colleagues shows that high-intensity interval exercise (HIIE) causes a smaller fall in blood glucose than iso-mechanical moderate continuous exercise (CONT) in adults with type 1 diabetes, with the greatest protection observed in the postabsorptive state.
• The study extends prior work suggesting that intermittent or sprint-containing exercise can blunt exercise-associated glycaemic decline by amplifying counterregulatory responses, but also shows that insulin-on-board materially modifies this effect.
• Across the broader literature, glycaemic outcomes during exercise in type 1 diabetes are determined not only by exercise modality, but also by timing relative to meals, circulating insulin levels, pre-exercise glucose, carbohydrate supplementation, and technology use.
• Clinically, the new findings support a more individualized exercise prescription: for adults at high risk of exercise-induced hypoglycaemia, HIIE performed in a lower-insulin, postabsorptive window may be a practical strategy when medically appropriate.

Background

Type 1 diabetes (T1D) creates a uniquely difficult metabolic environment for exercise. In individuals without diabetes, aerobic exercise is accompanied by a coordinated reduction in endogenous insulin secretion and an increase in glucagon, catecholamines, cortisol, and growth hormone, allowing hepatic glucose output to better match muscle glucose uptake. In T1D, exogenous insulin levels cannot rapidly fall at exercise onset, especially when exercise is undertaken after a meal bolus. This mismatch favors accelerated peripheral glucose disposal, suppression of hepatic glucose production, and a decline in glycaemia during prolonged or moderate-intensity aerobic activity. As a result, fear of hypoglycaemia remains one of the most important barriers to exercise participation in T1D.

Over the last two decades, exercise science in T1D has moved beyond the simplistic distinction between “exercise is good” and “exercise causes hypoglycaemia.” Instead, a more nuanced physiology has emerged. Continuous moderate-intensity aerobic exercise often lowers glucose, resistance exercise may be more glycaemically stable, and high-intensity intervals or brief maximal sprints can attenuate glucose decline in some settings. However, the literature has also been heterogeneous. Seemingly discrepant results across studies often reflect differences in prandial state, insulin delivery method, fitness level, pre-exercise glucose, interval protocol, and recovery nutrition.

The randomized crossover trial by Melin et al. published in Diabetologia in 2026 is important because it addresses a clinically relevant but underexplored question: does the glycaemic advantage of HIIE over continuous moderate exercise depend on whether exercise is performed in the postprandial (PP) or postabsorptive (PA) state? The answer appears to be yes. Their data suggest that HIIE is not uniformly “protective”; rather, its protective effect is magnified when insulin-on-board is lower.

Key Content

The 2026 Melin et al. trial: design, findings, and why it matters

Melin A, Lespagnol E, Combes A, and colleagues conducted a randomized crossover study in 20 physically active adults with T1D, aged 18-55 years, with nine women included. Participants completed four 25-minute iso-mechanical cycling sessions in random order at the EURASPORT facility of Lille University: HIIE-PA, HIIE-PP, CONT-PA, and CONT-PP. HIIE consisted of alternating 1-minute bouts at 100% maximal aerobic power, whereas CONT was performed at 50% maximal aerobic power. The design is a key strength because matching mechanical workload reduces the common confounding that interval exercise simply “does less work” than continuous exercise.

The primary outcome was capillary glucose during exercise; secondary measures included lactate, oxygen consumption, heart rate, perceived exertion, enjoyment, and 24-hour post-exercise interstitial glucose with dietary and insulin tracking. Mixed-effects models evaluated modality, prandial state, and interaction effects.

The physiological signal was clear. HIIE generated higher oxygen uptake, heart rate, lactate concentration, and perceived exertion than CONT, confirming a stronger internal training stimulus despite similar mechanical load. Importantly, enjoyment was not reduced, which matters for adherence.

The glycaemic findings were the study’s central contribution. Glucose declined less in the postabsorptive than postprandial state, and less during HIIE than CONT. Most importantly, there was a significant interaction: the glucose-sparing effect of HIIE was more pronounced in the PA condition. Mean capillary glucose change during exercise was:

• HIIE-PA: -0.72 ± 1.71 mmol/L
• CONT-PA: -1.73 ± 1.48 mmol/L
• HIIE-PP: -2.33 ± 1.61 mmol/L
• CONT-PP: -2.61 ± 1.27 mmol/L

Thus, HIIE conferred the greatest relative benefit when performed in the lower-insulin postabsorptive state; in the postprandial state, the difference between HIIE and CONT narrowed considerably.

The recovery data further reinforce clinical relevance. Across 24 hours, adjusted time spent below 3.0 mmol/L was lower after HIIE than after CONT, despite lower dietary glycaemic load after HIIE. This suggests that the benefit was not simply due to more aggressive carbohydrate rescue. For clinicians, this is an important translational point: the intervention may reduce not only acute glucose decline during exercise but also clinically meaningful hypoglycaemia later in recovery.

Physiological rationale: why prandial state changes the effect of HIIE

The mechanistic interpretation of the Melin study is biologically plausible and aligns with earlier metabolic work. During moderate aerobic exercise, skeletal muscle glucose uptake rises sharply. In T1D, if circulating insulin levels remain high, hepatic glucose production is relatively constrained while peripheral glucose utilization is facilitated. The result is a fall in blood glucose.

HIIE may blunt this decline through several mechanisms:

• Greater catecholamine release, stimulating hepatic glycogenolysis and gluconeogenesis.
• Higher lactate generation, providing a gluconeogenic substrate.
• Transient reductions in glucose disposal during recovery phases of interval work.
• A stronger counterregulatory hormone response than steady-state moderate exercise.

However, high exogenous insulin levels in the postprandial period can overwhelm part of this protective physiology. Insulin suppresses hepatic glucose output and promotes muscle glucose uptake. Therefore, even if HIIE recruits a stronger counterregulatory response, the net glucose-preserving effect may be muted when insulin-on-board remains substantial. The Melin trial elegantly demonstrates this principle at the clinical level.

How the new trial fits with earlier evidence on intermittent high-intensity exercise

Earlier experimental studies in T1D had already suggested that adding intensity bursts to aerobic exercise can attenuate glucose decline. Classic work by Bussau and colleagues showed that a brief 10-second sprint performed before moderate exercise could reduce the fall in glycaemia, presumably through catecholamine-mediated hepatic glucose output. Related studies by Guelfi and collaborators found that intermittent high-intensity bouts during otherwise moderate exercise were associated with smaller reductions in glucose than continuous exercise alone.

These studies were mechanistically influential because they challenged the assumption that “harder exercise always means more hypoglycaemia.” Instead, they introduced the idea that exercise intensity pattern matters. Yet most of this earlier literature used tightly controlled laboratory paradigms, often in small samples, and did not systematically compare postabsorptive versus postprandial conditions using iso-work designs.

The next phase of evidence examined broader exercise categories rather than single sprint manipulations. Reviews and position statements increasingly distinguished aerobic exercise, resistance exercise, interval exercise, and mixed sessions. Resistance and interval formats often appeared less likely than prolonged aerobic exercise to cause immediate glucose decline, though with the caveat that delayed hyperglycaemia and subsequent nocturnal hypoglycaemia could still occur depending on insulin and nutrition adjustments.

Melin et al. build on that foundation in two ways. First, they confirm that higher-intensity interval exercise can be glucose-sparing even when total mechanical work is matched. Second, and more novel, they show that the size of this advantage is state-dependent: it is strongest when basal rather than prandial insulin predominates.

Evidence from reviews, consensus statements, and guideline-oriented literature

The 2017 international consensus statement on exercise management in T1D led by Riddell and colleagues remains a landmark synthesis. It emphasized that moderate aerobic exercise typically lowers glucose, that high-intensity and resistance exercise may produce more variable or smaller declines, and that insulin reductions and carbohydrate strategies should be tailored to exercise timing, intensity, and duration. Importantly, the consensus highlighted insulin-on-board as a major determinant of hypoglycaemia risk. The Melin trial can be viewed as a direct empirical confirmation of that principle in the specific context of HIIE versus continuous exercise.

Subsequent reviews by Zaharieva, Yardley, Moser, Rabasa-Lhoret, and others have similarly underscored that no exercise modality is universally safe without context. High-intensity intervals may reduce immediate glycaemic decline, but benefits depend on individual fitness, antecedent hypoglycaemia, sex-related hormonal influences, time of day, and pre-exercise insulin exposure. The newer literature around continuous glucose monitoring (CGM) and automated insulin delivery has also shown that technology improves safety, yet exercise remains a challenging scenario because insulin pharmacodynamics lag behind rapidly changing metabolic needs.

The American Diabetes Association and other professional organizations endorse regular aerobic and resistance exercise for people with T1D, while acknowledging the need for individualized insulin and carbohydrate adjustments. These documents rarely prescribe one single exercise mode as superior for all patients. Instead, they recommend matching the plan to patient goals, risks, and preferences. The Melin findings strengthen that individualized approach by suggesting that HIIE may be especially attractive for physically active adults who can exercise in a postabsorptive or low-bolus-insulin window.

Continuous moderate exercise versus HIIE: what the totality of evidence suggests

Viewed as a whole, the literature suggests several practical patterns.

First, continuous moderate aerobic exercise remains highly beneficial for cardiovascular fitness and overall health, but it is the exercise modality most consistently associated with acute glucose lowering in T1D. This effect is particularly evident when exercise follows a meal bolus or when basal insulin remains relatively high.

Second, HIIE and sprint-augmented exercise often produce a smaller immediate decline in glucose. The likely explanation is stronger sympathetic activation and greater hepatic glucose release. The Melin trial adds that this benefit is not merely theoretical but can be detected in a controlled crossover design with meaningful post-exercise outcomes.

Third, the superiority of HIIE is relative, not absolute. In the postprandial state, Melin et al. still observed substantial glucose declines with HIIE. Thus, HIIE should not be interpreted as eliminating the need for insulin adjustment or carbohydrate planning when exercise is undertaken soon after meals.

Fourth, recovery glycaemia matters as much as exercise-period glucose. Some patients tolerate exercise well during the activity but experience late-onset or nocturnal hypoglycaemia due to glycogen repletion and enhanced insulin sensitivity. The reduced time below 3.0 mmol/L over 24 hours after HIIE in the 2026 trial is therefore clinically meaningful and arguably more important than the intraworkout glucose slopes alone.

Methodological strengths and limitations of the Melin trial

The trial has several methodological strengths:

• Randomized crossover design, allowing each participant to serve as their own control.
• Four-condition structure directly comparing modality and prandial state.
• Iso-mechanical workload matching, which improves interpretability.
• Measurement of both exercise physiology and 24-hour glycaemic outcomes.
• Adjustment for carbohydrate intake and insulin dosing in recovery analyses.

There are also important limitations. The sample size was small, though typical for mechanistic exercise physiology studies. Participants were physically active adults, limiting generalizability to sedentary individuals, adolescents, older adults, or those with impaired awareness of hypoglycaemia. The exercise was laboratory-based cycling; other activities may produce different responses. The report summary does not indicate subgroup analyses by insulin delivery modality, sex hormones/menstrual phase, or fitness strata, all of which may influence glycaemic variability. Finally, the study examined acute responses rather than long-term training adaptations.

These limitations do not diminish the validity of the core finding, but they do caution against overgeneralization. The results should be seen as informing exercise prescription, not replacing individualized clinical judgment.

Translational implications for clinical practice

Several practice-oriented messages emerge from this body of evidence.

1. Timing relative to meals matters as much as exercise type.
The most consistent message from the 2026 trial is that lower insulin-on-board amplifies the glycaemic stability of HIIE. For adults with T1D who are already trained and medically cleared for vigorous exercise, scheduling HIIE in a postabsorptive or lower-bolus-insulin period may reduce glucose decline more effectively than performing the same session soon after a meal.

2. HIIE is not a free pass in the postprandial period.
Even under HIIE, postprandial exercise still produced marked declines in glucose. Patients should still consider meal bolus reductions, carbohydrate intake, or algorithmic pump strategies when exercising after food.

3. Recovery planning remains essential.
Lower late hypoglycaemia exposure after HIIE is encouraging, but it does not eliminate risk. CGM surveillance, individualized correction thresholds, bedtime snack considerations, and insulin dose adjustments may still be required, especially after afternoon or evening sessions.

4. Training stimulus and adherence matter.
HIIE induced a stronger cardiorespiratory and metabolic training stimulus without reducing enjoyment. This supports its role not only as a glycaemic strategy but also as an efficient fitness intervention for selected adults with T1D.

5. Patient selection is critical.
HIIE may not suit everyone. Individuals with proliferative retinopathy, uncontrolled hypertension, autonomic neuropathy, unstable cardiovascular disease, or very low fitness may require different exercise prescriptions or formal medical evaluation before engaging in vigorous interval work.

Expert Commentary

The Melin trial is a meaningful advance because it resolves a practical ambiguity that has long complicated exercise counseling in T1D. Previous research suggested that high-intensity intervals can be glucose-sparing, but real-world clinicians often observed that this was inconsistent. The new study helps explain why: the benefit of HIIE is conditional on the surrounding insulin milieu.

This insight has several broader implications. First, it shifts the discussion from “which exercise is safest?” to “which exercise is safest in which metabolic context?” That is a more sophisticated and clinically useful framing. Second, it aligns exercise counseling with the growing precision-medicine ethos in diabetes care. Exercise recommendations should be individualized according to insulin-on-board, recent meals, CGM trend arrows, time of day, prior hypoglycaemia, and patient goals. Third, it offers a plausible pathway to better adherence. Many adults with T1D avoid exercise because of hypoglycaemia fear, yet they also want efficient training. HIIE performed at an appropriate time may satisfy both concerns in selected patients.

There are, however, unresolved questions. The literature remains less clear on how these findings translate to people using hybrid closed-loop systems, where automated basal modulation may partially mitigate risk. It is also uncertain whether the same pattern holds in children, adolescents, less fit adults, or individuals with recurrent severe hypoglycaemia. Another open question is whether repeated training with HIIE changes counterregulatory responses over time, thereby altering the acute glycaemic phenotype. Longitudinal studies are needed.

A further issue is ecological validity. Laboratory-based exercise protocols isolate physiology but may not capture free-living sports, where stress, competition, ambient temperature, and mixed aerobic-anaerobic demands interact. Team sports, circuit training, and spontaneous recreational activity may behave more like HIIE than like steady cycling, but with greater variability. Future work should integrate wearable physiology, CGM, and standardized insulin algorithms to test whether the postabsorptive advantage of interval work persists outside the lab.

Finally, clinicians should avoid overinterpreting “less glucose decline” as necessarily “better” in every circumstance. Some individuals begin exercise hyperglycaemic, and a modest reduction may be desirable. Conversely, vigorous exercise can occasionally increase glucose transiently via catecholamine surges, especially when insulin levels are low. The ideal goal is not zero glucose movement but predictable, safe trajectories aligned with clinical context.

Conclusion

The 2026 randomized crossover trial by Melin and colleagues provides one of the clearest demonstrations to date that the glycaemic effects of exercise in T1D are jointly determined by exercise modality and metabolic state. Compared with iso-mechanical moderate continuous exercise, HIIE led to a smaller decline in capillary glucose and less clinically significant hypoglycaemia during 24-hour recovery. Crucially, this protective effect was substantially stronger in the postabsorptive than postprandial condition, highlighting the central role of insulin-on-board.

When integrated with prior mechanistic studies, consensus guidance, and the broader exercise literature in T1D, these findings support a practical clinical message: for appropriate adults with T1D, HIIE can be a useful strategy to reduce exercise-associated glycaemic decline, particularly when scheduled during periods of lower circulating insulin. Nevertheless, HIIE is not universally protective, especially after meals, and exercise planning must still incorporate insulin adjustment, carbohydrate strategy, CGM interpretation, and patient-specific risk assessment.

The next research frontier is clear: larger, more diverse trials should test HIIE versus other exercise modalities across insulin delivery systems, age groups, and real-world settings, with outcomes extending beyond acute glycaemia to adherence, quality of life, fitness adaptation, and severe hypoglycaemia prevention. For now, the Melin study meaningfully advances the science of precision exercise prescription in type 1 diabetes.

Table: Clinical interpretation of the current evidence

Question	Current evidence-based answer	Clinical implication
Does HIIE lower glucose less than moderate continuous exercise in T1D?	Often yes, especially in controlled studies and particularly when insulin-on-board is lower.	Consider HIIE as an option for patients prone to exercise-related hypoglycaemia.
Does prandial state matter?	Yes. The Melin 2026 trial shows the HIIE advantage is more pronounced in the postabsorptive than postprandial state.	Exercise timing relative to meals and bolus insulin should be planned deliberately.
Does HIIE eliminate hypoglycaemia risk?	No. Glucose can still fall substantially, especially after meals, and late hypoglycaemia remains possible.	Use CGM, insulin adjustment, and carbohydrate strategies as needed.
Is HIIE acceptable from an adherence perspective?	In the 2026 trial, enjoyment was maintained despite greater physiological strain.	HIIE may be both efficient and acceptable for selected adults.
Who should avoid or modify vigorous interval exercise?	Patients with certain complications or limited exercise capacity may need evaluation and individualized programs.	Screen for cardiovascular risk and diabetes complications before prescribing vigorous exercise.

References

• Melin A, Lespagnol E, Combes A, Berthoin S, Rabasa-Lhoret R, Savina Y, Tagougui S, Bautin N, Baudoux F, Descatoire A, Vambergue A, Heyman E. The protective effect of high-intensity interval exercise compared with continuous moderate exercise on glycaemic decline is more pronounced in the postabsorptive vs postprandial state in adults with type 1 diabetes: a randomised crossover trial. Diabetologia. 2026. PMID: 42262529.
• Riddell MC, Gallen IW, Smart CE, Taplin CE, Adolfsson P, Lumb AN, Kowalski A, Rabasa-Lhoret R, McCrimmon RJ, Hume C, Annan F, Fournier PA, Graham C, Bode B, Galassetti P, Jones TW, Millán IS, Heise T, Peters AL, Petz A, Laffel LM. Exercise management in type 1 diabetes: a consensus statement. Lancet Diabetes Endocrinol. 2017.
• Bussau VA, Ferreira LD, Jones TW, Fournier PA. The 10-s maximal sprint: a novel approach to counter an exercise-mediated fall in glycemia in individuals with type 1 diabetes. Diabetes Care. 2006.
• Guelfi KJ, Jones TW, Fournier PA. Intermittent high-intensity exercise does not increase the risk of early postexercise hypoglycemia in individuals with type 1 diabetes. Diabetes Care. 2005.
• Yardley JE, Kenny GP, Perkins BA, Riddell MC, Malcolm J, Boulay P, Khandwala F, Sigal RJ. Resistance versus aerobic exercise: acute effects on glycemia in type 1 diabetes. Diabetes Care. 2013.
• Zaharieva DP, Riddell MC. Prevention of exercise-associated dysglycemia: a case study-based approach in type 1 diabetes. Diabetes Spectrum. 2015.
• Moser O, Eckstein ML, McCarthy O, Deere R, Bain SC, Haahr H, Zijlstra E, Bracken RM. Performance and glucose control in individuals with type 1 diabetes: current insights and emerging technologies. Front Endocrinol. 2020.
• Rabasa-Lhoret R, Bourque J, Ducros F, Chiasson JL. Guidelines for premeal insulin dose reduction for postprandial exercise of different intensities and durations in type 1 diabetic subjects treated intensively with a basal-bolus insulin regimen. Diabetes Care. 2001.
• Colberg SR, Sigal RJ, Yardley JE, Riddell MC, Dunstan DW, Dempsey PC, Horton ES, Castorino K, Tate DF. Physical activity/exercise and diabetes: a position statement of the American Diabetes Association. Diabetes Care. 2016.