Summary: In 70 sedentary older adults with prediabetes, 12 weeks of low-volume HIIT (3 x 20-second cycle sprints, three times weekly) improved peripheral insulin sensitivity versus no training (Cederholm index estimated treatment difference 18.5%, 95% CI 7.4 to 28.3), but did not significantly improve whole-body insulin sensitivity (Matsuda index 17.0%, 95% CI -4.0 to 33.7). Walking alone showed the opposite pattern, improving whole-body but not peripheral insulin sensitivity.
PICO Summary
| Element | Detail |
|---|---|
| Population | 70 sedentary adults with prediabetes (36 women; mean age 60.8 years, BMI 31.6 kg/m2, fasting glucose 6.6 mmol/L, HbA1c 5.7% / 39 mmol/mol). Single-centre randomised controlled trial, Denmark. |
| Intervention | Low-volume HIIT (3 x 20-second all-out cycle sprints, three times weekly) for 12 weeks, as a HIIT-only arm and a HIIT-plus-walking arm (walking >10,000 steps/day). Four arms total of approximately 17 to 18 participants each. |
| Comparison | No-training control arm; a walking-only arm (>10,000 steps/day) provided an additional active comparator. |
| Outcome | Versus control, peripheral insulin sensitivity (Cederholm index) improved with HIIT (ETD 18.5%, 95% CI 7.4 to 28.3) and HIIT plus walking (15.7%, 95% CI 4.4 to 25.6), but not walking alone (9.4%, 95% CI -2.5 to 19.9). Whole-body insulin sensitivity (Matsuda index) improved with HIIT plus walking (28.0%, 95% CI 10.3 to 42.3) and walking alone (42.3%, 95% CI 28.3 to 53.5), but not HIIT alone (17.0%, 95% CI -4.0 to 33.7). Skeletal-muscle proteins for mitochondrial capacity and glucose uptake improved most with HIIT plus walking. No p values, absolute risk reduction, or NNT applicable (continuous mechanistic outcomes). |
HIIT and Insulin Sensitivity in Prediabetes
RCT · prediabetes · 12 weeks
Low-volume HIIT improved peripheral insulin sensitivity versus no training, but not whole-body sensitivity. Walking showed the opposite pattern, so HIIT was not superior to walking overall.
Expert Commentary
This small four-arm randomised trial offers a more nuanced message than the headline suggests. Low-volume HIIT improved peripheral (largely muscle) insulin sensitivity, yet failed to move the whole-body Matsuda index, whose confidence interval crossed zero. Walking alone produced the mirror image, improving whole-body sensitivity while leaving the peripheral index non-significant. The two modalities therefore appear to act through partly different mechanisms, and the combined arm captured both benefits along with the clearest skeletal-muscle adaptations. The verdict is that HIIT is a plausible, time-efficient option for peripheral insulin sensitivity in prediabetes, but it is not shown here to be superior to walking, and on whole-body sensitivity walking was arguably stronger. The principal limitation is the modest sample (roughly 17 per arm), which leaves the trial underpowered for between-arm comparisons and surrogate-marker outcomes rather than glycaemic or clinical endpoints. The work is from a Novo Nordisk Foundation-linked centre, an indirect sponsorship signal worth noting, though no implausibly large effects are reported. Can I use this with my patients? Cautiously yes, for a motivated, sedentary, lower-risk older adult with prediabetes who wants a short exercise option, while making clear that brisk daily walking remains at least as good for whole-body glucose handling. Larger trials with clinical glycaemic endpoints are needed before HIIT is positioned as the preferred modality.
References
Mensberg P, Frandsen C, Carl CS, Espersen E, Leineweber T, Larsen EL, et al. High-intensity interval training improves insulin sensitivity in individuals with prediabetes. Eur J Endocrinol. 2025;192(4):456-465. doi:10.1093/ejendo/lvaf004