SYNTH-25 · Guide
Menthol in the heat
The route may change whether it helps, and it can hide heat strain rather than remove it
Primary sources (abstracts checked 2026-09-27)
Jeffries & Waldron, 2019 — *The effects of menthol on exercise performance and thermal sensation: a meta-analysis.* J Sci Med Sport 22(6):707–715. DOI 10.1016/j.jsams.2018.12.002 · REF-961, A-
Gavel, Barreto, Hawke, Stellingwerff, James, Saunders & Logan-Sprenger, 2024 — *How cool is that? The effects of menthol mouth rinsing on exercise capacity and performance: a systematic review and meta-analysis.* Sports Med Open 10:18. DOI 10.1186/s40798-024-00679-8 · REF-962, A-
Wang, Fan, Li et al., 2026 — *Acute topical menthol application and exercise performance: a systematic review and meta-analysis.* J Therm Biol 140:104529. DOI 10.1016/j.jtherbio.2026.104529 · REF-963, A-
Zhu, Zhao, Yao, Wang & Yuan, 2026 — *Effects of different menthol administration routes on endurance performance and physiological responses in the heat: a network meta-analysis.* Front Nutr 13:1833420. DOI 10.3389/fnut.2026.1833420 · REF-964, A-
Barwood, Kupusarevic & Goodall, 2019 — repeated menthol spray during cycling in the heat. Int J Sports Physiol Perform 14(5):644–649. DOI 10.1123/ijspp.2018-0561 · REF-965, B
Barwood, Barwood, Davis et al., 2025 — whole-body menthol gel in trained triathletes. Int J Sports Physiol Perform 20(9):1271–1277. DOI 10.1123/ijspp.2025-0038 · REF-966, B
Riera, Trong, Sinnapah & Hue, 2014 — *Physical and perceptual cooling with beverages to increase cycle performance in a tropical climate.* PLoS ONE 9(8):e103718. DOI 10.1371/journal.pone.0103718 · REF-967, B
Rinaldi, Collado, Trong & Hue, 2023 — *One hour of internal precooling with cold water/menthol enhances cycling performance in a heat/wet stress environment: a pilot study.* Biol Sport. DOI 10.5114/biolsport.2023.116007 · REF-968, B
Key Findings
Menthol changes perception far more than physiology. It activates the TRPM-8 cold receptor chemically (mechanism from the pharmacology, not from these trials), so the nervous system reports "cold" while core temperature climbs as it would otherwise: thermal sensation fell (Hedges' g −0.54 across 11 studies, Jeffries 2019, REF-961, A-; g −0.74 for topical application, Wang 2026, REF-963, A-) while core temperature was unchanged (Wang 2026) and all routes had only modest, non-significant effects on core temperature and heart rate (Zhu 2026, REF-964, A-).
Ingested menthol drinks are the route with verified cycling trials here (the mouth-rinse trials quoted below are unverified). In 12 trained male cyclists/triathletes doing 20-km trials at 30.7 °C and 78% humidity, both a menthol aroma (0.5%) and a colder beverage improved performance, with ice-slush/menthol and cold/menthol the best conditions (Riera 2014, REF-967, B, crossover; the abstract gives no size for the gain).Further trial figures reported by the practitioner review that prompted this entry — +9% time to exhaustion at 65% peak power in 34 °C; +6.0% mean power over a 30-km time trial in trained women; +1.8% mean power in trained junior males over a 30-min race simulation with about two thirds of riders improving — are secondary reports not traced to a primary paper in this base and not verified against an abstract; do not quote them to an athlete as findings.
Ingestion rewards a long load. In 7 heat-acclimated trained cyclists, an ice-slush menthol drink taken through the full hour before a 1-h time trial in hot-humid conditions produced better performance than 30 minutes of the same precooling or a neutral drink, with lower rectal temperature during the rest period; 30 minutes was no better than the neutral drink (Rinaldi 2023, REF-968, B, crossover pilot; the abstract gives no size for the gain).The "~4 km" margin formerly quoted here is not in the abstract.
Topical is weaker and can work against you. In 8 trained cyclists at 35 °C / 22% humidity, repeated 0.20% menthol spraying to the torso extended time to exhaustion (4.6 vs 2.4 min) while significantly reducing sweat production (994 vs 1180 mL) and sweat rate (827 vs 941 mL/h), and its effect diminished with repeated application (Barwood 2019, REF-965, B, crossover).In 8 trained triathletes, whole-body 3.5% menthol gel before a 30-min swim plus 20-km cycling time trial improved perception but did not improve performance (39:36 vs 40:53 min, P = 0.34) and increased sweat production (1140 vs 961 mL) (Barwood 2025, REF-966, B, crossover) — so the sweat effect is not consistent across topical protocols.The 2026 topical meta-analysis (Wang 2026, REF-963, A-; 16 studies, 191 participants) found a small significant benefit (g = 0.27, P = 0.02), most evident in aerobic endurance tasks, after rub-on application and under temperate conditions; in hot environments (≥30 °C) the effect was smaller and not statistically significant (the abstract gives no effect size for that subgroup and warns that subgroup findings were sensitive to influential observations).
The meta-analyses disagree, and the disagreement is informative. Jeffries 2019 (REF-961, A-; 13 studies, all routes) found a small overall benefit (g = 0.33, P = 0.05) that was larger and significant for internal application (g = 0.40, P = 0.03); Gavel 2024 (REF-962, A-; 10 mouth-rinse studies) found no significant overall effect (SMD 0.12, P = 0.23) with no moderation by concentration (0.01% vs 0.1%), swill duration, number of swills or room temperature, noting a possible endurance-specific benefit; Zhu 2026 (REF-964, A-; 15 RCTs in ≥25 °C, network) ranked ingestion highest for endurance performance (SUCRA 76.1%) and mouth rinse for mean power (66.9%), with most pairwise comparisons non-significant, and ranked topical application weakest for thermoregulation (SUCRA 6.2%) with a possible slight upward trend in core temperature from obstructed heat loss.
Heat-specific? The base's reading is that menthol earns its place only on hot days: Zhu 2026 restricted itself to ≥25 °C, and a menthol rinse in thermoneutral 21 °C producing no benefit is a secondary-reported trial (not verified here).Against that, Gavel 2024's meta-regression found no influence of room temperature on the mouth-rinse effect (REF-962, A-), and the topical benefit was most evident under temperate conditions and not significant at ≥30 °C (REF-963, A-). Treat "only in the heat" as the base's practice rule, not a settled finding.
No additive effect on top of carbohydrate. Secondary-reported trials in which menthol was mixed into a carbohydrate swill or a commercial carb-menthol drink found nothing over the carbohydrate alone; these are not verified against a primary abstract here, and an independent rinse alongside normal fuelling was not tested.
Practical Application (the base's own protocol, built on the findings above)
In the heat, take menthol by mouth, not on the skin — and a cold menthol drink has more behind it than a rinse. Drink: internal application pooled g = 0.40, P = 0.03 (REF-961, A-); cold and ice-slush menthol drinks improved time trials in two small crossover trials in humid heat (REF-967, REF-968, B each; the second does not separate cold from menthol); ingestion ranked first for endurance performance in the heat, with no route significantly better than another (REF-964, A-).Rinse: pooled effect not significant, SMD 0.12 (REF-962, A-). Skin: g = 0.27 overall but smaller and not significant at ≥30 °C (REF-963, A-), and sweating moved in opposite directions in two trials (REF-965, REF-966, B each) — not recommended on a hot day.Drink as trialled in REF-967: 190 mL of cold (3 °C) or ice-slush menthol-aromatized beverage before the start and every 5 km. Rinse, if preferred (base practice, not a tested optimum): 0.01% L-menthol, 20–25 mL, swilled 5–10 s and spat out, repeated every 10–20 minutes (or every 5–10 km) through a hot effort.Gavel 2024 (REF-962) found no dose-response for concentration (0.01% vs 0.1%), swill duration or number of swills, so the low concentration and the repeat interval are practice, not measured optima; Riera 2014 dosed every 5 km, but with a drunk beverage, not a rinse.
More is not better. 2–5% on skin starts to irritate; above ~10% the sensation flips from cooling to burning — pharmacology carried from the practitioner review, not from these trials.
If there is a rest period, load long. An hour of ice-slush menthol before the start (Rinaldi 2023, REF-968, B — a pilot in 7 heat-acclimated cyclists that did not separate the cold from the menthol). Cold-water-plus-menthol immersion between efforts (+15.6% power in the second of two 20-min bouts) is a secondary-reported result not verified here — useful for criteriums, stage starts, triathlon transitions as an idea to test, not a number to promise.
Save it for genuinely hot days (base practice; see the heat-specific note above).
Treat commercial "cooling" carb drinks as carb drinks. The carbohydrate is why they work (inference from the null additive results).
Safety
The main risk IS the mechanism. Menthol makes an athlete feel cooler and less exerted without lowering the heat load, and Zhu 2026's own conclusion (REF-964, A-) is that it "may mask actual subjective thermal perception without alleviating objective physiological heat strain" and must be paired with physiological monitoring; topically it blunted sweating in one cycling trial (Barwood 2019, REF-965, B, whose authors conclude that care should be taken with its use) — it decouples how hot you feel from how hot you are.An expert panel convened ahead of the Tokyo Olympics flagging repeated mouth rinsing in elite or highly motivated athletes is reported by the practitioner review and not verified here.
No entry in the base reports adverse events or heat illness with menthol (the meta-analysis abstracts carry no safety outcome), so the base cannot say whether heat illness has been attributed to it; laboratory trials stop participants at a core-temperature ceiling — a safety net that does not exist at a race.
Menthol is not heat management. Hydration, pacing and honest reading of one's own state still do the work; menthol raises the bar for those rather than lowering it. Heat acclimation, cooling and hyperhydration are the countermeasures the heat-stress literature actually recommends (REF-002).
"No adverse effects at mouth-rinse concentrations (0.01%)" was formerly attributed here to the abstract of Gavel 2024; the abstract (REF-962) reports no safety outcome, so that statement is unverified and should not be quoted. Use food-grade commercial products rather than mixing raw crystals (base practice).
Limitations
Effects are small and inconsistent, and the one mouth-rinse meta-analysis held here (REF-962, A-) finds no significant overall effect (SMD 0.12). Individual response varies widely.
Several cycling figures in this entry (+9% TTE, +6.0% in women, +1.8% in juniors, +15.6% after immersion, the 21 °C null, the carbohydrate-additive nulls, the Tokyo panel) come from the practitioner review's reference list and are not held as entries in this base; they were not verified against a primary abstract on 2026-09-27 and are labelled as such above.
Whether an independent menthol rinse layered on top of normal carbohydrate fuelling adds anything is untested — the null results mixed menthol into the drink.
Provenance note: compiled from KnowledgeIsWatt #177 (Dr Jeff Rothschild, 27 Jul 2026), a practitioner review whose reference list carries full citations with DOIs. The eight primary papers listed above were checked against their abstracts on 2026-09-27; the remaining figures are marked as secondary reports.
How this entry was checked and changed
Why this level
the meta-analyses are real but small and inconsistent, the mouth-rinse meta-analysis is null, and the protocol below is the base's own practice
Content correction 2026-09-27
Numbers from the four meta-analyses and the two Barwood trials re-checked against abstracts and stated with their statistics (Jeffries g 0.33 / 0.40 / −0.54; Gavel SMD 0.12, P = 0.23, no moderation by concentration, swills or temperature; Wang g 0.27, 16 studies; Zhu SUCRA 76.1 / 66.9 / 6.2%; Barwood 2019 sweat 994 vs 1180 mL, TTE 4.6 vs 2.4 min).
Barwood 2025 result added: whole-body gel did not improve a 20-km TT after a swim (P = 0.34) and INCREASED sweat production — the "topical blunts sweating" claim now rests on Barwood 2019 only.
Riera 2014 and Rinaldi 2023 added as verified primaries for the internal-route and one-hour-precooling claims; the "~4 km" margin removed (not in the abstract).
Cycling trial figures without a primary paper in the base (+9%, +6.0%, +1.8%, +15.6%, 21 °C null, carbohydrate nulls, Tokyo panel) labelled as unverified secondary reports; "disappears at ≥30 °C" for topical labelled unconfirmed.
"Heat-specific" claim softened: Gavel 2024's meta-regression found no temperature moderation; the rule is now labelled base practice. Cross-reference to REF-002 reworded to what its corrected entry carries.
Verification 2026-10-01: only two sentences carry a refId (REF-529 tier framework — exists, Level A classification; REF-002 'heat acclimation, cooling and hyperhydration' — matches the entry's 'Mitigation strategies reviewed'). Both OK.
Verification 2026-10-01: ~16 further sentences cite author-year primaries that are NOT entries in the snapshot; they were not auditable and are counted neither OK nor wrong. Recommend creating REF entries for the 4 meta-analyses and 4 trials so the numbers become checkable.
Verification 2026-10-01: secondary-report figures are honestly labelled in every place they appear; no unlabelled secondary figure found.
Verification 2026-10-01: Applicability line ('a concrete protocol with a dose, an interval') oversells — dose and interval are declared practice in the protocol itself.
Revision 2026-10-01: the eight primary papers are now entries — Jeffries 2019 REF-961, Gavel 2024 REF-962, Wang 2026 REF-963, Zhu 2026 REF-964 (meta-analyses, A- each); Barwood 2019 REF-965, Barwood 2025 REF-966, Riera 2014 REF-967, Rinaldi 2023 REF-968 (small crossover trials, B each). Every author-year claim now carries its refId and level; the entries were added to Cross-References.
Revision 2026-10-01: recommendation rewritten to follow the entries. 'Rinse, don't spray' became 'by mouth, not on the skin — and a cold menthol drink has more behind it than a rinse': the ingestion evidence (REF-961 internal g = 0.40; REF-967, REF-968; REF-964 ranking) is stronger than the rinse evidence (REF-962 pooled null, SMD 0.12), and the topical g = 0.27 does not hold in the heat (REF-963: smaller and not significant at ≥30 °C).The rinse recipe stays, labelled practice; the drink protocol trialled in REF-967 was added. The lead line was changed the same way.
Revision 2026-10-01: numbers checked against the new entries' abstract-based text; effect sizes, P values, SUCRA values, sweat and time figures all match.Corrected: topical at ≥30 °C is 'smaller and not statistically significant' in the abstract (formerly 'could not be confirmed'/'disappears'); 'stricter criteria' (Gavel 2024) and 'the abstract of Gavel 2024 reports no adverse effects' are not in the abstract — removed and withdrawn; 'no better than nothing' is 'no better than the neutral drink'; '7-12 trained men' became 'athletes' (two abstracts do not state sex); 'one of the two mouth-rinse meta-analyses' became the one held.Riera and Rinaldi give no size for the gain in their abstracts, now said next to each.
Revision 2026-10-01: 'only in the heat' weakened further — besides REF-962's null temperature moderator, REF-963 found the topical benefit most evident under temperate conditions. Applicability line no longer promises 'a concrete protocol with a dose, an interval'. Secondary-report figures keep their existing labels (their sources are not held in the base). Level unchanged (B-).