The trained-athlete ceiling could move from 90 to 120 g/h
And the 120-200 g/h now circulating in the field is a practice the authors say the evidence does not yet support
Key Findings
The position they revise: the 2016 ACSM sport-nutrition guidelines recommend ≤90 g/h, from multiple-transportable carbohydrates (glucose/fructose mixtures), for exercise longer than ~2.5-3 h.
The proposal, in their words: "the upper limit of CHO intake could increase from 90 to 120 g/h (at least for trained participants)" — because both exogenous and whole-body carbohydrate oxidation rates can be increased at those higher ingestion rates.
120-200 g/h is reported as a field trend, explicitly NOT as a recommendation. Endurance athletes are consuming and experimenting with 120-200 g/h in training and racing, and "the efficacy of such doses is not yet substantiated by current scientific research."
The authors suggest such absolute doses may also modulate performance determinants such as durability and economy across cycling, marathon running and ultra-endurance exercise — stated as a possibility, not a finding.
Mechanism as the abstract frames it: the ergogenic effect is likely mediated by liver (and potentially muscle) glycogen sparing, maintained plasma glucose and whole-body carbohydrate oxidation, so the required intensity can be held longer.
The review covers factors affecting exogenous oxidation (blend, ratio, format, environment) and presents an updated model for personalising intake, including the authors' own practice insights.
Detail from the body of the review (full text, checked 2026-09-27): at ≤60 g/h a single source (glucose or a glucose polymer) is typically recommended; above that, a fructose-to-glucose ratio between 0.6 and 1.0 is given as optimal for exogenous availability and oxidation.
Detail from the body of the review (full text, checked 2026-09-27): in races with mean finish times of ~12-24 h reported intakes are 30-66 g/h (the review defines ultraendurance as >6 h), with elite case studies reaching 80-100 g/h in 100-mile and 24-h competitions (corrected 2026-10-12, full text).
Detail from the body of the review (full text, checked 2026-09-27): across 8 gut-training interventions, repeated carbohydrate feeding during exercise at 30-90 g/h reduced gut discomfort without measurable change in gastric emptying or epithelial injury markers, read by the authors as transporter (SGLT1/GLUT5) adaptation.
Detail from the body of the review (full text, checked 2026-09-27): the authors audit the underlying studies — >80% in Tier 1-2 participants, <5% in Tier 4-5, <3% in female-only cohorts.
Practical Application
This 2026 perspective review proposes raising the upper limit of carbohydrate intake from the 2016 ACSM ≤90 g/h to 120 g/h, at least for trained athletes, because exogenous and whole-body carbohydrate oxidation rise at those rates — while the 120-200 g/h some athletes now use is "not yet substantiated" by research. Quote 120 g/h as a proposed trained-athlete ceiling, and say plainly that higher intakes are what some athletes are doing, not what has been shown to work.
Pick the carbohydrate source by the rate, not by habit. full text, checked 2026-09-27: at or below ~60 g/h a single source is enough; above it, a fructose-to-glucose ratio of 0.6-1.0. Inference, not tested: that is nearer 1:1 than the 1:2 fructose:glucose blends still widely sold.
Earn the rate in training, repeatedly. full text, checked 2026-09-27: gut training at 30-90 g/h reduced discomfort in the reviewed interventions. Inference, not tested: an intestinal adaptation is unlikely to appear from a single session or to persist if skipped.
Drop the target for very long days. full text, checked 2026-09-27: reported intakes in races of ~12-24 h are 30-66 g/h, elite case studies in 100-mile and 24-h events 80-100 g/h (corrected 2026-10-12, full text) — prescribing 120 g/h for a day that long goes beyond what has been measured.
Name the population when the athlete is not in it. The abstract itself calls for research in both male and female athletes and in elite athletes under real-world conditions; for a woman or a world-class athlete this ceiling is an extrapolation and should be said as one.
Limitations
A perspective review, not a systematic one — no protocol, no risk-of-bias appraisal, no pooled effect sizes. The 120 g/h figure is the authors' proposed revision, not a meta-analytic estimate.
The revision is argued from oxidation, not from finishing time. Higher ingestion raises measured carbohydrate oxidation; the abstract does not claim a corresponding performance gain of 120 over 90 g/h.
The abstract flags the need for research in female and elite athletes under ecologically valid protocols; applying the ceiling upward to a world-class athlete or across to a woman is extrapolation.
Inference, not tested: individual oxidation capacity and body size vary, so a single grams-per-hour figure fits some athletes badly.
Silent on carbohydrate for short sessions and on the post-exercise window.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from A: narrative perspective review without systematic method; kept at B+ as an authoritative review from a leading carbohydrate-metabolism group
Content correction 2026-09-27
Level A -> B+ (perspective/narrative review without systematic method, authoritative group).
Title "ceiling moves" -> "could move", as the abstract words it.
"ACSM 30-60 g/h for 1-2.5 h" removed (not in the abstract); fructose ratio, ultra intakes, gut-training and the >80%/<5%/<3% participant audit labelled full text; "absolute rates come from male athletes" and "does not survive being skipped" removed or labelled inference.
Lead now states the paper's proposal with its numbers; cross-reference descriptions aligned with the corrected entries.
Lines labelled "full text; not checked against the abstract" were checked against the open-access full text (PMC13197957): every number in them appears there; relabelled "full text, checked 2026-09-27".
Full text (checked 2026-10-12)
The authors cite the 2016 ACSM guidelines as 30–60 g/h for endurance exercise and "stop and start" sports lasting 1–2.5 h and 90 g/h for exercise >2.5 to 3 h, and define ultraendurance as durations >6 h (full text).
To oxidize 90 g/h (1.5 g/min) athletes should likely ingest 90–120 g/h, because oxidation efficiency is not uniform; ingesting 120–144 g/h of multiple-transportable blends gave average peak exogenous oxidation of 1.60–1.75 g/min (96–105 g/h) (full text).
In a multisite trial of 51 recreationally trained cyclists and triathletes (0–120 g/h, 1:1:1 glucose, maltodextrin, fructose), the dose-response on performance was curvilinear with diminishing enhancement above 78 g/h (full text).
In male elite marathoners (PB faster than 2 h 30 min), 120 g/h gave a 3% better running economy than 60 g/h, but peak nausea, stomach fullness and abdominal cramps were greatest at 120 g/h (full text).
Exogenous CHO oxidation falls by 20%–30% in the heat and by ∼20%–50% above 4000 m, is unaffected by cold, and the authors still advise unchanged intake recommendations in all three conditions (full text).
Fluids, gels and chews at 108–120 g/h gave comparable exogenous oxidation, whereas a solid bar at 93 g/h gave lower oxidation late in exercise and more GI symptoms, so a mixed-format approach is advised (full text).
Cyclists over 70 kg oxidized more of a 90 g/h glucose dose than smaller cyclists (∼45 vs ∼33 g/h), and a personalized glucose dose of 65 ± 10 g/h gave peak exogenous oxidation indistinguishable from 90 g/h (full text).
Correction: the entry says field intakes "in ultra-endurance events (>6 h)" and "past 6 h" are 30-66 g/h; the paper gives "between 30 and 66 g/h" for races with mean finish times ∼12–24 h (full text).