Modelling Sodium Requirements of Athletes Across a Variety of Exercise Scenarios
Identifying When to Test and Target, or Season to Taste
Protocol
Design: a mathematical modelling study with no participants — the sodium intake needed to keep plasma [Na+] stable was estimated from fluid and electrolyte intakes and losses (from the abstract); the inputs varied and the competition scenarios are listed under Model below.
Outcome: the sodium replacement (as a proportion of losses) required in each modelled scenario.
Model
Target: keep plasma [Na+] stable for a given total body water; the sodium intake that achieves this can be estimated from fluid and electrolyte intakes and losses
Inputs varied: sweat [Na+] 20-80 mmol/L, sweat rate 0.5-2.5 L/h, fluid replacement 10-90% of losses; also exercise duration, body mass, baseline plasma [Na+], sweat [K+]
Competition scenarios: soccer match, elite marathon, 160 km ultramarathon; fluid replacement set to end at 2% total-body-water loss
Key Decision Framework
| Scenario (as modelled) | Targeted sodium / sweat testing needed? |
|---|---|
| Soccer match, elite marathon | No, in all realistic scenarios — season to taste |
| 160 km ultramarathon, sweat Na+ ≥40 mmol/L, >80% of fluid losses replaced | Yes — ≥47% sodium replacement required |
| Any scenario with <80% fluid replacement or sweat Na+ <40 mmol/L | Targeted replacement not indicated by the model |
Key Findings
Sodium requirements were driven by the proportion of fluid losses replaced and by sweat [Na+]; sweat rate, duration and body mass had minimal or no influence, baseline plasma [Na+] and sweat [K+] only a minor one
Only prolonged exercise combined with aggressive fluid replacement (>80%) and sweat [Na+] ≥40 mmol/L makes sweat testing and significant targeted sodium replacement appear necessary
Even in the ultramarathon, the required quantity stays below 50% of losses unless sweat rate ≥1.5 L/h, sweat [Na+] ≥40 mmol/L and fluid replacement ≥90%
Practical Application
The model finds sodium replacement unnecessary for a soccer match or an elite marathon in every realistic scenario; only a 160 km ultramarathon with sweat sodium ≥40 mmol/L and >80% of fluid losses replaced needed targeted sodium (≥47% of losses). For most cyclists: sodium to taste from normal sports drinks; a sweat test earns its cost only for very long events where you plan to drink back most of what you sweat.
Inference, not tested: the marathon and ultramarathon scenarios bracket typical road races and gran fondos on one side and gravel ultras on the other; the paper itself sets no hour threshold.
Do not chase sodium alone: the driver is the fraction of fluid replaced — drinking less than ~80% of losses keeps plasma sodium stable without targeted salt in the model.
Limitations
From the design as the abstract reports it: a mathematical model with no participants — the predictions were not tested against measured plasma sodium or health outcomes in athletes in this paper.
Only three competition scenarios were modelled (soccer match, elite marathon, 160 km ultramarathon running); no cycling event was modelled.
The model's answer depends on sweat [Na+] and the fraction of fluid losses replaced, which an athlete has to measure or estimate.
How this entry was checked and changed
Why this level
regraded 2026-09-27 from B: mathematical modelling of plasma-sodium balance, not validated against outcomes in athletes (rubric: modelling)
Content correction 2026-09-27
Decision table used a "4 hours" threshold and "events >4 h replacing >80%" -> the abstract defines scenarios by event (soccer, elite marathon, 160 km ultramarathon), not by hours; table rebuilt on the abstract's scenarios and thresholds (≥40 mmol/L, >80%, ≥47%, 1.5 L/h, ≥90%).
"Most cycling events (2-3 h) do not require testing", "gravel 6-8+ h with 1.5+ L/h intake: worth testing", "5-7 h borderline, >1.2 L/h" -> not in the abstract; replaced by a labelled inference.
"Over-supplementing sodium without matching fluid can worsen dehydration" -> not in the abstract; removed.
Journal line said "Sports Medicine, 52" -> the DOI is European Journal of Sport Science. Level B -> C per rubric (modelling study). Sport line "cycling" -> "general".
2026-09-28: added Protocol and Limitations from the abstract (Europe PMC PMID 35616504); n not applicable (modelling study, no participants).