The Hydrating Effects of Hypertonic, Isotonic and Hypotonic Sports Drinks and Waters on Central Hydration During Continuous Exercise: A Systematic Meta-Analysis
Design
Outcome: change in plasma volume (dPV) during 0-180 min of continuous exercise. Drink classes: hypertonic (>300 mOsmol/kg), isotonic (275-300), hypotonic (<275) carbohydrate-electrolyte drinks ([Na+] <50 mmol/L), and water/non-carbohydrate drinks (<40 mOsmol/kg). Random-effects meta-analysis adjusted for osmolality, ingestion rate, metabolic rate, with a weakly informative Bayesian prior.
Key Results
| Drink | Mean dPV during exercise (90% CL) | Rank |
|---|---|---|
| Hypotonic | −6.3% (−7.4, −5.3) | Best — smallest plasma-volume loss |
| Hypertonic | −7.4% (−8.5, −6.3) | Middle |
| Water | −7.5% (−8.5, −6.4) | Middle |
| Isotonic | −8.7% (−10.1, −7.4) | Worst |
Contrasts vs the smallest important effect (0.75%): hypotonic–isotonic 2.3% (p+ = 0.984, very likely superior); hypotonic–hypertonic 1.1% (p+ = 0.71) and hypotonic–water 1.1% (p+ = 0.72), likely superior; hypertonic–isotonic 1.2% (p+ = 0.74); water–isotonic 1.3% (p+ = 0.76); hypertonic–water 0.1% (p+ = 0.12, no difference).
Key Findings
Hypotonic carbohydrate-electrolyte drinks ingested continuously during exercise give the greatest hydration benefit (very likely better than isotonic, likely better than hypertonic and water)
Isotonic drinks — the standard commercial "sports drink" — came out worst for plasma-volume maintenance during exercise
Metabolic rate, ingestion rate, carbohydrate characteristics and electrolyte concentration were generally substantial modifiers of dPV
Window covered: 0-180 min of continuous exercise
Practical Application
Across 28 studies (68 drink treatments), hypotonic carbohydrate-electrolyte drinks preserved plasma volume best during 0-180 min of exercise (−6.3% vs isotonic −8.7%, hypertonic −7.4%, water −7.5%); hypotonic was very likely superior to isotonic and likely superior to hypertonic and water. For a cyclist: a lower-osmolality drink is the better hydration choice during the ride; an isotonic "sports drink" is not automatically the best for hydration.
Not from this paper; the base's own practice: keep the drink hypotonic by diluting a standard carbohydrate mix and adding sodium separately, and test gut tolerance in training before a race; the abstract gives no CHO percentage or sodium concentration recipe.
Hypertonic drinks were not worse than water for hydration in this analysis (hypertonic–water 0.1%, p+ = 0.12). Inference, not tested: any case against them during exercise rests on gut comfort, which this meta-analysis does not address.
Limitations
From the design as the abstract reports it: change in plasma volume (dPV) is a surrogate for hydration — no performance or thermoregulatory outcome was pooled.
The effects are model estimates across 28 studies, adjusted for drink osmolality, ingestion rate, metabolic rate and a weakly informative Bayesian prior; metabolic rate, ingestion rate, carbohydrate characteristics and electrolyte concentration were substantial modifiers of dPV.
Continuous exercise over 0-180 min only; intermittent exercise and longer events are outside the analysis.
Several contrasts are only "likely" (p+ = 0.71-0.76), and hypertonic vs water showed no difference (p+ = 0.12).
How this entry was checked and changed
Content correction 2026-09-27
Table had "Isotonic: moderate, Hypertonic: worst during exercise" -> the abstract's order is hypotonic −6.3% (best), hypertonic −7.4%, water −7.5%, isotonic −8.7% (worst); table rebuilt with the abstract's values and contrasts.
"Sodium in hypotonic drinks helps retain water", "CHO absorbed in the small intestine draws water into the body", "consistent across intensities" -> not in the abstract (metabolic rate was a modifier); removed.
Practical recipe "6-8% CHO + 20-30 mmol/L sodium", "~6% CHO + 20 mmol/L" -> not in the abstract; replaced by labelled house practice without those numbers. Sport line "cycling" -> "general".
2026-09-28: added Limitations from the abstract (Europe PMC PMID 34716905).