Sauna for type 2 diabetes mellitus
Evidence review
Overview
Heat therapy has been proposed to improve insulin sensitivity, but trials in type 2 diabetes have not shown glycemic benefit. Single sessions of infrared sauna or hot-water immersion did not improve glucose tolerance or insulin sensitivity in four small crossover trials, and heating close to a glucose load transiently raised glucose. Two uncontrolled studies of 8-10 hot-water immersions over 2 weeks reported improved fasting insulin sensitivity and lower systolic blood pressure without changes in glucose or endothelial function, and meta-analyses found no significant effect on HbA1c or fasting glucose. In a cross-sectional study of 1,297 patients, frequent hot-tub bathing was associated with slightly lower HbA1c.
Key findings
- Meta-analysis of 5 studies in T2D: HbA1c WMD -0.549% (95% CI -1.262 to 0.164; p=0.131), fasting glucose WMD -0.290 mmol/L (p=0.355), not significant (Sebők 2021).
- RCT meta-analysis (20 trials, mixed populations): no significant pooled effect of passive heating on fasting glucose or HbA1c (Hamaya 2025).
- Acute crossover trials in T2D (n=12-13): no change in glucose tolerance or insulin sensitivity after hot-water immersion (James 2021; Behzadi 2022); glucose iAUC higher after one infrared sauna session (17.7 vs 14.8 mmol/L/120 min; P<0.001; Schenaarts 2024).
- 8-10 hot-water immersions (104 F, 1 h) over 14 days, n=14, uncontrolled: fasting insulin sensitivity improved (QUICKI P=0.03), fasting glucose unchanged (P=0.83); systolic BP fell (P=0.03), FMD unchanged (P=0.43) (James 2023, 2024).
- Cross-sectional, 1,297 patients: HbA1c 7.10% with hot-tub bathing >=4/week vs 7.36% with <1/week (P=0.012) (Katsuyama 2022).
- Insulin-treated patients: sauna accelerated insulin absorption by 110% and lowered glucose by 3.0-3.3 mmol/L (54-59 mg/dL) vs control day (safety).
Every study we reviewed
6 found no effect · 3 found benefit · 3 mixed results · 1 found harm
Each study was checked on PubMed. The colored edge shows the direction of the result. The research library lists 32 studies on type 2 diabetes, across all heat and cold therapies, including case reports and older work not graded here.
A Single Sauna Session Does Not Improve Postprandial Blood Glucose Handling in Individuals with Type 2 Diabetes Mellitus: A Cross-Over, Randomized, Controlled Trial.
Schenaarts L et al. Exp Clin Endocrinol Diabetes. 2024. PubMed 39209309 · DOI · Free full text
- Who
- Older adults with T2D (mean age 69; BMI 27.5)
- What
- Single infrared sauna session (heated vs thermoneutral) before a 2-h OGTT
- Protocol
- 140 °F · 40 min · single session/wk
- Measured
- OGTT glucose iAUC, insulin, composite insulin sensitivity
- Result
- Glucose iAUC 17.7 vs 14.8 mmol/L/120 min (P<0.001, higher after heating); insulin 380 vs 376 pmol/L/120 min (P=0.93); insulin sensitivity index 4.5 vs 4.5 (P=0.67).
- Limitations
- Single session; small; does not address repeated use.
An acute bout of whole body passive hyperthermia increases plasma leptin, but does not alter glucose or insulin responses in obese type 2 diabetics and healthy adults.
Rivas E et al. J Therm Biol. 2016. PubMed 27264884 · DOI
- Who
- Obese adults with T2D not on insulin (n=9) and healthy controls (n=9)
- What
- Single 1-h head-out hot-water immersion (whole-body passive hyperthermia) vs control rest
- Protocol
- 103 (water) °F · 60 min · single session/wk
- Measured
- 75 g OGTT 24 h later: glucose, insulin, C-peptide, leptin
- Result
- No change in glucose, insulin or C-peptide AUC in either group; fasting leptin +17% in T2D (P<0.01).
- Limitations
- Hot-water immersion rather than sauna; single exposure; very small.
Non-acute effects of passive heating interventions on cardiometabolic risk and vascular health: systematic review and meta-analysis of randomized controlled trials.
Hamaya R et al. Am J Prev Cardiol. 2025. PubMed 41049507 · DOI · Free full text
- Who
- 20 RCTs of passive heating >=1 week in adults
- What
- Passive heating (hot water bathing, sauna, hot yoga, local heating), 2-15 weeks
- Protocol
- 2-15
- Measured
- Fasting glucose, HbA1c and other cardiometabolic markers
- Result
- No significant pooled effects on fasting glucose or HbA1c (estimates not given in abstract).
- Limitations
- Few trials in diabetic populations; mixed modalities.
The effects of repeated thermal therapy on quality of life in patients with type II diabetes mellitus.
Beever R et al. J Altern Complement Med. 2010. PubMed 20569036 · DOI
- Who
- Patients with T2D at a rural Canadian health center
- What
- Far-infrared sauna 20 min, 3 times/week for 3 months (interrupted time series)
- Protocol
- 20 min · 3/wk · 12 wk
- Measured
- SF-36v2 quality of life; stress and fatigue VAS
- Result
- Physical health, general health and social functioning SF-36 indices and stress and fatigue VAS improved (effect sizes not given in abstract).
- Limitations
- Uncontrolled; sample size not stated in abstract; no glycemic outcomes reported in abstract.
Repeated thermal therapy improves impaired vascular endothelial function in patients with coronary risk factors.
Imamura M et al. J Am Coll Cardiol. 2001. PubMed 11583886 · DOI · Free full text
- Who
- Men with >=1 coronary risk factor (hypercholesterolemia, hypertension, diabetes or smoking; n=25) and healthy controls (n=10)
- What
- Far-infrared sauna daily for 2 weeks (risk group only)
- Protocol
- 140 °F · 15 min · daily/wk · 2 wk
- Measured
- Brachial flow-mediated dilation
- Result
- FMD 4.0% to 5.8% (p<0.001) in the risk-factor group; nitroglycerin-mediated dilation unchanged.
- Limitations
- Not restricted to diabetes; no randomized control; young men (mean 38 y); surrogate endpoint.
The effect of repeated hot water immersion on insulin sensitivity, heat shock protein 70, and inflammation in individuals with type 2 diabetes mellitus.
James TJ et al. Am J Physiol Endocrinol Metab. 2023. PubMed 37938179 · DOI
- Who
- Adults with T2D
- What
- Repeated hot-water immersion, 8-10 one-hour sessions within 14 days (pre-post, no control group)
- Protocol
- 104 (water) °F · 60 min · 4-5/wk · 2 wk
- Measured
- Fasted and 2-h post-OGTT insulin sensitivity, glucose, insulin, eHSP70, inflammatory markers, resting metabolic rate
- Result
- Fasted insulin sensitivity (QUICKI) improved (P=0.03) and fasted insulin fell (P=0.04); fasting glucose unchanged (P=0.83); postprandial insulin sensitivity (P=0.19), glucose (P=0.40) and insulin (P=0.47) unchanged; IL-6 and IL-10 unchanged; resting metabolic rate -6.63% (P<0.05).
- Limitations
- Uncontrolled pre-post design; small; 2 weeks only; no HbA1c outcome.
The effect of repeated hot water immersion on vascular function, blood pressure and central haemodynamics in individuals with type 2 diabetes mellitus.
James TJ et al. J Therm Biol. 2024. PubMed 39642665 · DOI · Free full text
- Who
- Adults with T2D (same protocol and research group as James 2023)
- What
- Repeated hot-water immersion, 8-10 one-hour sessions within 14 days (pre-post, no control group)
- Protocol
- 104 (water) °F · 60 min · 4-5/wk · 2 wk
- Measured
- Brachial FMD, cutaneous microvascular function, blood pressure, cardiac index, stroke volume index, resting heart rate
- Result
- Systolic BP reduced (P=0.03), diastolic BP unchanged (P=0.56); resting heart rate, cardiac index and stroke volume index reduced (all P<0.02); brachial FMD (P=0.43) and forearm microvascular function (P=0.51-0.86) unchanged.
- Limitations
- Uncontrolled; small; likely overlapping participants with James 2023; effect sizes not given in abstract.
Acute effect of passive heat exposure on markers of cardiometabolic function in adults with type 2 diabetes mellitus.
Behzadi P et al. J Appl Physiol (1985). 2022. PubMed 35323077 · DOI
- Who
- Adults with T2D (mean age 62, BMI 30.1)
- What
- Single hot-water immersion vs thermoneutral immersion, randomized order
- Protocol
- 105.8 (water); control 93.2 °F · 90 (control); hot until core temperature >=101.3 F (38.5 C) held for 60 min min · single session/wk
- Measured
- Insulin sensitivity (oral minimal model) on OGTT 60 min after immersion; FMD; reactive hyperemia; HSP70/90; cytokines
- Result
- No acute change in insulin sensitivity (-0.3 dL/kg/min per uU/mL, 95% CI -0.9 to +0.2; P=0.18) or FMD (-1.0%, 95% CI -3.6 to +1.6; P=0.56); heart rate +34 beats/min and IL-6 +1.38 pg/mL vs thermoneutral.
- Limitations
- Single session; small.
Timing of acute passive heating on glucose tolerance and blood pressure in people with type 2 diabetes: a randomized, balanced crossover, control trial.
James TJ et al. J Appl Physiol (1985). 2021. PubMed 33411640 · DOI
- Who
- Adults with T2D
- What
- One-hour hot-water immersion 30 min before or 30 min after starting an OGTT vs OGTT alone (randomized, balanced crossover)
- Protocol
- 104 (water) °F · 60 min · single session/wk
- Measured
- Glucose AUC, insulin sensitivity, eHSP70, energy expenditure, blood pressure
- Result
- Glucose AUC unchanged (control 1,677, heat-before 1,797, heat-after 1,662 AU; P=0.28); insulin sensitivity unchanged (P=0.15); eHSP70 and energy expenditure increased; diastolic BP reduced when heating followed glucose ingestion (P<0.01).
- Limitations
- Acute single sessions; small.
Obese type 2 diabetics have a blunted hypotensive response to acute hyperthermia therapy that does not affect the perception of thermal stress or physiological strain compared to healthy adults.
Rivas E et al. Physiol Behav. 2016. PubMed 27570191 · DOI
- Who
- Obese adults with T2D (n=11, mean HbA1c 7.5%) and non-obese non-diabetic controls (n=9)
- What
- Single 60-min whole-body hot-water immersion (passive hyperthermia) vs 2-h resting control, random order
- Protocol
- 60 min · single session/wk
- Measured
- Blood pressure, thermal sensation, physiological strain
- Result
- Diastolic BP fell 27% in T2D and 33% in controls vs control condition; reduction in mean arterial pressure AUC attenuated in T2D (12%) vs controls (30%) (P<0.01); thermal sensation and physiological strain not different between groups.
- Limitations
- Single session; small; physiological endpoints only; likely overlapping sample with Rivas 2016 (J Therm Biol).
Habitual Hot-Tub Bathing and Cardiovascular Risk Factors in Patients With Type 2 Diabetes Mellitus: A Cross-Sectional Study.
Katsuyama H et al. Cardiol Res. 2022. PubMed 35836731 · DOI · Free full text
- Who
- Japanese outpatients with T2D
- What
- Self-reported habitual hot-tub bathing frequency (>=4/week vs 1-3/week vs <1/week)
- Protocol
- >=4 vs <1/wk
- Measured
- HbA1c, BMI, blood pressure
- Result
- HbA1c 7.10% vs 7.20% vs 7.36% (P=0.012); BMI 25.5 vs 26.0 vs 26.7 (P=0.025); diastolic BP 73 vs 75 vs 77 mmHg (P=0.001); bathing frequency remained a significant determinant after multivariable adjustment.
- Limitations
- Cross-sectional association only; self-reported exposure; residual confounding; small absolute differences.
Heat therapy shows benefit in patients with type 2 diabetes mellitus: a systematic review and meta-analysis.
Sebők J et al. Int J Hyperthermia. 2021. PubMed 34808071 · DOI · Free full text
- Who
- 5 interventional studies in patients with T2D
- What
- Passive heat therapy (mixed modalities)
- Measured
- Change in HbA1c, fasting glucose, triglycerides
- Result
- HbA1c WMD -0.549% (95% CI -1.262 to 0.164; p=0.131); fasting glucose WMD -0.290 mmol/L (95% CI -0.903 to 0.324; p=0.355); triglycerides WMD 0.035 mmol/L (95% CI -0.130 to 0.200; p=0.677). No significant differences.
- Limitations
- Only 5 small studies, pre-post comparisons; authors call for longer RCTs.
Passive heating and glycaemic control in non-diabetic and diabetic individuals: A systematic review and meta-analysis.
Maley MJ et al. PLoS One. 2019. PubMed 30901372 · DOI · Free full text
- Who
- 14 RCTs of passive heating with glycemic outcomes in non-diabetic and diabetic individuals
- What
- Passive heating (mixed modalities)
- Measured
- Glucose concentration during or after a glucose load
- Result
- When heating followed a glucose load, glucose was higher during heating in diabetic individuals (SMD 0.27, 95% CI 0.02-0.52; P=0.030) and non-diabetic individuals (SMD 0.75, 95% CI 0.48-1.02). Too few diabetic studies to assess other comparisons.
- Limitations
- Acute studies; heterogeneous; transient glucose intolerance rather than a long-term outcome.
Protocols used in the studies
Acute trials used one 40-min infrared sauna session at 140 F (60 C), or one hot-water immersion at 104-106 F (40-41 C) lasting 60-90 min with core temperature raised to about 101.3 F (38.5 C). The repeated hot-water immersion studies used 8-10 one-hour sessions at 104 F (40 C) within 14 days with rectal temperature held at 101.3-102.2 F (38.5-39.0 C). The quality-of-life study used far-infrared sauna for 20 min 3 times/week for 3 months; the endothelial study used 140 F far-infrared sauna for 15 min plus 30 min blanket rest daily for 2 weeks. The cross-sectional study recorded habitual home hot-tub bathing frequency.
What is still unknown
Effect of repeated sauna or hot-water immersion over months on HbA1c, insulin sensitivity and cardiovascular outcomes in randomized controlled trials; whether the cross-sectional HbA1c association reflects confounding; safe protocols for patients with neuropathy or autonomic dysfunction.
Safety
Cautions for sauna in this condition
- In insulin-treated patients, sauna (2 x 25 min at 185 F) accelerated subcutaneous insulin absorption by 110% and lowered blood glucose by 3.0-3.3 mmol/L vs a control day (PMID 7000239); hypoglycemia risk with insulin or sulfonylureas.
- A second-degree burn occurred in a patient with advanced diabetic neuropathy using a portable sauna (PMID 36203760); impaired sensation increases burn risk.
- Heating close to a meal or glucose load transiently raised glucose: higher post-meal glucose excursion after an infrared session (PMID 39209309) and SMD 0.27 in diabetic individuals in meta-analysis (PMID 30901372).
- One hour of hot-water immersion lowered diastolic BP by 27% in obese patients with T2D (PMID 27570191) and raised heart rate by about 34 beats/min (PMID 35323077); orthostatic hypotension and cardiac strain are relevant with antihypertensives or coronary disease.
For clinicians
Suggested rationale
Current trials do not show that heat therapy improves glycemic control in type 2 diabetes (Sebők 2021 meta-analysis; Schenaarts 2024); short hot-water immersion programs produced modest changes in fasting insulin sensitivity and systolic blood pressure without changes in glucose (James 2023, 2024). Any use would rest on cardiovascular rather than glycemic evidence, with attention to hypoglycemia with insulin and burn risk with neuropathy.
Suggested wording only. The treating clinician decides whether it reflects their clinical judgment for this patient.
Typical ICD-10 codes: E11.9, E11.65. Examples only; the clinician assigns the code.
What a letter of medical necessity should containWho can write it: physicians, NPs, PAs and others