Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2026 Jun 9;28(9):7825–7833. doi: 10.1111/dom.70958

Pre‐Exercise Galactose or Lactose Reduces Glycaemic Excursions During Exercise in Adults With Type 1 Diabetes: A Randomised, Double‐Blind, Crossover Trial

Karoline R Schmidt 1, Rakel F Johansen 1,✉, Simon B Riis 2, Mads S Larsen 2, Henrik Støvring 1,3, Gareth A Wallis 4, David Rowlands 5, Lars C Gormsen 6, Esben Søndergaard 1,3
PMCID: PMC13449015  PMID: 42264961

ABSTRACT

Aims

To test whether pre‐exercise galactose and lactose reduce glycaemic excursions during exercise compared with dextrose while maintaining protection against hypoglycaemia compared with placebo in adults with Type 1 diabetes

Methods

In a randomised, double‐blind, 4‐condition crossover trial, 14 adults with Type 1 diabetes completed four 60‐min cycling sessions. Thirty minutes before exercise, participants consumed 20 g of dextrose, lactose, galactose, or placebo. The primary outcome was time to hypoglycaemia. Secondary outcomes included time in, below, and above range and glucose excursions.

Results

Seven hypoglycaemic events occurred (none after dextrose, two after lactose, one after galactose, and four after placebo; p = 0.42), with no difference in time to hypoglycaemia. Time in range was lower for dextrose (48% [95% CI 40–57]) compared with lactose (70% [95% CI 62–78]), galactose (71% [95% CI 63–79]), and placebo (86% [95% CI 80–91]) (all p ≤ 0.01). Time below range was lower for dextrose (0% [95% CI 0–3]), lactose (1% [95% CI 0–5]), and galactose (1% [95% CI 0–5]) compared with placebo (8% [95% CI 4–14]) (all p < 0.05). Time above range was higher for DEX (52% [95% CI 44–60]), LAC (29% [95% CI 21–37]), and GAL (28% [95% CI 21–36]) compared with PLA (7% [95% CI 3–12], all p < 0.01).

Conclusions

Pre‐exercise galactose and lactose provide more stable glycaemic responses than dextrose and reduce time below range compared with placebo. These findings support their use as effective pre‐exercise nutritional strategies in individuals with Type 1 diabetes.

Trial Registration: ClinicalTrials.gov identifier: NCT05557227.

Keywords: clinical trial, exercise intervention, hypoglycaemia, Type 1 diabetes

1. Introduction

Exercise improves glycaemic control in individuals with Type 1 diabetes [1, 2, 3], and is associated with improvements in cardiovascular risk factors [4]. In addition, regular physical activity promotes well‐being and supports body weight management [1]. However, plasma glucose fluctuations and the fear of hypoglycaemia in relation to exercise are obstacles, which may prevent individuals with Type 1 diabetes from complying with exercise recommendations [5, 6, 7].

Consuming carbohydrates before exercise is recommended as a strategy to decrease the risk of exercise‐induced hypoglycaemia in Type 1 diabetes [1]. However, if carbohydrates are consumed with insulin, a rapid drop in blood glucose can occur during exercise, which may result in hypoglycaemia. If carbohydrates are consumed without insulin administration, this may result in hyperglycaemia prior to exercise, and a rapid drop in plasma glucose may still be seen during exercise [8]. This places individuals with Type 1 diabetes in a difficult position in regards to choosing the right insulin and carbohydrate strategy for exercise. As a result, there is a need for simple and flexible pre‐exercise strategies that can be applied without complex insulin adjustments and without the need for carbohydrate intake during exercise. One potential solution is consumption of foods or beverages with a low glycaemic index prior to exercise [9]. This approach has been supported by a non‐blinded cross‐over trial, which demonstrated that pre‐exercise fructose supplementation compared to water reduced the risk of hypoglycaemia while avoiding hyperglycaemia in individuals with Type 1 diabetes [10]. While fructose has been shown to be effective and widely used, alternative low‐glycaemic index carbohydrate sources such as galactose and lactose may also be of interest.

Lactose, a disaccharide composed of glucose and galactose, and the monosaccharide galactose, are potential alternative pre‐exercise low‐glycaemic index carbohydrate sources with estimated glycaemic indices of 47 and 20, respectively [11]. Intriguingly, galactose oxidation rates during exercise are only 50%–60% compared to glucose, presumably due to hepatic storage and/or a requirement for conversion to glucose in the liver before oxidation [12]. In addition, lactose ingestion prior to and during exercise allowed for higher fat oxidation and lower carbohydrate oxidation compared to sucrose in healthy subjects [13]. These metabolic features of milk‐derived sugars could potentially prevent pre‐exercise hyperglycaemia while providing more sustained protection against hypoglycaemia during and after exercise. To our knowledge, this is the first randomised controlled trial to examine the glycaemic effects of milk‐derived sugars during exercise in individuals with Type 1 diabetes.

Therefore, the aim of this study was to investigate the efficacy of lactose and galactose as low‐glycaemic index pre‐exercise carbohydrate sources for individuals with Type 1 diabetes. We hypothesised that lactose and galactose, when consumed 30 min prior to endurance exercise, would reduce plasma glucose excursions compared with dextrose while maintaining protection against hypoglycaemia compared with placebo.

2. Materials and Methods

2.1. Study Design and Participants

This double‐blind, randomised crossover trial (ClinicalTrials.gov NCT05557227) included 14 physically active adults with Type 1 diabetes. Participants were recruited from our outpatient clinic, a Facebook group for athletes with Type 1 diabetes, and from a regional database [14].

Inclusion criteria were age 18–65 years, diabetes duration > 2 years, HbA1c < 70 mmol/mol, and ≥ 150 min/week of moderate‐to‐vigorous physical activity. Exclusion criteria included diabetes complications, hypoglycaemia unawareness, glucose modifying medication other than insulin (e.g., SGLT2 inhibitors or GLP‐1 receptor agonists), lactose intolerance, galactosemia, heart disease, and other conditions precluding safe participation in the study. Screening included medical history, physical examination, electrocardiogram, and blood tests.

Participants completed four randomised study days, each involving a 60‐min cycling session at 50% W max, after consuming one of four study beverages: 20 g dextrose (d‐glucose; DEX), 20 g lactose (LAC), 20 g d‐galactose (GAL), or 15 g erythritol (noncaloric placebo; PLA), each dissolved in 300 mL of water. Study days were separated by 1–4 weeks.

Randomisation was achieved using computer‐generated random numbers with a 1:1:1:1 allocation ratio, without restrictions such as blocking or stratification [15]. Personnel enrolling participants had no access to the random allocation sequence.

The primary outcome was time to hypoglycaemia. Secondary outcomes included glycaemic excursions, time in range (3.9–10.0 mmol/L), time below range (< 3.9 mmol/L), and time above range (> 10.0 mmol/L), as well as glucose slopes, glucose infusion volume administered to maintain plasma glucose above 5 mmol/L following a hypoglycaemic event, and respiratory exchange ratio (RER).

The study was approved by the Central Denmark Region Ethics Committee, and all participants provided written informed consent. There were no changes to the trial protocol after study initiation.

2.2. Incremental Exercise Test for Determination of VO2max and W max

Participants completed a maximal incremental cycling test within 1 month before the first study day to determine VO2max and W max. The test was performed on a SRM Ergometer (SRM International, Jülich, Germany) and using a Quark CPET system (Cosmed, Rome, Italy) to measure respiratory gas exchange. After a 15‐min warm‐up, power output was increased by 25 W every 60 s, starting at 125 W for men and 75 W for women, until volitional exhaustion. Exercise intensity during study days was set at 50% W max.

2.3. Pre‐Study Day Preparation

Participants were instructed to avoid alcohol and strenuous exercise 48 h prior to each study day. They kept a food and insulin diary to replicate their food intake the day before the study. Minor deviations were allowed. On study days, participants using multiple daily injections (MDI) administered their usual basal insulin, while those using continuous subcutaneous insulin infusion (CSII) maintained their usual basal insulin settings without reductions. No bolus insulin was allowed within 4.5 h before study day initiation, and an overnight fast from midnight was mandatory. If hypoglycaemia occurred between midnight and study day initiation, participants were instructed to consume 4 dextrose tablets (12 g of dextrose) (Dextro Energy GmbH & Co, Krefeld, Germany), repeating the dose after 15 min if needed. Study days were rescheduled if plasma glucose was above 15 mmol/L at study day initiation or if severe hypoglycaemia (glucose < 2.8 mmol/L) had occurred within the last 24 h.

2.4. The Interventions

The intervention beverages contained erythritol (Zerose erythritol, Cargill, US), dextrose monohydrate (C‐Dex, Cargill, US), lactose (Variolac 992, Arla Foods Ingredients Group P/S, Aarhus, Denmark), or galactose (d(+)‐Galactose, AppliChem GmbH, Darmstadt, Germany), providing 20 g of carbohydrate (~335 kJ) for all caloric interventions.

For dextrose monohydrate, the contribution of water of crystallisation was accounted for to ensure equivalent carbohydrate content across conditions.

All beverages were pretested in a blinded manner and matched for taste and appearance. The placebo beverage was sweetened with 15 g erythritol to best match the taste.

Products were supplied as identical, numbered powdered sachets to maintain double blinding. Beverages were dissolved in 300 mL of water immediately before ingestion and administered by blinded study personnel.

2.5. Procedures

An overview of the study protocol is illustrated in Figure 1. The study was conducted at a clinical research unit at Steno Diabetes Center Aarhus, Denmark. Participants arrived between 6:30 and 8:00 AM after the overnight fast. The diary was reviewed to confirm protocol adherence. Catheters were placed in the antecubital veins for blood sampling and glucose infusion. The participants rested in bed, apart from during the exercise bout. Thirty min prior to exercise, participants consumed one of the four supplements. The 60‐min exercise session (50% W max) was then performed in a temperature‐controlled laboratory with fans, followed by another 90 min of rest.

FIGURE 1.

FIGURE 1

Open in a new tab

Study protocol. The protocol was repeated over four separate study days for each participant in a randomised, double‐blind, crossover design. On each study day, participants ingested one of four supplements: 20 g dextrose, lactose, galactose, or a noncaloric placebo dissolved in water, 30 min prior to a 60‐min cycling session performed at 50% W max. Resting phases before and after exercise are indicated.

Plasma glucose and lactate were measured every 10 min, and every 5 min if plasma glucose fell below 5 mmol/L. Free fatty acids (FFA), plasma galactose, and glycerol were measured at t = −40, 0, 15, 30, 45, 60, 90, 120, and 150 min. Heart rate was continuously monitored during exercise with a Polar H10 heart rate monitor (Polar Electro Oy, Kempele, Finland), and respiratory gas exchange was measured in 5‐min intervals at t = −40, 10, 15, 35, 55, and 90 min. Glucose infusion (5%) was initiated if plasma glucose ≤ 3.9 mmol/L and was continued to maintain levels just above 5 mmol/L during the rest of the study day. The Borg scale of perceived exertion (17) was assessed during exercise at 30 and 60 min. After exercise, participants completed a questionnaire concerning well‐being and adverse symptoms (Figure S1).

2.6. Laboratory Procedures

Laboratory analyses Plasma glucose and lactate were measured in real time using a YSI 2500 Glucose/Lactate Analyser (Boston Industries, Walpole, USA). Blood samples were placed on ice, centrifuged (4°C, 10 min), and stored at −80°C. Free fatty acids were measured using a NEFA‐HR (2) assay (Fujifilm Wako, Osaka, Japan). Galactose was determined using a colorimetric assay (Thermo Fisher Scientific, Waltham, USA), and glycerol using Randox reagents on an Indiko Plus analyser (Thermo Fisher, Waltham, USA).

2.7. Statistics

The sample size of 14 participants was based on a prior study investigating fructose as a pre‐exercise carbohydrate supplement (8), which estimated that 14 participants would provide 80% power to detect a 15‐min difference in time to hypoglycaemia between fructose and water (α = 0.05). Given the exploratory 4‐condition crossover design, this was considered sufficient for the primary outcome. All randomised participants and available study periods were included. No interim analyses or stopping rules were applied.

Analyses were performed in Stata 14 (College Station, TX, USA). Differences in hypoglycaemia occurrence were assessed using Fisher's exact test, and time to hypoglycaemia using the Wilcoxon rank‐sum test. Plasma outcomes were analysed using linear mixed‐effects models with fixed effects for intervention (dextrose, lactose, galactose, placebo), time (modelled as a piecewise‐linear spline with knots at −30, 10, 60, and 150 min), and their interaction, with participant as a random intercept. The baseline phase (−30 min) was included in the spline to allow modelling of baseline differences. Models were estimated by restricted maximum likelihood, and contrasts evaluated using Wald tests, assuming missing at random. Post hoc analyses were adjusted for hypoglycaemic events. No subgroup analyses were performed. Subjective ratings were analysed using the same linear mixed‐effects modelling approach as described for plasma outcomes. For glycaemic outcomes (TIR, TBR, TAR), pairwise comparisons were adjusted using Holm's method. A p < 0.05 was considered significant. Data are presented as mean ± SD, median (IQR), or model‐estimated means with 95% confidence intervals, as appropriate.

3. Results

3.1. Participants

Between January and September 2023, 17 individuals attended screening and 15 were randomised, and all study days and outcome assessments were completed within this period. One participant discontinued after the first study day because of excessive fatigue, leaving a final cohort of 14 participants (two women and 12 men). All analyses were based on these 14 participants, contributing a total of 56 study days (Figure S2). Baseline characteristics were: HbA1c 52.9 ± 8.7 mmol/mol (7.0% ± 0.8%), CGM time in range 69% ± 16%, VO2max 47 ± 7 mL/kg/min, and total daily insulin 0.47 ± 0.20 units/kg/day (Table 1). Three participants used CSII, and 11 used MDI. One participant used an automated insulin delivery pump that was operated in manual mode during study days.

TABLE 1.

Baseline characteristics of participants (n = 14).

Variable Value
Age (years) 49.0 (37.0–55.8)
Sex, male/female 12/2
BMI (kg/m2) 22.4 ± 1.8
Duration of Type 1 diabetes (years) 20.0 (3.0–23.7)
HbA1c, mmol/mol (%) 52.9 ± 8.7 (7.0 ± 0.8)
Time in range (%), n = 11 69.1 ± 16.1
Time below range (%), n = 11 3.0 (2–3.75)
Time above range (%), n = 11 26.0 ± 17.3
VO2max (mL/min/kg) 46.9 ± 7.0
Total insulin dose (units/kg) 0.47 ± 0.20
Basal insulin (% of total dose) 52.2 ± 18.1
Insulin regime (MDI/CSII) 11/3
Open in a new tab

Note: Data are presented as mean ± SD or median (interquartile range), as appropriate. Glucose sensor data (time in, below and above range) were obtained during the 2 weeks prior to the screening visit and were available for 11 participants. MDI, multiple daily injections; CSII, continuous subcutaneous insulin infusion.

3.2. Time to Hypoglycaemia

No difference was observed in time‐to hypoglycaemia between interventions (all p > 0.16). Hypoglycaemic events occurred zero times after dextrose (DEX), twice after lactose (LAC) (at t = 50 and 60 min), once after galactose (GAL) (at t = 35 min), and four times after placebo (PLA) (at t = 20, 25, 52, and 55 min). The small number of hypoglycaemic events (7 events across 56 study days) may have limited the ability to detect between‐intervention differences.

3.3. Time in Glycaemic Ranges

The TIR during exercise, assessed from 30‐min before until exercise completion, was lower for DEX (48% [95% CI 40–57]) compared with PLA (86% [95% CI 80–91], p < 0.001) (Figure 2). TIR was also lower for DEX compared with LAC (70% [95% CI 62–78], p = 0.01) and GAL (71% [95% CI 63–79], p = 0.01), with no significant differences observed between PLA, LAC, and GAL. Extending the analysis window to 30 min post‐exercise yielded similar results (data not shown).

FIGURE 2.

FIGURE 2

Open in a new tab

Time in glycemic ranges. Time in range (TIR; 3.9–10.0 mmol/L), time below range (TBR; < 3.9 mmol/L), and time above range (TAR; > 10.0 mmol/L) following ingestion of placebo (PLA), dextrose (DEX), lactose (LAC), or galactose (GAL). Metrics are based on plasma glucose measurements obtained every 10 min from 30 min before exercise (t = −30) to the end of the 60‐min exercise session (t = 60). For participants experiencing hypoglycaemia, all subsequent time points were categorised as TBR. DEX was associated with lower TIR and higher TAR compared with PLA, LAC, and GAL. TBR was higher for PLA compared with all carbohydrate interventions. No differences were observed between LAC and GAL.

TBR was lower for DEX (0% [95% CI 0–3]), LAC (1% [95% CI 0–5]), and GAL (1% [95% CI 0–5]) compared with PLA (8% [95% CI 4–14]; all p < 0.05), with no differences observed among the three carbohydrate interventions.

TAR was higher for DEX (52% [95% CI 44–60]), LAC (29% [95% CI 21–37]), and GAL (28% [95% CI 21–36]) compared with PLA (7% [95% CI 3–12], all p < 0.01). TAR was higher for DEX compared with LAC and GAL (both p = 0.006), whereas TAR did not differ between LAC and GAL (p = 0.95).

3.4. Plasma Glucose Concentration

Fasting plasma glucose concentrations were similar at baseline across all four study days (p = 0.44) (Figure 3A). Plasma glucose concentrations increased after ingestion of DEX, LAC, and GAL compared with PLA, and peaked after the onset of exercise (t = 10 min). The magnitude of this pre‐exercise rise differed between interventions, with the highest increase observed after DEX (4.9 mmol/L [95% CI 4.2–5.6]), followed by LAC (3.7 mmol/L [95% CI 2.9–4.4]) and then GAL (2.3 mmol/L [95% CI 1.6–3.0]), all pairwise p < 0.001. Following the peak, plasma glucose concentration decreased over the final 50 min of exercise. The decline was greater for DEX (2.4 mmol/L [95% CI: 1.7–3.1]) and LAC (2.2 mmol/L [95% CI 1.5–2.9]), and to a lesser extent GAL (1.1 mmol/L [95% CI 0.4–1.8]), compared with PLA (all p < 0.01). The slope of the decline from the peak was steeper after DEX compared to GAL (p < 0.01), while no difference was observed between LAC and GAL (p = 0.48) or DEX and LAC (p = 0.45). During the recovery period following exercise, plasma glucose concentrations plateaued across all groups around the end‐of‐exercise level, with no differences observed in slopes, all p > 0.83.

FIGURE 3.

FIGURE 3

Open in a new tab

Metabolites and respiratory exchange ratio. Plasma glucose (a), plasma galactose (b), plasma glycerol (c), plasma free fatty acids (FFA) (d), plasma lactate (e), and respiratory exchange ratio (RER) (f) are shown. Symbols represent the interventions: Placebo (PLA; black circles), dextrose (DEX; yellow squares), lactose (LAC; blue triangles), and galactose (GAL; pink inverted triangles). Data are presented as mean ± SEM.

3.5. Glucose Requirement

During exercise the total amount of glucose administered to maintain plasma glucose just above 5 mmol/L in participants who developed hypoglycaemia was 4.0 and 0 g after LAC, 8.5 g after GAL, and 10.3 , 9.8, 5.8 and 1 g after PLA. Post‐exercise glucose infusion was 21.5 and 11.4 g after LAC, 9.7 g after GAL, and 14.3, 9.7, 7.4 and 7.0 g after PLA with no differences between the interventions (all p > 0.16).

3.6. Galactose and Other Metabolites

Mean plasma galactose concentration increased from baseline after GAL ingestion, peaking at 0.4 mmol/L [95% CI: 0.3–0.6, p < 0.001]. A smaller, non‐significant increase was observed after LAC ingestion, with plasma galactose peaking at 0.1 mmol/L [95% CI: 0.0–0.2, p = 0.53] (Figure 3B).

Plasma glycerol concentrations were similar throughout the study day with an increase during exercise and a return to baseline in the post‐exercise resting phase (Figure 3C), all p > 0.27. Plasma FFA levels decreased after ingestion of all the supplements. During exercise, FFA levels were stable and increased after the cessation of exercise (Figure 3D). There were no differences in plasma FFA levels, all p > 0.27.

Plasma lactate increased from 0.9 mmol/L [95% CI 0.9–1.0] at baseline to 3.8 mmol/L [95% CI 3.5–4.0] at 60 min (p < 0.001) with similar peak values across interventions (all p > 0.1) (Figure 3E).

3.7. Indirect Calorimetry

The RER increased from a baseline mean of 0.79 [95% CI 0.78–0.80] to 0.86 [95% CI 0.85–0.88] during the exercise phase. No difference in RER was observed between the four interventions (p = 0.65) (Figure 3F).

3.8. Exercise Intensity and Perceived Level of Exertion

Energy expenditure for the 60‐min cycling session was similar for all interventions.

The participants exercised at approximately 80% [95% CI 78–83] of maximal HR determined during the VO2max test, 67% [95% CI 65–70] of VO2max, and at a mean Borg Rating of Perceived Exertion of 12.9 [95% CI 12.4–13.4], with no significant differences between interventions. The participants completed the 60‐min exercise bout at a workload of 2.0 W/kg [95% CI 1.2–2.8].

3.9. Participant‐Reported Tolerability and Symptoms

Hunger was reported lower after GAL compared with PLA and LAC (p < 0.01), with estimated means (95% CI) of 3.4 (2.5–4.4) for DEX, 4.6 (3.7–5.5) for LAC, 3.0 (2.1–4.0) for GAL, and 5.2 (4.2–6.2) for PLA. No differences were observed in other symptom scores (abdominal discomfort, stomach cramps, bloating, nausea, dizziness, belching or acid reflux, urge to urinate or defecate, or physical fatigue) (Figure S3). No serious adverse events or unintended effects were observed after any intervention during the study.

4. Discussion

Pre‐exercise ingestion of galactose and lactose improved glycaemic stability compared with dextrose and placebo during 60‐min aerobic cycling in adults with Type 1 diabetes. Galactose and lactose reduced time below range compared with placebo and attenuated glycaemic excursions compared with dextrose, which produced the steepest rise and subsequent fall in plasma glucose. These findings suggest that lactose and galactose, when ingested at the tested dose and timing, may help balance the challenge of preventing hypoglycaemia without inducing hyperglycaemia during exercise in this population.

We did not observe a difference in time to hypoglycaemia, the primary outcome. Hypoglycaemic events were less frequent than anticipated (7 out of 56 study days), which limited the ability to detect statistically significant differences between interventions. Therefore, the possibility of false negative findings for the primary outcome cannot be excluded. Several factors may explain the low incidence of hypoglycaemia, including the morning timing of the exercise sessions. Previous studies have shown that morning fasted exercise in individuals with Type 1 diabetes has been associated with stable or even increased plasma glucose concentrations [16, 17, 18]. This protective effect may be mediated by elevated FFA [19] and cortisol levels [17], both of which contribute to increased insulin resistance [20, 21]. Additionally, exercise intensity could also influence hypoglycaemia risk since higher‐intensity exercise may increase counterregulatory hormones and reduce glucose disposal [22]. However, the mean exercise intensity in our study was 67% of VO2max, similar to the intensity at which glucose disposal peaks in individuals with Type 1 diabetes (65% of VO2max) [23], suggesting that intensity was unlikely to explain the low hypoglycaemia rate. Nevertheless, exercise intensity may still have contributed to increased counterregulatory responses and reduced hypoglycaemia risk in some participants.

The attenuation of glycaemic excursions with galactose and lactose aligns with previous findings for other low‐glycaemic index carbohydrates such as isomaltulose [24, 25]. Furthermore, the more stable glucose levels achieved with galactose may have implications beyond Type 1 diabetes, as reduced rebound hypoglycaemia has been reported following pre‐exercise galactose ingestion compared with dextrose in healthy individuals with preserved insulin secretion [26]. In addition, a study has shown that a low‐carbohydrate, high‐protein meal reduces glycaemic excursions compared with a high‐carbohydrate, low‐protein meal in individuals with Type 1 diabetes in relation to exercise [27]. Therefore, it would be relevant to test combinations of low‐glycaemic index carbohydrates together with protein to identify the optimal pre‐exercise supplement for individuals with Type 1 diabetes.

Galactose ingestion increased plasma galactose concentrations, peaking at ~0.4 mmol/L, whereas lactose ingestion resulted in a smaller and non‐significant increase, peaking at ~0.1 mmol/L. This may reflect both the lower galactose exposure with lactose (~50% of the ingested carbohydrate provided as galactose) and more efficient hepatic clearance when co‐ingested with glucose, as previously suggested [28, 29, 30].

We did not observe differences in substrate metabolism (glycerol, free fatty acids, or respiratory exchange ratio), which contrasts to findings in healthy individuals, where galactose ingestion was associated with higher FFA levels and lipid oxidation during exercise compared to dextrose or sucrose [13, 31]. This may reflect differences related to the absence of endogenous insulin secretion in Type 1 diabetes. However, the relatively small sample size warrants cautious interpretation of these findings.

An interesting observation was the potential effect of galactose on appetite regulation. Participants reported greater post‐exercise satiety following galactose ingestion compared with other interventions, consistent with Stevenson et al. [32]. Whether this is a true effect or just a chance finding remains unclear and more studies are warranted.

Several limitations should be considered. The low incidence of hypoglycaemia limited our ability to detect differences in the primary outcome, and the study may have been underpowered given the small sample size and substantial inter‐individual variability in glycaemic responses to exercise. The study was conducted during a 1‐h morning aerobic cycling session, which may limit generalisability to other exercise modalities, intensities, durations, or times of day. In addition, a fixed carbohydrate dose (20 g) and timing strategy (30 min before exercise) were used, which may limit generalisability to other dosing and timing approaches. Carbohydrate dose was not adjusted for body weight, which may have contributed to interindividual variation in glycaemic responses. Furthermore, only physically active individuals meeting current activity recommendations were included, which may limit generalisability to less active individuals with Type 1 diabetes. Finally, although a crossover design with washout periods of 1–4 weeks was applied, residual effects of previous interventions cannot be entirely excluded, although this is considered unlikely given the acute nature of the interventions.

In conclusion, although pre‐exercise ingestion of galactose and lactose did not reduce time to hypoglycaemia, these carbohydrates reduced TBR compared to placebo and at the same time reduced TAR compared to dextrose. These findings support the potential role of galactose and lactose, as well as other low‐glycaemic index carbohydrates, as pre‐exercise supplements in Type 1 diabetes to optimise glucose management and minimise the risk of both hypo‐ and hyperglycaemia during exercise. Further research is needed to evaluate these findings in larger cohorts, across different exercise modalities and times of day, and to determine optimal dosing and timing strategies for low‐glycaemic index carbohydrates in individuals with Type 1 diabetes.

Author Contributions

R.F.J. and E.S. designed the study. R.F.J. and K.R.S. conducted the study. S.B.R. and M.S.L. contributed to study design and provided scientific input. K.R.S. and H.S. analysed data. K.R.S. drafted the manuscript and R.F.J. and E.S. critically revised it. G.A.W., D.R. and L.C.G. contributed to the study and reviewed the manuscript. All authors approved the final version. E.S. is guarantor.

Funding

This work was supported by Novo Nordisk Fonden (NNF22OC0078217), Poul og Erna Sehested Hansens Fond, and Danmarks Frie Forskningsfond (10.46540/2034‐00331B).

Conflicts of Interest

R.F.J. has received speaking fees from Rubin Medical, Sanofi, and Novo Nordisk. D.R. has received research funding from Arla Foods Ingredients. G.A.W. has received research funding and/or consultancy fees from GlaxoSmithKline Ltd., Sugar Nutrition UK, Lucozade Ribena Suntory Ltd., Gatorade Sports Science Institute, and Volac International Ltd. H.S. has received personal fees for statistical consulting from Novartis, MSD, Vesper Bio ApS, Neumirna, Oslo Economics, and NHTA. S.B.R. and M.S.L. are employed by Arla Foods Ingredients Group P/S. The other authors declare no conflicts of interest.

Supporting information

Data S1: CONSORT 2025 checklist item description.

DOM-28-7825-s002.docx (32.1KB, docx)

Figure S1: Gastrointestinal and exercise‐related symptoms questionnaire.

DOM-28-7825-s001.docx (46.2KB, docx)

Figure S2: CONSORT flow diagram.

DOM-28-7825-s004.pdf (101.2KB, pdf)

Figure S3: Mean ratings and 95% confidence intervals of questionnaire results.

DOM-28-7825-s003.pdf (69.8KB, pdf)

Acknowledgements

We thank the participants and laboratory staff for their contribution to the study. The study was supported by the Novo Nordisk Foundation (NNF22OC0078217), Arla Foods Ingredients Group P/S, Independent Research Fund Denmark (10.46540/2034‐00331B), and Poul og Erna Sehested Hansens Fund. Arla Foods Ingredients Group P/S provided the carbohydrate supplements. The funders had no role in study design, data collection, data analysis, data interpretation, or the decision to submit the manuscript for publication. The authors used ChatGPT (OpenAI) to assist with language editing.

Schmidt K. R., Johansen R. F., Riis S. B., et al., “Pre‐Exercise Galactose or Lactose Reduces Glycaemic Excursions During Exercise in Adults With Type 1 Diabetes: A Randomised, Double‐Blind, Crossover Trial,” Diabetes, Obesity and Metabolism 28, no. 9 (2026): 7825–7833, 10.1111/dom.70958.

Handling Editor: Nigel Irwin

Data Availability Statement

Data, protocol, and statistical analysis plan are available from the corresponding author upon reasonable request, subject to data protection regulations.

References

  • 1. Riddell M. C., Gallen I. W., Smart C. E., et al., “Exercise Management in Type 1 Diabetes: A Consensus Statement,” Lancet Diabetes and Endocrinology 5, no. 5 (2017): 377–390, 10.1016/S2213-8587(17)30014-1. [DOI] [PubMed] [Google Scholar]
  • 2. Codella R., Terruzzi I., and Luzi L., “Why Should People With Type 1 Diabetes Exercise Regularly?,” Acta Diabetologica 54, no. 7 (2017): 615–630, 10.1007/s00592-017-0978-x. [DOI] [PubMed] [Google Scholar]
  • 3. Quirk H., Blake H., Tennyson R., Randell T. L., and Glazebrook C., “Physical Activity Interventions in Children and Young People With Type 1 Diabetes Mellitus: A Systematic Review With Meta‐Analysis,” Diabetic Medicine 31, no. 10 (2014): 1163–1173, 10.1111/dme.12531. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4. Wu N., Bredin S. S. D., Guan Y., et al., “Cardiovascular Health Benefits of Exercise Training in Persons Living With Type 1 Diabetes: A Systematic Review and Meta‐Analysis,” Journal of Clinical Medicine 8, no. 2 (2019): 253, 10.3390/jcm8020253. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Brazeau A. S., Rabasa‐Lhoret R., Strychar I., and Mircescu H., “Barriers to Physical Activity Among Patients With Type 1 Diabetes,” Diabetes Care 31, no. 11 (2008): 2108–2109, 10.2337/dc08-0720. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Yardley J. E. and Sigal R. J., “Exercise Strategies for Hypoglycemia Prevention in Individuals With Type 1 Diabetes,” Diabetes Spectrum: A Publication of the American Diabetes Association 28, no. 1 (2015): 32–38, 10.2337/diaspect.28.1.32. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Johansen R. F., Caunt S., Heller S., et al., “Factors Influencing Physical Activity Level in Adults With Type 1 Diabetes: A Cross‐Sectional Study,” Canadian Journal of Diabetes 48 (2024): 431–438, 10.1016/j.jcjd.2024.06.002. [DOI] [PubMed] [Google Scholar]
  • 8. Riddell M. C., Zaharieva D. P., Tansey M., et al., “Individual Glucose Responses to Prolonged Moderate Intensity Aerobic Exercise in Adolescents With Type 1 Diabetes: The Higher They Start, the Harder They Fall,” Pediatric Diabetes 20, no. 1 (2019): 99–106, 10.1111/pedi.12799. [DOI] [PubMed] [Google Scholar]
  • 9. Atkinson F. S., Brand‐Miller J. C., Foster‐Powell K., Buyken A. E., and Goletzke J., “International Tables of Glycemic Index and Glycemic Load Values 2021: A Systematic Review,” American Journal of Clinical Nutrition 114, no. 5 (2021): 1625–1632, 10.1093/ajcn/nqab233. [DOI] [PubMed] [Google Scholar]
  • 10. Kosinski C., Herzig D., Laesser C. I., et al., “A Single Load of Fructose Attenuates the Risk of Exercise‐Induced Hypoglycemia in Adults With Type 1 Diabetes on Ultra‐Long‐Acting Basal Insulin: A Randomized, Open‐Label, Crossover Proof‐Of‐Principle Study,” Diabetes Care 43, no. 9 (2020): 2010–2016, 10.2337/dc19-2250. [DOI] [PubMed] [Google Scholar]
  • 11. Rytz A., Adeline D., Le K. A., et al., “Predicting Glycemic Index and Glycemic Load From Macronutrients to Accelerate Development of Foods and Beverages With Lower Glucose Responses,” Nutrients 11, no. 5 (2019): 1–11, 10.3390/nu11051172. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12. Leijssen D. P., Saris W. H., Jeukendrup A. E., and Wagenmakers A. J., “Oxidation of Exogenous [13C]Galactose and [13C]Glucose During Exercise,” Journal of Applied Physiology 79, no. 3 (1995): 720–725, 10.1152/jappl.1995.79.3.720. [DOI] [PubMed] [Google Scholar]
  • 13. Odell O. J., Podlogar T., and Wallis G. A., “Comparable Exogenous Carbohydrate Oxidation From Lactose or Sucrose During Exercise,” Medicine and Science in Sports and Exercise 52, no. 12 (2020): 2663–2672, 10.1249/MSS.0000000000002426. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. Bjerg L., Dalsgaard E. M., Norman K., Isaksen A. A., and Sandbaek A., “Cohort Profile: Health in Central Denmark (HICD) Cohort ‐ a Register‐Based Questionnaire Survey on Diabetes and Related Complications in the Central Denmark Region,” BMJ Open 12, no. 7 (2022): e060410, 10.1136/bmjopen-2021-060410. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15. Urbaniak G. C. and Plous S., “Research Randomizer.” (2013).
  • 16. Scott S. N., Cocks M., Andrews R. C., et al., “Fasted High‐Intensity Interval and Moderate‐Intensity Exercise Do Not Lead to Detrimental 24‐Hour Blood Glucose Profiles,” Journal of Clinical Endocrinology and Metabolism 104, no. 1 (2019): 111–117, 10.1210/jc.2018-01308. [DOI] [PubMed] [Google Scholar]
  • 17. Ruegemer J. J., Squires R. W., Marsh H. M., et al., “Differences Between Prebreakfast and Late Afternoon Glycemic Responses to Exercise in IDDM Patients,” Diabetes Care 13, no. 2 (1990): 104–110, 10.2337/diacare.13.2.104. [DOI] [PubMed] [Google Scholar]
  • 18. Toghi‐Eshghi S. R. and Yardley J. E., “Morning (Fasting) vs Afternoon Resistance Exercise in Individuals With Type 1 Diabetes: A Randomized Crossover Study,” Journal of Clinical Endocrinology and Metabolism 104, no. 11 (2019): 5217–5224, 10.1210/jc.2018-02384. [DOI] [PubMed] [Google Scholar]
  • 19. Roden M., Krssak M., Stingl H., et al., “Rapid Impairment of Skeletal Muscle Glucose Transport/Phosphorylation by Free Fatty Acids in Humans,” Diabetes 48, no. 2 (1999): 358–364, 10.2337/diabetes.48.2.358. [DOI] [PubMed] [Google Scholar]
  • 20. Plat L., Byrne M. M., Sturis J., et al., “Effects of Morning Cortisol Elevation on Insulin Secretion and Glucose Regulation in Humans,” American Journal of Physiology 270, no. 1 Pt 1 (1996): E36–E42, 10.1152/ajpendo.1996.270.1.E36. [DOI] [PubMed] [Google Scholar]
  • 21. Gormsen L. C., Nielsen C., Jessen N., Jorgensen J. O., and Moller N., “Time‐Course Effects of Physiological Free Fatty Acid Surges on Insulin Sensitivity in Humans,” Acta Physiologica (Oxford, England) 201, no. 3 (2011): 349–356, 10.1111/j.1748-1716.2010.02181.x. [DOI] [PubMed] [Google Scholar]
  • 22. Bally L., Zueger T., Buehler T., et al., “Metabolic and Hormonal Response to Intermittent High‐Intensity and Continuous Moderate Intensity Exercise in Individuals With Type 1 Diabetes: A Randomised Crossover Study,” Diabetologia 59, no. 4 (2016): 776–784, 10.1007/s00125-015-3854-7. [DOI] [PubMed] [Google Scholar]
  • 23. Shetty V. B., Fournier P. A., Davey R. J., et al., “Effect of Exercise Intensity on Glucose Requirements to Maintain Euglycemia During Exercise in Type 1 Diabetes,” Journal of Clinical Endocrinology and Metabolism 101, no. 3 (2016): 972–980, 10.1210/jc.2015-4026. [DOI] [PubMed] [Google Scholar]
  • 24. Bracken R. M., Page R., Gray B., et al., “Isomaltulose Improves Glycemia and Maintains Run Performance in Type 1 Diabetes,” Medicine and Science in Sports and Exercise 44, no. 5 (2012): 800–808, 10.1249/MSS.0b013e31823f6557. [DOI] [PubMed] [Google Scholar]
  • 25. West D. J., Stephens J. W., Bain S. C., et al., “A Combined Insulin Reduction and Carbohydrate Feeding Strategy 30 Min Before Running Best Preserves Blood Glucose Concentration After Exercise Through Improved Fuel Oxidation in Type 1 Diabetes Mellitus,” Journal of Sports Sciences 29, no. 3 (2011): 279–289, 10.1080/02640414.2010.531753. [DOI] [PubMed] [Google Scholar]
  • 26. Jentjens R. L. and Jeukendrup A. E., “Effects of Pre‐Exercise Ingestion of Trehalose, Galactose and Glucose on Subsequent Metabolism and Cycling Performance,” European Journal of Applied Physiology 88, no. 4–5 (2003): 459–465, 10.1007/s00421-002-0729-7. [DOI] [PubMed] [Google Scholar]
  • 27. Kristensen K. B., Ranjan A. G., McCarthy O. M., et al., “Effects of a Low‐Carbohydrate‐High‐Protein Pre‐Exercise Meal in Type 1 Diabetes – A Randomized Crossover Trial,” Journal of Clinical Endocrinology and Metabolism 109, no. 1 (2023): 208–216, 10.1210/clinem/dgad427. [DOI] [PubMed] [Google Scholar]
  • 28. Odell O. J., Impey S. G., Shad B. J., et al., “Oxidation of Independent and Combined Ingested Galactose and Glucose During Exercise,” Journal of Applied Physiology (1985) 133, no. 5 (2022): 1166–1174, 10.1152/japplphysiol.00105.2022. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. Williams C. A., Phillips T., and Macdonald I., “The Influence of Glucose on Serum Galactose Levels in Man,” Metabolism, Clinical and Experimental 32, no. 3 (1983): 250–256, 10.1016/0026-0495(83)90189-0. [DOI] [PubMed] [Google Scholar]
  • 30. Ercan N., Nuttall F. Q., Gannon M. C., Redmon J. B., and Sheridan K. J., “Effects of Glucose, Galactose, and Lactose Ingestion on the Plasma Glucose and Insulin Response in Persons With Non‐Insulin‐Dependent Diabetes Mellitus,” Metabolism 42, no. 12 (1993): 1560–1567, 10.1016/0026-0495(93)90151-d. [DOI] [PubMed] [Google Scholar]
  • 31. Stannard S. R., Hawke E. J., and Schnell N., “The Effect of Galactose Supplementation on Endurance Cycling Performance,” European Journal of Clinical Nutrition 63, no. 2 (2009): 209–214, 10.1038/sj.ejcn.1602924. [DOI] [PubMed] [Google Scholar]
  • 32. Duckworth L. C., Backhouse S. H., O'Hara J. P., and Stevenson E. J., “Effect of Galactose Ingestion Before and During Exercise on Substrate Oxidation, Postexercise Satiety, and Subsequent Energy Intake in Females,” Journal of the American College of Nutrition 35, no. 1 (2016): 1–12, 10.1080/07315724.2014.994790. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Data S1: CONSORT 2025 checklist item description.

DOM-28-7825-s002.docx (32.1KB, docx)

Figure S1: Gastrointestinal and exercise‐related symptoms questionnaire.

DOM-28-7825-s001.docx (46.2KB, docx)

Figure S2: CONSORT flow diagram.

DOM-28-7825-s004.pdf (101.2KB, pdf)

Figure S3: Mean ratings and 95% confidence intervals of questionnaire results.

DOM-28-7825-s003.pdf (69.8KB, pdf)

Data Availability Statement

Data, protocol, and statistical analysis plan are available from the corresponding author upon reasonable request, subject to data protection regulations.

Articles from Diabetes, Obesity & Metabolism are provided here courtesy of Wiley

RESOURCES