Pre-exercise chickpea milk consumption supports glycemic stability during endurance exercise in female runners

Abstract

[Purpose]

This study investigated the effects of pre-exercise chickpea milk consumption on glycemic stability and physiological responses during endurance exercise in trained female runners.

[Methods]

Fourteen eumenorrheic female half-marathon runners (25-35 years) completed a randomized, double-blind, counterbalanced crossover trial. Participants consumed 250 mL of chickpea milk or soy milk or a maltodextrin placebo 30 min before exercise, with each trial conducted during the follicular phase and separated by a washout period of at least one week. The exercise protocol consisted of treadmill running at 60% of maximal heart rate for 60 min, followed by running at 80% of maximal heart rate until volitional exhaustion. Blood glucose, blood lactate, free fatty acids, creatine kinase, respiratory exchange ratio, rating of perceived exertion, and time to exhaustion were assessed.

[Results]

Blood glucose concentrations differed significantly between the beverage conditions (p < 0.05). Compared with the placebo, chickpea milk produced a more stable glycemic response during exercise. Blood glucose variability, assessed using the coefficient of variation, differed significantly among beverage conditions (p = 0.031), with lower values following chickpea milk (13.07 ± 4.33%), and soy milk (12.37 ± 4.62%) compared with the placebo (16.07 ± 4.89%). No significant differences were observed among the beverage conditions for blood lactate, free fatty acids, creatine kinase, respiratory exchange ratio, rating of perceived exertion, or time to exhaustion (all p > 0.05).

[Conclusion]

Pre-exercise chickpea milk consumption supported glycemic stability during prolonged endurance exercise in trained female runners, but did not improve exercise performance or other physiological outcomes. Larger, adequately powered studies are warranted to determine whether this metabolic response translates into meaningful improvements in endurance performance.

INTRODUCTION

Carbohydrate availability plays an important role in supporting the metabolic demands and sustaining performance during endurance exercise [1-3]. During prolonged moderate-to-high-intensity exercise, carbohydrate oxidation contributes substantially to energy production, whereas reductions in carbohydrate availability may contribute to fatigue and impaired exercise capacity [1-4]. Accordingly, nutritional strategies providing carbohydrates before exercise are commonly used to support carbohydrate availability for subsequent endurance exercise [1-3]. Importantly, the amount, type, and digestibility of carbohydrate consumed before exercise can influence pre-exercise and exercise-related glycemic responses, with different carbohydrate sources producing distinct patterns of glucose availability [1,2]. Therefore, selecting an appropriate pre-exercise carbohydrate source may be relevant not only for providing exogenous energy, but also for supporting glycemic stability during prolonged endurance exercise.

Female endurance athletes present specific physiological considerations that may influence their metabolic and perceptual responses to exercise. For instance, fluctuations in ovarian hormones across the menstrual cycle have been shown to influence substrate metabolism, thermoregulation, perceived exertion, and exercise performance. However, the magnitude and consistency of these effects vary considerably among individuals and across studies [5-9]. These variations highlight the importance of considering menstrual status and cycle phase when designing and interpreting exercise studies involving female participants [5]. Nutritional intake may also interact with menstrual cycle-related physiological changes, further emphasizing the need to account for both nutritional and reproductive factors in female athletes [10]. Accordingly, controlling the timing of experimental trials within a consistent menstrual cycle phase may help reduce potential within-participant variability when evaluating metabolic responses to nutritional interventions [5].

Menstrual cycle symptoms and physiological fluctuations influence perceived exertion, training experiences, and exercise responses in female athletes, despite substantial interindividual variability [6-10]. Consequently, nutritional and metabolic studies in women should account for menstrual cycle characteristics [5,10]. In the present study, eumenorrheic female runners were recruited, and all experimental trials were scheduled during the follicular phase to reduce potential within-participant variability associated with menstrual cycle timing. Within this controlled framework, investigating the nutritional strategies that support metabolic stability during endurance exercises may provide relevant information for female endurance athletes.

Pre-exercise nutrition supports carbohydrate availability and metabolic demands during endurance exercise [1-3]. Specifically, the digestion and absorption rates of ingested carbohydrates influence glycemic responses before and during subsequent exercise [1,2]. Carbohydrate sources that are digested and absorbed more gradually may produce a more sustained glycemic response than rapidly available carbohydrate sources, potentially limiting pronounced fluctuations in circulating glucose concentrations [1,2]. Such regulation may be particularly relevant during prolonged endurance exercise, in which maintaining carbohydrate availability is important for meeting ongoing metabolic demands [1-4]. Accordingly, the selection of pre-exercise carbohydrate sources with favorable glycemic characteristics represents a potential nutritional strategy for supporting glycemic stability during endurance exercise.

Pre-exercise nutrition supports carbohydrate availability and metabolic demands during endurance exercise [1-3]. Specifically, the digestion and absorption rates of ingested carbohydrates influence glycemic responses before and during subsequent exercise [1,2]. Carbohydrate sources that are digested and absorbed more gradually may produce a more sustained glycemic response than rapidly available carbohydrate sources, potentially limiting pronounced fluctuations in circulating glucose concentrations [1,2]. Such regulation may be particularly relevant during prolonged endurance exercise, in which maintaining carbohydrate availability is important for meeting ongoing metabolic demands [1-4]. Accordingly, the selection of pre-exercise carbohydrate sources with favorable glycemic characteristics represents a potential nutritional strategy for supporting glycemic stability during endurance exercise.

Plant-based nutritional strategies have received increasing attention because they can provide carbohydrates, proteins, dietary fiber, and other bioactive components for diverse dietary patterns [11-13]. Legume-based foods are of particular interest because their nutritional composition and food structure can influence carbohydrate digestibility and the postprandial glycemic response [16,17]. Among legumes, chickpea (Cicer arietinum L.) is a nutrient-dense food containing carbohydrates, proteins, dietary fiber, and resistant starch, with characteristics that may contribute to a relatively gradual glycemic response [16,17]. These properties provide a rationale for investigating chickpea-derived beverages as a pre-exercise nutritional strategy to support glycemic stability during endurance exercises. In the present study, soy milk was included as a plant-based comparator, allowing the evaluation of acute metabolic responses to chickpea milk relative to other legume-based beverages. The investigation was limited to the acute effects of a single pre-exercise serving. Therefore, the potential physiological adaptations associated with habitual chickpea consumption were beyond the scope of this study.

Despite these nutritional and functional properties, limited research has examined chickpea-derived beverages as a pre-exercise nutritional strategy for endurance athletes. Furthermore, comparative evidence between chickpea milk and commonly consumed plant-based alternatives, such as soy milk, remains scarce, particularly in female populations under controlled menstrual conditions.

Therefore, this study aimed to investigate the effects of pre-exercise chickpea milk consumption on glycemic stability and physiological responses during endurance exercises in trained female runners. We hypothesized that pre-exercise chickpea milk consumption would support greater glycemic stability during prolonged endurance exercise than soy milk or a maltodextrin placebo.

METHODS

Participants

Fourteen trained female half-marathon runners (mean ± SD: age, 31.14 ± 3.63 years; body mass, 58.44 ± 7.47 kg; height, 162.21 ± 3.29 cm; VO₂ max, 41.71 ± 1.54 mL·kg⁻ ¹·min⁻¹) voluntarily participated in this study. All participants were regularly engaged in endurance training, at least three times per week.

Participants were eligible if they were eumenorrheic, defined as having regular menstrual cycles of approximately 21-35 days, and did not use hormonal contraceptives. Each experimental trial was scheduled during the early-to-mid follicular phase based on self-reported menstrual history and calendar tracking to reduce the potential within-participant variability associated with menstrual cycle timing. The menstrual cycle phase was not confirmed using hormonal measurements.

Exclusion criteria included underlying medical conditions that could affect exercise participation, glucose-6-phosphate dehydrogenase deficiency, known allergies to chickpeas or soybeans, and musculoskeletal injuries that could interfere with exercise performance.

All the experimental procedures were conducted at the Exercise and Sports Performance Laboratory (ESPL), Faculty of Sports Science, Chulalongkorn University, Thailand. The study protocol was approved by the Institutional Review Board of Chulalongkorn University (COA No. 200/67) on September 10, 2024, before participant enrollment began on November 4, 2024. Written informed consent was obtained from all participants prior to their participation. The study was retrospectively registered with the Thai Clinical Trials Registry (TCTR20260515005) on May 15, 2026, after completion of data collection.

Study Design

This study used a randomized, double-blind, counterbalanced crossover design. Each participant completed three experimental trials involving the consumption of chickpea milk (CH), soy milk (SO), and maltodextrin placebo (PL), with the order of the three beverage conditions counterbalanced across participants. Each trial was separated by a washout period of at least one week. All experimental trials were scheduled during the follicular phase of the menstrual cycle to reduce the potential within-participant variability associated with menstrual cycle timing.

Beverage Preparation

Chickpea and soy milk were prepared using standardized procedures. Dried chickpeas and soybeans were soaked separately in water, rinsed, blended with potable water, and filtered to remove insoluble residues. The resulting filtrates were pasteurized at 80 °C for 30 min and stored under refrigerated conditions until further use. The placebo beverage was prepared by dissolving maltodextrin in water and was matched to the experimental beverages in terms of energy content and appearance. All beverages were provided in 250 mL servings in identical opaque bottles to maintain participant blinding.

Figure 1. Preparation procedures for chickpea milk, soy milk, and maltodextrin placebo.

Top panel: Preparation of chickpea and soy milk, including soaking, rinsing, blending, filtration, and pasteurization to obtain the final beverages. Bottom panel: Preparation of the maltodextrin placebo by dissolving maltodextrin in water.

Experimental Protocol

The participants were instructed to fast for 6 h before each experimental trial and to refrain from strenuous exercise and caffeine consumption for at least 24 h before testing. Baseline measurements, including body composition and resting blood samples, were obtained upon arrival at the laboratory. The participants consumed the assigned beverage 30 min before the exercise test.

The exercise protocol consisted of treadmill running at 60% of maximal heart rate (HRmax) for 60 min, followed by an increase in exercise intensity to 80% HRmax until volitional exhaustion. The heart rate was monitored continuously throughout the exercise protocol.

Physiological and Biochemical Measurements

Capillary blood samples were collected from the fingertips at rest, immediately before exercise (0 min), during exercise (30 and 60 min), at the end of exercise, and during recovery to determine blood glucose and blood lactate concentrations using validated portable analyzers. Venous blood samples were collected at rest and during recovery to assess free fatty acid (FFA) and creatine kinase (CK) concentrations.

The respiratory exchange ratio (RER) was measured continuously during the exercise using a metabolic gas analysis system. The rating of perceived exertion (RPE) was recorded at standardized time points using the Borg scale. Time to exhaustion (TTE) was defined as the duration of the high-intensity exercise phase at 80% HRmax from the onset of this phase until volitional exhaustion.

Statistical Analysis

All data are presented as mean ± SD. Normality of data distribution was assessed using the Shapiro-Wilk test.

Variables measured repeatedly over time, including heart rate, blood glucose, blood lactate, RER, and RPE, were analyzed using a two-way repeated-measures analysis of variance (ANOVA) to examine the main effects of the beverage condition (CH, SO, and PL) and time, as well as the condition × time interaction. FFA and CK levels, which were measured at two time points (rest and recovery), were also analyzed using two-way repeated-measures ANOVA (condition × time). When significant effects were detected, Bonferroni-adjusted post hoc comparisons were performed.

TTE, which was measured once under each beverage condition, was analyzed using a one-way repeated-measures ANOVA. To quantitatively evaluate glycemic stability, the coefficient of variation (CV) of blood glucose concentration across all measurement time points was calculated for each participant and beverage condition as: (standard deviation/mean) × 100. Differences in CV among beverage conditions were analyzed using one-way repeated-measures ANOVA, followed by Bonferroni-adjusted pairwise comparisons when appropriate.

The blood glucose response was designated as the primary outcome. Blood lactate, FFA, CK, RER, RPE, and TTE were evaluated as secondary outcomes to provide complementary information regarding metabolic and physiological responses to the interventions. Because these secondary outcomes represented distinct physiological domains and were considered supportive rather than confirmatory endpoints, no additional multiplicity corrections across different outcome variables were applied.

Statistical significance was set at p < 0.05. All statistical analyses were performed using IBM SPSS Statistics version 28.0 (IBM Corp., Armonk, NY, USA).

Figure 2. Schematic overview of the randomized, double-blind, counterbalanced crossover study design and experimental protocol.

Fourteen trained female half-marathon runners completed three experimental trials involving chickpea milk (CH), soy milk (SO), and maltodextrin placebo (PL), with each trial separated by a washout period of at least one week. Participants consumed 250 mL of the assigned beverage 30 min before exercise. The exercise protocol consisted of treadmill running at 60% of maximal heart rate (HRmax) for 60 min, followed by running at 80% HRmax until volitional exhaustion. All experimental trials were scheduled during the follicular phase of the menstrual cycle.

RESULTS

Participant Characteristics

All fourteen participants completed three experimental trials. The participant characteristics are presented in Table 1. The participants had a mean age of 31.14 ± 3.63 years, body mass of 58.44 ± 7.47 kg, height of 162.21 ± 3.29 cm, and VO₂ max of 41.71 ± 1.54 mL·kg⁻¹·min⁻¹. All experimental trials were scheduled during the follicular phase, based on self-reported menstrual history and calendar tracking.

Table 1.

Participant characteristics (n = 14)

Variable	Mean ± SD
Age (years)	31.14 ± 3.63
Body mass (kg)	58.44 ± 7.47
Height (cm)	162.21 ± 3.29
Body mass index (kg/m2)	22.18 ± 2.51
Body fat (%)	16.11 ± 4.55
Maximal running speed (km·h-1)	10.83 ± 0.72
Maximal oxygen consumption (VO2max) (mL·kg-¹·min-¹)	41.71 ± 1.54
Menstrual cycle length (days)	29.39 ± 3.70
Day of menstrual cycle at testing (days)	10.98 ± 0.85

Note. Values are presented as mean ± SD. VO₂max, maximal oxygen consumption.

Heart Rate Responses

The heart rate responses across the experimental conditions are listed in Table 2. Heart rate increased progressively over time during exercise in all three beverage conditions (CH, SO, and PL). No significant differences in heart rate were observed among beverage conditions (p > 0.05).

Table 2.

Metabolic and physiological responses across beverage conditions and time points (n = 14)

Variable	Condition	Rest	0 min	30 min	60 min	End	Recovery
Heart rate (bpm)	CH	65.14 ± 7.91	78.78 ± 8.61	146.78 ± 14.24	152.21 ± 14.60	171.21 ± 8.65	89.86 ± 10.46
	SO	65.43 ± 6.28	77.21 ± 6.98	144.07 ± 14.79	153.57 ± 14.71	168.14 ± 11.54	89.57 ± 10.89
	PL	60.35 ±6.27	77.28 ± 9.21	143.50 ± 11.52	150.50 ± 10.87	168.64 ± 7.55	83.21 ±9.99
Blood glucose (mg·dL-¹)	CH	90.28 ± 5.09	101.42 ± 7.50*#	96.57 ± 11.77#	100.57 ± 13.48	120.86 ± 18.76	118.21 ± 12.66
	SO	93.71 ± 5.64	94.28 ± 6.33*†	96.07 ± 8.60†	99.93 ± 10.50†	119.50 ± 11.99	118.28 ± 12.61
	PL	92.36 ± 6.26	118.21 ± 19.89#†	84.43 ± 6.76#†	93.71 ± 8.04†	110.36 ± 17.75	111.78 ± 18.02
Blood lactate (mmol·L-¹)	CH	0.82 ± 0.57	2.18 ± 0.95	3.16 ± 1.64	3.59 ± 2.04	6.05 ± 2.12	2.84 ± 1.06
	SO	0.59 ± 0.22	1.78 ± 0.84	2.38 ± 1.48	3.07 ± 1.64	6.27 ± 3.26	2.77 ± 1.71
	PL	0.53 ± 0.14	1.98 ± 0.61	2.80 ± 1.18	3.00 ± 1.31	5.34 ± 2.03	2.65 ± 1.40
Free fatty acids (mmol·L-¹)	CH	0.043 ± 0.054	—	—	—	—	0.232 ± 0.142
	SO	0.031 ± 0.045	—	—	—	—	0.260 ± 0.138
	PL	0.024 ± 0.027	—	—	—	—	0.199 ± 0.121
Creatine kinase (U·L-¹)	CH	108.36 ± 65.96	—	—	—	—	141.78 ± 91.61
	SO	125.14 ± 101.93	—	—	—	—	152.71 ± 125.65
	PL	109.28 ± 54.29	—	—	—	—	134.21 ± 62.45
Respiratory exchange ratio (RER)	CH	—	0.87 ± 0.03	0.91 ± 0.04	0.93 ± 0.05	1.06 ± 0.07	—
	SO	—	0.86 ± 0.05	0.91 ± 0.05	0.94 ± 0.05	1.04 ± 0.07	—
	PL	—	0.90 ± 0.05	0.92 ± 0.03	0.95 ± 0.03	1.05 ± 0.07	—
Rating of perceived exertion (RPE)	CH	—	6.28 ± 0.61	11.36 ± 1.82	12.64 ± 2.20	15.93 ± 2.01	—
	SO	—	6.78 ± 0.97	10.57 ± 1.99	12.93 ± 2.33	15.28 ± 1.32	—
	PL	—	6.71 ± 0.82	10.57 ± 1.91	13.14 ± 2.35	15.64 ± 2.30	—

Note. Values are presented as mean ± SD. Dashes (—) indicate that the variable was not measured at the corresponding time point. CH, chickpea milk; SO, soy milk; PL, maltodextrin placebo.

^*p < 0.05 vs. rest within condition;

^#p < 0.05 vs. placebo;

^†p < 0.05 vs. soy milk.

Blood Glucose Responses

Beverage consumption significantly influenced blood glucose concentration (p < 0.05). Compared with the placebo, chickpea milk ingestion resulted in significantly higher blood glucose concentrations at the selected time points, while no significant differences were observed between the chickpea and soy milk conditions.

Consequently, each of the three beverage conditions demonstrated distinct glycemic response patterns. In the placebo condition, blood glucose levels increased markedly immediately before exercise (0 min) and subsequently declined during the first 30 min of exercise, producing a pattern consistent with a reactive hypoglycemia-like response. In contrast, blood glucose concentrations after chickpea milk ingestion changed more gradually across the exercise protocol, indicating a more stable glycemic response.

The CV of blood glucose concentrations across all measurement time points was used to quantify glycemic variability. Beverage consumption had a significant effect on blood glucose CV (F(2,28) = 3.964, p = 0.031). The placebo condition exhibited the greatest glycemic variability (16.07 ± 4.89%), whereas lower CV values were observed following chickpea milk (13.07 ± 4.33%) and soy milk (12.37 ± 4.62%). These findings indicate that glycemic variability differed among beverage conditions, with lower overall variability observed following both plant-based beverages than following the maltodextrin placebo.

Blood Lactate Responses

Blood lactate concentrations increased progressively during exercise and decreased during recovery under all three beverage conditions. No significant differences in blood lactate concentrations were observed between the beverage conditions (p > 0.05).

Free fatty acids (FFAs) and creatine kinase (CK)

No significant differences in FFA or CK concentrations were observed between the three beverage conditions (p > 0.05).

Respiratory exchange ratio (RER)

RER increased with increasing exercise intensity across all three beverage conditions. No significant differences in the RER were observed between the beverage conditions (p > 0.05).

Rating of perceived exertion (RPE)

The RPE increased progressively over time across all three beverage conditions. No significant differences in the RPE were observed between the beverage conditions (p > 0.05).

Time to exhaustion (TTE)

TTE did not differ significantly between the three beverage conditions (p = 0.075). Mean TTE following chickpea milk, soy milk, and placebo consumption was 66.40 ± 3.58, 65.65 ± 3.17, and 64.23 ± 2.46 min, respectively.

Figure 3. Metabolic and physiological responses across beverage conditions and time points (n = 14).

(A) Heart rate, (B) blood glucose, (C) blood lactate, and (D) free fatty acids. CH, chickpea milk; SO, soy milk; PL, maltodextrin placebo. Values are presented as mean ± SD. *p < 0.05 vs. rest within condition; ^#p < 0.05 vs. placebo; ^†p < 0.05 vs. soy milk.

Figure 4. Metabolic and physiological responses across beverage conditions and time points (n = 14).

(A) Creatine kinase, (B) respiratory exchange ratio, and (C) rating of perceived exertion. CH, chickpea milk; SO, soy milk; PL, maltodextrin placebo. Values are presented as mean ± SD.

Table 3.

Comparison of time to exhaustion among beverage conditions (n = 14)

Condition	Time to exhaustion (min) Mean ± SD	F (2, 26)	p value
Chickpea milk (CH)	66.40 ± 3.58
Soy milk (SO)	65.65 ± 3.17	2.87	.075
Placebo (PL)	64.23 ± 2.46

Note. Values are presented as mean ± SD. Differences among beverage conditions were analyzed using one-way repeated-measures ANOVA. No significant difference in time to exhaustion was observed among beverage conditions (p = .075). Statistical significance was set at p < .05.

DISCUSSION

The present study examined the effects of pre-exercise chickpea milk consumption on glycemic and physiological responses during endurance exercise in trained female runners, with all experimental trials scheduled during the follicular phase of the menstrual cycle. The primary finding was that chickpea milk produced a more stable glycemic response pattern than maltodextrin placebo, supported by lower overall blood glucose variability across the exercise protocol. However, this metabolic response was not accompanied by significant differences in the TTE or other physiological outcomes, including blood lactate, FFAs, CK, RER, RPE, and heart rate. These findings suggest that the primary acute effect observed following pre-exercise chickpea milk consumption is related to glycemic regulation rather than an improvement in endurance performance.

Our findings indicate that pre-exercise chickpea milk consumption is associated with a more stable glycemic response during prolonged endurance exercise. Following maltodextrin placebo administration, blood glucose levels increased markedly immediately before exercise and subsequently declined during the early stages of exercise, producing a pattern consistent with a reactive hypoglycemia-like response. Rapidly available carbohydrate consumed before exercise can produce pronounced changes in circulating glucose and insulin concentrations, which may influence the subsequent glycemic response during exercisec[1,2]. In contrast, chickpea milk produced a more gradual glycemic profile across the exercise protocol, suggesting that the characteristics of the pre-exercise carbohydrate source may influence the pattern of glucose availability during subsequent endurance exercises [1,2].

To further quantify glycemic stability, the CV of blood glucose concentration was calculated across all sampling time points. Glycemic variability differed significantly among beverage conditions, with the maltodextrin placebo exhibiting the greatest variability, and both chickpea milk and soy milk exhibiting lower overall variability. This quantitative assessment supports the observed differences in glycemic response patterns and indicates that the type of pre-exercise beverage influences the stability of circulating glucose concentrations during the exercise protocol. The lower glycemic variability observed following consumption of plant-based beverages may be related, at least in part, to differences in carbohydrate composition and digestibility compared with the rapidly available carbohydrates provided by maltodextrin [1,2].

The lower glycemic variability observed following chickpea milk ingestion may be partly explained by the carbohydrate characteristics of chickpeas. Chickpeas contain dietary fiber and resistant starch, and their food matrix can contribute to slower starch digestion and a gradual postprandial glycemic response [16,17]. These characteristics may help attenuate rapid fluctuations in circulating glucose levels compared with rapidly available carbohydrate sources. Accordingly, the more gradual glycemic profile observed following chickpea milk ingestion in the present study is biologically plausible; however, carbohydrate digestion, insulin responses, and gastrointestinal absorption were not directly assessed; therefore, the underlying mechanisms cannot be established from the present findings.

Despite the more stable glycemic response observed following chickpea milk ingestion, no significant differences were detected in the TTE or other physiological outcomes. Thus, the observed improvement in glycemic stability should not be interpreted as evidence of enhanced endurance performance. Although the mean TTE differed numerically among the beverage conditions, these differences did not reach statistical significance and should therefore be interpreted cautiously. The relatively small sample size may have limited the statistical power to detect modest differences in performance outcomes. Larger, adequately powered studies are needed to determine whether improved glycemic stability translates into meaningful changes in endurance performance.

The methodological strength of the present study lies in the scheduling of all experimental trials within the follicular phase of the menstrual cycle. Ovarian hormone fluctuations across the menstrual cycle have been proposed to influence metabolic and perceptual responses to exercise; however, the magnitude and consistency of these effects vary among individuals and across studies [5,8,9]. Conducting repeated experimental trials within a consistent cycle phase may help reduce potential within-participant variability when evaluating responses to nutritional interventions in female athletes [5]. Nevertheless, the menstrual cycle phase in the present study was determined using self-reported menstrual history and calendar tracking rather than hormonal verification. Therefore, the possibility of phase misclassification cannot be excluded.

Our findings should be interpreted in the context of acute nutritional intervention. Because the intervention consisted of a single pre-exercise serving, the observed glycemic response reflected an acute metabolic effect and should not be interpreted as evidence of long-term physiological adapta-tion. Although the carbohydrate characteristics of chickpeas, including dietary fiber and resistant starch, provide a plausible explanation for the more gradual glycemic response [16,17], the present study did not directly assess carbohydrate digestion or absorption. Furthermore, chickpea and soy milk produced comparable responses across the measured metabolic and physiological outcomes, suggesting that, under the present experimental conditions, chickpea milk may serve as an alternative legume-based pre-exercise beverage with acute responses comparable to those of soy milk.

This study had several limitations. First, the relatively small sample size may have limited the statistical power to detect modest differences in physiological and performance outcomes and may restrict the generalizability of the findings. Second, the menstrual cycle phase was determined using self-reported menstrual history and calendar tracking without hormonal verification, such as luteinizing hormone testing or serum ovarian hormone measurements. Consequently, some degree of menstrual-phase misclassification could not be excluded5. Third, as all experimental trials were scheduled during the follicular phase, the findings may not be generalizable to other phases of the menstrual cycle. Finally, the present study did not assess insulin responses, carbohydrate digestion, or intestinal glucose absorption; therefore, the mechanisms underlying the observed differences in glycemic variability could not be determined directly from the present data. Future studies should include larger, adequately powered samples, incorporate hormonal confirmation of the menstrual cycle phase, examine responses across different menstrual cycle phases, and include mechanistic measurements to clarify the physiological basis and potential performance relevance of the observed glycemic responses.

In conclusion, pre-exercise chickpea milk consumption was associated with a more stable glycemic response during prolonged endurance exercise than maltodextrin placebo, while no significant improvements were observed in endurance performance or other physiological outcomes. These findings suggest that chickpea milk may represent a plantbased pre-exercise nutritional option that supports acute glycemic stability in trained female runners. However, the observed glycemic response should not be interpreted as evidence of improved exercise performance. Larger, adequately powered studies that incorporate mechanistic measurements are warranted to determine whether these acute metabolic responses translate into meaningful performance benefits across different exercise conditions and athletic populations.