Abstract
tudy examined nutritional intake, body composition, menstrual health, and performance in elite female trail runners.Methods: A cross-sectional multivariate anal- ysis was conducted on 35 athletes (14 eumenorrheic, 21 amenorrheic/oligomenorrheic). Nu- tritional intake was assessed through 7-day and 24 h food records; anthropometry followed ISAK standards; performance was evaluated via ITRA and UTMB rankings. Statistical analyses includedt-tests, MANCOVA, regression models, and Random Forest, adjusting for body composition and covariates.Results: Although energy availability (EA) did not differ significantly between groups, 94.3% of athletes had clinically low EA (<30 kcal/kg FFM/day). Amenorrheic athletes consumed more simple carbohydrates (21.8±5.7% vs. 17.2±3.1%), protein (2.5±0.6 vs. 1.7±0.2 g/kg/day), fiber, and lipids, while eumenor- rheic athletes consumed more complex carbohydrates (129.7±27.0 vs. 82.5±33.3 g/day) and most vitamins. Both groups had inadequate calcium and iron intake. Low
energy availability (EA) did not differ significantly between groups, 94.3% of athletes had clinically low EA (<30 kcal/kg FFM/day). Amenorrheic athletes consumed more simple carbohydrates (21.8±5.7% vs. 17.2±3.1%), protein (2.5±0.6 vs. 1.7±0.2 g/kg/day), fiber, and lipids, while eumenor- rheic athletes consumed more complex carbohydrates (129.7±27.0 vs. 82.5±33.3 g/day) and most vitamins. Both groups had inadequate calcium and iron intake. Low EA was moderately associated with an ectomorphic somatotype (r = 0.418). Performance negatively correlated with simple carbohydrates (r =−0.624) and positively with complex carbohy- drates, total energy, protein, polyunsaturated fats, and zinc (r = 0.300–0.580). No significant performance differences were found between menstrual status groups.Conclusions: Men- strual irregularities did not affect performance, but nutritional patterns strongly influenced both performance and energy availability. Personalized nutrition strategies are essential for optimizing performance and safeguarding health in elite female trail runners. Keywords:female athletes; relative energy deficiency in sport; low energyavailability; menstrual dysfunction; sports medicine; ultra-endurance running; anthropometry; endurance sports 1. Introduction Until relatively few years ago, research in female physiology has primarily focused on reproduction, ignoring the non-reproductive functions of ovarian hormones such as estrogen and progesterone, whose cyclical fluctuations are regulated by pituitary gonadotropins—follicle-stimulating hormone (FSH) and luteinizing hormone (LH)—and J. Funct. Morphol. Kinesiol.2025,10, 482
J. Funct. Morphol. Kinesiol.2025,10, 482 2 of 20 not only ensure menstrual and reproductive function but also modulate bone, cardio- vascular, thermoregulatory, neuromuscular, and metabolic processes that may influence endurance performance in female athletes [1–4]. For many years, women have rarely been included as participants in sports science research, and even when they are, the applied methodologies often fail to consider their specific physiological needs [5]. A recent analysis of sports medicine and exercise science publications found a marked underrepresenta- tion of female participants, with only about 4% to 13% of articles focused exclusively on women [6,7]. Even when women are included in sports research, their hormonal fluctu- ations are often overlooked, or tests are frequently conducted during menstrual phases of low hormone levels to minimize potential hormonal interference [7,8]. These research gaps and omissions can be attributed to several factors [5]: a lack of progress in the de- velopment of women’s sports compared to men’s (often accompanied by discouraging attitudes toward female competitive participation), a shortage of researchers with expertise in female endocrinology and exercise physiology, the greater financial and time investment required for high-quality research on female athletes, and significantly less funding for women-focused research projects compared to those focused on men. These historical limi- tations have resulted in a dearth high-quality evidence to properly guide female athletes in their sporting pursuits. In competitive sports, female athletes are particularly vulnerable to adverse health effects often linked to low energy availability—LEA [9]. Energy availability is defined as the amount of energy remaining for physiological processes after subtracting the energy expended during exercise [10]. When energy availability is insufficient, critical processes such as cellular maintenance, thermoregulation, and reproductive function can be compro- mised [11]. In many cases, this state occurs even in otherwise healthy athletes due to a lack of awareness of their daily energy requirements relative to training loads and competitive demands. However, if this energy deficiency persists over time, it can lead to a number of adverse consequences, including menstrual and reproductive dysfunctions [12], com- promised bone health [11], impaired vitamin D absorption [13], immune dysfunctions [14], reduced
even in otherwise healthy athletes due to a lack of awareness of their daily energy requirements relative to training loads and competitive demands. However, if this energy deficiency persists over time, it can lead to a number of adverse consequences, including menstrual and reproductive dysfunctions [12], com- promised bone health [11], impaired vitamin D absorption [13], immune dysfunctions [14], reduced protein synthesis [15], increased cardiovascular risk [16], and diminished athletic performance [17]. In the most extreme cases, particularly among athletes who deliber- ately and dramatically restrict their caloric intake, dangerous eating behaviors can emerge, potentially progressing to eating disorders [18]. In recent decades, the exponential rise in endurance and ultra-endurance events has led to a surge in the popularity of trail running—TR—[19]. However, research on this discipline remains limited compared to other sports, and studies focusing specifically on female trail runners are even scarcer [20–23]. Although some studies indicate that sex-based hormonal differences can influence the performance of female trail runners [19,24,25], other biological factors differentiating female and male athletes remain insufficiently explored in this context [20,21,26,27]. In terms of sports nutrition, only a few studies to date have analyzed female trail runners’ nutritional intake and hydration strategies during competition [28–31], and no study has yet thoroughly examined their daily dietary habits. This lack of research is concerning, since elite female trail runners—due to their high training loads and competitive demands—constitute a high-risk population for developing metabolic and hormonal disorders. In fact, a significant prevalence of secondary amen- orrhea and oligomenorrhea has been reported in this population, underscoring the need to characterize their dietary habits and body composition with respect to menstrual cycle status [32]. This study aims to provide a detailed characterization of the nutritional pat- terns, body composition profile, and athletic performance of international-level female trail runners, analysing how nutrition and body composition relate to both athletic performance
J. Funct. Morphol. Kinesiol.2025,10, 482 3 of 20 and menstrual cycle status, as well as examining the influence of the menstrual cycle on athletic performance. 2. Materials and Methods 2.1. Ethical Approval This study was approved by the Ethics Committee for Clinical Research of the Catalan Sports Council (0099/S690/2013). All participants provided written informed consent and received detailed information about the procedures. To ensure the protection of participants’ rights, the study complied with the ethical principles of the Declaration of Helsinki, the Nuremberg Code, and the Belmont Report, as well as all applicable national and regional regulations. 2.2. Trial Design This investigation was designed as a descriptive, comparative, and correlational cross-sectional study using a non-probabilistic convenience sample. Participants were stratified into two groups according to menstrual status (eumenorrheic vs. secondary amenorrheic/oligomenorrheic), allowing for both group-based comparisons and multi- variate exploration. Four main assessment domains were included, each composed of multiple independent variables: (1) Nutritional assessment—encompassing metrics related to total energy intake, energy availability, macronutrient distribution, fiber, and intake of key micronutrients; (2) Anthropometric assessment—including anthropometric charac- teristics, body composition estimates, and somatotype components; (3) Menstrual health assessment—evaluating indicators of reproductive function based on questionnaire re- sponses; and (4) Performance assessment—based on international competitive rankings. The design allowed for the identification of differences between menstrual status groups across all domains, while also supporting the examination of interrelationships among nu- tritional, physiological, and performance-related variables. Menstrual status was modeled as the main outcome of interest to determine which factors were most strongly associated with menstrual dysfunction in this athletic population. Additionally, the influence of nutri- tional intake on sports performance was analyzed, accounting for relevant anthropometric characteristics. The analytical framework incorporated both group-level comparisons and integrated multivariate modeling, with consideration of covariates and interactions across variables, enabling a comprehensive understanding of how energy availability, dietary habits, and physical characteristics relate to menstrual health and athletic performance in elite female trail runners. 2.3. Participants A total of 35 elite female trail runners, all members of the Spanish national team, voluntarily participated in the present study. To ensure adequate statistical power, an a priori
and interactions across variables, enabling a comprehensive understanding of how energy availability, dietary habits, and physical characteristics relate to menstrual health and athletic performance in elite female trail runners. 2.3. Participants A total of 35 elite female trail runners, all members of the Spanish national team, voluntarily participated in the present study. To ensure adequate statistical power, an a priori sample size estimation was performed using G*Power v3.1.9.6, based on a two-group comparison with an assumed effect size of d = 0.85, a power of 72.0%, and a significance level ofα= 0.05. The analysis indicated a minimum required sample size ofn= 32 (n= 11for group 1 andn= 21 for group 2). Recruitment was conducted in collaboration with the Spanish Federation for Mountain and Climbing Sports (FEDME). At the time of data collection, all female athletes were in the general preparatory phase of their annual training cycle and had competed at the international level either during the current or previous competitive season. General characteristics and training backgrounds of the participants are presented in Table, according to the final grouping based on menstrual status: eumenorrheic (n= 14) and amenorrheic/oligomenorrheic (n= 21).
J. Funct. Morphol. Kinesiol.2025,10, 482 4 of 20 Table 1.Chronological age and training characteristics of mountain runners according to whether they report menstrual cycle dysfunctions or not. Total Group (n= 35) Eumenorrhea (n= 12) Secondary Amenor- rhea/Oligomenorrhea (n= 23) Difference (%) t-Test (p) Effect-Size ( d) Age (years) 33.7 ±7.5 37.1 ±6.5 31.9 ±7.5 15.6 −2.049 (0.048) * 0.69 Training volume (h/week) 12.9±3.7 12.7 ±3.1 13.0 ±4.0 2.3 0.219 (0.828) 0.08 Previous experience (years) 8.4±5.1 9.9 ±4.9 7.6 ±5.1 27.1 −1.263 (0.216) 0.45 Competitive ranking (ITRA-UTMB points) 656.9±63.6 663.8±36.7 653.3 ±74.5 1.6 −0.554 (0.583) 0.17 Values are expressed as mean±standard deviation. ITRA-UTMB: International Trail Running Association–Ultra- Trail du Mont-Blanc performance index. ES: Effect Size (Cohen’s d); * indicates statistically significant differences (p< 0.05);p: Statistical significance level (p-value). Inclusion criteria required participants to be adult female athletes actively engaged in structured training and international competition. Exclusion criteria comprised a prior diagnosis of anorexia nervosa (as defined by DSM-V), diabetes mellitus, corticosteroid therapy, hyperparathyroidism, gastrointestinal conditions involving malabsorption, or other disorders potentially affecting energy metabolism. To minimize confounding effects on body composition, athletes were also excluded if they had performed physical activity within 48 h prior to testing or reported the use of dietary supplements known to alter tissue dynamics. 2.4. Procedures 2.4.1. Nutritional Assessment Each participant completed a dual dietary assessment: a 7-day food diary and a 24 h recall, capturing all food and beverage intake. Multi-day records are widely used in sports nutrition to account for intake variability [33], with 3–7 days considered a practical and valid window for athletes [34]. To improve accuracy, participants received visual guides with portion-size images of common foods. Such aids enhance the validity of self-reported intake [35]. Individual instructions emphasized precise reporting of quantities, preparation methods, and timing. Supplement and medication use (e.g., iron or oral contraceptives) was also recorded, given their dietary relevance [36]. All records were reviewed one-on-one by a sports nutritionist to resolve ambiguities and ensure completeness—an approach that minimizes reporting errors [37]. Food diaries were analyzed using CESNID Nutritional Analysis Software ® (version 1.0) [38], developed by the University of
preparation methods, and timing. Supplement and medication use (e.g., iron or oral contraceptives) was also recorded, given their dietary relevance [36]. All records were reviewed one-on-one by a sports nutritionist to resolve ambiguities and ensure completeness—an approach that minimizes reporting errors [37]. Food diaries were analyzed using CESNID Nutritional Analysis Software ® (version 1.0) [38], developed by the University of Barcelona (Barcelona, Spain), based on a Spanish-specific food composition database. Nutrient and energy intakes were compared to recommended guidelines for athletes [39]. To assess energy adequacy, total intake was compared to individualized requirements: basal metabolic rate was calculated via the Harris-Benedict equation with a 1.3 activity factor for light daily activity [40], and training-related expenditure was added using MET values from the Compendium of Physical Activities [41]. Total energy intake was expressed relative to each athlete’s estimated needs, and all nutritional variables were reported as percentages of optimal recommendations [39], facilitating the evaluation of dietary adequacy and potential performance or health implications. 2.4.2. Anthropometric Assessment A full anthropometric profile was assessed in accordance with the International Soci- ety for the Advancement of Kinanthropometry (ISAK) standards [42]. All measurements were conducted by a single accredited anthropometrist (ISAK Level 3) who was part of the research team, ensuring consistency and minimizing inter-evaluator variability. The protocol included basic measurements (body mass, stretch stature, sitting height, and arm span); eight skinfolds (triceps, subscapular, biceps, iliac crest, supraspinale, ab-
J. Funct. Morphol. Kinesiol.2025,10, 482 5 of 20 dominal, thigh, and calf); girths (head, neck, arm relaxed, arm flexed and tensed, fore- arm, wrist, chest, waist, hip, thigh 1 cm gluteal, thigh middle, calf, and ankle); lengths and heights (acromiale–radiale, radiale–stylion, midstylion–dactylion, iliospinale height, trochanterion height, trochanterion–tibiale laterale, tibiale laterale height, foot, and tibiale mediale–sphyriontibiale); and breadths (biacromial, antero-posterior abdominal depth, biiliocristal, transverse chest, antero-posterior chest depth, humerus, bi-styloid, femur, and bimalleolar). All anthropometric variables were measured twice using non-consecutive measurements. A third measurement was taken if the difference between the first two exceeded 1% for all variables, or 5% in the case of skinfolds [42]. Final values were calcu- lated as the mean of the first two measurements, or the median of the three if a third was required. The intra-evaluator technical error of measurement was notably low: 0.021% for basic measurements, 1.37% for skinfolds, 0.83% for girths, 0.32% for lengths and heights, and 0.45% for breadths and depths, confirming a high level of precision. Anthropometric instruments included a Seca 220 ® telescopic stadiometer (Seca GmbH & Co. KG, Hamburg, Germany) (range: 85–200 cm; precision: 0.1 cm); a pre-calibrated Seca 710 ® mechanical scale (Seca GmbH & Co. KG, Hamburg, Germany) (capacity: 200 kg; precision: 0.05 kg); a Realmet©anthropometric tape (Realmet, Nottingham, UK) (precision: 0.1 cm); a Realmet© segmometer (Realmet, Nottingham, UK) (precision: 0.1 cm); a Harpenden ® skinfold caliper (Baty International, Burgess Hill, UK) (range: 0–90 mm; precision: 0.2 mm; constant pres- sure: 10 g/mm 2 ); and Holtain ® large and small sliding bone breadth calipers (Holtain Ltd., Crymych, UK) (range: 0–25 cm; precision: 0.1 cm). From the full ISAK-compliant profile, several derived indicators were calculated. Body proportionality indices and muscular cross-sectional areas were used to assess structural relationships among anthropometric dimensions. Body composition estimates—including fat mass [43], muscle mass [44], and bone mass [45]—were computed using validated anthropometric equations for athletic pop- ulations. Additionally, the Heath-Carter somatotype method [46] was used to determine individual morphological profiles, expressed in terms of endomorphy (relative fatness), mesomorphy (muscularity), and ectomorphy (linearity). 2.4.3. Menstrual Health Assessment Menstrual health was assessed
structural relationships among anthropometric dimensions. Body composition estimates—including fat mass [43], muscle mass [44], and bone mass [45]—were computed using validated anthropometric equations for athletic pop- ulations. Additionally, the Heath-Carter somatotype method [46] was used to determine individual morphological profiles, expressed in terms of endomorphy (relative fatness), mesomorphy (muscularity), and ectomorphy (linearity). 2.4.3. Menstrual Health Assessment Menstrual health was assessed through a two-step approach. First, each athlete completed a detailed ad hoc menstrual history questionnaire, developed based on es- tablished literature definitions to classify their cycle status as eumenorrheic or amenor- rheic/oligomenorrheic. Amenorrhea was conservatively defined as the absence of menstru- ation for over 90 days, in accordance with established clinical criteria [47]. Oligomenorrhea was defined as menstrual cycles longer than 35 days (approximately 5–7 periods per year), often associated with hypothalamic-pituitary-ovarian axis dysfunction in female athletes [48]. Second, an adapted version of the Low Energy Availability in Females Questionnaire (LEAF-Q) was administered, focusing exclusively on the reproductive and injury-related items. This adaptation was chosen to target key indicators of Relative Energy Deficiency in Sport (RED-S) [49], while avoiding redundancy, as gastrointestinal symptoms were assessed more precisely through comprehensive dietary intake data obtained from the 7-day food diary and 24 h recall. The LEAF-Q’s proven sensitivity, ease of administra- tion, and validation across athletic populations supported its use in this elite endurance cohort [10,50]. This approach allowed a focused and efficient assessment of menstrual disturbances and injury history relevant to RED-S risk without overlap with detailed nutritional data.
J. Funct. Morphol. Kinesiol.2025,10, 482 6 of 20 2.4.4. Performance Assessment Athletic performance was assessed using scores from two internationally recognized ranking systems: the ITRA ® Performance Index (International Trail Running Association) and the UTMB ® Index (Ultra-Trail du Mont-Blanc). Both systems are widely used in the trail running field to quantify competitive level, assigning standardized scores ranging from 0 to 1000 points through proprietary algorithms developed by their respective orga- nizations. These indices integrate objective parameters such as race distance, elevation gain, and finishing time, alongside contextual factors like terrain type and environmental conditions. The algorithms draw on extensive databases of historical race results, enabling normalization and cross-event comparability. To increase the robustness and representa- tiveness of the performance assessment, both scores were combined by calculating their arithmetic mean. This approach yielded a coherent and reliable indicator of each athlete’s competitive status. 2.5. Statistical Analysis The distribution of each variable was examined using the Kolmogorov–Smirnov test to assess normality. Descriptive statistics were reported as means and standard deviations. Between-group comparisons (eumenorrheic vs. amenorrheic/oligomenorrheic) were con- ducted using either the independent samples Student’st-test or the Mann–Whitney U test, depending on whether the assumptions of normality were met. Effect sizes were calculated using Cohen’s d, with the following thresholds: 0.21–0.49 (small), 0.50–0.70 (moderate), and≥0.71 (large). For correlation coefficients, effect size thresholds were interpreted as |r| = 0.10–0.29(small), |r| = 0.30–0.49 (moderate), and |r|≥0.50 (large) [51]. Associ- ations between anthropometric parameters and indicators of LEA were analyzed using Pearson or Spearman correlation coefficients, depending on data distribution. Additionally, Random Forest models were implemented as a non-linear classification technique, incorpo- rating bootstrap validation (100 iterations) and a train/test data partitioning strategy to enhance model robustness. To account for potential confounding factors and to evaluate ad- justed group differences, a Multivariate Analysis of Covariance (MANCOVA) was applied. The relationship between nutritional intake and athletic performance was explored using correlation matrices, stepwise multiple linear regression models, and univariate regression analyses. To reduce the likelihood of Type I error,p-values were corrected using the False Discovery Rate (FDR) procedure. Multicollinearity was assessed using the Variance Infla- tion
evaluate ad- justed group differences, a Multivariate Analysis of Covariance (MANCOVA) was applied. The relationship between nutritional intake and athletic performance was explored using correlation matrices, stepwise multiple linear regression models, and univariate regression analyses. To reduce the likelihood of Type I error,p-values were corrected using the False Discovery Rate (FDR) procedure. Multicollinearity was assessed using the Variance Infla- tion Factor (VIF), and partial correlations were computed with adjustment for covariates. All statistical procedures were performed in RStudio (version 4.4.3, RStudio, PBC, Boston, MA, USA), with the alpha level for statistical significance set atp≤0.050. 3. Results 3.1. Nutritional Results This section presents the results related to the nutritional status of the female trail runners included in the study (Table).
Description
The research explores the relationship between nutrition and performance in female trail runners.