Creatine and HMB supplements with full-body exercise improve function and metabolism in active older adults

This randomized, double-blind, placebo-controlled crossover trial was conducted with 30 physically active older adults (62.7 ± 5.3 years; range: 60–82; 20 men, 10 women) recruited from senior centers and community sports programs in Tenerife, Spain. The study protocol was approved by the University of Burgos Ethics Committee (IR 24/2023) and prospectively registered at ClinicalTrials.gov (NCT05951439↗). All participants provided written informed consent.

During each 6-week intervention phase, participants received either 3 g/day of CRE combined with 3 g/day of HMB (free acid form) or an isocaloric placebo consisting of 6 g/day of inulin. All supplements were provided in identical, opaque sachets containing 6 g of powder per day to ensure complete blinding [18, 35–37, 45–47].

Participants were instructed to dissolve one sachet daily in yogurt or fruit juice and to consume it approximately 30 min before bedtime, a schedule previously shown to optimize gastrointestinal tolerance and adherence in older adults [29]. All nutritional guidance and supplementation instructions were delivered individually by a registered dietitian-nutritionist. Supplement packages containing 42 sachets (covering each 6-week phase) were distributed at the T1 and T3 assessments.

Compliance was monitored weekly through participant logs, direct supervision during exercise sessions, and sachet counts at each assessment. Reported adherence exceeded 95%, and no supplementation-related adverse events were observed.

The selected dosage and duration were based on previous studies demonstrating the safety and efficacy of 3 g/day of CRE and 3 g/day of HMB in both older and athletic populations [28, 35, 45]. This moderate, loading-free dosing strategy has been shown to produce measurable improvements in functional strength, metabolic efficiency, and recovery capacity within 6–8 weeks, while maintaining excellent tolerability and compliance [44].

Participants were instructed to maintain their habitual dietary patterns throughout the study. In line with current nutritional guidelines for healthy ageing and exercise performance in older adults [40], they were specifically encouraged to achieve a daily protein intake of more than 1.2 g/kg/day and an energy intake of approximately 35–40 kcal/kg/day. These recommendations were reinforced during familiarisation sessions and periodically reviewed by a registered dietitian–nutritionist.

Dietary intake was monitored using a validated, semi-quantitative Food Frequency Questionnaire (FFQ) administered at the end of each 6-week intervention phase [51]. The FFQ assessed 24 grouped food categories (dairy, meat, fish, cereals, fruits, vegetables, healthy fats, beverages, and processed foods), providing a comprehensive overview of habitual energy and macronutrient intake. This instrument has demonstrated good test–retest reliability in Spanish adults (r ≈ 0.69–0.81 for energy and macronutrients) and sensitivity to dietary variation while minimizing participant burden [51].

All participants followed the same individualized IPC programme during both intervention phases. The IPC was designed to enhance overall physical fitness and counteract age-related declines in muscle strength, power, endurance, balance, and mobility, in accordance with the American College of Sports Medicine (ACSM) guidelines for exercise prescription in older adults [54]. The programme consisted of four supervised 60-minute sessions per week, with overall adherence exceeding 90%. The training model integrated multiple components within each session to reflect the multidimensional nature of daily physical tasks, a strategy demonstrated to produce superior functional transfer compared to single-modality protocols [55, 56].

Functional performance was evaluated at baseline and post-intervention using a battery of standardized tests recommended for the assessment of physical performance in older adults by recent international expert consensus [60] All assessments were conducted under controlled laboratory conditions by trained evaluators blinded to treatment allocation.

All assessments were conducted in a temperature-controlled room at the same time of day pre- and post-intervention, following an overnight fast and 24-hour abstinence from alcohol and vigorous exercise, to minimize variability from circadian rhythms, nutrition, and physical activity.

Safety and adherence were rigorously monitored throughout the trial using a multi-method approach. Weekly check-ins and structured monitoring forms documented supplement intake, exercise attendance, perceived well-being, and any potential adverse events. Supplement adherence was verified through returned sachet counts, participant logs, and direct supervision during exercise sessions. Exercise adherence was tracked via attendance sheets and training records.

The primary endpoint for sample size estimation was gait speed, following EWGSOP2 recommendations [1]. A confirmatory family of functional performance measures (4-m gait speed, 5R-STS, TUG, and 400-m walk) were designated as co-primary outcomes. Secondary outcomes included metabolic/physiological parameters and inflammatory status (EPCR). Exploratory sex-specific analyses were conducted to examine potential differences in the response to the intervention; however, the study was not powered a priori to detect sex-by-treatment interactions.

At PRE (pooled T1/T3), no between-group differences emerged for any functional performance test (all p > 0.05). As anticipated in this high-functioning cohort, SPPB showed a marked ceiling effect, with the vast majority scoring 12/12, limiting discriminatory power at this stage. Main PRE and POST values by condition are presented in Table 1.

Significant time×group interactions were detected for 4-m gait speed, 5R-STS, TUG, and the 400-m walk in the total sample and in men (all p < 0.001), indicating superior improvement with CRE + HMB versus placebo. No interaction was observed for SPPB or static balance, consistent with ceiling effects.

At POST, the CRE + HMB phase outperformed placebo in gait speed, 5R-STS, TUG, and 400-m walk in the total sample and in men (all p < 0.05). Among women, POST differences favored CRE + HMB in gait speed, TUG, and 400-m walk (all p < 0.05), with no differences in SPPB or balance.

Within-phase changes corroborated these patterns: the CRE + HMB phase improved from PRE to POST across all functional outcomes in the total sample (SPPB, balance, gait speed, 5R-STS, TUG, 400-m; all p < 0.05). In men, all outcomes improved except balance; in women, all improved except SPPB. The placebo phase showed no significant within-phase changes.

Given SPPB ceiling effects, the primary inferences rely on sensitive mobility/endurance endpoints (gait speed, 5R-STS, TUG, 400-m). To mitigate potential carry-over, all inferential models adjusted for PRE values (pooled T1/T3), focusing on within-subject change.

At PRE, no between-group differences were detected in physiological, metabolic, or inflammatory measures (all p > 0.05) (Table 2).

Significant time×group interactions were observed for BMR in the total sample (p < 0.001), metabolic rate index in the total sample (p = 0.026), DBP in women (p = 0.020), and EPCR in men (p = 0.021). Within-phase analyses showed significant improvements in MEP during the CRE + HMB phase for the total sample and specifically for women (both p < 0.05), with no significant changes during placebo.

Between-group POST comparisons indicated higher BMR and a favorable trend for lower visceral fat during CRE + HMB versus placebo. DBP and MEP improvements were more evident in women, suggesting potential cardiovascular and respiratory benefits. However, the reduction in diastolic blood pressure, although statistically significant (p = 0.015), was clinically modest (< 5 mmHg), indicating a mild physiological adaptation rather than a therapeutic effect. No significant effects were identified for SBP, heart rate, or SpO₂.

Analysis of the inflammatory biomarker EPCR revealed a sex-specific response. While the time×group interaction was not significant for the total sample (p = 0.920), a significant interaction in men reflected divergent changes between conditions. The modest rise in EPCR during the CRE + HMB phase remained within the physiological range and was not accompanied by adverse changes in blood pressure or systemic inflammation, suggesting a context of adaptive endothelial modulation rather than dysfunction.

Exploratory regression models identified several independent predictors of functional performance (Table 3). The supplementation group (CRE + HMB) consistently emerged as a primary determinant of improved outcomes, particularly for gait speed, 5R-STS, and TUG, confirming the direct contribution of the intervention.

For balance, additional predictors included sex (better in women), higher BMR, lower metabolic age, lower DBP, and higher MEP. For 5R-STS, supplementation was the main predictor, with ΔSpO₂ providing additional support. For TUG, supplementation again predicted better performance, alongside lower metabolic age, higher BMR, and higher SpO₂. For the 400-m walk, lower metabolic age and higher BMR predicted faster times, while group allocation did not remain significant. Across models, adjusted R² ranged from 0.13 to 0.36, underscoring the clinical relevance of supplementation as the principal driver of functional improvements, complemented by the contribution of metabolic and physiological factors.

Exploratory ANCOVA models were conducted to formally test for potential carryover and sequence effects by including period (T1 vs. T3), treatment sequence (Placebo→CRE + HMB vs. CRE + HMB→Placebo), and their interaction, adjusted for age. As detailed in Table 4, the sequence factor was not a significant source of variation for any functional, metabolic, physiological, or inflammatory outcome in the total sample or when stratified by sex. Although a small number of isolated period × sequence interactions reached statistical significance (e.g., for 5R-STS and 400-m walk), these effects were inconsistent across outcomes and sex strata and did not influence the primary findings related to supplementation. Collectively, these analyses support the adequacy of the 3-week washout period and confirm that carryover or order effects are unlikely to explain the observed benefits of supplementation.

The present study demonstrates that six weeks of combined creatine plus β-hydroxy-β-methylbutyrate (CRE + HMB) supplementation, integrated into a supervised multicomponent training program, produced meaningful improvements in functional performance and metabolic efficiency in physically active older adults. These results expand current evidence supporting the complementary mechanisms of both supplements in enhancing muscular energetics, recovery capacity, and functional outcomes in aging populations [45, 50, 66].

Rather than reiterating well-established mechanisms of CRE and HMB, this discussion focuses on how their integration with structured multicomponent exercise translates into measurable improvements in gait speed, strength, and metabolic parameters. The magnitude of these effects—supported by large partial η² values (0.17–0.42)—indicates physiological and clinical relevance. Similar magnitudes of improvement have been reported in recent meta-analyses of CRE supplementation during resistance or multicomponent training in older adults [20, 42, 67].

From a translational perspective, the functional outcomes that showed the largest effect sizes—particularly gait speed, chair-stand performance, TUG, and walking endurance—are well-established indicators of mobility status and predictors of adverse health events in older adults [1–3, 31, 32]. Impairments in these measures have been associated with a higher risk of falls, mobility limitation, and loss of functional independence [2, 3, 31, 32]. Therefore, the magnitude of the observed effects likely reflects changes that extend beyond statistical significance and may be clinically meaningful, particularly in terms of mobility preservation and maintenance of independence in physically active older adults. Although clinical endpoints such as falls or hospitalizations were not directly assessed, the observed improvements in these functional domains suggest a potential for favorable downstream effects on clinically relevant outcomes that are known to be function-dependent in older adults.

Moreover, our findings align with current consensus guidelines emphasizing the integration of nutritional support and progressive exercise for the prevention of frailty and maintenance of functional independence in aging [56, 60]. The combined strategy tested here exemplifies a feasible and translational approach to enhance both muscular and metabolic resilience, complementing the growing body of literature supporting multimodal interventions for healthy aging [14, 15, 66].

Building on these findings, the present study complements our previous crossover trial in the same population [29], which demonstrated improvements in muscle strength and endurance independent of mass gains. Together, both datasets provide a coherent picture of how CRE + HMB supplementation enhances functional capacity across multiple performance domains—strength, mobility, and metabolic health—in physically active older adults. This dual evidence reinforces the concept that CRE + HMB supplementation enhances muscle quality and functional efficiency, even in individuals without overt sarcopenia or frailty.

In line with this functional enhancement, the improvements observed in gait speed, 5R-STS , TUG , and the 400-m walk are clinically meaningful and align with previous studies reporting beneficial effects of CRE and HMB—administered individually or in combination with exercise—on functional capacity in older adults [18–21, 27, 28]. In contrast to studies limited to resistance training, our IPC model integrated resistance, endurance, balance, and coordination, providing a more comprehensive stimulus for both functional and metabolic adaptation.

The absence of significant changes in the SPPB likely reflects a ceiling effect associated with the high baseline functional status of the participants, rather than a lack of intervention efficacy. Ceiling effects in composite functional scales such as the SPPB have been widely reported in physically active or non-frail older adults, in whom baseline scores are often close to the upper limit of the scale, limiting the sensitivity of these instruments to detect further improvements following an intervention [68, 69]. In such contexts, additional physiological and neuromuscular adaptations may occur without translating into measurable changes in global performance scores. Accordingly, the lack of significant SPPB changes in the present study should be interpreted as a limitation of measurement sensitivity rather than as an absence of functional benefit.

From a study design perspective, these findings suggest that the SPPB may be more sensitive for detecting intervention effects in frailer populations or individuals with lower baseline function, where greater scope for improvement exists [70]. Conversely, future trials targeting higher-functioning older adults may benefit from incorporating more demanding or continuous functional assessments—such as extended gait tests, power-based measures, or endurance tasks—to better capture intervention-related adaptations.

According to partial eta-squared (η²p) thresholds proposed by Richardson (2011) [71], most functional outcomes—particularly gait speed, chair-stand performance, TUG, and the 400-m walk—showed large effect sizes (η²p = 0.15–0.29). These magnitudes confirm that the observed functional gains are not only statistically significant but also clinically meaningful, reinforcing the practical relevance of the intervention for maintaining mobility and independence in older adults.

The functional improvements observed in this study likely arise from convergent muscular, metabolic, and cardiovascular adaptations elicited by the combined CRE + HMB and exercise intervention. Both supplements act through complementary mechanisms that enhance muscle energetics and tissue efficiency. Creatine supports rapid ATP resynthesis through the phosphocreatine system, sustaining contractile performance and delaying fatigue during repeated efforts (for instance [20, 37, 42, 45, 50],. HMB, in turn, has been reported to attenuate proteolysis and enhance protein turnover and recovery, which may contribute to improved muscle integrity and functional performance in older adults [24, 27, 36]. Together, these actions are consistent with improvements in muscle quality—that is, greater functional output per unit of muscle—described in older populations [8]. The observed increases in BMR and metabolic rate index are compatible with an upregulation of energy turnover and bioenergetic efficiency during training and recovery. In older adults, CRE supplementation alongside resistance-type exercise has been associated with enhanced training adaptations (e.g., strength, lean mass), anabolic signaling, and muscle remodeling [19–21, 41, 42]and may modulate exercise-related oxidative and inflammatory responses [72]. These effects, integrated with supervised multicomponent training, provide a plausible physiological basis for the mobility improvements reported here [12, 14, 15, 55, 64].

Furthermore, the concomitant reductions in visceral fat index and diastolic blood pressure—particularly in women—suggest systemic metabolic and vascular benefits of the combined strategy. In women, the statistically significant yet clinically modest reduction in diastolic blood pressure (< 5 mmHg) likely reflects a mild enhancement in vascular compliance or autonomic regulation rather than a true antihypertensive effect. Prior evidence indicates that exercise-based programs in older adults can improve body composition and cardiometabolic health [12, 14–17, 55, 64] and attenuate low-grade inflammation linked to aging (“inflammaging”) [7, 72, 73]. Collectively, these findings support the concept that integrating targeted supplementation with structured multicomponent exercise can help sustain mobility and functional independence with aging [1, 10, 12, 14, 15].

In addition, the modest increase in circulating EPCR observed after the intervention should be interpreted with caution. While elevated EPCR may reflect enhanced endothelial turnover and adaptive remodeling secondary to exercise and improved perfusion, it can also indicate compensatory activation of the protein C pathway in response to subclinical vascular stress [33, 34]. The absence of concomitant rises in diastolic blood pressure or inflammatory markers, together with improved functional and metabolic parameters, supports the interpretation of this response as part of a favorable endothelial adaptation rather than endothelial dysfunction. While this adaptive interpretation appears most plausible, a transient pro-inflammatory response cannot be entirely ruled out. Nevertheless, future studies incorporating direct measures of vascular reactivity and soluble EPCR isoforms are warranted to clarify the precise clinical significance of this finding in aging populations.

Collectively, these findings illustrate that improvements in muscle energetics and vascular efficiency may underpin the enhanced functional outcomes observed. These physiological adaptations are further elaborated in Sect. 4.6, where mechanistic pathways underlying the observed synergy between CRE and HMB are discussed in greater depth.

Regarding the methodological design, although reviewers highlighted the relatively short 6-week intervention and 3-week washout, several methodological and physiological arguments justify this design. First, previous trials have shown that moderate, loading-free CRE doses (3 g/day) achieve measurable functional and metabolic benefits within 4–6 weeks [27, 29, 42]. Second, under this dosing strategy, intramuscular creatine declines by approximately 1.6–1.7% per day after cessation, returning to baseline within 3–4 weeks [42, 43]. Third, formal analyses of potential carryover effects—including comparison of T1 vs. T3 values and ANCOVA models incorporating period and sequence terms (Table 4)—confirmed the absence of residual effects. These results, consistent with prior studies in older adults using similar protocols, validate the adequacy of the washout period and the robustness of the crossover design.

In addition to this methodological rigor, the present analysis specifically addressed several statistical considerations raised by reviewers. Partial eta-squared (η²p) values were interpreted using established thresholds (small ≥ 0.01, medium ≥ 0.06, large ≥ 0.14) [71], providing a clearer understanding of the magnitude and practical relevance of the observed effects. The inclusion of Table 4 allowed formal testing for potential period, sequence, and carryover effects, thereby strengthening the internal validity and robustness of the crossover design. Furthermore, the regression analyses extended the inferential depth by identifying key metabolic (BMR, metabolic age) and physiological (SpO₂, DBP) predictors of functional gains.

Observed effect sizes for the primary outcomes (η²p = 0.17–0.42) were notably higher than the medium threshold assumed in the a priori power analysis, suggesting that the achieved sample size and statistical approach provided sufficient power to detect meaningful effects while minimizing type II error risk. This coherence between a priori estimates and observed effects supports the methodological soundness and reproducibility of the experimental design. Collectively, these results highlight the integrative nature of functional performance in aging, where metabolic efficiency and cardiorespiratory health jointly sustain mobility and independence.

From a mechanistic perspective, both supplements act through complementary mechanisms that plausibly explain the observed synergy. Creatine increases intramuscular phosphocreatine availability, supports rapid ATP resynthesis, enhances high-intensity exercise capacity, and stimulates anabolic pathways such as mTOR and satellite cell proliferation [18, 21, 41]. HMB, on the other hand, attenuates proteolysis through inhibition of the ubiquitin–proteasome pathway, stabilizes cell membranes, and promotes recovery from exercise-induced muscle damage [24–26, 72] Together, these actions enhance muscle contractile quality and reduce fatigue during repeated submaximal efforts, leading to improved mobility and endurance. Furthermore, both nutrients have been linked to improved mitochondrial biogenesis, oxidative metabolism, and endothelial function [19, 25, 35] which may partly explain the concurrent improvements in BMR, oxygen utilization, and cardiopulmonary function observed here.

Several limitations of the present study should be acknowledged. First, the relatively short intervention period (six weeks) limits the ability to assess the persistence of benefits or long-term adaptations following supplementation cessation. Second, body composition and basal metabolic rate were estimated using multifrequency bioelectrical impedance rather than gold-standard methods such as dual-energy X-ray absorptiometry or indirect calorimetry, which may reduce the precision of absolute values despite the method’s strong reliability for within-person changes. Third, the fixed dosing strategy (3 g/day of CRE and 3 g/day of HMB) was not normalized to body mass and could have contributed to interindividual variability, including the sex-specific differences observed. Moreover, biochemical verification of supplement adherence and single-supplement comparison arms (CRE-only or HMB-only) were not included, precluding a direct evaluation of additive versus synergistic effects. Although sex-related differences in fat-free mass (FFM) were evident, no statistical adjustment for this covariate was applied to avoid model overfitting given the modest sample size. These differences likely reflect physiological dimorphism rather than treatment-specific effects, yet future studies with larger cohorts should explore FFM-adjusted analyses to better disentangle compositional from functional influences. In addition, sex-specific analyses were exploratory in nature and the study was not powered a priori to formally test sex-by-intervention interactions. Furthermore, confidence intervals were not systematically reported for all effect size estimates, which may limit the precision of interpretation regarding the range of plausible effects. Furthermore, the high baseline functional status of the participants may have limited the sensitivity of the SPPB to detect further improvements, consistent with a ceiling effect in well-functioning older adults.

Nonetheless, several aspects mitigate these limitations: (i) the 3-week washout period exceeded the time generally required for reversal of exercise-induced functional gains in older adults; (ii) baseline equivalence (T1 vs. T3) and the absence of significant sequence effects (Table 4) support the adequacy of this control; and (iii) participants were instructed to maintain their habitual routines throughout the study, minimizing unsupervised variability. Collectively, these factors reinforce confidence that the observed effects primarily reflect the supplementation–exercise interaction rather than residual or external influences. Accordingly, no clinically relevant sequence or period effects were observed (Table 4), confirming the validity of the pooled PRE/POST analysis. Moreover, the training-induced baseline improvement across phases might have attenuated some between-condition contrasts.

Despite these constraints, this study presents several notable strengths. The randomized, double-blind, crossover design maximized statistical power and minimized interindividual variability. The intervention achieved excellent adherence (> 90%) and was conducted under direct supervision, ensuring methodological rigor and ecological validity. Moreover, the combined evaluation of functional, metabolic, and physiological outcomes provides a comprehensive understanding of the effects of CRE + HMB supplementation integrated with structured multicomponent training. Observed effect sizes exceeded a priori expectations, supporting the adequacy of the sample size and reinforcing the robustness of the study’s crossover design.

Future studies should extend these findings by incorporating longer follow-up periods, body mass–adjusted dosing, biochemical adherence verification, and advanced biomarkers of muscle metabolism and endothelial function. Such approaches will help clarify the mechanistic pathways and long-term sustainability of these benefits. Overall, the present results support the integration of targeted nutritional supplementation with multicomponent exercise as a feasible and safe strategy to enhance mobility, metabolic efficiency, and healthy aging in older adults.

This study did not receive any specific grants from funding agencies in the public, commercial, or not-for-profit sectors.

R.R.-H. and J.M.-A. conceived and designed the study; oversaw data collection, analysis, and interpretation; and drafted the manuscript. A.M.-O., M.M.-F., M.S.-R., M.S.-P., and N.B. contributed to data acquisition, intervention supervision, and critical revision of the manuscript for important intellectual content. All authors participated in the interpretation of the results, reviewed and approved the final version of the manuscript, and agreed to be accountable for all aspects of the work.

The authors declare no competing interests.

All procedures performed in this study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards. The study protocol was approved by the Ethics Committee of the University of Burgos (approval code: IR 24/2023) and registered at ClinicalTrials.gov (Identifier: NCT05951439↗). Written informed consent was obtained from all participants prior to enrollment.

Generative AI was not used to create or alter figures or data. Language editing assistance (if any) was performed under human oversight; authors take full responsibility for the content.

The datasets generated and analyzed during the current study are available from the corresponding author on reasonable request.