Beyond angina: Ranolazine is associated with increased forced-swimming endurance and altered echocardiographic parameters in healthy rats: a comparative study with Trimetazidine

Abstract

Background

Anti-anginal agents that modulate myocardial substrate metabolism, such as ranolazine and trimetazidine, are pharmacologically capable of altering exercise performance independently of their licensed indication in ischemic heart disease. Trimetazidine has been included on the World Anti-Doping Agency (WADA) Prohibited List since 2014 on the basis of comparable evidence. Whether ranolazine, a mechanistically distinct late sodium current inhibitor, produces analogous performance-modifying effects in the absence of myocardial ischemia has not been systematically examined. This exploratory, hypothesis-generating study characterized the functional exercise and structural echocardiographic correlates of chronic ranolazine and trimetazidine administration in healthy rats.

Methods

Twenty-four male Sprague–Dawley rats were randomized in a blinded, parallel-group design to ranolazine (50 mg/kg/day), trimetazidine (10 mg/kg/day), or saline control (n = 8 per group) for 30 days. Two control-group rats died under ketamine-xylazine anesthesia during baseline (day 0) echocardiography, before any study data were collected, and were excluded from all subsequent analyses, yielding a final analytic sample of 22 rats (control, n = 6; trimetazidine, n = 8; ranolazine, n = 8). Forced swimming, rotarod testing, and transthoracic echocardiography were performed at baseline, day 15, and day 30. Because Shapiro–Wilk testing indicated non-normal distributions, non-parametric statistics (Kruskal–Wallis, Friedman, Wilcoxon signed-rank with Bonferroni-adjusted post-hoc thresholds) were applied a priori, and exact p-values are reported throughout.

Results

Total swimming time differed significantly between groups (ranolazine 39,668 s [33,794–43,838]; trimetazidine 35,694 s [30,776–39,516]; control 32,426 s [29,389–34,453]; p = 0.006), with ranolazine exceeding trimetazidine in sessions 16–18 (p < 0.001). Rotarod performance was unchanged across groups and time (p > 0.05 for all comparisons). Echocardiography showed small but statistically significant increases in interventricular septal thickness and calculated left ventricular mass in all three groups, including controls, with ejection fraction and fractional shortening preserved throughout.

Conclusions

Chronic ranolazine administration was associated with exercise-performance changes that paralleled, and in later sessions exceeded, those produced by trimetazidine, together with modest echocardiographic changes that are compatible with, but not proof of, physiological cardiac remodeling. Because the present design did not include biochemical, histological, electrocardiographic, or direct oxygen-consumption confirmation, and because the structural changes occurred in controls as well as treated animals, these observations should be regarded as hypothesis-generating rather than mechanistically conclusive. They identify ranolazine as a candidate worthy of further, more deeply mechanistic investigation rather than establishing, on their own, a basis for anti-doping action.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12872-026-06339-x.

Introduction

The pursuit of enhanced physical performance in competitive sport has long driven the exploration of pharmacological agents capable of optimizing cardiac energy metabolism and cardiovascular efficiency. Substances deliberately used to elevate athletic performance beyond an individual's physiological baseline are conventionally termed ergogenic aids, a designation derived from the Greek ergon (work) and gennan (to produce); the term encompasses nutritional, mechanical, and pharmacological interventions and is central to the regulatory framework applied by anti-doping authorities. Anti-anginal drugs, traditionally prescribed to alleviate symptoms in coronary artery disease (CAD) by improving myocardial efficiency, have garnered particular attention for their off-label ergogenic potential in healthy individuals [5, 20]. CAD, characterized by atherosclerotic plaque accumulation leading to myocardial ischemia, manifests as stable angina pectoris in many patients who do not achieve full symptomatic relief despite advances in percutaneous revascularization [1, 2]. In this clinical context, anti-anginal therapies such as ranolazine and trimetazidine are used to suppress angina and improve quality of life [3]. However, because their principal mechanisms of action involve the modulation of myocardial energy substrate utilization rather than coronary vasodilation per se, their pharmacological effects are not necessarily confined to ischemic myocardium and could, in principle, extend to the non-diseased heart.

Trimetazidine, a metabolic modulator, inhibits long-chain 3-ketoacyl-CoA thiolase, shifting adenosine triphosphate (ATP) production away from fatty acid oxidation toward the comparatively more oxygen-efficient pathway of glucose oxidation [4]. This agent was added to the World Anti-Doping Agency (WADA) Prohibited List in 2014 following evidence of performance enhancement, including prolonged exercise duration in rodent models when combined with structured training [5]. Subsequent detection of trimetazidine in samples from Olympic-level athletes has underscored the practical relevance of this risk [6]. Ranolazine, a piperazine derivative recommended by the European Society of Cardiology for the treatment of chronic stable angina [7], acts principally as a selective inhibitor of the late sodium current (INaL) in cardiomyocytes [8]. By reducing intracellular sodium accumulation and the consequent secondary rise in intracellular calcium, ranolazine is understood to mitigate diastolic wall tension, to favor coronary microvascular perfusion, and to improve myocardial relaxation without materially altering heart rate or arterial blood pressure [9, 10]. Landmark clinical trials, MARISA, CARISA, and ERICA, have established its anti-ischemic efficacy in patients with CAD, demonstrating improved exercise tolerance and reduced angina frequency [11–13]. More recent investigations in models of pulmonary hypertension and in patients with non-ST-elevation acute coronary syndromes have suggested additional cardioprotective properties, including suppression of atrial arrhythmia and improvement of right ventricular performance [14–16]; it should be emphasized, however, that ranolazine's regulatory labeling and the bulk of its supporting arrhythmia literature pertain specifically to supraventricular rather than ventricular arrhythmic substrates, reflecting differential expression of the underlying ion channels between the atrial and ventricular myocardium [8, 9].

Despite this substantial body of clinical evidence in diseased hearts, the behavior of ranolazine in healthy, non-ischemic myocardium subjected to physical training remains largely unexplored. Exercise physiology indicates that sustained aerobic training induces a coordinated set of cardiovascular adaptations, including increased cardiac output and maximal oxygen utilization (VO2max), to meet the heightened metabolic demand of working skeletal muscle [17, 18]. During strenuous exercise, skeletal muscle blood flow may increase from a resting baseline of approximately 2–4 mL/min/100 g to as much as 100 mL/min/100 g, a redistribution facilitated by local vasodilation and sympathetic activation [19]. Pharmacological agents capable of augmenting these processes, or of allowing an athlete to sustain a higher relative workload for a given perceived effort, could in principle confer an unfair competitive advantage, a concern that aligns with WADA's general criteria for prohibition [20]. Given the precedent established by trimetazidine, a systematic, hypothesis-generating evaluation of ranolazine's effects on exercise capacity and cardiac structure in a healthy animal model represents a reasonable and timely first step toward determining whether its ergogenic potential merits more detailed mechanistic and, eventually, clinical scrutiny.

Accordingly, this study aimed to characterize, in healthy rats, the functional exercise and structural echocardiographic correlates of chronic ranolazine administration, using trimetazidine as an active comparator with established ergogenic properties and saline as a negative control. By combining forced swimming as an index of aerobic exercise capacity, rotarod testing as an index of motor coordination, and serial echocardiography as an index of cardiac structure and function, we sought to describe whether ranolazine produces performance-related adaptations that parallel those of a known performance-enhancing agent. We wish to state explicitly, in light of the substantial methodological limitations detailed later in this manuscript, that the present investigation was conceived and should be read as an exploratory, hypothesis-generating pilot characterization rather than as a mechanistic study; it was not designed, and is not adequate, to establish causality between any specific metabolic mechanism and the functional outcomes observed. This research was conducted and is reported in compliance with the ARRIVE 2.0 guidelines for the reporting of animal research, in the interest of transparency regarding animal welfare, experimental design, and outcome reporting, and to facilitate independent reproducibility and ethical scrutiny [21].

Methods

Study design and ethical considerations

This randomized, controlled, parallel-group, exploratory study was conducted at the Health Sciences University Hamidiye School of Medicine Physiology Laboratory in Istanbul, Türkiye, between January and March 2024. All procedures adhered to the ARRIVE guidelines (version 2.0) for the reporting of animal research, with particular attention to the principles of replacement, reduction, and refinement [21]. The protocol was approved by the Institutional Animal Care and Use Committee (IACUC) of Health Sciences University (approval date: 18 March 2023; approval number: 03/04), in compliance with Turkish national regulations governing animal experimentation. Sample size was determined a priori by power analysis (G*Power software, version 3.1; Heinrich Heine University, Düsseldorf, Germany), based on previously published rat exercise data, targeting 80% power to detect a 20% between-group difference in swimming duration (effect size d = 1.2, α = 0.05), which yielded a requirement of n = 6 per group [5]. No dedicated pilot study preceded the present investigation; expectations regarding effect size were instead informed by the cited historical data, a methodological choice that we acknowledge in the Limitations section. Animals were acclimatized for seven days prior to the start of the experimental protocol. At the conclusion of the 30-day experimental protocol, all animals were euthanized under deep anaesthesia to ensure unconsciousness prior to the application of the physical method. Anaesthesia was induced by intraperitoneal administration of an overdose of ketamine hydrochloride (150 mg/kg) combined with xylazine hydrochloride (15 mg/kg), agents identical to those used for echocardiographic imaging sessions but administered at supra-anaesthetic doses. Adequate depth of anaesthesia and confirmed unconsciousness were verified by the absence of the pedal withdrawal reflex, loss of righting reflex, and unresponsiveness to noxious stimulation applied to the plantar surface of the hindpaw. Cervical dislocation was then performed as the definitive method of euthanasia by a trained investigator (C.D., DVM) using a single, rapid manual technique. Death was confirmed by cessation of cardiac activity and permanent cessation of respiration. This two-stage approach, chemical induction of unconsciousness followed by physical euthanasia, was selected in accordance with the AVMA Guidelines for the Euthanasia of Animals (2020 edition), which permits cervical dislocation in rodents provided that prior sedation or anaesthesia is employed to ensure the animal is unconscious at the time the physical method is applied, and was consistent with the protocol approved by the Institutional Animal Care and Use Committee of Health Sciences University (approval number: 03/04). The overall experimental design is depicted in Fig. 1. We emphasize at the outset of the Methods that this investigation was designed as a hypothesis-generating, functional and structural characterization study; it was not powered, nor was it designed, to interrogate the biochemical or molecular mechanisms underlying any observed between-group differences, a point elaborated upon in the Discussion and Limitations.

Fig. 1.

Overall experimental design, depicting the randomization of twenty-four rats to control, trimetazidine, and ranolazine groups (two control-group rats subsequently died under anesthesia during baseline echocardiography and were excluded before further data collection, yielding a final analytic sample of 22 rats), the timeline of daily drug administration and exercise testing across the 30-day protocol, and the time points (days 0, 15, and 30) at which echocardiography and rotarod testing were performed

Animals and housing

Twenty-four adult male Sprague–Dawley rats (Rattus norvegicus), aged 8–10 weeks and weighing 250–300 g (mean 275 ± 15 g) at enrollment, were obtained from a certified commercial breeder (Kobay Experimental Animals Laboratory, Ankara, Türkiye). This age range corresponds to young, sexually mature adult rats and is widely used in rodent exercise and cardiovascular pharmacology studies because it balances developmental maturity against the practical constraints of study duration and animal supply; we nonetheless acknowledge, and discuss further in the Limitations section, that rats in this age window have not yet reached full somatic maturity and continue to gain body and organ mass independently of any experimental intervention, a consideration directly relevant to the interpretation of the echocardiographic findings reported below. Only male rats were used in order to avoid the confounding influence of estrous-cycle-related hormonal variability on exercise performance; we recognize that this design choice precludes any assessment of sex-based differences in the responses to ranolazine or trimetazidine and discuss this explicitly as a limitation. Rats were housed in groups of four in polycarbonate cages (40 × 30 × 20 cm) under controlled environmental conditions (temperature 22 ± 2 °C; relative humidity 50–60%; 12:12-h light:dark cycle, lights on at 07:00), with ad libitum access to standard rodent chow (15% protein, 60% carbohydrate, 25% fat; energy density 3.0 kcal/g; Kobay Experimental Animals Laboratory, Ankara, Türkiye) and tap water. Aspen-shaving bedding was changed twice weekly, and environmental enrichment consisted of polyvinyl chloride tunnels and nesting material to reduce stress. Daily visual health monitoring assessed for signs of distress, including weight loss exceeding 10% of baseline and lethargy, with veterinary oversight provided throughout. Two rats allocated to the control group died under ketamine-xylazine anesthesia during the baseline (day 0) echocardiographic examination, before any body-weight, rotarod, or swimming data had been collected; death was attributed to anesthesia-related cardiorespiratory depression rather than to any study procedure other than the anesthesia itself. Both animals were excluded from all subsequent data collection and analysis, and no further adverse events, unplanned deaths, or exclusions occurred for the remainder of the study. The final analytic sample therefore comprised 22 rats: control, n = 6; trimetazidine, n = 8; ranolazine, n = 8.

Randomization and blinding

Rats were allocated to three groups of eight using a computer-generated random sequence (random.org; Randomness and Integrity Services Ltd., Dublin, Ireland): control (saline vehicle), ranolazine (50 mg/kg/day), and trimetazidine (10 mg/kg/day). Allocation was concealed in sealed, opaque envelopes until the moment of dosing. Investigators who performed behavioral testing and echocardiographic acquisition and analysis were blinded to group assignment through the use of coded animal identifiers; drug preparation and administration were performed by a separate, unblinded technician not involved in outcome assessment. Blinding was maintained until the completion of all statistical analyses.

Drug administration

Ranolazine (Latixa; Menarini) and trimetazidine (Vastarel; Servier) were dissolved in 0.9% saline and administered by oral gavage using a 16-gauge curved feeding needle (administered volume: 1 mL/100 g body weight). The selected doses, 50 mg/kg/day for ranolazine and 10 mg/kg/day for trimetazidine, were based on prior rodent pharmacology studies demonstrating anti-anginal and metabolic efficacy at these dose levels without overt toxicity [5, 22, 23]. Specifically, the ranolazine dose follows the experimental range validated by Wang and colleagues across multiple rodent models of angina [22], in which doses in this order of magnitude were required to achieve measurable anti-ischemic and electrophysiological effects given the comparatively rapid hepatic clearance of ranolazine in rodents relative to humans; the corresponding human therapeutic plasma concentration range was used as a secondary benchmark to confirm pharmacological plausibility. The trimetazidine dose of 10 mg/kg/day corresponds to that used in the rodent exercise-capacity study by Xie and colleagues, in which trimetazidine combined with exercise training produced measurable improvements in exhaustive swimming time and mitochondrial quality control parameters in mice [5], and is consistent with anesthesia- and metabolism-related rodent dosing considerations described by Giroux and colleagues [23]. We acknowledge that neither drug was subjected to a dose-ranging or dose–response evaluation within the present study, and that the single dose levels selected, while pharmacologically justified by precedent, do not allow conclusions regarding dose-dependency of the observed effects; this constitutes an explicit limitation discussed later in this manuscript. Administration occurred daily at 09:00 for 30 consecutive days, one hour prior to each scheduled exercise session, timed to approximate the pharmacokinetic peaks of each agent, ranolazine reaches peak plasma concentration (Tmax) at approximately 4–5 h after oral modified-release dosing in humans [24, 25], while trimetazidine reaches Tmax at approximately 2–4 h after oral dosing [4, 25]; because rodent absorption and elimination kinetics differ quantitatively from those in humans, this timing should be understood as a pragmatic approximation rather than a pharmacokinetically validated peak-exposure window in the rat, a further limitation we acknowledge explicitly. Control animals received an equivalent volume of saline by the same route and schedule. Body weights were recorded weekly and gavage volumes adjusted accordingly to maintain consistent mg/kg dosing throughout the study.

Exercise protocols

Forced swimming test

Aerobic exercise capacity was assessed using a progressive forced swimming paradigm conducted in a cylindrical tank (50 cm diameter, 80 cm height) filled to a depth of 60 cm with water maintained at 30 ± 1 °C [26, 27]. An initial adaptation phase (days 1–3) consisted of unloaded five-minute swims. From day 4 onward, a tail load equivalent to 8% of body weight was applied using lead weights to standardize the exhaustion endpoint [28]. Twenty timed sessions were conducted on weekdays from day 4 to day 30, with each session continuing until exhaustion, operationally defined as submersion lasting longer than three seconds without resurfacing [29]. Swimming duration was recorded independently by two blinded observers using a stopwatch (inter-rater reliability: intraclass correlation coefficient [ICC] = 0.98). Following each session, animals were dried and returned to their home cages, and tank water was replaced daily to minimize contamination and maintain thermal consistency. We note, and discuss at length in the Discussion, that the forced swimming test, although a long-established and validated index of aerobic endurance in rodents [26, 28], is influenced by factors beyond aerobic capacity alone, including acute stress reactivity, motivational state, and thermoregulatory demand, and that these confounders cannot be fully disentangled from genuine metabolic or cardiovascular adaptation using this paradigm in isolation.

Rotarod test

Motor coordination and balance were assessed using an accelerating rotarod apparatus (7-cm diameter rod, accelerating from 4 to 40 rpm over 300 s; Ugo Basile S.r.l., Gemonio, Italy) [30]. Animals underwent a three-day acclimatization period (three trials per day) to establish stable baseline performance prior to the start of drug administration. Formal testing was performed on days 0, 15, and 30, with three trials per session separated by a 60-s inter-trial rest interval; latency to fall was averaged across the three trials for each session. The test endpoint was defined as either falling from the rod or passive rotation (clinging to the rod without active walking).

Echocardiographic assessment

Transthoracic echocardiography was performed on days 0, 15, and 30 using a Vivid S6 ultrasound system equipped with a 12-MHz linear-array probe (GE Healthcare, Chicago, Illinois, USA) [31]. Anaesthesia for echocardiographic imaging was induced by intraperitoneal administration of ketamine hydrochloride (80 mg/kg) combined with xylazine hydrochloride (10 mg/kg). The combined intraperitoneal injection volume was calculated individually per animal based on body weight recorded on the morning of each imaging session to ensure consistent mg/kg dosing. Depth of anaesthesia was confirmed prior to commencing image acquisition by absence of the pedal withdrawal reflex; echocardiographic acquisition was initiated only after satisfactory anaesthetic depth was verified. This ketamine–xylazine combination is a well-established and validated regimen for transthoracic echocardiography in rats, producing stable, moderate-depth surgical anaesthesia with acceptable cardiovascular depression at these dosage levels, and its use in this context is consistent with previously published normal echocardiographic reference values for adult male Sprague–Dawley rats [32]. Atropine sulphate (0.05 mg/kg intraperitoneally) was available for administration if clinically significant bradycardia (heart rate below 150 beats per minute) was observed during imaging, in accordance with institutional veterinary guidance, though administration was not required in any animal during the study. Body temperature was maintained throughout each imaging session using a thermostatically controlled heated platform (37 ± 0.5 °C), and respiratory rate was monitored visually by a dedicated observer throughout the procedure. Animals recovered from anaesthesia in a warm environment under direct observation and were returned to their home cages only after full righting reflex had returned. Animals were positioned in dorsal recumbency on a heated imaging platform to maintain normothermia during image acquisition. Measurements followed European Society of Cardiology position paper for rodent echocardiography [33]: interventricular septal thickness in diastole and systole (IVSD/IVSS), left ventricular internal dimensions (LVIDd/LVIDs), posterior wall thickness (LVPWd/LVPWs), aortic dimensions (AoD/AoS), valvular peak velocities (AV Vmax, PV Vmax), ejection fraction (EF), fractional shortening (FS), end-diastolic and end-systolic volumes (EDV/ESV), and left ventricular (LV) mass. LV mass was not measured directly by gravimetric or histological means at any point in this protocol; rather, it was estimated non-invasively from two-dimensional and M-mode echocardiographic linear measurements using the cube-function Devereux formula (LV mass = 1.04 × [(IVSD + LVIDd + LVPWd)3 − LVIDd3] − 13.6) [34]. We make this distinction explicit because it bears directly on the interpretation of the structural findings reported in the Results: an echocardiographically estimated increase in calculated LV mass is not equivalent to, and should not be conflated with, a directly measured or histologically confirmed increase in ventricular myocardial mass, and the absence of terminal heart-weight or histopathological corroboration is discussed as an explicit limitation. Three consecutive cardiac cycles were averaged for each parameter by a single blinded cardiologist, with intra-observer variability below 5%. Heart rate was recorded from the echocardiographic Doppler trace at the time of each imaging session; because this represents a brief, intermittent measurement obtained under ketamine-xylazine anesthesia rather than continuous ambulatory electrocardiographic or Holter monitoring, it is not equivalent to formal arrhythmia surveillance.

Biological sample handling

The present protocol was designed exclusively around in vivo functional (forced swimming, rotarod) and structural (echocardiographic) endpoints. No additional tissue, blood, or plasma samples were collected or banked for subsequent biochemical, metabolomic, or histological analysis at the time of euthanasia. We recognize that this represents a substantive limitation, particularly with respect to the inability to corroborate the proposed metabolic mechanisms of ranolazine and trimetazidine with direct biochemical evidence (e.g., lactate, ATP, or substrate-utilization assays) or to validate the echocardiographic structural findings against heart weight or histopathology; we address this point explicitly in the Limitations section and consider tissue banking and gravimetric/histological heart assessment to be a priority addition for any confirmatory follow-up study.

Statistical analysis

Data distribution was assessed using the Shapiro–Wilk test; because the majority of variables departed from normality, non-parametric methods were applied throughout: the Kruskal–Wallis test for between-group comparisons at each time point, the Friedman test for within-group comparisons across the three repeated time points, and the Wilcoxon signed-rank test for pairwise post-hoc comparisons. For each repeated-measures factor with three time points, three pairwise post-hoc comparisons are possible (day 0 vs. day 15, day 15 vs. day 30, and day 0 vs. day 30); a Bonferroni-adjusted significance threshold of p < 0.017 (0.05/3) was applied specifically to these three pairwise comparisons to control the family-wise error rate within each repeated-measures factor. This Bonferroni-adjusted threshold is a feature of the post-hoc pairwise testing procedure alone and does not alter the overall significance threshold of α = 0.05 used for the primary Kruskal–Wallis and Friedman omnibus tests; to avoid the ambiguity, exact p-values are reported throughout the Results section for every comparison. Continuous variables are reported as medians with interquartile ranges (IQR) in parentheses, consistent with the non-parametric analytical approach. All analyses were performed using SPSS Statistics, version 27.0 (IBM Corp., Armonk, New York, USA). For the forced swimming test, total swimming time and average per-session duration across all 21 sessions were pre-specified as the primary summary endpoints; the 21 individual within-session between-group comparisons are reported as exploratory, descriptive analyses given the number of repeated tests across sessions and are not additionally adjusted for multiplicity beyond the family-wise correction described above for post-hoc pairwise testing, and should be interpreted with corresponding caution. For these two primary summary endpoints specifically, pairwise between-group differences (ranolazine vs. control, trimetazidine vs. control, ranolazine vs. trimetazidine) were further characterized using the Hodges-Lehmann estimator of the median difference, together with its exact distribution-free 95% confidence interval, and Cliff's delta (numerically equivalent to the rank-biserial correlation) as a rank-based effect-size measure, interpreted using the conventional thresholds of negligible (< 0.147), small (< 0.33), medium (< 0.474), and large (≥ 0.474) effect. Beyond the two control-group deaths under baseline anesthesia described above (see Animals and Housing), no further animals or data points were excluded from analysis, and all 22 remaining rats completed the full 30-day protocol.

Results

Two of the twenty-four randomized rats, both allocated to the control group, died under anesthesia during baseline (day 0) echocardiography, before any study data were collected, and were excluded from all analyses; the remaining 22 rats (control, n = 6; trimetazidine, n = 8; ranolazine, n = 8) completed the full 30-day protocol without further mortality or exclusion. Baseline characteristics, including body weight, echocardiographic parameters, and rotarod latency, were statistically indistinguishable between groups (p > 0.05 for all comparisons), confirming adequate randomization.

Body weight changes

Body weight increased over the course of the study in all three groups, with no significant between-group differences at any individual time point (Kruskal–Wallis: day 0, p = 0.85; day 15, p = 0.72; day 30, p = 0.68). Within-group Friedman testing detected significant change over time in all groups (control, p = 0.012; trimetazidine, p = 0.008; ranolazine, p = 0.010); however, post-hoc Wilcoxon comparisons did not reach the Bonferroni-adjusted pairwise threshold of p < 0.017 in any group (e.g., control day 0 vs. day 15, p = 0.021). Median weights are summarized in Table 1. The pattern, an initial plateau or modest decline between day 0 and day 15, followed by a clear increase between day 15 and day 30, was observed in all three groups, including controls, and most plausibly reflects a combination of exercise-related catabolic stress in the early phase of training superimposed on ongoing somatic growth, given that the rats were 8–10 weeks of age at enrollment and therefore not yet fully grown. Because this pattern did not differ by treatment group, it cannot be attributed specifically to either ranolazine or trimetazidine.

Table 1.

Body Weight Across Groups at Days 0, 15, and 30 (median [IQR], grams). Control, n = 6; trimetazidine, n = 8; ranolazine, n = 8

Group	Day 0 (g)	Day 15 (g)	Day 30 (g)	p* (Friedman)
Control	320 (310–330)	310 (300–330)	335 (320–350)	0.012
Trimetazidine	325 (300–340)	323 (300–330)	335 (310–345)	0.008
Ranolazine	320 (315–325)	315 (300–320)	328 (310–345)	0.010
p† (Kruskal–Wallis)	0.85	0.72	0.68	—

^*Friedman test (within-group); †Kruskal–Wallis test (between-group)

Forced swimming test performance

Total swimming time and average per-session duration across all 21 sessions were the pre-specified primary summary endpoints for the forced swimming test; the session-by-session comparisons that follow are reported as exploratory, descriptive analyses given the number of repeated tests across sessions and should be interpreted with corresponding caution. Swimming duration increased progressively across sessions in all three groups, consistent with a training effect common to all animals irrespective of treatment. Statistically significant between-group differences emerged from session 3 onward, with exact Kruskal–Wallis p-values of 0.048 (session 3), 0.043 (session 5), 0.046 (session 9), 0.032 (session 11), 0.023 (session 12), 0.032 (session 13), 0.002 (session 15), < 0.001 (session 16), < 0.001 (session 17), < 0.001 (session 18), 0.002 (session 19), and 0.002 (session 20); all other individual sessions did not reach significance (Table 2). Post-hoc analysis indicated that both the ranolazine and trimetazidine groups outperformed controls across these sessions. For example, at session 15, median swimming duration was 1743 s (1672–1797) in the control group, 1987 s (1910–2463) in the trimetazidine group, and 2017 s (1842–2293) in the ranolazine group (p = 0.002).

Table 2.

Forced Swimming Test: Swimming Duration by Session and Group (median [IQR], seconds). Control, n = 6; trimetazidine, n = 8; ranolazine, n = 8

Session	Control	Trimetazidine	Ranolazine	p (Kruskal–Wallis)
1	178 (165–214)	177 (144–224)	175 (144–237)	0.907
2	316 (198–510)	330 (220–590)	436 (304–648)	0.182
3	1156 (966–1245)	1110 (745–1365)	1332 (565–1370)	0.048
4	1089 (814–1230)	1162 (880–1370)	1267 (780–1613)	0.107
5	1069 (765–1540)	1810 (975–2188)	1608 (1260–2205)	0.043
6	1315 (930–1510)	1603 (950–1995)	1780 (830–2160)	0.078
7	1863 (1257–2100)	1840 (1463–2250)	1785 (1020–2350)	0.824
8	1678 (1380–1995)	2103 (1786–2460)	2141 (1242–2530)	0.066
9	1677 (1384–1945)	2103 (1782–2460)	2248 (1500–3000)	0.046
10	1833 (1734–2046)	1997 (1688–2310)	2127 (1610–2787)	0.204
11	1946 (1630–2163)	1912 (1790–2280)	2140 (1967–2364)	0.032
12	1819 (1740–1990)	2010 (1845–2225)	2117 (1801–2407)	0.023
13	1763 (1723–1911)	2012 (1473–2323)	2235 (1811–2449)	0.032
14	1876 (1750–2035)	2117 (1743–2253)	2077 (1718–2509)	0.159
15	1743 (1672–1797)	1987 (1910–2463)	2017 (1842–2293)	0.002
16	1711 (1590–1817)	2121 (1795–2463)	2447 (2350–2698)	< 0.001
17	1795 (1695–1868)	2040 (1845–2293)	2465 (2382–2525)	< 0.001
18	1841 (1798–1889)	2010 (1729–2274)	2515 (2456–2652)	< 0.001
19	1909 (1873–1935)	2035 (1887–2237)	2156 (2113–2469)	0.002
20	1654 (1576–1718)	1947 (1867–2132)	1919 (1879–2187)	0.002
21	2030 (2003–2121)	2240 (1894–2413)	2181 (1984–2449)	0.241
**Total**	**32,426 (29,389–34,453)**	**35,694 (30,776–39516)**	**39,668 (33,794–43,838)**	**0.006**
**Average**	**1548 (1399–1641)**	**1700 (1466–1881)**	**1889 (1609–2087)**	**0.006**

Bold rows denote total and average swimming time across all 21 sessions and constitute the pre-specified primary summary endpoints; rows 1–21 are exploratory, descriptive session-level comparisons. Exact Kruskal–Wallis p-values are reported for every session

Pairwise Hodges-Lehmann shift (95% CI) and Cliff's δ for the primary summary endpoint (total/average swimming time): ranolazine vs. control, 7360 s (2253–10,181), δ = 0.92 (large); trimetazidine vs. control, 3716 s (1387–6821), δ = 0.79 (large); ranolazine vs. trimetazidine, 2884 s (−1208–6568), δ = 0.41 (medium, not statistically significant at the Bonferroni-adjusted threshold of p < 0.0167)

The temporal pattern of the between-group divergence is notable: ranolazine and trimetazidine separated from control as early as session 3 (day 4 of dosing, p = 0.048), well before any echocardiographically detectable structural change was observed (the earliest significant structural change occurred at the day 15 assessment). Ranolazine produced a more pronounced effect than trimetazidine specifically in sessions 16, 17, and 18 (each p < 0.001), corresponding to days 22–26 of the protocol. Total swimming time across all 21 sessions was 39,668 s (33,794–43,838) for ranolazine, 35,694 s (30,776–39,516) for trimetazidine, and 32,426 s (29,389–34,453) for control (p = 0.006); average per-session duration followed the same ordering: ranolazine 1889 s (1609–2087), trimetazidine 1700 s (1466–1881), control 1548 s (1399–1641) (p = 0.006). Because these two primary summary endpoints are monotonic transformations of one another (average duration = total duration/21), pairwise post-hoc comparisons are numerically identical for both and are reported once: for total (and average) swimming time, ranolazine exceeded control with a large effect (Hodges-Lehmann shift 7360 s, or 350.5 s per session; 95% CI 2253–10,181 s; exact Mann–Whitney p = 0.003; Cliff's δ = 0.92), trimetazidine exceeded control with a large effect (shift 3716 s, or 176.9 s per session; 95% CI 1387–6821 s; p = 0.013; Cliff's δ = 0.79), and ranolazine did not differ significantly from trimetazidine (shift 2884 s, or 137.3 s per session; 95% CI −1208–6568 s; p = 0.195; Cliff's δ = 0.41, a medium effect), against a Bonferroni-adjusted threshold of p < 0.0167 for these three pairwise comparisons. These effect-size estimates are summarized in Table 2. These data are summarized in Table 2 and displayed graphically in Fig. 2.

Fig. 2.

Graphical representation of forced-swimming duration across all 21 sessions for the control, trimetazidine, and ranolazine groups, derived from the data presented in Table 2

Rotarod test performance

No significant between-group differences in fall latency were observed at any time point (Kruskal–Wallis: day 0, p = 0.804; day 15, p = 0.748; day 30, p = 0.576). Within-group change over time was likewise non-significant in all groups (Friedman: control, p = 0.846; trimetazidine, p = 0.223; ranolazine, p = 0.072). Rotarod durations are summarized in Table 3. Taken together with the absence of any significant between-group or within-group effect at the conventional or Bonferroni-adjusted threshold, these data indicate that neither drug measurably altered motor coordination or balance over the 30-day protocol.

Table 3.

Rotarod Test Duration by Group at Days 0, 15, and 30 (median [IQR], seconds). Control, n = 6; trimetazidine, n = 8; ranolazine, n = 8

Group	Day 0 (s)	Day 15 (s)	Day 30 (s)	p (Friedman)
Control	69 (25–92)	85 (41–90)	75 (34–87)	0.846
Trimetazidine	48 (19–102)	83 (43–112)	82 (51–139)	0.223
Ranolazine	69 (11–97)	64 (36–134)	85 (46–117)	0.072
p (Kruskal–Wallis)	0.804	0.748	0.576	—

Echocardiographic parameters

Baseline echocardiographic parameters did not differ between groups (p > 0.05 for all comparisons). Within-group (Friedman) analysis identified statistically significant change over the 30-day protocol in interventricular septal thickness (IVSD, IVSS) and calculated LV mass in every group, including controls: control (IVSD p = 0.008, IVSS p = 0.004, LV mass p = 0.008), trimetazidine (IVSD p = 0.009, IVSS p < 0.001, LV mass p = 0.007), and ranolazine (IVSD p = 0.007, IVSS p < 0.001, LV mass p = 0.010). The ranolazine group additionally showed significant change in LVIDS (p = 0.030), AoS (p = 0.048), AV Vmax (p = 0.030), EF (p = 0.008), FS (p = 0.008), and ESV (p = 0.030). Post-hoc Wilcoxon testing (Bonferroni-adjusted threshold p < 0.017) confirmed significant pairwise increases in IVSS for both the trimetazidine group (day 0 vs. day 15, p = 0.007; day 0 vs. day 30, p = 0.001) and the ranolazine group (day 0 vs. day 15, p = 0.005; day 0 vs. day 30, p = 0.010), and a significant pairwise increase in IVSD for the ranolazine group (day 0 vs. day 30, p = 0.015). In the control group, the omnibus Friedman tests for IVSD, IVSS, and LV mass were significant, but no individual pairwise post-hoc comparison reached the Bonferroni-adjusted threshold of p < 0.017; we describe this pattern as a directionally consistent but not individually significant increase.

Across all three groups, ejection fraction and fractional shortening either remained stable or increased numerically (most notably in the ranolazine group), with no instance of decline, arguing against pathological systolic dysfunction accompanying the observed structural changes. Heart rate, recorded from the echocardiographic Doppler trace at each time point, did not change significantly within or between any group (Friedman p = 0.115 control, p = 0.135 trimetazidine, p = 0.417 ranolazine), providing reassurance against a gross chronotropic effect of either drug, although this intermittent, anesthetized measurement does not substitute for continuous electrocardiographic or Holter-based arrhythmia surveillance. Complete echocardiographic data for the control (n = 6), trimetazidine (n = 8), and ranolazine (n = 8) groups are presented in Tables 4, 5, and 6, respectively; representative M-mode and Doppler images are shown in Fig. 3, and the trajectories of IVSD, IVSS, and calculated LV mass across groups and time points are displayed graphically in Fig. 4.

Table 4.

Echocardiographic Parameters, Control Group (median [IQR]). n = 6

Parameter	Day 0	Day 15	Day 30	p (Friedman)
Heart Rate (bpm)	252 (245–297)	274 (245–287)	258 (232–281)	0.115
IVSD (mm)	1.7 (1.7–2.0)	1.9 (1.8–2.0)	2.0 (1.8–2.1)	0.008
IVSS (mm)	2.6 (2.3–2.7)	2.7 (2.7–2.9)	2.9 (2.8–3.0)	0.004
LVIDD (mm)	7.0 (6.2–7.5)	6.8 (6.2–7.5)	7.2 (6.5–7.6)	0.032
LVIDS (mm)	4.6 (4.3–5.6)	4.5 (4.1–5.6)	5.0 (4.1–5.7)	0.028
LVPWD (mm)	2.0 (1.8–2.2)	1.85 (1.7–2.0)	2.0 (1.7–2.2)	0.449
LVPWS (mm)	2.6 (2.5–2.6)	2.6 (2.5–2.8)	2.65 (2.5–2.7)	0.305
AoS (mm)	3.6 (3.5–3.7)	3.7 (3.4–3.8)	3.65 (3.6–3.9)	0.179
AoD (mm)	3.2 (3.1–3.3)	3.2 (3.1–3.3)	3.15 (3.1–3.3)	1.000
Aortic Stiffness (%)	12.5 (9.0–16.1)	12.3 (9.6–22.5)	16.1 (12.5–14.8)	0.309
AV Vmax (m/s)	1.22 (1.09–1.39)	1.27 (1.15–1.41)	1.26 (1.06–1.83)	0.846
PV Vmax (m/s)	0.70 (0.67–0.72)	0.84 (0.74–1.05)	0.79 (0.68–0.88)	0.032
EF (%)	67 (58–77)	71 (51–78)	67 (54–81)	0.846
FS (%)	31 (25.3–39.4)	33.6 (21.1–39.4)	31 (23–43)	0.846
Aortic Decel. Time (ms)	70 (55–88)	52 (46–74)	61 (50–66)	0.048
Pulm. Decel. Time (ms)	54 (44–60)	42 (37–61)	48 (44–63)	0.094
Pulm. Accel. Time (ms)	37 (33–43)	34 (24–51)	36 (27–45)	0.580
EDV (mL)	0.36 (0.25–0.44)	0.34 (0.25–0.44)	0.40 (0.29–0.46)	0.032
ESV (mL)	0.10 (0.08–0.18)	0.09 (0.07–0.18)	0.13 (0.07–0.19)	0.028
Cardiac Output (mL/min)	65 (49–73)	54.4 (46.7–82.6)	71.3 (54.9–77.8)	0.311
LV Mass (g)	1.89 (1.82–1.92)	1.93 (1.89–1.99)	1.97 (1.94–1.99)	0.008

Table 5.

Echocardiographic Parameters, Trimetazidine Group (median [IQR]). n = 8

Parameter	Day 0	Day 15	Day 30	p (Friedman)
Heart Rate (bpm)	252 (216–279)	252 (182–315)	260 (245–301)	0.135
IVSD (mm)	1.8 (1.7–2.0)	1.95 (1.8–2.0)	1.95 (1.8–2.1)	0.009
IVSS (mm)	2.7 (2.6–2.9)	2.8 (2.7–3.0)	2.8 (2.7–3.1)	< 0.001
LVIDD (mm)	6.8 (6.4–7.5)	7.1 (6.4–8.0)	7.25 (6.3–8.4)	0.250
LVIDS (mm)	4.75 (4.5–5.2)	4.7 (4.4–5.4)	5.15 (4.1–5.5)	0.639
LVPWD (mm)	2.1 (1.7–2.2)	1.95 (1.8–2.3)	2.0 (1.8–2.2)	0.887
LVPWS (mm)	2.6 (2.3–2.7)	2.6 (2.4–2.8)	2.6 (2.3–2.8)	0.549
AoS (mm)	3.6 (3.4–3.9)	3.55 (3.4–4.0)	3.6 (3.5–3.8)	0.852
AoD (mm)	3.15 (3.0–3.4)	3.15 (2.9–3.5)	3.2 (3.0–3.3)	0.898
Aortic Stiffness (%)	13.1 (9.6–18.7)	13.8 (6.1–20.6)	13.6 (9.1–23.3)	0.968
AV Vmax (m/s)	1.32 (1.06–1.87)	1.17 (1.0–1.38)	1.46 (1.04–1.82)	0.140
PV Vmax (m/s)	0.71 (0.54–0.79)	0.68 (0.52–0.81)	0.60 (0.56–0.80)	0.905
EF (%)	66 (58–74)	68 (58–75)	70 (53–75)	0.607
FS (%)	30.2 (25.0–36.5)	32.0 (25.0–37.5)	33.2 (22.5–37.0)	0.607
Aortic Decel. Time (ms)	62.5 (48–70)	63.5 (44–83)	50.0 (39–61)	0.122
Pulm. Decel. Time (ms)	53.5 (37–74)	48.0 (37–59)	50.5 (39–63)	0.508
Pulm. Accel. Time (ms)	40.5 (25–48)	32.0 (25–48)	35.0 (26–42)	0.687
EDV (mL)	0.33 (0.27–0.44)	0.37 (0.27–0.54)	0.40 (0.26–0.62)	0.250
ESV (mL)	0.11 (0.09–0.14)	0.11 (0.09–0.17)	0.14 (0.08–0.17)	0.639
Cardiac Output (mL/min)	55.5 (44.4–78.6)	62.8 (46.0–90.0)	65.0 (56.5–116.5)	0.197
LV Mass (g)	1.91 (1.82–1.97)	1.94 (1.90–2.02)	1.97 (1.87–2.02)	0.007

Table 6.

Echocardiographic Parameters, Ranolazine Group (median [IQR]). n = 8

Parameter	Day 0	Day 15	Day 30	p (Friedman)
Heart Rate (bpm)	282 (244–318)	280 (201–322)	257 (232–302)	0.417
IVSD (mm)	1.9 (1.7–1.9)	1.95 (1.8–2.0)	1.95 (1.9–2.1)	0.007
IVSS (mm)	2.65 (2.5–2.8)	2.85 (2.7–3.0)	2.85 (2.8–3.0)	< 0.001
LVIDD (mm)	7.25 (6.5–7.3)	7.35 (6.3–7.7)	7.1 (6.1–8.0)	0.798
LVIDS (mm)	5.05 (4.3–5.2)	4.65 (3.9–5.0)	4.45 (3.8–5.9)	0.030
LVPWD (mm)	1.9 (1.9–2.0)	1.95 (1.8–2.2)	1.9 (1.8–2.1)	0.432
LVPWS (mm)	2.5 (2.5–2.6)	2.55 (2.2–3.0)	2.6 (2.3–2.8)	0.905
AoS (mm)	3.5 (3.4–3.7)	3.5 (3.3–3.7)	3.55 (3.4–3.9)	0.048
AoD (mm)	3.1 (3.0–3.4)	3.05 (2.9–3.2)	3.1 (2.7–3.2)	0.167
Aortic Stiffness (%)	12.7 (8.8–16.6)	13.3 (10.0–24.0)	15.8 (9.6–25.9)	0.227
AV Vmax (m/s)	1.30 (1.0–1.41)	1.32 (1.2–1.56)	1.58 (1.1–1.78)	0.030
PV Vmax (m/s)	0.66 (0.50–0.83)	0.71 (0.56–1.29)	0.75 (0.61–0.86)	0.417
EF (%)	66 (63–71)	74 (72–76)	75 (51–85)	0.008
FS (%)	30.1 (28.1–33.8)	35.8 (34.3–38.0)	37.1 (21.3–46.8)	0.008
Aortic Decel. Time (ms)	61 (48–77)	50 (48–66)	57 (42–68)	0.227
Pulm. Decel. Time (ms)	52 (44–63)	50 (36–70)	55 (41–63)	0.657
Pulm. Accel. Time (ms)	37 (25–44)	33 (33–51)	32 (20–42)	0.875
EDV (mL)	0.39 (0.28–0.41)	0.41 (0.26–0.48)	0.37 (0.23–0.54)	0.798
ESV (mL)	0.13 (0.08–0.14)	0.11 (0.06–0.13)	0.09 (0.05–0.21)	0.030
Cardiac Output (mL/min)	71 (57–87)	73 (53–114)	62 (48–132)	0.223
LV Mass (g)	1.89 (1.84–1.92)	1.96 (1.87–2.04)	1.97 (1.89–1.99)	0.010

Abbreviations: IVSD/IVSS interventricular septal thickness in diastole/systole, LVIDD/LVIDS left ventricular internal diastolic/systolic diameter, LVPWD/LVPWS left ventricular posterior wall thickness in diastole/systole, AoS/AoD aortic systolic/diastolic diameter, AV Vmax/PV Vmax Aortic/pulmonary valve peak transvalvular velocity, EF Ejection fraction, FS Fractional shortening, EDV/ESV End-diastolic/end-systolic volume, LV Left ventricular. p-values from the Friedman test (within-group, across the three time points)

Fig. 3.

Representative transthoracic echocardiographic images. A M-mode image used to derive interventricular septal thickness, left ventricular internal dimensions, and posterior wall thickness. B Pulsed-wave Doppler tracing used to derive aortic and pulmonary valve peak transvalvular velocities. Images shown are representative single examples rather than per-group composites; the acquisition of full per-group, four-chamber recordings is identified as a priority addition in the Limitations section

Fig. 4.

Interventricular septal thickness in diastole (IVSD) and systole (IVSS) and calculated left ventricular mass (median, error bars = interquartile range) across the control, trimetazidine, and ranolazine groups at days 0, 15, and 30, derived from the data presented in Tables 4, 5 and 6

Discussion

The findings reported above should therefore be read as a hypothesis-generating functional and structural characterization of chronic ranolazine and trimetazidine administration in healthy rats, rather than as confirmation of any specific biochemical pathway underlying the effects observed. With that framing established, this study demonstrates that both ranolazine and trimetazidine were associated with greater forced-swimming endurance than saline control, with ranolazine producing a numerically larger effect in the later sessions of the protocol, and that all three groups, including controls, showed small increases in septal thickness and calculated LV mass over the 30-day observation period. These findings extend prior descriptions of trimetazidine's performance-modifying properties in rodents [5] to ranolazine and are broadly consistent with the wider literature describing cardioprotective and right-ventricular benefits of ranolazine outside the setting of overt ischemia [14–16]; however, we are careful not to overstate the strength of this inference given the limitations enumerated below.

The forced swimming test is a long-established and validated paradigm for assessing aerobic endurance in rodents [26, 28], and the progressive improvement observed across sessions in all groups is consistent with a generalized training effect, plausibly involving increases in maximal oxygen utilization and skeletal muscle efficiency [17, 18]. That treated groups outperformed controls is compatible with trimetazidine's established shift toward glucose oxidation [4] and with ranolazine's reduction of late-sodium-current-mediated calcium overload, which is mechanistically linked to preserved diastolic function and coronary microvascular flow in prior experimental and clinical work [8–10, 22]. We emphasize, however, that the present study measured none of these proposed mechanisms directly: no biochemical markers of substrate utilization (lactate, glucose, free fatty acids), no measures of myocardial ATP content or mitochondrial function, and no direct assessment of maximal oxygen consumption (VO2max) were obtained, despite VO2max being invoked conceptually in our framing of the exercise-physiology background. This is a direct and important limitation, and the causal chain linking ranolazine's known electrophysiological mechanism to the swimming-performance differences observed here remains, on the basis of the present data alone, inferential rather than demonstrated.

Mechanisms of performance enhancement: a working hypothesis, not a tested mechanism

A pattern in our data that we consider worth highlighting, while stressing that it constitutes a hypothesis derived from the temporal sequence of effects rather than a directly tested mechanism, is the timing of the divergence between treated and control animals. Significant differences in swimming duration emerged as early as session 3, several days before any echocardiographically detectable structural change was apparent; the earliest significant within-group structural changes were detected only at the day-15 assessment. This temporal dissociation is at least compatible with an early, non-structural (metabolic or hemodynamic) contribution to the initial performance difference, with structural remodeling becoming detectable only later in the protocol. Ranolazine is reported to approach steady-state plasma concentrations within approximately three days in mammals, and its principal mechanism, inhibition of the late sodium current, is mechanistically linked to reduced intracellular calcium overload, lower diastolic wall tension, and improved microvascular perfusion even without overt ischemia [8, 9]; a parallel acute mechanism for trimetazidine involves the substrate shift toward glucose oxidation [4]. We propose, as a hypothesis for future, adequately instrumented study rather than as an established conclusion, that an initial period dominated by metabolic and hemodynamic effects (days 1–14) may be followed by a second period (days 15–30) in which structural adaptation, increased septal thickness and calculated LV mass without decline in systolic function, becomes superimposed on, and potentially reinforces, the ongoing functional benefit (Fig. 5). We reiterate that this biphasic model is not validated by direct biochemical or histological data in the present study and should be treated as a framework to be tested, not a finding to be cited as established.

Fig. 5.

Proposed biphasic model of performance modification by ranolazine and trimetazidine, presented as a working hypothesis derived from the temporal pattern of the present functional and structural observations rather than as a directly tested mechanism. Phase 1 (days 1–14) depicts hypothesized acute metabolic and hemodynamic effects; Phase 2 (days 15–30) depicts the structural echocardiographic changes (increased interventricular septal thickness and calculated LV mass, with preserved ejection fraction and fractional shortening) observed in this study. INaL, late sodium current; 3-KAT, 3-ketoacyl-CoA thiolase; IVSD/IVSS, interventricular septal thickness in diastole/systole; LV, left ventricle; EF, ejection fraction; FS, fractional shortening; ATP, adenosine triphosphate

Structural echocardiographic changes and the growth-phase confound

The increases in IVSD, IVSS, and calculated LV mass observed in our cohort are, in isolation, compatible with physiological cardiac remodeling of the type described in trained human athletes ("athlete's heart") [35]. However, a critical observation that must temper this interpretation is that statistically significant within-group increases in these same parameters were also detected in the saline-treated control group. Because the rats were 8–10 weeks of age at study entry and therefore had not yet reached full skeletal and cardiac maturity, a meaningful component of the structural change observed across all three groups, including controls, is plausibly attributable to normal somatic and cardiac growth occurring in parallel with the experimental protocol rather than to exercise training or drug treatment specifically. We cannot, on the basis of the present design, partition the observed increase in calculated LV mass into growth-related, exercise-related, and drug-related components; doing so would require either an additional untreated, unexercised, age-matched comparator group or serial measurements extended well beyond the point of skeletal maturity. We regard this as one of the more consequential limitations of the present study and recommend that any confirmatory follow-up investigation either employ fully mature animals or incorporate a sedentary, age-matched control arm capable of isolating the contribution of ongoing growth.

With that important caveat stated, we note that the magnitude of the increase in IVSS and calculated LV mass was numerically larger in the ranolazine and trimetazidine groups than in controls, and that ranolazine additionally showed significant change in EF, FS, LVIDS, AoS, AV Vmax, and ESV that was not mirrored in the control group, raising the possibility of a treatment-related contribution superimposed on the background of growth-related change. No instance of EF or FS decline was observed in any group, arguing against pathological or maladaptive remodeling [34], and the synergy between pharmacological substrate modulation and training-induced angiogenesis or reduced oxidative stress represents one plausible, but again untested, explanation for the additional structural change seen in the treated groups [19, 36]. We did not measure heart weight gravimetrically or perform histological assessment of cardiomyocyte size or interstitial fibrosis at necropsy; in the absence of such corroboration, the echocardiographic findings reported here should be interpreted as evidence of functional cardiac remodeling detectable by ultrasound, not as definitive proof of cellular hypertrophy.

Heart rate, arrhythmia, and species-specific electrophysiology

Heart rate, sampled from the echocardiographic Doppler trace at each of the three imaging sessions, did not differ significantly within or between groups at any time point, consistent with ranolazine's established lack of clinically significant chronotropic effect in humans [9, 24] and providing some reassurance against a gross drug-induced change in resting heart rate in this model. We did not, however, perform continuous electrocardiographic or Holter monitoring at any point in the protocol, and the absence of a heart-rate change on intermittent, anesthetized echocardiographic sampling cannot exclude transient or exercise-induced arrhythmia that would only be detectable with continuous ambulatory recording; we regard the lack of ECG/Holter data as one of the more important omissions of the present design. We also wish to address explicitly the related question of whether our rat model is mechanistically well suited to detect ranolazine-associated arrhythmia at all. Ranolazine's regulatory labeling and the preponderance of its supporting arrhythmia literature relate specifically to supraventricular arrhythmic substrates (for example, atrial fibrillation), reflecting the relatively atrial-selective expression of certain of its target currents, whereas data supporting a comparable ventricular antiarrhythmic or proarrhythmic effect are considerably more limited [8, 9]. Furthermore, the molecular composition of cardiac potassium and sodium channel isoforms, and their relative density between the atrial and ventricular myocardium, differs quantitatively between the rat and the human heart, as does the overall repolarization reserve of the rodent ventricle [37]. These species- and chamber-specific electrophysiological differences mean that an absence of detectable arrhythmia in this rat model, even had continuous monitoring been performed, would not necessarily generalize to the human, or specifically the human athlete's heart, and conversely that any arrhythmic signal detected in rats would require independent confirmation in a species with human-like repolarization physiology before clinical or regulatory inference could be drawn.

Translational considerations: how well does a forced-exercise rat model approximate a doping athlete?

A further question concerns how well the present model approximates the scenario of an athlete using ranolazine for performance enhancement, and we consider this an important caveat rather than a minor technical point. The forced swimming paradigm used here is an aversive, escape-motivated stimulus rather than the voluntary, goal-directed training undertaken by competitive athletes [28, 29], and the resulting performance measure, time to exhaustion under an externally imposed load, is influenced by acute stress reactivity, motivational state, and thermoregulatory demand in addition to genuine aerobic capacity; these confounders cannot be fully separated using the forced swimming test in isolation, and we acknowledge this directly in response to the specific concern that this paradigm may not isolate pure aerobic performance. In addition, drug absorption, distribution, metabolism, and excretion differ quantitatively between rats and humans, body-mass-normalized dosing does not guarantee equivalent target-tissue exposure between species, and the rats in this study were healthy, sedentary-bred laboratory animals rather than highly trained athletes with established cardiovascular adaptations. For these reasons, we regard the present findings as informative for hypothesis generation regarding ranolazine's biological plausibility as an ergogenic agent, but as insufficient, on their own, to support direct extrapolation to human athletic doping risk.

Motor coordination and body weight

Rotarod performance showed no significant change within or between groups at any time point, suggesting that neither drug measurably impaired motor coordination or balance, consistent with ranolazine's established central nervous system safety profile [15, 38] and with its reportedly minimal direct effects on astrocytes and neurons in primary culture [38]. This pattern indicates that the performance differences observed in the forced swimming test are more plausibly cardiovascular or metabolic than neuromuscular in origin, although, as above, this inference is not directly tested by the present data. Body weight increased over the study period in all groups without significant between-group differences (Table 1); the early plateau followed by later gain most likely reflects a combination of exercise-related metabolic demand and ongoing somatic growth in these still-maturing animals [39, 40], rather than a specific drug effect.

Finally, we note that all behavioral and imaging procedures were performed during the light phase (09:00–11:00), the inactive period for nocturnal rats. While this timing was applied uniformly across all three groups, such that any resulting suppression of absolute performance should affect all groups equally and should not, in principle, bias the relative between-group comparisons that form the basis of our conclusions, we cannot fully exclude a drug-by-circadian-phase interaction, and we did not test ranolazine or trimetazidine administration during the rats' active dark phase. We regard this as an additional, if secondary, limitation.

Anti-doping relevance: a hypothesis-generating signal, not a basis for regulatory action

The present results indicate that ranolazine produced exercise-performance changes in healthy rats that paralleled, and in later sessions exceeded, those of trimetazidine, an agent already prohibited by WADA on the basis of analogous rodent and human evidence [5, 6, 20], together with modest, partially growth-confounded echocardiographic changes. We consider this pattern sufficiently interesting to merit further, more deeply mechanistic investigation, but we no longer characterize it as direct evidence warranting anti-doping scrutiny of ranolazine; rather, we frame it explicitly as a hypothesis-generating signal that should motivate confirmatory studies, incorporating biochemical, histological, electrocardiographic, and direct oxygen-consumption endpoints, and ideally extending to mature animals or human pharmacodynamic studies, before any conclusions relevant to anti-doping policy can reasonably be drawn.

Limitations

Several limitations warrant explicit acknowledgement. Most fundamentally, although this was a randomized, blinded, controlled experimental study rather than an observational one, it remains non-mechanistic in scope: no biochemical or metabolic data (lactate, ATP content, substrate utilization, or maximal oxygen consumption) were obtained, so the presumed metabolic basis of the functional differences observed cannot be confirmed. Echocardiographically estimated LV mass was not corroborated by gravimetric heart weight or histological assessment of cardiomyocyte size or fibrosis, and no tissue was banked at euthanasia for retrospective analysis; the structural findings should therefore be interpreted as evidence of ultrasound-detectable remodeling rather than confirmed cellular hypertrophy. Because the rats were 8–10 weeks old at enrollment and had not reached full somatic and cardiac maturity, and because significant increases in IVSD, IVSS, and calculated LV mass were also observed in the saline control group, a meaningful component of the structural change is plausibly attributable to normal growth rather than to exercise or drug treatment; the present design cannot partition growth-related, exercise-related, and drug-related contributions, and we regard this as one of the more consequential limitations of the study. No electrocardiographic, Holter, or telemetric monitoring was performed at any point; heart rate was sampled only intermittently from the echocardiographic Doppler trace under ketamine-xylazine anesthesia, which is inadequate to exclude exercise- or drug-induced arrhythmia, a particularly relevant gap given ranolazine’s electrophysiological activity. Only a single dose of each drug was evaluated, selected from precedent in the rodent pharmacology literature rather than from a dose-ranging study performed within the present protocol, so no conclusions regarding dose-dependency of the observed effects can be drawn.

Several further limitations bear on the generalizability of these findings. The forced swimming test, although an established and validated index of aerobic endurance, is influenced by acute stress reactivity, motivational state, and thermoregulatory demand in addition to genuine aerobic capacity, and reflects an aversive, escape-motivated form of exercise that differs qualitatively from the voluntary training undertaken by human athletes; these confounders cannot be fully disentangled from true metabolic or cardiovascular adaptation using this paradigm alone. Only male rats were studied, to avoid the confounding influence of estrous-cycle variability on exercise performance, which precludes any assessment of sex-based differences that may be clinically relevant given known sex differences in cardiac electrophysiology and drug pharmacokinetics. All behavioral testing and imaging were performed during the light (inactive) phase for these nocturnal animals; although this timing was applied uniformly across groups, a drug-by-circadian-phase interaction cannot be fully excluded. Species differences in cardiac ion channel expression and density between the rat and human heart, including differences in repolarizing potassium currents between the atrial and ventricular myocardium, limit the degree to which electrophysiological findings in this model can be generalized to humans, and the rat model itself differs from human athletic doping scenarios in training modality, drug pharmacokinetics, and baseline cardiovascular conditioning. Finally, this was a single-center study with a modest, pilot-level sample size (trimetazidine and ranolazine, n = 8 each; control, n = 6, after the unplanned loss of two animals to anesthesia-related mortality described above), and the resulting unequal group sizes modestly reduce the power of the control-group comparisons relative to the originally planned n = 8 per group; non-parametric statistical methods were applied a priori based on Shapiro–Wilk testing, which we consider statistically appropriate given the observed data distributions, although we acknowledge that a parametric sensitivity analysis (for example, repeated-measures ANOVA with Bonferroni correction) could be performed on the underlying individual-animal dataset. Cumulatively, these limitations mean that the present findings should be regarded as hypothesis-generating and exploratory rather than as a definitive or mechanistically validated characterization of ranolazine’s ergogenic potential.

Conclusion

In this exploratory, hypothesis-generating study, chronic ranolazine administration in healthy rats was associated with greater forced-swimming endurance than saline control, with a numerically larger effect than trimetazidine in the later sessions of a 30-day protocol, together with modest echocardiographic increases in septal thickness and calculated LV mass that were also detectable, to a lesser degree, in untreated control animals and that were not corroborated by biochemical, histological, or electrocardiographic data. Motor coordination, heart rate, and systolic function were preserved throughout. These observations are compatible with, but do not establish, a metabolic and structural basis for ranolazine's ergogenic potential, and they should be understood as motivating, rather than constituting, a case for further mechanistic investigation. We recommend that future studies incorporate biochemical markers of substrate metabolism, gravimetric and histological validation of cardiac structure, continuous electrocardiographic monitoring, direct measurement of maximal oxygen consumption, fully mature animals or an additional sedentary age-matched control arm, and, ideally, both sexes, before any inference regarding ranolazine's relevance to human athletic performance or anti-doping policy is drawn.

Abbreviations

AoD

Aortic diastolic diameter

AoS

Aortic systolic diameter

ATP

Adenosine triphosphate

AV Vmax

Aortic valve peak transvalvular velocity

AVMA

American Veterinary Medical Association

CAD

Coronary artery disease

EDV

End-diastolic volume

EF

Ejection fraction

ESV

End-systolic volume

FDA

Food and Drug Administration

FS

Fractional shortening

IACUC

Institutional Animal Care and Use Committee

ICC

Intraclass correlation coefficient

INaL

Late sodium current

IVSD

Interventricular septal diastolic thickness

IVSS

Interventricular septal systolic thickness

LV

Left ventricular

LVIDD

Left ventricular internal diastolic diameter

LVIDS

Left ventricular internal systolic diameter

LVPWD

Left ventricular posterior wall diastolic thickness

LVPWS

Left ventricular posterior wall systolic thickness

PV Vmax

Pulmonary valve peak transvalvular velocity

VO2max

Maximum oxygen utilization

WADA

World Anti-Doping Agency

Authors’ contributions

Conceptualization: Şahhan KILIÇ, Akın TORUN, Mehmet UZUN. Methodology: Şahhan KILIÇ, Süha ASAL, Cumaali DEMİRTAŞ, Serkan DİLMEN. Investigation: Şahhan KILIÇ, Süha ASAL, Akın TORUN. Data Curation: Cumaali DEMİRTAŞ, Serkan DİLMEN. Formal Analysis: Akın TORUN, Ahmet Lütfullah ORHAN, Mehmet UZUN. Writing – Original Draft: Şahhan KILIÇ, Süha ASAL. Writing – Review & Editing: Tufan ÇINAR, Mehmet UZUN. Supervision: Tufan ÇINAR, Mehmet UZUN, Ahmet Lütfullah ORHAN. Project Administration: Şahhan KILIÇ.

Funding

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

Data availability

The individual-animal dataset supporting the findings of this study, including the raw values underlying the medians and interquartile ranges reported throughout, is available from the corresponding author upon reasonable request. No biological tissue samples were banked during the original protocol; this is addressed explicitly in the Limitations section.

Declarations

Ethics approval and consent to participate

The study was approved by the Institutional Animal Care and Use Committee (IACUC) of Health Sciences University (approval date: 18 March 2023; approval number: 03/04) and was conducted in accordance with the ARRIVE guidelines (version 2.0) for ethical animal research.

Competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.