Article View

Age is a major biological determinant of reproductive development in stallions, but its effects are not necessarily expressed uniformly across testicular size, ejaculatory output, conventional semen characteristics, sperm morphology, and objective sperm-motion variables. This issue is particularly relevant in the Mangalarga Marchador, an important Brazilian horse breed for which age-related testicular growth has been described, but an integrated field-based characterisation of gonadal, seminal, and computer-assisted sperm-analysis traits remains limited. The objective of this cross-sectional study was therefore to characterise age-related reproductive remodelling in Mangalarga Marchador stallions by integrating body condition, testicular biometry, expected daily sperm production, conventional semen evaluation, sperm morphology, portable computer-assisted sperm analysis (CASA), and principal component analysis (PCA). We hypothesised that age-related differences would be expressed more strongly in gonadal development and ejaculatory output than in all sperm functional domains, supporting a multidimensional rather than uniformly linear model of reproductive maturation. The study included 443 provisionally or definitively registered Mangalarga Marchador stallions evaluated under commercial field conditions in different regions of Brazil during 2024. Stallions ranged from 23 to 316 months of age and were classified into four biologically and breed-relevant age classes: <36 months (n = 52), 36–59 months (n = 168), 60–83 months (n = 77), and ≥84 months (n = 146). All andrological and biometric examinations were performed by a single trained examiner. Body condition score (BCS), body weight, total testicular width, total testicular volume, and expected daily sperm production were available for the entire population. Semen was collected using a Colorado-model artificial vagina. Total and gel-free ejaculate volumes were determined, and conventional microscopic assessment included total motility, progressive motility, vigour, and sperm concentration. Sperm morphology was evaluated by quantifying major defects, minor defects, total defects, and morphologically normal spermatozoa. A 10-µL aliquot was additionally analysed with the portable Mace®/Macebox® CASA platform, which generated total and progressive motility, software-derived vigour, concentration, curvilinear velocity (VCL), average path velocity (VAP), straight-line velocity (VSL), straightness (STR), linearity (LIN), amplitude of lateral head displacement (ALH), and beat-cross frequency (BCF). Because the dataset originated from routine commercial field evaluations, endpoint availability differed among stallions; consequently, the maximum number of valid observations was retained for each analysis and variable-specific sample sizes were reported transparently (Table 1). Age class was treated as the principal explanatory factor. Distributional assumptions were evaluated using the Shapiro–Wilk test and graphical inspection. Non-Gaussian variables were summarised as medians and interquartile ranges and compared using the Kruskal–Wallis test, with Dunn’s adjusted pairwise comparisons when appropriate. Approximately Gaussian variables were compared by one-way analysis of variance followed by Tukey’s test, and categorical variables were assessed using chi-square or Fisher’s exact tests, as applicable. PCA was performed on standardised variables with sufficient completeness to explore the latent organisation of the seminal phenotype. Statistical significance was set at P < 0.05. Body condition was comparable among age classes. Median BCS was 5.0 in every group, and no significant age-class effect was detected (P = 0.452), despite a progressive increase in body weight from a median of 390.0 kg in stallions <36 months to 436.0 kg in stallions ≥84 months (P < 0.001). This comparability indicates that systematic differences in body reserves were unlikely to account for the reproductive patterns attributed to age. In contrast, gonadal measurements showed a clear age-related gradient (Table 2). Median total testicular width increased from 9.2 in the youngest group to 10.4 in the oldest group, and median total testicular volume increased from 252.4 to 338.5 across the same age classes (both P < 0.001). Expected daily sperm production followed a similar pattern, increasing from 4.80 in stallions <36 months to 6.86 in stallions ≥84 months (P < 0.001). These findings demonstrate progressive expansion of testicular dimensions and estimated spermatogenic capacity across the age structure evaluated. The most pronounced seminal differences were observed in ejaculatory output (Table 3; Figure 1). Median total ejaculate volume increased progressively from 30.0 mL in stallions <36 months to 38.0 mL at 36–59 months, 54.5 mL at 60–83 months, and 64.5 mL at ≥84 months (P < 0.001). Gel-free ejaculate volume showed the same general pattern, with medians of 28.2, 33.5, 50.5, and 50.0 mL, respectively (P < 0.001). The increase in the gel-free fraction indicates that the age-associated expansion of ejaculatory output was not attributable solely to variation in the gel component. Conventional total motility showed a modest global age-class effect, with medians ranging from 75.0% to 80.0% (P = 0.045), but no broad parallel improvement was detected in the remaining routine semen-quality variables. Progressive motility did not differ significantly among age classes (P = 0.072), and conventional vigour remained at a median of 3.5 in all groups (P = 0.504). Sperm concentration was also similar, with medians of 203.8, 206.2, 210.0, and 165.0 ×10⁶ spermatozoa/mL from the youngest to the oldest age class (P = 0.512). Thus, older stallions produced larger ejaculates without a corresponding increase in sperm concentration per millilitre. Sperm morphology was comparatively stable throughout the age range (Table 4). No significant differences were detected in major defects (P = 0.593), minor defects (P = 0.985), total defects (P = 0.759), or morphologically normal spermatozoa (P = 0.759). Median total sperm defects ranged from 18.2% to 22.0%, while median normal morphology ranged from 78.0% to 81.8%. These results indicate that the age-related increase in gonadal dimensions and ejaculatory output was not accompanied by a general deterioration or improvement in sperm structural quality. Objective CASA-derived variables also remained statistically stable across age classes (Table 5; Figure 1). CASA total motility (P = 0.909), progressive motility (P = 0.801), software-derived vigour (P = 0.260), VCL (P = 0.367), VAP (P = 0.449), VSL (P = 0.439), STR (P = 0.899), LIN (P = 0.096), ALH (P = 0.309), and BCF (P = 0.325) showed no significant age-class effects. Although numerical dispersion was greater for several variables in the youngest group, particularly for velocity-based measurements, the overall architecture of sperm motion was conserved. The contrast between marked increases in ejaculate volume and the stability of morphology and CASA-derived kinematics provides direct support for a domain-dependent model of reproductive maturation. PCA was conducted in 155 stallions with complete observations for gel-free ejaculate volume, sperm concentration, CASA progressive motility, VAP, LIN, and major sperm defects (Table 6; Figure 2). Bartlett’s test of sphericity was significant (χ² = 140.43, 15 degrees of freedom, P < 0.001), whereas the Kaiser–Meyer–Olkin index was 0.500, indicating that the PCA should be regarded as an exploratory representation of the correlation structure. Three biologically interpretable components jointly explained 72.6% of the total variance. PC1 explained 30.4% and was driven mainly by CASA progressive motility and VAP, defining a sperm-motion efficiency domain. PC2 explained 24.4% and was characterised by a positive loading for gel-free ejaculate volume and an opposite loading for sperm concentration, defining an ejaculatory-output or dilution domain. PC3 explained 17.9% and was dominated by major sperm defects, representing a structural-abnormality domain. The separation of these axes confirms that ejaculatory output, sperm motion, and sperm morphology constitute related but partially independent dimensions of the seminal phenotype. Taken together, the findings show that age-related reproductive remodelling in Mangalarga Marchador stallions was expressed principally through progressive increases in testicular dimensions, estimated spermatogenic capacity, and ejaculatory output. These changes occurred in animals with comparable body condition and were not accompanied by broad age-related differences in sperm concentration, progressive motility, vigour, morphology, or CASA-derived kinematic architecture. Because the study was cross-sectional, the results describe differences among age classes rather than longitudinal changes within the same stallions. Nevertheless, the large field-based population and integrated analytical approach demonstrate that reproductive maturation in this breed should not be interpreted as a uniform improvement in overall semen quality. Instead, age selectively influenced the output domain, while structural and motion-related sperm domains remained comparatively conserved. Field breeding-soundness evaluation of Mangalarga Marchador stallions should therefore integrate age, body condition, testicular biometry, ejaculate volume, sperm concentration, conventional motility, morphology, and, when available, objective CASA-derived variables rather than relying on any single endpoint.